From 784bdf3bb694b256fcd6120b93e8947a84249a3a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 29 Jul 2016 16:32:30 +0200 Subject: [PATCH 001/235] futex: Assume all mappings are private on !MMU systems To quote Rick why there is no need for shared mapping on !MMU systems: |With MMU, shared futex keys need to identify the physical backing for |a memory address because it may be mapped at different addresses in |different processes (or even multiple times in the same process). |Without MMU this cannot happen. You only have physical addresses. So |the "private futex" behavior of using the virtual address as the key |is always correct (for both shared and private cases) on nommu |systems. This patch disables the FLAGS_SHARED in a way that allows the compiler to remove that code. [bigeasy: Added changelog ] Reported-by: Rich Felker Signed-off-by: Thomas Gleixner Signed-off-by: Sebastian Andrzej Siewior Cc: Andrew Morton Link: http://lkml.kernel.org/r/20160729143230.GA21715@linutronix.de Signed-off-by: Thomas Gleixner --- kernel/futex.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/kernel/futex.c b/kernel/futex.c index 33664f70e2d2..46cb3a301bc1 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -179,7 +179,15 @@ int __read_mostly futex_cmpxchg_enabled; * Futex flags used to encode options to functions and preserve them across * restarts. */ -#define FLAGS_SHARED 0x01 +#ifdef CONFIG_MMU +# define FLAGS_SHARED 0x01 +#else +/* + * NOMMU does not have per process address space. Let the compiler optimize + * code away. + */ +# define FLAGS_SHARED 0x00 +#endif #define FLAGS_CLOCKRT 0x02 #define FLAGS_HAS_TIMEOUT 0x04 @@ -405,6 +413,16 @@ static void get_futex_key_refs(union futex_key *key) if (!key->both.ptr) return; + /* + * On MMU less systems futexes are always "private" as there is no per + * process address space. We need the smp wmb nevertheless - yes, + * arch/blackfin has MMU less SMP ... + */ + if (!IS_ENABLED(CONFIG_MMU)) { + smp_mb(); /* explicit smp_mb(); (B) */ + return; + } + switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { case FUT_OFF_INODE: ihold(key->shared.inode); /* implies smp_mb(); (B) */ @@ -436,6 +454,9 @@ static void drop_futex_key_refs(union futex_key *key) return; } + if (!IS_ENABLED(CONFIG_MMU)) + return; + switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { case FUT_OFF_INODE: iput(key->shared.inode); From 3b2c1710fac7fb278b760d1545e637cbb5ea5b5b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= Date: Wed, 13 Jul 2016 16:32:03 +0300 Subject: [PATCH 002/235] drm/i915: Wait up to 3ms for the pcu to ack the cdclk change request on SKL MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Bspec tells us to keep bashing the PCU for up to 3ms when trying to inform it about an upcoming change in the cdclk frequency. Currently we only keep at it for 15*10usec (+ whatever delays gets added by the sandybridge_pcode_read() itself). Let's change the limit to 3ms. I decided to keep 10 usec delay per iteration for now, even though the spec doesn't really tell us to do that. Cc: stable@vger.kernel.org Fixes: 5d96d8afcfbb ("drm/i915/skl: Deinit/init the display at suspend/resume") Cc: David Weinehall Signed-off-by: Ville Syrjälä Link: http://patchwork.freedesktop.org/patch/msgid/1468416723-23440-1-git-send-email-ville.syrjala@linux.intel.com Tested-by: David Weinehall Reviewed-by: Chris Wilson (cherry picked from commit 848496e5902833600f7992f4faa82dc1546051ba) Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_display.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 9f56ac11975f..e5c46a6ec65b 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5691,15 +5691,7 @@ static bool skl_cdclk_pcu_ready(struct drm_i915_private *dev_priv) static bool skl_cdclk_wait_for_pcu_ready(struct drm_i915_private *dev_priv) { - unsigned int i; - - for (i = 0; i < 15; i++) { - if (skl_cdclk_pcu_ready(dev_priv)) - return true; - udelay(10); - } - - return false; + return _wait_for(skl_cdclk_pcu_ready(dev_priv), 3000, 10) == 0; } static void skl_set_cdclk(struct drm_i915_private *dev_priv, int cdclk, int vco) From a7b4667a00025ac28300737c868bd4818b6d8c4d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= Date: Mon, 23 May 2016 17:42:48 +0300 Subject: [PATCH 003/235] drm/i915: Never fully mask the the EI up rps interrupt on SNB/IVB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit SNB (and IVB too I suppose) starts to misbehave if the GPU gets stuck in an infinite batch buffer loop. The GPU apparently hogs something critical and CPUs start to lose interrupts and whatnot. We can keep the system limping along by unmasking some interrupts in GEN6_PMINTRMSK. The EI up interrupt has been previously chosen for that task, so let's never mask it. v2: s/gen6_rps_pm_mask/gen6_sanitize_rps_pm_mask/ (Chris) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93122 Signed-off-by: Ville Syrjälä Reviewed-by: Chris Wilson Signed-off-by: Chris Wilson Link: http://patchwork.freedesktop.org/patch/msgid/1464014568-4529-1-git-send-email-ville.syrjala@linux.intel.com Cc: stable@vger.kernel.org (cherry picked from commit 12c100bfa5d9103b6c4d43636fee09c31e75605a) Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_pm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 5a8ee0c76593..5ca68d2ed7d4 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -4892,7 +4892,8 @@ void gen6_rps_idle(struct drm_i915_private *dev_priv) else gen6_set_rps(dev_priv, dev_priv->rps.idle_freq); dev_priv->rps.last_adj = 0; - I915_WRITE(GEN6_PMINTRMSK, 0xffffffff); + I915_WRITE(GEN6_PMINTRMSK, + gen6_sanitize_rps_pm_mask(dev_priv, ~0)); } mutex_unlock(&dev_priv->rps.hw_lock); From 0596a9048bf2aca2a74b312493f39e4d5ac3b653 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 14 Jun 2016 14:18:27 +0100 Subject: [PATCH 004/235] Btrfs: add missing check for writeback errors on fsync When we start an fsync we start ordered extents for all delalloc ranges. However before attempting to log the inode, we only wait for those ordered extents if we are not doing a full sync (bit BTRFS_INODE_NEEDS_FULL_SYNC is set in the inode's flags). This means that if an ordered extent completes with an IO error before we check if we can skip logging the inode, we will not catch and report the IO error to user space. This is because on an IO error, when the ordered extent completes we do not update the inode, so if the inode was not previously updated by the current transaction we end up not logging it through calls to fsync and therefore not check its mapping flags for the presence of IO errors. Fix this by checking for errors in the flags of the inode's mapping when we notice we can skip logging the inode. This caused sporadic failures in the test generic/331 (which explicitly tests for IO errors during an fsync call). Signed-off-by: Filipe Manana Reviewed-by: Liu Bo --- fs/btrfs/file.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index bcfb4a27ddd4..6f049109fffe 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2033,6 +2033,14 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); + /* + * An ordered extent might have started before and completed + * already with io errors, in which case the inode was not + * updated and we end up here. So check the inode's mapping + * flags for any errors that might have happened while doing + * writeback of file data. + */ + ret = btrfs_inode_check_errors(inode); inode_unlock(inode); goto out; } From 801bec365e0e19f2ba066cd3e25a67dee21b4aae Mon Sep 17 00:00:00 2001 From: Robbie Ko Date: Tue, 23 Jun 2015 18:39:46 +0800 Subject: [PATCH 005/235] Btrfs: send, fix failure to move directories with the same name around When doing an incremental send we can end up not moving directories that have the same name. This happens when the same parent directory has different child directories with the same name in the parent and send snapshots. For example, consider the following scenario: Parent snapshot: . (ino 256) |---- d/ (ino 257) | |--- p1/ (ino 258) | |---- p1/ (ino 259) Send snapshot: . (ino 256) |--- d/ (ino 257) |--- p1/ (ino 259) |--- p1/ (ino 258) The directory named "d" (inode 257) has in both snapshots an entry with the name "p1" but it refers to different inodes in both snapshots (inode 258 in the parent snapshot and inode 259 in the send snapshot). When attempting to move inode 258, the operation is delayed because its new parent, inode 259, was not yet moved/renamed (as the stream is currently processing inode 258). Then when processing inode 259, we also end up delaying its move/rename operation so that it happens after inode 258 is moved/renamed. This decision to delay the move/rename rename operation of inode 259 is due to the fact that the new parent inode (257) still has inode 258 as its child, which has the same name has inode 259. So we end up with inode 258 move/rename operation waiting for inode's 259 move/rename operation, which in turn it waiting for inode's 258 move/rename. This results in ending the send stream without issuing move/rename operations for inodes 258 and 259 and generating the following warnings in syslog/dmesg: [148402.979747] ------------[ cut here ]------------ [148402.980588] WARNING: CPU: 14 PID: 4117 at fs/btrfs/send.c:6177 btrfs_ioctl_send+0xe03/0xe51 [btrfs] [148402.981928] Modules linked in: btrfs crc32c_generic xor raid6_pq acpi_cpufreq tpm_tis ppdev tpm parport_pc psmouse parport sg pcspkr i2c_piix4 i2c_core evdev processor serio_raw button loop autofs4 ext4 crc16 jbd2 mbcache sr_mod cdrom sd_mod ata_generic virtio_scsi ata_piix libata virtio_pci virtio_ring virtio e1000 scsi_mod floppy [last unloaded: btrfs] [148402.986999] CPU: 14 PID: 4117 Comm: btrfs Tainted: G W 4.6.0-rc7-btrfs-next-31+ #1 [148402.988136] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS by qemu-project.org 04/01/2014 [148402.988136] 0000000000000000 ffff88022139fca8 ffffffff8126b42c 0000000000000000 [148402.988136] 0000000000000000 ffff88022139fce8 ffffffff81052b14 000018212139fac8 [148402.988136] ffff88022b0db400 0000000000000000 0000000000000001 0000000000000000 [148402.988136] Call Trace: [148402.988136] [] dump_stack+0x67/0x90 [148402.988136] [] __warn+0xc2/0xdd [148402.988136] [] warn_slowpath_null+0x1d/0x1f [148402.988136] [] btrfs_ioctl_send+0xe03/0xe51 [btrfs] [148402.988136] [] btrfs_ioctl+0x14f/0x1f81 [btrfs] [148402.988136] [] ? arch_local_irq_save+0x9/0xc [148402.988136] [] ? __lock_is_held+0x3c/0x57 [148402.988136] [] vfs_ioctl+0x18/0x34 [148402.988136] [] do_vfs_ioctl+0x550/0x5be [148402.988136] [] ? __fget+0x6b/0x77 [148402.988136] [] ? __fget_light+0x62/0x71 [148402.988136] [] SyS_ioctl+0x57/0x79 [148402.988136] [] entry_SYSCALL_64_fastpath+0x18/0xa8 [148402.988136] [] ? trace_hardirqs_off_caller+0x3f/0xaa [148403.011373] ---[ end trace a4539270c8056f8b ]--- [148403.012296] ------------[ cut here ]------------ [148403.013071] WARNING: CPU: 14 PID: 4117 at fs/btrfs/send.c:6194 btrfs_ioctl_send+0xe19/0xe51 [btrfs] [148403.014447] Modules linked in: btrfs crc32c_generic xor raid6_pq acpi_cpufreq tpm_tis ppdev tpm parport_pc psmouse parport sg pcspkr i2c_piix4 i2c_core evdev processor serio_raw button loop autofs4 ext4 crc16 jbd2 mbcache sr_mod cdrom sd_mod ata_generic virtio_scsi ata_piix libata virtio_pci virtio_ring virtio e1000 scsi_mod floppy [last unloaded: btrfs] [148403.019708] CPU: 14 PID: 4117 Comm: btrfs Tainted: G W 4.6.0-rc7-btrfs-next-31+ #1 [148403.020104] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS by qemu-project.org 04/01/2014 [148403.020104] 0000000000000000 ffff88022139fca8 ffffffff8126b42c 0000000000000000 [148403.020104] 0000000000000000 ffff88022139fce8 ffffffff81052b14 000018322139fac8 [148403.020104] ffff88022b0db400 0000000000000000 0000000000000001 0000000000000000 [148403.020104] Call Trace: [148403.020104] [] dump_stack+0x67/0x90 [148403.020104] [] __warn+0xc2/0xdd [148403.020104] [] warn_slowpath_null+0x1d/0x1f [148403.020104] [] btrfs_ioctl_send+0xe19/0xe51 [btrfs] [148403.020104] [] btrfs_ioctl+0x14f/0x1f81 [btrfs] [148403.020104] [] ? arch_local_irq_save+0x9/0xc [148403.020104] [] ? __lock_is_held+0x3c/0x57 [148403.020104] [] vfs_ioctl+0x18/0x34 [148403.020104] [] do_vfs_ioctl+0x550/0x5be [148403.020104] [] ? __fget+0x6b/0x77 [148403.020104] [] ? __fget_light+0x62/0x71 [148403.020104] [] SyS_ioctl+0x57/0x79 [148403.020104] [] entry_SYSCALL_64_fastpath+0x18/0xa8 [148403.020104] [] ? trace_hardirqs_off_caller+0x3f/0xaa [148403.038981] ---[ end trace a4539270c8056f8c ]--- There's another issue caused by similar (but more complex) changes in the directory hierarchy that makes move/rename operations fail, described with the following example: Parent snapshot: . |---- a/ (ino 262) | |---- c/ (ino 268) | |---- d/ (ino 263) |---- ance/ (ino 267) |---- e/ (ino 264) |---- f/ (ino 265) |---- ance/ (ino 266) Send snapshot: . |---- a/ (ino 262) |---- c/ (ino 268) | |---- ance/ (ino 267) | |---- d/ (ino 263) | |---- ance/ (ino 266) | |---- f/ (ino 265) |---- e/ (ino 264) When the inode 265 is processed, the path for inode 267 is computed, which at that time corresponds to "d/ance", and it's stored in the names cache. Later on when processing inode 266, we end up orphanizing (renaming to a name matching the pattern o--) inode 267 because it has the same name as inode 266 and it's currently a child of the new parent directory (inode 263) for inode 266. After the orphanization and while we are still processing inode 266, a rename operation for inode 266 is generated. However the source path for that rename operation is incorrect because it ends up using the old, pre-orphanization, name of inode 267. The no longer valid name for inode 267 was previously cached when processing inode 265 and it remains usable and considered valid until the inode currently being processed has a number greater than 267. This resulted in the receiving side failing with the following error: ERROR: rename d/ance/ance -> d/ance failed: No such file or directory So fix these issues by detecting such circular dependencies for rename operations and by clearing the cached name of an inode once the inode is orphanized. A test case for fstests will follow soon. Signed-off-by: Robbie Ko Signed-off-by: Filipe Manana [Rewrote change log to be more detailed and organized, and improved comments] Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 100 +++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 95 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index b71dd298385c..993e1bab0a6b 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -231,7 +231,6 @@ struct pending_dir_move { u64 parent_ino; u64 ino; u64 gen; - bool is_orphan; struct list_head update_refs; }; @@ -1861,7 +1860,8 @@ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen, * was already unlinked/moved, so we can safely assume that we will not * overwrite anything at this point in time. */ - if (other_inode > sctx->send_progress) { + if (other_inode > sctx->send_progress || + is_waiting_for_move(sctx, other_inode)) { ret = get_inode_info(sctx->parent_root, other_inode, NULL, who_gen, NULL, NULL, NULL, NULL); if (ret < 0) @@ -3047,7 +3047,6 @@ static int add_pending_dir_move(struct send_ctx *sctx, pm->parent_ino = parent_ino; pm->ino = ino; pm->gen = ino_gen; - pm->is_orphan = is_orphan; INIT_LIST_HEAD(&pm->list); INIT_LIST_HEAD(&pm->update_refs); RB_CLEAR_NODE(&pm->node); @@ -3113,6 +3112,48 @@ static struct pending_dir_move *get_pending_dir_moves(struct send_ctx *sctx, return NULL; } +static int path_loop(struct send_ctx *sctx, struct fs_path *name, + u64 ino, u64 gen, u64 *ancestor_ino) +{ + int ret = 0; + u64 parent_inode = 0; + u64 parent_gen = 0; + u64 start_ino = ino; + + *ancestor_ino = 0; + while (ino != BTRFS_FIRST_FREE_OBJECTID) { + fs_path_reset(name); + + if (is_waiting_for_rm(sctx, ino)) + break; + if (is_waiting_for_move(sctx, ino)) { + if (*ancestor_ino == 0) + *ancestor_ino = ino; + ret = get_first_ref(sctx->parent_root, ino, + &parent_inode, &parent_gen, name); + } else { + ret = __get_cur_name_and_parent(sctx, ino, gen, + &parent_inode, + &parent_gen, name); + if (ret > 0) { + ret = 0; + break; + } + } + if (ret < 0) + break; + if (parent_inode == start_ino) { + ret = 1; + if (*ancestor_ino == 0) + *ancestor_ino = ino; + break; + } + ino = parent_inode; + gen = parent_gen; + } + return ret; +} + static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) { struct fs_path *from_path = NULL; @@ -3123,6 +3164,8 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) u64 parent_ino, parent_gen; struct waiting_dir_move *dm = NULL; u64 rmdir_ino = 0; + u64 ancestor; + bool is_orphan; int ret; name = fs_path_alloc(); @@ -3135,9 +3178,10 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) dm = get_waiting_dir_move(sctx, pm->ino); ASSERT(dm); rmdir_ino = dm->rmdir_ino; + is_orphan = dm->orphanized; free_waiting_dir_move(sctx, dm); - if (pm->is_orphan) { + if (is_orphan) { ret = gen_unique_name(sctx, pm->ino, pm->gen, from_path); } else { @@ -3155,6 +3199,22 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) goto out; sctx->send_progress = sctx->cur_ino + 1; + ret = path_loop(sctx, name, pm->ino, pm->gen, &ancestor); + if (ret) { + LIST_HEAD(deleted_refs); + ASSERT(ancestor > BTRFS_FIRST_FREE_OBJECTID); + ret = add_pending_dir_move(sctx, pm->ino, pm->gen, ancestor, + &pm->update_refs, &deleted_refs, + is_orphan); + if (ret < 0) + goto out; + if (rmdir_ino) { + dm = get_waiting_dir_move(sctx, pm->ino); + ASSERT(dm); + dm->rmdir_ino = rmdir_ino; + } + goto out; + } fs_path_reset(name); to_path = name; name = NULL; @@ -3325,6 +3385,7 @@ static int wait_for_dest_dir_move(struct send_ctx *sctx, u64 left_gen; u64 right_gen; int ret = 0; + struct waiting_dir_move *wdm; if (RB_EMPTY_ROOT(&sctx->waiting_dir_moves)) return 0; @@ -3383,7 +3444,8 @@ static int wait_for_dest_dir_move(struct send_ctx *sctx, goto out; } - if (is_waiting_for_move(sctx, di_key.objectid)) { + wdm = get_waiting_dir_move(sctx, di_key.objectid); + if (wdm && !wdm->orphanized) { ret = add_pending_dir_move(sctx, sctx->cur_ino, sctx->cur_inode_gen, @@ -3643,11 +3705,26 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); goto out; if (ret) { struct name_cache_entry *nce; + struct waiting_dir_move *wdm; ret = orphanize_inode(sctx, ow_inode, ow_gen, cur->full_path); if (ret < 0) goto out; + + /* + * If ow_inode has its rename operation delayed + * make sure that its orphanized name is used in + * the source path when performing its rename + * operation. + */ + if (is_waiting_for_move(sctx, ow_inode)) { + wdm = get_waiting_dir_move(sctx, + ow_inode); + ASSERT(wdm); + wdm->orphanized = true; + } + /* * Make sure we clear our orphanized inode's * name from the name cache. This is because the @@ -3663,6 +3740,19 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); name_cache_delete(sctx, nce); kfree(nce); } + + /* + * ow_inode might currently be an ancestor of + * cur_ino, therefore compute valid_path (the + * current path of cur_ino) again because it + * might contain the pre-orphanization name of + * ow_inode, which is no longer valid. + */ + fs_path_reset(valid_path); + ret = get_cur_path(sctx, sctx->cur_ino, + sctx->cur_inode_gen, valid_path); + if (ret < 0) + goto out; } else { ret = send_unlink(sctx, cur->full_path); if (ret < 0) From 7969e77a73164b418da851cbae35cdd6c1b43fee Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 17 Jun 2016 17:13:36 +0100 Subject: [PATCH 006/235] Btrfs: send, add missing error check for calls to path_loop() The function path_loop() can return a negative integer, signaling an error, 0 if there's no path loop and 1 if there's a path loop. We were treating any non zero values as meaning that a path loop exists. Fix this by explicitly checking for errors and gracefully return them to user space. Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 993e1bab0a6b..0dc05bb856ff 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -3200,6 +3200,8 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) sctx->send_progress = sctx->cur_ino + 1; ret = path_loop(sctx, name, pm->ino, pm->gen, &ancestor); + if (ret < 0) + goto out; if (ret) { LIST_HEAD(deleted_refs); ASSERT(ancestor > BTRFS_FIRST_FREE_OBJECTID); From 4122ea64f877f632390bca1886c8a6f666f28398 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 27 Jun 2016 16:54:44 +0100 Subject: [PATCH 007/235] Btrfs: incremental send, fix invalid paths for rename operations Example scenario: Parent snapshot: . (ino 277) |---- tmp/ (ino 278) |---- pre/ (ino 280) | |---- wait_dir/ (ino 281) | |---- desc/ (ino 282) |---- ance/ (ino 283) | |---- below_ance/ (ino 279) | |---- other_dir/ (ino 284) Send snapshot: . (ino 277) |---- tmp/ (ino 278) |---- other_dir/ (ino 284) |---- below_ance/ (ino 279) | |---- pre/ (ino 280) | |---- wait_dir/ (ino 281) |---- desc/ (ino 282) |---- ance/ (ino 283) While computing the send stream the following steps happen: 1) While processing inode 279 we end up delaying its rename operation because its new parent in the send snapshot, inode 284, was not yet processed and therefore not yet renamed; 2) Later when processing inode 280 we end up renaming it immediately to "ance/below_once/pre" and not delay its rename operation because its new parent (inode 279 in the send snapshot) has its rename operation delayed and inode 280 is not an encestor of inode 279 (its parent in the send snapshot) in the parent snapshot; 3) When processing inode 281 we end up delaying its rename operation because its new parent in the send snapshot, inode 284, was not yet processed and therefore not yet renamed; 4) When processing inode 282 we do not delay its rename operation because its parent in the send snapshot, inode 281, already has its own rename operation delayed and our current inode (282) is not an ancestor of inode 281 in the parent snapshot. Therefore inode 282 is renamed to "ance/below_ance/pre/wait_dir"; 5) When processing inode 283 we realize that we can rename it because one of its ancestors in the send snapshot, inode 281, has its rename operation delayed and inode 283 is not an ancestor of inode 281 in the parent snapshot. So a rename operation to rename inode 283 to "ance/below_ance/pre/wait_dir/desc/ance" is issued. This path is invalid due to a missing path building loop that was undetected by the incremental send implementation, as inode 283 ends up getting included twice in the path (once with its path in the parent snapshot). Therefore its rename operation must wait before the ancestor inode 284 is renamed. Fix this by not terminating the rename dependency checks when we find an ancestor, in the send snapshot, that has its rename operation delayed. So that we continue doing the same checks if the current inode is not an ancestor, in the parent snapshot, of an ancestor in the send snapshot we are processing in the loop. The problem and reproducer were reported by Robbie Ko, as part of a patch titled "Btrfs: incremental send, avoid ancestor rename to descendant". However the fix was unnecessarily complicated and can be addressed with much less code and effort. Reported-by: Robbie Ko Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 0dc05bb856ff..3d6a4dde8fe4 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -3534,7 +3534,8 @@ static int wait_for_parent_move(struct send_ctx *sctx, ret = is_ancestor(sctx->parent_root, sctx->cur_ino, sctx->cur_inode_gen, ino, path_before); - break; + if (ret) + break; } fs_path_reset(path_before); From 99ea42ddb13bad6188d4f183f8866cacd14fa141 Mon Sep 17 00:00:00 2001 From: Robbie Ko Date: Tue, 23 Jun 2015 18:39:49 +0800 Subject: [PATCH 008/235] Btrfs: incremental send, fix premature rmdir operations Under certain situations, an incremental send operation can contain a rmdir operation that will make the receiving end fail when attempting to execute it, because the target directory is not yet empty. Consider the following example: Parent snapshot: . (ino 256) |--- a/ (ino 257) | |--- c/ (ino 260) | |--- del/ (ino 259) |--- tmp/ (ino 258) |--- x/ (ino 261) Send snapshot: . (ino 256) |--- a/ (ino 257) | |--- x/ (ino 261) | |--- c/ (ino 260) |--- tmp/ (ino 258) 1) When processing inode 258, we delay its rename operation because inode 260 is its new parent in the send snapshot and it was not yet renamed (since 260 > 258, that is, beyond the current progress); 2) When processing inode 259, we realize we can not yet send an rmdir operation (against inode 259) because inode 258 was still not yet renamed/moved away from inode 259. Therefore we update data structures so that after inode 258 is renamed, we try again to see if we can finally send an rmdir operation for inode 259; 3) When we process inode 260, we send a rename operation for it followed by a rename operation for inode 258. Once we send the rename operation for inode 258 we then check if we can finally issue an rmdir for its previous parent, inode 259, by calling the can_rmdir() function with a value of sctx->cur_ino + 1 (260 + 1 = 261) for its "progress" argument. This makes can_rmdir() return true (value 1) because even though there's still a child inode of inode 259 that was not yet renamed/moved, which is inode 261, the given value of progress (261) is not lower then 261 (that is, not lower than the inode number of some child of inode 259). So we end up sending a rmdir operation for inode 259 before its child inode 261 is processed and renamed. So fix this by passing the correct progress value to the call to can_rmdir() from within apply_dir_move() (where we issue delayed rename operations), which should match stcx->cur_ino (the number of the inode currently being processed) and not sctx->cur_ino + 1. A test case for fstests follows soon. Signed-off-by: Robbie Ko Signed-off-by: Filipe Manana [Rewrote change log to be more detailed, clear and well formatted] Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 3d6a4dde8fe4..4115aba001ad 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -3236,7 +3236,7 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) /* already deleted */ goto finish; } - ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino + 1); + ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino); if (ret < 0) goto out; if (!ret) From 443f9d266c4848a3a9ea386f66c4bed2f5997940 Mon Sep 17 00:00:00 2001 From: Robbie Ko Date: Mon, 22 Jun 2015 17:08:45 +0800 Subject: [PATCH 009/235] Btrfs: send, fix warning due to late freeing of orphan_dir_info structures Under certain situations, when doing an incremental send, we can end up not freeing orphan_dir_info structures as soon as they are no longer needed. Instead we end up freeing them only after finishing the send stream, which causes a warning to be emitted: [282735.229200] ------------[ cut here ]------------ [282735.229968] WARNING: CPU: 9 PID: 10588 at fs/btrfs/send.c:6298 btrfs_ioctl_send+0xe2f/0xe51 [btrfs] [282735.231282] Modules linked in: btrfs crc32c_generic xor raid6_pq acpi_cpufreq tpm_tis ppdev tpm parport_pc psmouse parport sg pcspkr i2c_piix4 i2c_core evdev processor serio_raw button loop autofs4 ext4 crc16 jbd2 mbcache sr_mod cdrom sd_mod ata_generic virtio_scsi ata_piix libata virtio_pci virtio_ring virtio e1000 scsi_mod floppy [last unloaded: btrfs] [282735.237130] CPU: 9 PID: 10588 Comm: btrfs Tainted: G W 4.6.0-rc7-btrfs-next-31+ #1 [282735.239309] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS by qemu-project.org 04/01/2014 [282735.240160] 0000000000000000 ffff880224273ca8 ffffffff8126b42c 0000000000000000 [282735.240160] 0000000000000000 ffff880224273ce8 ffffffff81052b14 0000189a24273ac8 [282735.240160] ffff8802210c9800 0000000000000000 0000000000000001 0000000000000000 [282735.240160] Call Trace: [282735.240160] [] dump_stack+0x67/0x90 [282735.240160] [] __warn+0xc2/0xdd [282735.240160] [] warn_slowpath_null+0x1d/0x1f [282735.240160] [] btrfs_ioctl_send+0xe2f/0xe51 [btrfs] [282735.240160] [] btrfs_ioctl+0x14f/0x1f81 [btrfs] [282735.240160] [] ? arch_local_irq_save+0x9/0xc [282735.240160] [] vfs_ioctl+0x18/0x34 [282735.240160] [] do_vfs_ioctl+0x550/0x5be [282735.240160] [] ? __fget+0x6b/0x77 [282735.240160] [] ? __fget_light+0x62/0x71 [282735.240160] [] SyS_ioctl+0x57/0x79 [282735.240160] [] entry_SYSCALL_64_fastpath+0x18/0xa8 [282735.240160] [] ? time_hardirqs_off+0x9/0x14 [282735.240160] [] ? trace_hardirqs_off_caller+0x1f/0xaa [282735.256343] ---[ end trace a4539270c8056f93 ]--- Consider the following example: Parent snapshot: . (ino 256) |--- a/ (ino 257) | |--- c/ (ino 260) | |--- del/ (ino 259) |--- tmp/ (ino 258) |--- x/ (ino 261) |--- y/ (ino 262) Send snapshot: . (ino 256) |--- a/ (ino 257) | |--- x/ (ino 261) | |--- y/ (ino 262) | |--- c/ (ino 260) |--- tmp/ (ino 258) 1) When processing inode 258, we end up delaying its rename operation because it has an ancestor (in the send snapshot) that has a higher inode number (inode 260) which was also renamed in the send snapshot, therefore we delay the rename of inode 258 so that it happens after inode 260 is renamed; 2) When processing inode 259, we end up delaying its deletion (rmdir operation) because it has a child inode (258) that has its rename operation delayed. At this point we allocate an orphan_dir_info structure and tag inode 258 so that we later attempt to see if we can delete (rmdir) inode 259 once inode 258 is renamed; 3) When we process inode 260, after renaming it we finally do the rename operation for inode 258. Once we issue the rename operation for inode 258 we notice that this inode was tagged so that we attempt to see if at this point we can delete (rmdir) inode 259. But at this point we can not still delete inode 259 because it has 2 children, inodes 261 and 262, that were not yet processed and therefore not yet moved (renamed) away from inode 259. We end up not freeing the orphan_dir_info structure allocated in step 2; 4) We process inodes 261 and 262, and once we move/rename inode 262 we issue the rmdir operation for inode 260; 5) We finish the send stream and notice that red black tree that contains orphan_dir_info structures is not empty, so we emit a warning and then free any orphan_dir_structures left. So fix this by freeing an orphan_dir_info structure once we try to apply a pending rename operation if we can not delete yet the tagged directory. A test case for fstests follows soon. Signed-off-by: Robbie Ko Signed-off-by: Filipe Manana [Modified changelog to be more detailed and easier to understand] --- fs/btrfs/send.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 4115aba001ad..e552c33e72c0 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -2947,6 +2947,10 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen, } if (loc.objectid > send_progress) { + struct orphan_dir_info *odi; + + odi = get_orphan_dir_info(sctx, dir); + free_orphan_dir_info(sctx, odi); ret = 0; goto out; } From 764433a12e78e1a2161793c02eceafdfed2d652b Mon Sep 17 00:00:00 2001 From: Robbie Ko Date: Tue, 23 Jun 2015 18:39:50 +0800 Subject: [PATCH 010/235] Btrfs: send, fix invalid leaf accesses due to incorrect utimes operations During an incremental send, if we have delayed rename operations for inodes that were children of directories which were removed in the send snapshot, we can end up accessing incorrect items in a leaf or accessing beyond the last item of the leaf due to issuing utimes operations for the removed inodes. Consider the following example: Parent snapshot: . (ino 256) |--- a/ (ino 257) | |--- c/ (ino 262) | |--- b/ (ino 258) | |--- d/ (ino 263) | |--- del/ (ino 261) |--- x/ (ino 259) |--- y/ (ino 260) Send snapshot: . (ino 256) |--- a/ (ino 257) | |--- b/ (ino 258) | |--- c/ (ino 262) | |--- y/ (ino 260) | |--- d/ (ino 263) |--- x/ (ino 259) 1) When processing inodes 259 and 260, we end up delaying their rename operations because their parents, inodes 263 and 262 respectively, were not yet processed and therefore not yet renamed; 2) When processing inode 262, its rename operation is issued and right after the rename operation for inode 260 is issued. However right after issuing the rename operation for inode 260, at send.c:apply_dir_move(), we issue utimes operations for all current and past parents of inode 260. This means we try to send a utimes operation for its old parent, inode 261 (deleted in the send snapshot), which does not cause any immediate and deterministic failure, because when the target inode is not found in the send snapshot, the send.c:send_utimes() function ignores it and uses the leaf region pointed to by path->slots[0], which can be any unrelated item (belonging to other inode) or it can be a region outside the leaf boundaries, if the leaf is full and path->slots[0] matches the number of items in the leaf. So we end up either successfully sending a utimes operation, which is fine and irrelevant because the old parent (inode 261) will end up being deleted later, or we end up doing an invalid memory access tha crashes the kernel. So fix this by making apply_dir_move() issue utimes operations only for parents that still exist in the send snapshot. In a separate patch we will make send_utimes() return an error (-ENOENT) if the given inode does not exists in the send snapshot. Signed-off-by: Robbie Ko Signed-off-by: Filipe Manana [Rewrote change log to be more detailed and better organized] Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index e552c33e72c0..8b653967b163 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -3270,8 +3270,18 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) * and old parent(s). */ list_for_each_entry(cur, &pm->update_refs, list) { - if (cur->dir == rmdir_ino) + /* + * The parent inode might have been deleted in the send snapshot + */ + ret = get_inode_info(sctx->send_root, cur->dir, NULL, + NULL, NULL, NULL, NULL, NULL); + if (ret == -ENOENT) { + ret = 0; continue; + } + if (ret < 0) + goto out; + ret = send_utimes(sctx, cur->dir, cur->dir_gen); if (ret < 0) goto out; From 15b253eace1f98dabbc6e03f6793fcf8603b1655 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Sat, 2 Jul 2016 05:43:46 +0100 Subject: [PATCH 011/235] Btrfs: send, avoid incorrect leaf accesses when sending utimes operations The caller of send_utimes() is supposed to be sure that the inode number it passes to this function does actually exists in the send snapshot. However due to logic/algorithm bugs (such as the one fixed by the patch titled "Btrfs: send, fix invalid leaf accesses due to incorrect utimes operations"), this might not be the case and when that happens it makes send_utimes() access use an unrelated leaf item as the target inode item or access beyond a leaf's boundaries (when the leaf is full and path->slots[0] matches the number of items in the leaf). So if the call to btrfs_search_slot() done by send_utimes() does not find the inode item, just make sure send_utimes() returns -ENOENT and does not silently accesses unrelated leaf items or does invalid leaf accesses, also allowing us to easialy and deterministically catch such algorithmic/logic bugs. Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 8b653967b163..2db8dc89ca41 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -2502,6 +2502,8 @@ verbose_printk("btrfs: send_utimes %llu\n", ino); key.type = BTRFS_INODE_ITEM_KEY; key.offset = 0; ret = btrfs_search_slot(NULL, sctx->send_root, &key, path, 0, 0); + if (ret > 0) + ret = -ENOENT; if (ret < 0) goto out; From 951555856b88aa47bc238de6b4c6e97bfd9d36df Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 1 Aug 2016 01:50:37 +0100 Subject: [PATCH 012/235] Btrfs: send, don't bug on inconsistent snapshots When doing an incremental send, if we find a new/modified/deleted extent, reference or xattr without having previously processed the corresponding inode item we end up exexuting a BUG_ON(). This is because whenever an extent, xattr or reference is added, modified or deleted, we always expect to have the corresponding inode item updated. However there are situations where this will not happen due to transient -ENOMEM or -ENOSPC errors when doing delayed inode updates. For example, when punching holes we can succeed in deleting and modifying (shrinking) extents but later fail to do the delayed inode update. So after such failure we close our transaction handle and right after a snapshot of the fs/subvol tree can be made and used later for a send operation. The same thing can happen during truncate, link, unlink, and xattr related operations. So instead of executing a BUG_ON, make send return an -EIO error and print an informative error message do dmesg/syslog. Signed-off-by: Filipe Manana --- fs/btrfs/send.c | 48 +++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 45 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 2db8dc89ca41..efe129fe2678 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -273,6 +273,39 @@ struct name_cache_entry { char name[]; }; +static void inconsistent_snapshot_error(struct send_ctx *sctx, + enum btrfs_compare_tree_result result, + const char *what) +{ + const char *result_string; + + switch (result) { + case BTRFS_COMPARE_TREE_NEW: + result_string = "new"; + break; + case BTRFS_COMPARE_TREE_DELETED: + result_string = "deleted"; + break; + case BTRFS_COMPARE_TREE_CHANGED: + result_string = "updated"; + break; + case BTRFS_COMPARE_TREE_SAME: + ASSERT(0); + result_string = "unchanged"; + break; + default: + ASSERT(0); + result_string = "unexpected"; + } + + btrfs_err(sctx->send_root->fs_info, + "Send: inconsistent snapshot, found %s %s for inode %llu without updated inode item, send root is %llu, parent root is %llu", + result_string, what, sctx->cmp_key->objectid, + sctx->send_root->root_key.objectid, + (sctx->parent_root ? + sctx->parent_root->root_key.objectid : 0)); +} + static int is_waiting_for_move(struct send_ctx *sctx, u64 ino); static struct waiting_dir_move * @@ -5711,7 +5744,10 @@ static int changed_ref(struct send_ctx *sctx, { int ret = 0; - BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid); + if (sctx->cur_ino != sctx->cmp_key->objectid) { + inconsistent_snapshot_error(sctx, result, "reference"); + return -EIO; + } if (!sctx->cur_inode_new_gen && sctx->cur_ino != BTRFS_FIRST_FREE_OBJECTID) { @@ -5736,7 +5772,10 @@ static int changed_xattr(struct send_ctx *sctx, { int ret = 0; - BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid); + if (sctx->cur_ino != sctx->cmp_key->objectid) { + inconsistent_snapshot_error(sctx, result, "xattr"); + return -EIO; + } if (!sctx->cur_inode_new_gen && !sctx->cur_inode_deleted) { if (result == BTRFS_COMPARE_TREE_NEW) @@ -5760,7 +5799,10 @@ static int changed_extent(struct send_ctx *sctx, { int ret = 0; - BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid); + if (sctx->cur_ino != sctx->cmp_key->objectid) { + inconsistent_snapshot_error(sctx, result, "extent"); + return -EIO; + } if (!sctx->cur_inode_new_gen && !sctx->cur_inode_deleted) { if (result != BTRFS_COMPARE_TREE_DELETED) From 67710892ec983aa79ad1e2a2642fe8e3a4a194ea Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 6 Jun 2016 11:51:25 +0100 Subject: [PATCH 013/235] Btrfs: be more precise on errors when getting an inode from disk When we attempt to read an inode from disk, we end up always returning an -ESTALE error to the caller regardless of the actual failure reason, which can be an out of memory problem (when allocating a path), some error found when reading from the fs/subvolume btree (like a genuine IO error) or the inode does not exists. So lets start returning the real error code to the callers so that they don't treat all -ESTALE errors as meaning that the inode does not exists (such as during orphan cleanup). This will also be needed for a subsequent patch in the same series dealing with a special fsync case. Signed-off-by: Filipe Manana --- fs/btrfs/inode.c | 27 ++++++++++++++++++--------- 1 file changed, 18 insertions(+), 9 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 3e61bd1f3f65..f9686541997b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3428,10 +3428,10 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) found_key.offset = 0; inode = btrfs_iget(root->fs_info->sb, &found_key, root, NULL); ret = PTR_ERR_OR_ZERO(inode); - if (ret && ret != -ESTALE) + if (ret && ret != -ENOENT) goto out; - if (ret == -ESTALE && root == root->fs_info->tree_root) { + if (ret == -ENOENT && root == root->fs_info->tree_root) { struct btrfs_root *dead_root; struct btrfs_fs_info *fs_info = root->fs_info; int is_dead_root = 0; @@ -3467,7 +3467,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) * Inode is already gone but the orphan item is still there, * kill the orphan item. */ - if (ret == -ESTALE) { + if (ret == -ENOENT) { trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { ret = PTR_ERR(trans); @@ -3626,7 +3626,7 @@ static noinline int acls_after_inode_item(struct extent_buffer *leaf, /* * read an inode from the btree into the in-memory inode */ -static void btrfs_read_locked_inode(struct inode *inode) +static int btrfs_read_locked_inode(struct inode *inode) { struct btrfs_path *path; struct extent_buffer *leaf; @@ -3645,14 +3645,19 @@ static void btrfs_read_locked_inode(struct inode *inode) filled = true; path = btrfs_alloc_path(); - if (!path) + if (!path) { + ret = -ENOMEM; goto make_bad; + } memcpy(&location, &BTRFS_I(inode)->location, sizeof(location)); ret = btrfs_lookup_inode(NULL, root, path, &location, 0); - if (ret) + if (ret) { + if (ret > 0) + ret = -ENOENT; goto make_bad; + } leaf = path->nodes[0]; @@ -3805,11 +3810,12 @@ static void btrfs_read_locked_inode(struct inode *inode) } btrfs_update_iflags(inode); - return; + return 0; make_bad: btrfs_free_path(path); make_bad_inode(inode); + return ret; } /* @@ -5595,7 +5601,9 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, return ERR_PTR(-ENOMEM); if (inode->i_state & I_NEW) { - btrfs_read_locked_inode(inode); + int ret; + + ret = btrfs_read_locked_inode(inode); if (!is_bad_inode(inode)) { inode_tree_add(inode); unlock_new_inode(inode); @@ -5604,7 +5612,8 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, } else { unlock_new_inode(inode); iput(inode); - inode = ERR_PTR(-ESTALE); + ASSERT(ret < 0); + inode = ERR_PTR(ret < 0 ? ret : -ESTALE); } } From 44f714dae50a2e795d3268a6831762aa6fa54f55 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 6 Jun 2016 16:11:13 +0100 Subject: [PATCH 014/235] Btrfs: improve performance on fsync against new inode after rename/unlink With commit 56f23fdbb600 ("Btrfs: fix file/data loss caused by fsync after rename and new inode") we got simple fix for a functional issue when the following sequence of actions is done: at transaction N create file A at directory D at transaction N + M (where M >= 1) move/rename existing file A from directory D to directory E create a new file named A at directory D fsync the new file power fail The solution was to simply detect such scenario and fallback to a full transaction commit when we detect it. However this turned out to had a significant impact on throughput (and a bit on latency too) for benchmarks using the dbench tool, which simulates real workloads from smbd (Samba) servers. For example on a test vm (with a debug kernel): Unpatched: Throughput 19.1572 MB/sec 32 clients 32 procs max_latency=1005.229 ms Patched: Throughput 23.7015 MB/sec 32 clients 32 procs max_latency=809.206 ms The patched results (this patch is applied) are similar to the results of a kernel with the commit 56f23fdbb600 ("Btrfs: fix file/data loss caused by fsync after rename and new inode") reverted. This change avoids the fallback to a transaction commit and instead makes sure all the names of the conflicting inode (the one that had a name in a past transaction that matches the name of the new file in the same parent directory) are logged so that at log replay time we don't lose neither the new file nor the old file, and the old file gets the name it was renamed to. This also ends up avoiding a full transaction commit for a similar case that involves an unlink instead of a rename of the old file: at transaction N create file A at directory D at transaction N + M (where M >= 1) remove file A create a new file named A at directory D fsync the new file power fail Signed-off-by: Filipe Manana --- fs/btrfs/inode.c | 19 +++++++++- fs/btrfs/tree-log.c | 85 ++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 95 insertions(+), 9 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index f9686541997b..453b9d0da5f1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4203,6 +4203,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) int err = 0; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_trans_handle *trans; + u64 last_unlink_trans; if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) return -ENOTEMPTY; @@ -4225,11 +4226,27 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) if (err) goto out; + last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; + /* now the directory is empty */ err = btrfs_unlink_inode(trans, root, dir, d_inode(dentry), dentry->d_name.name, dentry->d_name.len); - if (!err) + if (!err) { btrfs_i_size_write(inode, 0); + /* + * Propagate the last_unlink_trans value of the deleted dir to + * its parent directory. This is to prevent an unrecoverable + * log tree in the case we do something like this: + * 1) create dir foo + * 2) create snapshot under dir foo + * 3) delete the snapshot + * 4) rmdir foo + * 5) mkdir foo + * 6) fsync foo or some file inside foo + */ + if (last_unlink_trans >= trans->transid) + BTRFS_I(dir)->last_unlink_trans = last_unlink_trans; + } out: btrfs_end_transaction(trans, root); btrfs_btree_balance_dirty(root); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index c05f69a8ec42..5fa624cd815d 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -4469,7 +4469,8 @@ static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans, static int btrfs_check_ref_name_override(struct extent_buffer *eb, const int slot, const struct btrfs_key *key, - struct inode *inode) + struct inode *inode, + u64 *other_ino) { int ret; struct btrfs_path *search_path; @@ -4528,7 +4529,16 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb, search_path, parent, name, this_name_len, 0); if (di && !IS_ERR(di)) { - ret = 1; + struct btrfs_key di_key; + + btrfs_dir_item_key_to_cpu(search_path->nodes[0], + di, &di_key); + if (di_key.type == BTRFS_INODE_ITEM_KEY) { + ret = 1; + *other_ino = di_key.objectid; + } else { + ret = -EAGAIN; + } goto out; } else if (IS_ERR(di)) { ret = PTR_ERR(di); @@ -4718,16 +4728,71 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, if ((min_key.type == BTRFS_INODE_REF_KEY || min_key.type == BTRFS_INODE_EXTREF_KEY) && BTRFS_I(inode)->generation == trans->transid) { + u64 other_ino = 0; + ret = btrfs_check_ref_name_override(path->nodes[0], path->slots[0], - &min_key, inode); + &min_key, inode, + &other_ino); if (ret < 0) { err = ret; goto out_unlock; } else if (ret > 0) { - err = 1; - btrfs_set_log_full_commit(root->fs_info, trans); - goto out_unlock; + struct btrfs_key inode_key; + struct inode *other_inode; + + if (ins_nr > 0) { + ins_nr++; + } else { + ins_nr = 1; + ins_start_slot = path->slots[0]; + } + ret = copy_items(trans, inode, dst_path, path, + &last_extent, ins_start_slot, + ins_nr, inode_only, + logged_isize); + if (ret < 0) { + err = ret; + goto out_unlock; + } + ins_nr = 0; + btrfs_release_path(path); + inode_key.objectid = other_ino; + inode_key.type = BTRFS_INODE_ITEM_KEY; + inode_key.offset = 0; + other_inode = btrfs_iget(root->fs_info->sb, + &inode_key, root, + NULL); + /* + * If the other inode that had a conflicting dir + * entry was deleted in the current transaction, + * we don't need to do more work nor fallback to + * a transaction commit. + */ + if (IS_ERR(other_inode) && + PTR_ERR(other_inode) == -ENOENT) { + goto next_key; + } else if (IS_ERR(other_inode)) { + err = PTR_ERR(other_inode); + goto out_unlock; + } + /* + * We are safe logging the other inode without + * acquiring its i_mutex as long as we log with + * the LOG_INODE_EXISTS mode. We're safe against + * concurrent renames of the other inode as well + * because during a rename we pin the log and + * update the log with the new name before we + * unpin it. + */ + err = btrfs_log_inode(trans, root, other_inode, + LOG_INODE_EXISTS, + 0, LLONG_MAX, ctx); + iput(other_inode); + if (err) + goto out_unlock; + else + goto next_key; } } @@ -4795,7 +4860,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, ins_nr = 0; } btrfs_release_path(path); - +next_key: if (min_key.offset < (u64)-1) { min_key.offset++; } else if (min_key.type < max_key.type) { @@ -4989,8 +5054,12 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) break; - if (IS_ROOT(parent)) + if (IS_ROOT(parent)) { + inode = d_inode(parent); + if (btrfs_must_commit_transaction(trans, inode)) + ret = 1; break; + } parent = dget_parent(parent); dput(old_parent); From f005bd7e3b84a353475a2895e2c7686a66297d87 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 1 Aug 2016 10:54:15 +0100 Subject: [PATCH 015/235] clocksource/arm_arch_timer: Force per-CPU interrupt to be level-triggered The ARM architected timer produces level-triggered interrupts (this is mandated by the architecture). Unfortunately, a number of device-trees get this wrong, and expose an edge-triggered interrupt. Until now, this wasn't too much an issue, as the programming of the trigger would fail (the corresponding PPI cannot be reconfigured), and the kernel would be happy with this. But we're about to change this, and trust DT a lot if the driver doesn't provide its own trigger information. In that context, the timer breaks badly. While we do need to fix the DTs, there is also some userspace out there (kvmtool) that generates the same kind of broken DT on the fly, and that will completely break with newer kernels. As a safety measure, and to keep buggy software alive as well as buying us some time to fix DTs all over the place, let's check what trigger configuration has been given us by the firmware. If this is not a level configuration, then we know that the DT/ACPI configuration is bust, and we pick some defaults which won't be worse than the existing setup. Signed-off-by: Marc Zyngier Cc: Andrew Lunn Cc: Liu Gang Cc: Mark Rutland Cc: Masahiro Yamada Cc: Wenbin Song Cc: Mingkai Hu Cc: Florian Fainelli Cc: Kevin Hilman Cc: Daniel Lezcano Cc: Michal Simek Cc: Jon Hunter Cc: arm@kernel.org Cc: bcm-kernel-feedback-list@broadcom.com Cc: linux-arm-kernel@lists.infradead.org Cc: Sebastian Hesselbarth Cc: Jason Cooper Cc: Ray Jui Cc: "Hou Zhiqiang" Cc: Tirumalesh Chalamarla Cc: linux-samsung-soc@vger.kernel.org Cc: Yuan Yao Cc: Jan Glauber Cc: Gregory Clement Cc: linux-amlogic@lists.infradead.org Cc: soren.brinkmann@xilinx.com Cc: Rajesh Bhagat Cc: Scott Branden Cc: Duc Dang Cc: Kukjin Kim Cc: Carlo Caione Cc: Dinh Nguyen Link: http://lkml.kernel.org/r/1470045256-9032-2-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- drivers/clocksource/arm_arch_timer.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c index 28bce3f4f81d..57700541f951 100644 --- a/drivers/clocksource/arm_arch_timer.c +++ b/drivers/clocksource/arm_arch_timer.c @@ -8,6 +8,9 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ + +#define pr_fmt(fmt) "arm_arch_timer: " fmt + #include #include #include @@ -370,16 +373,33 @@ static bool arch_timer_has_nonsecure_ppi(void) arch_timer_ppi[PHYS_NONSECURE_PPI]); } +static u32 check_ppi_trigger(int irq) +{ + u32 flags = irq_get_trigger_type(irq); + + if (flags != IRQF_TRIGGER_HIGH && flags != IRQF_TRIGGER_LOW) { + pr_warn("WARNING: Invalid trigger for IRQ%d, assuming level low\n", irq); + pr_warn("WARNING: Please fix your firmware\n"); + flags = IRQF_TRIGGER_LOW; + } + + return flags; +} + static int arch_timer_starting_cpu(unsigned int cpu) { struct clock_event_device *clk = this_cpu_ptr(arch_timer_evt); + u32 flags; __arch_timer_setup(ARCH_CP15_TIMER, clk); - enable_percpu_irq(arch_timer_ppi[arch_timer_uses_ppi], 0); + flags = check_ppi_trigger(arch_timer_ppi[arch_timer_uses_ppi]); + enable_percpu_irq(arch_timer_ppi[arch_timer_uses_ppi], flags); - if (arch_timer_has_nonsecure_ppi()) - enable_percpu_irq(arch_timer_ppi[PHYS_NONSECURE_PPI], 0); + if (arch_timer_has_nonsecure_ppi()) { + flags = check_ppi_trigger(arch_timer_ppi[PHYS_NONSECURE_PPI]); + enable_percpu_irq(arch_timer_ppi[PHYS_NONSECURE_PPI], flags); + } arch_counter_set_user_access(); if (evtstrm_enable) From 7f555c8e5a84b348c2b76f4ca78eae7222354c03 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 29 Jul 2016 18:03:42 -0400 Subject: [PATCH 016/235] drm/amdgpu/gmc7: add missing mullins case Looks like this got missed when we ported the code from radeon. Reviewed-by: Edward O'Callaghan Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index d24a82bd0c7a..0b0f08641eed 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -144,6 +144,7 @@ static int gmc_v7_0_init_microcode(struct amdgpu_device *adev) break; case CHIP_KAVERI: case CHIP_KABINI: + case CHIP_MULLINS: return 0; default: BUG(); } From 903f75c8958e21f0fba2520467c11eaca824226d Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 29 Jul 2016 18:14:39 -0400 Subject: [PATCH 017/235] drm/amdgpu/ci: add mullins to default case for smc ucode It's already covered by the default case, but add it for consistency. Reviewed-by: Alexandre Demers Reviewed-by: Edward O'Callaghan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/ci_dpm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c index e2f0e5d58d5c..a5c94b482459 100644 --- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c @@ -5779,6 +5779,7 @@ static int ci_dpm_init_microcode(struct amdgpu_device *adev) break; case CHIP_KAVERI: case CHIP_KABINI: + case CHIP_MULLINS: default: BUG(); } From 4f7ec157cb6dfb1e2cd18abf21e6cb1107842c5f Mon Sep 17 00:00:00 2001 From: Eric Huang Date: Thu, 28 Jul 2016 17:25:01 -0400 Subject: [PATCH 018/235] drm/amd/amdgpu: change pptable output format from ASCII to binary Reviewed-by: Edward O'Callaghan Reviewed-by: Alex Deucher Signed-off-by: Eric Huang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index ff63b88b0ffa..5cc7052e391d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -305,7 +305,7 @@ static ssize_t amdgpu_get_pp_table(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; char *table = NULL; - int size, i; + int size; if (adev->pp_enabled) size = amdgpu_dpm_get_pp_table(adev, &table); @@ -315,10 +315,7 @@ static ssize_t amdgpu_get_pp_table(struct device *dev, if (size >= PAGE_SIZE) size = PAGE_SIZE - 1; - for (i = 0; i < size; i++) { - sprintf(buf + i, "%02x", table[i]); - } - sprintf(buf + i, "\n"); + memcpy(buf, table, size); return size; } From 1f4c17a03ba7f430d63dba8c8e08ff1e2712581d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 1 Aug 2016 13:36:08 -0400 Subject: [PATCH 019/235] SUNRPC: Handle EADDRNOTAVAIL on connection failures If the connect attempt immediately fails with an EADDRNOTAVAIL error, then that means our choice of source port number was bad. This error is expected when we set the SO_REUSEPORT socket option and we have 2 sockets sharing the same source and destination address and port combinations. Signed-off-by: Trond Myklebust Fixes: 402e23b4ed9ed ("SUNRPC: Fix stupid typo in xs_sock_set_reuseport") Cc: stable@vger.kernel.org # v4.0+ --- net/sunrpc/xprtsock.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 111767ab124a..c6b1d48c4319 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2295,6 +2295,10 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) /* SYN_SENT! */ if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; + break; + case -EADDRNOTAVAIL: + /* Source port number is unavailable. Try a new one! */ + transport->srcport = 0; } out: return ret; From ad3331acb1464024d20a184d3358f3cecc373b54 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 2 Aug 2016 13:47:43 -0400 Subject: [PATCH 020/235] SUNRPC: Fix up socket autodisconnect Ensure that we don't forget to set up the disconnection timer for the case when a connect request is fulfilled after the RPC request that initiated it has timed out or been interrupted. Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 8313960cac52..ea244b29138b 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -680,6 +680,20 @@ void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie) spin_unlock_bh(&xprt->transport_lock); } +static bool +xprt_has_timer(const struct rpc_xprt *xprt) +{ + return xprt->idle_timeout != 0; +} + +static void +xprt_schedule_autodisconnect(struct rpc_xprt *xprt) + __must_hold(&xprt->transport_lock) +{ + if (list_empty(&xprt->recv) && xprt_has_timer(xprt)) + mod_timer(&xprt->timer, xprt->last_used + xprt->idle_timeout); +} + static void xprt_init_autodisconnect(unsigned long data) { @@ -688,6 +702,8 @@ xprt_init_autodisconnect(unsigned long data) spin_lock(&xprt->transport_lock); if (!list_empty(&xprt->recv)) goto out_abort; + /* Reset xprt->last_used to avoid connect/autodisconnect cycling */ + xprt->last_used = jiffies; if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) goto out_abort; spin_unlock(&xprt->transport_lock); @@ -725,6 +741,7 @@ void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie) goto out; xprt->snd_task =NULL; xprt->ops->release_xprt(xprt, NULL); + xprt_schedule_autodisconnect(xprt); out: spin_unlock_bh(&xprt->transport_lock); wake_up_bit(&xprt->state, XPRT_LOCKED); @@ -888,11 +905,6 @@ static void xprt_timer(struct rpc_task *task) spin_unlock_bh(&xprt->transport_lock); } -static inline int xprt_has_timer(struct rpc_xprt *xprt) -{ - return xprt->idle_timeout != 0; -} - /** * xprt_prepare_transmit - reserve the transport before sending a request * @task: RPC task about to send a request @@ -1280,9 +1292,7 @@ void xprt_release(struct rpc_task *task) if (!list_empty(&req->rq_list)) list_del(&req->rq_list); xprt->last_used = jiffies; - if (list_empty(&xprt->recv) && xprt_has_timer(xprt)) - mod_timer(&xprt->timer, - xprt->last_used + xprt->idle_timeout); + xprt_schedule_autodisconnect(xprt); spin_unlock_bh(&xprt->transport_lock); if (req->rq_buffer) xprt->ops->buf_free(req->rq_buffer); From 1fabaddd153e9e2f7e5695266abb56c934629718 Mon Sep 17 00:00:00 2001 From: Alim Akhtar Date: Mon, 11 Jul 2016 09:34:18 +0200 Subject: [PATCH 021/235] arm64: dts: Fix RTC by providing rtc_src clock Add RTC source clock as Exynos7 needs source (32.768KHz) clock for RTC block. Without this currently S3C RTC driver probe is broken on this SoC. Signed-off-by: Alim Akhtar Reviewed-by: Javier Martinez Canillas Signed-off-by: Krzysztof Kozlowski Signed-off-by: Arnd Bergmann --- arch/arm64/boot/dts/exynos/exynos7-espresso.dts | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm64/boot/dts/exynos/exynos7-espresso.dts b/arch/arm64/boot/dts/exynos/exynos7-espresso.dts index 299f3ce969ab..c528dd52ba2d 100644 --- a/arch/arm64/boot/dts/exynos/exynos7-espresso.dts +++ b/arch/arm64/boot/dts/exynos/exynos7-espresso.dts @@ -12,6 +12,7 @@ /dts-v1/; #include "exynos7.dtsi" #include +#include / { model = "Samsung Exynos7 Espresso board based on EXYNOS7"; @@ -43,6 +44,8 @@ &serial_2 { &rtc { status = "okay"; + clocks = <&clock_ccore PCLK_RTC>, <&s2mps15_osc S2MPS11_CLK_AP>; + clock-names = "rtc", "rtc_src"; }; &watchdog { From d97ffe788184313fd83ff7da4b6587156f8388f0 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 2 Aug 2016 09:44:33 +0200 Subject: [PATCH 022/235] ARM: do away with final ARCH_REQUIRE_GPIOLIB A new user of the Kconfig selection of ARCH_REQUIRE_GPIOLIB has appeared. Replace with just selecting GPIOLIB. Signed-off-by: Linus Walleij Signed-off-by: Olof Johansson --- arch/arm/mach-clps711x/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-clps711x/Kconfig b/arch/arm/mach-clps711x/Kconfig index dc7c6edeab39..61284b9389cf 100644 --- a/arch/arm/mach-clps711x/Kconfig +++ b/arch/arm/mach-clps711x/Kconfig @@ -1,13 +1,13 @@ menuconfig ARCH_CLPS711X bool "Cirrus Logic EP721x/EP731x-based" depends on ARCH_MULTI_V4T - select ARCH_REQUIRE_GPIOLIB select AUTO_ZRELADDR select CLKSRC_OF select CLPS711X_TIMER select COMMON_CLK select CPU_ARM720T select GENERIC_CLOCKEVENTS + select GPIOLIB select MFD_SYSCON select OF_IRQ select USE_OF From 4d2bf027e8ca499a52224e508f52a59d3858094a Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Tue, 26 Jul 2016 10:00:48 +0200 Subject: [PATCH 023/235] MAINTAINER: alpine: add a mailing list Add the linux-arm-kernel mailing list for the Alpine SoCs. Signed-off-by: Antoine Tenart Signed-off-by: Olof Johansson --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 25f43204014d..b5c04dfeab22 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -999,6 +999,7 @@ N: meson ARM/Annapurna Labs ALPINE ARCHITECTURE M: Tsahee Zidenberg M: Antoine Tenart +L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Maintained F: arch/arm/mach-alpine/ F: arch/arm/boot/dts/alpine* From 869ec056fa8450184423c8076e0a342db127795c Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 30 Jun 2016 14:25:19 +0200 Subject: [PATCH 024/235] ARM: shmobile: don't call platform_can_secondary_boot on UP For rcar-gen2, we build the SMP files even for UP configurations, and that just broke: arch/arm/mach-shmobile/built-in.o: In function `shmobile_smp_init_fallback_ops': pm-rcar-gen2.c:(.init.text+0x40c): undefined reference to `platform_can_secondary_boot' This adds an compile-time check before the call to platform_can_secondary_boot, turning the function into an empty stub for UP configurations. Signed-off-by: Arnd Bergmann Fixes: c21af444eace ("ARM: shmobile: smp: Add function to prioritize DT SMP") Acked-by: Geert Uytterhoeven Signed-off-by: Olof Johansson --- arch/arm/mach-shmobile/platsmp.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm/mach-shmobile/platsmp.c b/arch/arm/mach-shmobile/platsmp.c index f3dba6f356e2..02e21bceb085 100644 --- a/arch/arm/mach-shmobile/platsmp.c +++ b/arch/arm/mach-shmobile/platsmp.c @@ -40,5 +40,8 @@ bool shmobile_smp_cpu_can_disable(unsigned int cpu) bool __init shmobile_smp_init_fallback_ops(void) { /* fallback on PSCI/smp_ops if no other DT based method is detected */ + if (!IS_ENABLED(CONFIG_SMP)) + return false; + return platform_can_secondary_boot() ? true : false; } From fe85f07f582751da0a65d4ba7e3b36ebcbe55019 Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Tue, 2 Aug 2016 11:21:35 +0800 Subject: [PATCH 025/235] drm/amdgpu: update golden setting of iceland Signed-off-by: Huang Rui Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index bff8668e9e6d..a4983237baa0 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -409,6 +409,7 @@ static const u32 golden_settings_iceland_a11[] = mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000, mmPA_SC_RASTER_CONFIG, 0x3f3fffff, 0x00000002, mmPA_SC_RASTER_CONFIG_1, 0x0000003f, 0x00000000, + mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0000003c, mmSQ_RANDOM_WAVE_PRI, 0x001fffff, 0x000006fd, mmTA_CNTL_AUX, 0x000f000f, 0x000b0000, mmTCC_CTRL, 0x00100000, 0xf31fff7f, From 3a494b588f724fde00ae3ca0aa1c7e6cef1d0d51 Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Tue, 2 Aug 2016 11:28:47 +0800 Subject: [PATCH 026/235] drm/amdgpu: update golden setting of carrizo Signed-off-by: Huang Rui Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index a4983237baa0..85854ba2cf0c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -506,8 +506,10 @@ static const u32 cz_golden_settings_a11[] = mmGB_GPU_ID, 0x0000000f, 0x00000000, mmPA_SC_ENHANCE, 0xffffffff, 0x00000001, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000, + mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0000003c, mmSQ_RANDOM_WAVE_PRI, 0x001fffff, 0x000006fd, mmTA_CNTL_AUX, 0x000f000f, 0x00010000, + mmTCC_CTRL, 0x00100000, 0xf31fff7f, mmTCC_EXE_DISABLE, 0x00000002, 0x00000002, mmTCP_ADDR_CONFIG, 0x0000000f, 0x000000f3, mmTCP_CHAN_STEER_LO, 0xffffffff, 0x00001302 From 9761bc53c8e1c52f3f0aa4334f38fd87b0cf1613 Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Tue, 2 Aug 2016 11:26:34 +0800 Subject: [PATCH 027/235] drm/amdgpu: update golden setting of polaris11 Signed-off-by: Huang Rui Reviewed-by: Alex Deucher Reviewed-by: Ken Wang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 85854ba2cf0c..3b7f6e87d244 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -270,7 +270,8 @@ static const u32 tonga_mgcg_cgcg_init[] = static const u32 golden_settings_polaris11_a11[] = { - mmCB_HW_CONTROL, 0xfffdf3cf, 0x00006208, + mmCB_HW_CONTROL, 0x0000f3cf, 0x00007208, + mmCB_HW_CONTROL_2, 0x0f000000, 0x0f000000, mmCB_HW_CONTROL_3, 0x000001ff, 0x00000040, mmDB_DEBUG2, 0xf00fffff, 0x00000400, mmPA_SC_ENHANCE, 0xffffffff, 0x20000001, @@ -279,7 +280,7 @@ static const u32 golden_settings_polaris11_a11[] = mmPA_SC_RASTER_CONFIG_1, 0x0000003f, 0x00000000, mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0001003c, mmRLC_CGCG_CGLS_CTRL_3D, 0xffffffff, 0x0001003c, - mmSQ_CONFIG, 0x07f80000, 0x07180000, + mmSQ_CONFIG, 0x07f80000, 0x01180000, mmTA_CNTL_AUX, 0x000f000f, 0x000b0000, mmTCC_CTRL, 0x00100000, 0xf31fff7f, mmTCP_ADDR_CONFIG, 0x000003ff, 0x000000f3, From 6d51c813b172b4374fe7a6b732b6666f8d77bfea Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Tue, 2 Aug 2016 11:30:05 +0800 Subject: [PATCH 028/235] drm/amdgpu: update golden setting of stoney Signed-off-by: Huang Rui Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 717359d3ba8c..2aee2c6f3cd5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -103,6 +103,11 @@ static const u32 stoney_mgcg_cgcg_init[] = mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; +static const u32 golden_settings_stoney_common[] = +{ + mmMC_HUB_RDREQ_UVD, MC_HUB_RDREQ_UVD__PRESCALE_MASK, 0x00000004, + mmMC_RD_GRP_OTH, MC_RD_GRP_OTH__UVD_MASK, 0x00600000 +}; static void gmc_v8_0_init_golden_registers(struct amdgpu_device *adev) { @@ -142,6 +147,9 @@ static void gmc_v8_0_init_golden_registers(struct amdgpu_device *adev) amdgpu_program_register_sequence(adev, stoney_mgcg_cgcg_init, (const u32)ARRAY_SIZE(stoney_mgcg_cgcg_init)); + amdgpu_program_register_sequence(adev, + golden_settings_stoney_common, + (const u32)ARRAY_SIZE(golden_settings_stoney_common)); break; default: break; From a5a5e3084ec457f234298392dfa2c55db47b603f Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Tue, 2 Aug 2016 13:15:12 +0800 Subject: [PATCH 029/235] drm/amdgpu: update golden setting of polaris10 Signed-off-by: Huang Rui Reviewed-by: Ken Wang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 3b7f6e87d244..b8184617ca25 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -302,8 +302,8 @@ static const u32 polaris11_golden_common_all[] = static const u32 golden_settings_polaris10_a11[] = { mmATC_MISC_CG, 0x000c0fc0, 0x000c0200, - mmCB_HW_CONTROL, 0xfffdf3cf, 0x00007208, - mmCB_HW_CONTROL_2, 0, 0x0f000000, + mmCB_HW_CONTROL, 0x0001f3cf, 0x00007208, + mmCB_HW_CONTROL_2, 0x0f000000, 0x0f000000, mmCB_HW_CONTROL_3, 0x000001ff, 0x00000040, mmDB_DEBUG2, 0xf00fffff, 0x00000400, mmPA_SC_ENHANCE, 0xffffffff, 0x20000001, From 7a376ac11fc2109dfd86442ff79982ecf16dcd6d Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 3 Aug 2016 11:07:44 +0200 Subject: [PATCH 030/235] Revert "ARM: aspeed: adapt defconfigs for new CONFIG_PRINTK_TIME" The patch that this was preparing for made it into neither v4.7 nor v4.8, so we should back this out as well to avoid the opposite warning: arch/arm/configs/aspeed_g5_defconfig:62:warning: symbol value '1' invalid for PRINTK_TIME arch/arm/configs/aspeed_g4_defconfig:61:warning: symbol value '1' invalid for PRINTK_TIME Sorry for not catching this earlier. Signed-off-by: Arnd Bergmann Fixes: 0ef659a30055 ("ARM: aspeed: adapt defconfigs for new CONFIG_PRINTK_TIME") Cc: stable@vger.kernel.org # v4.7 --- arch/arm/configs/aspeed_g4_defconfig | 2 +- arch/arm/configs/aspeed_g5_defconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/configs/aspeed_g4_defconfig b/arch/arm/configs/aspeed_g4_defconfig index b6e54ee9bdbd..ca39c04fec6b 100644 --- a/arch/arm/configs/aspeed_g4_defconfig +++ b/arch/arm/configs/aspeed_g4_defconfig @@ -58,7 +58,7 @@ CONFIG_SERIAL_OF_PLATFORM=y # CONFIG_IOMMU_SUPPORT is not set CONFIG_FIRMWARE_MEMMAP=y CONFIG_FANOTIFY=y -CONFIG_PRINTK_TIME=1 +CONFIG_PRINTK_TIME=y CONFIG_DYNAMIC_DEBUG=y CONFIG_STRIP_ASM_SYMS=y CONFIG_PAGE_POISONING=y diff --git a/arch/arm/configs/aspeed_g5_defconfig b/arch/arm/configs/aspeed_g5_defconfig index 892605167357..4f366b0370e9 100644 --- a/arch/arm/configs/aspeed_g5_defconfig +++ b/arch/arm/configs/aspeed_g5_defconfig @@ -59,7 +59,7 @@ CONFIG_SERIAL_OF_PLATFORM=y # CONFIG_IOMMU_SUPPORT is not set CONFIG_FIRMWARE_MEMMAP=y CONFIG_FANOTIFY=y -CONFIG_PRINTK_TIME=1 +CONFIG_PRINTK_TIME=y CONFIG_DYNAMIC_DEBUG=y CONFIG_STRIP_ASM_SYMS=y CONFIG_PAGE_POISONING=y From e6571499336e10f93a77c51a35fd1a96828eea71 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 3 Aug 2016 11:01:32 +0100 Subject: [PATCH 031/235] Btrfs: remove unused function btrfs_add_delayed_qgroup_reserve() No longer used as of commit 5846a3c26873 ("btrfs: qgroup: Fix a race in delayed_ref which leads to abort trans"). Signed-off-by: Filipe Manana --- fs/btrfs/delayed-ref.c | 27 --------------------------- fs/btrfs/delayed-ref.h | 3 --- 2 files changed, 30 deletions(-) diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 430b3689b112..02dd2ec7d534 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -861,33 +861,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, return 0; } -int btrfs_add_delayed_qgroup_reserve(struct btrfs_fs_info *fs_info, - struct btrfs_trans_handle *trans, - u64 ref_root, u64 bytenr, u64 num_bytes) -{ - struct btrfs_delayed_ref_root *delayed_refs; - struct btrfs_delayed_ref_head *ref_head; - int ret = 0; - - if (!fs_info->quota_enabled || !is_fstree(ref_root)) - return 0; - - delayed_refs = &trans->transaction->delayed_refs; - - spin_lock(&delayed_refs->lock); - ref_head = find_ref_head(&delayed_refs->href_root, bytenr, 0); - if (!ref_head) { - ret = -ENOENT; - goto out; - } - WARN_ON(ref_head->qgroup_reserved || ref_head->qgroup_ref_root); - ref_head->qgroup_ref_root = ref_root; - ref_head->qgroup_reserved = num_bytes; -out: - spin_unlock(&delayed_refs->lock); - return ret; -} - int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info, struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes, diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 5fca9534a271..43f3629760e9 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -250,9 +250,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, u64 parent, u64 ref_root, u64 owner, u64 offset, u64 reserved, int action, struct btrfs_delayed_extent_op *extent_op); -int btrfs_add_delayed_qgroup_reserve(struct btrfs_fs_info *fs_info, - struct btrfs_trans_handle *trans, - u64 ref_root, u64 bytenr, u64 num_bytes); int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info, struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes, From 0b857b44b5e445dc850cd91c45ce6edeb7797480 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sun, 31 Jul 2016 00:27:39 -0700 Subject: [PATCH 032/235] nvme-rdma: Don't leak uninitialized memory in connect request private data Zero out the full nvme_rdma_cm_req structure before sending it. Otherwise we end up leaking kernel memory in the reserved field, which might break forward compatibility in the future. Signed-off-by: Roland Dreier Reviewed-by: Christoph Hellwig Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 3e3ce2b0424e..b96b88369871 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1269,7 +1269,7 @@ static int nvme_rdma_route_resolved(struct nvme_rdma_queue *queue) { struct nvme_rdma_ctrl *ctrl = queue->ctrl; struct rdma_conn_param param = { }; - struct nvme_rdma_cm_req priv; + struct nvme_rdma_cm_req priv = { }; int ret; param.qp_num = queue->qp->qp_num; From 5f372eb3e76317b4fe4ba53ad1547f39fc883350 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 31 Jul 2016 18:43:15 +0300 Subject: [PATCH 033/235] nvme-rdma: Queue ns scanning after a sucessful reconnection On an ordered target shutdown, the target can send a AEN on a namespace removal, this will trigger the host to queue ns-list query. The shutdown will trigger error recovery which will attepmt periodic reconnect. We can hit a race where the ns rescanning fails (error recovery kicked in and we're not connected) causing removing all the namespaces and when we reconnect we won't see any namespaces for this controller. So, queue a namespace rescan after we successfully reconnected to the target. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index b96b88369871..7fd1d735ced5 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -748,8 +748,10 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); WARN_ON_ONCE(!changed); - if (ctrl->queue_count > 1) + if (ctrl->queue_count > 1) { nvme_start_queues(&ctrl->ctrl); + nvme_queue_scan(&ctrl->ctrl); + } dev_info(ctrl->ctrl.device, "Successfully reconnected\n"); From 57de5a0a40db97bb390d3ac1f4c2e74b9f3515c3 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 14 Jul 2016 17:39:47 +0300 Subject: [PATCH 034/235] nvme-rdma: Fix device removal handling Device removal sequence may have crashed because the controller (and admin queue space) was freed before we destroyed the admin queue resources. Thus we want to destroy the admin queue and only then queue controller deletion and wait for it to complete. More specifically we: 1. own the controller deletion (make sure we are not competing with another deletion). 2. get rid of inflight reconnects if exists (which also destroy and create queues). 3. destroy the queue. 4. safely queue controller deletion (and wait for it to complete). Reported-by: Steve Wise Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 45 ++++++++++++++++++++-------------------- 1 file changed, 23 insertions(+), 22 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 7fd1d735ced5..3ffec3728abe 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -169,7 +169,6 @@ MODULE_PARM_DESC(register_always, static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *event); static void nvme_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc); -static int __nvme_rdma_del_ctrl(struct nvme_rdma_ctrl *ctrl); /* XXX: really should move to a generic header sooner or later.. */ static inline void put_unaligned_le24(u32 val, u8 *p) @@ -1320,37 +1319,39 @@ static int nvme_rdma_route_resolved(struct nvme_rdma_queue *queue) * that caught the event. Since we hold the callout until the controller * deletion is completed, we'll deadlock if the controller deletion will * call rdma_destroy_id on this queue's cm_id. Thus, we claim ownership - * of destroying this queue before-hand, destroy the queue resources - * after the controller deletion completed with the exception of destroying - * the cm_id implicitely by returning a non-zero rc to the callout. + * of destroying this queue before-hand, destroy the queue resources, + * then queue the controller deletion which won't destroy this queue and + * we destroy the cm_id implicitely by returning a non-zero rc to the callout. */ static int nvme_rdma_device_unplug(struct nvme_rdma_queue *queue) { struct nvme_rdma_ctrl *ctrl = queue->ctrl; - int ret, ctrl_deleted = 0; + int ret; - /* First disable the queue so ctrl delete won't free it */ - if (!test_and_clear_bit(NVME_RDMA_Q_CONNECTED, &queue->flags)) - goto out; + /* Own the controller deletion */ + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_DELETING)) + return 0; - /* delete the controller */ - ret = __nvme_rdma_del_ctrl(ctrl); - if (!ret) { - dev_warn(ctrl->ctrl.device, - "Got rdma device removal event, deleting ctrl\n"); - flush_work(&ctrl->delete_work); + dev_warn(ctrl->ctrl.device, + "Got rdma device removal event, deleting ctrl\n"); + + /* Get rid of reconnect work if its running */ + cancel_delayed_work_sync(&ctrl->reconnect_work); + + /* Disable the queue so ctrl delete won't free it */ + if (test_and_clear_bit(NVME_RDMA_Q_CONNECTED, &queue->flags)) { + /* Free this queue ourselves */ + nvme_rdma_stop_queue(queue); + nvme_rdma_destroy_queue_ib(queue); /* Return non-zero so the cm_id will destroy implicitly */ - ctrl_deleted = 1; - - /* Free this queue ourselves */ - rdma_disconnect(queue->cm_id); - ib_drain_qp(queue->qp); - nvme_rdma_destroy_queue_ib(queue); + ret = 1; } -out: - return ctrl_deleted; + /* Queue controller deletion */ + queue_work(nvme_rdma_wq, &ctrl->delete_work); + flush_work(&ctrl->delete_work); + return ret; } static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id, From 2461a8dd38bea3cb5b1c1f0323794483292fb03f Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 24 Jul 2016 09:29:51 +0300 Subject: [PATCH 035/235] nvme-rdma: Remove duplicate call to nvme_remove_namespaces nvme_uninit_ctrl already does that for us. Note that we reordered nvme_rdma_shutdown_ctrl and nvme_uninit_ctrl, this is perfectly fine because we actually want ctrl uninit (aen, scan cancellation and namespaces removal) to happen before we shutdown the rdma resources. Also, centralize the deletion work and the dead controller removal work code duplication into __nvme_rdma_shutdown_ctrl that accepts a shutdown boolean. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 3ffec3728abe..1279bc292253 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1660,15 +1660,20 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl) nvme_rdma_destroy_admin_queue(ctrl); } +static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown) +{ + nvme_uninit_ctrl(&ctrl->ctrl); + if (shutdown) + nvme_rdma_shutdown_ctrl(ctrl); + nvme_put_ctrl(&ctrl->ctrl); +} + static void nvme_rdma_del_ctrl_work(struct work_struct *work) { struct nvme_rdma_ctrl *ctrl = container_of(work, struct nvme_rdma_ctrl, delete_work); - nvme_remove_namespaces(&ctrl->ctrl); - nvme_rdma_shutdown_ctrl(ctrl); - nvme_uninit_ctrl(&ctrl->ctrl); - nvme_put_ctrl(&ctrl->ctrl); + __nvme_rdma_remove_ctrl(ctrl, true); } static int __nvme_rdma_del_ctrl(struct nvme_rdma_ctrl *ctrl) @@ -1701,9 +1706,7 @@ static void nvme_rdma_remove_ctrl_work(struct work_struct *work) struct nvme_rdma_ctrl *ctrl = container_of(work, struct nvme_rdma_ctrl, delete_work); - nvme_remove_namespaces(&ctrl->ctrl); - nvme_uninit_ctrl(&ctrl->ctrl); - nvme_put_ctrl(&ctrl->ctrl); + __nvme_rdma_remove_ctrl(ctrl, false); } static void nvme_rdma_reset_ctrl_work(struct work_struct *work) From a34ca17a9717fe607cd58285a1704cb6526cf561 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 24 Jul 2016 09:22:19 +0300 Subject: [PATCH 036/235] nvme-rdma: Free the I/O tags when we delete the controller If we wait until we free the controller (free_ctrl) we might lose our rdma device without any notification while we still have open resources (tags mrs and dma mappings). Instead, destroy the tags with their rdma resources once we delete the device and not when freeing it. Note that we don't do that in nvme_rdma_shutdown_ctrl because controller reset uses it as well and we want to give active I/O a chance to complete successfully. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 1279bc292253..6378dc94aeaf 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -686,11 +686,6 @@ static void nvme_rdma_free_ctrl(struct nvme_ctrl *nctrl) list_del(&ctrl->list); mutex_unlock(&nvme_rdma_ctrl_mutex); - if (ctrl->ctrl.tagset) { - blk_cleanup_queue(ctrl->ctrl.connect_q); - blk_mq_free_tag_set(&ctrl->tag_set); - nvme_rdma_dev_put(ctrl->device); - } kfree(ctrl->queues); nvmf_free_options(nctrl->opts); free_ctrl: @@ -1665,6 +1660,13 @@ static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown) nvme_uninit_ctrl(&ctrl->ctrl); if (shutdown) nvme_rdma_shutdown_ctrl(ctrl); + + if (ctrl->ctrl.tagset) { + blk_cleanup_queue(ctrl->ctrl.connect_q); + blk_mq_free_tag_set(&ctrl->tag_set); + nvme_rdma_dev_put(ctrl->device); + } + nvme_put_ctrl(&ctrl->ctrl); } From a159c64d936eb0d1da29d8ad384183d8984899c9 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 24 Jul 2016 09:32:08 +0300 Subject: [PATCH 037/235] nvme-loop: Remove duplicate call to nvme_remove_namespaces nvme_uninit_ctrl already does that for us. Note that we reordered nvme_loop_shutdown_ctrl with nvme_uninit_ctrl but its safe because we want controller uninit to happen before we shutdown the transport resources. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/loop.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c index 94e782987cc9..7affd40a6b33 100644 --- a/drivers/nvme/target/loop.c +++ b/drivers/nvme/target/loop.c @@ -414,9 +414,8 @@ static void nvme_loop_del_ctrl_work(struct work_struct *work) struct nvme_loop_ctrl *ctrl = container_of(work, struct nvme_loop_ctrl, delete_work); - nvme_remove_namespaces(&ctrl->ctrl); - nvme_loop_shutdown_ctrl(ctrl); nvme_uninit_ctrl(&ctrl->ctrl); + nvme_loop_shutdown_ctrl(ctrl); nvme_put_ctrl(&ctrl->ctrl); } @@ -501,7 +500,6 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work) nvme_loop_destroy_admin_queue(ctrl); out_disable: dev_warn(ctrl->ctrl.device, "Removing after reset failure\n"); - nvme_remove_namespaces(&ctrl->ctrl); nvme_uninit_ctrl(&ctrl->ctrl); nvme_put_ctrl(&ctrl->ctrl); } From 45862ebcc4883b1b6bc0701cd15cb2b68b140c5d Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 24 Jul 2016 09:26:16 +0300 Subject: [PATCH 038/235] nvme-rdma: Make sure to shutdown the controller if we can Relying on ctrl state in nvme_rdma_shutdown_ctrl is wrong because it will never be NVME_CTRL_LIVE (delete_ctrl or reset_ctrl invoked it). Instead, check that the admin queue is connected. Note that it is safe because we can never see a copmeting thread trying to destroy the admin queue (reset or delete controller). Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 6378dc94aeaf..f4b836869d2c 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1646,7 +1646,7 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl) nvme_rdma_free_io_queues(ctrl); } - if (ctrl->ctrl.state == NVME_CTRL_LIVE) + if (test_bit(NVME_RDMA_Q_CONNECTED, &ctrl->queues[0].flags)) nvme_shutdown_ctrl(&ctrl->ctrl); blk_mq_stop_hw_queues(ctrl->ctrl.admin_q); From c45eb4fed12d278d3619f1904885bd0d7bcbf036 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 13 Jul 2016 18:34:45 +0100 Subject: [PATCH 039/235] drm/i915/fbdev: Check for the framebuffer before use If the fbdev probing fails, and in our error path we fail to clear the dev_priv->fbdev, then we can try and use a dangling fbdev pointer, and in particular a NULL fb. This could also happen in pathological cases where we try to operate on the fbdev prior to it being probed. Reported-by: Maarten Lankhorst Signed-off-by: Chris Wilson Cc: Maarten Lankhorst Cc: Daniel Vetter Cc: Mika Kuoppala Link: http://patchwork.freedesktop.org/patch/msgid/1468431285-28264-2-git-send-email-chris@chris-wilson.co.uk Reviewed-by: Daniel Vetter Reviewed-by: Mika Kuoppala (cherry picked from commit 6bc265424df02f8162f4a17a37e4982e1c64460e) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_fbdev.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 86b00c6db1a6..3e3632c18733 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -782,7 +782,7 @@ void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous struct intel_fbdev *ifbdev = dev_priv->fbdev; struct fb_info *info; - if (!ifbdev) + if (!ifbdev || !ifbdev->fb) return; info = ifbdev->helper.fbdev; @@ -827,31 +827,28 @@ void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous void intel_fbdev_output_poll_changed(struct drm_device *dev) { - struct drm_i915_private *dev_priv = to_i915(dev); - if (dev_priv->fbdev) - drm_fb_helper_hotplug_event(&dev_priv->fbdev->helper); + struct intel_fbdev *ifbdev = to_i915(dev)->fbdev; + + if (ifbdev && ifbdev->fb) + drm_fb_helper_hotplug_event(&ifbdev->helper); } void intel_fbdev_restore_mode(struct drm_device *dev) { - int ret; - struct drm_i915_private *dev_priv = to_i915(dev); - struct intel_fbdev *ifbdev = dev_priv->fbdev; - struct drm_fb_helper *fb_helper; + struct intel_fbdev *ifbdev = to_i915(dev)->fbdev; if (!ifbdev) return; intel_fbdev_sync(ifbdev); + if (!ifbdev->fb) + return; - fb_helper = &ifbdev->helper; - - ret = drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper); - if (ret) { + if (drm_fb_helper_restore_fbdev_mode_unlocked(&ifbdev->helper)) { DRM_DEBUG("failed to restore crtc mode\n"); } else { - mutex_lock(&fb_helper->dev->struct_mutex); + mutex_lock(&dev->struct_mutex); intel_fb_obj_invalidate(ifbdev->fb->obj, ORIGIN_GTT); - mutex_unlock(&fb_helper->dev->struct_mutex); + mutex_unlock(&dev->struct_mutex); } } From d8f7750a08968b105056328652d2c332bdfa062d Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 19 May 2016 15:24:55 +0300 Subject: [PATCH 040/235] nvmet-rdma: Correctly handle RDMA device hot removal When configuring a device attached listener, we may see device removal events. In this case we return a non-zero return code from the cm event handler which implicitly destroys the cm_id. It is possible that in the future the user will remove this listener and by that trigger a second call to rdma_destroy_id on an already destroyed cm_id -> BUG. In addition, when a queue bound (active session) cm_id generates a DEVICE_REMOVAL event we must guarantee all resources are cleaned up by the time we return from the event handler. Introduce nvmet_rdma_device_removal which addresses (or at least attempts to) both scenarios. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/rdma.c | 87 ++++++++++++++++++++++++++++++-------- 1 file changed, 70 insertions(+), 17 deletions(-) diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index e06d504bdf0c..48c811850c29 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -77,6 +77,7 @@ enum nvmet_rdma_queue_state { NVMET_RDMA_Q_CONNECTING, NVMET_RDMA_Q_LIVE, NVMET_RDMA_Q_DISCONNECTING, + NVMET_RDMA_IN_DEVICE_REMOVAL, }; struct nvmet_rdma_queue { @@ -984,7 +985,10 @@ static void nvmet_rdma_release_queue_work(struct work_struct *w) struct nvmet_rdma_device *dev = queue->dev; nvmet_rdma_free_queue(queue); - rdma_destroy_id(cm_id); + + if (queue->state != NVMET_RDMA_IN_DEVICE_REMOVAL) + rdma_destroy_id(cm_id); + kref_put(&dev->ref, nvmet_rdma_free_dev); } @@ -1233,8 +1237,9 @@ static void __nvmet_rdma_queue_disconnect(struct nvmet_rdma_queue *queue) switch (queue->state) { case NVMET_RDMA_Q_CONNECTING: case NVMET_RDMA_Q_LIVE: - disconnect = true; queue->state = NVMET_RDMA_Q_DISCONNECTING; + case NVMET_RDMA_IN_DEVICE_REMOVAL: + disconnect = true; break; case NVMET_RDMA_Q_DISCONNECTING: break; @@ -1272,6 +1277,62 @@ static void nvmet_rdma_queue_connect_fail(struct rdma_cm_id *cm_id, schedule_work(&queue->release_work); } +/** + * nvme_rdma_device_removal() - Handle RDMA device removal + * @queue: nvmet rdma queue (cm id qp_context) + * @addr: nvmet address (cm_id context) + * + * DEVICE_REMOVAL event notifies us that the RDMA device is about + * to unplug so we should take care of destroying our RDMA resources. + * This event will be generated for each allocated cm_id. + * + * Note that this event can be generated on a normal queue cm_id + * and/or a device bound listener cm_id (where in this case + * queue will be null). + * + * we claim ownership on destroying the cm_id. For queues we move + * the queue state to NVMET_RDMA_IN_DEVICE_REMOVAL and for port + * we nullify the priv to prevent double cm_id destruction and destroying + * the cm_id implicitely by returning a non-zero rc to the callout. + */ +static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id, + struct nvmet_rdma_queue *queue) +{ + unsigned long flags; + + if (!queue) { + struct nvmet_port *port = cm_id->context; + + /* + * This is a listener cm_id. Make sure that + * future remove_port won't invoke a double + * cm_id destroy. use atomic xchg to make sure + * we don't compete with remove_port. + */ + if (xchg(&port->priv, NULL) != cm_id) + return 0; + } else { + /* + * This is a queue cm_id. Make sure that + * release queue will not destroy the cm_id + * and schedule all ctrl queues removal (only + * if the queue is not disconnecting already). + */ + spin_lock_irqsave(&queue->state_lock, flags); + if (queue->state != NVMET_RDMA_Q_DISCONNECTING) + queue->state = NVMET_RDMA_IN_DEVICE_REMOVAL; + spin_unlock_irqrestore(&queue->state_lock, flags); + nvmet_rdma_queue_disconnect(queue); + flush_scheduled_work(); + } + + /* + * We need to return 1 so that the core will destroy + * it's own ID. What a great API design.. + */ + return 1; +} + static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *event) { @@ -1294,20 +1355,11 @@ static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, break; case RDMA_CM_EVENT_ADDR_CHANGE: case RDMA_CM_EVENT_DISCONNECTED: - case RDMA_CM_EVENT_DEVICE_REMOVAL: case RDMA_CM_EVENT_TIMEWAIT_EXIT: - /* - * We can get the device removal callback even for a - * CM ID that we aren't actually using. In that case - * the context pointer is NULL, so we shouldn't try - * to disconnect a non-existing queue. But we also - * need to return 1 so that the core will destroy - * it's own ID. What a great API design.. - */ - if (queue) - nvmet_rdma_queue_disconnect(queue); - else - ret = 1; + nvmet_rdma_queue_disconnect(queue); + break; + case RDMA_CM_EVENT_DEVICE_REMOVAL: + ret = nvmet_rdma_device_removal(cm_id, queue); break; case RDMA_CM_EVENT_REJECTED: case RDMA_CM_EVENT_UNREACHABLE: @@ -1396,9 +1448,10 @@ static int nvmet_rdma_add_port(struct nvmet_port *port) static void nvmet_rdma_remove_port(struct nvmet_port *port) { - struct rdma_cm_id *cm_id = port->priv; + struct rdma_cm_id *cm_id = xchg(&port->priv, NULL); - rdma_destroy_id(cm_id); + if (cm_id) + rdma_destroy_id(cm_id); } static struct nvmet_fabrics_ops nvmet_rdma_ops = { From 40e64e07213201710a51e270595d6e6c028f9502 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 28 Jul 2016 18:04:09 +0300 Subject: [PATCH 041/235] nvmet-rdma: Don't use the inline buffer in order to avoid allocation for small reads Under extreme conditions this might cause data corruptions. By doing that we we repost the buffer and then post this buffer for the device to send. If we happen to use shared receive queues the device might write to the buffer before it sends it (there is no ordering between send and recv queues). Without SRQs we probably won't get that if the host doesn't mis-behave and send more than we allowed it, but relying on that is not really a good idea. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/rdma.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 48c811850c29..b4d648536c3e 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -616,15 +616,10 @@ static u16 nvmet_rdma_map_sgl_keyed(struct nvmet_rdma_rsp *rsp, if (!len) return 0; - /* use the already allocated data buffer if possible */ - if (len <= NVMET_RDMA_INLINE_DATA_SIZE && rsp->queue->host_qid) { - nvmet_rdma_use_inline_sg(rsp, len, 0); - } else { - status = nvmet_rdma_alloc_sgl(&rsp->req.sg, &rsp->req.sg_cnt, - len); - if (status) - return status; - } + status = nvmet_rdma_alloc_sgl(&rsp->req.sg, &rsp->req.sg_cnt, + len); + if (status) + return status; ret = rdma_rw_ctx_init(&rsp->rw, cm_id->qp, cm_id->port_num, rsp->req.sg, rsp->req.sg_cnt, 0, addr, key, From 28b89118539da03f4b188763e1b2fd1aec0f580a Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 4 Aug 2016 11:18:49 +0300 Subject: [PATCH 042/235] nvmet: Fix controller serial number inconsistency The host is allowed to issue identify as many times as it wants, we need to stay consistent when reporting the serial number for a given controller. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/admin-cmd.c | 6 +----- drivers/nvme/target/core.c | 4 ++++ drivers/nvme/target/nvmet.h | 1 + 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c index 2fac17a5ad53..47c564b5a289 100644 --- a/drivers/nvme/target/admin-cmd.c +++ b/drivers/nvme/target/admin-cmd.c @@ -13,7 +13,6 @@ */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include -#include #include #include "nvmet.h" @@ -83,7 +82,6 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) { struct nvmet_ctrl *ctrl = req->sq->ctrl; struct nvme_id_ctrl *id; - u64 serial; u16 status = 0; id = kzalloc(sizeof(*id), GFP_KERNEL); @@ -96,10 +94,8 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->vid = 0; id->ssvid = 0; - /* generate a random serial number as our controllers are ephemeral: */ - get_random_bytes(&serial, sizeof(serial)); memset(id->sn, ' ', sizeof(id->sn)); - snprintf(id->sn, sizeof(id->sn), "%llx", serial); + snprintf(id->sn, sizeof(id->sn), "%llx", ctrl->serial); memset(id->mn, ' ', sizeof(id->mn)); strncpy((char *)id->mn, "Linux", sizeof(id->mn)); diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 8a891ca53367..6559d5afa7bf 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -13,6 +13,7 @@ */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include +#include #include "nvmet.h" static struct nvmet_fabrics_ops *nvmet_transports[NVMF_TRTYPE_MAX]; @@ -728,6 +729,9 @@ u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, memcpy(ctrl->subsysnqn, subsysnqn, NVMF_NQN_SIZE); memcpy(ctrl->hostnqn, hostnqn, NVMF_NQN_SIZE); + /* generate a random serial number as our controllers are ephemeral: */ + get_random_bytes(&ctrl->serial, sizeof(ctrl->serial)); + kref_init(&ctrl->ref); ctrl->subsys = subsys; diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index 57dd6d834c28..76b6eedccaf9 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -113,6 +113,7 @@ struct nvmet_ctrl { struct mutex lock; u64 cap; + u64 serial; u32 cc; u32 csts; From 3ef1b4b298d98ccb3cc895baf1b18f7f9d073bee Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 4 Aug 2016 13:46:19 +0300 Subject: [PATCH 043/235] nvme-rdma: start async event handler after reconnecting to a controller When we reset or reconnect to a controller, we are cancelling the async event handler so we can safely re-establish resources, but we need to remember to start it again when we successfully reconnect. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index f4b836869d2c..e82434ab3a28 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -745,6 +745,7 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) if (ctrl->queue_count > 1) { nvme_start_queues(&ctrl->ctrl); nvme_queue_scan(&ctrl->ctrl); + nvme_queue_async_events(&ctrl->ctrl); } dev_info(ctrl->ctrl.device, "Successfully reconnected\n"); @@ -1747,6 +1748,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work) if (ctrl->queue_count > 1) { nvme_start_queues(&ctrl->ctrl); nvme_queue_scan(&ctrl->ctrl); + nvme_queue_async_events(&ctrl->ctrl); } return; From e3266378bdbca82c2854fc612fa9a391eba1f173 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 4 Aug 2016 17:37:40 +0300 Subject: [PATCH 044/235] nvme-rdma: Remove unused includes Signed-off-by: Sagi Grimberg Reviewed-by: Steve Wise --- drivers/nvme/host/rdma.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index e82434ab3a28..8d2875b4c56d 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -12,13 +12,11 @@ * more details. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#include #include #include #include #include #include -#include #include #include #include @@ -26,7 +24,6 @@ #include #include #include -#include #include #include From 9130b8dbc6ac20f2dc5846e1647f5b60eafab6e3 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Wed, 3 Aug 2016 20:19:48 -0400 Subject: [PATCH 045/235] SUNRPC: allow for upcalls for same uid but different gss service It's possible to have simultaneous upcalls for the same UIDs but different GSS service. In that case, we need to allow for the upcall to gssd to proceed so that not the same context is used by two different GSS services. Some servers lock the use of context to the GSS service. Signed-off-by: Olga Kornievskaia Cc: stable@vger.kernel.org # v3.9+ Signed-off-by: Trond Myklebust --- net/sunrpc/auth_gss/auth_gss.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index 23c8e7c39656..976c7812bbd5 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -340,12 +340,14 @@ gss_release_msg(struct gss_upcall_msg *gss_msg) } static struct gss_upcall_msg * -__gss_find_upcall(struct rpc_pipe *pipe, kuid_t uid) +__gss_find_upcall(struct rpc_pipe *pipe, kuid_t uid, const struct gss_auth *auth) { struct gss_upcall_msg *pos; list_for_each_entry(pos, &pipe->in_downcall, list) { if (!uid_eq(pos->uid, uid)) continue; + if (auth && pos->auth->service != auth->service) + continue; atomic_inc(&pos->count); dprintk("RPC: %s found msg %p\n", __func__, pos); return pos; @@ -365,7 +367,7 @@ gss_add_msg(struct gss_upcall_msg *gss_msg) struct gss_upcall_msg *old; spin_lock(&pipe->lock); - old = __gss_find_upcall(pipe, gss_msg->uid); + old = __gss_find_upcall(pipe, gss_msg->uid, gss_msg->auth); if (old == NULL) { atomic_inc(&gss_msg->count); list_add(&gss_msg->list, &pipe->in_downcall); @@ -714,7 +716,7 @@ gss_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) err = -ENOENT; /* Find a matching upcall */ spin_lock(&pipe->lock); - gss_msg = __gss_find_upcall(pipe, uid); + gss_msg = __gss_find_upcall(pipe, uid, NULL); if (gss_msg == NULL) { spin_unlock(&pipe->lock); goto err_put_ctx; From d88e4d82efc7aca47f1e31808717c9718e466d93 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 4 Aug 2016 16:24:28 +1000 Subject: [PATCH 046/235] SUNRPC: disable the use of IPv6 temporary addresses. If the net.ipv6.conf.*.use_temp_addr sysctl is set to '2', then TCP connections over IPv6 will prefer a 'private' source address. These eventually expire and become invalid, typically after a week, but the time is configurable. When the local address becomes invalid the client will not be able to receive replies from the server. Eventually the connection will timeout or break and a new connection will be established, but this can take half an hour (typically TCP connection break time). RFC 4941, which describes private IPv6 addresses, acknowledges that some applications might not work well with them and that the application may explicitly a request non-temporary (i.e. "public") address. I believe this is correct for SUNRPC clients. Without this change, a client will occasionally experience a long delay if private addresses have been enabled. The privacy offered by private addresses is of little value for an NFS server which requires client authentication. For NFSv3 this will often not be a problem because idle connections are closed after 5 minutes. For NFSv4 connections never go idle due to the period RENEW (or equivalent) request. Signed-off-by: NeilBrown Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index c6b1d48c4319..bf79212fdbf8 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2236,6 +2236,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) unsigned int keepcnt = xprt->timeout->to_retries + 1; unsigned int opt_on = 1; unsigned int timeo; + unsigned int addr_pref = IPV6_PREFER_SRC_PUBLIC; /* TCP Keepalive options */ kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE, @@ -2247,6 +2248,16 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) kernel_setsockopt(sock, SOL_TCP, TCP_KEEPCNT, (char *)&keepcnt, sizeof(keepcnt)); + /* Avoid temporary address, they are bad for long-lived + * connections such as NFS mounts. + * RFC4941, section 3.6 suggests that: + * Individual applications, which have specific + * knowledge about the normal duration of connections, + * MAY override this as appropriate. + */ + kernel_setsockopt(sock, SOL_IPV6, IPV6_ADDR_PREFERENCES, + (char *)&addr_pref, sizeof(addr_pref)); + /* TCP user timeout (see RFC5482) */ timeo = jiffies_to_msecs(xprt->timeout->to_initval) * (xprt->timeout->to_retries + 1); From 206b3bb57481758b5fe98e90f47abb0f3fc0954e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Aug 2016 12:16:19 -0400 Subject: [PATCH 047/235] NFSv4.2: LAYOUTSTATS may return NFS4ERR_ADMIN/DELEG_REVOKED We should handle those errors in the same way we handle the other stateid errors: by invalidating the faulty layout stateid. Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index 33da841a21bb..6f4752734804 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -338,6 +338,8 @@ nfs42_layoutstat_done(struct rpc_task *task, void *calldata) case 0: break; case -NFS4ERR_EXPIRED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: case -NFS4ERR_STALE_STATEID: case -NFS4ERR_OLD_STATEID: case -NFS4ERR_BAD_STATEID: From 02910177aede34d6f49e2dc14b1c5c6cd468d94f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 4 Aug 2016 00:00:33 -0400 Subject: [PATCH 048/235] SUNRPC: Fix reconnection timeouts When the connect attempt fails and backs off, we should start the clock at the last connection attempt, not time at which we queue up the reconnect job. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index bf79212fdbf8..04b0c4190dd7 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2173,6 +2173,8 @@ static void xs_udp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) write_unlock_bh(&sk->sk_callback_lock); } xs_udp_do_set_buffer_size(xprt); + + xprt->stat.connect_start = jiffies; } static void xs_udp_setup_socket(struct work_struct *work) @@ -2384,6 +2386,16 @@ static void xs_tcp_setup_socket(struct work_struct *work) xprt_wake_pending_tasks(xprt, status); } +static unsigned long xs_reconnect_delay(const struct rpc_xprt *xprt) +{ + unsigned long start, now = jiffies; + + start = xprt->stat.connect_start + xprt->reestablish_timeout; + if (time_after(start, now)) + return start - now; + return 0; +} + /** * xs_connect - connect a socket to a remote endpoint * @xprt: pointer to transport structure @@ -2401,6 +2413,7 @@ static void xs_tcp_setup_socket(struct work_struct *work) static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + unsigned long delay = 0; WARN_ON_ONCE(!xprt_lock_connect(xprt, task, transport)); @@ -2412,19 +2425,19 @@ static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task) /* Start by resetting any existing state */ xs_reset_transport(transport); - queue_delayed_work(xprtiod_workqueue, - &transport->connect_worker, - xprt->reestablish_timeout); + delay = xs_reconnect_delay(xprt); + xprt->reestablish_timeout <<= 1; if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; if (xprt->reestablish_timeout > XS_TCP_MAX_REEST_TO) xprt->reestablish_timeout = XS_TCP_MAX_REEST_TO; - } else { + } else dprintk("RPC: xs_connect scheduled xprt %p\n", xprt); - queue_delayed_work(xprtiod_workqueue, - &transport->connect_worker, 0); - } + + queue_delayed_work(xprtiod_workqueue, + &transport->connect_worker, + delay); } /** From 34b58355ad1d9987267f071265a7de6c8e00662a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Fri, 5 Aug 2016 18:36:10 +0900 Subject: [PATCH 049/235] drm/ttm: Wait for a BO to become idle before unbinding it from GTT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes hangs under memory pressure, e.g. running the piglit test tex3d-maxsize concurrently with other tests. Fixes: 17d33bc9d6ef ("drm/ttm: drop waiting for idle in ttm_bo_evict.") Reviewed-by: Christian König Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 4 ++-- drivers/gpu/drm/nouveau/nouveau_bo.c | 4 ++-- drivers/gpu/drm/radeon/radeon_ttm.c | 4 ++-- drivers/gpu/drm/ttm/ttm_bo.c | 3 ++- drivers/gpu/drm/ttm/ttm_bo_util.c | 10 +++++++++- include/drm/ttm/ttm_bo_driver.h | 3 ++- 6 files changed, 19 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index b7742e62972a..9b61c8ba7aaf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -335,7 +335,7 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, if (unlikely(r)) { goto out_cleanup; } - r = ttm_bo_move_ttm(bo, true, no_wait_gpu, new_mem); + r = ttm_bo_move_ttm(bo, true, interruptible, no_wait_gpu, new_mem); out_cleanup: ttm_bo_mem_put(bo, &tmp_mem); return r; @@ -368,7 +368,7 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, if (unlikely(r)) { return r; } - r = ttm_bo_move_ttm(bo, true, no_wait_gpu, &tmp_mem); + r = ttm_bo_move_ttm(bo, true, interruptible, no_wait_gpu, &tmp_mem); if (unlikely(r)) { goto out_cleanup; } diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 528bdeffb339..6190035edfea 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -1151,7 +1151,7 @@ nouveau_bo_move_flipd(struct ttm_buffer_object *bo, bool evict, bool intr, if (ret) goto out; - ret = ttm_bo_move_ttm(bo, true, no_wait_gpu, new_mem); + ret = ttm_bo_move_ttm(bo, true, intr, no_wait_gpu, new_mem); out: ttm_bo_mem_put(bo, &tmp_mem); return ret; @@ -1179,7 +1179,7 @@ nouveau_bo_move_flips(struct ttm_buffer_object *bo, bool evict, bool intr, if (ret) return ret; - ret = ttm_bo_move_ttm(bo, true, no_wait_gpu, &tmp_mem); + ret = ttm_bo_move_ttm(bo, true, intr, no_wait_gpu, &tmp_mem); if (ret) goto out; diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index ffdad81ef964..0c00e192c845 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -346,7 +346,7 @@ static int radeon_move_vram_ram(struct ttm_buffer_object *bo, if (unlikely(r)) { goto out_cleanup; } - r = ttm_bo_move_ttm(bo, true, no_wait_gpu, new_mem); + r = ttm_bo_move_ttm(bo, true, interruptible, no_wait_gpu, new_mem); out_cleanup: ttm_bo_mem_put(bo, &tmp_mem); return r; @@ -379,7 +379,7 @@ static int radeon_move_ram_vram(struct ttm_buffer_object *bo, if (unlikely(r)) { return r; } - r = ttm_bo_move_ttm(bo, true, no_wait_gpu, &tmp_mem); + r = ttm_bo_move_ttm(bo, true, interruptible, no_wait_gpu, &tmp_mem); if (unlikely(r)) { goto out_cleanup; } diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 4054d804fe06..42c074a9c955 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -354,7 +354,8 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo, if (!(old_man->flags & TTM_MEMTYPE_FLAG_FIXED) && !(new_man->flags & TTM_MEMTYPE_FLAG_FIXED)) - ret = ttm_bo_move_ttm(bo, evict, no_wait_gpu, mem); + ret = ttm_bo_move_ttm(bo, evict, interruptible, no_wait_gpu, + mem); else if (bdev->driver->move) ret = bdev->driver->move(bo, evict, interruptible, no_wait_gpu, mem); diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 2df602a35f92..f157a9efd220 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -45,7 +45,7 @@ void ttm_bo_free_old_node(struct ttm_buffer_object *bo) } int ttm_bo_move_ttm(struct ttm_buffer_object *bo, - bool evict, + bool evict, bool interruptible, bool no_wait_gpu, struct ttm_mem_reg *new_mem) { struct ttm_tt *ttm = bo->ttm; @@ -53,6 +53,14 @@ int ttm_bo_move_ttm(struct ttm_buffer_object *bo, int ret; if (old_mem->mem_type != TTM_PL_SYSTEM) { + ret = ttm_bo_wait(bo, interruptible, no_wait_gpu); + + if (unlikely(ret != 0)) { + if (ret != -ERESTARTSYS) + pr_err("Failed to expire sync object before unbinding TTM\n"); + return ret; + } + ttm_tt_unbind(ttm); ttm_bo_free_old_node(bo); ttm_flag_masked(&old_mem->placement, TTM_PL_FLAG_SYSTEM, diff --git a/include/drm/ttm/ttm_bo_driver.h b/include/drm/ttm/ttm_bo_driver.h index 4348d6d5877a..99c6d01d24f2 100644 --- a/include/drm/ttm/ttm_bo_driver.h +++ b/include/drm/ttm/ttm_bo_driver.h @@ -962,6 +962,7 @@ void ttm_mem_io_free(struct ttm_bo_device *bdev, * * @bo: A pointer to a struct ttm_buffer_object. * @evict: 1: This is an eviction. Don't try to pipeline. + * @interruptible: Sleep interruptible if waiting. * @no_wait_gpu: Return immediately if the GPU is busy. * @new_mem: struct ttm_mem_reg indicating where to move. * @@ -976,7 +977,7 @@ void ttm_mem_io_free(struct ttm_bo_device *bdev, */ extern int ttm_bo_move_ttm(struct ttm_buffer_object *bo, - bool evict, bool no_wait_gpu, + bool evict, bool interruptible, bool no_wait_gpu, struct ttm_mem_reg *new_mem); /** From 3851f1cdb2b8d507b10395fc110d4c37d6121285 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 4 Aug 2016 00:08:45 -0400 Subject: [PATCH 050/235] SUNRPC: Limit the reconnect backoff timer to the max RPC message timeout ...and ensure that we propagate it to new transports on the same client. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 ++- net/sunrpc/clnt.c | 3 +++ net/sunrpc/xprtsock.c | 18 ++++++++++++------ 3 files changed, 17 insertions(+), 7 deletions(-) diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 5e3e1b63dbb3..a16070dd03ee 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -218,7 +218,8 @@ struct rpc_xprt { struct work_struct task_cleanup; struct timer_list timer; unsigned long last_used, - idle_timeout; + idle_timeout, + max_reconnect_timeout; /* * Send stuff diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index cb49898a5a58..faac5472d14d 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -2638,6 +2638,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt, { struct rpc_xprt_switch *xps; struct rpc_xprt *xprt; + unsigned long reconnect_timeout; unsigned char resvport; int ret = 0; @@ -2649,6 +2650,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt, return -EAGAIN; } resvport = xprt->resvport; + reconnect_timeout = xprt->max_reconnect_timeout; rcu_read_unlock(); xprt = xprt_create_transport(xprtargs); @@ -2657,6 +2659,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt, goto out_put_switch; } xprt->resvport = resvport; + xprt->max_reconnect_timeout = reconnect_timeout; rpc_xprt_switch_set_roundrobin(xps); if (setup) { diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 04b0c4190dd7..8ede3bc52481 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -177,7 +177,6 @@ static struct ctl_table sunrpc_table[] = { * increase over time if the server is down or not responding. */ #define XS_TCP_INIT_REEST_TO (3U * HZ) -#define XS_TCP_MAX_REEST_TO (5U * 60 * HZ) /* * TCP idle timeout; client drops the transport socket if it is idle @@ -2396,6 +2395,15 @@ static unsigned long xs_reconnect_delay(const struct rpc_xprt *xprt) return 0; } +static void xs_reconnect_backoff(struct rpc_xprt *xprt) +{ + xprt->reestablish_timeout <<= 1; + if (xprt->reestablish_timeout > xprt->max_reconnect_timeout) + xprt->reestablish_timeout = xprt->max_reconnect_timeout; + if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) + xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; +} + /** * xs_connect - connect a socket to a remote endpoint * @xprt: pointer to transport structure @@ -2426,12 +2434,8 @@ static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task) xs_reset_transport(transport); delay = xs_reconnect_delay(xprt); + xs_reconnect_backoff(xprt); - xprt->reestablish_timeout <<= 1; - if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) - xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; - if (xprt->reestablish_timeout > XS_TCP_MAX_REEST_TO) - xprt->reestablish_timeout = XS_TCP_MAX_REEST_TO; } else dprintk("RPC: xs_connect scheduled xprt %p\n", xprt); @@ -2989,6 +2993,8 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args) xprt->ops = &xs_tcp_ops; xprt->timeout = &xs_tcp_default_timeout; + xprt->max_reconnect_timeout = xprt->timeout->to_maxval; + INIT_WORK(&transport->recv_worker, xs_tcp_data_receive_workfn); INIT_DELAYED_WORK(&transport->connect_worker, xs_tcp_setup_socket); From fb10fb67ad2ce43d5e5b8ad22d2ba826844acc56 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Aug 2016 19:13:08 -0400 Subject: [PATCH 051/235] NFSv4: Cleanup the setting of the nfs4 lease period Make a helper function nfs4_set_lease_period() and have nfs41_setup_state_renewal() and nfs4_do_fsinfo() use it. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 4 ++++ fs/nfs/nfs4proc.c | 9 +++------ fs/nfs/nfs4renewd.c | 17 +++++++++++++++++ fs/nfs/nfs4state.c | 9 +++------ 4 files changed, 27 insertions(+), 12 deletions(-) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 4be567a54958..74a0e34e5ded 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -395,6 +395,10 @@ extern void nfs4_schedule_state_renewal(struct nfs_client *); extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); extern void nfs4_kill_renewd(struct nfs_client *); extern void nfs4_renew_state(struct work_struct *); +extern void nfs4_set_lease_period(struct nfs_client *clp, + unsigned long lease, + unsigned long lastrenewed); + /* nfs4state.c */ struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index da5c9e58e907..b9e18679af50 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4237,12 +4237,9 @@ static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, str err = _nfs4_do_fsinfo(server, fhandle, fsinfo); trace_nfs4_fsinfo(server, fhandle, fsinfo->fattr, err); if (err == 0) { - struct nfs_client *clp = server->nfs_client; - - spin_lock(&clp->cl_lock); - clp->cl_lease_time = fsinfo->lease_time * HZ; - clp->cl_last_renewal = now; - spin_unlock(&clp->cl_lock); + nfs4_set_lease_period(server->nfs_client, + fsinfo->lease_time * HZ, + now); break; } err = nfs4_handle_exception(server, err, &exception); diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index e1ba58c3d1ad..062029ab344a 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -136,6 +136,23 @@ nfs4_kill_renewd(struct nfs_client *clp) cancel_delayed_work_sync(&clp->cl_renewd); } +/** + * nfs4_set_lease_period - Sets the lease period on a nfs_client + * + * @clp: pointer to nfs_client + * @lease: new value for lease period + * @lastrenewed: time at which lease was last renewed + */ +void nfs4_set_lease_period(struct nfs_client *clp, + unsigned long lease, + unsigned long lastrenewed) +{ + spin_lock(&clp->cl_lock); + clp->cl_lease_time = lease; + clp->cl_last_renewal = lastrenewed; + spin_unlock(&clp->cl_lock); +} + /* * Local variables: * c-basic-offset: 8 diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 834b875900d6..cada00aa5096 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -277,20 +277,17 @@ static int nfs41_setup_state_renewal(struct nfs_client *clp) { int status; struct nfs_fsinfo fsinfo; + unsigned long now; if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { nfs4_schedule_state_renewal(clp); return 0; } + now = jiffies; status = nfs4_proc_get_lease_time(clp, &fsinfo); if (status == 0) { - /* Update lease time and schedule renewal */ - spin_lock(&clp->cl_lock); - clp->cl_lease_time = fsinfo.lease_time * HZ; - clp->cl_last_renewal = jiffies; - spin_unlock(&clp->cl_lock); - + nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now); nfs4_schedule_state_renewal(clp); } From 8d480326c3d6921ff5f1cc988c993bd572248deb Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Aug 2016 19:03:31 -0400 Subject: [PATCH 052/235] NFSv4: Cap the transport reconnection timer at 1/2 lease period We don't want to miss a lease period renewal due to the TCP connection failing to reconnect in a timely fashion. To ensure this doesn't happen, cap the reconnection timer so that we retry the connection attempt at least every 1/2 lease period. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4renewd.c | 3 +++ include/linux/sunrpc/clnt.h | 2 ++ net/sunrpc/clnt.c | 21 +++++++++++++++++++++ 3 files changed, 26 insertions(+) diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index 062029ab344a..82e77198d17e 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -151,6 +151,9 @@ void nfs4_set_lease_period(struct nfs_client *clp, clp->cl_lease_time = lease; clp->cl_last_renewal = lastrenewed; spin_unlock(&clp->cl_lock); + + /* Cap maximum reconnect timeout at 1/2 lease period */ + rpc_cap_max_reconnect_timeout(clp->cl_rpcclient, lease >> 1); } /* diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index b6810c92b8bb..5c02b0691587 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -195,6 +195,8 @@ int rpc_clnt_add_xprt(struct rpc_clnt *, struct xprt_create *, struct rpc_xprt *, void *), void *data); +void rpc_cap_max_reconnect_timeout(struct rpc_clnt *clnt, + unsigned long timeo); const char *rpc_proc_name(const struct rpc_task *task); #endif /* __KERNEL__ */ diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index faac5472d14d..7f79fb7dc6a0 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -2676,6 +2676,27 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt, } EXPORT_SYMBOL_GPL(rpc_clnt_add_xprt); +static int +rpc_xprt_cap_max_reconnect_timeout(struct rpc_clnt *clnt, + struct rpc_xprt *xprt, + void *data) +{ + unsigned long timeout = *((unsigned long *)data); + + if (timeout < xprt->max_reconnect_timeout) + xprt->max_reconnect_timeout = timeout; + return 0; +} + +void +rpc_cap_max_reconnect_timeout(struct rpc_clnt *clnt, unsigned long timeo) +{ + rpc_clnt_iterate_for_each_xprt(clnt, + rpc_xprt_cap_max_reconnect_timeout, + &timeo); +} +EXPORT_SYMBOL_GPL(rpc_cap_max_reconnect_timeout); + #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) static void rpc_show_header(void) { From 8e85946777d3f667085f298c0acfec12fe105078 Mon Sep 17 00:00:00 2001 From: Akshay Adiga Date: Thu, 4 Aug 2016 20:59:17 +0530 Subject: [PATCH 053/235] cpufreq: powernv: Fix crash in gpstate_timer_handler() Commit 09ca4c9b5958 (cpufreq: powernv: Replacing pstate_id with frequency table index) changes calc_global_pstate() to use cpufreq_table index instead of pstate_id. But in gpstate_timer_handler(), pstate_id was being passed instead of cpufreq_table index, which caused index_to_pstate() to access out of bound indices, leading to this crash. Adding sanity check for index and pstate, to ensure only valid pstate and index values are returned. Call Trace: [c00000078d66b130] [c00000000011d224] __free_irq+0x234/0x360 (unreliable) [c00000078d66b1c0] [c00000000011d44c] free_irq+0x6c/0xa0 [c00000078d66b1f0] [c00000000006c4f8] opal_event_shutdown+0x88/0xd0 [c00000078d66b230] [c000000000067a4c] opal_shutdown+0x1c/0x90 [c00000078d66b260] [c000000000063a00] pnv_shutdown+0x20/0x40 [c00000078d66b280] [c000000000021538] machine_restart+0x38/0x90 [c0000000078d66b310] [c000000000965ea0] panic+0x284/0x300 [c00000078d66b3a0] [c00000000001f508] die+0x388/0x450 [c00000078d66b430] [c000000000045a50] bad_page_fault+0xd0/0x140 [c00000078d66b4a0] [c000000000008964] handle_page_fault+0x2c/0x30 interrupt: 300 at gpstate_timer_handler+0x150/0x260 LR = gpstate_timer_handler+0x130/0x260 [c00000078d66b7f0] [c000000000132b58] call_timer_fn+0x58/0x1c0 [c00000078d66b880] [c000000000132e20] expire_timers+0x130/0x1d0 [c00000078d66b8f0] [c000000000133068] run_timer_softirq+0x1a8/0x230 [c00000078d66b980] [c0000000000b535c] __do_softirq+0x18c/0x400 [c00000078d66ba70] [c0000000000b5828] irq_exit+0xc8/0x100 [c00000078d66ba90] [c00000000001e214] timer_interrupt+0xa4/0xe0 [c00000078d66bac0] [c0000000000027d0] decrementer_common+0x150/0x180 interrupt: 901 at arch_local_irq_restore+0x74/0x90 0] [c000000000106b34] call_cpuidle+0x44/0x90 [c00000078d66be50] [c00000000010708c] cpu_startup_entry+0x38c/0x460 [c00000078d66bf20] [c00000000003d930] start_secondary+0x330/0x380 [c00000078d66bf90] [c000000000008e6c] start_secondary_prolog+0x10/0x14 Fixes: 09ca4c9b5958 (cpufreq: powernv: Replacing pstate_id with frequency table index) Reported-by: Madhavan Srinivasan Signed-off-by: Akshay Adiga Acked-by: Viresh Kumar Tested-by: Andrew Donnellan Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/powernv-cpufreq.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/powernv-cpufreq.c b/drivers/cpufreq/powernv-cpufreq.c index 0f138b050e9a..2fcc879d2b97 100644 --- a/drivers/cpufreq/powernv-cpufreq.c +++ b/drivers/cpufreq/powernv-cpufreq.c @@ -145,11 +145,30 @@ static struct powernv_pstate_info { /* Use following macros for conversions between pstate_id and index */ static inline int idx_to_pstate(unsigned int i) { + if (unlikely(i >= powernv_pstate_info.nr_pstates)) { + pr_warn_once("index %u is out of bound\n", i); + return powernv_freqs[powernv_pstate_info.nominal].driver_data; + } + return powernv_freqs[i].driver_data; } static inline unsigned int pstate_to_idx(int pstate) { + int min = powernv_freqs[powernv_pstate_info.min].driver_data; + int max = powernv_freqs[powernv_pstate_info.max].driver_data; + + if (min > 0) { + if (unlikely((pstate < max) || (pstate > min))) { + pr_warn_once("pstate %d is out of bound\n", pstate); + return powernv_pstate_info.nominal; + } + } else { + if (unlikely((pstate > max) || (pstate < min))) { + pr_warn_once("pstate %d is out of bound\n", pstate); + return powernv_pstate_info.nominal; + } + } /* * abs() is deliberately used so that is works with * both monotonically increasing and decreasing @@ -594,7 +613,7 @@ void gpstate_timer_handler(unsigned long data) } else { gpstate_idx = calc_global_pstate(gpstates->elapsed_time, gpstates->highest_lpstate_idx, - freq_data.pstate_id); + gpstates->last_lpstate_idx); } /* From 0b98027122d0715aa2dc39882e4c295112275785 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Fri, 29 Jul 2016 14:26:53 +1000 Subject: [PATCH 054/235] m68knommu: fix user a5 register being overwritten On no-MMU systems the application a5 register can be overwitten with the address of the process data segment when processing application signals. For flat format applications compiled with full absolute relocation this effectively corrupts the a5 register on signal processing - and this very quickly leads to process crash and often takes out the whole system with a panic as well. This has no effect on flat format applications compiled with the more common PIC methods (such as -msep-data). These format applications reserve a5 for the pointer to the data segment anyway - so it doesn't change it. A long time ago the a5 register was used in the code packed into the user stack to enable signal return processing. And so it had to be restored on end of signal cleanup processing back to the original a5 user value. This was historically done by saving away a5 in the sigcontext structure. At some point (a long time back it seems) the a5 restore process was changed and it was hard coded to put the user data segment address directly into a5. Which is ok for the common PIC compiled application case, but breaks the full relocation application code. We no longer use this type of signal handling mechanism and so we don't need to do anything special to save and restore a5 at all now. So remove the code that hard codes a5 to the address of the user data segment. Signed-off-by: Greg Ungerer --- arch/m68k/kernel/signal.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/m68k/kernel/signal.c b/arch/m68k/kernel/signal.c index 2dcee3a88867..9202f82dfce6 100644 --- a/arch/m68k/kernel/signal.c +++ b/arch/m68k/kernel/signal.c @@ -213,7 +213,6 @@ static inline int frame_extra_sizes(int f) static inline void adjustformat(struct pt_regs *regs) { - ((struct switch_stack *)regs - 1)->a5 = current->mm->start_data; /* * set format byte to make stack appear modulo 4, which it will * be when doing the rte From 176b1ec21306f0775e6d7a03f42d82a0b144ef0e Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Thu, 23 Jun 2016 15:02:46 -0700 Subject: [PATCH 055/235] thermal: intel_pch_thermal: Add suspend/resume callback Added suspend/resume callback to disable/enable PCH thermal sensor respectively. If the sensor is enabled by the BIOS, then the sensor status will not be changed during suspend/resume. Signed-off-by: Srinivas Pandruvada Signed-off-by: Zhang Rui --- drivers/thermal/intel_pch_thermal.c | 60 ++++++++++++++++++++++++++++- 1 file changed, 59 insertions(+), 1 deletion(-) diff --git a/drivers/thermal/intel_pch_thermal.c b/drivers/thermal/intel_pch_thermal.c index 6a6ec1c95a7a..9b4815e81b0d 100644 --- a/drivers/thermal/intel_pch_thermal.c +++ b/drivers/thermal/intel_pch_thermal.c @@ -21,6 +21,7 @@ #include #include #include +#include /* Intel PCH thermal Device IDs */ #define PCH_THERMAL_DID_WPT 0x9CA4 /* Wildcat Point */ @@ -65,6 +66,7 @@ struct pch_thermal_device { unsigned long crt_temp; int hot_trip_id; unsigned long hot_temp; + bool bios_enabled; }; static int pch_wpt_init(struct pch_thermal_device *ptd, int *nr_trips) @@ -75,8 +77,10 @@ static int pch_wpt_init(struct pch_thermal_device *ptd, int *nr_trips) *nr_trips = 0; /* Check if BIOS has already enabled thermal sensor */ - if (WPT_TSS_TSDSS & readb(ptd->hw_base + WPT_TSS)) + if (WPT_TSS_TSDSS & readb(ptd->hw_base + WPT_TSS)) { + ptd->bios_enabled = true; goto read_trips; + } tsel = readb(ptd->hw_base + WPT_TSEL); /* @@ -130,9 +134,39 @@ static int pch_wpt_get_temp(struct pch_thermal_device *ptd, int *temp) return 0; } +static int pch_wpt_suspend(struct pch_thermal_device *ptd) +{ + u8 tsel; + + if (ptd->bios_enabled) + return 0; + + tsel = readb(ptd->hw_base + WPT_TSEL); + + writeb(tsel & 0xFE, ptd->hw_base + WPT_TSEL); + + return 0; +} + +static int pch_wpt_resume(struct pch_thermal_device *ptd) +{ + u8 tsel; + + if (ptd->bios_enabled) + return 0; + + tsel = readb(ptd->hw_base + WPT_TSEL); + + writeb(tsel | WPT_TSEL_ETS, ptd->hw_base + WPT_TSEL); + + return 0; +} + struct pch_dev_ops { int (*hw_init)(struct pch_thermal_device *ptd, int *nr_trips); int (*get_temp)(struct pch_thermal_device *ptd, int *temp); + int (*suspend)(struct pch_thermal_device *ptd); + int (*resume)(struct pch_thermal_device *ptd); }; @@ -140,6 +174,8 @@ struct pch_dev_ops { static const struct pch_dev_ops pch_dev_ops_wpt = { .hw_init = pch_wpt_init, .get_temp = pch_wpt_get_temp, + .suspend = pch_wpt_suspend, + .resume = pch_wpt_resume, }; static int pch_thermal_get_temp(struct thermal_zone_device *tzd, int *temp) @@ -269,6 +305,22 @@ static void intel_pch_thermal_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static int intel_pch_thermal_suspend(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct pch_thermal_device *ptd = pci_get_drvdata(pdev); + + return ptd->ops->suspend(ptd); +} + +static int intel_pch_thermal_resume(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct pch_thermal_device *ptd = pci_get_drvdata(pdev); + + return ptd->ops->resume(ptd); +} + static struct pci_device_id intel_pch_thermal_id[] = { { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_WPT) }, { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_SKL) }, @@ -276,11 +328,17 @@ static struct pci_device_id intel_pch_thermal_id[] = { }; MODULE_DEVICE_TABLE(pci, intel_pch_thermal_id); +static const struct dev_pm_ops intel_pch_pm_ops = { + .suspend = intel_pch_thermal_suspend, + .resume = intel_pch_thermal_resume, +}; + static struct pci_driver intel_pch_thermal_driver = { .name = "intel_pch_thermal", .id_table = intel_pch_thermal_id, .probe = intel_pch_thermal_probe, .remove = intel_pch_thermal_remove, + .driver.pm = &intel_pch_pm_ops, }; module_pci_driver(intel_pch_thermal_driver); From 70c50ee72eebb3d1e3aae7450269fd8d8074c610 Mon Sep 17 00:00:00 2001 From: Petr Mladek Date: Fri, 5 Aug 2016 15:20:41 +0200 Subject: [PATCH 056/235] thermal/powerclamp: Prevent division by zero when counting interval I have got a zero division error when disabling the forced idle injection from the intel powerclamp. I did echo 0 >/sys/class/thermal/cooling_device48/cur_state and got [ 986.072632] divide error: 0000 [#1] PREEMPT SMP [ 986.078989] Modules linked in: [ 986.083618] CPU: 17 PID: 24967 Comm: kidle_inject/17 Not tainted 4.7.0-1-default+ #3055 [ 986.093781] Hardware name: Intel Corporation S2600CP/S2600CP, BIOS RMLSDP.86I.R3.27.D685.1305151734 05/15/2013 [ 986.106227] task: ffff880430e1c080 task.stack: ffff880427ef0000 [ 986.114122] RIP: 0010:[] [] clamp_thread+0x1d9/0x600 [ 986.124609] RSP: 0018:ffff880427ef3e20 EFLAGS: 00010246 [ 986.131860] RAX: 0000000000000258 RBX: 0000000000000006 RCX: 0000000000000001 [ 986.141179] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000018 [ 986.150478] RBP: ffff880427ef3ec8 R08: ffff880427ef0000 R09: 0000000000000002 [ 986.159779] R10: 0000000000003df2 R11: 0000000000000018 R12: 0000000000000002 [ 986.169089] R13: 0000000000000000 R14: ffff880427ef0000 R15: ffff880427ef0000 [ 986.178388] FS: 0000000000000000(0000) GS:ffff880435940000(0000) knlGS:0000000000000000 [ 986.188785] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 986.196559] CR2: 00007f1d0caf0000 CR3: 0000000002006000 CR4: 00000000001406e0 [ 986.205909] Stack: [ 986.209524] ffff8802be897b00 ffff880430e1c080 0000000000000011 0000006a35959780 [ 986.219236] 0000000000000011 ffff880427ef0008 0000000000000000 ffff8804359503d0 [ 986.228966] 0000000100029d93 ffffffff81794140 0000000000000000 ffffffff05000011 [ 986.238686] Call Trace: [ 986.242825] [] ? pkg_state_counter+0x80/0x80 [ 986.250866] [] ? powerclamp_set_cur_state+0x180/0x180 [ 986.259797] [] kthread+0xc9/0xe0 [ 986.266682] [] ret_from_fork+0x1f/0x40 [ 986.274142] [] ? kthread_create_on_node+0x180/0x180 [ 986.282869] Code: d1 ea 48 89 d6 80 3d 6a d0 d4 00 00 ba 64 00 00 00 89 d8 41 0f 45 f5 0f af c2 42 8d 14 2e be 31 00 00 00 83 fa 31 0f 42 f2 31 d2 f6 48 8b 15 9e 07 87 00 48 8b 3d 97 07 87 00 48 63 f0 83 e8 [ 986.307806] RIP [] clamp_thread+0x1d9/0x600 [ 986.315871] RSP RIP points to the following lines: compensation = get_compensation(target_ratio); interval = duration_jiffies*100/(target_ratio+compensation); A solution would be to switch the following two commands in powerclamp_set_cur_state(): set_target_ratio = 0; end_power_clamp(); But I think that the zero division might happen also when target_ratio is non-zero because the compensation might be negative. Therefore we also check the sum of target_ratio and compensation explicitly. Also the compensated_ratio variable is always set. Therefore there is no need to initialize it. Signed-off-by: Petr Mladek Acked-by: Jacob Pan Signed-off-by: Zhang Rui --- drivers/thermal/intel_powerclamp.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/thermal/intel_powerclamp.c b/drivers/thermal/intel_powerclamp.c index 015ce2eb6eb7..0e4dc0afcfd2 100644 --- a/drivers/thermal/intel_powerclamp.c +++ b/drivers/thermal/intel_powerclamp.c @@ -388,7 +388,7 @@ static int clamp_thread(void *arg) int sleeptime; unsigned long target_jiffies; unsigned int guard; - unsigned int compensation = 0; + unsigned int compensated_ratio; int interval; /* jiffies to sleep for each attempt */ unsigned int duration_jiffies = msecs_to_jiffies(duration); unsigned int window_size_now; @@ -409,8 +409,11 @@ static int clamp_thread(void *arg) * c-states, thus we need to compensate the injected idle ratio * to achieve the actual target reported by the HW. */ - compensation = get_compensation(target_ratio); - interval = duration_jiffies*100/(target_ratio+compensation); + compensated_ratio = target_ratio + + get_compensation(target_ratio); + if (compensated_ratio <= 0) + compensated_ratio = 1; + interval = duration_jiffies * 100 / compensated_ratio; /* align idle time */ target_jiffies = roundup(jiffies, interval); @@ -647,8 +650,8 @@ static int powerclamp_set_cur_state(struct thermal_cooling_device *cdev, goto exit_set; } else if (set_target_ratio > 0 && new_target_ratio == 0) { pr_info("Stop forced idle injection\n"); - set_target_ratio = 0; end_power_clamp(); + set_target_ratio = 0; } else /* adjust currently running */ { set_target_ratio = new_target_ratio; /* make new set_target_ratio visible to other cpus */ From d0b7306d203c82e7c04d6eb066ca4898f016ebdd Mon Sep 17 00:00:00 2001 From: Michele Di Giorgio Date: Thu, 2 Jun 2016 15:25:31 +0100 Subject: [PATCH 057/235] thermal: fix race condition when updating cooling device When multiple thermal zones are bound to the same cooling device, multiple kernel threads may want to update the cooling device state by calling thermal_cdev_update(). Having cdev not protected by a mutex can lead to a race condition. Consider the following situation with two kernel threads k1 and k2: Thread k1 Thread k2 || || call thermal_cdev_update() || ... || set_cur_state(cdev, target); call power_actor_set_power() || ... || instance->target = state; || cdev->updated = false; || || cdev->updated = true; || // completes execution call thermal_cdev_update() || // cdev->updated == true || return; || \/ time k2 has already looped through the thermal instances looking for the deepest cooling device state and is preempted right before setting cdev->updated to true. Now, k1 runs, modifies the thermal instance state and sets cdev->updated to false. Then, k1 is preempted and k2 continues the execution by setting cdev->updated to true, therefore preventing k1 from performing the update. Notice that this is not an issue if k2 looks at the instance->target modified by k1 "after" it is assigned by k1. In fact, in this case the update will happen anyway and k1 can safely return immediately from thermal_cdev_update(). This may lead to a situation where a thermal governor never updates the cooling device. For example, this is the case for the step_wise governor: when calling the function thermal_zone_trip_update(), the governor may always get a new state equal to the old one (which, however, wasn't notified to the cooling device) and will therefore skip the update. CC: Zhang Rui CC: Eduardo Valentin CC: Peter Feuerer Reported-by: Toby Huang Signed-off-by: Michele Di Giorgio Reviewed-by: Javi Merino Signed-off-by: Zhang Rui --- drivers/thermal/fair_share.c | 2 ++ drivers/thermal/gov_bang_bang.c | 2 ++ drivers/thermal/power_allocator.c | 2 ++ drivers/thermal/step_wise.c | 2 ++ drivers/thermal/thermal_core.c | 14 +++++++++----- 5 files changed, 17 insertions(+), 5 deletions(-) diff --git a/drivers/thermal/fair_share.c b/drivers/thermal/fair_share.c index 34fe36504a55..68bd1b569118 100644 --- a/drivers/thermal/fair_share.c +++ b/drivers/thermal/fair_share.c @@ -116,7 +116,9 @@ static int fair_share_throttle(struct thermal_zone_device *tz, int trip) instance->target = get_target_state(tz, cdev, percentage, cur_trip_level); + mutex_lock(&instance->cdev->lock); instance->cdev->updated = false; + mutex_unlock(&instance->cdev->lock); thermal_cdev_update(cdev); } return 0; diff --git a/drivers/thermal/gov_bang_bang.c b/drivers/thermal/gov_bang_bang.c index fc52016d4e85..bb118a152cbb 100644 --- a/drivers/thermal/gov_bang_bang.c +++ b/drivers/thermal/gov_bang_bang.c @@ -71,7 +71,9 @@ static void thermal_zone_trip_update(struct thermal_zone_device *tz, int trip) dev_dbg(&instance->cdev->device, "target=%d\n", (int)instance->target); + mutex_lock(&instance->cdev->lock); instance->cdev->updated = false; /* cdev needs update */ + mutex_unlock(&instance->cdev->lock); } mutex_unlock(&tz->lock); diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 2f1a863a8e15..b4d3116cfdaf 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -529,7 +529,9 @@ static void allow_maximum_power(struct thermal_zone_device *tz) continue; instance->target = 0; + mutex_lock(&instance->cdev->lock); instance->cdev->updated = false; + mutex_unlock(&instance->cdev->lock); thermal_cdev_update(instance->cdev); } } diff --git a/drivers/thermal/step_wise.c b/drivers/thermal/step_wise.c index ea9366ad3e6b..bcef2e7c4ec9 100644 --- a/drivers/thermal/step_wise.c +++ b/drivers/thermal/step_wise.c @@ -175,7 +175,9 @@ static void thermal_zone_trip_update(struct thermal_zone_device *tz, int trip) update_passive_instance(tz, trip_type, -1); instance->initialized = true; + mutex_lock(&instance->cdev->lock); instance->cdev->updated = false; /* cdev needs update */ + mutex_unlock(&instance->cdev->lock); } mutex_unlock(&tz->lock); diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c index 5133cd1e10b7..e2fc6161dded 100644 --- a/drivers/thermal/thermal_core.c +++ b/drivers/thermal/thermal_core.c @@ -1093,7 +1093,9 @@ int power_actor_set_power(struct thermal_cooling_device *cdev, return ret; instance->target = state; + mutex_lock(&cdev->lock); cdev->updated = false; + mutex_unlock(&cdev->lock); thermal_cdev_update(cdev); return 0; @@ -1623,11 +1625,13 @@ void thermal_cdev_update(struct thermal_cooling_device *cdev) struct thermal_instance *instance; unsigned long target = 0; - /* cooling device is updated*/ - if (cdev->updated) - return; - mutex_lock(&cdev->lock); + /* cooling device is updated*/ + if (cdev->updated) { + mutex_unlock(&cdev->lock); + return; + } + /* Make sure cdev enters the deepest cooling state */ list_for_each_entry(instance, &cdev->thermal_instances, cdev_node) { dev_dbg(&cdev->device, "zone%d->target=%lu\n", @@ -1637,9 +1641,9 @@ void thermal_cdev_update(struct thermal_cooling_device *cdev) if (instance->target > target) target = instance->target; } - mutex_unlock(&cdev->lock); cdev->ops->set_cur_state(cdev, target); cdev->updated = true; + mutex_unlock(&cdev->lock); trace_cdev_update(cdev, target); dev_dbg(&cdev->device, "set to state %lu\n", target); } From f4c592439b144ef1de66eac764140643a54099eb Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 4 Jul 2016 07:19:32 +0000 Subject: [PATCH 058/235] thermal: hwmon: EXPORT_SYMBOL_GPL for thermal hwmon sysfs thermal_add_hwmon_sysfs()/thermal_remove_hwmon_sysfs() need EXPORT_SYMBOL_GPL(). Otherwise we will have ERROR >> ERROR: "thermal_remove_hwmon_sysfs" [drivers/thermal/rcar_thermal.ko] undefined! >> ERROR: "thermal_add_hwmon_sysfs" [drivers/thermal/rcar_thermal.ko] undefined! Signed-off-by: Kuninori Morimoto Signed-off-by: Zhang Rui --- drivers/thermal/thermal_hwmon.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/thermal/thermal_hwmon.c b/drivers/thermal/thermal_hwmon.c index 06fd2ed9ef9d..c41c7742903a 100644 --- a/drivers/thermal/thermal_hwmon.c +++ b/drivers/thermal/thermal_hwmon.c @@ -232,6 +232,7 @@ int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) return result; } +EXPORT_SYMBOL_GPL(thermal_add_hwmon_sysfs); void thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz) { @@ -270,3 +271,4 @@ void thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz) hwmon_device_unregister(hwmon->device); kfree(hwmon); } +EXPORT_SYMBOL_GPL(thermal_remove_hwmon_sysfs); From 165989a5b667b90589f21c8affe496ad21f08591 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 25 Jul 2016 07:01:19 +0000 Subject: [PATCH 059/235] thermal: clock_cooling: Fix missing mutex_init() The driver allocates the mutex but not initialize it. Use mutex_init() on it to initialize it correctly. This is detected by Coccinelle semantic patch. Signed-off-by: Wei Yongjun Signed-off-by: Zhang Rui --- drivers/thermal/clock_cooling.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/thermal/clock_cooling.c b/drivers/thermal/clock_cooling.c index 1b4ff0f4c716..ed5dd0e88657 100644 --- a/drivers/thermal/clock_cooling.c +++ b/drivers/thermal/clock_cooling.c @@ -426,6 +426,7 @@ clock_cooling_register(struct device *dev, const char *clock_name) if (!ccdev) return ERR_PTR(-ENOMEM); + mutex_init(&ccdev->lock); ccdev->dev = dev; ccdev->clk = devm_clk_get(dev, clock_name); if (IS_ERR(ccdev->clk)) From 29986cc8a66fcc4b8ea8a01216a72f06b865360b Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Wed, 3 Aug 2016 23:31:36 +0300 Subject: [PATCH 060/235] drm: rcar-du: Link HDMI encoder with bridge The conversion of the rcar-du driver from the I2C slave encoder to the DRM bridge API left the HDMI encoder's bridge pointer NULL, preventing the bridge from being handled automatically by the DRM core. Fix it. Fixes: 1d926114d8f4 ("drm: rcar-du: Remove i2c slave encoder interface for hdmi encoder") Signed-off-by: Laurent Pinchart Signed-off-by: Dave Airlie --- drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c b/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c index 4de3ff0dbebd..e03004f4588d 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c @@ -125,6 +125,7 @@ int rcar_du_hdmienc_init(struct rcar_du_device *rcdu, /* Link drm_bridge to encoder */ bridge->encoder = encoder; + encoder->bridge = bridge; ret = drm_bridge_attach(rcdu->ddev, bridge); if (ret) { From 5c6c201ccbaf9d3901f829441d457293f7ca8ef4 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 4 Aug 2016 18:35:48 +0200 Subject: [PATCH 061/235] drm: Paper over locking inversion after registration rework drm_connector_register_all requires a few too many locks because our connector_list locking is busted. Add another FIXME+hack to work around this. This should address the below lockdep splat: ====================================================== [ INFO: possible circular locking dependency detected ] 4.7.0-rc5+ #524 Tainted: G O ------------------------------------------------------- kworker/u8:0/6 is trying to acquire lock: (&dev->mode_config.mutex){+.+.+.}, at: [] drm_modeset_lock_all+0x40/0x120 but task is already holding lock: ((fb_notifier_list).rwsem){++++.+}, at: [] __blocking_notifier_call_chain+0x35/0x70 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 ((fb_notifier_list).rwsem){++++.+}: [] lock_acquire+0xb1/0x200 [] down_write+0x44/0x80 [] blocking_notifier_chain_register+0x21/0xb0 [] fb_register_client+0x18/0x20 [] backlight_device_register+0x136/0x260 [] intel_backlight_device_register+0xa2/0x160 [i915] [] intel_connector_register+0xe/0x10 [i915] [] intel_dp_connector_register+0x1b/0x80 [i915] [] drm_connector_register+0x4a/0x80 [] drm_connector_register_all+0x64/0xf0 [] drm_modeset_register_all+0x174/0x1c0 [] drm_dev_register+0xc2/0xd0 [] i915_driver_load+0x1547/0x2200 [i915] [] i915_pci_probe+0x4f/0x70 [i915] [] local_pci_probe+0x45/0xa0 [] pci_device_probe+0xdb/0x130 [] driver_probe_device+0x223/0x440 [] __driver_attach+0xd5/0x100 [] bus_for_each_dev+0x66/0xa0 [] driver_attach+0x1e/0x20 [] bus_add_driver+0x1ee/0x280 [] driver_register+0x60/0xe0 [] __pci_register_driver+0x60/0x70 [] i915_init+0x5b/0x62 [i915] [] do_one_initcall+0x3d/0x150 [] do_init_module+0x5f/0x1d9 [] load_module+0x20e6/0x27e0 [] SYSC_finit_module+0xc3/0xf0 [] SyS_finit_module+0xe/0x10 [] entry_SYSCALL_64_fastpath+0x1c/0xac -> #0 (&dev->mode_config.mutex){+.+.+.}: [] __lock_acquire+0x10fc/0x1260 [] lock_acquire+0xb1/0x200 [] mutex_lock_nested+0x67/0x3c0 [] drm_modeset_lock_all+0x40/0x120 [] drm_fb_helper_restore_fbdev_mode_unlocked+0x2b/0x80 [] drm_fb_helper_set_par+0x2d/0x50 [] intel_fbdev_set_par+0x1a/0x60 [i915] [] fbcon_init+0x586/0x610 [] visual_init+0xca/0x130 [] do_bind_con_driver+0x1c1/0x3a0 [] do_take_over_console+0x116/0x180 [] do_fbcon_takeover+0x57/0xb0 [] fbcon_event_notify+0x658/0x750 [] notifier_call_chain+0x3e/0xb0 [] __blocking_notifier_call_chain+0x4d/0x70 [] blocking_notifier_call_chain+0x16/0x20 [] fb_notifier_call_chain+0x1b/0x20 [] register_framebuffer+0x251/0x330 [] drm_fb_helper_initial_config+0x25f/0x3f0 [] intel_fbdev_initial_config+0x18/0x30 [i915] [] async_run_entry_fn+0x48/0x150 [] process_one_work+0x1e7/0x750 [] worker_thread+0x4b/0x4f0 [] kthread+0xef/0x110 [] ret_from_fork+0x1f/0x40 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock((fb_notifier_list).rwsem); lock(&dev->mode_config.mutex); lock((fb_notifier_list).rwsem); lock(&dev->mode_config.mutex); *** DEADLOCK *** 6 locks held by kworker/u8:0/6: #0: ("events_unbound"){.+.+.+}, at: [] process_one_work+0x169/0x750 #1: ((&entry->work)){+.+.+.}, at: [] process_one_work+0x169/0x750 #2: (registration_lock){+.+.+.}, at: [] register_framebuffer+0x27/0x330 #3: (console_lock){+.+.+.}, at: [] register_framebuffer+0x26e/0x330 #4: (&fb_info->lock){+.+.+.}, at: [] lock_fb_info+0x1d/0x40 #5: ((fb_notifier_list).rwsem){++++.+}, at: [] __blocking_notifier_call_chain+0x35/0x70 stack backtrace: CPU: 2 PID: 6 Comm: kworker/u8:0 Tainted: G O 4.7.0-rc5+ #524 Hardware name: Intel Corp. Broxton P/NOTEBOOK, BIOS APLKRVPA.X64.0138.B33.1606250842 06/25/2016 Workqueue: events_unbound async_run_entry_fn 0000000000000000 ffff8800758577f0 ffffffff814507a5 ffffffff828b9900 ffffffff828b9900 ffff880075857830 ffffffff810dc6fa ffff880075857880 ffff88007584d688 0000000000000005 0000000000000006 ffff88007584d6b0 Call Trace: [] dump_stack+0x67/0x92 [] print_circular_bug+0x1aa/0x200 [] __lock_acquire+0x10fc/0x1260 [] lock_acquire+0xb1/0x200 [] ? drm_modeset_lock_all+0x40/0x120 [] ? drm_modeset_lock_all+0x40/0x120 [] mutex_lock_nested+0x67/0x3c0 [] ? drm_modeset_lock_all+0x40/0x120 [] ? rcu_read_lock_sched_held+0x7f/0x90 [] ? kmem_cache_alloc_trace+0x248/0x2b0 [] ? drm_modeset_lock_all+0x25/0x120 [] drm_modeset_lock_all+0x40/0x120 [] drm_fb_helper_restore_fbdev_mode_unlocked+0x2b/0x80 [] drm_fb_helper_set_par+0x2d/0x50 [] intel_fbdev_set_par+0x1a/0x60 [i915] [] fbcon_init+0x586/0x610 [] visual_init+0xca/0x130 [] do_bind_con_driver+0x1c1/0x3a0 [] do_take_over_console+0x116/0x180 [] do_fbcon_takeover+0x57/0xb0 [] fbcon_event_notify+0x658/0x750 [] notifier_call_chain+0x3e/0xb0 [] __blocking_notifier_call_chain+0x4d/0x70 [] blocking_notifier_call_chain+0x16/0x20 [] fb_notifier_call_chain+0x1b/0x20 [] register_framebuffer+0x251/0x330 [] ? vga_switcheroo_client_fb_set+0x5d/0x70 [] drm_fb_helper_initial_config+0x25f/0x3f0 [] intel_fbdev_initial_config+0x18/0x30 [i915] [] async_run_entry_fn+0x48/0x150 [] process_one_work+0x1e7/0x750 [] ? process_one_work+0x169/0x750 [] worker_thread+0x4b/0x4f0 [] ? process_one_work+0x750/0x750 [] kthread+0xef/0x110 [] ret_from_fork+0x1f/0x40 [] ? kthread_stop+0x2e0/0x2e0 v2: Rebase onto the right branch (hand-editing patches ftw) and add more reporters. Reported-by: Imre Deak Cc: Imre Deak Cc: Chris Wilson Acked-by: Chris Wilson Reported-by: Jiri Kosina Cc: Jiri Kosina Signed-off-by: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_crtc.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index f1d9f0569d7f..b1dbb60af99f 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -1121,16 +1121,14 @@ static int drm_connector_register_all(struct drm_device *dev) struct drm_connector *connector; int ret; - mutex_lock(&dev->mode_config.mutex); - - drm_for_each_connector(connector, dev) { + /* FIXME: taking the mode config mutex ends up in a clash with + * fbcon/backlight registration */ + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { ret = drm_connector_register(connector); if (ret) goto err; } - mutex_unlock(&dev->mode_config.mutex); - return 0; err: From 416f37d0816b9720b8227953e55954d81456f991 Mon Sep 17 00:00:00 2001 From: Darren Stevens Date: Wed, 27 Jul 2016 15:41:29 +0100 Subject: [PATCH 062/235] powerpc/pasemi: Fix coherent_dma_mask for dma engine Commit 817820b0226a ("powerpc/iommu: Support "hybrid" iommu/direct DMA ops for coherent_mask < dma_mask) adds a check of coherent_dma_mask for dma allocations. Unfortunately current PASemi code does not set this value for the DMA engine, which ends up with the default value of 0xffffffff, the result is on a PASemi system with >2Gb ram and iommu enabled the the onboard ethernet stops working due to an inability to allocate memory. Add an initialisation to pci_dma_dev_setup_pasemi(). Signed-off-by: Darren Stevens Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/pasemi/iommu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/powerpc/platforms/pasemi/iommu.c b/arch/powerpc/platforms/pasemi/iommu.c index 309d9ccccd50..c61667e8bb06 100644 --- a/arch/powerpc/platforms/pasemi/iommu.c +++ b/arch/powerpc/platforms/pasemi/iommu.c @@ -187,6 +187,11 @@ static void pci_dma_dev_setup_pasemi(struct pci_dev *dev) if (dev->vendor == 0x1959 && dev->device == 0xa007 && !firmware_has_feature(FW_FEATURE_LPAR)) { dev->dev.archdata.dma_ops = &dma_direct_ops; + /* + * Set the coherent DMA mask to prevent the iommu + * being used unnecessarily + */ + dev->dev.coherent_dma_mask = DMA_BIT_MASK(44); return; } #endif From 68202c9f0ad6e16ee806fbadbc5838d55fe5aa5c Mon Sep 17 00:00:00 2001 From: Ross Zwisler Date: Fri, 29 Jul 2016 14:59:12 -0600 Subject: [PATCH 063/235] libnvdimm, nd_blk: mask off reserved status bits The "NVDIMM Block Window Driver Writer's Guide": http://pmem.io/documents/NVDIMM_DriverWritersGuide-July-2016.pdf ...defines the layout of the block window status register. For the July 2016 version of the spec linked to above, this happens in Figure 4 on page 26. The only bits defined in this spec are bits 31, 5, 4, 2, 1 and 0. The rest of the bits in the status register are reserved, and there is a warning following the diagram that says: Note: The driver cannot assume the value of the RESERVED bits in the status register are zero. These reserved bits need to be masked off, and the driver must avoid checking the state of those bits. This change ensures that for hardware implementations that set these reserved bits in the status register, the driver won't incorrectly fail the block I/Os. Cc: #v4.2+ Reviewed-by: Lee, Chun-Yi Signed-off-by: Ross Zwisler Signed-off-by: Dan Williams --- drivers/acpi/nfit/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c index 8c234dd9b8bc..80cc7c089a15 100644 --- a/drivers/acpi/nfit/core.c +++ b/drivers/acpi/nfit/core.c @@ -1527,11 +1527,12 @@ static u32 read_blk_stat(struct nfit_blk *nfit_blk, unsigned int bw) { struct nfit_blk_mmio *mmio = &nfit_blk->mmio[DCR]; u64 offset = nfit_blk->stat_offset + mmio->size * bw; + const u32 STATUS_MASK = 0x80000037; if (mmio->num_lines) offset = to_interleave_offset(offset, mmio); - return readl(mmio->addr.base + offset); + return readl(mmio->addr.base + offset) & STATUS_MASK; } static void write_blk_ctl(struct nfit_blk *nfit_blk, unsigned int bw, From abe8b4e3cef88b8202641d63f5ad58141b970b0f Mon Sep 17 00:00:00 2001 From: Vishal Verma Date: Wed, 27 Jul 2016 16:38:59 -0600 Subject: [PATCH 064/235] nvdimm, btt: add a size attribute for BTTs To be consistent with other namespaces, expose a 'size' attribute for BTT devices also. Cc: Dan Williams Reported-by: Linda Knippers Signed-off-by: Vishal Verma Signed-off-by: Dan Williams --- drivers/nvdimm/btt.c | 1 + drivers/nvdimm/btt_devs.c | 20 ++++++++++++++++++++ drivers/nvdimm/nd.h | 1 + 3 files changed, 22 insertions(+) diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c index 88e91666f145..368795aad5c9 100644 --- a/drivers/nvdimm/btt.c +++ b/drivers/nvdimm/btt.c @@ -1269,6 +1269,7 @@ static int btt_blk_init(struct btt *btt) } } set_capacity(btt->btt_disk, btt->nlba * btt->sector_size >> 9); + btt->nd_btt->size = btt->nlba * (u64)btt->sector_size; revalidate_disk(btt->btt_disk); return 0; diff --git a/drivers/nvdimm/btt_devs.c b/drivers/nvdimm/btt_devs.c index 3fa7919f94a8..97dd2925ed6e 100644 --- a/drivers/nvdimm/btt_devs.c +++ b/drivers/nvdimm/btt_devs.c @@ -140,10 +140,30 @@ static ssize_t namespace_store(struct device *dev, } static DEVICE_ATTR_RW(namespace); +static ssize_t size_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nd_btt *nd_btt = to_nd_btt(dev); + ssize_t rc; + + device_lock(dev); + if (dev->driver) + rc = sprintf(buf, "%llu\n", nd_btt->size); + else { + /* no size to convey if the btt instance is disabled */ + rc = -ENXIO; + } + device_unlock(dev); + + return rc; +} +static DEVICE_ATTR_RO(size); + static struct attribute *nd_btt_attributes[] = { &dev_attr_sector_size.attr, &dev_attr_namespace.attr, &dev_attr_uuid.attr, + &dev_attr_size.attr, NULL, }; diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h index 40476399d227..8024a0ef86d3 100644 --- a/drivers/nvdimm/nd.h +++ b/drivers/nvdimm/nd.h @@ -143,6 +143,7 @@ struct nd_btt { struct nd_namespace_common *ndns; struct btt *btt; unsigned long lbasize; + u64 size; u8 *uuid; int id; }; From 65a97a67a7d7b7a953492573d5740d32510b9daa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Pali=20Roh=C3=A1r?= Date: Wed, 27 Jul 2016 09:34:42 +0200 Subject: [PATCH 065/235] dell-wmi: Ignore WMI event 0xe00e MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit WMI event 0xe00e is received when battery was removed or inserted. Signed-off-by: Pali Rohár Signed-off-by: Darren Hart --- drivers/platform/x86/dell-wmi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c index d2bc092defd7..da2fe18162e1 100644 --- a/drivers/platform/x86/dell-wmi.c +++ b/drivers/platform/x86/dell-wmi.c @@ -110,8 +110,8 @@ static const struct key_entry dell_wmi_keymap_type_0000[] __initconst = { /* BIOS error detected */ { KE_IGNORE, 0xe00d, { KEY_RESERVED } }, - /* Unknown, defined in ACPI DSDT */ - /* { KE_IGNORE, 0xe00e, { KEY_RESERVED } }, */ + /* Battery was removed or inserted */ + { KE_IGNORE, 0xe00e, { KEY_RESERVED } }, /* Wifi Catcher */ { KE_KEY, 0xe011, { KEY_PROG2 } }, From c22e853a2ed19321d00c1eae339ffdc4f5e7757e Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Sat, 30 Jul 2016 00:37:57 +0000 Subject: [PATCH 066/235] libceph: fix return value check in alloc_msg_with_page_vector() In case of error, the function ceph_alloc_page_vector() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Fixes: 1907920324f1 ('libceph: support for sending notifies') Signed-off-by: Wei Yongjun Signed-off-by: Ilya Dryomov --- net/ceph/osd_client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index b5ec09612ff7..a97e7b506612 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -4220,7 +4220,7 @@ static struct ceph_msg *alloc_msg_with_page_vector(struct ceph_msg_header *hdr) pages = ceph_alloc_page_vector(calc_pages_for(0, data_len), GFP_NOIO); - if (!pages) { + if (IS_ERR(pages)) { ceph_msg_put(m); return NULL; } From f52ec33cbd848632559c87c9305a70fb6eb97f18 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Sat, 30 Jul 2016 00:37:31 +0000 Subject: [PATCH 067/235] libceph: make cancel_generic_request() static Fixes the following sparse warning: net/ceph/mon_client.c:577:6: warning: symbol 'cancel_generic_request' was not declared. Should it be static? Signed-off-by: Wei Yongjun Signed-off-by: Ilya Dryomov --- net/ceph/mon_client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c index c83326c5ba58..ef34a02719d7 100644 --- a/net/ceph/mon_client.c +++ b/net/ceph/mon_client.c @@ -574,7 +574,7 @@ static void complete_generic_request(struct ceph_mon_generic_request *req) put_generic_request(req); } -void cancel_generic_request(struct ceph_mon_generic_request *req) +static void cancel_generic_request(struct ceph_mon_generic_request *req) { struct ceph_mon_client *monc = req->monc; struct ceph_mon_generic_request *lookup_req; From 864364a29c26ed83b3eeca5fa278468dc3ae9ed4 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Sat, 30 Jul 2016 00:38:26 +0000 Subject: [PATCH 068/235] libceph: using kfree_rcu() to simplify the code The callback function of call_rcu() just calls a kfree(), so we can use kfree_rcu() instead of call_rcu() + callback function. Signed-off-by: Wei Yongjun Signed-off-by: Ilya Dryomov --- net/ceph/string_table.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/net/ceph/string_table.c b/net/ceph/string_table.c index ca53c8319209..22fb96efcf34 100644 --- a/net/ceph/string_table.c +++ b/net/ceph/string_table.c @@ -84,12 +84,6 @@ struct ceph_string *ceph_find_or_create_string(const char* str, size_t len) } EXPORT_SYMBOL(ceph_find_or_create_string); -static void ceph_free_string(struct rcu_head *head) -{ - struct ceph_string *cs = container_of(head, struct ceph_string, rcu); - kfree(cs); -} - void ceph_release_string(struct kref *ref) { struct ceph_string *cs = container_of(ref, struct ceph_string, kref); @@ -101,7 +95,7 @@ void ceph_release_string(struct kref *ref) } spin_unlock(&string_tree_lock); - call_rcu(&cs->rcu, ceph_free_string); + kfree_rcu(cs, rcu); } EXPORT_SYMBOL(ceph_release_string); From e4d2b16a445f85b7dc3fc67b21756eca515e0c74 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Thu, 4 Aug 2016 08:43:33 +0800 Subject: [PATCH 069/235] ceph: fix null pointer dereference in ceph_flush_snaps() Signed-off-by: Yan, Zheng --- fs/ceph/caps.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 99115cae1652..16e6ded0b7f2 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1347,9 +1347,12 @@ void ceph_flush_snaps(struct ceph_inode_info *ci, { struct inode *inode = &ci->vfs_inode; struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; - struct ceph_mds_session *session = *psession; + struct ceph_mds_session *session = NULL; int mds; + dout("ceph_flush_snaps %p\n", inode); + if (psession) + session = *psession; retry: spin_lock(&ci->i_ceph_lock); if (!(ci->i_ceph_flags & CEPH_I_FLUSH_SNAPS)) { From 6b6dddbe11b13bb00e0f9a1af2021e266811be85 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Fri, 5 Aug 2016 16:15:38 +0200 Subject: [PATCH 070/235] rbd: destroy header_oloc in rbd_dev_release() Purely cosmetic at this point, as rbd doesn't use RADOS namespaces and hence rbd_dev->header_oloc->pool_ns is always NULL. Signed-off-by: Ilya Dryomov --- drivers/block/rbd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 1a04af6d2421..07668a6f0607 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -3950,6 +3950,7 @@ static void rbd_dev_release(struct device *dev) bool need_put = !!rbd_dev->opts; ceph_oid_destroy(&rbd_dev->header_oid); + ceph_oloc_destroy(&rbd_dev->header_oloc); rbd_put_client(rbd_dev->rbd_client); rbd_spec_put(rbd_dev->spec); From e4630fdd47637168927905983205d7b7c5c08c09 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 8 Aug 2016 15:31:31 +0200 Subject: [PATCH 071/235] x86/power/64: Always create temporary identity mapping correctly The low-level resume-from-hibernation code on x86-64 uses kernel_ident_mapping_init() to create the temoprary identity mapping, but that function assumes that the offset between kernel virtual addresses and physical addresses is aligned on the PGD level. However, with a randomized identity mapping base, it may be aligned on the PUD level and if that happens, the temporary identity mapping created by set_up_temporary_mappings() will not reflect the actual kernel identity mapping and the image restoration will fail as a result (leading to a kernel panic most of the time). To fix this problem, rework kernel_ident_mapping_init() to support unaligned offsets between KVA and PA up to the PMD level and make set_up_temporary_mappings() use it as approprtiate. Reported-and-tested-by: Thomas Garnier Reported-by: Borislav Petkov Suggested-by: Yinghai Lu Signed-off-by: Rafael J. Wysocki Acked-by: Yinghai Lu --- arch/x86/include/asm/init.h | 4 ++-- arch/x86/mm/ident_map.c | 19 +++++++++++-------- arch/x86/power/hibernate_64.c | 2 +- 3 files changed, 14 insertions(+), 11 deletions(-) diff --git a/arch/x86/include/asm/init.h b/arch/x86/include/asm/init.h index 223042086f4e..737da62bfeb0 100644 --- a/arch/x86/include/asm/init.h +++ b/arch/x86/include/asm/init.h @@ -5,10 +5,10 @@ struct x86_mapping_info { void *(*alloc_pgt_page)(void *); /* allocate buf for page table */ void *context; /* context for alloc_pgt_page */ unsigned long pmd_flag; /* page flag for PMD entry */ - bool kernel_mapping; /* kernel mapping or ident mapping */ + unsigned long offset; /* ident mapping offset */ }; int kernel_ident_mapping_init(struct x86_mapping_info *info, pgd_t *pgd_page, - unsigned long addr, unsigned long end); + unsigned long pstart, unsigned long pend); #endif /* _ASM_X86_INIT_H */ diff --git a/arch/x86/mm/ident_map.c b/arch/x86/mm/ident_map.c index ec21796ac5fd..4473cb4f8b90 100644 --- a/arch/x86/mm/ident_map.c +++ b/arch/x86/mm/ident_map.c @@ -3,15 +3,17 @@ * included by both the compressed kernel and the regular kernel. */ -static void ident_pmd_init(unsigned long pmd_flag, pmd_t *pmd_page, +static void ident_pmd_init(struct x86_mapping_info *info, pmd_t *pmd_page, unsigned long addr, unsigned long end) { addr &= PMD_MASK; for (; addr < end; addr += PMD_SIZE) { pmd_t *pmd = pmd_page + pmd_index(addr); - if (!pmd_present(*pmd)) - set_pmd(pmd, __pmd(addr | pmd_flag)); + if (pmd_present(*pmd)) + continue; + + set_pmd(pmd, __pmd((addr - info->offset) | info->pmd_flag)); } } @@ -30,13 +32,13 @@ static int ident_pud_init(struct x86_mapping_info *info, pud_t *pud_page, if (pud_present(*pud)) { pmd = pmd_offset(pud, 0); - ident_pmd_init(info->pmd_flag, pmd, addr, next); + ident_pmd_init(info, pmd, addr, next); continue; } pmd = (pmd_t *)info->alloc_pgt_page(info->context); if (!pmd) return -ENOMEM; - ident_pmd_init(info->pmd_flag, pmd, addr, next); + ident_pmd_init(info, pmd, addr, next); set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE)); } @@ -44,14 +46,15 @@ static int ident_pud_init(struct x86_mapping_info *info, pud_t *pud_page, } int kernel_ident_mapping_init(struct x86_mapping_info *info, pgd_t *pgd_page, - unsigned long addr, unsigned long end) + unsigned long pstart, unsigned long pend) { + unsigned long addr = pstart + info->offset; + unsigned long end = pend + info->offset; unsigned long next; int result; - int off = info->kernel_mapping ? pgd_index(__PAGE_OFFSET) : 0; for (; addr < end; addr = next) { - pgd_t *pgd = pgd_page + pgd_index(addr) + off; + pgd_t *pgd = pgd_page + pgd_index(addr); pud_t *pud; next = (addr & PGDIR_MASK) + PGDIR_SIZE; diff --git a/arch/x86/power/hibernate_64.c b/arch/x86/power/hibernate_64.c index f0b5f2d402af..a3e3ccc87138 100644 --- a/arch/x86/power/hibernate_64.c +++ b/arch/x86/power/hibernate_64.c @@ -87,7 +87,7 @@ static int set_up_temporary_mappings(void) struct x86_mapping_info info = { .alloc_pgt_page = alloc_pgt_page, .pmd_flag = __PAGE_KERNEL_LARGE_EXEC, - .kernel_mapping = true, + .offset = __PAGE_OFFSET, }; unsigned long mstart, mend; pgd_t *pgd; From c8952a707556e04374d7b2fdb3a079d63ddf6f2f Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Mon, 8 Aug 2016 16:16:23 -0600 Subject: [PATCH 072/235] vfio/pci: Fix NULL pointer oops in error interrupt setup handling There are multiple cases in vfio_pci_set_ctx_trigger_single() where we assume we can safely read from our data pointer without actually checking whether the user has passed any data via the count field. VFIO_IRQ_SET_DATA_NONE in particular is entirely broken since we attempt to pull an int32_t file descriptor out before even checking the data type. The other data types assume the data pointer contains one element of their type as well. In part this is good news because we were previously restricted from doing much sanitization of parameters because it was missed in the past and we didn't want to break existing users. Clearly DATA_NONE is completely broken, so it must not have any users and we can fix it up completely. For DATA_BOOL and DATA_EVENTFD, we'll just protect ourselves, returning error when count is zero since we previously would have oopsed. Signed-off-by: Alex Williamson Reported-by: Chris Thompson Cc: stable@vger.kernel.org Reviewed-by: Eric Auger --- drivers/vfio/pci/vfio_pci_intrs.c | 85 ++++++++++++++++++------------- 1 file changed, 49 insertions(+), 36 deletions(-) diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c index 15ecfc9c5f6c..152b43822ef1 100644 --- a/drivers/vfio/pci/vfio_pci_intrs.c +++ b/drivers/vfio/pci/vfio_pci_intrs.c @@ -564,67 +564,80 @@ static int vfio_pci_set_msi_trigger(struct vfio_pci_device *vdev, } static int vfio_pci_set_ctx_trigger_single(struct eventfd_ctx **ctx, - uint32_t flags, void *data) + unsigned int count, uint32_t flags, + void *data) { - int32_t fd = *(int32_t *)data; - - if (!(flags & VFIO_IRQ_SET_DATA_TYPE_MASK)) - return -EINVAL; - /* DATA_NONE/DATA_BOOL enables loopback testing */ if (flags & VFIO_IRQ_SET_DATA_NONE) { - if (*ctx) - eventfd_signal(*ctx, 1); - return 0; + if (*ctx) { + if (count) { + eventfd_signal(*ctx, 1); + } else { + eventfd_ctx_put(*ctx); + *ctx = NULL; + } + return 0; + } } else if (flags & VFIO_IRQ_SET_DATA_BOOL) { - uint8_t trigger = *(uint8_t *)data; + uint8_t trigger; + + if (!count) + return -EINVAL; + + trigger = *(uint8_t *)data; if (trigger && *ctx) eventfd_signal(*ctx, 1); + + return 0; + } else if (flags & VFIO_IRQ_SET_DATA_EVENTFD) { + int32_t fd; + + if (!count) + return -EINVAL; + + fd = *(int32_t *)data; + if (fd == -1) { + if (*ctx) + eventfd_ctx_put(*ctx); + *ctx = NULL; + } else if (fd >= 0) { + struct eventfd_ctx *efdctx; + + efdctx = eventfd_ctx_fdget(fd); + if (IS_ERR(efdctx)) + return PTR_ERR(efdctx); + + if (*ctx) + eventfd_ctx_put(*ctx); + + *ctx = efdctx; + } return 0; } - /* Handle SET_DATA_EVENTFD */ - if (fd == -1) { - if (*ctx) - eventfd_ctx_put(*ctx); - *ctx = NULL; - return 0; - } else if (fd >= 0) { - struct eventfd_ctx *efdctx; - efdctx = eventfd_ctx_fdget(fd); - if (IS_ERR(efdctx)) - return PTR_ERR(efdctx); - if (*ctx) - eventfd_ctx_put(*ctx); - *ctx = efdctx; - return 0; - } else - return -EINVAL; + return -EINVAL; } static int vfio_pci_set_err_trigger(struct vfio_pci_device *vdev, unsigned index, unsigned start, unsigned count, uint32_t flags, void *data) { - if (index != VFIO_PCI_ERR_IRQ_INDEX) + if (index != VFIO_PCI_ERR_IRQ_INDEX || start != 0 || count > 1) return -EINVAL; - /* - * We should sanitize start & count, but that wasn't caught - * originally, so this IRQ index must forever ignore them :-( - */ - - return vfio_pci_set_ctx_trigger_single(&vdev->err_trigger, flags, data); + return vfio_pci_set_ctx_trigger_single(&vdev->err_trigger, + count, flags, data); } static int vfio_pci_set_req_trigger(struct vfio_pci_device *vdev, unsigned index, unsigned start, unsigned count, uint32_t flags, void *data) { - if (index != VFIO_PCI_REQ_IRQ_INDEX || start != 0 || count != 1) + if (index != VFIO_PCI_REQ_IRQ_INDEX || start != 0 || count > 1) return -EINVAL; - return vfio_pci_set_ctx_trigger_single(&vdev->req_trigger, flags, data); + return vfio_pci_set_ctx_trigger_single(&vdev->req_trigger, + count, flags, data); } int vfio_pci_set_irqs_ioctl(struct vfio_pci_device *vdev, uint32_t flags, From e10aec652f31ec61d6a0b4d00d8ef8d2b66fa0fd Mon Sep 17 00:00:00 2001 From: Mario Kleiner Date: Wed, 6 Jul 2016 12:05:44 +0200 Subject: [PATCH 073/235] drm/edid: Add 6 bpc quirk for display AEO model 0. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Bugzilla https://bugzilla.kernel.org/show_bug.cgi?id=105331 reports that the "AEO model 0" display is driven with 8 bpc without dithering by default, which looks bad because that panel is apparently a 6 bpc DP panel with faulty EDID. A fix for this was made by commit 013dd9e03872 ("drm/i915/dp: fall back to 18 bpp when sink capability is unknown"). That commit triggers new regressions in precision for DP->DVI and DP->VGA displays. A patch is out to revert that commit, but it will revert video output for the AEO model 0 panel to 8 bpc without dithering. The EDID 1.3 of that panel, as decoded from the xrandr output attached to that bugzilla bug report, is somewhat faulty, and beyond other problems also sets the "DFP 1.x compliant TMDS" bit, which according to DFP spec means to drive the panel with 8 bpc and no dithering in absence of other colorimetry information. Try to make the original bug reporter happy despite the faulty EDID by adding a quirk to mark that panel as 6 bpc, so 6 bpc output with dithering creates a nice picture. Tested by injecting the edid from the fdo bug into a DP connector via drm_kms_helper.edid_firmware and verifying the 6 bpc + dithering is selected. This patch should be backported to stable. Signed-off-by: Mario Kleiner Cc: stable@vger.kernel.org Cc: Jani Nikula Cc: Ville Syrjälä Cc: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_edid.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 7df26d4b7ad8..2cb472b9976a 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -74,6 +74,8 @@ #define EDID_QUIRK_FORCE_8BPC (1 << 8) /* Force 12bpc */ #define EDID_QUIRK_FORCE_12BPC (1 << 9) +/* Force 6bpc */ +#define EDID_QUIRK_FORCE_6BPC (1 << 10) struct detailed_mode_closure { struct drm_connector *connector; @@ -100,6 +102,9 @@ static struct edid_quirk { /* Unknown Acer */ { "ACR", 2423, EDID_QUIRK_FIRST_DETAILED_PREFERRED }, + /* AEO model 0 reports 8 bpc, but is a 6 bpc panel */ + { "AEO", 0, EDID_QUIRK_FORCE_6BPC }, + /* Belinea 10 15 55 */ { "MAX", 1516, EDID_QUIRK_PREFER_LARGE_60 }, { "MAX", 0x77e, EDID_QUIRK_PREFER_LARGE_60 }, @@ -4082,6 +4087,9 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid) drm_add_display_info(edid, &connector->display_info, connector); + if (quirks & EDID_QUIRK_FORCE_6BPC) + connector->display_info.bpc = 6; + if (quirks & EDID_QUIRK_FORCE_8BPC) connector->display_info.bpc = 8; From 196f954e250943df414efd3d632254c29be38e59 Mon Sep 17 00:00:00 2001 From: Mario Kleiner Date: Wed, 6 Jul 2016 12:05:45 +0200 Subject: [PATCH 074/235] drm/i915/dp: Revert "drm/i915/dp: fall back to 18 bpp when sink capability is unknown" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 013dd9e03872 ("drm/i915/dp: fall back to 18 bpp when sink capability is unknown") This commit introduced a regression into stable kernels, as it reduces output color depth to 6 bpc for any video sink connected to a Displayport connector if that sink doesn't report a specific color depth via EDID, or if our EDID parser doesn't actually recognize the proper bpc from EDID. Affected are active DisplayPort->VGA converters and active DisplayPort->DVI converters. Both should be able to handle 8 bpc, but are degraded to 6 bpc with this patch. The reverted commit was meant to fix Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=105331 A followup patch implements a fix for that specific bug, which is caused by a faulty EDID of the affected DP panel by adding a new EDID quirk for that panel. DP 18 bpp fallback handling and other improvements to DP sink bpc detection will be handled for future kernels in a separate series of patches. Please backport to stable. Signed-off-by: Mario Kleiner Acked-by: Jani Nikula Cc: stable@vger.kernel.org Cc: Ville Syrjälä Cc: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/i915/intel_display.c | 20 +++++--------------- 1 file changed, 5 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index baeb75388dbe..dcf93b3d4fb6 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -12106,21 +12106,11 @@ connected_sink_compute_bpp(struct intel_connector *connector, pipe_config->pipe_bpp = connector->base.display_info.bpc*3; } - /* Clamp bpp to default limit on screens without EDID 1.4 */ - if (connector->base.display_info.bpc == 0) { - int type = connector->base.connector_type; - int clamp_bpp = 24; - - /* Fall back to 18 bpp when DP sink capability is unknown. */ - if (type == DRM_MODE_CONNECTOR_DisplayPort || - type == DRM_MODE_CONNECTOR_eDP) - clamp_bpp = 18; - - if (bpp > clamp_bpp) { - DRM_DEBUG_KMS("clamping display bpp (was %d) to default limit of %d\n", - bpp, clamp_bpp); - pipe_config->pipe_bpp = clamp_bpp; - } + /* Clamp bpp to 8 on screens without EDID 1.4 */ + if (connector->base.display_info.bpc == 0 && bpp > 24) { + DRM_DEBUG_KMS("clamping display bpp (was %d) to default limit of 24\n", + bpp); + pipe_config->pipe_bpp = 24; } } From 210a021dab639694600450c14b877bf3e3240adc Mon Sep 17 00:00:00 2001 From: Mario Kleiner Date: Wed, 6 Jul 2016 12:05:48 +0200 Subject: [PATCH 075/235] drm/edid: Set 8 bpc color depth for displays with "DFP 1.x compliant TMDS". MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit According to E-EDID spec 1.3, table 3.9, a digital video sink with the "DFP 1.x compliant TMDS" bit set is "signal compatible with VESA DFP 1.x TMDS CRGB, 1 pixel / clock, up to 8 bits / color MSB aligned". For such displays, the DFP spec 1.0, section 3.10 "EDID support" says: "If the DFP monitor only supports EDID 1.X (1.1, 1.2, etc.) without extensions, the host will make the following assumptions: 1. 24-bit MSB-aligned RGB TFT 2. DE polarity is active high 3. H and V syncs are active high 4. Established CRT timings will be used 5. Dithering will not be enabled on the host" So if we don't know the bit depth of the display from additional colorimetry info we should assume 8 bpc / 24 bpp by default. This patch adds info->bpc = 8 assignement for that case. Signed-off-by: Mario Kleiner Cc: Jani Nikula Cc: Ville Syrjälä Cc: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_edid.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 2cb472b9976a..637a0aa4d3a0 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -3867,6 +3867,20 @@ static void drm_add_display_info(struct edid *edid, /* HDMI deep color modes supported? Assign to info, if so */ drm_assign_hdmi_deep_color_info(edid, info, connector); + /* + * Digital sink with "DFP 1.x compliant TMDS" according to EDID 1.3? + * + * For such displays, the DFP spec 1.0, section 3.10 "EDID support" + * tells us to assume 8 bpc color depth if the EDID doesn't have + * extensions which tell otherwise. + */ + if ((info->bpc == 0) && (edid->revision < 4) && + (edid->input & DRM_EDID_DIGITAL_TYPE_DVI)) { + info->bpc = 8; + DRM_DEBUG("%s: Assigning DFP sink color depth as %d bpc.\n", + connector->name, info->bpc); + } + /* Only defined for 1.4 with digital displays */ if (edid->revision < 4) return; From d26e94149276f8c3d4911aa2c8395ba99b613c8d Mon Sep 17 00:00:00 2001 From: Emese Revfy Date: Fri, 17 Jun 2016 23:28:34 -0700 Subject: [PATCH 076/235] kbuild: no gcc-plugins during cc-option tests The gcc-plugins arguments should not be included when performing cc-option tests. Steps to reproduce: 1) make mrproper 2) make defconfig 3) enable GCC_PLUGINS, GCC_PLUGIN_CYC_COMPLEXITY 4) enable FUNCTION_TRACER (it will select other options as well) 5) make && make modules Build errors: MODPOST 18 modules ERROR: "__fentry__" [net/netfilter/xt_nat.ko] undefined! ERROR: "__fentry__" [net/netfilter/xt_mark.ko] undefined! ERROR: "__fentry__" [net/netfilter/xt_addrtype.ko] undefined! ERROR: "__fentry__" [net/netfilter/xt_LOG.ko] undefined! ERROR: "__fentry__" [net/netfilter/nf_nat_sip.ko] undefined! ERROR: "__fentry__" [net/netfilter/nf_nat_irc.ko] undefined! ERROR: "__fentry__" [net/netfilter/nf_nat_ftp.ko] undefined! ERROR: "__fentry__" [net/netfilter/nf_nat.ko] undefined! Reported-by: Laura Abbott Signed-off-by: Emese Revfy [kees: renamed variable, clarified commit message] Signed-off-by: Kees Cook --- scripts/Kbuild.include | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index 15b196fc2f49..179219845dfc 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -108,16 +108,20 @@ as-option = $(call try-run,\ as-instr = $(call try-run,\ printf "%b\n" "$(1)" | $(CC) $(KBUILD_AFLAGS) -c -x assembler -o "$$TMP" -,$(2),$(3)) +# Do not attempt to build with gcc plugins during cc-option tests. +# (And this uses delayed resolution so the flags will be up to date.) +CC_OPTION_CFLAGS = $(filter-out $(GCC_PLUGINS_CFLAGS),$(KBUILD_CFLAGS)) + # cc-option # Usage: cflags-y += $(call cc-option,-march=winchip-c6,-march=i586) cc-option = $(call try-run,\ - $(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",$(1),$(2)) + $(CC) $(KBUILD_CPPFLAGS) $(CC_OPTION_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",$(1),$(2)) # cc-option-yn # Usage: flag := $(call cc-option-yn,-march=winchip-c6) cc-option-yn = $(call try-run,\ - $(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",y,n) + $(CC) $(KBUILD_CPPFLAGS) $(CC_OPTION_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",y,n) # cc-option-align # Prefix align with either -falign or -malign @@ -127,7 +131,7 @@ cc-option-align = $(subst -functions=0,,\ # cc-disable-warning # Usage: cflags-y += $(call cc-disable-warning,unused-but-set-variable) cc-disable-warning = $(call try-run,\ - $(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1))) + $(CC) $(KBUILD_CPPFLAGS) $(CC_OPTION_CFLAGS) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1))) # cc-name # Expands to either gcc or clang From ed58c0e9eefef517aa5a547b78658e2ab4422232 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 17 Jun 2016 23:11:12 -0700 Subject: [PATCH 077/235] gcc-plugins: abort builds cleanly when not supported When the compiler doesn't support gcc plugins (either due to missing headers or too old a version), report the problem and abort the build instead of emitting a warning and letting the build founder with arcane compiler errors. Signed-off-by: Kees Cook --- Makefile | 7 ------- scripts/Makefile.gcc-plugins | 38 +++++++++++++++++++++++++----------- scripts/gcc-plugin.sh | 14 +++++++++++++ 3 files changed, 41 insertions(+), 18 deletions(-) diff --git a/Makefile b/Makefile index 70de1448c571..8c504f324154 100644 --- a/Makefile +++ b/Makefile @@ -635,13 +635,6 @@ endif # Tell gcc to never replace conditional load with a non-conditional one KBUILD_CFLAGS += $(call cc-option,--param=allow-store-data-races=0) -PHONY += gcc-plugins -gcc-plugins: scripts_basic -ifdef CONFIG_GCC_PLUGINS - $(Q)$(MAKE) $(build)=scripts/gcc-plugins -endif - @: - include scripts/Makefile.gcc-plugins ifdef CONFIG_READABLE_ASM diff --git a/scripts/Makefile.gcc-plugins b/scripts/Makefile.gcc-plugins index 5e22b60589c1..bbca1f46f6e4 100644 --- a/scripts/Makefile.gcc-plugins +++ b/scripts/Makefile.gcc-plugins @@ -23,21 +23,37 @@ ifdef CONFIG_GCC_PLUGINS export PLUGINCC GCC_PLUGINS_CFLAGS GCC_PLUGIN SANCOV_PLUGIN - ifeq ($(PLUGINCC),) - ifneq ($(GCC_PLUGINS_CFLAGS),) - ifeq ($(call cc-ifversion, -ge, 0405, y), y) - PLUGINCC := $(shell $(CONFIG_SHELL) -x $(srctree)/scripts/gcc-plugin.sh "$(__PLUGINCC)" "$(HOSTCXX)" "$(CC)") - $(warning warning: your gcc installation does not support plugins, perhaps the necessary headers are missing?) - else - $(warning warning: your gcc version does not support plugins, you should upgrade it to gcc 4.5 at least) - endif - endif - else + ifneq ($(PLUGINCC),) # SANCOV_PLUGIN can be only in CFLAGS_KCOV because avoid duplication. GCC_PLUGINS_CFLAGS := $(filter-out $(SANCOV_PLUGIN), $(GCC_PLUGINS_CFLAGS)) endif KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS) GCC_PLUGIN := $(gcc-plugin-y) - endif + +# If plugins aren't supported, abort the build before hard-to-read compiler +# errors start getting spewed by the main build. +PHONY += gcc-plugins-check +gcc-plugins-check: FORCE +ifdef CONFIG_GCC_PLUGINS + ifeq ($(PLUGINCC),) + ifneq ($(GCC_PLUGINS_CFLAGS),) + ifeq ($(call cc-ifversion, -ge, 0405, y), y) + $(Q)$(srctree)/scripts/gcc-plugin.sh --show-error "$(__PLUGINCC)" "$(HOSTCXX)" "$(CC)" || true + @echo "Cannot use CONFIG_GCC_PLUGINS: your gcc installation does not support plugins, perhaps the necessary headers are missing?" >&2 && exit 1 + else + @echo "Cannot use CONFIG_GCC_PLUGINS: your gcc version does not support plugins, you should upgrade it to at least gcc 4.5" >&2 && exit 1 + endif + endif + endif +endif + @: + +# Actually do the build, if requested. +PHONY += gcc-plugins +gcc-plugins: scripts_basic gcc-plugins-check +ifdef CONFIG_GCC_PLUGINS + $(Q)$(MAKE) $(build)=scripts/gcc-plugins +endif + @: diff --git a/scripts/gcc-plugin.sh b/scripts/gcc-plugin.sh index fb9207565471..b65224bfb847 100755 --- a/scripts/gcc-plugin.sh +++ b/scripts/gcc-plugin.sh @@ -1,5 +1,12 @@ #!/bin/sh srctree=$(dirname "$0") + +SHOW_ERROR= +if [ "$1" = "--show-error" ] ; then + SHOW_ERROR=1 + shift || true +fi + gccplugins_dir=$($3 -print-file-name=plugin) plugincc=$($1 -E -x c++ - -o /dev/null -I"${srctree}"/gcc-plugins -I"${gccplugins_dir}"/include 2>&1 <&2 + fi exit 1 fi @@ -48,4 +58,8 @@ then echo "$2" exit 0 fi + +if [ -n "$SHOW_ERROR" ] ; then + echo "${plugincc}" >&2 +fi exit 1 From 65d59ec8ad99e9d932ccca4daabb8e3fdafeb11b Mon Sep 17 00:00:00 2001 From: Emese Revfy Date: Mon, 20 Jun 2016 20:40:24 +0200 Subject: [PATCH 078/235] gcc-plugins: Add support for passing plugin arguments The latent_entropy plugin needs to pass arguments, so this adds the support. Signed-off-by: Emese Revfy Signed-off-by: Kees Cook --- scripts/Makefile.gcc-plugins | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/Makefile.gcc-plugins b/scripts/Makefile.gcc-plugins index bbca1f46f6e4..ffb928b9c74b 100644 --- a/scripts/Makefile.gcc-plugins +++ b/scripts/Makefile.gcc-plugins @@ -19,7 +19,7 @@ ifdef CONFIG_GCC_PLUGINS endif endif - GCC_PLUGINS_CFLAGS := $(addprefix -fplugin=$(objtree)/scripts/gcc-plugins/, $(gcc-plugin-y)) + GCC_PLUGINS_CFLAGS := $(strip $(addprefix -fplugin=$(objtree)/scripts/gcc-plugins/, $(gcc-plugin-y)) $(gcc-plugin-cflags-y)) export PLUGINCC GCC_PLUGINS_CFLAGS GCC_PLUGIN SANCOV_PLUGIN From 7040c83bfbaf2340d2f336dc7641ce909c8c2b4c Mon Sep 17 00:00:00 2001 From: Emese Revfy Date: Sun, 26 Jun 2016 17:36:43 +0200 Subject: [PATCH 079/235] gcc-plugins: Automate make rule generation There's no reason to repeat the same names in the Makefile when the .so files have already been listed. The .o list can be generated from them. Reported-by: PaX Team Signed-off-by: Emese Revfy [kees: clarified commit message] Signed-off-by: Kees Cook --- scripts/gcc-plugins/Makefile | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/scripts/gcc-plugins/Makefile b/scripts/gcc-plugins/Makefile index 88c8ec47232b..ee0a75d2a1c7 100644 --- a/scripts/gcc-plugins/Makefile +++ b/scripts/gcc-plugins/Makefile @@ -21,7 +21,6 @@ endif $(HOSTLIBS)-y := $(GCC_PLUGIN) always := $($(HOSTLIBS)-y) -cyc_complexity_plugin-objs := cyc_complexity_plugin.o -sancov_plugin-objs := sancov_plugin.o +$(foreach p,$($(HOSTLIBS)-y:%.so=%),$(eval $(p)-objs := $(p).o)) clean-files += *.so From caefd8c9a9fb06811e07bf3571e5d4450846b16a Mon Sep 17 00:00:00 2001 From: Emese Revfy Date: Sun, 26 Jun 2016 17:38:20 +0200 Subject: [PATCH 080/235] gcc-plugins: Add support for plugin subdirectories This adds support for building more complex gcc plugins that live in a subdirectory instead of just in a single source file. Reported-by: PaX Team Signed-off-by: Emese Revfy [kees: clarified commit message] Signed-off-by: Kees Cook --- scripts/Makefile.gcc-plugins | 3 ++- scripts/gcc-plugins/Makefile | 9 ++++++--- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/scripts/Makefile.gcc-plugins b/scripts/Makefile.gcc-plugins index ffb928b9c74b..61f0e6db909b 100644 --- a/scripts/Makefile.gcc-plugins +++ b/scripts/Makefile.gcc-plugins @@ -21,7 +21,7 @@ ifdef CONFIG_GCC_PLUGINS GCC_PLUGINS_CFLAGS := $(strip $(addprefix -fplugin=$(objtree)/scripts/gcc-plugins/, $(gcc-plugin-y)) $(gcc-plugin-cflags-y)) - export PLUGINCC GCC_PLUGINS_CFLAGS GCC_PLUGIN SANCOV_PLUGIN + export PLUGINCC GCC_PLUGINS_CFLAGS GCC_PLUGIN GCC_PLUGIN_SUBDIR SANCOV_PLUGIN ifneq ($(PLUGINCC),) # SANCOV_PLUGIN can be only in CFLAGS_KCOV because avoid duplication. @@ -30,6 +30,7 @@ ifdef CONFIG_GCC_PLUGINS KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS) GCC_PLUGIN := $(gcc-plugin-y) + GCC_PLUGIN_SUBDIR := $(gcc-plugin-subdir-y) endif # If plugins aren't supported, abort the build before hard-to-read compiler diff --git a/scripts/gcc-plugins/Makefile b/scripts/gcc-plugins/Makefile index ee0a75d2a1c7..8b29dc17c73c 100644 --- a/scripts/gcc-plugins/Makefile +++ b/scripts/gcc-plugins/Makefile @@ -12,15 +12,18 @@ else export HOST_EXTRACXXFLAGS endif -export GCCPLUGINS_DIR HOSTLIBS - ifneq ($(CFLAGS_KCOV), $(SANCOV_PLUGIN)) GCC_PLUGIN := $(filter-out $(SANCOV_PLUGIN), $(GCC_PLUGIN)) endif -$(HOSTLIBS)-y := $(GCC_PLUGIN) +export HOSTLIBS + +$(HOSTLIBS)-y := $(foreach p,$(GCC_PLUGIN),$(if $(findstring /,$(p)),,$(p))) always := $($(HOSTLIBS)-y) $(foreach p,$($(HOSTLIBS)-y:%.so=%),$(eval $(p)-objs := $(p).o)) +subdir-y := $(GCC_PLUGIN_SUBDIR) +subdir- += $(GCC_PLUGIN_SUBDIR) + clean-files += *.so From 36e9d08b58f44c3a02974c405ccaaa6ecfaf05b8 Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Tue, 9 Aug 2016 02:16:18 +0200 Subject: [PATCH 081/235] drm/cirrus: Fix NULL pointer dereference when registering the fbdev cirrus_modeset_init() is initializing/registering the emulated fbdev and, since commit c61b93fe51b1 ("drm/atomic: Fix remaining places where !funcs->best_encoder is valid"), DRM internals can access/test some of the fields in mode_config->funcs as part of the fbdev registration process. Make sure dev->mode_config.funcs is properly set to avoid dereferencing a NULL pointer. Reported-by: Mike Marshall Reported-by: Eric W. Biederman Signed-off-by: Boris Brezillon Fixes: c61b93fe51b1 ("drm/atomic: Fix remaining places where !funcs->best_encoder is valid") Signed-off-by: Dave Airlie --- drivers/gpu/drm/cirrus/cirrus_main.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c index 80446e2d3ab6..76bcb43e7c06 100644 --- a/drivers/gpu/drm/cirrus/cirrus_main.c +++ b/drivers/gpu/drm/cirrus/cirrus_main.c @@ -185,14 +185,23 @@ int cirrus_driver_load(struct drm_device *dev, unsigned long flags) goto out; } + /* + * cirrus_modeset_init() is initializing/registering the emulated fbdev + * and DRM internals can access/test some of the fields in + * mode_config->funcs as part of the fbdev registration process. + * Make sure dev->mode_config.funcs is properly set to avoid + * dereferencing a NULL pointer. + * FIXME: mode_config.funcs assignment should probably be done in + * cirrus_modeset_init() (that's a common pattern seen in other DRM + * drivers). + */ + dev->mode_config.funcs = &cirrus_mode_funcs; r = cirrus_modeset_init(cdev); if (r) { dev_err(&dev->pdev->dev, "Fatal error during modeset init: %d\n", r); goto out; } - dev->mode_config.funcs = (void *)&cirrus_mode_funcs; - return 0; out: cirrus_driver_unload(dev); From d2cf5be07ff7c7cde8bef8551a30e8e86e2037a7 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 4 Aug 2016 16:38:15 +1000 Subject: [PATCH 082/235] crypto: crc32c-vpmsum - Convert to CPU feature based module autoloading This patch utilises the GENERIC_CPU_AUTOPROBE infrastructure to automatically load the crc32c-vpmsum module if the CPU supports it. Signed-off-by: Anton Blanchard Signed-off-by: Michael Ellerman --- arch/powerpc/crypto/crc32c-vpmsum_glue.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/crypto/crc32c-vpmsum_glue.c b/arch/powerpc/crypto/crc32c-vpmsum_glue.c index bfe3d37a24ef..9fa046d56eba 100644 --- a/arch/powerpc/crypto/crc32c-vpmsum_glue.c +++ b/arch/powerpc/crypto/crc32c-vpmsum_glue.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #define CHKSUM_BLOCK_SIZE 1 @@ -157,7 +158,7 @@ static void __exit crc32c_vpmsum_mod_fini(void) crypto_unregister_shash(&alg); } -module_init(crc32c_vpmsum_mod_init); +module_cpu_feature_match(PPC_MODULE_FEATURE_VEC_CRYPTO, crc32c_vpmsum_mod_init); module_exit(crc32c_vpmsum_mod_fini); MODULE_AUTHOR("Anton Blanchard "); From 880a3d6afd068682d6386a0528be1217541d3d8e Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 2 Aug 2016 12:39:43 +1000 Subject: [PATCH 083/235] powerpc/xics: Properly set Edge/Level type and enable resend This sets the type of the interrupt appropriately. We set it as follow: - If not mapped from the device-tree, we use edge. This is the case of the virtual interrupts and PCI MSIs for example. - If mapped from the device-tree and #interrupt-cells is 2 (PAPR compliant), we use the second cell to set the appropriate type - If mapped from the device-tree and #interrupt-cells is 1 (current OPAL on P8 does that), we assume level sensitive since those are typically going to be the PSI LSIs which are level sensitive. Additionally, we mark the interrupts requested via the opal_interrupts property all level. This is a bit fishy but the best we can do until we fix OPAL to properly expose them with a complete descriptor. It is also correct for the current HW anyway as OPAL interrupts are currently PCI error and PSI interrupts which are level. Finally now that edge interrupts are properly identified, we can enable CONFIG_HARDIRQS_SW_RESEND which will make the core re-send them if they occur while masked, which some drivers rely upon. This fixes issues with lost interrupts on some Mellanox adapters. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/xics.h | 2 + arch/powerpc/platforms/powernv/opal-irqchip.c | 3 +- arch/powerpc/sysdev/xics/Kconfig | 1 + arch/powerpc/sysdev/xics/ics-opal.c | 4 +- arch/powerpc/sysdev/xics/ics-rtas.c | 4 +- arch/powerpc/sysdev/xics/xics-common.c | 59 ++++++++++++++++--- 6 files changed, 63 insertions(+), 10 deletions(-) diff --git a/arch/powerpc/include/asm/xics.h b/arch/powerpc/include/asm/xics.h index f5f729c11578..f0b238516e9b 100644 --- a/arch/powerpc/include/asm/xics.h +++ b/arch/powerpc/include/asm/xics.h @@ -159,6 +159,8 @@ extern void xics_teardown_cpu(void); extern void xics_kexec_teardown_cpu(int secondary); extern void xics_migrate_irqs_away(void); extern void icp_native_eoi(struct irq_data *d); +extern int xics_set_irq_type(struct irq_data *d, unsigned int flow_type); +extern int xics_retrigger(struct irq_data *data); #ifdef CONFIG_SMP extern int xics_get_irq_server(unsigned int virq, const struct cpumask *cpumask, unsigned int strict_check); diff --git a/arch/powerpc/platforms/powernv/opal-irqchip.c b/arch/powerpc/platforms/powernv/opal-irqchip.c index e505223b4ec5..ed8bba68a162 100644 --- a/arch/powerpc/platforms/powernv/opal-irqchip.c +++ b/arch/powerpc/platforms/powernv/opal-irqchip.c @@ -228,7 +228,8 @@ int __init opal_event_init(void) } /* Install interrupt handler */ - rc = request_irq(virq, opal_interrupt, 0, "opal", NULL); + rc = request_irq(virq, opal_interrupt, IRQF_TRIGGER_LOW, + "opal", NULL); if (rc) { irq_dispose_mapping(virq); pr_warn("Error %d requesting irq %d (0x%x)\n", diff --git a/arch/powerpc/sysdev/xics/Kconfig b/arch/powerpc/sysdev/xics/Kconfig index 0031eda320c3..385e7aa9e273 100644 --- a/arch/powerpc/sysdev/xics/Kconfig +++ b/arch/powerpc/sysdev/xics/Kconfig @@ -1,6 +1,7 @@ config PPC_XICS def_bool n select PPC_SMP_MUXED_IPI + select HARDIRQS_SW_RESEND config PPC_ICP_NATIVE def_bool n diff --git a/arch/powerpc/sysdev/xics/ics-opal.c b/arch/powerpc/sysdev/xics/ics-opal.c index 27c936c080a6..1c6bf4b66f56 100644 --- a/arch/powerpc/sysdev/xics/ics-opal.c +++ b/arch/powerpc/sysdev/xics/ics-opal.c @@ -156,7 +156,9 @@ static struct irq_chip ics_opal_irq_chip = { .irq_mask = ics_opal_mask_irq, .irq_unmask = ics_opal_unmask_irq, .irq_eoi = NULL, /* Patched at init time */ - .irq_set_affinity = ics_opal_set_affinity + .irq_set_affinity = ics_opal_set_affinity, + .irq_set_type = xics_set_irq_type, + .irq_retrigger = xics_retrigger, }; static int ics_opal_map(struct ics *ics, unsigned int virq); diff --git a/arch/powerpc/sysdev/xics/ics-rtas.c b/arch/powerpc/sysdev/xics/ics-rtas.c index 3854dd41558d..78ee5c778ef8 100644 --- a/arch/powerpc/sysdev/xics/ics-rtas.c +++ b/arch/powerpc/sysdev/xics/ics-rtas.c @@ -163,7 +163,9 @@ static struct irq_chip ics_rtas_irq_chip = { .irq_mask = ics_rtas_mask_irq, .irq_unmask = ics_rtas_unmask_irq, .irq_eoi = NULL, /* Patched at init time */ - .irq_set_affinity = ics_rtas_set_affinity + .irq_set_affinity = ics_rtas_set_affinity, + .irq_set_type = xics_set_irq_type, + .irq_retrigger = xics_retrigger, }; static int ics_rtas_map(struct ics *ics, unsigned int virq) diff --git a/arch/powerpc/sysdev/xics/xics-common.c b/arch/powerpc/sysdev/xics/xics-common.c index a795a5f0301c..9d530f479588 100644 --- a/arch/powerpc/sysdev/xics/xics-common.c +++ b/arch/powerpc/sysdev/xics/xics-common.c @@ -328,8 +328,12 @@ static int xics_host_map(struct irq_domain *h, unsigned int virq, pr_devel("xics: map virq %d, hwirq 0x%lx\n", virq, hw); - /* They aren't all level sensitive but we just don't really know */ - irq_set_status_flags(virq, IRQ_LEVEL); + /* + * Mark interrupts as edge sensitive by default so that resend + * actually works. The device-tree parsing will turn the LSIs + * back to level. + */ + irq_clear_status_flags(virq, IRQ_LEVEL); /* Don't call into ICS for IPIs */ if (hw == XICS_IPI) { @@ -351,13 +355,54 @@ static int xics_host_xlate(struct irq_domain *h, struct device_node *ct, irq_hw_number_t *out_hwirq, unsigned int *out_flags) { - /* Current xics implementation translates everything - * to level. It is not technically right for MSIs but this - * is irrelevant at this point. We might get smarter in the future - */ *out_hwirq = intspec[0]; - *out_flags = IRQ_TYPE_LEVEL_LOW; + /* + * If intsize is at least 2, we look for the type in the second cell, + * we assume the LSB indicates a level interrupt. + */ + if (intsize > 1) { + if (intspec[1] & 1) + *out_flags = IRQ_TYPE_LEVEL_LOW; + else + *out_flags = IRQ_TYPE_EDGE_RISING; + } else + *out_flags = IRQ_TYPE_LEVEL_LOW; + + return 0; +} + +int xics_set_irq_type(struct irq_data *d, unsigned int flow_type) +{ + /* + * We only support these. This has really no effect other than setting + * the corresponding descriptor bits mind you but those will in turn + * affect the resend function when re-enabling an edge interrupt. + * + * Set set the default to edge as explained in map(). + */ + if (flow_type == IRQ_TYPE_DEFAULT || flow_type == IRQ_TYPE_NONE) + flow_type = IRQ_TYPE_EDGE_RISING; + + if (flow_type != IRQ_TYPE_EDGE_RISING && + flow_type != IRQ_TYPE_LEVEL_LOW) + return -EINVAL; + + irqd_set_trigger_type(d, flow_type); + + return IRQ_SET_MASK_OK_NOCOPY; +} + +int xics_retrigger(struct irq_data *data) +{ + /* + * We need to push a dummy CPPR when retriggering, since the subsequent + * EOI will try to pop it. Passing 0 works, as the function hard codes + * the priority value anyway. + */ + xics_push_cppr(0); + + /* Tell the core to do a soft retrigger */ return 0; } From 54a94fcf2f00c9deb9491d5ab2b5a4022332f47e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 4 Aug 2016 08:37:03 +0300 Subject: [PATCH 084/235] powerpc/cell: Add missing error code in spufs_mkgang() We should return -ENOMEM if alloc_spu_gang() fails. Signed-off-by: Dan Carpenter Acked-by: Arnd Bergmann Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/cell/spufs/inode.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 5be15cff758d..2975754c65ea 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -496,8 +496,10 @@ spufs_mkgang(struct inode *dir, struct dentry *dentry, umode_t mode) gang = alloc_spu_gang(); SPUFS_I(inode)->i_ctx = NULL; SPUFS_I(inode)->i_gang = gang; - if (!gang) + if (!gang) { + ret = -ENOMEM; goto out_iput; + } inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; From 4d9021957b5218310e28767f25551ca8f5eac797 Mon Sep 17 00:00:00 2001 From: Alexey Kardashevskiy Date: Wed, 3 Aug 2016 18:40:45 +1000 Subject: [PATCH 085/235] powerpc/powernv/ioda: Fix TCE invalidate to work in real mode again Commit fd141d1a99a3 ("powerpc/powernv/pci: Rework accessing the TCE invalidate register") broke TCE invalidation on IODA2/PHB3 for real mode. This makes invalidate work again. Fixes: fd141d1a99a3 ("powerpc/powernv/pci: Rework accessing the TCE invalidate register") Signed-off-by: Alexey Kardashevskiy Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/powernv/pci-ioda.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 6b9528307f62..dedbbe9785ba 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -1877,7 +1877,7 @@ static void pnv_pci_phb3_tce_invalidate(struct pnv_ioda_pe *pe, bool rm, unsigned shift, unsigned long index, unsigned long npages) { - __be64 __iomem *invalidate = pnv_ioda_get_inval_reg(pe->phb, false); + __be64 __iomem *invalidate = pnv_ioda_get_inval_reg(pe->phb, rm); unsigned long start, end, inc; /* We'll invalidate DMA address in PE scope */ From e325d76f8bd2d222a1f577aba00dfb43cece4cbc Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Fri, 5 Aug 2016 19:13:12 +0530 Subject: [PATCH 086/235] powerpc/powernv: Load correct TOC pointer while waking up from winkle. The function pnv_restore_hyp_resource() loads the TOC into r2 from the invalid PACA pointer before fixing r13 value. This do not affect POWER ISA 3.0 but it does have an impact on POWER ISA 2.07 or less leading CPU to get stuck forever. login: [ 471.830433] Processor 120 is stuck. This can be easily reproducible using following steps: - Turn off SMT $ ppc64_cpu --smt=off - offline/online any online cpu (Thread 0 of any core which is online) $ echo 0 > /sys/devices/system/cpu/cpu/online $ echo 1 > /sys/devices/system/cpu/cpu/online For POWER ISA 2.07 or less, the last bit of HSPRG0 is set indicating that thread is waking up from winkle. Hence, the last bit of HSPRG0(r13) needs to be clear before accessing it as PACA to avoid loading invalid values from invalid PACA pointer. Fix this by loading TOC after r13 register is corrected. Fixes: bcef83a00dc4 ("powerpc/powernv: Add platform support for stop instruction") Signed-off-by: Mahesh Salgaonkar Acked-by: Vaidyanathan Srinivasan Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/idle_book3s.S | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S index ba79d15f4ddd..07a330ed7022 100644 --- a/arch/powerpc/kernel/idle_book3s.S +++ b/arch/powerpc/kernel/idle_book3s.S @@ -363,8 +363,8 @@ _GLOBAL(power9_idle_stop) * cr3 - set to gt if waking up with partial/complete hypervisor state loss */ _GLOBAL(pnv_restore_hyp_resource) - ld r2,PACATOC(r13); BEGIN_FTR_SECTION + ld r2,PACATOC(r13); /* * POWER ISA 3. Use PSSCR to determine if we * are waking up from deep idle state @@ -395,6 +395,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300) */ clrldi r5,r13,63 clrrdi r13,r13,1 + + /* Now that we are sure r13 is corrected, load TOC */ + ld r2,PACATOC(r13); cmpwi cr4,r5,1 mtspr SPRN_HSPRG0,r13 From 98d8821a47f3fd7354d3ab87adb50b10c9adb937 Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Fri, 5 Aug 2016 17:34:04 +0530 Subject: [PATCH 087/235] powerpc/powernv: Move IDLE_STATE_ENTER_SEQ macro to cpuidle.h Move IDLE_STATE_ENTER_SEQ macro to cpuidle.h so that MCE handler changes in subsequent patch can use it. No functionality change. Signed-off-by: Mahesh Salgaonkar Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/cpuidle.h | 13 +++++++++++++ arch/powerpc/kernel/idle_book3s.S | 12 ------------ 2 files changed, 13 insertions(+), 12 deletions(-) diff --git a/arch/powerpc/include/asm/cpuidle.h b/arch/powerpc/include/asm/cpuidle.h index 3d7fc06532a1..01b8a13f0224 100644 --- a/arch/powerpc/include/asm/cpuidle.h +++ b/arch/powerpc/include/asm/cpuidle.h @@ -19,4 +19,17 @@ extern u64 pnv_first_deep_stop_state; #endif +/* Idle state entry routines */ +#ifdef CONFIG_PPC_P7_NAP +#define IDLE_STATE_ENTER_SEQ(IDLE_INST) \ + /* Magic NAP/SLEEP/WINKLE mode enter sequence */ \ + std r0,0(r1); \ + ptesync; \ + ld r0,0(r1); \ +1: cmp cr0,r0,r0; \ + bne 1b; \ + IDLE_INST; \ + b . +#endif /* CONFIG_PPC_P7_NAP */ + #endif diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S index 07a330ed7022..2265c6398a17 100644 --- a/arch/powerpc/kernel/idle_book3s.S +++ b/arch/powerpc/kernel/idle_book3s.S @@ -44,18 +44,6 @@ PSSCR_PSLL_MASK | PSSCR_TR_MASK | \ PSSCR_MTL_MASK -/* Idle state entry routines */ - -#define IDLE_STATE_ENTER_SEQ(IDLE_INST) \ - /* Magic NAP/SLEEP/WINKLE mode enter sequence */ \ - std r0,0(r1); \ - ptesync; \ - ld r0,0(r1); \ -1: cmp cr0,r0,r0; \ - bne 1b; \ - IDLE_INST; \ - b . - .text /* From 7627e40c66b5547e12b6c5673646ceea84797a74 Mon Sep 17 00:00:00 2001 From: Piotr Karasinski Date: Sat, 6 Aug 2016 21:23:05 +0200 Subject: [PATCH 088/235] ALSA: usb-audio: Add a sample rate quirk for Creative Live! Cam Socialize HD (VF0610) VF0610 does not support reading the sample rate which leads to many lines of "cannot get freq at ep 0x82". This patch adds the USB ID (0x041E:4080) to snd_usb_get_sample_rate_quirk() list. Signed-off-by: Piotr Karasinski Cc: Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 6adde457b602..b70d4dc4ef10 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1128,6 +1128,7 @@ bool snd_usb_get_sample_rate_quirk(struct snd_usb_audio *chip) { /* devices which do not support reading the sample rate. */ switch (chip->usb_id) { + case USB_ID(0x041E, 0x4080): /* Creative Live Cam VF0610 */ case USB_ID(0x045E, 0x075D): /* MS Lifecam Cinema */ case USB_ID(0x045E, 0x076D): /* MS Lifecam HD-5000 */ case USB_ID(0x045E, 0x076E): /* MS Lifecam HD-5001 */ From 41f5e3bdbf706a9e98194bf0c4b62a875c02f170 Mon Sep 17 00:00:00 2001 From: "Vittorio Gambaletta (VittGam)" Date: Mon, 8 Aug 2016 12:35:40 +0200 Subject: [PATCH 089/235] ALSA: usb-audio: Add quirk for ELP HD USB Camera The ELP HD USB Camera (05a3:9420) needs this quirk for suppressing the unsupported sample rate inquiry. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=98481 Cc: Signed-off-by: Vittorio Gambaletta Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index b70d4dc4ef10..6cf1f3597455 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1139,6 +1139,7 @@ bool snd_usb_get_sample_rate_quirk(struct snd_usb_audio *chip) case USB_ID(0x047F, 0xAA05): /* Plantronics DA45 */ case USB_ID(0x04D8, 0xFEEA): /* Benchmark DAC1 Pre */ case USB_ID(0x0556, 0x0014): /* Phoenix Audio TMX320VC */ + case USB_ID(0x05A3, 0x9420): /* ELP HD USB Camera */ case USB_ID(0x074D, 0x3553): /* Outlaw RR2150 (Micronas UAC3553B) */ case USB_ID(0x1de7, 0x0013): /* Phoenix Audio MT202exe */ case USB_ID(0x1de7, 0x0014): /* Phoenix Audio TMX320 */ From bc14c49195e49b3231c01e4c44e3e5456c940b94 Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Fri, 5 Aug 2016 17:34:13 +0530 Subject: [PATCH 090/235] powerpc/powernv: Fix MCE handler to avoid trashing CR0/CR1 registers. The current implementation of MCE early handling modifies CR0/1 registers without saving its old values. Fix this by moving early check for powersaving mode to machine_check_handle_early(). The power architecture 2.06 or later allows the possibility of getting machine check while in nap/sleep/winkle. The last bit of HSPRG0 is set to 1, if thread is woken up from winkle. Hence, clear the last bit of HSPRG0 (r13) before MCE handler starts using it as paca pointer. Also, the current code always puts the thread into nap state irrespective of whatever idle state it woke up from. Fix that by looking at paca->thread_idle_state and put the thread back into same state where it came from. Fixes: 1c51089f777b ("powerpc/book3s: Return from interrupt if coming from evil context.") Reported-by: Paul Mackerras Signed-off-by: Mahesh Salgaonkar Reviewed-by: Shreyas B. Prabhu Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/exceptions-64s.S | 69 ++++++++++++++++------------ 1 file changed, 40 insertions(+), 29 deletions(-) diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 41091fdf9bd8..df6d45eb4115 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -144,29 +144,14 @@ machine_check_pSeries_1: * vector */ SET_SCRATCH0(r13) /* save r13 */ -#ifdef CONFIG_PPC_P7_NAP -BEGIN_FTR_SECTION - /* Running native on arch 2.06 or later, check if we are - * waking up from nap. We only handle no state loss and - * supervisor state loss. We do -not- handle hypervisor - * state loss at this time. + /* + * Running native on arch 2.06 or later, we may wakeup from winkle + * inside machine check. If yes, then last bit of HSPGR0 would be set + * to 1. Hence clear it unconditionally. */ - mfspr r13,SPRN_SRR1 - rlwinm. r13,r13,47-31,30,31 - OPT_GET_SPR(r13, SPRN_CFAR, CPU_FTR_CFAR) - beq 9f - - mfspr r13,SPRN_SRR1 - rlwinm. r13,r13,47-31,30,31 - /* waking up from powersave (nap) state */ - cmpwi cr1,r13,2 - /* Total loss of HV state is fatal. let's just stay stuck here */ - OPT_GET_SPR(r13, SPRN_CFAR, CPU_FTR_CFAR) - bgt cr1,. -9: - OPT_SET_SPR(r13, SPRN_CFAR, CPU_FTR_CFAR) -END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) -#endif /* CONFIG_PPC_P7_NAP */ + GET_PACA(r13) + clrrdi r13,r13,1 + SET_PACA(r13) EXCEPTION_PROLOG_0(PACA_EXMC) BEGIN_FTR_SECTION b machine_check_powernv_early @@ -1273,25 +1258,51 @@ machine_check_handle_early: * Check if thread was in power saving mode. We come here when any * of the following is true: * a. thread wasn't in power saving mode - * b. thread was in power saving mode with no state loss or - * supervisor state loss + * b. thread was in power saving mode with no state loss, + * supervisor state loss or hypervisor state loss. * - * Go back to nap again if (b) is true. + * Go back to nap/sleep/winkle mode again if (b) is true. */ rlwinm. r11,r12,47-31,30,31 /* Was it in power saving mode? */ beq 4f /* No, it wasn;t */ /* Thread was in power saving mode. Go back to nap again. */ cmpwi r11,2 - bne 3f - /* Supervisor state loss */ + blt 3f + /* Supervisor/Hypervisor state loss */ li r0,1 stb r0,PACA_NAPSTATELOST(r13) 3: bl machine_check_queue_event MACHINE_CHECK_HANDLER_WINDUP GET_PACA(r13) ld r1,PACAR1(r13) - li r3,PNV_THREAD_NAP - b pnv_enter_arch207_idle_mode + /* + * Check what idle state this CPU was in and go back to same mode + * again. + */ + lbz r3,PACA_THREAD_IDLE_STATE(r13) + cmpwi r3,PNV_THREAD_NAP + bgt 10f + IDLE_STATE_ENTER_SEQ(PPC_NAP) + /* No return */ +10: + cmpwi r3,PNV_THREAD_SLEEP + bgt 2f + IDLE_STATE_ENTER_SEQ(PPC_SLEEP) + /* No return */ + +2: + /* + * Go back to winkle. Please note that this thread was woken up in + * machine check from winkle and have not restored the per-subcore + * state. Hence before going back to winkle, set last bit of HSPGR0 + * to 1. This will make sure that if this thread gets woken up + * again at reset vector 0x100 then it will get chance to restore + * the subcore state. + */ + ori r13,r13,1 + SET_PACA(r13) + IDLE_STATE_ENTER_SEQ(PPC_WINKLE) + /* No return */ 4: #endif /* From 9dc512819e4b723856773f5a018865c20441072f Mon Sep 17 00:00:00 2001 From: Alastair D'Silva Date: Mon, 1 Aug 2016 16:32:51 +1000 Subject: [PATCH 091/235] powerpc: Fix unused function warning 'lmb_to_memblock' This patch fixes the following warning: arch/powerpc/platforms/pseries/hotplug-memory.c:323:29: error: 'lmb_to_memblock' defined but not used [-Werror=unused-function] static struct memory_block *lmb_to_memblock(struct of_drconf_cell *lmb) ^~~~~~~~~~~~~~~ The only consumer of this function is 'dlpar_remove_lmb', which is enabled with CONFIG_MEMORY_HOTREMOVE, so move it into the same ifdef block. Signed-off-by: Michael Ellerman --- .../platforms/pseries/hotplug-memory.c | 26 +++++++++---------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/powerpc/platforms/pseries/hotplug-memory.c b/arch/powerpc/platforms/pseries/hotplug-memory.c index 43f7beb2902d..76ec104e88be 100644 --- a/arch/powerpc/platforms/pseries/hotplug-memory.c +++ b/arch/powerpc/platforms/pseries/hotplug-memory.c @@ -320,19 +320,6 @@ static int dlpar_remove_device_tree_lmb(struct of_drconf_cell *lmb) return dlpar_update_device_tree_lmb(lmb); } -static struct memory_block *lmb_to_memblock(struct of_drconf_cell *lmb) -{ - unsigned long section_nr; - struct mem_section *mem_sect; - struct memory_block *mem_block; - - section_nr = pfn_to_section_nr(PFN_DOWN(lmb->base_addr)); - mem_sect = __nr_to_section(section_nr); - - mem_block = find_memory_block(mem_sect); - return mem_block; -} - #ifdef CONFIG_MEMORY_HOTREMOVE static int pseries_remove_memblock(unsigned long base, unsigned int memblock_size) { @@ -420,6 +407,19 @@ static bool lmb_is_removable(struct of_drconf_cell *lmb) static int dlpar_add_lmb(struct of_drconf_cell *); +static struct memory_block *lmb_to_memblock(struct of_drconf_cell *lmb) +{ + unsigned long section_nr; + struct mem_section *mem_sect; + struct memory_block *mem_block; + + section_nr = pfn_to_section_nr(PFN_DOWN(lmb->base_addr)); + mem_sect = __nr_to_section(section_nr); + + mem_block = find_memory_block(mem_sect); + return mem_block; +} + static int dlpar_remove_lmb(struct of_drconf_cell *lmb) { struct memory_block *mem_block; From 546c4402c57497a6ffdf5373aff75bc89f0866bc Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 6 Aug 2016 15:54:12 -0700 Subject: [PATCH 092/235] powerpc/vdso: Add missing include file Some powerpc builds fail with the following buld error. In file included from ./arch/powerpc/include/asm/mmu_context.h:11:0, from arch/powerpc/kernel/vdso.c:28: arch/powerpc/include/asm/cputhreads.h: In function 'get_tensr': arch/powerpc/include/asm/cputhreads.h:101:2: error: implicit declaration of function 'cpu_has_feature' Fixes: b92a226e5284 ("powerpc: Move cpu_has_feature() to a separate file") Signed-off-by: Guenter Roeck Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/vdso.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index 6767605ea8da..4111d30badfa 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -22,6 +22,7 @@ #include #include +#include #include #include #include From cbd74e1bc8129efb9908f130a8a6e60fd95d2106 Mon Sep 17 00:00:00 2001 From: Philippe Bergheaud Date: Fri, 5 Aug 2016 14:02:00 +0200 Subject: [PATCH 093/235] cxl: Use fixed width predefined types in data structure. This patch fixes a regression introduced by commit b810253bd934 ("cxl: Add mechanism for delivering AFU driver specific events"). It changes the type u8 to __u8 in the uapi header cxl.h, because the former is a kernel internal type, and may not be defined in userland build environments, in particular when cross-compiling libcxl on x86_64 linux machines (RHEL6.7 and Ubuntu 16.04). This patch also changes the size of the field data_size, and makes it constant, to support 32-bit userland applications running on big-endian ppc64 kernels transparently. mpe: This is an ABI change, however the ABI was only added during the 4.8 merge window so has never been part of a released kernel - therefore we give ourselves permission to change it. Fixes: b810253bd934 ("cxl: Add mechanism for delivering AFU driver specific events") Signed-off-by: Philippe Bergheaud Reviewed-by: Frederic Barrat Signed-off-by: Michael Ellerman --- include/uapi/misc/cxl.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/uapi/misc/cxl.h b/include/uapi/misc/cxl.h index cbae529b7ce0..180d526a55c3 100644 --- a/include/uapi/misc/cxl.h +++ b/include/uapi/misc/cxl.h @@ -136,8 +136,8 @@ struct cxl_event_afu_driver_reserved { * * Of course the contents will be ABI, but that's up the AFU driver. */ - size_t data_size; - u8 data[]; + __u32 data_size; + __u8 data[]; }; struct cxl_event { From 164793379ad3b7ef5fc5a28260c111358892dff3 Mon Sep 17 00:00:00 2001 From: Andrew Donnellan Date: Thu, 28 Jul 2016 15:39:41 +1000 Subject: [PATCH 094/235] cxl: Fix NULL dereference in cxl_context_init() on PowerVM guests Commit f67a6722d650 ("cxl: Workaround PE=0 hardware limitation in Mellanox CX4") added a "min_pe" field to struct cxl_service_layer_ops, to allow us to work around a Mellanox CX-4 hardware limitation. When allocating the PE number in cxl_context_init(), we read from ctx->afu->adapter->native->sl_ops->min_pe to get the minimum PE number. Unsurprisingly, in a PowerVM guest ctx->afu->adapter->native is NULL, and guests don't have a cxl_service_layer_ops struct anywhere. Move min_pe from struct cxl_service_layer_ops to struct cxl so it's accessible in both native and PowerVM environments. For the Mellanox CX-4, set the min_pe value in set_sl_ops(). Fixes: f67a6722d650 ("cxl: Workaround PE=0 hardware limitation in Mellanox CX4") Reported-by: Frederic Barrat Signed-off-by: Andrew Donnellan Acked-by: Ian Munsie Reviewed-by: Frederic Barrat Signed-off-by: Michael Ellerman --- drivers/misc/cxl/context.c | 3 +-- drivers/misc/cxl/cxl.h | 2 +- drivers/misc/cxl/pci.c | 3 ++- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c index bdee9a01ef35..c466ee2b0c97 100644 --- a/drivers/misc/cxl/context.c +++ b/drivers/misc/cxl/context.c @@ -90,8 +90,7 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master, */ mutex_lock(&afu->contexts_lock); idr_preload(GFP_KERNEL); - i = idr_alloc(&ctx->afu->contexts_idr, ctx, - ctx->afu->adapter->native->sl_ops->min_pe, + i = idr_alloc(&ctx->afu->contexts_idr, ctx, ctx->afu->adapter->min_pe, ctx->afu->num_procs, GFP_NOWAIT); idr_preload_end(); mutex_unlock(&afu->contexts_lock); diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h index de090533f18c..344a0ff8f8c7 100644 --- a/drivers/misc/cxl/cxl.h +++ b/drivers/misc/cxl/cxl.h @@ -561,7 +561,6 @@ struct cxl_service_layer_ops { u64 (*timebase_read)(struct cxl *adapter); int capi_mode; bool needs_reset_before_disable; - int min_pe; }; struct cxl_native { @@ -603,6 +602,7 @@ struct cxl { struct bin_attribute cxl_attr; int adapter_num; int user_irqs; + int min_pe; u64 ps_size; u16 psl_rev; u16 base_image; diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c index d152e2de8c93..1d0347c36e6d 100644 --- a/drivers/misc/cxl/pci.c +++ b/drivers/misc/cxl/pci.c @@ -1521,14 +1521,15 @@ static const struct cxl_service_layer_ops xsl_ops = { .write_timebase_ctrl = write_timebase_ctrl_xsl, .timebase_read = timebase_read_xsl, .capi_mode = OPAL_PHB_CAPI_MODE_DMA, - .min_pe = 1, /* Workaround for Mellanox CX4 HW bug */ }; static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev) { if (dev->vendor == PCI_VENDOR_ID_MELLANOX && dev->device == 0x1013) { + /* Mellanox CX-4 */ dev_info(&adapter->dev, "Device uses an XSL\n"); adapter->native->sl_ops = &xsl_ops; + adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */ } else { dev_info(&adapter->dev, "Device uses a PSL\n"); adapter->native->sl_ops = &psl_ops; From 6fd40f192a9dba391b2d84882f1ed3169c52b714 Mon Sep 17 00:00:00 2001 From: Andrew Donnellan Date: Fri, 22 Jul 2016 19:01:36 +1000 Subject: [PATCH 095/235] cxl: Fix sparse warnings Make native_irq_wait() static and use NULL rather than 0 to initialise phb->cfg_data in cxl_pci_vphb_add() to remove sparse warnings. Signed-off-by: Andrew Donnellan Reviewed-by: Matthew R. Ochs Acked-by: Ian Munsie Signed-off-by: Michael Ellerman --- drivers/misc/cxl/native.c | 2 +- drivers/misc/cxl/vphb.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c index 3bcdaee11ba1..e606fdc4bc9c 100644 --- a/drivers/misc/cxl/native.c +++ b/drivers/misc/cxl/native.c @@ -924,7 +924,7 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data) return fail_psl_irq(afu, &irq_info); } -void native_irq_wait(struct cxl_context *ctx) +static void native_irq_wait(struct cxl_context *ctx) { u64 dsisr; int timeout = 1000; diff --git a/drivers/misc/cxl/vphb.c b/drivers/misc/cxl/vphb.c index dee8def1c193..7ada5f1b7bb6 100644 --- a/drivers/misc/cxl/vphb.c +++ b/drivers/misc/cxl/vphb.c @@ -221,7 +221,7 @@ int cxl_pci_vphb_add(struct cxl_afu *afu) /* Setup the PHB using arch provided callback */ phb->ops = &cxl_pcie_pci_ops; phb->cfg_addr = NULL; - phb->cfg_data = 0; + phb->cfg_data = NULL; phb->private_data = afu; phb->controller_ops = cxl_pci_controller_ops; From 10560b9afc8abf349843dff88c45dd43223e803e Mon Sep 17 00:00:00 2001 From: "Guilherme G. Piccoli" Date: Fri, 22 Jul 2016 14:05:29 -0300 Subject: [PATCH 096/235] powerpc/eeh: Switch to conventional PCI address output in EEH log This is a very minor/trivial fix for the output of PCI address on EEH logs. The PCI address on "OF node" field currently is using ":" as a separator for the function, but the usual separator is ".". This patch changes the separator to dot, so the PCI address is printed as usual. Signed-off-by: Guilherme G. Piccoli Reviewed-by: Gavin Shan Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/eeh.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c index c9bc78e9c610..7429556eb8df 100644 --- a/arch/powerpc/kernel/eeh.c +++ b/arch/powerpc/kernel/eeh.c @@ -168,10 +168,10 @@ static size_t eeh_dump_dev_log(struct eeh_dev *edev, char *buf, size_t len) int n = 0, l = 0; char buffer[128]; - n += scnprintf(buf+n, len-n, "%04x:%02x:%02x:%01x\n", + n += scnprintf(buf+n, len-n, "%04x:%02x:%02x.%01x\n", edev->phb->global_number, pdn->busno, PCI_SLOT(pdn->devfn), PCI_FUNC(pdn->devfn)); - pr_warn("EEH: of node=%04x:%02x:%02x:%01x\n", + pr_warn("EEH: of node=%04x:%02x:%02x.%01x\n", edev->phb->global_number, pdn->busno, PCI_SLOT(pdn->devfn), PCI_FUNC(pdn->devfn)); From 61e8a0d5a0270b91581f6c715036844b2ea98da1 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 5 Aug 2016 16:40:56 +1000 Subject: [PATCH 097/235] powerpc/pci: Fix endian bug in fixed PHB numbering The recent commit 63a72284b159 ("powerpc/pci: Assign fixed PHB number based on device-tree properties"), added code to read a 64-bit property from the device tree, and if not found read a 32-bit property (reg). There was a bug in the 32-bit case, on big endian machines, due to the use of the 64-bit value to read the 32-bit property. The cast of &prop means we end up writing to the high 32-bit of prop, leaving the low 32-bits containing whatever junk was on the stack. If that junk value was non-zero, and < MAX_PHBS, we would end up using it as the PHB id. This results in users seeing what appear to be random PHB ids. Fix it by reading into a u32 property and then assigning that to the u64 value, letting the CPU do the correct conversions for us. Fixes: 63a72284b159 ("powerpc/pci: Assign fixed PHB number based on device-tree properties") Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/pci-common.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index a5c0153ede37..7fdf324d5b51 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -78,6 +78,7 @@ EXPORT_SYMBOL(get_pci_dma_ops); static int get_phb_number(struct device_node *dn) { int ret, phb_id = -1; + u32 prop_32; u64 prop; /* @@ -86,8 +87,10 @@ static int get_phb_number(struct device_node *dn) * reading "ibm,opal-phbid", only present in OPAL environment. */ ret = of_property_read_u64(dn, "ibm,opal-phbid", &prop); - if (ret) - ret = of_property_read_u32_index(dn, "reg", 1, (u32 *)&prop); + if (ret) { + ret = of_property_read_u32_index(dn, "reg", 1, &prop_32); + prop = prop_32; + } if (!ret) phb_id = (int)(prop & (MAX_PHBS - 1)); From 46c8f0b077a838eb1f6169bb370aab8ed98f7630 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Mon, 8 Aug 2016 16:29:07 -0400 Subject: [PATCH 098/235] timers: Fix get_next_timer_interrupt() computation The tick_nohz_stop_sched_tick() routine is not properly canceling the sched timer when nothing is pending, because get_next_timer_interrupt() is no longer returning KTIME_MAX in that case. This causes periodic interrupts when none are needed. When determining the next interrupt time, we first use __next_timer_interrupt() to get the first expiring timer in the timer wheel. If no timer is found, we return the base clock value plus NEXT_TIMER_MAX_DELTA to indicate there is no timer in the timer wheel. Back in get_next_timer_interrupt(), we set the "expires" value by converting the timer wheel expiry (in ticks) to a nsec value. But we don't want to do this if the timer wheel expiry value indicates no timer; we want to return KTIME_MAX. Prior to commit 500462a9de65 ("timers: Switch to a non-cascading wheel") we checked base->active_timers to see if any timers were active, and if not, we didn't touch the expiry value and so properly returned KTIME_MAX. Now we don't have active_timers. To fix this, we now just check the timer wheel expiry value to see if it is "now + NEXT_TIMER_MAX_DELTA", and if it is, we don't try to compute a new value based on it, but instead simply let the KTIME_MAX value in expires remain. Fixes: 500462a9de65 "timers: Switch to a non-cascading wheel" Signed-off-by: Chris Metcalf Cc: Frederic Weisbecker Cc: Christoph Lameter Cc: John Stultz Link: http://lkml.kernel.org/r/1470688147-22287-1-git-send-email-cmetcalf@mellanox.com Signed-off-by: Thomas Gleixner --- kernel/time/timer.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 555670a5143c..32bf6f75a8fe 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -1496,6 +1496,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]); u64 expires = KTIME_MAX; unsigned long nextevt; + bool is_max_delta; /* * Pretend that there is no timer pending if the cpu is offline. @@ -1506,6 +1507,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) spin_lock(&base->lock); nextevt = __next_timer_interrupt(base); + is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA); base->next_expiry = nextevt; /* * We have a fresh next event. Check whether we can forward the base: @@ -1519,7 +1521,8 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) expires = basem; base->is_idle = false; } else { - expires = basem + (nextevt - basej) * TICK_NSEC; + if (!is_max_delta) + expires = basem + (nextevt - basej) * TICK_NSEC; /* * If we expect to sleep more than a tick, mark the base idle: */ From c74dd88e77d3ecbc9e55c78796d82c9aa21cabad Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Tue, 9 Aug 2016 10:39:13 +0530 Subject: [PATCH 099/235] powerpc/book3s: Fix MCE console messages for unrecoverable MCE. When machine check occurs with MSR(RI=0), it means MC interrupt is unrecoverable and kernel goes down to panic path. But the console message still shows it as recovered. This patch fixes the MCE console messages. Fixes: 36df96f8acaf ("powerpc/book3s: Decode and save machine check event.") Signed-off-by: Mahesh Salgaonkar Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/mce.c | 3 ++- arch/powerpc/platforms/powernv/opal.c | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/mce.c b/arch/powerpc/kernel/mce.c index ef267fd9dd22..5e7ece0fda9f 100644 --- a/arch/powerpc/kernel/mce.c +++ b/arch/powerpc/kernel/mce.c @@ -92,7 +92,8 @@ void save_mce_event(struct pt_regs *regs, long handled, mce->in_use = 1; mce->initiator = MCE_INITIATOR_CPU; - if (handled) + /* Mark it recovered if we have handled it and MSR(RI=1). */ + if (handled && (regs->msr & MSR_RI)) mce->disposition = MCE_DISPOSITION_RECOVERED; else mce->disposition = MCE_DISPOSITION_NOT_RECOVERED; diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c index 8b4fc68cebcb..6c9a65b52e63 100644 --- a/arch/powerpc/platforms/powernv/opal.c +++ b/arch/powerpc/platforms/powernv/opal.c @@ -399,6 +399,7 @@ static int opal_recover_mce(struct pt_regs *regs, if (!(regs->msr & MSR_RI)) { /* If MSR_RI isn't set, we cannot recover */ + pr_err("Machine check interrupt unrecoverable: MSR(RI=0)\n"); recovered = 0; } else if (evt->disposition == MCE_DISPOSITION_RECOVERED) { /* Platform corrected itself */ From 5958d19a143eb229e9ece20bd4c781ad41cb7d24 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 8 Jul 2016 15:55:43 +1000 Subject: [PATCH 100/235] powerpc/pnv/pci: Fix incorrect PE reservation attempt on some 64-bit BARs The generic allocation code may sometimes decide to assign a prefetchable 64-bit BAR to the M32 window. In fact it may also decide to allocate a 64-bit non-prefetchable BAR to the M64 one ! So using the resource flags as a test to decide which window was used for PE allocation is just wrong and leads to insane PE numbers. Instead, compare the addresses to figure it out. Signed-off-by: Benjamin Herrenschmidt [mpe: Rename the function as agreed by Ben & Gavin] Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/powernv/pci-ioda.c | 30 +++++++++++++++-------- 1 file changed, 20 insertions(+), 10 deletions(-) diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index dedbbe9785ba..fd9444f9fb0c 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -111,10 +111,17 @@ static int __init iommu_setup(char *str) } early_param("iommu", iommu_setup); -static inline bool pnv_pci_is_mem_pref_64(unsigned long flags) +static inline bool pnv_pci_is_m64(struct pnv_phb *phb, struct resource *r) { - return ((flags & (IORESOURCE_MEM_64 | IORESOURCE_PREFETCH)) == - (IORESOURCE_MEM_64 | IORESOURCE_PREFETCH)); + /* + * WARNING: We cannot rely on the resource flags. The Linux PCI + * allocation code sometimes decides to put a 64-bit prefetchable + * BAR in the 32-bit window, so we have to compare the addresses. + * + * For simplicity we only test resource start. + */ + return (r->start >= phb->ioda.m64_base && + r->start < (phb->ioda.m64_base + phb->ioda.m64_size)); } static struct pnv_ioda_pe *pnv_ioda_init_pe(struct pnv_phb *phb, int pe_no) @@ -229,7 +236,7 @@ static void pnv_ioda_reserve_dev_m64_pe(struct pci_dev *pdev, sgsz = phb->ioda.m64_segsize; for (i = 0; i <= PCI_ROM_RESOURCE; i++) { r = &pdev->resource[i]; - if (!r->parent || !pnv_pci_is_mem_pref_64(r->flags)) + if (!r->parent || !pnv_pci_is_m64(phb, r)) continue; start = _ALIGN_DOWN(r->start - base, sgsz); @@ -2863,7 +2870,7 @@ static void pnv_pci_ioda_fixup_iov_resources(struct pci_dev *pdev) res = &pdev->resource[i + PCI_IOV_RESOURCES]; if (!res->flags || res->parent) continue; - if (!pnv_pci_is_mem_pref_64(res->flags)) { + if (!pnv_pci_is_m64(phb, res)) { dev_warn(&pdev->dev, "Don't support SR-IOV with" " non M64 VF BAR%d: %pR. \n", i, res); @@ -2958,7 +2965,7 @@ static void pnv_ioda_setup_pe_res(struct pnv_ioda_pe *pe, index++; } } else if ((res->flags & IORESOURCE_MEM) && - !pnv_pci_is_mem_pref_64(res->flags)) { + !pnv_pci_is_m64(phb, res)) { region.start = res->start - phb->hose->mem_offset[0] - phb->ioda.m32_pci_base; @@ -3083,9 +3090,12 @@ static resource_size_t pnv_pci_window_alignment(struct pci_bus *bus, bridge = bridge->bus->self; } - /* We fail back to M32 if M64 isn't supported */ - if (phb->ioda.m64_segsize && - pnv_pci_is_mem_pref_64(type)) + /* + * We fall back to M32 if M64 isn't supported. We enforce the M64 + * alignment for any 64-bit resource, PCIe doesn't care and + * bridges only do 64-bit prefetchable anyway. + */ + if (phb->ioda.m64_segsize && (type & IORESOURCE_MEM_64)) return phb->ioda.m64_segsize; if (type & IORESOURCE_MEM) return phb->ioda.m32_segsize; @@ -3125,7 +3135,7 @@ static void pnv_pci_fixup_bridge_resources(struct pci_bus *bus, w = NULL; if (r->flags & type & IORESOURCE_IO) w = &hose->io_resource; - else if (pnv_pci_is_mem_pref_64(r->flags) && + else if (pnv_pci_is_m64(phb, r) && (type & IORESOURCE_PREFETCH) && phb->ioda.m64_segsize) w = &hose->mem_resources[1]; From 50ee91bdef41c15b671dcd9446ee007a1d2f5ab7 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Tue, 9 Aug 2016 10:30:49 +0800 Subject: [PATCH 101/235] arm64: Support hard limit of cpu count by nr_cpus Enable the hard limit of cpu count by set boot options nr_cpus=x on arm64, and make a minor change about message when total number of cpu exceeds the limit. Reviewed-by: Suzuki K Poulose Reported-by: Shiyuan Hu Signed-off-by: Kefeng Wang Signed-off-by: Will Deacon --- arch/arm64/kernel/smp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 76a6d9263908..d93d43352504 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -661,9 +661,9 @@ void __init smp_init_cpus(void) acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_INTERRUPT, acpi_parse_gic_cpu_interface, 0); - if (cpu_count > NR_CPUS) - pr_warn("no. of cores (%d) greater than configured maximum of %d - clipping\n", - cpu_count, NR_CPUS); + if (cpu_count > nr_cpu_ids) + pr_warn("Number of cores (%d) exceeds configured maximum of %d - clipping\n", + cpu_count, nr_cpu_ids); if (!bootcpu_valid) { pr_err("missing boot CPU MPIDR, not enabling secondaries\n"); @@ -677,7 +677,7 @@ void __init smp_init_cpus(void) * with entries in cpu_logical_map while initializing the cpus. * If the cpu set-up fails, invalidate the cpu_logical_map entry. */ - for (i = 1; i < NR_CPUS; i++) { + for (i = 1; i < nr_cpu_ids; i++) { if (cpu_logical_map(i) != INVALID_HWID) { if (smp_cpu_setup(i)) cpu_logical_map(i) = INVALID_HWID; From 58625edf9e2515ed41dac2a24fa8004030a87b87 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 2 Aug 2016 14:16:31 +0000 Subject: [PATCH 102/235] virtio: fix memory leak in virtqueue_add() When using the indirect buffers feature, 'desc' is allocated in virtqueue_add() but isn't freed before leaving on a ring full error, causing a memory leak. For example, it seems rather clear that this can trigger with virtio net if mergeable buffers are not used. Cc: stable@vger.kernel.org Signed-off-by: Wei Yongjun Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio_ring.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 114a0c88afb8..5ed228ddadba 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -327,6 +327,8 @@ static inline int virtqueue_add(struct virtqueue *_vq, * host should service the ring ASAP. */ if (out_sgs) vq->notify(&vq->vq); + if (indirect) + kfree(desc); END_USE(vq); return -ENOSPC; } From 3cc36f6e34bd2d92d23be7b598ba5e639c47b01a Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 3 Aug 2016 07:18:51 +0300 Subject: [PATCH 103/235] virtio: fix error handling for debug builds On error, virtqueue_add calls START_USE but not END_USE. Thankfully that's normally empty anyway, but might not be when debugging. Fix it up. Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio_ring.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 5ed228ddadba..e383ecdaca59 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -428,6 +428,7 @@ static inline int virtqueue_add(struct virtqueue *_vq, if (indirect) kfree(desc); + END_USE(vq); return -EIO; } From 1b8553c04bf95180eb91be94f089a1e8b38cfd62 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 3 Aug 2016 19:59:47 +0300 Subject: [PATCH 104/235] 9p/trans_virtio: use kvfree() for iov_iter_get_pages_alloc() The memory allocated by iov_iter_get_pages_alloc() can be allocated with vmalloc() if kmalloc() failed -- see get_pages_array(). In that case we need to free it with vfree(), so let's use kvfree(). The bug manifests like this: BUG: unable to handle kernel paging request at ffffeb0400072da0 IP: [] kfree+0x4b/0x140 PGD 0 Oops: 0000 [#1] PREEMPT SMP KASAN CPU: 2 PID: 675 Comm: trinity-c2 Not tainted 4.7.0-rc7+ #14 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014 task: ffff8800badef2c0 ti: ffff880069208000 task.ti: ffff880069208000 RIP: 0010:[] [] kfree+0x4b/0x140 RSP: 0000:ffff88006920f3f0 EFLAGS: 00010282 RAX: ffffea0000000000 RBX: ffffc90001cb6000 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000246 RDI: ffffc90001cb6000 RBP: ffff88006920f410 R08: 0000000000000000 R09: dffffc0000000000 R10: ffff8800badefa30 R11: 0000056a3d3b0d9f R12: ffff88006920f620 R13: ffffeb0400072d80 R14: ffff8800baa94078 R15: 0000000000000000 FS: 00007fbd2b437700(0000) GS:ffff88011af00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffeb0400072da0 CR3: 000000006926d000 CR4: 00000000000006e0 Stack: 0000000000000001 ffff88006920f620 ffffed001755280f ffff8800baa94078 ffff88006920f6a8 ffffffff8310442b dffffc0000000000 ffff8800badefa30 ffff8800badefa28 ffff88011af1fba0 1ffff1000d241e98 ffff8800ba892150 Call Trace: [] p9_virtio_zc_request+0x72b/0xdb0 [] p9_client_zc_rpc.constprop.8+0x246/0xb10 [] p9_client_read+0x4c9/0x750 [] v9fs_fid_readpage+0x14c/0x320 [] v9fs_vfs_readpage+0x36/0x50 [] filemap_fault+0x9a3/0xe60 [] __do_fault+0x158/0x300 [] handle_mm_fault+0x1cf1/0x3c80 [] __do_page_fault+0x30a/0x8e0 [] do_page_fault+0x2f/0x80 [] do_async_page_fault+0x27/0xa0 [] async_page_fault+0x28/0x30 Code: 00 80 41 54 53 49 01 fd 48 0f 42 05 b0 39 67 02 48 89 fb 49 01 c5 48 b8 00 00 00 00 00 ea ff ff 49 c1 ed 0c 49 c1 e5 06 49 01 c5 <49> 8b 45 20 48 8d 50 ff a8 01 4c 0f 45 ea 49 8b 55 20 48 8d 42 RIP [] kfree+0x4b/0x140 RSP CR2: ffffeb0400072da0 ---[ end trace f3d59a04bafec038 ]--- Cc: Al Viro Signed-off-by: Vegard Nossum Signed-off-by: Michael S. Tsirkin --- net/9p/trans_virtio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 4acb1d5417aa..f24b25c25106 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -507,8 +507,8 @@ p9_virtio_zc_request(struct p9_client *client, struct p9_req_t *req, /* wakeup anybody waiting for slots to pin pages */ wake_up(&vp_wq); } - kfree(in_pages); - kfree(out_pages); + kvfree(in_pages); + kvfree(out_pages); return err; } From 3fda5d6e580193fa005014355b3a61498f1b3ae0 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Thu, 4 Aug 2016 14:52:53 +0100 Subject: [PATCH 105/235] vhost/vsock: fix vhost virtio_vsock_pkt use-after-free Stash the packet length in a local variable before handing over ownership of the packet to virtio_transport_recv_pkt() or virtio_transport_free_pkt(). This patch solves the use-after-free since pkt is no longer guaranteed to be alive. Reported-by: Dan Carpenter Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- drivers/vhost/vsock.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c index 0ddf3a2dbfc4..e3b30ea9ece5 100644 --- a/drivers/vhost/vsock.c +++ b/drivers/vhost/vsock.c @@ -307,6 +307,8 @@ static void vhost_vsock_handle_tx_kick(struct vhost_work *work) vhost_disable_notify(&vsock->dev, vq); for (;;) { + u32 len; + if (!vhost_vsock_more_replies(vsock)) { /* Stop tx until the device processes already * pending replies. Leave tx virtqueue @@ -334,13 +336,15 @@ static void vhost_vsock_handle_tx_kick(struct vhost_work *work) continue; } + len = pkt->len; + /* Only accept correctly addressed packets */ if (le64_to_cpu(pkt->hdr.src_cid) == vsock->guest_cid) virtio_transport_recv_pkt(pkt); else virtio_transport_free_pkt(pkt); - vhost_add_used(vq, head, sizeof(pkt->hdr) + pkt->len); + vhost_add_used(vq, head, sizeof(pkt->hdr) + len); added = true; } From 28ad55578b8a76390d966b09da8c7fa3644f5140 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Fri, 5 Aug 2016 13:52:09 +0100 Subject: [PATCH 106/235] virtio-vsock: fix include guard typo Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- include/uapi/linux/virtio_vsock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/virtio_vsock.h b/include/uapi/linux/virtio_vsock.h index 6b011c19b50f..1d57ed3d84d2 100644 --- a/include/uapi/linux/virtio_vsock.h +++ b/include/uapi/linux/virtio_vsock.h @@ -32,7 +32,7 @@ */ #ifndef _UAPI_LINUX_VIRTIO_VSOCK_H -#define _UAPI_LINUX_VIRTIO_VOSCK_H +#define _UAPI_LINUX_VIRTIO_VSOCK_H #include #include From 347a529398e8e723338cca5d8a8ae2d9e7e93448 Mon Sep 17 00:00:00 2001 From: Minfei Huang Date: Tue, 9 Aug 2016 16:39:20 +0800 Subject: [PATCH 107/235] virtio_blk: Fix a slient kernel panic We do a lot of memory allocation in function init_vq, and don't handle the allocation failure properly. Then this function will return 0, although initialization fails due to lacking memory. At that moment, kernel will panic in guest machine, if virtio is used to drive disk. To fix this bug, we should take care of allocation failure, and return correct value to let caller know what happen. Tested-by: Chao Fan Signed-off-by: Minfei Huang Signed-off-by: Minfei Huang Reviewed-by: Cornelia Huck Reviewed-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- drivers/block/virtio_blk.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 1523e05c46fc..93b1aaa5ba3b 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -391,23 +391,17 @@ static int init_vq(struct virtio_blk *vblk) num_vqs = 1; vblk->vqs = kmalloc(sizeof(*vblk->vqs) * num_vqs, GFP_KERNEL); - if (!vblk->vqs) { + if (!vblk->vqs) + return -ENOMEM; + + names = kmalloc(sizeof(*names) * num_vqs, GFP_KERNEL); + callbacks = kmalloc(sizeof(*callbacks) * num_vqs, GFP_KERNEL); + vqs = kmalloc(sizeof(*vqs) * num_vqs, GFP_KERNEL); + if (!names || !callbacks || !vqs) { err = -ENOMEM; goto out; } - names = kmalloc(sizeof(*names) * num_vqs, GFP_KERNEL); - if (!names) - goto err_names; - - callbacks = kmalloc(sizeof(*callbacks) * num_vqs, GFP_KERNEL); - if (!callbacks) - goto err_callbacks; - - vqs = kmalloc(sizeof(*vqs) * num_vqs, GFP_KERNEL); - if (!vqs) - goto err_vqs; - for (i = 0; i < num_vqs; i++) { callbacks[i] = virtblk_done; snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req.%d", i); @@ -417,7 +411,7 @@ static int init_vq(struct virtio_blk *vblk) /* Discover virtqueues and write information to configuration. */ err = vdev->config->find_vqs(vdev, num_vqs, vqs, callbacks, names); if (err) - goto err_find_vqs; + goto out; for (i = 0; i < num_vqs; i++) { spin_lock_init(&vblk->vqs[i].lock); @@ -425,16 +419,12 @@ static int init_vq(struct virtio_blk *vblk) } vblk->num_vqs = num_vqs; - err_find_vqs: +out: kfree(vqs); - err_vqs: kfree(callbacks); - err_callbacks: kfree(names); - err_names: if (err) kfree(vblk->vqs); - out: return err; } From 2ab0d56aadbcd120b8fa524b4a1142e8b06e13c8 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Thu, 7 Jul 2016 17:07:56 +0200 Subject: [PATCH 108/235] virtio/s390: keep early_put_chars In case the registration of the hvc tty never happens AND the kernel thinks that hvc0 is the preferred console we should keep the early printk function to avoid a kernel panic due to code being removed. Signed-off-by: Christian Borntraeger Signed-off-by: Jing Liu Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin --- drivers/s390/virtio/kvm_virtio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/virtio/kvm_virtio.c b/drivers/s390/virtio/kvm_virtio.c index 1d060fd293a3..b0a849f02df3 100644 --- a/drivers/s390/virtio/kvm_virtio.c +++ b/drivers/s390/virtio/kvm_virtio.c @@ -482,7 +482,7 @@ static int __init kvm_devices_init(void) } /* code for early console output with virtio_console */ -static __init int early_put_chars(u32 vtermno, const char *buf, int count) +static int early_put_chars(u32 vtermno, const char *buf, int count) { char scratch[17]; unsigned int len = count; From 3b2fbb3f06efe5bd2dfdce2a1db703e23c1a78af Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 7 Jul 2016 17:07:57 +0200 Subject: [PATCH 109/235] virtio/s390: deprecate old transport There only ever have been two host implementations of the old s390-virtio (pre-ccw) transport: the experimental kuli userspace, and qemu. As qemu switched its default to ccw with 2.4 (with most users having used ccw well before that) and removed the old transport entirely in 2.6, s390-virtio probably hasn't been in active use for quite some time and is therefore likely to bitrot. Let's start the slow march towards removing the code by deprecating it. Note that this also deprecates the early virtio console code, which has been causing trouble in the guest without being wired up in any relevant hypervisor code. Acked-by: Christian Borntraeger Reviewed-by: Dong Jia Shi Reviewed-by: Sascha Silbe Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin --- arch/s390/Kconfig | 13 +++++++++++++ drivers/s390/virtio/Makefile | 6 +++++- drivers/s390/virtio/kvm_virtio.c | 2 ++ 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 9e607bf2d640..5a2907ee8c93 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -871,4 +871,17 @@ config S390_GUEST Select this option if you want to run the kernel as a guest under the KVM hypervisor. +config S390_GUEST_OLD_TRANSPORT + def_bool y + prompt "Guest support for old s390 virtio transport (DEPRECATED)" + depends on S390_GUEST + help + Enable this option to add support for the old s390-virtio + transport (i.e. virtio devices NOT based on virtio-ccw). This + type of virtio devices is only available on the experimental + kuli userspace or with old (< 2.6) qemu. If you are running + with a modern version of qemu (which supports virtio-ccw since + 1.4 and uses it by default since version 2.4), you probably won't + need this. + endmenu diff --git a/drivers/s390/virtio/Makefile b/drivers/s390/virtio/Makefile index 241891a57caf..df40692a9011 100644 --- a/drivers/s390/virtio/Makefile +++ b/drivers/s390/virtio/Makefile @@ -6,4 +6,8 @@ # it under the terms of the GNU General Public License (version 2 only) # as published by the Free Software Foundation. -obj-$(CONFIG_S390_GUEST) += kvm_virtio.o virtio_ccw.o +s390-virtio-objs := virtio_ccw.o +ifdef CONFIG_S390_GUEST_OLD_TRANSPORT +s390-virtio-objs += kvm_virtio.o +endif +obj-$(CONFIG_S390_GUEST) += $(s390-virtio-objs) diff --git a/drivers/s390/virtio/kvm_virtio.c b/drivers/s390/virtio/kvm_virtio.c index b0a849f02df3..5e5c11f37b24 100644 --- a/drivers/s390/virtio/kvm_virtio.c +++ b/drivers/s390/virtio/kvm_virtio.c @@ -458,6 +458,8 @@ static int __init kvm_devices_init(void) if (test_devices_support(total_memory_size) < 0) return -ENODEV; + pr_warn("The s390-virtio transport is deprecated. Please switch to a modern host providing virtio-ccw.\n"); + rc = vmem_add_mapping(total_memory_size, PAGE_SIZE); if (rc) return rc; From 97b1d23f7bcbee00a5b85bf2c022c612ea67429d Mon Sep 17 00:00:00 2001 From: James Hogan Date: Fri, 5 Aug 2016 12:08:36 +0100 Subject: [PATCH 110/235] metag: Drop show_mem() from mem_init() The recent commit 599d0c954f91 ("mm, vmscan: move LRU lists to node"), changed memory management code so that show_mem() is no longer safe to call prior to setup_per_cpu_pageset(), as pgdat->per_cpu_nodestats will still be NULL. This causes an oops on metag due to the call to show_mem() from mem_init(): node_page_state_snapshot(...) + 0x48 pgdat_reclaimable(struct pglist_data * pgdat = 0x402517a0) show_free_areas(unsigned int filter = 0) + 0x2cc show_mem(unsigned int filter = 0) + 0x18 mem_init() mm_init() start_kernel() + 0x204 This wasn't a problem before with zone_reclaimable() as zone_pcp_init() was already setting zone->pageset to &boot_pageset, via setup_arch() and paging_init(), which happens before mm_init(): zone_pcp_init(...) free_area_init_core(...) + 0x138 free_area_init_node(int nid = 0, ...) + 0x1a0 free_area_init_nodes(...) + 0x440 paging_init(unsigned long mem_end = 0x4fe00000) + 0x378 setup_arch(char ** cmdline_p = 0x4024e038) + 0x2b8 start_kernel() + 0x54 No other arches appear to call show_mem() during boot, and it doesn't really add much value to the log, so lets just drop it from mem_init(). Signed-off-by: James Hogan Acked-by: Mel Gorman Cc: linux-metag@vger.kernel.org --- arch/metag/mm/init.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/metag/mm/init.c b/arch/metag/mm/init.c index 11fa51c89617..c0ec116b3993 100644 --- a/arch/metag/mm/init.c +++ b/arch/metag/mm/init.c @@ -390,7 +390,6 @@ void __init mem_init(void) free_all_bootmem(); mem_init_print_info(NULL); - show_mem(0); } void free_initmem(void) From bcdc09af3ef30ef071677544ce23a1c8873a2dda Mon Sep 17 00:00:00 2001 From: Brendan Gregg Date: Wed, 3 Aug 2016 02:47:49 +0000 Subject: [PATCH 111/235] perf script: Add 'bpf-output' field to usage message This adds the 'bpf-output' field to the perf script usage message, and docs. Signed-off-by: Brendan Gregg Cc: Alexander Shishkin Cc: Alexei Starovoitov Cc: Peter Zijlstra Cc: Wang Nan Link: http://lkml.kernel.org/r/1470192469-11910-4-git-send-email-bgregg@netflix.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/perf-script.txt | 4 ++-- tools/perf/builtin-script.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt index 1f6c70594f0f..053bbbd84ece 100644 --- a/tools/perf/Documentation/perf-script.txt +++ b/tools/perf/Documentation/perf-script.txt @@ -116,8 +116,8 @@ OPTIONS --fields:: Comma separated list of fields to print. Options are: comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff, - srcline, period, iregs, brstack, brstacksym, flags. - Field list can be prepended with the type, trace, sw or hw, + srcline, period, iregs, brstack, brstacksym, flags, bpf-output, + callindent. Field list can be prepended with the type, trace, sw or hw, to indicate to which event type the field list applies. e.g., -F sw:comm,tid,time,ip,sym and -F trace:time,cpu,trace diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 971ff91b16cb..9c640a8081c7 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -2116,7 +2116,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused) "Valid types: hw,sw,trace,raw. " "Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso," "addr,symoff,period,iregs,brstack,brstacksym,flags," - "callindent", parse_output_fields), + "bpf-output,callindent", parse_output_fields), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]", From 887fa86d6fd7a45cee2d0f9d5f75026786d61df2 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 5 Aug 2016 12:37:21 -0300 Subject: [PATCH 112/235] perf hists: Trim libtraceevent trace_seq buffers When we use libtraceevent to format trace event fields into printable strings to use in hist entries it is important to trim it from the default 4 KiB it starts with to what is really used, to reduce the memory footprint, so use realloc(seq.buffer, seq.len + 1) when returning the seq.buffer formatted with the fields contents. Reported-and-Tested-by: Wang Nan Cc: Adrian Hunter Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Link: http://lkml.kernel.org/n/tip-t3hl7uxmilrkigzmc90rlhk2@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/sort.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 947d21f38398..3d3cb8392c86 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -588,7 +588,11 @@ static char *get_trace_output(struct hist_entry *he) } else { pevent_event_info(&seq, evsel->tp_format, &rec); } - return seq.buffer; + /* + * Trim the buffer, it starts at 4KB and we're not going to + * add anything more to this buffer. + */ + return realloc(seq.buffer, seq.len + 1); } static int64_t From 8e34189b347d76acf48ce05831176582201b664d Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Sat, 6 Aug 2016 19:29:48 +0900 Subject: [PATCH 113/235] perf probe: Adjust map->reloc offset when finding kernel symbol from map Adjust map->reloc offset for the unmapped address when finding alternative symbol address from map, because KASLR can relocate the kernel symbol address. The same adjustment has been done when finding appropriate kernel symbol address from map which was introduced by commit f90acac75713 ("perf probe: Find given address from offline dwarf") Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Masami Hiramatsu Cc: Alexei Starovoitov Cc: Wang Nan Link: http://lkml.kernel.org/r/20160806192948.e366f3fbc4b194de600f8326@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/probe-event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 953dc1ab2ed7..d5ccb656fd81 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -385,7 +385,7 @@ static int find_alternative_probe_point(struct debuginfo *dinfo, if (uprobes) address = sym->start; else - address = map->unmap_ip(map, sym->start); + address = map->unmap_ip(map, sym->start) - map->reloc; break; } if (!address) { From cb3f3378cd09aa3fe975b4ad5ee0229dc76315bb Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Fri, 5 Aug 2016 15:22:36 +0300 Subject: [PATCH 114/235] perf probe: Fix module name matching If module is "module" then dso->short_name is "[module]". Substring comparing is't enough: "raid10" matches to "[raid1]". This patch also checks terminating zero in module name. Signed-off-by: Konstantin Khlebnikov Acked-by: Masami Hiramatsu Link: http://lkml.kernel.org/r/147039975648.715620.12985971832789032159.stgit@buzz Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/probe-event.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index d5ccb656fd81..1201f73ca723 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -170,8 +170,10 @@ static struct map *kernel_get_module_map(const char *module) module = "kernel"; for (pos = maps__first(maps); pos; pos = map__next(pos)) { + /* short_name is "[module]" */ if (strncmp(pos->dso->short_name + 1, module, - pos->dso->short_name_len - 2) == 0) { + pos->dso->short_name_len - 2) == 0 && + module[pos->dso->short_name_len - 2] == '\0') { return pos; } } From 3df33eff2ba96be4f1535db4f672013d756dc9b1 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Tue, 9 Aug 2016 14:04:29 +0100 Subject: [PATCH 115/235] perf stat: Avoid skew when reading events When we don't have a tracee (i.e. we're attaching to a task or CPU), counters can still be running after our workload finishes, and can still be running as we read their values. As we read events one-by-one, there can be arbitrary skew between values of events, even within a group. This means that ratios within an event group are not reliable. This skew can be seen if measuring a group of identical events, e.g: # perf stat -a -C0 -e '{cycles,cycles}' sleep 1 To avoid this, we must stop groups from counting before we read the values of any constituent events. This patch adds and makes use of a new disable_counters() helper, which disables group leaders (and thus each group as a whole). This mirrors the use of enable_counters() for starting event groups in the absence of a tracee. Closing a group leader splits the group, and without a disabled group leader the newly split events will begin counting. Thus to ensure counts are reliable we must defer closing group leaders until all counts have been read. To do so this patch removes the event closing logic from the read_counters() helper, explicitly closes the events using perf_evlist__close(), which also aids legibility. Signed-off-by: Mark Rutland Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1470747869-3567-1-git-send-email-mark.rutland@arm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-stat.c | 31 +++++++++++++++++++++++-------- 1 file changed, 23 insertions(+), 8 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 0c16d20d7e32..3c7452b39f57 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -331,7 +331,7 @@ static int read_counter(struct perf_evsel *counter) return 0; } -static void read_counters(bool close_counters) +static void read_counters(void) { struct perf_evsel *counter; @@ -341,11 +341,6 @@ static void read_counters(bool close_counters) if (perf_stat_process_counter(&stat_config, counter)) pr_warning("failed to process counter %s\n", counter->name); - - if (close_counters) { - perf_evsel__close_fd(counter, perf_evsel__nr_cpus(counter), - thread_map__nr(evsel_list->threads)); - } } } @@ -353,7 +348,7 @@ static void process_interval(void) { struct timespec ts, rs; - read_counters(false); + read_counters(); clock_gettime(CLOCK_MONOTONIC, &ts); diff_timespec(&rs, &ts, &ref_time); @@ -380,6 +375,17 @@ static void enable_counters(void) perf_evlist__enable(evsel_list); } +static void disable_counters(void) +{ + /* + * If we don't have tracee (attaching to task or cpu), counters may + * still be running. To get accurate group ratios, we must stop groups + * from counting before reading their constituent counters. + */ + if (!target__none(&target)) + perf_evlist__disable(evsel_list); +} + static volatile int workload_exec_errno; /* @@ -657,11 +663,20 @@ static int __run_perf_stat(int argc, const char **argv) } } + disable_counters(); + t1 = rdclock(); update_stats(&walltime_nsecs_stats, t1 - t0); - read_counters(true); + /* + * Closing a group leader splits the group, and as we only disable + * group leaders, results in remaining events becoming enabled. To + * avoid arbitrary skew, we must read all counters before closing any + * group leaders. + */ + read_counters(); + perf_evlist__close(evsel_list); return WEXITSTATUS(status); } From c87edb36118664f1fa275107c1138b6f47793240 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Fri, 5 Aug 2016 12:41:52 -0400 Subject: [PATCH 116/235] tracing: Fix tick_stop tracepoint symbols for user export The symbols used in the tick_stop tracepoint were not being converted properly into integers in the trace_stop format file. Instead we had this: print fmt: "success=%d dependency=%s", REC->success, __print_symbolic(REC->dependency, { 0, "NONE" }, { (1 << TICK_DEP_BIT_POSIX_TIMER), "POSIX_TIMER" }, { (1 << TICK_DEP_BIT_PERF_EVENTS), "PERF_EVENTS" }, { (1 << TICK_DEP_BIT_SCHED), "SCHED" }, { (1 << TICK_DEP_BIT_CLOCK_UNSTABLE), "CLOCK_UNSTABLE" }) User space tools have no idea how to parse "TICK_DEP_BIT_SCHED" or the other symbols used to do the bit shifting. The reason is that the conversion was done with using the TICK_DEP_MASK_* symbols which are just macros that convert to the BIT shift itself (with the exception of NONE, which was converted properly, because it doesn't use bits, and is defined as zero). The TICK_DEP_BIT_* needs to be denoted by TRACE_DEFINE_ENUM() in order to have this properly converted for user space tools to parse this event. Cc: stable@vger.kernel.org Cc: Frederic Weisbecker Fixes: e6e6cc22e067 ("nohz: Use enum code for tick stop failure tracing message") Reported-by: Luiz Capitulino Tested-by: Luiz Capitulino Signed-off-by: Steven Rostedt --- include/trace/events/timer.h | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/include/trace/events/timer.h b/include/trace/events/timer.h index 51440131d337..28c5da6fdfac 100644 --- a/include/trace/events/timer.h +++ b/include/trace/events/timer.h @@ -330,24 +330,32 @@ TRACE_EVENT(itimer_expire, #ifdef CONFIG_NO_HZ_COMMON #define TICK_DEP_NAMES \ - tick_dep_name(NONE) \ + tick_dep_mask_name(NONE) \ tick_dep_name(POSIX_TIMER) \ tick_dep_name(PERF_EVENTS) \ tick_dep_name(SCHED) \ tick_dep_name_end(CLOCK_UNSTABLE) #undef tick_dep_name +#undef tick_dep_mask_name #undef tick_dep_name_end -#define tick_dep_name(sdep) TRACE_DEFINE_ENUM(TICK_DEP_MASK_##sdep); -#define tick_dep_name_end(sdep) TRACE_DEFINE_ENUM(TICK_DEP_MASK_##sdep); +/* The MASK will convert to their bits and they need to be processed too */ +#define tick_dep_name(sdep) TRACE_DEFINE_ENUM(TICK_DEP_BIT_##sdep); \ + TRACE_DEFINE_ENUM(TICK_DEP_MASK_##sdep); +#define tick_dep_name_end(sdep) TRACE_DEFINE_ENUM(TICK_DEP_BIT_##sdep); \ + TRACE_DEFINE_ENUM(TICK_DEP_MASK_##sdep); +/* NONE only has a mask defined for it */ +#define tick_dep_mask_name(sdep) TRACE_DEFINE_ENUM(TICK_DEP_MASK_##sdep); TICK_DEP_NAMES #undef tick_dep_name +#undef tick_dep_mask_name #undef tick_dep_name_end #define tick_dep_name(sdep) { TICK_DEP_MASK_##sdep, #sdep }, +#define tick_dep_mask_name(sdep) { TICK_DEP_MASK_##sdep, #sdep }, #define tick_dep_name_end(sdep) { TICK_DEP_MASK_##sdep, #sdep } #define show_tick_dep_name(val) \ From 19f00b011729417f69e4df53cc3fe5ecc25134a4 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Tue, 9 Aug 2016 11:40:08 +0900 Subject: [PATCH 117/235] perf probe: Support signedness casting The 'perf probe' tool detects a variable's type and use the detected type to add a new probe. Then, kprobes prints its variable in hexadecimal format if the variable is unsigned and prints in decimal if it is signed. We sometimes want to see unsigned variable in decimal format (i.e. sector_t or size_t). In that case, we need to investigate the variable's size manually to specify just signedness. This patch add signedness casting support. By specifying "s" or "u" as a type, perf-probe will investigate variable size as usual and use the specified signedness. E.g. without this: $ perf probe -a 'submit_bio bio->bi_iter.bi_sector' Added new event: probe:submit_bio (on submit_bio with bi_sector=bio->bi_iter.bi_sector) You can now use it in all perf tools, such as: perf record -e probe:submit_bio -aR sleep 1 $ cat trace_pipe|head dbench-9692 [003] d..1 971.096633: submit_bio: (submit_bio+0x0/0x140) bi_sector=0x3a3d00 dbench-9692 [003] d..1 971.096685: submit_bio: (submit_bio+0x0/0x140) bi_sector=0x1a3d80 dbench-9692 [003] d..1 971.096687: submit_bio: (submit_bio+0x0/0x140) bi_sector=0x3a3d80 ... // need to investigate the variable size $ perf probe -a 'submit_bio bio->bi_iter.bi_sector:s64' Added new event: probe:submit_bio (on submit_bio with bi_sector=bio->bi_iter.bi_sector:s64) You can now use it in all perf tools, such as: perf record -e probe:submit_bio -aR sleep 1 With this: // just use "s" to cast its signedness $ perf probe -v -a 'submit_bio bio->bi_iter.bi_sector:s' Added new event: probe:submit_bio (on submit_bio with bi_sector=bio->bi_iter.bi_sector:s) You can now use it in all perf tools, such as: perf record -e probe:submit_bio -aR sleep 1 $ cat trace_pipe|head dbench-9689 [001] d..1 1212.391237: submit_bio: (submit_bio+0x0/0x140) bi_sector=128 dbench-9689 [001] d..1 1212.391252: submit_bio: (submit_bio+0x0/0x140) bi_sector=131072 dbench-9697 [006] d..1 1212.398611: submit_bio: (submit_bio+0x0/0x140) bi_sector=30208 This commit also update perf-probe.txt to describe "types". Most parts are based on existing documentation: Documentation/trace/kprobetrace.txt Committer note: Testing using 'perf trace': # perf probe -a 'submit_bio bio->bi_iter.bi_sector' Added new event: probe:submit_bio (on submit_bio with bi_sector=bio->bi_iter.bi_sector) You can now use it in all perf tools, such as: perf record -e probe:submit_bio -aR sleep 1 # trace --no-syscalls --ev probe:submit_bio 0.000 probe:submit_bio:(ffffffffac3aee00) bi_sector=0xc133c0) 3181.861 probe:submit_bio:(ffffffffac3aee00) bi_sector=0x6cffb8) 3181.881 probe:submit_bio:(ffffffffac3aee00) bi_sector=0x6cffc0) 3184.488 probe:submit_bio:(ffffffffac3aee00) bi_sector=0x6cffc8) 4717.927 probe:submit_bio:(ffffffffac3aee00) bi_sector=0x4dc7a88) 4717.970 probe:submit_bio:(ffffffffac3aee00) bi_sector=0x4dc7880) ^C[root@jouet ~]# Now, using this new feature: [root@jouet ~]# perf probe -a 'submit_bio bio->bi_iter.bi_sector:s' Added new event: probe:submit_bio (on submit_bio with bi_sector=bio->bi_iter.bi_sector:s) You can now use it in all perf tools, such as: perf record -e probe:submit_bio -aR sleep 1 [root@jouet ~]# trace --no-syscalls --ev probe:submit_bio 0.000 probe:submit_bio:(ffffffffac3aee00) bi_sector=7145704) 0.017 probe:submit_bio:(ffffffffac3aee00) bi_sector=7145712) 0.019 probe:submit_bio:(ffffffffac3aee00) bi_sector=7145720) 2.567 probe:submit_bio:(ffffffffac3aee00) bi_sector=7145728) 5631.919 probe:submit_bio:(ffffffffac3aee00) bi_sector=0) 5631.941 probe:submit_bio:(ffffffffac3aee00) bi_sector=8) 5631.945 probe:submit_bio:(ffffffffac3aee00) bi_sector=16) 5631.948 probe:submit_bio:(ffffffffac3aee00) bi_sector=24) ^C# With callchains: # trace --no-syscalls --ev probe:submit_bio/max-stack=10/ 0.000 probe:submit_bio:(ffffffffac3aee00) bi_sector=50662544) submit_bio+0xa8200001 ([kernel.kallsyms]) submit_bh+0xa8200013 ([kernel.kallsyms]) jbd2_journal_commit_transaction+0xa8200691 ([kernel.kallsyms]) kjournald2+0xa82000ca ([kernel.kallsyms]) kthread+0xa82000d8 ([kernel.kallsyms]) ret_from_fork+0xa820001f ([kernel.kallsyms]) 0.023 probe:submit_bio:(ffffffffac3aee00) bi_sector=50662552) submit_bio+0xa8200001 ([kernel.kallsyms]) submit_bh+0xa8200013 ([kernel.kallsyms]) jbd2_journal_commit_transaction+0xa8200691 ([kernel.kallsyms]) kjournald2+0xa82000ca ([kernel.kallsyms]) kthread+0xa82000d8 ([kernel.kallsyms]) ret_from_fork+0xa820001f ([kernel.kallsyms]) 0.027 probe:submit_bio:(ffffffffac3aee00) bi_sector=50662560) submit_bio+0xa8200001 ([kernel.kallsyms]) submit_bh+0xa8200013 ([kernel.kallsyms]) jbd2_journal_commit_transaction+0xa8200691 ([kernel.kallsyms]) kjournald2+0xa82000ca ([kernel.kallsyms]) kthread+0xa82000d8 ([kernel.kallsyms]) ret_from_fork+0xa820001f ([kernel.kallsyms]) 2.593 probe:submit_bio:(ffffffffac3aee00) bi_sector=50662568) submit_bio+0xa8200001 ([kernel.kallsyms]) submit_bh+0xa8200013 ([kernel.kallsyms]) journal_submit_commit_record+0xa82001ac ([kernel.kallsyms]) jbd2_journal_commit_transaction+0xa82012e8 ([kernel.kallsyms]) kjournald2+0xa82000ca ([kernel.kallsyms]) kthread+0xa82000d8 ([kernel.kallsyms]) ret_from_fork+0xa820001f ([kernel.kallsyms]) ^C# Signed-off-by: Naohiro Aota Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Hemant Kumar Cc: Masami Hiramatsu Cc: Peter Zijlstra Cc: Wang Nan Link: http://lkml.kernel.org/r/1470710408-23515-1-git-send-email-naohiro.aota@hgst.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/perf-probe.txt | 10 +++++++++- tools/perf/util/probe-finder.c | 15 ++++++++++++--- 2 files changed, 21 insertions(+), 4 deletions(-) diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt index 736da44596e4..b303bcdd8ed1 100644 --- a/tools/perf/Documentation/perf-probe.txt +++ b/tools/perf/Documentation/perf-probe.txt @@ -176,10 +176,18 @@ Each probe argument follows below syntax. 'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.) '$vars' and '$params' special arguments are also available for NAME, '$vars' is expanded to the local variables (including function parameters) which can access at given probe point. '$params' is expanded to only the function parameters. -'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type. +'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. Currently, basic types (u8/u16/u32/u64/s8/s16/s32/s64), signedness casting (u/s), "string" and bitfield are supported. (see TYPES for detail) On x86 systems %REG is always the short form of the register: for example %AX. %RAX or %EAX is not valid. +TYPES +----- +Basic types (u8/u16/u32/u64/s8/s16/s32/s64) are integer types. Prefix 's' and 'u' means those types are signed and unsigned respectively. Traced arguments are shown in decimal (signed) or hex (unsigned). You can also use 's' or 'u' to specify only signedness and leave its size auto-detected by perf probe. +String type is a special type, which fetches a "null-terminated" string from kernel space. This means it will fail and store NULL if the string container has been paged out. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type. +Bitfield is another special type, which takes 3 parameters, bit-width, bit-offset, and container-size (usually 32). The syntax is; + + b@/ + LINE SYNTAX ----------- Line range is described by following syntax. diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index f2d9ff064e2d..5c290c682afe 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -297,10 +297,13 @@ static int convert_variable_type(Dwarf_Die *vr_die, char sbuf[STRERR_BUFSIZE]; int bsize, boffs, total; int ret; + char sign; /* TODO: check all types */ - if (cast && strcmp(cast, "string") != 0) { + if (cast && strcmp(cast, "string") != 0 && + strcmp(cast, "s") != 0 && strcmp(cast, "u") != 0) { /* Non string type is OK */ + /* and respect signedness cast */ tvar->type = strdup(cast); return (tvar->type == NULL) ? -ENOMEM : 0; } @@ -361,6 +364,13 @@ static int convert_variable_type(Dwarf_Die *vr_die, return (tvar->type == NULL) ? -ENOMEM : 0; } + if (cast && (strcmp(cast, "u") == 0)) + sign = 'u'; + else if (cast && (strcmp(cast, "s") == 0)) + sign = 's'; + else + sign = die_is_signed_type(&type) ? 's' : 'u'; + ret = dwarf_bytesize(&type); if (ret <= 0) /* No size ... try to use default type */ @@ -373,8 +383,7 @@ static int convert_variable_type(Dwarf_Die *vr_die, dwarf_diename(&type), MAX_BASIC_TYPE_BITS); ret = MAX_BASIC_TYPE_BITS; } - ret = snprintf(buf, 16, "%c%d", - die_is_signed_type(&type) ? 's' : 'u', ret); + ret = snprintf(buf, 16, "%c%d", sign, ret); formatted: if (ret < 0 || ret >= 16) { From bebfb730125863aac74607dff24f1bdb9fb02a90 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 9 Aug 2016 11:21:57 -0300 Subject: [PATCH 118/235] tools: Sync cpufeatures.h and vmx.h with the kernel There were changes related to the deprecation of the "pcommit" instruction: fd1d961dd681 ("x86/insn: remove pcommit") dfa169bbee00 ("Revert "KVM: x86: add pcommit support"") No need to update anything in the tools, as "pcommit" wasn't being listed on the VMX_EXIT_REASONS in the tools/perf/arch/x86/util/kvm-stat.c file. Just grab fresh copies of these files to silence the file cache coherency detector: $ make -C tools/perf O=/tmp/build/perf install-bin make: Entering directory '/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j4' parallel build Warning: tools/arch/x86/include/asm/cpufeatures.h differs from kernel Warning: tools/arch/x86/include/uapi/asm/vmx.h differs from kernel INSTALL GTK UI # Cc: Adrian Hunter Cc: Dan Williams Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Paolo Bonzini Cc: Ross Zwisler Cc: Wang Nan Cc: Xiao Guangrong Link: http://lkml.kernel.org/n/tip-07pmcc1ysydhyyxbmp1vt0l4@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/asm/cpufeatures.h | 9 +++------ tools/arch/x86/include/uapi/asm/vmx.h | 4 +--- 2 files changed, 4 insertions(+), 9 deletions(-) diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h index 4a413485f9eb..92a8308b96f6 100644 --- a/tools/arch/x86/include/asm/cpufeatures.h +++ b/tools/arch/x86/include/asm/cpufeatures.h @@ -225,7 +225,6 @@ #define X86_FEATURE_RDSEED ( 9*32+18) /* The RDSEED instruction */ #define X86_FEATURE_ADX ( 9*32+19) /* The ADCX and ADOX instructions */ #define X86_FEATURE_SMAP ( 9*32+20) /* Supervisor Mode Access Prevention */ -#define X86_FEATURE_PCOMMIT ( 9*32+22) /* PCOMMIT instruction */ #define X86_FEATURE_CLFLUSHOPT ( 9*32+23) /* CLFLUSHOPT instruction */ #define X86_FEATURE_CLWB ( 9*32+24) /* CLWB instruction */ #define X86_FEATURE_AVX512PF ( 9*32+26) /* AVX-512 Prefetch */ @@ -301,10 +300,6 @@ #define X86_BUG_FXSAVE_LEAK X86_BUG(6) /* FXSAVE leaks FOP/FIP/FOP */ #define X86_BUG_CLFLUSH_MONITOR X86_BUG(7) /* AAI65, CLFLUSH required before MONITOR */ #define X86_BUG_SYSRET_SS_ATTRS X86_BUG(8) /* SYSRET doesn't fix up SS attrs */ -#define X86_BUG_NULL_SEG X86_BUG(9) /* Nulling a selector preserves the base */ -#define X86_BUG_SWAPGS_FENCE X86_BUG(10) /* SWAPGS without input dep on GS */ - - #ifdef CONFIG_X86_32 /* * 64-bit kernels don't use X86_BUG_ESPFIX. Make the define conditional @@ -312,5 +307,7 @@ */ #define X86_BUG_ESPFIX X86_BUG(9) /* "" IRET to 16-bit SS corrupts ESP/RSP high bits */ #endif - +#define X86_BUG_NULL_SEG X86_BUG(10) /* Nulling a selector preserves the base */ +#define X86_BUG_SWAPGS_FENCE X86_BUG(11) /* SWAPGS without input dep on GS */ +#define X86_BUG_MONITOR X86_BUG(12) /* IPI required to wake up remote CPU */ #endif /* _ASM_X86_CPUFEATURES_H */ diff --git a/tools/arch/x86/include/uapi/asm/vmx.h b/tools/arch/x86/include/uapi/asm/vmx.h index 5b15d94a33f8..37fee272618f 100644 --- a/tools/arch/x86/include/uapi/asm/vmx.h +++ b/tools/arch/x86/include/uapi/asm/vmx.h @@ -78,7 +78,6 @@ #define EXIT_REASON_PML_FULL 62 #define EXIT_REASON_XSAVES 63 #define EXIT_REASON_XRSTORS 64 -#define EXIT_REASON_PCOMMIT 65 #define VMX_EXIT_REASONS \ { EXIT_REASON_EXCEPTION_NMI, "EXCEPTION_NMI" }, \ @@ -127,8 +126,7 @@ { EXIT_REASON_INVVPID, "INVVPID" }, \ { EXIT_REASON_INVPCID, "INVPCID" }, \ { EXIT_REASON_XSAVES, "XSAVES" }, \ - { EXIT_REASON_XRSTORS, "XRSTORS" }, \ - { EXIT_REASON_PCOMMIT, "PCOMMIT" } + { EXIT_REASON_XRSTORS, "XRSTORS" } #define VMX_ABORT_SAVE_GUEST_MSR_FAIL 1 #define VMX_ABORT_LOAD_HOST_MSR_FAIL 4 From 791cceb89f7987c0375ff2e8971928a47f62ccae Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 9 Aug 2016 11:48:07 -0300 Subject: [PATCH 119/235] toops: Sync tools/include/uapi/linux/bpf.h with the kernel The way we're using kernel headers in tools/ now, with a copy that is made to the same path prefixed by "tools/" plus checking if that copy got stale, i.e. if the kernel counterpart changed, helps in keeping track with new features that may be useful for tools to exploit. For instance, looking at all the changes to bpf.h since it was last copied to tools/include brings this to toolers' attention: Need to investigate this one to check how to run a program via perf, setting up a BPF event, that will take advantage of the way perf already calls clang/LLVM, sets up the event and runs the workload in a single command line, helping in debugging such semi cooperative programs: 96ae52279594 ("bpf: Add bpf_probe_write_user BPF helper to be called in tracers") This one needs further investigation about using the feature it improves in 'perf trace' to do some tcpdumpin' mixed with syscalls, tracepoints, probe points, callgraphs, etc: 555c8a8623a3 ("bpf: avoid stack copy and use skb ctx for event output") Add tracing just packets that are related to some container to that mix: 4a482f34afcc ("cgroup: bpf: Add bpf_skb_in_cgroup_proto") 4ed8ec521ed5 ("cgroup: bpf: Add BPF_MAP_TYPE_CGROUP_ARRAY") Definetely needs to have example programs accessing task_struct from a bpf proggie started from 'perf trace': 606274c5abd8 ("bpf: introduce bpf_get_current_task() helper") Core networking related, XDP: 6ce96ca348a9 ("bpf: add XDP_TX xdp_action for direct forwarding") 6a773a15a1e8 ("bpf: add XDP prog type for early driver filter") 13c5c240f789 ("bpf: add bpf_get_hash_recalc helper") d2485c4242a8 ("bpf: add bpf_skb_change_type helper") 6578171a7ff0 ("bpf: add bpf_skb_change_proto helper") Changes detected by the tools build system: $ make -C tools/perf O=/tmp/build/perf install-bin make: Entering directory '/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j4' parallel build Warning: tools/include/uapi/linux/bpf.h differs from kernel INSTALL GTK UI CC /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o $ Cc: Adrian Hunter Cc: Alexei Starovoitov Cc: Brenden Blanco Cc: Daniel Borkmann Cc: David Ahern Cc: David S. Miller Cc: Jiri Olsa Cc: Martin KaFai Lau Cc: Namhyung Kim Cc: Sargun Dhillon Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-difq4ts1xvww6eyfs9e7zlft@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/bpf.h | 86 +++++++++++++++++++++++++++++++++- 1 file changed, 85 insertions(+), 1 deletion(-) diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 406459b935a2..da218fec6056 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -84,6 +84,7 @@ enum bpf_map_type { BPF_MAP_TYPE_PERCPU_HASH, BPF_MAP_TYPE_PERCPU_ARRAY, BPF_MAP_TYPE_STACK_TRACE, + BPF_MAP_TYPE_CGROUP_ARRAY, }; enum bpf_prog_type { @@ -93,6 +94,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_SCHED_CLS, BPF_PROG_TYPE_SCHED_ACT, BPF_PROG_TYPE_TRACEPOINT, + BPF_PROG_TYPE_XDP, }; #define BPF_PSEUDO_MAP_FD 1 @@ -313,6 +315,66 @@ enum bpf_func_id { */ BPF_FUNC_skb_get_tunnel_opt, BPF_FUNC_skb_set_tunnel_opt, + + /** + * bpf_skb_change_proto(skb, proto, flags) + * Change protocol of the skb. Currently supported is + * v4 -> v6, v6 -> v4 transitions. The helper will also + * resize the skb. eBPF program is expected to fill the + * new headers via skb_store_bytes and lX_csum_replace. + * @skb: pointer to skb + * @proto: new skb->protocol type + * @flags: reserved + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_proto, + + /** + * bpf_skb_change_type(skb, type) + * Change packet type of skb. + * @skb: pointer to skb + * @type: new skb->pkt_type type + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_type, + + /** + * bpf_skb_in_cgroup(skb, map, index) - Check cgroup2 membership of skb + * @skb: pointer to skb + * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type + * @index: index of the cgroup in the bpf_map + * Return: + * == 0 skb failed the cgroup2 descendant test + * == 1 skb succeeded the cgroup2 descendant test + * < 0 error + */ + BPF_FUNC_skb_in_cgroup, + + /** + * bpf_get_hash_recalc(skb) + * Retrieve and possibly recalculate skb->hash. + * @skb: pointer to skb + * Return: hash + */ + BPF_FUNC_get_hash_recalc, + + /** + * u64 bpf_get_current_task(void) + * Returns current task_struct + * Return: current + */ + BPF_FUNC_get_current_task, + + /** + * bpf_probe_write_user(void *dst, void *src, int len) + * safely attempt to write to a location + * @dst: destination address in userspace + * @src: source address on stack + * @len: number of bytes to copy + * Return: 0 on success or negative error + */ + BPF_FUNC_probe_write_user, + __BPF_FUNC_MAX_ID, }; @@ -347,9 +409,11 @@ enum bpf_func_id { #define BPF_F_ZERO_CSUM_TX (1ULL << 1) #define BPF_F_DONT_FRAGMENT (1ULL << 2) -/* BPF_FUNC_perf_event_output flags. */ +/* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK +/* BPF_FUNC_perf_event_output for sk_buff input context. */ +#define BPF_F_CTXLEN_MASK (0xfffffULL << 32) /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure @@ -386,4 +450,24 @@ struct bpf_tunnel_key { __u32 tunnel_label; }; +/* User return codes for XDP prog type. + * A valid XDP program must return one of these defined values. All other + * return codes are reserved for future use. Unknown return codes will result + * in packet drop. + */ +enum xdp_action { + XDP_ABORTED = 0, + XDP_DROP, + XDP_PASS, + XDP_TX, +}; + +/* user accessible metadata for XDP packet hook + * new fields must be added to the end of this structure + */ +struct xdp_md { + __u32 data; + __u32 data_end; +}; + #endif /* _UAPI__LINUX_BPF_H__ */ From 840b49ba554b40fa8301ad2716abd2fe3d9e382a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 9 Aug 2016 11:56:33 -0300 Subject: [PATCH 120/235] tools: Sync cpufeatures headers with the kernel Due to: 1e61f78baf89 ("x86/cpufeature: Make sure DISABLED/REQUIRED macros are updated") No changes to tools using those headers (tools/arch/x86/lib/mem{set,cpu}_64.S) seems necessary. Detected by the tools build header drift checker: $ make -C tools/perf O=/tmp/build/perf make: Entering directory '/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j4' parallel build GEN /tmp/build/perf/common-cmds.h Warning: tools/arch/x86/include/asm/disabled-features.h differs from kernel Warning: tools/arch/x86/include/asm/required-features.h differs from kernel Warning: tools/arch/x86/include/asm/cpufeatures.h differs from kernel CC /tmp/build/perf/util/probe-finder.o CC /tmp/build/perf/builtin-help.o ^C$ Cc: Adrian Hunter Cc: Dave Hansen Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-ja75m7zk8j0jkzmrv16i5ehw@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/asm/disabled-features.h | 2 ++ tools/arch/x86/include/asm/required-features.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/tools/arch/x86/include/asm/disabled-features.h b/tools/arch/x86/include/asm/disabled-features.h index 911e9358ceb1..85599ad4d024 100644 --- a/tools/arch/x86/include/asm/disabled-features.h +++ b/tools/arch/x86/include/asm/disabled-features.h @@ -56,5 +56,7 @@ #define DISABLED_MASK14 0 #define DISABLED_MASK15 0 #define DISABLED_MASK16 (DISABLE_PKU|DISABLE_OSPKE) +#define DISABLED_MASK17 0 +#define DISABLED_MASK_CHECK BUILD_BUG_ON_ZERO(NCAPINTS != 18) #endif /* _ASM_X86_DISABLED_FEATURES_H */ diff --git a/tools/arch/x86/include/asm/required-features.h b/tools/arch/x86/include/asm/required-features.h index 4916144e3c42..fac9a5c0abe9 100644 --- a/tools/arch/x86/include/asm/required-features.h +++ b/tools/arch/x86/include/asm/required-features.h @@ -99,5 +99,7 @@ #define REQUIRED_MASK14 0 #define REQUIRED_MASK15 0 #define REQUIRED_MASK16 0 +#define REQUIRED_MASK17 0 +#define REQUIRED_MASK_CHECK BUILD_BUG_ON_ZERO(NCAPINTS != 18) #endif /* _ASM_X86_REQUIRED_FEATURES_H */ From d820456dc70b231d62171ba46b43db0045e9bd57 Mon Sep 17 00:00:00 2001 From: Ravi Bangoria Date: Tue, 9 Aug 2016 01:23:24 -0500 Subject: [PATCH 121/235] perf probe: Add function to post process kernel trace events Instead of inline code, introduce function to post process kernel probe trace events. Signed-off-by: Ravi Bangoria Acked-by: Masami Hiramatsu Cc: Alexander Shishkin Cc: Ananth N Mavinakayanahalli Cc: Balbir Singh Cc: Namhyung Kim Cc: Naveen N. Rao Cc: Peter Zijlstra Cc: Wang Nan Link: http://lkml.kernel.org/r/1470723805-5081-1-git-send-email-ravi.bangoria@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/probe-event.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 1201f73ca723..234fbfb5c2ed 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -666,22 +666,14 @@ static int add_module_to_probe_trace_events(struct probe_trace_event *tevs, return ret; } -/* Post processing the probe events */ -static int post_process_probe_trace_events(struct probe_trace_event *tevs, - int ntevs, const char *module, - bool uprobe) +static int +post_process_kernel_probe_trace_events(struct probe_trace_event *tevs, + int ntevs) { struct ref_reloc_sym *reloc_sym; char *tmp; int i, skipped = 0; - if (uprobe) - return add_exec_to_probe_trace_events(tevs, ntevs, module); - - /* Note that currently ref_reloc_sym based probe is not for drivers */ - if (module) - return add_module_to_probe_trace_events(tevs, ntevs, module); - reloc_sym = kernel_get_ref_reloc_sym(); if (!reloc_sym) { pr_warning("Relocated base symbol is not found!\n"); @@ -713,6 +705,21 @@ static int post_process_probe_trace_events(struct probe_trace_event *tevs, return skipped; } +/* Post processing the probe events */ +static int post_process_probe_trace_events(struct probe_trace_event *tevs, + int ntevs, const char *module, + bool uprobe) +{ + if (uprobe) + return add_exec_to_probe_trace_events(tevs, ntevs, module); + + if (module) + /* Currently ref_reloc_sym based probe is not for drivers */ + return add_module_to_probe_trace_events(tevs, ntevs, module); + + return post_process_kernel_probe_trace_events(tevs, ntevs); +} + /* Try to find perf_probe_event with debuginfo */ static int try_to_find_probe_trace_events(struct perf_probe_event *pev, struct probe_trace_event **tevs) From 99e608b5954c9e1ebadbf9660b74697d9dfd9f20 Mon Sep 17 00:00:00 2001 From: Ravi Bangoria Date: Tue, 9 Aug 2016 01:23:25 -0500 Subject: [PATCH 122/235] perf probe ppc64le: Fix probe location when using DWARF Powerpc has Global Entry Point and Local Entry Point for functions. LEP catches call from both the GEP and the LEP. Symbol table of ELF contains GEP and Offset from which we can calculate LEP, but debuginfo does not have LEP info. Currently, perf prioritize symbol table over dwarf to probe on LEP for ppc64le. But when user tries to probe with function parameter, we fall back to using dwarf(i.e. GEP) and when function called via LEP, probe will never hit. For example: $ objdump -d vmlinux ... do_sys_open(): c0000000002eb4a0: e8 00 4c 3c addis r2,r12,232 c0000000002eb4a4: 60 00 42 38 addi r2,r2,96 c0000000002eb4a8: a6 02 08 7c mflr r0 c0000000002eb4ac: d0 ff 41 fb std r26,-48(r1) $ sudo ./perf probe do_sys_open $ sudo cat /sys/kernel/debug/tracing/kprobe_events p:probe/do_sys_open _text+3060904 $ sudo ./perf probe 'do_sys_open filename:string' $ sudo cat /sys/kernel/debug/tracing/kprobe_events p:probe/do_sys_open _text+3060896 filename_string=+0(%gpr4):string For second case, perf probed on GEP. So when function will be called via LEP, probe won't hit. $ sudo ./perf record -a -e probe:do_sys_open ls [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.195 MB perf.data ] To resolve this issue, let's not prioritize symbol table, let perf decide what it wants to use. Perf is already converting GEP to LEP when it uses symbol table. When perf uses debuginfo, let it find LEP offset form symbol table. This way we fall back to probe on LEP for all cases. After patch: $ sudo ./perf probe 'do_sys_open filename:string' $ sudo cat /sys/kernel/debug/tracing/kprobe_events p:probe/do_sys_open _text+3060904 filename_string=+0(%gpr4):string $ sudo ./perf record -a -e probe:do_sys_open ls [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.197 MB perf.data (11 samples) ] Signed-off-by: Ravi Bangoria Acked-by: Masami Hiramatsu Cc: Alexander Shishkin Cc: Ananth N Mavinakayanahalli Cc: Balbir Singh Cc: Namhyung Kim Cc: Naveen N. Rao Cc: Peter Zijlstra Cc: Wang Nan Link: http://lkml.kernel.org/r/1470723805-5081-2-git-send-email-ravi.bangoria@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/arch/powerpc/util/sym-handling.c | 27 ++++++++++++--- tools/perf/util/probe-event.c | 37 ++++++++++++--------- tools/perf/util/probe-event.h | 6 +++- 3 files changed, 49 insertions(+), 21 deletions(-) diff --git a/tools/perf/arch/powerpc/util/sym-handling.c b/tools/perf/arch/powerpc/util/sym-handling.c index c6d0f91731a1..8d4dc97d80ba 100644 --- a/tools/perf/arch/powerpc/util/sym-handling.c +++ b/tools/perf/arch/powerpc/util/sym-handling.c @@ -54,10 +54,6 @@ int arch__compare_symbol_names(const char *namea, const char *nameb) #endif #if defined(_CALL_ELF) && _CALL_ELF == 2 -bool arch__prefers_symtab(void) -{ - return true; -} #ifdef HAVE_LIBELF_SUPPORT void arch__sym_update(struct symbol *s, GElf_Sym *sym) @@ -100,4 +96,27 @@ void arch__fix_tev_from_maps(struct perf_probe_event *pev, tev->point.offset += lep_offset; } } + +void arch__post_process_probe_trace_events(struct perf_probe_event *pev, + int ntevs) +{ + struct probe_trace_event *tev; + struct map *map; + struct symbol *sym = NULL; + struct rb_node *tmp; + int i = 0; + + map = get_target_map(pev->target, pev->uprobes); + if (!map || map__load(map, NULL) < 0) + return; + + for (i = 0; i < ntevs; i++) { + tev = &pev->tevs[i]; + map__for_each_symbol(map, sym, tmp) { + if (map->unmap_ip(map, sym->start) == tev->point.address) + arch__fix_tev_from_maps(pev, tev, map, sym); + } + } +} + #endif diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 234fbfb5c2ed..28733962cd80 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -180,7 +180,7 @@ static struct map *kernel_get_module_map(const char *module) return NULL; } -static struct map *get_target_map(const char *target, bool user) +struct map *get_target_map(const char *target, bool user) { /* Init maps of given executable or kernel */ if (user) @@ -705,19 +705,32 @@ post_process_kernel_probe_trace_events(struct probe_trace_event *tevs, return skipped; } +void __weak +arch__post_process_probe_trace_events(struct perf_probe_event *pev __maybe_unused, + int ntevs __maybe_unused) +{ +} + /* Post processing the probe events */ -static int post_process_probe_trace_events(struct probe_trace_event *tevs, +static int post_process_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event *tevs, int ntevs, const char *module, bool uprobe) { + int ret; + if (uprobe) - return add_exec_to_probe_trace_events(tevs, ntevs, module); - - if (module) + ret = add_exec_to_probe_trace_events(tevs, ntevs, module); + else if (module) /* Currently ref_reloc_sym based probe is not for drivers */ - return add_module_to_probe_trace_events(tevs, ntevs, module); + ret = add_module_to_probe_trace_events(tevs, ntevs, module); + else + ret = post_process_kernel_probe_trace_events(tevs, ntevs); - return post_process_kernel_probe_trace_events(tevs, ntevs); + if (ret >= 0) + arch__post_process_probe_trace_events(pev, ntevs); + + return ret; } /* Try to find perf_probe_event with debuginfo */ @@ -758,7 +771,7 @@ static int try_to_find_probe_trace_events(struct perf_probe_event *pev, if (ntevs > 0) { /* Succeeded to find trace events */ pr_debug("Found %d probe_trace_events.\n", ntevs); - ret = post_process_probe_trace_events(*tevs, ntevs, + ret = post_process_probe_trace_events(pev, *tevs, ntevs, pev->target, pev->uprobes); if (ret < 0 || ret == ntevs) { clear_probe_trace_events(*tevs, ntevs); @@ -2945,8 +2958,6 @@ static int try_to_find_absolute_address(struct perf_probe_event *pev, return err; } -bool __weak arch__prefers_symtab(void) { return false; } - /* Concatinate two arrays */ static void *memcat(void *a, size_t sz_a, void *b, size_t sz_b) { @@ -3167,12 +3178,6 @@ static int convert_to_probe_trace_events(struct perf_probe_event *pev, if (ret > 0 || pev->sdt) /* SDT can be found only in the cache */ return ret == 0 ? -ENOENT : ret; /* Found in probe cache */ - if (arch__prefers_symtab() && !perf_probe_event_need_dwarf(pev)) { - ret = find_probe_trace_events_from_map(pev, tevs); - if (ret > 0) - return ret; /* Found in symbol table */ - } - /* Convert perf_probe_event with debuginfo */ ret = try_to_find_probe_trace_events(pev, tevs); if (ret != 0) diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h index e18ea9fe6385..f4f45db77c1c 100644 --- a/tools/perf/util/probe-event.h +++ b/tools/perf/util/probe-event.h @@ -158,7 +158,6 @@ int show_line_range(struct line_range *lr, const char *module, bool user); int show_available_vars(struct perf_probe_event *pevs, int npevs, struct strfilter *filter); int show_available_funcs(const char *module, struct strfilter *filter, bool user); -bool arch__prefers_symtab(void); void arch__fix_tev_from_maps(struct perf_probe_event *pev, struct probe_trace_event *tev, struct map *map, struct symbol *sym); @@ -173,4 +172,9 @@ int e_snprintf(char *str, size_t size, const char *format, ...) int copy_to_probe_trace_arg(struct probe_trace_arg *tvar, struct perf_probe_arg *pvar); +struct map *get_target_map(const char *target, bool user); + +void arch__post_process_probe_trace_events(struct perf_probe_event *pev, + int ntevs); + #endif /*_PROBE_EVENT_H */ From d8734849d8007dacaa40b31ba7319ed28077141d Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Mon, 8 Aug 2016 15:24:02 +0200 Subject: [PATCH 123/235] rbd: nuke the 32-bit pool id check ceph_file_layout::pool_id is now s64. rbd_add_get_pool_id() and ceph_pg_poolid_by_name() both return an int, so it's bogus anyway. Signed-off-by: Ilya Dryomov Reviewed-by: Alex Elder --- drivers/block/rbd.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 07668a6f0607..6c6519f6492a 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -5337,15 +5337,6 @@ static ssize_t do_rbd_add(struct bus_type *bus, } spec->pool_id = (u64)rc; - /* The ceph file layout needs to fit pool id in 32 bits */ - - if (spec->pool_id > (u64)U32_MAX) { - rbd_warn(NULL, "pool id too large (%llu > %u)", - (unsigned long long)spec->pool_id, U32_MAX); - rc = -EIO; - goto err_out_client; - } - rbd_dev = rbd_dev_create(rbdc, spec, rbd_opts); if (!rbd_dev) { rc = -ENOMEM; From 4eacd4cb3a7c4794688ef4fad5d01f3a532a58e9 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Tue, 9 Aug 2016 16:12:09 +0200 Subject: [PATCH 124/235] ceph: initialize pathbase in the !dentry case in encode_caps_cb() pathbase is the base inode; set it to 0 if we've got no path. Coverity-id: 146348 Signed-off-by: Ilya Dryomov Reviewed-by: Alex Elder --- fs/ceph/mds_client.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index fa59a85226b2..f72d4ae303b2 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2759,6 +2759,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, } else { path = NULL; pathlen = 0; + pathbase = 0; } spin_lock(&ci->i_ceph_lock); From a026bb12cc57d758e045126a252e12e868076cb4 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Wed, 3 Aug 2016 18:08:55 +0100 Subject: [PATCH 125/235] drivers/perf: arm-pmu: convert arm_pmu_mutex to spinlock arm_pmu_mutex is never held long and we don't want to sleep while the lock is being held as it's executed in the context of hotplug notifiers. So it can be converted to a simple spinlock instead. Without this patch we get the following warning: BUG: sleeping function called from invalid context at kernel/locking/mutex.c:620 in_atomic(): 1, irqs_disabled(): 128, pid: 0, name: swapper/2 no locks held by swapper/2/0. irq event stamp: 381314 hardirqs last enabled at (381313): _raw_spin_unlock_irqrestore+0x7c/0x88 hardirqs last disabled at (381314): cpu_die+0x28/0x48 softirqs last enabled at (381294): _local_bh_enable+0x28/0x50 softirqs last disabled at (381293): irq_enter+0x58/0x78 CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.7.0 #12 Call trace: dump_backtrace+0x0/0x220 show_stack+0x24/0x30 dump_stack+0xb4/0xf0 ___might_sleep+0x1d8/0x1f0 __might_sleep+0x5c/0x98 mutex_lock_nested+0x54/0x400 arm_perf_starting_cpu+0x34/0xb0 cpuhp_invoke_callback+0x88/0x3d8 notify_cpu_starting+0x78/0x98 secondary_start_kernel+0x108/0x1a8 This patch converts the mutex to spinlock to eliminate the above warnings. This constraints pmu->reset to be non-blocking call which is the case with all the ARM PMU backends. Cc: Stephen Boyd Fixes: 37b502f121ad ("arm/perf: Fix hotplug state machine conversion") Acked-by: Mark Rutland Signed-off-by: Sudeep Holla Signed-off-by: Will Deacon --- drivers/perf/arm_pmu.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c index 6ccb994bdfcb..4c9a537a1265 100644 --- a/drivers/perf/arm_pmu.c +++ b/drivers/perf/arm_pmu.c @@ -688,7 +688,7 @@ static int cpu_pmu_request_irq(struct arm_pmu *cpu_pmu, irq_handler_t handler) return 0; } -static DEFINE_MUTEX(arm_pmu_mutex); +static DEFINE_SPINLOCK(arm_pmu_lock); static LIST_HEAD(arm_pmu_list); /* @@ -701,7 +701,7 @@ static int arm_perf_starting_cpu(unsigned int cpu) { struct arm_pmu *pmu; - mutex_lock(&arm_pmu_mutex); + spin_lock(&arm_pmu_lock); list_for_each_entry(pmu, &arm_pmu_list, entry) { if (!cpumask_test_cpu(cpu, &pmu->supported_cpus)) @@ -709,7 +709,7 @@ static int arm_perf_starting_cpu(unsigned int cpu) if (pmu->reset) pmu->reset(pmu); } - mutex_unlock(&arm_pmu_mutex); + spin_unlock(&arm_pmu_lock); return 0; } @@ -821,9 +821,9 @@ static int cpu_pmu_init(struct arm_pmu *cpu_pmu) if (!cpu_hw_events) return -ENOMEM; - mutex_lock(&arm_pmu_mutex); + spin_lock(&arm_pmu_lock); list_add_tail(&cpu_pmu->entry, &arm_pmu_list); - mutex_unlock(&arm_pmu_mutex); + spin_unlock(&arm_pmu_lock); err = cpu_pm_pmu_register(cpu_pmu); if (err) @@ -859,9 +859,9 @@ static int cpu_pmu_init(struct arm_pmu *cpu_pmu) return 0; out_unregister: - mutex_lock(&arm_pmu_mutex); + spin_lock(&arm_pmu_lock); list_del(&cpu_pmu->entry); - mutex_unlock(&arm_pmu_mutex); + spin_unlock(&arm_pmu_lock); free_percpu(cpu_hw_events); return err; } @@ -869,9 +869,9 @@ static int cpu_pmu_init(struct arm_pmu *cpu_pmu) static void cpu_pmu_destroy(struct arm_pmu *cpu_pmu) { cpu_pm_pmu_unregister(cpu_pmu); - mutex_lock(&arm_pmu_mutex); + spin_lock(&arm_pmu_lock); list_del(&cpu_pmu->entry); - mutex_unlock(&arm_pmu_mutex); + spin_unlock(&arm_pmu_lock); free_percpu(cpu_pmu->hw_events); } From 7f1d642fbb5c356519617c24757a0cbed7f800a8 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 19 Jul 2016 15:39:02 +0100 Subject: [PATCH 126/235] drivers/perf: arm-pmu: Fix handling of SPI lacking "interrupt-affinity" property Patch 19a469a58720 ("drivers/perf: arm-pmu: Handle per-interrupt affinity mask") added support for partitionned PPI setups, but inadvertently broke setups using SPIs without the "interrupt-affinity" property (which is the case for UP platforms). This patch restore the broken functionnality by testing whether the interrupt is percpu or not instead of relying on the using_spi flag that really means "SPI *and* interrupt-affinity property". Acked-by: Mark Rutland Reported-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven Fixes: 19a469a58720 ("drivers/perf: arm-pmu: Handle per-interrupt affinity mask") Signed-off-by: Marc Zyngier Signed-off-by: Will Deacon --- drivers/perf/arm_pmu.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c index 4c9a537a1265..c494613c1909 100644 --- a/drivers/perf/arm_pmu.c +++ b/drivers/perf/arm_pmu.c @@ -967,11 +967,12 @@ static int of_pmu_irq_cfg(struct arm_pmu *pmu) /* If we didn't manage to parse anything, try the interrupt affinity */ if (cpumask_weight(&pmu->supported_cpus) == 0) { - if (!using_spi) { - /* If using PPIs, check the affinity of the partition */ - int ret, irq; + int irq = platform_get_irq(pdev, 0); + + if (irq_is_percpu(irq)) { + /* If using PPIs, check the affinity of the partition */ + int ret; - irq = platform_get_irq(pdev, 0); ret = irq_get_percpu_devid_partition(irq, &pmu->supported_cpus); if (ret) { kfree(irqs); From c4159a75b64c0e67caededf4d7372c1b58a5f42a Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Mon, 8 Aug 2016 23:03:12 +0300 Subject: [PATCH 127/235] mm: memcontrol: only mark charged pages with PageKmemcg To distinguish non-slab pages charged to kmemcg we mark them PageKmemcg, which sets page->_mapcount to -512. Currently, we set/clear PageKmemcg in __alloc_pages_nodemask()/free_pages_prepare() for any page allocated with __GFP_ACCOUNT, including those that aren't actually charged to any cgroup, i.e. allocated from the root cgroup context. To avoid overhead in case cgroups are not used, we only do that if memcg_kmem_enabled() is true. The latter is set iff there are kmem-enabled memory cgroups (online or offline). The root cgroup is not considered kmem-enabled. As a result, if a page is allocated with __GFP_ACCOUNT for the root cgroup when there are kmem-enabled memory cgroups and is freed after all kmem-enabled memory cgroups were removed, e.g. # no memory cgroups has been created yet, create one mkdir /sys/fs/cgroup/memory/test # run something allocating pages with __GFP_ACCOUNT, e.g. # a program using pipe dmesg | tail # remove the memory cgroup rmdir /sys/fs/cgroup/memory/test we'll get bad page state bug complaining about page->_mapcount != -1: BUG: Bad page state in process swapper/0 pfn:1fd945c page:ffffea007f651700 count:0 mapcount:-511 mapping: (null) index:0x0 flags: 0x1000000000000000() To avoid that, let's mark with PageKmemcg only those pages that are actually charged to and hence pin a non-root memory cgroup. Fixes: 4949148ad433 ("mm: charge/uncharge kmemcg from generic page allocator paths") Reported-and-tested-by: Eric Dumazet Signed-off-by: Vladimir Davydov Signed-off-by: Linus Torvalds --- fs/pipe.c | 4 +--- mm/memcontrol.c | 14 ++++++++++++-- mm/page_alloc.c | 14 +++++--------- 3 files changed, 18 insertions(+), 14 deletions(-) diff --git a/fs/pipe.c b/fs/pipe.c index 4b32928f5426..4ebe6b2e5217 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -144,10 +144,8 @@ static int anon_pipe_buf_steal(struct pipe_inode_info *pipe, struct page *page = buf->page; if (page_count(page) == 1) { - if (memcg_kmem_enabled()) { + if (memcg_kmem_enabled()) memcg_kmem_uncharge(page, 0); - __ClearPageKmemcg(page); - } __SetPageLocked(page); return 0; } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 66beca1ad92f..e74d7080ec9e 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2337,8 +2337,11 @@ int memcg_kmem_charge(struct page *page, gfp_t gfp, int order) return 0; memcg = get_mem_cgroup_from_mm(current->mm); - if (!mem_cgroup_is_root(memcg)) + if (!mem_cgroup_is_root(memcg)) { ret = memcg_kmem_charge_memcg(page, gfp, order, memcg); + if (!ret) + __SetPageKmemcg(page); + } css_put(&memcg->css); return ret; } @@ -2365,6 +2368,11 @@ void memcg_kmem_uncharge(struct page *page, int order) page_counter_uncharge(&memcg->memsw, nr_pages); page->mem_cgroup = NULL; + + /* slab pages do not have PageKmemcg flag set */ + if (PageKmemcg(page)) + __ClearPageKmemcg(page); + css_put_many(&memcg->css, nr_pages); } #endif /* !CONFIG_SLOB */ @@ -5537,8 +5545,10 @@ static void uncharge_list(struct list_head *page_list) else nr_file += nr_pages; pgpgout++; - } else + } else { nr_kmem += 1 << compound_order(page); + __ClearPageKmemcg(page); + } page->mem_cgroup = NULL; } while (next != page_list); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fb975cec3518..ee744fa3b93d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1008,10 +1008,8 @@ static __always_inline bool free_pages_prepare(struct page *page, } if (PageMappingFlags(page)) page->mapping = NULL; - if (memcg_kmem_enabled() && PageKmemcg(page)) { + if (memcg_kmem_enabled() && PageKmemcg(page)) memcg_kmem_uncharge(page, order); - __ClearPageKmemcg(page); - } if (check_free) bad += free_pages_check(page); if (bad) @@ -3756,12 +3754,10 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, } out: - if (memcg_kmem_enabled() && (gfp_mask & __GFP_ACCOUNT) && page) { - if (unlikely(memcg_kmem_charge(page, gfp_mask, order))) { - __free_pages(page, order); - page = NULL; - } else - __SetPageKmemcg(page); + if (memcg_kmem_enabled() && (gfp_mask & __GFP_ACCOUNT) && page && + unlikely(memcg_kmem_charge(page, gfp_mask, order) != 0)) { + __free_pages(page, order); + page = NULL; } if (kmemcheck_enabled && page) From a3d1ddd932bc86f0f7027079754ab0aefd259109 Mon Sep 17 00:00:00 2001 From: Brian King Date: Mon, 8 Aug 2016 17:53:12 -0500 Subject: [PATCH 128/235] ipr: Fix sync scsi scan Commit b195d5e2bffd ("ipr: Wait to do async scan until scsi host is initialized") fixed async scan for ipr, but broke sync scan for ipr. This fixes sync scan back up. Signed-off-by: Brian King Reported-and-tested-by: Michael Ellerman Signed-off-by: Linus Torvalds --- drivers/scsi/ipr.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index bf85974be862..17d04c702e1b 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -10410,8 +10410,11 @@ static int ipr_probe(struct pci_dev *pdev, const struct pci_device_id *dev_id) __ipr_remove(pdev); return rc; } + spin_lock_irqsave(ioa_cfg->host->host_lock, flags); + ioa_cfg->scan_enabled = 1; + schedule_work(&ioa_cfg->work_q); + spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags); - scsi_scan_host(ioa_cfg->host); ioa_cfg->iopoll_weight = ioa_cfg->chip_cfg->iopoll_weight; if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) { @@ -10421,10 +10424,8 @@ static int ipr_probe(struct pci_dev *pdev, const struct pci_device_id *dev_id) } } - spin_lock_irqsave(ioa_cfg->host->host_lock, flags); - ioa_cfg->scan_enabled = 1; - schedule_work(&ioa_cfg->work_q); - spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags); + scsi_scan_host(ioa_cfg->host); + return 0; } From a0cba2179ea4c1820fce2ee046b6ed90ecc56196 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 9 Aug 2016 10:48:18 -0700 Subject: [PATCH 129/235] Revert "printk: create pr_ functions" This reverts commit 874f9c7da9a4acbc1b9e12ca722579fb50e4d142. Geert Uytterhoeven reports: "This change seems to have an (unintendent?) side-effect. Before, pr_*() calls without a trailing newline characters would be printed with a newline character appended, both on the console and in the output of the dmesg command. After this commit, no new line character is appended, and the output of the next pr_*() call of the same type may be appended, like in: - Truncating RAM at 0x0000000040000000-0x00000000c0000000 to -0x0000000070000000 - Ignoring RAM at 0x0000000200000000-0x0000000240000000 (!CONFIG_HIGHMEM) + Truncating RAM at 0x0000000040000000-0x00000000c0000000 to -0x0000000070000000Ignoring RAM at 0x0000000200000000-0x0000000240000000 (!CONFIG_HIGHMEM)" Joe Perches says: "No, that is not intentional. The newline handling code inside vprintk_emit is a bit involved and for now I suggest a revert until this has all the same behavior as earlier" Reported-by: Geert Uytterhoeven Requested-by: Joe Perches Cc: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/printk.h | 48 +++++++++++++--------------------------- kernel/printk/internal.h | 16 +++++--------- kernel/printk/nmi.c | 13 ++--------- kernel/printk/printk.c | 25 +++------------------ 4 files changed, 26 insertions(+), 76 deletions(-) diff --git a/include/linux/printk.h b/include/linux/printk.h index 8dc155dab3ed..696a56be7d3e 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -266,39 +266,21 @@ extern asmlinkage void dump_stack(void) __cold; * and other debug macros are compiled out unless either DEBUG is defined * or CONFIG_DYNAMIC_DEBUG is set. */ - -#ifdef CONFIG_PRINTK - -asmlinkage __printf(1, 2) __cold void __pr_emerg(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_alert(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_crit(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_err(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_warn(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_notice(const char *fmt, ...); -asmlinkage __printf(1, 2) __cold void __pr_info(const char *fmt, ...); - -#define pr_emerg(fmt, ...) __pr_emerg(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_alert(fmt, ...) __pr_alert(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_crit(fmt, ...) __pr_crit(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_err(fmt, ...) __pr_err(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_warn(fmt, ...) __pr_warn(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_notice(fmt, ...) __pr_notice(pr_fmt(fmt), ##__VA_ARGS__) -#define pr_info(fmt, ...) __pr_info(pr_fmt(fmt), ##__VA_ARGS__) - -#else - -#define pr_emerg(fmt, ...) printk(KERN_EMERG pr_fmt(fmt), ##__VA_ARGS__) -#define pr_alert(fmt, ...) printk(KERN_ALERT pr_fmt(fmt), ##__VA_ARGS__) -#define pr_crit(fmt, ...) printk(KERN_CRIT pr_fmt(fmt), ##__VA_ARGS__) -#define pr_err(fmt, ...) printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__) -#define pr_warn(fmt, ...) printk(KERN_WARNING pr_fmt(fmt), ##__VA_ARGS__) -#define pr_notice(fmt, ...) printk(KERN_NOTICE pr_fmt(fmt), ##__VA_ARGS__) -#define pr_info(fmt, ...) printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__) - -#endif - -#define pr_warning pr_warn - +#define pr_emerg(fmt, ...) \ + printk(KERN_EMERG pr_fmt(fmt), ##__VA_ARGS__) +#define pr_alert(fmt, ...) \ + printk(KERN_ALERT pr_fmt(fmt), ##__VA_ARGS__) +#define pr_crit(fmt, ...) \ + printk(KERN_CRIT pr_fmt(fmt), ##__VA_ARGS__) +#define pr_err(fmt, ...) \ + printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__) +#define pr_warning(fmt, ...) \ + printk(KERN_WARNING pr_fmt(fmt), ##__VA_ARGS__) +#define pr_warn pr_warning +#define pr_notice(fmt, ...) \ + printk(KERN_NOTICE pr_fmt(fmt), ##__VA_ARGS__) +#define pr_info(fmt, ...) \ + printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__) /* * Like KERN_CONT, pr_cont() should only be used when continuing * a line with no newline ('\n') enclosed. Otherwise it defaults diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h index 5d4505f30083..7fd2838fa417 100644 --- a/kernel/printk/internal.h +++ b/kernel/printk/internal.h @@ -16,11 +16,9 @@ */ #include -typedef __printf(2, 0) int (*printk_func_t)(int level, const char *fmt, - va_list args); +typedef __printf(1, 0) int (*printk_func_t)(const char *fmt, va_list args); -__printf(2, 0) -int vprintk_default(int level, const char *fmt, va_list args); +int __printf(1, 0) vprintk_default(const char *fmt, va_list args); #ifdef CONFIG_PRINTK_NMI @@ -33,10 +31,9 @@ extern raw_spinlock_t logbuf_lock; * via per-CPU variable. */ DECLARE_PER_CPU(printk_func_t, printk_func); -__printf(2, 0) -static inline int vprintk_func(int level, const char *fmt, va_list args) +static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) { - return this_cpu_read(printk_func)(level, fmt, args); + return this_cpu_read(printk_func)(fmt, args); } extern atomic_t nmi_message_lost; @@ -47,10 +44,9 @@ static inline int get_nmi_message_lost(void) #else /* CONFIG_PRINTK_NMI */ -__printf(2, 0) -static inline int vprintk_func(int level, const char *fmt, va_list args) +static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) { - return vprintk_default(level, fmt, args); + return vprintk_default(fmt, args); } static inline int get_nmi_message_lost(void) diff --git a/kernel/printk/nmi.c b/kernel/printk/nmi.c index bc3eeb1ae6da..b69eb8a2876f 100644 --- a/kernel/printk/nmi.c +++ b/kernel/printk/nmi.c @@ -58,7 +58,7 @@ static DEFINE_PER_CPU(struct nmi_seq_buf, nmi_print_seq); * one writer running. But the buffer might get flushed from another * CPU, so we need to be careful. */ -static int vprintk_nmi(int level, const char *fmt, va_list args) +static int vprintk_nmi(const char *fmt, va_list args) { struct nmi_seq_buf *s = this_cpu_ptr(&nmi_print_seq); int add = 0; @@ -79,16 +79,7 @@ static int vprintk_nmi(int level, const char *fmt, va_list args) if (!len) smp_rmb(); - if (level != LOGLEVEL_DEFAULT) { - add = snprintf(s->buffer + len, sizeof(s->buffer) - len, - KERN_SOH "%c", '0' + level); - add += vsnprintf(s->buffer + len + add, - sizeof(s->buffer) - len - add, - fmt, args); - } else { - add = vsnprintf(s->buffer + len, sizeof(s->buffer) - len, - fmt, args); - } + add = vsnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, args); /* * Do it once again if the buffer has been flushed in the meantime. diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index a37fc8cf8e84..eea6dbc2d8cf 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -1930,26 +1930,7 @@ asmlinkage int printk_emit(int facility, int level, } EXPORT_SYMBOL(printk_emit); -#define define_pr_level(func, loglevel) \ -asmlinkage __visible void func(const char *fmt, ...) \ -{ \ - va_list args; \ - \ - va_start(args, fmt); \ - vprintk_default(loglevel, fmt, args); \ - va_end(args); \ -} \ -EXPORT_SYMBOL(func) - -define_pr_level(__pr_emerg, LOGLEVEL_EMERG); -define_pr_level(__pr_alert, LOGLEVEL_ALERT); -define_pr_level(__pr_crit, LOGLEVEL_CRIT); -define_pr_level(__pr_err, LOGLEVEL_ERR); -define_pr_level(__pr_warn, LOGLEVEL_WARNING); -define_pr_level(__pr_notice, LOGLEVEL_NOTICE); -define_pr_level(__pr_info, LOGLEVEL_INFO); - -int vprintk_default(int level, const char *fmt, va_list args) +int vprintk_default(const char *fmt, va_list args) { int r; @@ -1959,7 +1940,7 @@ int vprintk_default(int level, const char *fmt, va_list args) return r; } #endif - r = vprintk_emit(0, level, NULL, 0, fmt, args); + r = vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, 0, fmt, args); return r; } @@ -1992,7 +1973,7 @@ asmlinkage __visible int printk(const char *fmt, ...) int r; va_start(args, fmt); - r = vprintk_func(LOGLEVEL_DEFAULT, fmt, args); + r = vprintk_func(fmt, args); va_end(args); return r; From 51350ea0d7f355dfc03deb343a665802d3d5cbba Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Thu, 4 Aug 2016 21:36:05 +0300 Subject: [PATCH 130/235] mm, writeback: flush plugged IO in wakeup_flusher_threads() I've found funny live-lock between raid10 barriers during resync and memory controller hard limits. Inside mpage_readpages() task holds on to its plug bio which blocks the barrier in raid10. Its memory cgroup have no free memory thus the task goes into reclaimer but all reclaimable pages are dirty and cannot be written because raid10 is rebuilding and stuck on the barrier. Common flush of such IO in schedule() never happens, because the caller doesn't go to sleep. Lock is 'live' because changing memory limit or killing tasks which holds that stuck bio unblock whole progress. That was what happened in 3.18.x but I see no difference in upstream logic. Theoretically this might happen even without memory cgroup. Signed-off-by: Konstantin Khlebnikov Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 4d09d4441e3e..05713a5da083 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1949,6 +1949,12 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) { struct backing_dev_info *bdi; + /* + * If we are expecting writeback progress we must submit plugged IO. + */ + if (blk_needs_flush_plug(current)) + blk_schedule_flush_plug(current); + if (!nr_pages) nr_pages = get_nr_dirty_pages(); From c6d2ee09c2fffd3efdd31be2b2811d081a45bb99 Mon Sep 17 00:00:00 2001 From: Frederic Barrat Date: Mon, 8 Aug 2016 11:57:48 +0200 Subject: [PATCH 131/235] cxl: Set psl_fir_cntl to production environment value Switch the setting of psl_fir_cntl from debug to production environment recommended value. It mostly affects the PSL behavior when an error is raised in psl_fir1/2. Tested with cxlflash. Signed-off-by: Frederic Barrat Reviewed-by: Uma Krishnan Signed-off-by: Michael Ellerman --- drivers/misc/cxl/pci.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c index 1d0347c36e6d..6f0c4ac4b649 100644 --- a/drivers/misc/cxl/pci.c +++ b/drivers/misc/cxl/pci.c @@ -379,7 +379,7 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_dev *dev) { - u64 psl_dsnctl; + u64 psl_dsnctl, psl_fircntl; u64 chipid; u64 capp_unit_id; int rc; @@ -398,8 +398,11 @@ static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_ cxl_p1_write(adapter, CXL_PSL_RESLCKTO, 0x20000000200ULL); /* snoop write mask */ cxl_p1_write(adapter, CXL_PSL_SNWRALLOC, 0x00000000FFFFFFFFULL); - /* set fir_accum */ - cxl_p1_write(adapter, CXL_PSL_FIR_CNTL, 0x0800000000000000ULL); + /* set fir_cntl to recommended value for production env */ + psl_fircntl = (0x2ULL << (63-3)); /* ce_report */ + psl_fircntl |= (0x1ULL << (63-6)); /* FIR_report */ + psl_fircntl |= 0x1ULL; /* ce_thresh */ + cxl_p1_write(adapter, CXL_PSL_FIR_CNTL, psl_fircntl); /* for debugging with trace arrays */ cxl_p1_write(adapter, CXL_PSL_TRACE, 0x0000FF7C00000000ULL); From 1bc8b816cb8058c31f61fe78442f10a43209e582 Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Tue, 2 Aug 2016 10:07:05 +0200 Subject: [PATCH 132/235] powerpc/32: Fix csum_partial_copy_generic() Commit 7aef4136566b0 ("powerpc32: rewrite csum_partial_copy_generic() based on copy_tofrom_user()") introduced a bug when destination address is odd and initial csum is not null In that (rare) case the initial csum value has to be rotated one byte as well as the resulting value is This patch also fixes related comments Fixes: 7aef4136566b0 ("powerpc32: rewrite csum_partial_copy_generic() based on copy_tofrom_user()") Signed-off-by: Christophe Leroy Signed-off-by: Michael Ellerman --- arch/powerpc/lib/checksum_32.S | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/lib/checksum_32.S b/arch/powerpc/lib/checksum_32.S index d90870a66b60..0a57fe6d49cc 100644 --- a/arch/powerpc/lib/checksum_32.S +++ b/arch/powerpc/lib/checksum_32.S @@ -127,8 +127,9 @@ _GLOBAL(csum_partial_copy_generic) stw r7,12(r1) stw r8,8(r1) - andi. r0,r4,1 /* is destination address even ? */ - cmplwi cr7,r0,0 + rlwinm r0,r4,3,0x8 + rlwnm r6,r6,r0,0,31 /* odd destination address: rotate one byte */ + cmplwi cr7,r0,0 /* is destination address even ? */ addic r12,r6,0 addi r6,r4,-4 neg r0,r4 @@ -237,7 +238,7 @@ _GLOBAL(csum_partial_copy_generic) 66: addze r3,r12 addi r1,r1,16 beqlr+ cr7 - rlwinm r3,r3,8,0,31 /* swap bytes for odd destination */ + rlwinm r3,r3,8,0,31 /* odd destination address: rotate one byte */ blr /* read fault */ From c7a318ba868c61fc9be710a4970172d8c2eeb8b9 Mon Sep 17 00:00:00 2001 From: Cyril Bur Date: Wed, 10 Aug 2016 15:44:46 +1000 Subject: [PATCH 133/235] powerpc/ptrace: Fix coredump since ptrace TM changes Commit 8d460f6156cd ("powerpc/process: Add the function flush_tmregs_to_thread") added flush_tmregs_to_thread() and included the assumption that it would only be called for a task which is not current. Although this is correct for ptrace, when generating a core dump, some of the routines which call flush_tmregs_to_thread() are called. This leads to a WARNing such as: Not expecting ptrace on self: TM regs may be incorrect ------------[ cut here ]------------ WARNING: CPU: 123 PID: 7727 at arch/powerpc/kernel/process.c:1088 flush_tmregs_to_thread+0x78/0x80 CPU: 123 PID: 7727 Comm: libvirtd Not tainted 4.8.0-rc1-gcc6x-g61e8a0d #1 task: c000000fe631b600 task.stack: c000000fe63b0000 NIP: c00000000001a1a8 LR: c00000000001a1a4 CTR: c000000000717780 REGS: c000000fe63b3420 TRAP: 0700 Not tainted (4.8.0-rc1-gcc6x-g61e8a0d) MSR: 900000010282b033 CR: 28004222 XER: 20000000 ... NIP [c00000000001a1a8] flush_tmregs_to_thread+0x78/0x80 LR [c00000000001a1a4] flush_tmregs_to_thread+0x74/0x80 Call Trace: flush_tmregs_to_thread+0x74/0x80 (unreliable) vsr_get+0x64/0x1a0 elf_core_dump+0x604/0x1430 do_coredump+0x5fc/0x1200 get_signal+0x398/0x740 do_signal+0x54/0x2b0 do_notify_resume+0x98/0xb0 ret_from_except_lite+0x70/0x74 So fix flush_tmregs_to_thread() to detect the case where it is called on current, and a transaction is active, and in that case flush the TM regs to the thread_struct. This patch also moves flush_tmregs_to_thread() into ptrace.c as it is only called from that file. Fixes: 8d460f6156cd ("powerpc/process: Add the function flush_tmregs_to_thread") Signed-off-by: Cyril Bur [mpe: Flesh out change log] Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/switch_to.h | 8 -------- arch/powerpc/kernel/process.c | 20 -------------------- arch/powerpc/kernel/ptrace.c | 19 +++++++++++++++++++ 3 files changed, 19 insertions(+), 28 deletions(-) diff --git a/arch/powerpc/include/asm/switch_to.h b/arch/powerpc/include/asm/switch_to.h index 0a74ebe934e1..17c8380673a6 100644 --- a/arch/powerpc/include/asm/switch_to.h +++ b/arch/powerpc/include/asm/switch_to.h @@ -75,14 +75,6 @@ static inline void disable_kernel_spe(void) static inline void __giveup_spe(struct task_struct *t) { } #endif -#ifdef CONFIG_PPC_TRANSACTIONAL_MEM -extern void flush_tmregs_to_thread(struct task_struct *); -#else -static inline void flush_tmregs_to_thread(struct task_struct *t) -{ -} -#endif - static inline void clear_task_ebb(struct task_struct *t) { #ifdef CONFIG_PPC_BOOK3S_64 diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 58ccf86415b4..9ee2623e0f67 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -1074,26 +1074,6 @@ static inline void restore_sprs(struct thread_struct *old_thread, #endif } -#ifdef CONFIG_PPC_TRANSACTIONAL_MEM -void flush_tmregs_to_thread(struct task_struct *tsk) -{ - /* - * Process self tracing is not yet supported through - * ptrace interface. Ptrace generic code should have - * prevented this from happening in the first place. - * Warn once here with the message, if some how it - * is attempted. - */ - WARN_ONCE(tsk == current, - "Not expecting ptrace on self: TM regs may be incorrect\n"); - - /* - * If task is not current, it should have been flushed - * already to it's thread_struct during __switch_to(). - */ -} -#endif - struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *new) { diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c index 4f3c5756cc09..bf91658a8a40 100644 --- a/arch/powerpc/kernel/ptrace.c +++ b/arch/powerpc/kernel/ptrace.c @@ -38,6 +38,7 @@ #include #include #include +#include #define CREATE_TRACE_POINTS #include @@ -118,6 +119,24 @@ static const struct pt_regs_offset regoffset_table[] = { REG_OFFSET_END, }; +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM +static void flush_tmregs_to_thread(struct task_struct *tsk) +{ + /* + * If task is not current, it will have been flushed already to + * it's thread_struct during __switch_to(). + * + * A reclaim flushes ALL the state. + */ + + if (tsk == current && MSR_TM_SUSPENDED(mfmsr())) + tm_reclaim_current(TM_CAUSE_SIGNAL); + +} +#else +static inline void flush_tmregs_to_thread(struct task_struct *tsk) { } +#endif + /** * regs_query_register_offset() - query register offset from its name * @name: the name of a register From 7d70c63c7132eb95e428e945242360233ab23d82 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 10 Aug 2016 17:29:29 +1000 Subject: [PATCH 134/235] powerpc: Print the kernel load address at the end of prom_init() This makes it easier to debug crashes that happen very early before the kernel takes over Open Firmware by allowing us to relate the OF reported crashing addresses to offsets within the kernel. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/prom_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 6ee4b72cda42..4e74fc588a3f 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -2940,7 +2940,7 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, /* Don't print anything after quiesce under OPAL, it crashes OFW */ if (of_platform != PLATFORM_OPAL) { - prom_printf("Booting Linux via __start() ...\n"); + prom_printf("Booting Linux via __start() @ 0x%lx ...\n", kbase); prom_debug("->dt_header_start=0x%x\n", hdr); } From f9cc1d1f808dbdfd56978259d262b879504efaed Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 10 Aug 2016 17:32:38 +1000 Subject: [PATCH 135/235] powerpc: Update obsolete comment in setup_32.c about early_init() We don't identify the machine type anymore... Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/setup_32.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index c3e861df4b20..f10a975baa5e 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -93,10 +93,8 @@ notrace unsigned long __init early_init(unsigned long dt_ptr) * and we are running with enough of the MMU enabled to have our * proper kernel virtual addresses * - * Find out what kind of machine we're on and save any data we need - * from the early boot process (devtree is copied on pmac by prom_init()). - * This is called very early on the boot process, after a minimal - * MMU environment has been set up but before MMU_init is called. + * We do the initial parsing of the flat device-tree and prepares + * for the MMU to be fully initialized. */ extern unsigned int memset_nocache_branch; /* Insn to be replaced by NOP */ From 97f6e0cc35026a2a09147a6da636d901525e1969 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 10 Aug 2016 17:27:34 +1000 Subject: [PATCH 136/235] powerpc/32: Fix crash during static key init We cannot do those initializations from apply_feature_fixups() as this function runs in a very restricted environment on 32-bit where the kernel isn't running at its linked address and the PTRRELOC() macro must be used for any global accesss. Instead, split them into a separtate steup_feature_keys() function which is called in a more suitable spot on ppc32. Fixes: 309b315b6ec6 ("powerpc: Call jump_label_init() in apply_feature_fixups()") Reported-and-tested-by: Christian Kujau Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/feature-fixups.h | 1 + arch/powerpc/kernel/setup_32.c | 3 +++ arch/powerpc/kernel/setup_64.c | 1 + arch/powerpc/lib/feature-fixups.c | 3 +++ 4 files changed, 8 insertions(+) diff --git a/arch/powerpc/include/asm/feature-fixups.h b/arch/powerpc/include/asm/feature-fixups.h index 57fec8ac7b92..ddf54f5bbdd1 100644 --- a/arch/powerpc/include/asm/feature-fixups.h +++ b/arch/powerpc/include/asm/feature-fixups.h @@ -186,6 +186,7 @@ label##3: \ #ifndef __ASSEMBLY__ void apply_feature_fixups(void); +void setup_feature_keys(void); #endif #endif /* __ASM_POWERPC_FEATURE_FIXUPS_H */ diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index f10a975baa5e..24ec3ea4b3a2 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -100,6 +100,9 @@ extern unsigned int memset_nocache_branch; /* Insn to be replaced by NOP */ notrace void __init machine_init(u64 dt_ptr) { + /* Configure static keys first, now that we're relocated. */ + setup_feature_keys(); + /* Enable early debugging if any specified (see udbg.h) */ udbg_early_init(); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index eafb9a79e011..7ac8e6eaab5b 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -300,6 +300,7 @@ void __init early_setup(unsigned long dt_ptr) /* Apply all the dynamic patching */ apply_feature_fixups(); + setup_feature_keys(); /* Initialize the hash table or TLB handling */ early_init_mmu(); diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index 74145f02ad41..043415f0bdb1 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -188,7 +188,10 @@ void __init apply_feature_fixups(void) &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup); #endif do_final_fixups(); +} +void __init setup_feature_keys(void) +{ /* * Initialise jump label. This causes all the cpu/mmu_has_feature() * checks to take on their correct polarity based on the current set of From 1a9e4c564ab174e53ed86def922804a5ddc63e7d Mon Sep 17 00:00:00 2001 From: Nicolai Stange Date: Thu, 14 Jul 2016 17:22:54 +0200 Subject: [PATCH 137/235] x86/timers/apic: Fix imprecise timer interrupts by eliminating TSC clockevents frequency roundoff error I noticed the following bug/misbehavior on certain Intel systems: with a single task running on a NOHZ CPU on an Intel Haswell, I recognized that I did not only get the one expected local_timer APIC interrupt, but two per second at minimum. (!) Further tracing showed that the first one precedes the programmed deadline by up to ~50us and hence, it did nothing except for reprogramming the TSC deadline clockevent device to trigger shortly thereafter again. The reason for this is imprecise calibration, the timeout we program into the APIC results in 'too short' timer interrupts. The core (hr)timer code notices this (because it has a precise ktime source and sees the short interrupt) and fixes it up by programming an additional very short interrupt period. This is obviously suboptimal. The reason for the imprecise calibration is twofold, and this patch fixes the first reason: In setup_APIC_timer(), the registered clockevent device's frequency is calculated by first dividing tsc_khz by TSC_DIVISOR and multiplying it with 1000 afterwards: (tsc_khz / TSC_DIVISOR) * 1000 The multiplication with 1000 is done for converting from kHz to Hz and the division by TSC_DIVISOR is carried out in order to make sure that the final result fits into an u32. However, with the order given in this calculation, the roundoff error introduced by the division gets magnified by a factor of 1000 by the following multiplication. To fix it, reversing the order of the division and the multiplication a la: (tsc_khz * 1000) / TSC_DIVISOR ... reduces the roundoff error already. Furthermore, if TSC_DIVISOR divides 1000, associativity holds: (tsc_khz * 1000) / TSC_DIVISOR = tsc_khz * (1000 / TSC_DIVISOR) and thus, the roundoff error even vanishes and the whole operation can be carried out within 32 bits. The powers of two that divide 1000 are 2, 4 and 8. A value of 8 for TSC_DIVISOR still allows for TSC frequencies up to 2^32 / 10^9ns * 8 = 34.4GHz which is way larger than anything to expect in the next years. Thus we also replace the current TSC_DIVISOR value of 32 by 8. Reverse the order of the divison and the multiplication in the calculation of the registered clockevent device's frequency. Signed-off-by: Nicolai Stange Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Paolo Bonzini Acked-by: Thomas Gleixner Cc: Adrian Hunter Cc: Borislav Petkov Cc: Christopher S. Hall Cc: H. Peter Anvin Cc: Hidehiro Kawai Cc: Len Brown Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Viresh Kumar Link: http://lkml.kernel.org/r/20160714152255.18295-2-nicstange@gmail.com [ Improved changelog. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/apic.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index ac8d8ad8b009..a315dc404756 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -313,7 +313,7 @@ int lapic_get_maxlvt(void) /* Clock divisor */ #define APIC_DIVISOR 16 -#define TSC_DIVISOR 32 +#define TSC_DIVISOR 8 /* * This function sets up the local APIC timer, with a timeout of @@ -565,7 +565,7 @@ static void setup_APIC_timer(void) CLOCK_EVT_FEAT_DUMMY); levt->set_next_event = lapic_next_deadline; clockevents_config_and_register(levt, - (tsc_khz / TSC_DIVISOR) * 1000, + tsc_khz * (1000 / TSC_DIVISOR), 0xF, ~0UL); } else clockevents_register_device(levt); From 6731b0d611a1274f9e785fa0189ac2aeeabd0591 Mon Sep 17 00:00:00 2001 From: Nicolai Stange Date: Thu, 14 Jul 2016 17:22:55 +0200 Subject: [PATCH 138/235] x86/timers/apic: Inform TSC deadline clockevent device about recalibration This patch eliminates a source of imprecise APIC timer interrupts, which imprecision may result in double interrupts or even late interrupts. The TSC deadline clockevent devices' configuration and registration happens before the TSC frequency calibration is refined in tsc_refine_calibration_work(). This results in the TSC clocksource and the TSC deadline clockevent devices being configured with slightly different frequencies: the former gets the refined one and the latter are configured with the inaccurate frequency detected earlier by means of the "Fast TSC calibration using PIT". Within the APIC code, introduce the notifier function lapic_update_tsc_freq() which reconfigures all per-CPU TSC deadline clockevent devices with the current tsc_khz. Call it from the TSC code after TSC calibration refinement has happened. Signed-off-by: Nicolai Stange Signed-off-by: Peter Zijlstra (Intel) Acked-by: Thomas Gleixner Cc: Adrian Hunter Cc: Borislav Petkov Cc: Christopher S. Hall Cc: H. Peter Anvin Cc: Hidehiro Kawai Cc: Len Brown Cc: Linus Torvalds Cc: Paolo Bonzini Cc: Peter Zijlstra Cc: Viresh Kumar Link: http://lkml.kernel.org/r/20160714152255.18295-3-nicstange@gmail.com [ Pushed #ifdef CONFIG_X86_LOCAL_APIC into header, improved changelog. ] Signed-off-by: Ingo Molnar --- arch/x86/include/asm/apic.h | 2 ++ arch/x86/kernel/apic/apic.c | 24 ++++++++++++++++++++++++ arch/x86/kernel/tsc.c | 4 ++++ 3 files changed, 30 insertions(+) diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index f5befd4945f2..124357773ffa 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -135,6 +135,7 @@ extern void init_apic_mappings(void); void register_lapic_address(unsigned long address); extern void setup_boot_APIC_clock(void); extern void setup_secondary_APIC_clock(void); +extern void lapic_update_tsc_freq(void); extern int APIC_init_uniprocessor(void); #ifdef CONFIG_X86_64 @@ -170,6 +171,7 @@ static inline void init_apic_mappings(void) { } static inline void disable_local_APIC(void) { } # define setup_boot_APIC_clock x86_init_noop # define setup_secondary_APIC_clock x86_init_noop +static inline void lapic_update_tsc_freq(void) { } #endif /* !CONFIG_X86_LOCAL_APIC */ #ifdef CONFIG_X86_X2APIC diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index a315dc404756..0fd3d659f13c 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -571,6 +571,30 @@ static void setup_APIC_timer(void) clockevents_register_device(levt); } +/* + * Install the updated TSC frequency from recalibration at the TSC + * deadline clockevent devices. + */ +static void __lapic_update_tsc_freq(void *info) +{ + struct clock_event_device *levt = this_cpu_ptr(&lapic_events); + + if (!this_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) + return; + + clockevents_update_freq(levt, tsc_khz * (1000 / TSC_DIVISOR)); +} + +void lapic_update_tsc_freq(void) +{ + /* + * The clockevent device's ->mult and ->shift can both be + * changed. In order to avoid races, schedule the frequency + * update code on each CPU. + */ + on_each_cpu(__lapic_update_tsc_freq, NULL, 0); +} + /* * In this functions we calibrate APIC bus clocks to the external timer. * diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index a804b5ab32d0..8fb4b6abac0e 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -22,6 +22,7 @@ #include #include #include +#include unsigned int __read_mostly cpu_khz; /* TSC clocks / usec, not used here */ EXPORT_SYMBOL(cpu_khz); @@ -1249,6 +1250,9 @@ static void tsc_refine_calibration_work(struct work_struct *work) (unsigned long)tsc_khz / 1000, (unsigned long)tsc_khz % 1000); + /* Inform the TSC deadline clockevent devices about the recalibration */ + lapic_update_tsc_freq(); + out: if (boot_cpu_has(X86_FEATURE_ART)) art_related_clocksource = &clocksource_tsc; From 0b8f1e2e26bfc6b9abe3f0f3faba2cb0eecb9fb9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 4 Aug 2016 14:37:24 +0200 Subject: [PATCH 139/235] perf/core: Fix sideband list-iteration vs. event ordering NULL pointer deference crash Vegard Nossum reported that perf fuzzing generates a NULL pointer dereference crash: > Digging a bit deeper into this, it seems the event itself is getting > created by perf_event_open() and it gets added to the pmu_event_list > through: > > perf_event_open() > - perf_event_alloc() > - account_event() > - account_pmu_sb_event() > - attach_sb_event() > > so at this point the event is being attached but its ->ctx is still > NULL. It seems like ->ctx is set just a bit later in > perf_event_open(), though. > > But before that, __schedule() comes along and creates a stack trace > similar to the one above: > > __schedule() > - __perf_event_task_sched_out() > - perf_iterate_sb() > - perf_iterate_sb_cpu() > - event_filter_match() > - perf_cgroup_match() > - __get_cpu_context() > - (dereference ctx which is NULL) > > So I guess the question is... should the event be attached (= put on > the list) before ->ctx gets set? Or should the cgroup code check for a > NULL ->ctx? The latter seems like the simplest solution. Moving the list-add later creates a bit of a mess. Reported-by: Vegard Nossum Tested-by: Vegard Nossum Tested-by: Vince Weaver Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: David Carrillo-Cisneros Cc: Jiri Olsa Cc: Kan Liang Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Fixes: f2fb6bef9251 ("perf/core: Optimize side-band event delivery") Link: http://lkml.kernel.org/r/20160804123724.GN6862@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/events/core.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index a19550d80ab1..87d02b8cb87e 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1716,8 +1716,8 @@ static inline int pmu_filter_match(struct perf_event *event) static inline int event_filter_match(struct perf_event *event) { - return (event->cpu == -1 || event->cpu == smp_processor_id()) - && perf_cgroup_match(event) && pmu_filter_match(event); + return (event->cpu == -1 || event->cpu == smp_processor_id()) && + perf_cgroup_match(event) && pmu_filter_match(event); } static void @@ -1737,8 +1737,8 @@ event_sched_out(struct perf_event *event, * maintained, otherwise bogus information is return * via read() for time_enabled, time_running: */ - if (event->state == PERF_EVENT_STATE_INACTIVE - && !event_filter_match(event)) { + if (event->state == PERF_EVENT_STATE_INACTIVE && + !event_filter_match(event)) { delta = tstamp - event->tstamp_stopped; event->tstamp_running += delta; event->tstamp_stopped = tstamp; @@ -2236,10 +2236,15 @@ perf_install_in_context(struct perf_event_context *ctx, lockdep_assert_held(&ctx->mutex); - event->ctx = ctx; if (event->cpu != -1) event->cpu = cpu; + /* + * Ensures that if we can observe event->ctx, both the event and ctx + * will be 'complete'. See perf_iterate_sb_cpu(). + */ + smp_store_release(&event->ctx, ctx); + if (!task) { cpu_function_call(cpu, __perf_install_in_context, event); return; @@ -5969,6 +5974,14 @@ static void perf_iterate_sb_cpu(perf_iterate_f output, void *data) struct perf_event *event; list_for_each_entry_rcu(event, &pel->list, sb_list) { + /* + * Skip events that are not fully formed yet; ensure that + * if we observe event->ctx, both event and ctx will be + * complete enough. See perf_install_in_context(). + */ + if (!smp_load_acquire(&event->ctx)) + continue; + if (event->state < PERF_EVENT_STATE_INACTIVE) continue; if (!event_filter_match(event)) From db4a835601b73cf8d6cd8986381d966b8e13d2d9 Mon Sep 17 00:00:00 2001 From: David Carrillo-Cisneros Date: Tue, 2 Aug 2016 00:48:12 -0700 Subject: [PATCH 140/235] perf/core: Set cgroup in CPU contexts for new cgroup events There's a perf stat bug easy to observer on a machine with only one cgroup: $ perf stat -e cycles -I 1000 -C 0 -G / # time counts unit events 1.000161699 cycles / 2.000355591 cycles / 3.000565154 cycles / 4.000951350 cycles / We'd expect some output there. The underlying problem is that there is an optimization in perf_cgroup_sched_{in,out}() that skips the switch of cgroup events if the old and new cgroups in a task switch are the same. This optimization interacts with the current code in two ways that cause a CPU context's cgroup (cpuctx->cgrp) to be NULL even if a cgroup event matches the current task. These are: 1. On creation of the first cgroup event in a CPU: In current code, cpuctx->cpu is only set in perf_cgroup_sched_in, but due to the aforesaid optimization, perf_cgroup_sched_in will run until the next cgroup switches in that CPU. This may happen late or never happen, depending on system's number of cgroups, CPU load, etc. 2. On deletion of the last cgroup event in a cpuctx: In list_del_event, cpuctx->cgrp is set NULL. Any new cgroup event will not be sched in because cpuctx->cgrp == NULL until a cgroup switch occurs and perf_cgroup_sched_in is executed (updating cpuctx->cgrp). This patch fixes both problems by setting cpuctx->cgrp in list_add_event, mirroring what list_del_event does when removing a cgroup event from CPU context, as introduced in: commit 68cacd29167b ("perf_events: Fix stale ->cgrp pointer in update_cgrp_time_from_cpuctx()") With this patch, cpuctx->cgrp is always set/clear when installing/removing the first/last cgroup event in/from the CPU context. With cpuctx->cgrp correctly set, event_filter_match works as intended when events are sched in/out. After the fix, the output is as expected: $ perf stat -e cycles -I 1000 -a -G / # time counts unit events 1.004699159 627342882 cycles / 2.007397156 615272690 cycles / 3.010019057 616726074 cycles / Signed-off-by: David Carrillo-Cisneros Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Kan Liang Cc: Linus Torvalds Cc: Paul Turner Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vegard Nossum Cc: Vince Weaver Link: http://lkml.kernel.org/r/1470124092-113192-1-git-send-email-davidcc@google.com Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 4 +++ kernel/events/core.c | 54 +++++++++++++++++++++++++------------- 2 files changed, 40 insertions(+), 18 deletions(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 8ed4326164cc..2b6b43cc0dd5 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -743,7 +743,9 @@ struct perf_event_context { u64 parent_gen; u64 generation; int pin_count; +#ifdef CONFIG_CGROUP_PERF int nr_cgroups; /* cgroup evts */ +#endif void *task_ctx_data; /* pmu specific data */ struct rcu_head rcu_head; }; @@ -769,7 +771,9 @@ struct perf_cpu_context { unsigned int hrtimer_active; struct pmu *unique_pmu; +#ifdef CONFIG_CGROUP_PERF struct perf_cgroup *cgrp; +#endif }; struct perf_output_handle { diff --git a/kernel/events/core.c b/kernel/events/core.c index 87d02b8cb87e..1903b8f3a705 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -843,6 +843,32 @@ perf_cgroup_mark_enabled(struct perf_event *event, } } } + +/* + * Update cpuctx->cgrp so that it is set when first cgroup event is added and + * cleared when last cgroup event is removed. + */ +static inline void +list_update_cgroup_event(struct perf_event *event, + struct perf_event_context *ctx, bool add) +{ + struct perf_cpu_context *cpuctx; + + if (!is_cgroup_event(event)) + return; + + if (add && ctx->nr_cgroups++) + return; + else if (!add && --ctx->nr_cgroups) + return; + /* + * Because cgroup events are always per-cpu events, + * this will always be called from the right CPU. + */ + cpuctx = __get_cpu_context(ctx); + cpuctx->cgrp = add ? event->cgrp : NULL; +} + #else /* !CONFIG_CGROUP_PERF */ static inline bool @@ -920,6 +946,13 @@ perf_cgroup_mark_enabled(struct perf_event *event, struct perf_event_context *ctx) { } + +static inline void +list_update_cgroup_event(struct perf_event *event, + struct perf_event_context *ctx, bool add) +{ +} + #endif /* @@ -1392,6 +1425,7 @@ ctx_group_list(struct perf_event *event, struct perf_event_context *ctx) static void list_add_event(struct perf_event *event, struct perf_event_context *ctx) { + lockdep_assert_held(&ctx->lock); WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT); @@ -1412,8 +1446,7 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx) list_add_tail(&event->group_entry, list); } - if (is_cgroup_event(event)) - ctx->nr_cgroups++; + list_update_cgroup_event(event, ctx, true); list_add_rcu(&event->event_entry, &ctx->event_list); ctx->nr_events++; @@ -1581,8 +1614,6 @@ static void perf_group_attach(struct perf_event *event) static void list_del_event(struct perf_event *event, struct perf_event_context *ctx) { - struct perf_cpu_context *cpuctx; - WARN_ON_ONCE(event->ctx != ctx); lockdep_assert_held(&ctx->lock); @@ -1594,20 +1625,7 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) event->attach_state &= ~PERF_ATTACH_CONTEXT; - if (is_cgroup_event(event)) { - ctx->nr_cgroups--; - /* - * Because cgroup events are always per-cpu events, this will - * always be called from the right CPU. - */ - cpuctx = __get_cpu_context(ctx); - /* - * If there are no more cgroup events then clear cgrp to avoid - * stale pointer in update_cgrp_time_from_cpuctx(). - */ - if (!ctx->nr_cgroups) - cpuctx->cgrp = NULL; - } + list_update_cgroup_event(event, ctx, false); ctx->nr_events--; if (event->attr.inherit_stat) From a23eadfae2fd45536a355b785d5a1533e1955c22 Mon Sep 17 00:00:00 2001 From: Tommaso Cucinotta Date: Tue, 19 Jul 2016 11:44:50 +0200 Subject: [PATCH 141/235] sched/deadline: Fix wrap-around in DL heap Current code in cpudeadline.c has a bug in re-heapifying when adding a new element at the end of the heap, because a deadline value of 0 is temporarily set in the new elem, then cpudl_change_key() is called with the actual elem deadline as param. However, the function compares the new deadline to set with the one previously in the elem, which is 0. So, if current absolute deadlines grew so much to have negative values as s64, the comparison in cpudl_change_key() makes the wrong decision. Instead, as from dl_time_before(), the kernel should handle correctly abs deadlines wrap-arounds. This patch fixes the problem with a minimally invasive change that forces cpudl_change_key() to heapify up in this case. Signed-off-by: Tommaso Cucinotta Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Luca Abeni Cc: Juri Lelli Cc: Juri Lelli Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1468921493-10054-2-git-send-email-tommaso.cucinotta@sssup.it Signed-off-by: Ingo Molnar --- kernel/sched/cpudeadline.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c index 5be58820465c..d4184498c9f5 100644 --- a/kernel/sched/cpudeadline.c +++ b/kernel/sched/cpudeadline.c @@ -168,7 +168,7 @@ void cpudl_set(struct cpudl *cp, int cpu, u64 dl, int is_valid) if (old_idx == IDX_INVALID) { cp->size++; - cp->elements[cp->size - 1].dl = 0; + cp->elements[cp->size - 1].dl = dl; cp->elements[cp->size - 1].cpu = cpu; cp->elements[cpu].idx = cp->size - 1; cpudl_change_key(cp, cp->size - 1, dl); From b8922125e4790fa237a8a4204562ecf457ef54bb Mon Sep 17 00:00:00 2001 From: Xunlei Pang Date: Sat, 9 Jul 2016 15:54:22 +0800 Subject: [PATCH 142/235] sched/fair: Fix typo in sync_throttle() We should update cfs_rq->throttled_clock_task, not pcfs_rq->throttle_clock_task. The effects of this bug was probably occasionally erratic group scheduling, particularly in cgroups-intense workloads. Signed-off-by: Xunlei Pang [ Added changelog. ] Signed-off-by: Peter Zijlstra (Intel) Acked-by: Konstantin Khlebnikov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 55e16d30bd99 ("sched/fair: Rework throttle_count sync") Link: http://lkml.kernel.org/r/1468050862-18864-1-git-send-email-xlpang@redhat.com Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 4088eedea763..039de34f1521 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4269,7 +4269,7 @@ static void sync_throttle(struct task_group *tg, int cpu) pcfs_rq = tg->parent->cfs_rq[cpu]; cfs_rq->throttle_count = pcfs_rq->throttle_count; - pcfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu)); + cfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu)); } /* conditionally throttle active cfs_rq's from put_prev_entity() */ From 6075620b0590eaf22f10ce88833eb20a57f760d6 Mon Sep 17 00:00:00 2001 From: Giovanni Gherdovich Date: Fri, 5 Aug 2016 10:21:56 +0200 Subject: [PATCH 143/235] sched/cputime: Mitigate performance regression in times()/clock_gettime() Commit: 6e998916dfe3 ("sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency") fixed a problem whereby clock_nanosleep() followed by clock_gettime() could allow a task to wake early. It addressed the problem by calling the scheduling classes update_curr() when the cputimer starts. Said change induced a considerable performance regression on the syscalls times() and clock_gettimes(CLOCK_PROCESS_CPUTIME_ID). There are some debuggers and applications that monitor their own performance that accidentally depend on the performance of these specific calls. This patch mitigates the performace loss by prefetching data in the CPU cache, as stalls due to cache misses appear to be where most time is spent in our benchmarks. Here are the performance gain of this patch over v4.7-rc7 on a Sandy Bridge box with 32 logical cores and 2 NUMA nodes. The test is repeated with a variable number of threads, from 2 to 4*num_cpus; the results are in seconds and correspond to the average of 10 runs; the percentage gain is computed with (before-after)/before so a positive value is an improvement (it's faster). The improvement varies between a few percents for 5-20 threads and more than 10% for 2 or >20 threads. pound_clock_gettime: threads 4.7-rc7 patched 4.7-rc7 [num] [secs] [secs (percent)] 2 3.48 3.06 ( 11.83%) 5 3.33 3.25 ( 2.40%) 8 3.37 3.26 ( 3.30%) 12 3.32 3.37 ( -1.60%) 21 4.01 3.90 ( 2.74%) 30 3.63 3.36 ( 7.41%) 48 3.71 3.11 ( 16.27%) 79 3.75 3.16 ( 15.74%) 110 3.81 3.25 ( 14.80%) 128 3.88 3.31 ( 14.76%) pound_times: threads 4.7-rc7 patched 4.7-rc7 [num] [secs] [secs (percent)] 2 3.65 3.25 ( 11.03%) 5 3.45 3.17 ( 7.92%) 8 3.52 3.22 ( 8.69%) 12 3.29 3.36 ( -2.04%) 21 4.07 3.92 ( 3.78%) 30 3.87 3.40 ( 12.17%) 48 3.79 3.16 ( 16.61%) 79 3.88 3.28 ( 15.42%) 110 3.90 3.38 ( 13.35%) 128 4.00 3.38 ( 15.45%) pound_clock_gettime and pound_clock_gettime are two benchmarks included in the MMTests framework. They launch a given number of threads which repeatedly call times() or clock_gettimes(). The results above can be reproduced with cloning MMTests from github.com and running the "poundtime" workload: $ git clone https://github.com/gormanm/mmtests.git $ cd mmtests $ cp configs/config-global-dhp__workload_poundtime config $ ./run-mmtests.sh --run-monitor $(uname -r) The above will run "poundtime" measuring the kernel currently running on the machine; Once a new kernel is installed and the machine rebooted, running again $ cd mmtests $ ./run-mmtests.sh --run-monitor $(uname -r) will produce results to compare with. A comparison table will be output with: $ cd mmtests/work/log $ ../../compare-kernels.sh the table will contain a lot of entries; grepping for "Amean" (as in "arithmetic mean") will give the tables presented above. The source code for the two benchmarks is reported at the end of this changelog for clairity. The cache misses addressed by this patch were found using a combination of `perf top`, `perf record` and `perf annotate`. The incriminated lines were found to be struct sched_entity *curr = cfs_rq->curr; and delta_exec = now - curr->exec_start; in the function update_curr() from kernel/sched/fair.c. This patch prefetches the data from memory just before update_curr is called in the interested execution path. A comparison of the total number of cycles before and after the patch follows; the data is obtained using `perf stat -r 10 -ddd ` running over the same sequence of number of threads used above (a positive gain is an improvement): threads cycles before cycles after gain 2 19,699,563,964 +-1.19% 17,358,917,517 +-1.85% 11.88% 5 47,401,089,566 +-2.96% 45,103,730,829 +-0.97% 4.85% 8 80,923,501,004 +-3.01% 71,419,385,977 +-0.77% 11.74% 12 112,326,485,473 +-0.47% 110,371,524,403 +-0.47% 1.74% 21 193,455,574,299 +-0.72% 180,120,667,904 +-0.36% 6.89% 30 315,073,519,013 +-1.64% 271,222,225,950 +-1.29% 13.92% 48 321,969,515,332 +-1.48% 273,353,977,321 +-1.16% 15.10% 79 337,866,003,422 +-0.97% 289,462,481,538 +-1.05% 14.33% 110 338,712,691,920 +-0.78% 290,574,233,170 +-0.77% 14.21% 128 348,384,794,006 +-0.50% 292,691,648,206 +-0.66% 15.99% A comparison of cache miss vs total cache loads ratios, before and after the patch (again from the `perf stat -r 10 -ddd ` tables): threads L1 misses/total*100 L1 misses/total*100 gain before after 2 7.43 +-4.90% 7.36 +-4.70% 0.94% 5 13.09 +-4.74% 13.52 +-3.73% -3.28% 8 13.79 +-5.61% 12.90 +-3.27% 6.45% 12 11.57 +-2.44% 8.71 +-1.40% 24.72% 21 12.39 +-3.92% 9.97 +-1.84% 19.53% 30 13.91 +-2.53% 11.73 +-2.28% 15.67% 48 13.71 +-1.59% 12.32 +-1.97% 10.14% 79 14.44 +-0.66% 13.40 +-1.06% 7.20% 110 15.86 +-0.50% 14.46 +-0.59% 8.83% 128 16.51 +-0.32% 15.06 +-0.78% 8.78% As a final note, the following shows the evolution of performance figures in the "poundtime" benchmark and pinpoints commit 6e998916dfe3 ("sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency") as a major source of degradation, mostly unaddressed to this day (figures expressed in seconds). pound_clock_gettime: threads parent of 6e998916dfe3 4.7-rc7 6e998916dfe3 itself 2 2.23 3.68 ( -64.56%) 3.48 (-55.48%) 5 2.83 3.78 ( -33.42%) 3.33 (-17.43%) 8 2.84 4.31 ( -52.12%) 3.37 (-18.76%) 12 3.09 3.61 ( -16.74%) 3.32 ( -7.17%) 21 3.14 4.63 ( -47.36%) 4.01 (-27.71%) 30 3.28 5.75 ( -75.37%) 3.63 (-10.80%) 48 3.02 6.05 (-100.56%) 3.71 (-22.99%) 79 2.88 6.30 (-118.90%) 3.75 (-30.26%) 110 2.95 6.46 (-119.00%) 3.81 (-29.24%) 128 3.05 6.42 (-110.08%) 3.88 (-27.04%) pound_times: threads parent of 6e998916dfe3 4.7-rc7 6e998916dfe3 itself 2 2.27 3.73 ( -64.71%) 3.65 (-61.14%) 5 2.78 3.77 ( -35.56%) 3.45 (-23.98%) 8 2.79 4.41 ( -57.71%) 3.52 (-26.05%) 12 3.02 3.56 ( -17.94%) 3.29 ( -9.08%) 21 3.10 4.61 ( -48.74%) 4.07 (-31.34%) 30 3.33 5.75 ( -72.53%) 3.87 (-16.01%) 48 2.96 6.06 (-105.04%) 3.79 (-28.10%) 79 2.88 6.24 (-116.83%) 3.88 (-34.81%) 110 2.98 6.37 (-114.08%) 3.90 (-31.12%) 128 3.10 6.35 (-104.61%) 4.00 (-28.87%) The source code of the two benchmarks follows. To compile the two: NR_THREADS=42 for FILE in pound_times pound_clock_gettime; do gcc -lrt -O2 -lpthread -DNUM_THREADS=$NR_THREADS $FILE.c -o $FILE done ==== BEGIN pound_times.c ==== struct tms start; void *pound (void *threadid) { struct tms end; int oldutime = 0; int utime; int i; for (i = 0; i < 5000000 / NUM_THREADS; i++) { times(&end); utime = ((int)end.tms_utime - (int)start.tms_utime); if (oldutime > utime) { printf("utime decreased, was %d, now %d!\n", oldutime, utime); } oldutime = utime; } pthread_exit(NULL); } int main() { pthread_t th[NUM_THREADS]; long i; times(&start); for (i = 0; i < NUM_THREADS; i++) { pthread_create (&th[i], NULL, pound, (void *)i); } pthread_exit(NULL); return 0; } ==== END pound_times.c ==== ==== BEGIN pound_clock_gettime.c ==== void *pound (void *threadid) { struct timespec ts; int rc, i; unsigned long prev = 0, this = 0; for (i = 0; i < 5000000 / NUM_THREADS; i++) { rc = clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &ts); if (rc < 0) perror("clock_gettime"); this = (ts.tv_sec * 1000000000) + ts.tv_nsec; if (0 && this < prev) printf("%lu ns timewarp at iteration %d\n", prev - this, i); prev = this; } pthread_exit(NULL); } int main() { pthread_t th[NUM_THREADS]; long rc, i; pid_t pgid; for (i = 0; i < NUM_THREADS; i++) { rc = pthread_create(&th[i], NULL, pound, (void *)i); if (rc < 0) perror("pthread_create"); } pthread_exit(NULL); return 0; } ==== END pound_clock_gettime.c ==== Suggested-by: Mike Galbraith Signed-off-by: Giovanni Gherdovich Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Stanislaw Gruszka Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1470385316-15027-2-git-send-email-ggherdovich@suse.cz Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 5c883fe8e440..2a906f20fba7 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -74,6 +74,7 @@ #include #include #include +#include #include #include @@ -2971,6 +2972,23 @@ DEFINE_PER_CPU(struct kernel_cpustat, kernel_cpustat); EXPORT_PER_CPU_SYMBOL(kstat); EXPORT_PER_CPU_SYMBOL(kernel_cpustat); +/* + * The function fair_sched_class.update_curr accesses the struct curr + * and its field curr->exec_start; when called from task_sched_runtime(), + * we observe a high rate of cache misses in practice. + * Prefetching this data results in improved performance. + */ +static inline void prefetch_curr_exec_start(struct task_struct *p) +{ +#ifdef CONFIG_FAIR_GROUP_SCHED + struct sched_entity *curr = (&p->se)->cfs_rq->curr; +#else + struct sched_entity *curr = (&task_rq(p)->cfs)->curr; +#endif + prefetch(curr); + prefetch(&curr->exec_start); +} + /* * Return accounted runtime for the task. * In case the task is currently running, return the runtime plus current's @@ -3005,6 +3023,7 @@ unsigned long long task_sched_runtime(struct task_struct *p) * thread, breaking clock_gettime(). */ if (task_current(rq, p) && task_on_rq_queued(p)) { + prefetch_curr_exec_start(p); update_rq_clock(rq); p->sched_class->update_curr(rq); } From c0c8c9fa210c9a042060435f17e40ba4a76d6d6f Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Thu, 4 Aug 2016 09:42:20 +0800 Subject: [PATCH 144/235] sched/deadline: Fix lock pinning warning during CPU hotplug The following warning can be triggered by hot-unplugging the CPU on which an active SCHED_DEADLINE task is running on: WARNING: CPU: 0 PID: 0 at kernel/locking/lockdep.c:3531 lock_release+0x690/0x6a0 releasing a pinned lock Call Trace: dump_stack+0x99/0xd0 __warn+0xd1/0xf0 ? dl_task_timer+0x1a1/0x2b0 warn_slowpath_fmt+0x4f/0x60 ? sched_clock+0x13/0x20 lock_release+0x690/0x6a0 ? enqueue_pushable_dl_task+0x9b/0xa0 ? enqueue_task_dl+0x1ca/0x480 _raw_spin_unlock+0x1f/0x40 dl_task_timer+0x1a1/0x2b0 ? push_dl_task.part.31+0x190/0x190 WARNING: CPU: 0 PID: 0 at kernel/locking/lockdep.c:3649 lock_unpin_lock+0x181/0x1a0 unpinning an unpinned lock Call Trace: dump_stack+0x99/0xd0 __warn+0xd1/0xf0 warn_slowpath_fmt+0x4f/0x60 lock_unpin_lock+0x181/0x1a0 dl_task_timer+0x127/0x2b0 ? push_dl_task.part.31+0x190/0x190 As per the comment before this code, its safe to drop the RQ lock here, and since we (potentially) change rq, unpin and repin to avoid the splat. Signed-off-by: Wanpeng Li [ Rewrote changelog. ] Signed-off-by: Peter Zijlstra (Intel) Cc: Juri Lelli Cc: Linus Torvalds Cc: Luca Abeni Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1470274940-17976-1-git-send-email-wanpeng.li@hotmail.com Signed-off-by: Ingo Molnar --- kernel/sched/deadline.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index fcb7f0217ff4..1ce8867283dc 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -658,8 +658,11 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) * * XXX figure out if select_task_rq_dl() deals with offline cpus. */ - if (unlikely(!rq->online)) + if (unlikely(!rq->online)) { + lockdep_unpin_lock(&rq->lock, rf.cookie); rq = dl_task_offline_migration(rq, p); + rf.cookie = lockdep_pin_lock(&rq->lock); + } /* * Queueing this task back might have overloaded rq, check if we need From 2db34e8bf9a22f4e38b29deccee57457bc0e7d74 Mon Sep 17 00:00:00 2001 From: pan xinhui Date: Mon, 18 Jul 2016 17:47:39 +0800 Subject: [PATCH 145/235] locking/qrwlock: Fix write unlock bug on big endian systems This patch aims to get rid of endianness in queued_write_unlock(). We want to set __qrwlock->wmode to NULL, however the address is not &lock->cnts in big endian machine. That causes queued_write_unlock() write NULL to the wrong field of __qrwlock. So implement __qrwlock_write_byte() which returns the correct __qrwlock->wmode address. Suggested-by: Peter Zijlstra (Intel) Signed-off-by: Pan Xinhui Signed-off-by: Peter Zijlstra (Intel) Cc: Andrew Morton Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Waiman.Long@hpe.com Cc: arnd@arndb.de Cc: boqun.feng@gmail.com Cc: will.deacon@arm.com Link: http://lkml.kernel.org/r/1468835259-4486-1-git-send-email-xinhui.pan@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- include/asm-generic/qrwlock.h | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) diff --git a/include/asm-generic/qrwlock.h b/include/asm-generic/qrwlock.h index 54a8e65e18b6..7d026bf27713 100644 --- a/include/asm-generic/qrwlock.h +++ b/include/asm-generic/qrwlock.h @@ -25,7 +25,20 @@ #include /* - * Writer states & reader shift and bias + * Writer states & reader shift and bias. + * + * | +0 | +1 | +2 | +3 | + * ----+----+----+----+----+ + * LE | 78 | 56 | 34 | 12 | 0x12345678 + * ----+----+----+----+----+ + * | wr | rd | + * +----+----+----+----+ + * + * ----+----+----+----+----+ + * BE | 12 | 34 | 56 | 78 | 0x12345678 + * ----+----+----+----+----+ + * | rd | wr | + * +----+----+----+----+ */ #define _QW_WAITING 1 /* A writer is waiting */ #define _QW_LOCKED 0xff /* A writer holds the lock */ @@ -133,13 +146,23 @@ static inline void queued_read_unlock(struct qrwlock *lock) (void)atomic_sub_return_release(_QR_BIAS, &lock->cnts); } +/** + * __qrwlock_write_byte - retrieve the write byte address of a queue rwlock + * @lock : Pointer to queue rwlock structure + * Return: the write byte address of a queue rwlock + */ +static inline u8 *__qrwlock_write_byte(struct qrwlock *lock) +{ + return (u8 *)lock + 3 * IS_BUILTIN(CONFIG_CPU_BIG_ENDIAN); +} + /** * queued_write_unlock - release write lock of a queue rwlock * @lock : Pointer to queue rwlock structure */ static inline void queued_write_unlock(struct qrwlock *lock) { - smp_store_release((u8 *)&lock->cnts, 0); + smp_store_release(__qrwlock_write_byte(lock), 0); } /* From 229ce631574761870a2ac938845fadbd07f35caa Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Thu, 14 Jul 2016 16:15:56 +0800 Subject: [PATCH 146/235] locking/pvqspinlock: Fix double hash race When the lock holder vCPU is racing with the queue head: CPU 0 (lock holder) CPU1 (queue head) =================== ================= spin_lock(); spin_lock(); pv_kick_node(): pv_wait_head_or_lock(): if (!lp) { lp = pv_hash(lock, pn); xchg(&l->locked, _Q_SLOW_VAL); } WRITE_ONCE(pn->state, vcpu_halted); cmpxchg(&pn->state, vcpu_halted, vcpu_hashed); WRITE_ONCE(l->locked, _Q_SLOW_VAL); (void)pv_hash(lock, pn); In this case, lock holder inserts the pv_node of queue head into the hash table and set _Q_SLOW_VAL unnecessary. This patch avoids it by restoring/setting vcpu_hashed state after failing adaptive locking spinning. Signed-off-by: Wanpeng Li Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Pan Xinhui Cc: Andrew Morton Cc: Davidlohr Bueso Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Waiman Long Link: http://lkml.kernel.org/r/1468484156-4521-1-git-send-email-wanpeng.li@hotmail.com Signed-off-by: Ingo Molnar --- kernel/locking/qspinlock_paravirt.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h index 37649e69056c..8a99abf58080 100644 --- a/kernel/locking/qspinlock_paravirt.h +++ b/kernel/locking/qspinlock_paravirt.h @@ -450,7 +450,7 @@ pv_wait_head_or_lock(struct qspinlock *lock, struct mcs_spinlock *node) goto gotlock; } } - WRITE_ONCE(pn->state, vcpu_halted); + WRITE_ONCE(pn->state, vcpu_hashed); qstat_inc(qstat_pv_wait_head, true); qstat_inc(qstat_pv_wait_again, waitcnt); pv_wait(&l->locked, _Q_SLOW_VAL); From c2ace36b884de9330c4149064ae8d212d2e0d9ee Mon Sep 17 00:00:00 2001 From: Pan Xinhui Date: Wed, 13 Jul 2016 18:23:34 +0800 Subject: [PATCH 147/235] locking/pvqspinlock: Fix a bug in qstat_read() It's obviously wrong to set stat to NULL. So lets remove it. Otherwise it is always zero when we check the latency of kick/wake. Signed-off-by: Pan Xinhui Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Waiman Long Cc: Andrew Morton Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1468405414-3700-1-git-send-email-xinhui.pan@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/locking/qspinlock_stat.h | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/locking/qspinlock_stat.h b/kernel/locking/qspinlock_stat.h index 22e025309845..b9d031516254 100644 --- a/kernel/locking/qspinlock_stat.h +++ b/kernel/locking/qspinlock_stat.h @@ -153,7 +153,6 @@ static ssize_t qstat_read(struct file *file, char __user *user_buf, */ if ((counter == qstat_pv_latency_kick) || (counter == qstat_pv_latency_wake)) { - stat = 0; if (kicks) stat = DIV_ROUND_CLOSEST_ULL(stat, kicks); } From 469f00231278da68062a809306df0bac95a27507 Mon Sep 17 00:00:00 2001 From: Alexander Potapenko Date: Fri, 15 Jul 2016 11:42:43 +0200 Subject: [PATCH 148/235] x86, kasan, ftrace: Put APIC interrupt handlers into .irqentry.text Dmitry Vyukov has reported unexpected KASAN stackdepot growth: https://github.com/google/kasan/issues/36 ... which is caused by the APIC handlers not being present in .irqentry.text: When building with CONFIG_FUNCTION_GRAPH_TRACER=y or CONFIG_KASAN=y, put the APIC interrupt handlers into the .irqentry.text section. This is needed because both KASAN and function graph tracer use __irqentry_text_start and __irqentry_text_end to determine whether a function is an IRQ entry point. Reported-by: Dmitry Vyukov Signed-off-by: Alexander Potapenko Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: aryabinin@virtuozzo.com Cc: kasan-dev@googlegroups.com Cc: kcc@google.com Cc: rostedt@goodmis.org Link: http://lkml.kernel.org/r/1468575763-144889-1-git-send-email-glider@google.com [ Minor edits. ] Signed-off-by: Ingo Molnar --- arch/x86/entry/entry_64.S | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index b846875aeea6..9f85827db24e 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -601,9 +601,20 @@ apicinterrupt3 \num trace(\sym) smp_trace(\sym) .endm #endif +/* Make sure APIC interrupt handlers end up in the irqentry section: */ +#if defined(CONFIG_FUNCTION_GRAPH_TRACER) || defined(CONFIG_KASAN) +# define PUSH_SECTION_IRQENTRY .pushsection .irqentry.text, "ax" +# define POP_SECTION_IRQENTRY .popsection +#else +# define PUSH_SECTION_IRQENTRY +# define POP_SECTION_IRQENTRY +#endif + .macro apicinterrupt num sym do_sym +PUSH_SECTION_IRQENTRY apicinterrupt3 \num \sym \do_sym trace_apicinterrupt \num \sym +POP_SECTION_IRQENTRY .endm #ifdef CONFIG_SMP From 22cc1ca3c5469cf17e149be232817b9223afa5e4 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 9 Aug 2016 21:54:53 +0200 Subject: [PATCH 149/235] x86/hpet: Fix /dev/rtc breakage caused by RTC cleanup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ville Syrjälä reports "The first time I run hwclock after rebooting I get this: open("/dev/rtc", O_RDONLY) = 3 ioctl(3, PHN_SET_REGS or RTC_UIE_ON, 0) = 0 select(4, [3], NULL, NULL, {10, 0}) = 0 (Timeout) ioctl(3, PHN_NOT_OH or RTC_UIE_OFF, 0) = 0 close(3) = 0 On all subsequent runs I get this: open("/dev/rtc", O_RDONLY) = 3 ioctl(3, PHN_SET_REGS or RTC_UIE_ON, 0) = -1 EINVAL (Invalid argument) ioctl(3, RTC_RD_TIME, 0x7ffd76b3ae70) = -1 EINVAL (Invalid argument) close(3) = 0" This was caused by a stupid typo in a patch that should have been a simple rename to move around contents of a header file, but accidentally wrote zeroes into the rtc rather than reading from it: 463a86304cae ("char/genrtc: x86: remove remnants of asm/rtc.h") Reported-by: Ville Syrjälä Tested-by: Jarkko Nikula Tested-by: Ville Syrjälä Signed-off-by: Arnd Bergmann Cc: Alessandro Zummo Cc: Alexandre Belloni Cc: Borislav Petkov Cc: Geert Uytterhoeven Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: rtc-linux@googlegroups.com Fixes: 463a86304cae ("char/genrtc: x86: remove remnants of asm/rtc.h") Link: http://lkml.kernel.org/r/20160809195528.1604312-1-arnd@arndb.de Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index ed16e58658a4..c6dfd801df97 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -1242,7 +1242,7 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id) memset(&curr_time, 0, sizeof(struct rtc_time)); if (hpet_rtc_flags & (RTC_UIE | RTC_AIE)) - mc146818_set_time(&curr_time); + mc146818_get_time(&curr_time); if (hpet_rtc_flags & RTC_UIE && curr_time.tm_sec != hpet_prev_update_sec) { From c7d2361f7524f365c1ae42f47880e3fa9efb2c2a Mon Sep 17 00:00:00 2001 From: Thomas Garnier Date: Tue, 9 Aug 2016 10:11:04 -0700 Subject: [PATCH 150/235] x86/mm/KASLR: Fix physical memory calculation on KASLR memory randomization Initialize KASLR memory randomization after max_pfn is initialized. Also ensure the size is rounded up. It could create problems on machines with more than 1Tb of memory on certain random addresses. Signed-off-by: Thomas Garnier Cc: Aleksey Makarov Cc: Andrew Morton Cc: Andy Lutomirski Cc: Baoquan He Cc: Borislav Petkov Cc: Borislav Petkov Cc: Brian Gerst Cc: Christian Borntraeger Cc: Dan Williams Cc: Dave Hansen Cc: Dave Young Cc: Denys Vlasenko Cc: Fabian Frederick Cc: H. Peter Anvin Cc: Joerg Roedel Cc: Josh Poimboeuf Cc: Kees Cook Cc: Linus Torvalds Cc: Lv Zheng Cc: Mark Salter Cc: Peter Zijlstra Cc: Rafael J . Wysocki Cc: Thomas Gleixner Cc: Toshi Kani Cc: kernel-hardening@lists.openwall.com Fixes: 021182e52fe0 ("Enable KASLR for physical mapping memory regions") Link: http://lkml.kernel.org/r/1470762665-88032-1-git-send-email-thgarnie@google.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 8 ++++++-- arch/x86/mm/kaslr.c | 2 +- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 991b77986d57..95cf31c9f4ec 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -936,8 +936,6 @@ void __init setup_arch(char **cmdline_p) x86_init.oem.arch_setup(); - kernel_randomize_memory(); - iomem_resource.end = (1ULL << boot_cpu_data.x86_phys_bits) - 1; setup_memory_map(); parse_setup_data(); @@ -1055,6 +1053,12 @@ void __init setup_arch(char **cmdline_p) max_possible_pfn = max_pfn; + /* + * Define random base addresses for memory sections after max_pfn is + * defined and before each memory section base is used. + */ + kernel_randomize_memory(); + #ifdef CONFIG_X86_32 /* max_low_pfn get updated here */ find_low_pfn_range(); diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index 26dccd6c0df1..ec8654f117d8 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -97,7 +97,7 @@ void __init kernel_randomize_memory(void) * add padding if needed (especially for memory hotplug support). */ BUG_ON(kaslr_regions[0].base != &page_offset_base); - memory_tb = ((max_pfn << PAGE_SHIFT) >> TB_SHIFT) + + memory_tb = DIV_ROUND_UP(max_pfn << PAGE_SHIFT, 1UL << TB_SHIFT) + CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING; /* Adapt phyiscal memory region size based on available memory */ From fb754f958f8e46202c1efd7f66d5b3db1208117d Mon Sep 17 00:00:00 2001 From: Thomas Garnier Date: Tue, 9 Aug 2016 10:11:05 -0700 Subject: [PATCH 151/235] x86/mm/KASLR: Increase BRK pages for KASLR memory randomization Default implementation expects 6 pages maximum are needed for low page allocations. If KASLR memory randomization is enabled, the worse case of e820 layout would require 12 pages (no large pages). It is due to the PUD level randomization and the variable e820 memory layout. This bug was found while doing extensive testing of KASLR memory randomization on different type of hardware. Signed-off-by: Thomas Garnier Cc: Aleksey Makarov Cc: Andrew Morton Cc: Andy Lutomirski Cc: Baoquan He Cc: Borislav Petkov Cc: Borislav Petkov Cc: Brian Gerst Cc: Christian Borntraeger Cc: Dan Williams Cc: Dave Hansen Cc: Dave Young Cc: Denys Vlasenko Cc: Fabian Frederick Cc: H. Peter Anvin Cc: Joerg Roedel Cc: Josh Poimboeuf Cc: Kees Cook Cc: Linus Torvalds Cc: Lv Zheng Cc: Mark Salter Cc: Peter Zijlstra Cc: Rafael J . Wysocki Cc: Thomas Gleixner Cc: Toshi Kani Cc: kernel-hardening@lists.openwall.com Fixes: 021182e52fe0 ("Enable KASLR for physical mapping memory regions") Link: http://lkml.kernel.org/r/1470762665-88032-2-git-send-email-thgarnie@google.com Signed-off-by: Ingo Molnar --- arch/x86/mm/init.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 620928903be3..d28a2d741f9e 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -122,8 +122,18 @@ __ref void *alloc_low_pages(unsigned int num) return __va(pfn << PAGE_SHIFT); } -/* need 3 4k for initial PMD_SIZE, 3 4k for 0-ISA_END_ADDRESS */ -#define INIT_PGT_BUF_SIZE (6 * PAGE_SIZE) +/* + * By default need 3 4k for initial PMD_SIZE, 3 4k for 0-ISA_END_ADDRESS. + * With KASLR memory randomization, depending on the machine e820 memory + * and the PUD alignment. We may need twice more pages when KASLR memory + * randomization is enabled. + */ +#ifndef CONFIG_RANDOMIZE_MEMORY +#define INIT_PGD_PAGE_COUNT 6 +#else +#define INIT_PGD_PAGE_COUNT 12 +#endif +#define INIT_PGT_BUF_SIZE (INIT_PGD_PAGE_COUNT * PAGE_SIZE) RESERVE_BRK(early_pgt_alloc, INIT_PGT_BUF_SIZE); void __init early_alloc_pgt_buf(void) { From 164af597ce945751e2dcd53d0a86e84203a6d117 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 9 Aug 2016 22:43:46 +1000 Subject: [PATCH 152/235] powerpc/Makefile: Use cflags-y/aflags-y for setting endian options When we introduced the little endian support, we added the endian flags to CC directly using override. I don't know the history of why we did that, I suspect no one does. Although this mostly works, it has one bug, which is that CROSS32CC doesn't get -mbig-endian. That means when the compiler is little endian by default and the user is building big endian, vdso32 is incorrectly compiled as little endian and the kernel fails to build. Instead we can add the endian flags to cflags-y/aflags-y, and then append those to KBUILD_CFLAGS/KBUILD_AFLAGS. This has the advantage of being 1) less ugly, 2) the documented way of adding flags in the arch Makefile and 3) it fixes building vdso32 with a LE toolchain. Signed-off-by: Michael Ellerman --- arch/powerpc/Makefile | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index ca254546cd05..1934707bf321 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -66,29 +66,28 @@ endif UTS_MACHINE := $(OLDARCH) ifeq ($(CONFIG_CPU_LITTLE_ENDIAN),y) -override CC += -mlittle-endian -ifneq ($(cc-name),clang) -override CC += -mno-strict-align -endif -override AS += -mlittle-endian override LD += -EL -override CROSS32CC += -mlittle-endian override CROSS32AS += -mlittle-endian LDEMULATION := lppc GNUTARGET := powerpcle MULTIPLEWORD := -mno-multiple KBUILD_CFLAGS_MODULE += $(call cc-option,-mno-save-toc-indirect) else -ifeq ($(call cc-option-yn,-mbig-endian),y) -override CC += -mbig-endian -override AS += -mbig-endian -endif override LD += -EB LDEMULATION := ppc GNUTARGET := powerpc MULTIPLEWORD := -mmultiple endif +cflags-$(CONFIG_CPU_BIG_ENDIAN) += $(call cc-option,-mbig-endian) +cflags-$(CONFIG_CPU_LITTLE_ENDIAN) += -mlittle-endian +ifneq ($(cc-name),clang) + cflags-$(CONFIG_CPU_LITTLE_ENDIAN) += -mno-strict-align +endif + +aflags-$(CONFIG_CPU_BIG_ENDIAN) += $(call cc-option,-mbig-endian) +aflags-$(CONFIG_CPU_LITTLE_ENDIAN) += -mlittle-endian + ifeq ($(HAS_BIARCH),y) override AS += -a$(CONFIG_WORD_SIZE) override LD += -m elf$(CONFIG_WORD_SIZE)$(LDEMULATION) @@ -232,6 +231,9 @@ cpu-as-$(CONFIG_E200) += -Wa,-me200 KBUILD_AFLAGS += $(cpu-as-y) KBUILD_CFLAGS += $(cpu-as-y) +KBUILD_AFLAGS += $(aflags-y) +KBUILD_CFLAGS += $(cflags-y) + head-y := arch/powerpc/kernel/head_$(CONFIG_WORD_SIZE).o head-$(CONFIG_8xx) := arch/powerpc/kernel/head_8xx.o head-$(CONFIG_40x) := arch/powerpc/kernel/head_40x.o From b9a4a0d02c5b8d9a1397c11d741d2a1a56381178 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Tue, 9 Aug 2016 22:17:29 +1000 Subject: [PATCH 153/235] powerpc/vdso: Fix build rules to rebuild vdsos correctly When using if_changed, we need to add FORCE as a dependency (see Documentation/kbuild/makefiles.txt) otherwise we don't get command line change checking amongst other things. This has resulted in vdsos not being rebuilt when switching between big and little endian. The vdso64/32ld commands have to be changed around to avoid pulling FORCE into the linker command line (code copied from x86). Signed-off-by: Nicholas Piggin Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/vdso32/Makefile | 6 +++--- arch/powerpc/kernel/vdso64/Makefile | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/powerpc/kernel/vdso32/Makefile b/arch/powerpc/kernel/vdso32/Makefile index cbabd143acae..78a7449bf489 100644 --- a/arch/powerpc/kernel/vdso32/Makefile +++ b/arch/powerpc/kernel/vdso32/Makefile @@ -30,7 +30,7 @@ CPPFLAGS_vdso32.lds += -P -C -Upowerpc $(obj)/vdso32_wrapper.o : $(obj)/vdso32.so # link rule for the .so file, .lds has to be first -$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) +$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) FORCE $(call if_changed,vdso32ld) # strip rule for the .so file @@ -39,12 +39,12 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # assembly rules for the .S files -$(obj-vdso32): %.o: %.S +$(obj-vdso32): %.o: %.S FORCE $(call if_changed_dep,vdso32as) # actual build commands quiet_cmd_vdso32ld = VDSO32L $@ - cmd_vdso32ld = $(CROSS32CC) $(c_flags) -Wl,-T $^ -o $@ + cmd_vdso32ld = $(CROSS32CC) $(c_flags) -o $@ -Wl,-T$(filter %.lds,$^) $(filter %.o,$^) quiet_cmd_vdso32as = VDSO32A $@ cmd_vdso32as = $(CROSS32CC) $(a_flags) -c -o $@ $< diff --git a/arch/powerpc/kernel/vdso64/Makefile b/arch/powerpc/kernel/vdso64/Makefile index c710802b8fb6..366ae09b14c1 100644 --- a/arch/powerpc/kernel/vdso64/Makefile +++ b/arch/powerpc/kernel/vdso64/Makefile @@ -23,7 +23,7 @@ CPPFLAGS_vdso64.lds += -P -C -U$(ARCH) $(obj)/vdso64_wrapper.o : $(obj)/vdso64.so # link rule for the .so file, .lds has to be first -$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) +$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) FORCE $(call if_changed,vdso64ld) # strip rule for the .so file @@ -32,12 +32,12 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # assembly rules for the .S files -$(obj-vdso64): %.o: %.S +$(obj-vdso64): %.o: %.S FORCE $(call if_changed_dep,vdso64as) # actual build commands quiet_cmd_vdso64ld = VDSO64L $@ - cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $^ -o $@ + cmd_vdso64ld = $(CC) $(c_flags) -o $@ -Wl,-T$(filter %.lds,$^) $(filter %.o,$^) quiet_cmd_vdso64as = VDSO64A $@ cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< From a52ff34e5ec61749c62c6618b76a9d6dbecee450 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 4 Aug 2016 22:38:36 +0200 Subject: [PATCH 154/235] ALSA: hda - Manage power well properly for resume For SKL and later Intel chips, we control the power well per codec basis via link_power callback since the commit [03b135cebc47: ALSA: hda - remove dependency on i915 power well for SKL]. However, there are a few exceptional cases where the gfx registers are accessed from the audio driver: namely the wakeup override bit toggling at (both system and runtime) resume. This seems causing a kernel warning when accessed during the power well down (and likely resulting in the bogus register accesses). This patch puts the proper power up / down sequence around the resume code so that the wakeup bit is fiddled properly while the power is up. (The other callback, sync_audio_rate, is used only in the PCM callback, so it's guaranteed in the power-on.) Also, by this proper power up/down, the instantaneous flip of wakeup bit in the resume callback that was introduced by the commit [033ea349a7cd: ALSA: hda - Fix Skylake codec timeout] becomes superfluous, as snd_hdac_display_power() already does it. So we can clean it up together. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96214 Fixes: 03b135cebc47 ('ALSA: hda - remove dependency on i915 power well for SKL') Cc: # v4.2+ Tested-by: Hans de Goede Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_intel.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 89dacf9b4e6c..160c7f713722 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -906,20 +906,23 @@ static int azx_resume(struct device *dev) struct snd_card *card = dev_get_drvdata(dev); struct azx *chip; struct hda_intel *hda; + struct hdac_bus *bus; if (!card) return 0; chip = card->private_data; hda = container_of(chip, struct hda_intel, chip); + bus = azx_bus(chip); if (chip->disabled || hda->init_failed || !chip->running) return 0; - if (chip->driver_caps & AZX_DCAPS_I915_POWERWELL - && hda->need_i915_power) { - snd_hdac_display_power(azx_bus(chip), true); - snd_hdac_i915_set_bclk(azx_bus(chip)); + if (chip->driver_caps & AZX_DCAPS_I915_POWERWELL) { + snd_hdac_display_power(bus, true); + if (hda->need_i915_power) + snd_hdac_i915_set_bclk(bus); } + if (chip->msi) if (pci_enable_msi(pci) < 0) chip->msi = 0; @@ -929,6 +932,11 @@ static int azx_resume(struct device *dev) hda_intel_init_chip(chip, true); + /* power down again for link-controlled chips */ + if ((chip->driver_caps & AZX_DCAPS_I915_POWERWELL) && + !hda->need_i915_power) + snd_hdac_display_power(bus, false); + snd_power_change_state(card, SNDRV_CTL_POWER_D0); trace_azx_resume(chip); @@ -1008,6 +1016,7 @@ static int azx_runtime_resume(struct device *dev) chip = card->private_data; hda = container_of(chip, struct hda_intel, chip); + bus = azx_bus(chip); if (chip->disabled || hda->init_failed) return 0; @@ -1015,15 +1024,9 @@ static int azx_runtime_resume(struct device *dev) return 0; if (chip->driver_caps & AZX_DCAPS_I915_POWERWELL) { - bus = azx_bus(chip); - if (hda->need_i915_power) { - snd_hdac_display_power(bus, true); + snd_hdac_display_power(bus, true); + if (hda->need_i915_power) snd_hdac_i915_set_bclk(bus); - } else { - /* toggle codec wakeup bit for STATESTS read */ - snd_hdac_set_codec_wakeup(bus, true); - snd_hdac_set_codec_wakeup(bus, false); - } } /* Read STATESTS before controller reset */ @@ -1043,6 +1046,11 @@ static int azx_runtime_resume(struct device *dev) azx_writew(chip, WAKEEN, azx_readw(chip, WAKEEN) & ~STATESTS_INT_MASK); + /* power down again for link-controlled chips */ + if ((chip->driver_caps & AZX_DCAPS_I915_POWERWELL) && + !hda->need_i915_power) + snd_hdac_display_power(bus, false); + trace_azx_runtime_resume(chip); return 0; } From ca49e64f0cb1368fc666a53b16b45d4505763d9c Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 29 Jul 2016 20:48:09 +1000 Subject: [PATCH 155/235] selftests/powerpc: Specify we expect to build with std=gnu99 We have some tests that assume we're using std=gnu99, which is fine on most compilers, but some old compilers use a different default. So make it explicit that we want to use std=gnu99. Signed-off-by: Michael Ellerman --- tools/testing/selftests/powerpc/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/powerpc/Makefile b/tools/testing/selftests/powerpc/Makefile index 3c40c9d0e6c7..1cc6d64c39b7 100644 --- a/tools/testing/selftests/powerpc/Makefile +++ b/tools/testing/selftests/powerpc/Makefile @@ -8,7 +8,7 @@ ifeq ($(ARCH),powerpc) GIT_VERSION = $(shell git describe --always --long --dirty || echo "unknown") -CFLAGS := -Wall -O2 -Wall -Werror -DGIT_VERSION='"$(GIT_VERSION)"' -I$(CURDIR) $(CFLAGS) +CFLAGS := -std=gnu99 -Wall -O2 -Wall -Werror -DGIT_VERSION='"$(GIT_VERSION)"' -I$(CURDIR) $(CFLAGS) export CFLAGS From 5cf0791da5c162ebc14b01eb01631cfa7ed4fa6e Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Fri, 5 Aug 2016 15:37:39 +0200 Subject: [PATCH 156/235] x86/mm: Disable preemption during CR3 read+write MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There's a subtle preemption race on UP kernels: Usually current->mm (and therefore mm->pgd) stays the same during the lifetime of a task so it does not matter if a task gets preempted during the read and write of the CR3. But then, there is this scenario on x86-UP: TaskA is in do_exit() and exit_mm() sets current->mm = NULL followed by: -> mmput() -> exit_mmap() -> tlb_finish_mmu() -> tlb_flush_mmu() -> tlb_flush_mmu_tlbonly() -> tlb_flush() -> flush_tlb_mm_range() -> __flush_tlb_up() -> __flush_tlb() -> __native_flush_tlb() At this point current->mm is NULL but current->active_mm still points to the "old" mm. Let's preempt taskA _after_ native_read_cr3() by taskB. TaskB has its own mm so CR3 has changed. Now preempt back to taskA. TaskA has no ->mm set so it borrows taskB's mm and so CR3 remains unchanged. Once taskA gets active it continues where it was interrupted and that means it writes its old CR3 value back. Everything is fine because userland won't need its memory anymore. Now the fun part: Let's preempt taskA one more time and get back to taskB. This time switch_mm() won't do a thing because oldmm (->active_mm) is the same as mm (as per context_switch()). So we remain with a bad CR3 / PGD and return to userland. The next thing that happens is handle_mm_fault() with an address for the execution of its code in userland. handle_mm_fault() realizes that it has a PTE with proper rights so it returns doing nothing. But the CPU looks at the wrong PGD and insists that something is wrong and faults again. And again. And one more time… This pagefault circle continues until the scheduler gets tired of it and puts another task on the CPU. It gets little difficult if the task is a RT task with a high priority. The system will either freeze or it gets fixed by the software watchdog thread which usually runs at RT-max prio. But waiting for the watchdog will increase the latency of the RT task which is no good. Fix this by disabling preemption across the critical code section. Signed-off-by: Sebastian Andrzej Siewior Acked-by: Peter Zijlstra (Intel) Acked-by: Rik van Riel Acked-by: Andy Lutomirski Cc: Borislav Petkov Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1470404259-26290-1-git-send-email-bigeasy@linutronix.de [ Prettified the changelog. ] Signed-off-by: Ingo Molnar --- arch/x86/include/asm/tlbflush.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index 4e5be94e079a..6fa85944af83 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -135,7 +135,14 @@ static inline void cr4_set_bits_and_update_boot(unsigned long mask) static inline void __native_flush_tlb(void) { + /* + * If current->mm == NULL then we borrow a mm which may change during a + * task switch and therefore we must not be preempted while we write CR3 + * back: + */ + preempt_disable(); native_write_cr3(native_read_cr3()); + preempt_enable(); } static inline void __native_flush_tlb_global_irq_disabled(void) From 3e035305875cfa8a58c1ca573d0cfa6a7f201f27 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 3 Aug 2016 19:14:29 +0200 Subject: [PATCH 157/235] x86/entry: Clarify the RF saving/restoring situation with SYSCALL/SYSRET Clarify why exactly RF cannot be restored properly by SYSRET to avoid confusion. No functionality change. Signed-off-by: Borislav Petkov Acked-by: Andy Lutomirski Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160803171429.GA2590@nazgul.tnic Signed-off-by: Ingo Molnar --- arch/x86/entry/entry_64.S | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 9f85827db24e..d172c619c449 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -288,11 +288,15 @@ return_from_SYSCALL_64: jne opportunistic_sysret_failed /* - * SYSRET can't restore RF. SYSRET can restore TF, but unlike IRET, - * restoring TF results in a trap from userspace immediately after - * SYSRET. This would cause an infinite loop whenever #DB happens - * with register state that satisfies the opportunistic SYSRET - * conditions. For example, single-stepping this user code: + * SYSCALL clears RF when it saves RFLAGS in R11 and SYSRET cannot + * restore RF properly. If the slowpath sets it for whatever reason, we + * need to restore it correctly. + * + * SYSRET can restore TF, but unlike IRET, restoring TF results in a + * trap from userspace immediately after SYSRET. This would cause an + * infinite loop whenever #DB happens with register state that satisfies + * the opportunistic SYSRET conditions. For example, single-stepping + * this user code: * * movq $stuck_here, %rcx * pushfq From 054f621fd5b1c7245710f5d3935c94ce6ae4b3b7 Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 1 Aug 2016 13:40:50 -0500 Subject: [PATCH 158/235] x86/platform/UV: Fix problem with UV4 Socket IDs not being contiguous The UV4 Socket IDs are not guaranteed to equate to Node values which can cause the GAM (Global Addressable Memory) table lookups to fail. Fix this by using an independent index into the GAM table instead of the Socket ID to reference the base address. Tested-by: Frank Ramsay Tested-by: John Estabrook Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Nathan Zimmer Cc: Alex Thorlton Cc: Andrew Banman Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Russ Anderson Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160801184050.048755337@asylum.americas.sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/x2apic_uv_x.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 09b59adaea3f..d9187336aa2e 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -325,7 +325,7 @@ static __init void build_uv_gr_table(void) struct uv_gam_range_entry *gre = uv_gre_table; struct uv_gam_range_s *grt; unsigned long last_limit = 0, ram_limit = 0; - int bytes, i, sid, lsid = -1; + int bytes, i, sid, lsid = -1, indx = 0, lindx = -1; if (!gre) return; @@ -356,11 +356,12 @@ static __init void build_uv_gr_table(void) } sid = gre->sockid - _min_socket; if (lsid < sid) { /* new range */ - grt = &_gr_table[sid]; - grt->base = lsid; + grt = &_gr_table[indx]; + grt->base = lindx; grt->nasid = gre->nasid; grt->limit = last_limit = gre->limit; lsid = sid; + lindx = indx++; continue; } if (lsid == sid && !ram_limit) { /* update range */ @@ -371,7 +372,7 @@ static __init void build_uv_gr_table(void) } if (!ram_limit) { /* non-contiguous ram range */ grt++; - grt->base = sid - 1; + grt->base = lindx; grt->nasid = gre->nasid; grt->limit = last_limit = gre->limit; continue; From e363d24c2b997c421476c6aa00547edadf678efe Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 1 Aug 2016 13:40:51 -0500 Subject: [PATCH 159/235] x86/platform/UV: Fix bug with iounmap() of the UV4 EFI System Table causing a crash Save the uv_systab::size field before doing the iounmap() of the struct pointer, to avoid a NULL dereference crash. Tested-by: Frank Ramsay Tested-by: John Estabrook Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Nathan Zimmer Cc: Alex Thorlton Cc: Andrew Banman Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Russ Anderson Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160801184050.250424783@asylum.americas.sgi.com Signed-off-by: Ingo Molnar --- arch/x86/platform/uv/bios_uv.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/x86/platform/uv/bios_uv.c b/arch/x86/platform/uv/bios_uv.c index 66b2166ea4a1..4e9fd1378aec 100644 --- a/arch/x86/platform/uv/bios_uv.c +++ b/arch/x86/platform/uv/bios_uv.c @@ -199,12 +199,14 @@ void uv_bios_init(void) return; } + /* Starting with UV4 the UV systab size is variable */ if (uv_systab->revision >= UV_SYSTAB_VERSION_UV4) { + int size = uv_systab->size; + iounmap(uv_systab); - uv_systab = ioremap(efi.uv_systab, uv_systab->size); + uv_systab = ioremap(efi.uv_systab, size); if (!uv_systab) { - pr_err("UV: UVsystab: ioremap(%d) failed!\n", - uv_systab->size); + pr_err("UV: UVsystab: ioremap(%d) failed!\n", size); return; } } From 22ac2bca92f2d92c6495248d65ff648182df428d Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 1 Aug 2016 13:40:52 -0500 Subject: [PATCH 160/235] x86/platform/UV: Fix problem with UV4 BIOS providing incorrect PXM values There are some circumstances where the UV4 BIOS cannot provide the correct Proximity Node values to associate with specific Sockets and Physical Nodes. The decision was made to remove these values from BIOS and for the kernel to get these values from the standard ACPI tables. Tested-by: Frank Ramsay Tested-by: John Estabrook Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Nathan Zimmer Cc: Alex Thorlton Cc: Andrew Banman Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Russ Anderson Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160801184050.414210079@asylum.americas.sgi.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/uv/bios.h | 5 +++-- arch/x86/kernel/apic/x2apic_uv_x.c | 28 ++++++++++------------------ 2 files changed, 13 insertions(+), 20 deletions(-) diff --git a/arch/x86/include/asm/uv/bios.h b/arch/x86/include/asm/uv/bios.h index c852590254d5..e652a7cc6186 100644 --- a/arch/x86/include/asm/uv/bios.h +++ b/arch/x86/include/asm/uv/bios.h @@ -79,7 +79,7 @@ struct uv_gam_range_entry { u16 nasid; /* HNasid */ u16 sockid; /* Socket ID, high bits of APIC ID */ u16 pnode; /* Index to MMR and GRU spaces */ - u32 pxm; /* ACPI proximity domain number */ + u32 unused2; u32 limit; /* PA bits 56:26 (UV_GAM_RANGE_SHFT) */ }; @@ -88,7 +88,8 @@ struct uv_gam_range_entry { #define UV_SYSTAB_VERSION_UV4 0x400 /* UV4 BIOS base version */ #define UV_SYSTAB_VERSION_UV4_1 0x401 /* + gpa_shift */ #define UV_SYSTAB_VERSION_UV4_2 0x402 /* + TYPE_NVRAM/WINDOW/MBOX */ -#define UV_SYSTAB_VERSION_UV4_LATEST UV_SYSTAB_VERSION_UV4_2 +#define UV_SYSTAB_VERSION_UV4_3 0x403 /* - GAM Range PXM Value */ +#define UV_SYSTAB_VERSION_UV4_LATEST UV_SYSTAB_VERSION_UV4_3 #define UV_SYSTAB_TYPE_UNUSED 0 /* End of table (offset == 0) */ #define UV_SYSTAB_TYPE_GAM_PARAMS 1 /* GAM PARAM conversions */ diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index d9187336aa2e..6aa0545879bb 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -1156,19 +1156,18 @@ static void __init decode_gam_rng_tbl(unsigned long ptr) for (; gre->type != UV_GAM_RANGE_TYPE_UNUSED; gre++) { if (!index) { pr_info("UV: GAM Range Table...\n"); - pr_info("UV: # %20s %14s %5s %4s %5s %3s %2s %3s\n", + pr_info("UV: # %20s %14s %5s %4s %5s %3s %2s\n", "Range", "", "Size", "Type", "NASID", - "SID", "PN", "PXM"); + "SID", "PN"); } pr_info( - "UV: %2d: 0x%014lx-0x%014lx %5luG %3d %04x %02x %02x %3d\n", + "UV: %2d: 0x%014lx-0x%014lx %5luG %3d %04x %02x %02x\n", index++, (unsigned long)lgre << UV_GAM_RANGE_SHFT, (unsigned long)gre->limit << UV_GAM_RANGE_SHFT, ((unsigned long)(gre->limit - lgre)) >> (30 - UV_GAM_RANGE_SHFT), /* 64M -> 1G */ - gre->type, gre->nasid, gre->sockid, - gre->pnode, gre->pxm); + gre->type, gre->nasid, gre->sockid, gre->pnode); lgre = gre->limit; if (sock_min > gre->sockid) @@ -1287,7 +1286,7 @@ static void __init build_socket_tables(void) _pnode_to_socket[i] = SOCK_EMPTY; /* fill in pnode/node/addr conversion list values */ - pr_info("UV: GAM Building socket/pnode/pxm conversion tables\n"); + pr_info("UV: GAM Building socket/pnode conversion tables\n"); for (; gre->type != UV_GAM_RANGE_TYPE_UNUSED; gre++) { if (gre->type == UV_GAM_RANGE_TYPE_HOLE) continue; @@ -1295,20 +1294,18 @@ static void __init build_socket_tables(void) if (_socket_to_pnode[i] != SOCK_EMPTY) continue; /* duplicate */ _socket_to_pnode[i] = gre->pnode; - _socket_to_node[i] = gre->pxm; i = gre->pnode - minpnode; _pnode_to_socket[i] = gre->sockid; pr_info( - "UV: sid:%02x type:%d nasid:%04x pn:%02x pxm:%2d pn2s:%2x\n", + "UV: sid:%02x type:%d nasid:%04x pn:%02x pn2s:%2x\n", gre->sockid, gre->type, gre->nasid, _socket_to_pnode[gre->sockid - minsock], - _socket_to_node[gre->sockid - minsock], _pnode_to_socket[gre->pnode - minpnode]); } - /* check socket -> node values */ + /* Set socket -> node values */ lnid = -1; for_each_present_cpu(cpu) { int nid = cpu_to_node(cpu); @@ -1319,14 +1316,9 @@ static void __init build_socket_tables(void) lnid = nid; apicid = per_cpu(x86_cpu_to_apicid, cpu); sockid = apicid >> uv_cpuid.socketid_shift; - i = sockid - minsock; - - if (nid != _socket_to_node[i]) { - pr_warn( - "UV: %02x: type:%d socket:%02x PXM:%02x != node:%2d\n", - i, sockid, gre->type, _socket_to_node[i], nid); - _socket_to_node[i] = nid; - } + _socket_to_node[sockid - minsock] = nid; + pr_info("UV: sid:%02x: apicid:%04x node:%2d\n", + sockid, apicid, nid); } /* Setup physical blade to pnode translation from GAM Range Table */ From 5a52e8f822374bebc702bb2688ed8b5515bbb55b Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 1 Aug 2016 13:40:53 -0500 Subject: [PATCH 161/235] x86/platform/UV: Fix kernel panic running RHEL kdump kernel on UV systems The latest UV kernel support panics when RHEL7 kexec's the kdump kernel to make a dumpfile. This patch fixes the problem by turning off all UV support if NUMA is off. Tested-by: Frank Ramsay Tested-by: John Estabrook Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Nathan Zimmer Cc: Alex Thorlton Cc: Andrew Banman Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Russ Anderson Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160801184050.577755634@asylum.americas.sgi.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/x2apic_uv_x.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 6aa0545879bb..cb0673c1e940 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -223,6 +223,11 @@ static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) if (strncmp(oem_id, "SGI", 3) != 0) return 0; + if (numa_off) { + pr_err("UV: NUMA is off, disabling UV support\n"); + return 0; + } + /* Setup early hub type field in uv_hub_info for Node 0 */ uv_cpu_info->p_uv_hub_info = &uv_hub_info_node0; From 5e44258d168b2bdee51d9e2e1f1f4726ff9775cd Mon Sep 17 00:00:00 2001 From: Valdis Kletnieks Date: Sun, 31 Jul 2016 23:24:50 -0400 Subject: [PATCH 162/235] x86/build: Reduce the W=1 warnings noise when compiling x86 syscall tables Building an X86_64 kernel with W=1 throws a total of 9,948 lines of warnings of this form for both 32-bit and 64-bit syscall tables. Given that the entire rest of the build for my config only generates 8,375 lines of output, this is a big reduction in the warnings generated. The warnings follow this pattern: ./arch/x86/include/generated/asm/syscalls_32.h:885:21: warning: initialized field overwritten [-Woverride-init] __SYSCALL_I386(379, compat_sys_pwritev2, ) ^ arch/x86/entry/syscall_32.c:13:46: note: in definition of macro '__SYSCALL_I386' #define __SYSCALL_I386(nr, sym, qual) [nr] = sym, ^~~ ./arch/x86/include/generated/asm/syscalls_32.h:885:21: note: (near initialization for 'ia32_sys_call_table[379]') __SYSCALL_I386(379, compat_sys_pwritev2, ) ^ arch/x86/entry/syscall_32.c:13:46: note: in definition of macro '__SYSCALL_I386' #define __SYSCALL_I386(nr, sym, qual) [nr] = sym, Since we intentionally build the syscall tables this way, ignore that one warning in the two files. Signed-off-by: Valdis Kletnieks Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/7464.1470021890@turing-police.cc.vt.edu Signed-off-by: Ingo Molnar --- arch/x86/entry/Makefile | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/entry/Makefile b/arch/x86/entry/Makefile index fe91c25092da..77f28ce9c646 100644 --- a/arch/x86/entry/Makefile +++ b/arch/x86/entry/Makefile @@ -5,6 +5,8 @@ OBJECT_FILES_NON_STANDARD_entry_$(BITS).o := y OBJECT_FILES_NON_STANDARD_entry_64_compat.o := y +CFLAGS_syscall_64.o += -Wno-override-init +CFLAGS_syscall_32.o += -Wno-override-init obj-y := entry_$(BITS).o thunk_$(BITS).o syscall_$(BITS).o obj-y += common.o From b79daf85899215d5ede3641806db2e2a77b776b4 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Wed, 27 Jul 2016 16:20:40 -0700 Subject: [PATCH 163/235] x86/mm/pkeys: Fix compact mode by removing protection keys' XSAVE buffer manipulation The Memory Protection Keys "rights register" (PKRU) is XSAVE-managed, and is saved/restored along with the FPU state. When kernel code accesses FPU regsisters, it does a delicate dance with preempt. Otherwise, the context switching code can get confused as to whether the most up-to-date state is in the registers themselves or in the XSAVE buffer. But, PKRU is not a normal FPU register. Using it does not generate the normal device-not-available (#NM) exceptions which means we can not manage it lazily, and the kernel completley disallows using lazy mode when it is enabled. The dance with preempt *only* occurs when managing the FPU lazily. Since we never manage PKRU lazily, we do not have to do the dance with preempt; we can access it directly. Doing it this way saves a ton of complicated code (and is faster too). Further, the XSAVES reenabling failed to patch a bit of code in fpu__xfeature_set_state() the checked for compacted buffers. That check caused fpu__xfeature_set_state() to silently refuse to work when the kernel is using compacted XSAVE buffers. This broke execute-only and future pkey_mprotect() support when using compact XSAVE buffers. But, removing fpu__xfeature_set_state() gets rid of this issue, in addition to the nice cleanup and speedup. This fixes the same thing as a fix that Sai posted: https://lkml.org/lkml/2016/7/25/637 The fix that he posted is a much more obviously correct, but I think we should just do this instead. Reported-by: Sai Praneeth Prakhya Signed-off-by: Dave Hansen Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Dave Hansen Cc: Dave Hansen Cc: Fenghua Yu Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Quentin Casasnovas Cc: Ravi Shankar Cc: Thomas Gleixner Cc: Yu-Cheng Yu Link: http://lkml.kernel.org/r/20160727232040.7D060DAD@viggo.jf.intel.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/fpu/xstate.c | 138 +++++------------------------------ 1 file changed, 17 insertions(+), 121 deletions(-) diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index 680049aa4593..01567aa87503 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -866,105 +866,17 @@ const void *get_xsave_field_ptr(int xsave_state) return get_xsave_addr(&fpu->state.xsave, xsave_state); } - -/* - * Set xfeatures (aka XSTATE_BV) bit for a feature that we want - * to take out of its "init state". This will ensure that an - * XRSTOR actually restores the state. - */ -static void fpu__xfeature_set_non_init(struct xregs_state *xsave, - int xstate_feature_mask) -{ - xsave->header.xfeatures |= xstate_feature_mask; -} - -/* - * This function is safe to call whether the FPU is in use or not. - * - * Note that this only works on the current task. - * - * Inputs: - * @xsave_state: state which is defined in xsave.h (e.g. XFEATURE_MASK_FP, - * XFEATURE_MASK_SSE, etc...) - * @xsave_state_ptr: a pointer to a copy of the state that you would - * like written in to the current task's FPU xsave state. This pointer - * must not be located in the current tasks's xsave area. - * Output: - * address of the state in the xsave area or NULL if the state - * is not present or is in its 'init state'. - */ -static void fpu__xfeature_set_state(int xstate_feature_mask, - void *xstate_feature_src, size_t len) -{ - struct xregs_state *xsave = ¤t->thread.fpu.state.xsave; - struct fpu *fpu = ¤t->thread.fpu; - void *dst; - - if (!boot_cpu_has(X86_FEATURE_XSAVE)) { - WARN_ONCE(1, "%s() attempted with no xsave support", __func__); - return; - } - - /* - * Tell the FPU code that we need the FPU state to be in - * 'fpu' (not in the registers), and that we need it to - * be stable while we write to it. - */ - fpu__current_fpstate_write_begin(); - - /* - * This method *WILL* *NOT* work for compact-format - * buffers. If the 'xstate_feature_mask' is unset in - * xcomp_bv then we may need to move other feature state - * "up" in the buffer. - */ - if (xsave->header.xcomp_bv & xstate_feature_mask) { - WARN_ON_ONCE(1); - goto out; - } - - /* find the location in the xsave buffer of the desired state */ - dst = __raw_xsave_addr(&fpu->state.xsave, xstate_feature_mask); - - /* - * Make sure that the pointer being passed in did not - * come from the xsave buffer itself. - */ - WARN_ONCE(xstate_feature_src == dst, "set from xsave buffer itself"); - - /* put the caller-provided data in the location */ - memcpy(dst, xstate_feature_src, len); - - /* - * Mark the xfeature so that the CPU knows there is state - * in the buffer now. - */ - fpu__xfeature_set_non_init(xsave, xstate_feature_mask); -out: - /* - * We are done writing to the 'fpu'. Reenable preeption - * and (possibly) move the fpstate back in to the fpregs. - */ - fpu__current_fpstate_write_end(); -} - #define NR_VALID_PKRU_BITS (CONFIG_NR_PROTECTION_KEYS * 2) #define PKRU_VALID_MASK (NR_VALID_PKRU_BITS - 1) /* - * This will go out and modify the XSAVE buffer so that PKRU is - * set to a particular state for access to 'pkey'. - * - * PKRU state does affect kernel access to user memory. We do - * not modfiy PKRU *itself* here, only the XSAVE state that will - * be restored in to PKRU when we return back to userspace. + * This will go out and modify PKRU register to set the access + * rights for @pkey to @init_val. */ int arch_set_user_pkey_access(struct task_struct *tsk, int pkey, unsigned long init_val) { - struct xregs_state *xsave = &tsk->thread.fpu.state.xsave; - struct pkru_state *old_pkru_state; - struct pkru_state new_pkru_state; + u32 old_pkru; int pkey_shift = (pkey * PKRU_BITS_PER_PKEY); u32 new_pkru_bits = 0; @@ -974,6 +886,15 @@ int arch_set_user_pkey_access(struct task_struct *tsk, int pkey, */ if (!boot_cpu_has(X86_FEATURE_OSPKE)) return -EINVAL; + /* + * For most XSAVE components, this would be an arduous task: + * brining fpstate up to date with fpregs, updating fpstate, + * then re-populating fpregs. But, for components that are + * never lazily managed, we can just access the fpregs + * directly. PKRU is never managed lazily, so we can just + * manipulate it directly. Make sure it stays that way. + */ + WARN_ON_ONCE(!use_eager_fpu()); /* Set the bits we need in PKRU: */ if (init_val & PKEY_DISABLE_ACCESS) @@ -984,37 +905,12 @@ int arch_set_user_pkey_access(struct task_struct *tsk, int pkey, /* Shift the bits in to the correct place in PKRU for pkey: */ new_pkru_bits <<= pkey_shift; - /* Locate old copy of the state in the xsave buffer: */ - old_pkru_state = get_xsave_addr(xsave, XFEATURE_MASK_PKRU); + /* Get old PKRU and mask off any old bits in place: */ + old_pkru = read_pkru(); + old_pkru &= ~((PKRU_AD_BIT|PKRU_WD_BIT) << pkey_shift); - /* - * When state is not in the buffer, it is in the init - * state, set it manually. Otherwise, copy out the old - * state. - */ - if (!old_pkru_state) - new_pkru_state.pkru = 0; - else - new_pkru_state.pkru = old_pkru_state->pkru; - - /* Mask off any old bits in place: */ - new_pkru_state.pkru &= ~((PKRU_AD_BIT|PKRU_WD_BIT) << pkey_shift); - - /* Set the newly-requested bits: */ - new_pkru_state.pkru |= new_pkru_bits; - - /* - * We could theoretically live without zeroing pkru.pad. - * The current XSAVE feature state definition says that - * only bytes 0->3 are used. But we do not want to - * chance leaking kernel stack out to userspace in case a - * memcpy() of the whole xsave buffer was done. - * - * They're in the same cacheline anyway. - */ - new_pkru_state.pad = 0; - - fpu__xfeature_set_state(XFEATURE_MASK_PKRU, &new_pkru_state, sizeof(new_pkru_state)); + /* Write old part along with new part: */ + write_pkru(old_pkru | new_pkru_bits); return 0; } From aec742e8e1cf05114530bdef8051238b6dbb4052 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 10 Aug 2016 08:45:11 -0700 Subject: [PATCH 164/235] get_maintainer: Don't check if STDIN exists in a VCS repository If get_maintainer is not given any filename arguments on the command line, the standard input is read for a patch. But checking if a VCS has a file named &STDIN is not a good idea and fails. Verify the nominal input file is not &STDIN before checking the VCS. Fixes: 4cad35a7ca69 ("get_maintainer.pl: reduce need for command-line option -f") Reported-by: Christopher Covington Signed-off-by: Joe Perches Signed-off-by: Linus Torvalds --- scripts/get_maintainer.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/get_maintainer.pl b/scripts/get_maintainer.pl index 122fcdaf42c8..49a00d54b835 100755 --- a/scripts/get_maintainer.pl +++ b/scripts/get_maintainer.pl @@ -432,7 +432,7 @@ foreach my $file (@ARGV) { die "$P: file '${file}' not found\n"; } } - if ($from_filename || vcs_file_exists($file)) { + if ($from_filename || ($file ne "&STDIN" && vcs_file_exists($file))) { $file =~ s/^\Q${cur_path}\E//; #strip any absolute path $file =~ s/^\Q${lk_path}\E//; #or the path to the lk tree push(@files, $file); From 7de249964f5578e67b99699c5f0b405738d820a2 Mon Sep 17 00:00:00 2001 From: Dave Weinstein Date: Thu, 28 Jul 2016 11:55:41 -0700 Subject: [PATCH 165/235] arm: oabi compat: add missing access checks Add access checks to sys_oabi_epoll_wait() and sys_oabi_semtimedop(). This fixes CVE-2016-3857, a local privilege escalation under CONFIG_OABI_COMPAT. Cc: stable@vger.kernel.org Reported-by: Chiachih Wu Reviewed-by: Kees Cook Reviewed-by: Nicolas Pitre Signed-off-by: Dave Weinstein Signed-off-by: Linus Torvalds --- arch/arm/kernel/sys_oabi-compat.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/arch/arm/kernel/sys_oabi-compat.c b/arch/arm/kernel/sys_oabi-compat.c index 087acb569b63..5f221acd21ae 100644 --- a/arch/arm/kernel/sys_oabi-compat.c +++ b/arch/arm/kernel/sys_oabi-compat.c @@ -279,8 +279,12 @@ asmlinkage long sys_oabi_epoll_wait(int epfd, mm_segment_t fs; long ret, err, i; - if (maxevents <= 0 || maxevents > (INT_MAX/sizeof(struct epoll_event))) + if (maxevents <= 0 || + maxevents > (INT_MAX/sizeof(*kbuf)) || + maxevents > (INT_MAX/sizeof(*events))) return -EINVAL; + if (!access_ok(VERIFY_WRITE, events, sizeof(*events) * maxevents)) + return -EFAULT; kbuf = kmalloc(sizeof(*kbuf) * maxevents, GFP_KERNEL); if (!kbuf) return -ENOMEM; @@ -317,6 +321,8 @@ asmlinkage long sys_oabi_semtimedop(int semid, if (nsops < 1 || nsops > SEMOPM) return -EINVAL; + if (!access_ok(VERIFY_READ, tsops, sizeof(*tsops) * nsops)) + return -EFAULT; sops = kmalloc(sizeof(*sops) * nsops, GFP_KERNEL); if (!sops) return -ENOMEM; From 549fba3a615a5dc6219bb41b3ee3951173fcf1e6 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Jun 2016 16:21:19 +0200 Subject: [PATCH 166/235] ARM: don't include removed directories Three platforms used to have header files in include/mach that are now all gone, but the removed directories are still being included, which leads to -Wmissing-include-dirs warnings. This removes the extra -I flags. Signed-off-by: Arnd Bergmann --- arch/arm/mach-mvebu/Makefile | 3 +-- arch/arm/mach-realview/Makefile | 3 +-- arch/arm/mach-s5pv210/Makefile | 2 +- 3 files changed, 3 insertions(+), 5 deletions(-) diff --git a/arch/arm/mach-mvebu/Makefile b/arch/arm/mach-mvebu/Makefile index e53c6cfcab51..6c6497e80a7b 100644 --- a/arch/arm/mach-mvebu/Makefile +++ b/arch/arm/mach-mvebu/Makefile @@ -1,5 +1,4 @@ -ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include \ - -I$(srctree)/arch/arm/plat-orion/include +ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/arch/arm/plat-orion/include AFLAGS_coherency_ll.o := -Wa,-march=armv7-a CFLAGS_pmsu.o := -march=armv7-a diff --git a/arch/arm/mach-realview/Makefile b/arch/arm/mach-realview/Makefile index dae8d86ef4cc..404882130956 100644 --- a/arch/arm/mach-realview/Makefile +++ b/arch/arm/mach-realview/Makefile @@ -1,8 +1,7 @@ # # Makefile for the linux kernel. # -ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include \ - -I$(srctree)/arch/arm/plat-versatile/include +ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/arch/arm/plat-versatile/include obj-y := core.o obj-$(CONFIG_REALVIEW_DT) += realview-dt.o diff --git a/arch/arm/mach-s5pv210/Makefile b/arch/arm/mach-s5pv210/Makefile index 72b9e9671507..fa7fb716e388 100644 --- a/arch/arm/mach-s5pv210/Makefile +++ b/arch/arm/mach-s5pv210/Makefile @@ -5,7 +5,7 @@ # # Licensed under GPLv2 -ccflags-$(CONFIG_ARCH_MULTIPLATFORM) += -I$(srctree)/$(src)/include -I$(srctree)/arch/arm/plat-samsung/include +ccflags-$(CONFIG_ARCH_MULTIPLATFORM) += -I$(srctree)/arch/arm/plat-samsung/include # Core From de8a06f674a6f9d781e5087e9217319c7b831821 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 9 Jun 2016 09:50:28 +0200 Subject: [PATCH 167/235] ARM: hide mach-*/ include for ARM_SINGLE_ARMV7M The machine specific header files are exported for traditional platforms, but not for the ones that use ARCH_MULTIPLATFORM, as they could conflict with one another. In case of ARM_SINGLE_ARMV7M, we end up also exporting them, but that appears to be a mistake, and we should treat it the same way as ARCH_MULTIPLATFORM here. 'make W=1' warns about this because it passes -Wmissing-includes to gcc and the directories are not actually present. Signed-off-by: Arnd Bergmann --- arch/arm/Makefile | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/Makefile b/arch/arm/Makefile index 56ea5c60b318..61f6ccc19cfa 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -260,12 +260,14 @@ machdirs := $(patsubst %,arch/arm/mach-%/,$(machine-y)) platdirs := $(patsubst %,arch/arm/plat-%/,$(sort $(plat-y))) ifneq ($(CONFIG_ARCH_MULTIPLATFORM),y) +ifneq ($(CONFIG_ARM_SINGLE_ARMV7M),y) ifeq ($(KBUILD_SRC),) KBUILD_CPPFLAGS += $(patsubst %,-I%include,$(machdirs) $(platdirs)) else KBUILD_CPPFLAGS += $(patsubst %,-I$(srctree)/%include,$(machdirs) $(platdirs)) endif endif +endif export TEXT_OFFSET GZFLAGS MMUEXT From af9d238c6a1d5c131104bb011cc4c55a6207b9a2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 6 Jul 2016 14:16:24 +0200 Subject: [PATCH 168/235] ARM: oxnas: select reset controller framework For unknown reasons, we have to enable three symbols for a platform to use a reset controller driver, otherwise we get a Kconfig warning: warning: (MACH_OX810SE) selects RESET_OXNAS which has unmet direct dependencies (RESET_CONTROLLER) This selects the other two symbols for oxnas. Signed-off-by: Arnd Bergmann Acked-by: Neil Armstrong --- arch/arm/mach-oxnas/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mach-oxnas/Kconfig b/arch/arm/mach-oxnas/Kconfig index 567496bd250a..29100beb2e7f 100644 --- a/arch/arm/mach-oxnas/Kconfig +++ b/arch/arm/mach-oxnas/Kconfig @@ -11,11 +11,13 @@ if ARCH_OXNAS config MACH_OX810SE bool "Support OX810SE Based Products" + select ARCH_HAS_RESET_CONTROLLER select COMMON_CLK_OXNAS select CPU_ARM926T select MFD_SYSCON select OXNAS_RPS_TIMER select PINCTRL_OXNAS + select RESET_CONTROLLER select RESET_OXNAS select VERSATILE_FPGA_IRQ help From 83e484fcbe44e91fe8efb41b33ed3434806eeaac Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Wed, 10 Aug 2016 11:38:24 +0200 Subject: [PATCH 169/235] ARM: dts: add syscon compatible string for CP syscon This syscon needs to be looked up by flash protection, CLCD display output settings and other consumers. Signed-off-by: Linus Walleij Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/integratorcp.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/integratorcp.dts b/arch/arm/boot/dts/integratorcp.dts index d43f15b4f79a..79430fbfec3b 100644 --- a/arch/arm/boot/dts/integratorcp.dts +++ b/arch/arm/boot/dts/integratorcp.dts @@ -94,7 +94,7 @@ timclk: timclk@1M { }; syscon { - compatible = "arm,integrator-cp-syscon"; + compatible = "arm,integrator-cp-syscon", "syscon"; reg = <0xcb000000 0x100>; }; From f2b54191f7bef2925db39eceb84e7e19ff6d8f56 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Wed, 10 Aug 2016 11:38:12 +0200 Subject: [PATCH 170/235] ARM: dts: add syscon compatible string for AP syscon This syscon needs to be looked up by clocks, flash protection and other consumers. Signed-off-by: Linus Walleij Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/integratorap.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/integratorap.dts b/arch/arm/boot/dts/integratorap.dts index cf06e32ee108..4b34b54e09a1 100644 --- a/arch/arm/boot/dts/integratorap.dts +++ b/arch/arm/boot/dts/integratorap.dts @@ -42,7 +42,7 @@ uartclk: uartclk@14.74M { }; syscon { - compatible = "arm,integrator-ap-syscon"; + compatible = "arm,integrator-ap-syscon", "syscon"; reg = <0x11000000 0x100>; interrupt-parent = <&pic>; /* These are the logical module IRQs */ From b5c86b7496d74f6e454bcab5166efa023e1f0459 Mon Sep 17 00:00:00 2001 From: Ralf Ramsauer Date: Mon, 18 Jul 2016 11:46:48 +0200 Subject: [PATCH 171/235] ARM: tegra: fix erroneous address in dts c90bb7b enabled the high speed UARTs of the Jetson TK1. Due to a merge quirk, wrong addresses were introduced. Fix it and use the correct addresses. Thierry let me know, that there is another patch (b5896f67ab3c in linux-next) in preparation which removes all the '0,' prefixes of unit addresses on Tegra124 and is planned to go upstream in 4.8, so this patch will get reverted then. But for the moment, this patch is necessary to fix current misbehaviour. Fixes: c90bb7b9b9 ("ARM: tegra: Add high speed UARTs to Jetson TK1 device tree") Signed-off-by: Ralf Ramsauer Acked-by: Thierry Reding Cc: stable@vger.kernel.org # v4.7 Cc: linux-tegra@vger.kernel.org Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/tegra124-jetson-tk1.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/tegra124-jetson-tk1.dts b/arch/arm/boot/dts/tegra124-jetson-tk1.dts index e52b82449a79..6403e0de540e 100644 --- a/arch/arm/boot/dts/tegra124-jetson-tk1.dts +++ b/arch/arm/boot/dts/tegra124-jetson-tk1.dts @@ -1382,7 +1382,7 @@ dsi_b { * Pin 41: BR_UART1_TXD * Pin 44: BR_UART1_RXD */ - serial@70006000 { + serial@0,70006000 { compatible = "nvidia,tegra124-hsuart", "nvidia,tegra30-hsuart"; status = "okay"; }; @@ -1394,7 +1394,7 @@ serial@70006000 { * Pin 71: UART2_CTS_L * Pin 74: UART2_RTS_L */ - serial@70006040 { + serial@0,70006040 { compatible = "nvidia,tegra124-hsuart", "nvidia,tegra30-hsuart"; status = "okay"; }; From a20303725ec31ea0fcf498f1885b1d4245a4ee56 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Wed, 10 Aug 2016 14:02:17 +0200 Subject: [PATCH 172/235] ARM: dts: realview: Fix PBX-A9 cache description Clearly QEMU is very permissive in how its PL310 model may be set up, but the real hardware turns out to be far more particular about things actually being correct. Fix up the DT description so that the real thing actually boots: - The arm,data-latency and arm,tag-latency properties need 3 cells to be valid, otherwise we end up retaining the default 8-cycle latencies which leads pretty quickly to lockup. - The arm,dirty-latency property is only relevant to L210/L220, so get rid of it. - The cache geometry override also leads to lockup and/or general misbehaviour. Irritatingly, the manual doesn't state the actual PL310 configuration, but based on the boardfile code and poking registers from the Boot Monitor, it would seem to be 8 sets of 16KB ways. With that, we can successfully boot to enjoy the fun of mismatched FPUs... Cc: stable@vger.kernel.org Signed-off-by: Robin Murphy Tested-by: Mark Rutland Signed-off-by: Linus Walleij Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/arm-realview-pbx-a9.dts | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/arch/arm/boot/dts/arm-realview-pbx-a9.dts b/arch/arm/boot/dts/arm-realview-pbx-a9.dts index db808f92dd79..90d00b407f85 100644 --- a/arch/arm/boot/dts/arm-realview-pbx-a9.dts +++ b/arch/arm/boot/dts/arm-realview-pbx-a9.dts @@ -70,13 +70,12 @@ L2: l2-cache { * associativity as these may be erroneously set * up by boot loader(s). */ - cache-size = <1048576>; // 1MB - cache-sets = <4096>; + cache-size = <131072>; // 128KB + cache-sets = <512>; cache-line-size = <32>; arm,parity-disable; - arm,tag-latency = <1>; - arm,data-latency = <1 1>; - arm,dirty-latency = <1>; + arm,tag-latency = <1 1 1>; + arm,data-latency = <1 1 1>; }; scu: scu@1f000000 { From 5a3f75a4b2dab54b6b0d645b500583abfc27b260 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Wed, 3 Aug 2016 15:29:33 +0100 Subject: [PATCH 173/235] arm64: Kconfig: select ALPINE_MSI only if PCI is selected Even when PCI is disabled, ARCH_ALPINE selects ALPINE_MSI triggerring the following config warning: warning: (ARCH_ALPINE) selects ALPINE_MSI which has unmet direct dependencies (PCI) This patch makes selection of ALPINE_MSI conditional on PCI. Cc: Arnd Bergmann Acked-by: Antoine Tenart Signed-off-by: Sudeep Holla Signed-off-by: Arnd Bergmann --- arch/arm64/Kconfig.platforms | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/Kconfig.platforms b/arch/arm64/Kconfig.platforms index bb2616b16157..fd3ee3a392be 100644 --- a/arch/arm64/Kconfig.platforms +++ b/arch/arm64/Kconfig.platforms @@ -8,7 +8,7 @@ config ARCH_SUNXI config ARCH_ALPINE bool "Annapurna Labs Alpine platform" - select ALPINE_MSI + select ALPINE_MSI if PCI help This enables support for the Annapurna Labs Alpine Soc family. From f9db43bc296833451cbe4b03eb5e1f633ad1f787 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Wed, 3 Aug 2016 15:29:34 +0100 Subject: [PATCH 174/235] arm64: Kconfig: select HISILICON_IRQ_MBIGEN only if PCI is selected Even when PCI is disabled, ARCH_HISI selects HISILICON_IRQ_MBIGEN triggerring the following config warning: warning: (ARM64 && HISILICON_IRQ_MBIGEN) selects ARM_GIC_V3_ITS which has unmet direct dependencies (PCI && PCI_MSI) This patch makes selection of HISILICON_IRQ_MBIGEN conditional on PCI. Cc: Ma Jun Cc: Arnd Bergmann Signed-off-by: Sudeep Holla Signed-off-by: Arnd Bergmann --- arch/arm64/Kconfig.platforms | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/Kconfig.platforms b/arch/arm64/Kconfig.platforms index fd3ee3a392be..be5d824ebdba 100644 --- a/arch/arm64/Kconfig.platforms +++ b/arch/arm64/Kconfig.platforms @@ -66,7 +66,7 @@ config ARCH_LG1K config ARCH_HISI bool "Hisilicon SoC Family" select ARM_TIMER_SP804 - select HISILICON_IRQ_MBIGEN + select HISILICON_IRQ_MBIGEN if PCI help This enables support for Hisilicon ARMv8 SoC family From d8d378fa1a0c98ecb50ca52c9bf3bc14e25aa2d2 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Wed, 10 Aug 2016 15:59:09 -0700 Subject: [PATCH 175/235] tools/testing/nvdimm: fix SIGTERM vs hotplug crash The unit tests crash when hotplug races the previous probe. This race requires that the loading of the nfit_test module be terminated with SIGTERM, and the module to be unloaded while the ars scan is still running. In contrast to the normal nfit driver, the unit test calls acpi_nfit_init() twice to simulate hotplug, whereas the nominal case goes through the acpi_nfit_notify() event handler. The acpi_nfit_notify() path is careful to flush the previous region registration before servicing the hotplug event. The unit test was missing this guarantee. BUG: unable to handle kernel NULL pointer dereference at (null) IP: [] pwq_activate_delayed_work+0x47/0x170 [..] Call Trace: [] pwq_dec_nr_in_flight+0x66/0xa0 [] process_one_work+0x2d0/0x680 [] ? process_one_work+0x171/0x680 [] worker_thread+0x4e/0x480 [] ? process_one_work+0x680/0x680 [] ? process_one_work+0x680/0x680 [] kthread+0xf3/0x110 [] ret_from_fork+0x1f/0x40 [] ? kthread_create_on_node+0x230/0x230 Cc: Signed-off-by: Dan Williams --- tools/testing/nvdimm/test/nfit.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c index 5404efa578a3..dd48f421844c 100644 --- a/tools/testing/nvdimm/test/nfit.c +++ b/tools/testing/nvdimm/test/nfit.c @@ -13,6 +13,7 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include #include +#include #include #include #include @@ -1474,6 +1475,7 @@ static int nfit_test_probe(struct platform_device *pdev) if (nfit_test->setup != nfit_test0_setup) return 0; + flush_work(&acpi_desc->work); nfit_test->setup_hotplug = 1; nfit_test->setup(nfit_test); From 73984137d32bb6e48646daea60ebe1457a01a061 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 10 Aug 2016 16:27:38 -0700 Subject: [PATCH 176/235] rapidio: dereferencing an error pointer Original patch: https://lkml.org/lkml/2016/8/4/32 If riocm_ch_alloc() fails then we end up dereferencing the error pointer. The problem is that we're not unwinding in the reverse order from how we allocate things so it gets confusing. I've changed this around so now "ch" is NULL when we are done with it after we call riocm_put_channel(). That way we can check if it's NULL and avoid calling riocm_put_channel() on it twice. I renamed err_nodev to err_put_new_ch so that it better reflects what the goto does. Then because we had flipping things around, it means we don't neeed to initialize the pointers to NULL and we can remove an if statement and pull things in an indent level. Link: http://lkml.kernel.org/r/20160805152406.20713-1-alexandre.bounine@idt.com Signed-off-by: Dan Carpenter Signed-off-by: Alexandre Bounine Cc: Matt Porter Cc: Andre van Herk Cc: Barry Wood Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rapidio/rio_cm.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/drivers/rapidio/rio_cm.c b/drivers/rapidio/rio_cm.c index cecc15a880de..3fa17ac8df54 100644 --- a/drivers/rapidio/rio_cm.c +++ b/drivers/rapidio/rio_cm.c @@ -1080,8 +1080,8 @@ static int riocm_send_ack(struct rio_channel *ch) static struct rio_channel *riocm_ch_accept(u16 ch_id, u16 *new_ch_id, long timeout) { - struct rio_channel *ch = NULL; - struct rio_channel *new_ch = NULL; + struct rio_channel *ch; + struct rio_channel *new_ch; struct conn_req *req; struct cm_peer *peer; int found = 0; @@ -1155,6 +1155,7 @@ static struct rio_channel *riocm_ch_accept(u16 ch_id, u16 *new_ch_id, spin_unlock_bh(&ch->lock); riocm_put_channel(ch); + ch = NULL; kfree(req); down_read(&rdev_sem); @@ -1172,7 +1173,7 @@ static struct rio_channel *riocm_ch_accept(u16 ch_id, u16 *new_ch_id, if (!found) { /* If peer device object not found, simply ignore the request */ err = -ENODEV; - goto err_nodev; + goto err_put_new_ch; } new_ch->rdev = peer->rdev; @@ -1184,15 +1185,16 @@ static struct rio_channel *riocm_ch_accept(u16 ch_id, u16 *new_ch_id, *new_ch_id = new_ch->id; return new_ch; + +err_put_new_ch: + spin_lock_bh(&idr_lock); + idr_remove(&ch_idr, new_ch->id); + spin_unlock_bh(&idr_lock); + riocm_put_channel(new_ch); + err_put: - riocm_put_channel(ch); -err_nodev: - if (new_ch) { - spin_lock_bh(&idr_lock); - idr_remove(&ch_idr, new_ch->id); - spin_unlock_bh(&idr_lock); - riocm_put_channel(new_ch); - } + if (ch) + riocm_put_channel(ch); *new_ch_id = 0; return ERR_PTR(err); } From a545de5ce2ef3abc4db0b9331840acf59c8f9efa Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 10 Aug 2016 16:27:41 -0700 Subject: [PATCH 177/235] revert "ARM: keystone: dts: add psci command definition" Revert commit 51d5d12b8f3d ("ARM: keystone: dts: add psci command definition"), which was inadvertently added twice. Cc: Russell King - ARM Linux Cc: Vitaly Andrianov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/boot/dts/keystone.dtsi | 8 -------- 1 file changed, 8 deletions(-) diff --git a/arch/arm/boot/dts/keystone.dtsi b/arch/arm/boot/dts/keystone.dtsi index 00cb314d5e4d..e23f46d15c80 100644 --- a/arch/arm/boot/dts/keystone.dtsi +++ b/arch/arm/boot/dts/keystone.dtsi @@ -70,14 +70,6 @@ psci { cpu_on = <0x84000003>; }; - psci { - compatible = "arm,psci"; - method = "smc"; - cpu_suspend = <0x84000001>; - cpu_off = <0x84000002>; - cpu_on = <0x84000003>; - }; - soc { #address-cells = <1>; #size-cells = <1>; From 3b33719c9b741066f7d2bc6036409752f8e0478d Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 10 Aug 2016 16:27:44 -0700 Subject: [PATCH 178/235] thp: move shmem_huge_enabled() outside of SYSFS ifdef The newly introduced shmem_huge_enabled() function has two definitions, but neither of them is visible if CONFIG_SYSFS is disabled, leading to a build error: mm/khugepaged.o: In function `khugepaged': khugepaged.c:(.text.khugepaged+0x3ca): undefined reference to `shmem_huge_enabled' This changes the #ifdef guards around the definition to match those that are used in the header file. Fixes: e496cf3d7821 ("thp: introduce CONFIG_TRANSPARENT_HUGE_PAGECACHE") Link: http://lkml.kernel.org/r/20160809123638.1357593-1-arnd@arndb.de Signed-off-by: Arnd Bergmann Acked-by: Kirill A. Shutemov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/shmem.c b/mm/shmem.c index 7f7748a0f9e1..fd8b2b5741b1 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3975,7 +3975,9 @@ static ssize_t shmem_enabled_store(struct kobject *kobj, struct kobj_attribute shmem_enabled_attr = __ATTR(shmem_enabled, 0644, shmem_enabled_show, shmem_enabled_store); +#endif /* CONFIG_TRANSPARENT_HUGE_PAGECACHE && CONFIG_SYSFS */ +#ifdef CONFIG_TRANSPARENT_HUGE_PAGECACHE bool shmem_huge_enabled(struct vm_area_struct *vma) { struct inode *inode = file_inode(vma->vm_file); @@ -4006,7 +4008,7 @@ bool shmem_huge_enabled(struct vm_area_struct *vma) return false; } } -#endif /* CONFIG_TRANSPARENT_HUGE_PAGECACHE && CONFIG_SYSFS */ +#endif /* CONFIG_TRANSPARENT_HUGE_PAGECACHE */ #else /* !CONFIG_SHMEM */ From 81cbcbc2d810c0ce49fba81f864302e1afe5ff27 Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Wed, 10 Aug 2016 16:27:46 -0700 Subject: [PATCH 179/235] mm/page_alloc.c: fix wrong initialization when sysctl_min_unmapped_ratio changes Before resetting min_unmapped_pages, we need to initialize min_unmapped_pages rather than min_slab_pages. Fixes: a5f5f91da6 (mm: convert zone_reclaim to node_reclaim) Link: http://lkml.kernel.org/r/1470724248-26780-1-git-send-email-iamjoonsoo.kim@lge.com Signed-off-by: Joonsoo Kim Acked-by: Mel Gorman Cc: Vlastimil Babka Cc: Johannes Weiner Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ee744fa3b93d..9a92718b1103 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -6854,7 +6854,7 @@ int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *table, int write, return rc; for_each_online_pgdat(pgdat) - pgdat->min_slab_pages = 0; + pgdat->min_unmapped_pages = 0; for_each_zone(zone) zone->zone_pgdat->min_unmapped_pages += (zone->managed_pages * From 6423aa8192c596848e1b23bd4193dc0924e7274d Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Wed, 10 Aug 2016 16:27:49 -0700 Subject: [PATCH 180/235] mm/page_alloc.c: recalculate some of node threshold when on/offline memory Some of node threshold depends on number of managed pages in the node. When memory is going on/offline, it can be changed and we need to adjust them. Add recalculation to appropriate places and clean-up related functions for better maintenance. Link: http://lkml.kernel.org/r/1470724248-26780-2-git-send-email-iamjoonsoo.kim@lge.com Signed-off-by: Joonsoo Kim Acked-by: Mel Gorman Cc: Vlastimil Babka Cc: Johannes Weiner Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 50 ++++++++++++++++++++++++++++++++++--------------- 1 file changed, 35 insertions(+), 15 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 9a92718b1103..ab2c0ff8c2e6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4757,6 +4757,8 @@ int local_memory_node(int node) } #endif +static void setup_min_unmapped_ratio(void); +static void setup_min_slab_ratio(void); #else /* CONFIG_NUMA */ static void set_zonelist_order(void) @@ -5878,9 +5880,6 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat) zone->managed_pages = is_highmem_idx(j) ? realsize : freesize; #ifdef CONFIG_NUMA zone->node = nid; - pgdat->min_unmapped_pages += (freesize*sysctl_min_unmapped_ratio) - / 100; - pgdat->min_slab_pages += (freesize * sysctl_min_slab_ratio) / 100; #endif zone->name = zone_names[j]; zone->zone_pgdat = pgdat; @@ -6801,6 +6800,12 @@ int __meminit init_per_zone_wmark_min(void) setup_per_zone_wmarks(); refresh_zone_stat_thresholds(); setup_per_zone_lowmem_reserve(); + +#ifdef CONFIG_NUMA + setup_min_unmapped_ratio(); + setup_min_slab_ratio(); +#endif + return 0; } core_initcall(init_per_zone_wmark_min) @@ -6842,16 +6847,10 @@ int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write, } #ifdef CONFIG_NUMA -int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *length, loff_t *ppos) +static void setup_min_unmapped_ratio(void) { - struct pglist_data *pgdat; + pg_data_t *pgdat; struct zone *zone; - int rc; - - rc = proc_dointvec_minmax(table, write, buffer, length, ppos); - if (rc) - return rc; for_each_online_pgdat(pgdat) pgdat->min_unmapped_pages = 0; @@ -6859,26 +6858,47 @@ int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *table, int write, for_each_zone(zone) zone->zone_pgdat->min_unmapped_pages += (zone->managed_pages * sysctl_min_unmapped_ratio) / 100; - return 0; } -int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write, + +int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *table, int write, void __user *buffer, size_t *length, loff_t *ppos) { - struct pglist_data *pgdat; - struct zone *zone; int rc; rc = proc_dointvec_minmax(table, write, buffer, length, ppos); if (rc) return rc; + setup_min_unmapped_ratio(); + + return 0; +} + +static void setup_min_slab_ratio(void) +{ + pg_data_t *pgdat; + struct zone *zone; + for_each_online_pgdat(pgdat) pgdat->min_slab_pages = 0; for_each_zone(zone) zone->zone_pgdat->min_slab_pages += (zone->managed_pages * sysctl_min_slab_ratio) / 100; +} + +int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) +{ + int rc; + + rc = proc_dointvec_minmax(table, write, buffer, length, ppos); + if (rc) + return rc; + + setup_min_slab_ratio(); + return 0; } #endif From c8efc390c1e0eca195ae59a2f7cec46773620e0c Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 10 Aug 2016 16:27:52 -0700 Subject: [PATCH 181/235] mm, rmap: fix false positive VM_BUG() in page_add_file_rmap() PageTransCompound() doesn't distinguish THP from from any other type of compound pages. This can lead to false-positive VM_BUG_ON() in page_add_file_rmap() if called on compound page from a driver[1]. I think we can exclude such cases by checking if the page belong to a mapping. The VM_BUG_ON_PAGE() is downgraded to VM_WARN_ON_ONCE(). This path should not cause any harm to non-THP page, but good to know if we step on anything else. [1] http://lkml.kernel.org/r/c711e067-0bff-a6cb-3c37-04dfe77d2db1@redhat.com Link: http://lkml.kernel.org/r/20160810161345.GA67522@black.fi.intel.com Signed-off-by: Kirill A. Shutemov Reported-by: Laura Abbott Tested-by: Laura Abbott Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/rmap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/mm/rmap.c b/mm/rmap.c index 709bc83703b1..d4f56060ba3f 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1284,8 +1284,9 @@ void page_add_file_rmap(struct page *page, bool compound) VM_BUG_ON_PAGE(!PageSwapBacked(page), page); __inc_node_page_state(page, NR_SHMEM_PMDMAPPED); } else { - if (PageTransCompound(page)) { - VM_BUG_ON_PAGE(!PageLocked(page), page); + if (PageTransCompound(page) && page_mapping(page)) { + VM_WARN_ON_ONCE(!PageLocked(page)); + SetPageDoubleMap(compound_head(page)); if (PageMlocked(page)) clear_page_mlock(compound_head(page)); From 57dea93ac42d341e1fe902528b348ef6763fa485 Mon Sep 17 00:00:00 2001 From: Steve Capper Date: Wed, 10 Aug 2016 16:27:55 -0700 Subject: [PATCH 182/235] rmap: fix compound check logic in page_remove_file_rmap In page_remove_file_rmap(.) we have the following check: VM_BUG_ON_PAGE(compound && !PageTransHuge(page), page); This is meant to check for either HugeTLB pages or THP when a compound page is passed in. Unfortunately, if one disables CONFIG_TRANSPARENT_HUGEPAGE, then PageTransHuge(.) will always return false, provoking BUGs when one runs the libhugetlbfs test suite. This patch replaces PageTransHuge(), with PageHead() which will work for both HugeTLB and THP. Fixes: dd78fedde4b9 ("rmap: support file thp") Link: http://lkml.kernel.org/r/1470838217-5889-1-git-send-email-steve.capper@arm.com Signed-off-by: Steve Capper Acked-by: Kirill A. Shutemov Cc: Huang Shijie Cc: Will Deacon Cc: Catalin Marinas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/rmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/rmap.c b/mm/rmap.c index d4f56060ba3f..1ef36404e7b2 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1304,7 +1304,7 @@ static void page_remove_file_rmap(struct page *page, bool compound) { int i, nr = 1; - VM_BUG_ON_PAGE(compound && !PageTransHuge(page), page); + VM_BUG_ON_PAGE(compound && !PageHead(page), page); lock_page_memcg(page); /* Hugepages are not counted in NR_FILE_MAPPED for now. */ From 6039892396d845b18228935561960441900cffca Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 10 Aug 2016 16:27:58 -0700 Subject: [PATCH 183/235] mm/slub.c: run free_partial() outside of the kmem_cache_node->list_lock With debugobjects enabled and using SLAB_DESTROY_BY_RCU, when a kmem_cache_node is destroyed the call_rcu() may trigger a slab allocation to fill the debug object pool (__debug_object_init:fill_pool). Everywhere but during kmem_cache_destroy(), discard_slab() is performed outside of the kmem_cache_node->list_lock and avoids a lockdep warning about potential recursion: ============================================= [ INFO: possible recursive locking detected ] 4.8.0-rc1-gfxbench+ #1 Tainted: G U --------------------------------------------- rmmod/8895 is trying to acquire lock: (&(&n->list_lock)->rlock){-.-...}, at: [] get_partial_node.isra.63+0x47/0x430 but task is already holding lock: (&(&n->list_lock)->rlock){-.-...}, at: [] __kmem_cache_shutdown+0x54/0x320 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&(&n->list_lock)->rlock); lock(&(&n->list_lock)->rlock); *** DEADLOCK *** May be due to missing lock nesting notation 5 locks held by rmmod/8895: #0: (&dev->mutex){......}, at: driver_detach+0x42/0xc0 #1: (&dev->mutex){......}, at: driver_detach+0x50/0xc0 #2: (cpu_hotplug.dep_map){++++++}, at: get_online_cpus+0x2d/0x80 #3: (slab_mutex){+.+.+.}, at: kmem_cache_destroy+0x3c/0x220 #4: (&(&n->list_lock)->rlock){-.-...}, at: __kmem_cache_shutdown+0x54/0x320 stack backtrace: CPU: 6 PID: 8895 Comm: rmmod Tainted: G U 4.8.0-rc1-gfxbench+ #1 Hardware name: Gigabyte Technology Co., Ltd. H87M-D3H/H87M-D3H, BIOS F11 08/18/2015 Call Trace: __lock_acquire+0x1646/0x1ad0 lock_acquire+0xb2/0x200 _raw_spin_lock+0x36/0x50 get_partial_node.isra.63+0x47/0x430 ___slab_alloc.constprop.67+0x1a7/0x3b0 __slab_alloc.isra.64.constprop.66+0x43/0x80 kmem_cache_alloc+0x236/0x2d0 __debug_object_init+0x2de/0x400 debug_object_activate+0x109/0x1e0 __call_rcu.constprop.63+0x32/0x2f0 call_rcu+0x12/0x20 discard_slab+0x3d/0x40 __kmem_cache_shutdown+0xdb/0x320 shutdown_cache+0x19/0x60 kmem_cache_destroy+0x1ae/0x220 i915_gem_load_cleanup+0x14/0x40 [i915] i915_driver_unload+0x151/0x180 [i915] i915_pci_remove+0x14/0x20 [i915] pci_device_remove+0x34/0xb0 __device_release_driver+0x95/0x140 driver_detach+0xb6/0xc0 bus_remove_driver+0x53/0xd0 driver_unregister+0x27/0x50 pci_unregister_driver+0x25/0x70 i915_exit+0x1a/0x1e2 [i915] SyS_delete_module+0x193/0x1f0 entry_SYSCALL_64_fastpath+0x1c/0xac Fixes: 52b4b950b507 ("mm: slab: free kmem_cache_node after destroy sysfs file") Link: http://lkml.kernel.org/r/1470759070-18743-1-git-send-email-chris@chris-wilson.co.uk Reported-by: Dave Gordon Signed-off-by: Chris Wilson Reviewed-by: Vladimir Davydov Acked-by: Christoph Lameter Cc: Pekka Enberg Cc: David Rientjes Cc: Joonsoo Kim Cc: Dmitry Safonov Cc: Daniel Vetter Cc: Dave Gordon Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slub.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/mm/slub.c b/mm/slub.c index cead06394e9e..9adae58462f8 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -3629,6 +3629,7 @@ static void list_slab_objects(struct kmem_cache *s, struct page *page, */ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n) { + LIST_HEAD(discard); struct page *page, *h; BUG_ON(irqs_disabled()); @@ -3636,13 +3637,16 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n) list_for_each_entry_safe(page, h, &n->partial, lru) { if (!page->inuse) { remove_partial(n, page); - discard_slab(s, page); + list_add(&page->lru, &discard); } else { list_slab_objects(s, page, "Objects remaining in %s on __kmem_cache_shutdown()"); } } spin_unlock_irq(&n->list_lock); + + list_for_each_entry_safe(page, h, &discard, lru) + discard_slab(s, page); } /* From 12beb346710b766b8e74a7a3ac8165835bd68def Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 10 Aug 2016 22:47:59 +0200 Subject: [PATCH 184/235] Merge tag 'pxa-fixes-v4.8' of https://github.com/rjarzmik/linux into randconfig-4.8 This is the pxa changes for v4.8 cycle. This is a tiny fix couple to enable changes in includes in gpio API without breaking pxa boards. * tag 'pxa-fixes-v4.8' of https://github.com/rjarzmik/linux: ARM: pxa: add module.h for corgi symbol_get/symbol_put usage ARM: pxa: add module.h for spitz symbol_get/symbol_put usage --- arch/arm/mach-pxa/corgi.c | 1 + arch/arm/mach-pxa/spitz.c | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c index dc109dc3a622..10bfdb169366 100644 --- a/arch/arm/mach-pxa/corgi.c +++ b/arch/arm/mach-pxa/corgi.c @@ -13,6 +13,7 @@ */ #include +#include /* symbol_get ; symbol_put */ #include #include #include diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c index 1080580b1343..2c150bfc0cd5 100644 --- a/arch/arm/mach-pxa/spitz.c +++ b/arch/arm/mach-pxa/spitz.c @@ -13,6 +13,7 @@ */ #include +#include /* symbol_get ; symbol_put */ #include #include #include From 62d16b5a3fca4d186e13215e0d7d2f6d36191796 Mon Sep 17 00:00:00 2001 From: Nicolas Iooss Date: Sat, 6 Aug 2016 12:20:39 +0200 Subject: [PATCH 185/235] x86/mm/kaslr: Fix -Wformat-security warning debug_putstr() is used to output strings without using printf-like formatting but debug_putstr(v) is defined as early_printk(v) in arch/x86/lib/kaslr.c. This makes clang reports the following warning when building with -Wformat-security: arch/x86/lib/kaslr.c:57:15: warning: format string is not a string literal (potentially insecure) [-Wformat-security] debug_putstr(purpose); ^~~~~~~ Fix this by using "%s" in early_printk(). Signed-off-by: Nicolas Iooss Acked-by: Kees Cook Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20160806102039.27221-1-nicolas.iooss_linux@m4x.org Signed-off-by: Ingo Molnar --- arch/x86/lib/kaslr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/lib/kaslr.c b/arch/x86/lib/kaslr.c index f7dfeda83e5c..121f59c6ee54 100644 --- a/arch/x86/lib/kaslr.c +++ b/arch/x86/lib/kaslr.c @@ -19,7 +19,7 @@ #include #include -#define debug_putstr(v) early_printk(v) +#define debug_putstr(v) early_printk("%s", v) #define has_cpuflag(f) boot_cpu_has(f) #define get_boot_seed() kaslr_offset() #endif From f9bcf1e0e0145323ba2cf72ecad5264ff3883eb1 Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Thu, 11 Aug 2016 13:36:35 +0800 Subject: [PATCH 186/235] sched/cputime: Fix steal time accounting Commit: 57430218317 ("sched/cputime: Count actually elapsed irq & softirq time") ... didn't take steal time into consideration with passing the noirqtime kernel parameter. As Paolo pointed out before: | Why not? If idle=poll, for example, any time the guest is suspended (and | thus cannot poll) does count as stolen time. This patch fixes it by reducing steal time from idle time accounting when the noirqtime parameter is true. The average idle time drops from 56.8% to 54.75% for nohz idle kvm guest(noirqtime, idle=poll, four vCPUs running on one pCPU). Signed-off-by: Wanpeng Li Cc: Frederic Weisbecker Cc: Linus Torvalds Cc: Paolo Bonzini Cc: Peter Zijlstra (Intel) Cc: Peter Zijlstra Cc: Radim Cc: Rik van Riel Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1470893795-3527-1-git-send-email-wanpeng.li@hotmail.com Signed-off-by: Ingo Molnar --- kernel/sched/cputime.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 1934f658c036..8b9bcc5a58fa 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -508,13 +508,20 @@ void account_process_tick(struct task_struct *p, int user_tick) */ void account_idle_ticks(unsigned long ticks) { - + cputime_t cputime, steal; if (sched_clock_irqtime) { irqtime_account_idle_ticks(ticks); return; } - account_idle_time(jiffies_to_cputime(ticks)); + cputime = cputime_one_jiffy; + steal = steal_account_process_time(cputime); + + if (steal >= cputime) + return; + + cputime -= steal; + account_idle_time(cputime); } /* From ace7fab7a6cdd363a615ec537f2aa94dbc761ee2 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Wed, 10 Aug 2016 10:23:25 -0700 Subject: [PATCH 187/235] x86/mm: Fix swap entry comment and macro A recent patch changed the format of a swap PTE. The comment explaining the format of the swap PTE is wrong about the bits used for the swap type field. Amusingly, the ASCII art and the patch description are correct, but the comment itself is wrong. As I was looking at this, I also noticed that the SWP_OFFSET_FIRST_BIT has an off-by-one error. This does not really hurt anything. It just wasted a bit of space in the PTE, giving us 2^59 bytes of addressable space in our swapfiles instead of 2^60. But, it doesn't match with the comments, and it wastes a bit of space, so fix it. Signed-off-by: Dave Hansen Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Dave Hansen Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Luis R. Rodriguez Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Toshi Kani Fixes: 00839ee3b299 ("x86/mm: Move swap offset/type up in PTE to work around erratum") Link: http://lkml.kernel.org/r/20160810172325.E56AD7DA@viggo.jf.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable_64.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index 7e8ec7ae10fa..1cc82ece9ac1 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -145,7 +145,7 @@ static inline int pgd_large(pgd_t pgd) { return 0; } * * | ... | 11| 10| 9|8|7|6|5| 4| 3|2|1|0| <- bit number * | ... |SW3|SW2|SW1|G|L|D|A|CD|WT|U|W|P| <- bit names - * | OFFSET (14->63) | TYPE (10-13) |0|X|X|X| X| X|X|X|0| <- swp entry + * | OFFSET (14->63) | TYPE (9-13) |0|X|X|X| X| X|X|X|0| <- swp entry * * G (8) is aliased and used as a PROT_NONE indicator for * !present ptes. We need to start storing swap entries above @@ -156,7 +156,7 @@ static inline int pgd_large(pgd_t pgd) { return 0; } #define SWP_TYPE_FIRST_BIT (_PAGE_BIT_PROTNONE + 1) #define SWP_TYPE_BITS 5 /* Place the offset above the type: */ -#define SWP_OFFSET_FIRST_BIT (SWP_TYPE_FIRST_BIT + SWP_TYPE_BITS + 1) +#define SWP_OFFSET_FIRST_BIT (SWP_TYPE_FIRST_BIT + SWP_TYPE_BITS) #define MAX_SWAPFILES_CHECK() BUILD_BUG_ON(MAX_SWAPFILES_SHIFT > SWP_TYPE_BITS) From 82ba4faca1bffad429f15c90c980ffd010366c25 Mon Sep 17 00:00:00 2001 From: Aaron Lu Date: Thu, 11 Aug 2016 15:44:30 +0800 Subject: [PATCH 188/235] x86/irq: Do not substract irq_tlb_count from irq_call_count Since commit: 52aec3308db8 ("x86/tlb: replace INVALIDATE_TLB_VECTOR by CALL_FUNCTION_VECTOR") the TLB remote shootdown is done through call function vector. That commit didn't take care of irq_tlb_count, which a later commit: fd0f5869724f ("x86: Distinguish TLB shootdown interrupts from other functions call interrupts") ... tried to fix. The fix assumes every increase of irq_tlb_count has a corresponding increase of irq_call_count. So the irq_call_count is always bigger than irq_tlb_count and we could substract irq_tlb_count from irq_call_count. Unfortunately this is not true for the smp_call_function_single() case. The IPI is only sent if the target CPU's call_single_queue is empty when adding a csd into it in generic_exec_single. That means if two threads are both adding flush tlb csds to the same CPU's call_single_queue, only one IPI is sent. In other words, the irq_call_count is incremented by 1 but irq_tlb_count is incremented by 2. Over time, irq_tlb_count will be bigger than irq_call_count and the substract will produce a very large irq_call_count value due to overflow. Considering that: 1) it's not worth to send more IPIs for the sake of accurate counting of irq_call_count in generic_exec_single(); 2) it's not easy to tell if the call function interrupt is for TLB shootdown in __smp_call_function_single_interrupt(). Not to exclude TLB shootdown from call function count seems to be the simplest fix and this patch just does that. This bug was found by LKP's cyclic performance regression tracking recently with the vm-scalability test suite. I have bisected to commit: 3dec0ba0be6a ("mm/rmap: share the i_mmap_rwsem") This commit didn't do anything wrong but revealed the irq_call_count problem. IIUC, the commit makes rwc->remap_one in rmap_walk_file concurrent with multiple threads. When remap_one is try_to_unmap_one(), then multiple threads could queue flush TLB to the same CPU but only one IPI will be sent. Since the commit was added in Linux v3.19, the counting problem only shows up from v3.19 onwards. Signed-off-by: Aaron Lu Cc: Alex Shi Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Davidlohr Bueso Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Huang Ying Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Tomoki Sekiyama Link: http://lkml.kernel.org/r/20160811074430.GA18163@aaronlu.sh.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/hardirq.h | 4 ---- arch/x86/kernel/irq.c | 3 +-- 2 files changed, 1 insertion(+), 6 deletions(-) diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index 7178043b0e1d..59405a248fc2 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -22,10 +22,6 @@ typedef struct { #ifdef CONFIG_SMP unsigned int irq_resched_count; unsigned int irq_call_count; - /* - * irq_tlb_count is double-counted in irq_call_count, so it must be - * subtracted from irq_call_count when displaying irq_call_count - */ unsigned int irq_tlb_count; #endif #ifdef CONFIG_X86_THERMAL_VECTOR diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 61521dc19c10..9f669fdd2010 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -102,8 +102,7 @@ int arch_show_interrupts(struct seq_file *p, int prec) seq_puts(p, " Rescheduling interrupts\n"); seq_printf(p, "%*s: ", prec, "CAL"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->irq_call_count - - irq_stats(j)->irq_tlb_count); + seq_printf(p, "%10u ", irq_stats(j)->irq_call_count); seq_puts(p, " Function call interrupts\n"); seq_printf(p, "%*s: ", prec, "TLB"); for_each_online_cpu(j) From 007b756053386af079ba963a8f5817ac651c7c59 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 10 Aug 2016 02:29:13 -0700 Subject: [PATCH 189/235] x86/boot: Run reserve_bios_regions() after we initialize the memory map reserve_bios_regions() is a quirk that reserves memory that we might otherwise think is available. There's no need to run it so early, and running it before we have the memory map initialized with its non-quirky inputs makes it hard to make reserve_bios_regions() more intelligent. Move it right after we populate the memblock state. Signed-off-by: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mario Limonciello Cc: Matt Fleming Cc: Matthew Garrett Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/59f58618911005c799c6c9979ce6ae4881d907c2.1470821230.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/head32.c | 2 -- arch/x86/kernel/head64.c | 1 - arch/x86/kernel/setup.c | 2 ++ 3 files changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 2dda0bc4576e..f16c55bfc090 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -25,8 +25,6 @@ static void __init i386_default_early_setup(void) /* Initialize 32bit specific setup functions */ x86_init.resources.reserve_resources = i386_reserve_resources; x86_init.mpparse.setup_ioapic_ids = setup_ioapic_ids_from_mpc; - - reserve_bios_regions(); } asmlinkage __visible void __init i386_start_kernel(void) diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 99d48e7d2974..54a2372f5dbb 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -183,7 +183,6 @@ void __init x86_64_start_reservations(char *real_mode_data) copy_bootdata(__va(real_mode_data)); x86_early_init_platform_quirks(); - reserve_bios_regions(); switch (boot_params.hdr.hardware_subarch) { case X86_SUBARCH_INTEL_MID: diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 95cf31c9f4ec..bf780e0f76ef 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1101,6 +1101,8 @@ void __init setup_arch(char **cmdline_p) efi_find_mirror(); } + reserve_bios_regions(); + /* * The EFI specification says that boot service code won't be called * after ExitBootServices(). This is, in fact, a lie. From 18bc7bd523e0fc5be8d76bf84bde733a97a8c375 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 10 Aug 2016 02:29:14 -0700 Subject: [PATCH 190/235] x86/boot: Synchronize trampoline_cr4_features and mmu_cr4_features directly The initialization process for trampoline_cr4_features and mmu_cr4_features was confusing. The intent is for mmu_cr4_features and *trampoline_cr4_features to stay in sync, but trampoline_cr4_features is NULL until setup_real_mode() runs. The old code synchronized *trampoline_cr4_features *twice*, once in setup_real_mode() and once in setup_arch(). It also initialized mmu_cr4_features in setup_real_mode(), which causes the actual value of mmu_cr4_features to potentially depend on when setup_real_mode() is called. With this patch, mmu_cr4_features is initialized directly in setup_arch(), and *trampoline_cr4_features is synchronized to mmu_cr4_features when the trampoline is set up. After this patch, it should be safe to defer setup_real_mode(). Signed-off-by: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mario Limonciello Cc: Matt Fleming Cc: Matthew Garrett Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/d48a263f9912389b957dd495a7127b009259ffe0.1470821230.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 17 ++++++++++------- arch/x86/realmode/init.c | 3 ++- 2 files changed, 12 insertions(+), 8 deletions(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index bf780e0f76ef..95c8c9c7dd6d 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1131,6 +1131,16 @@ void __init setup_arch(char **cmdline_p) early_trap_pf_init(); + /* + * Update mmu_cr4_features (and, indirectly, trampoline_cr4_features) + * with the current CR4 value. This may not be necessary, but + * auditing all the early-boot CR4 manipulation would be needed to + * rule it out. + */ + if (boot_cpu_data.cpuid_level >= 0) + /* A CPU has %cr4 if and only if it has CPUID. */ + mmu_cr4_features = __read_cr4(); + setup_real_mode(); memblock_set_current_limit(get_max_mapped()); @@ -1180,13 +1190,6 @@ void __init setup_arch(char **cmdline_p) kasan_init(); - if (boot_cpu_data.cpuid_level >= 0) { - /* A CPU has %cr4 if and only if it has CPUID */ - mmu_cr4_features = __read_cr4(); - if (trampoline_cr4_features) - *trampoline_cr4_features = mmu_cr4_features; - } - #ifdef CONFIG_X86_32 /* sync back kernel address range */ clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY, diff --git a/arch/x86/realmode/init.c b/arch/x86/realmode/init.c index 705e3fffb4a1..c5bdc4e473e7 100644 --- a/arch/x86/realmode/init.c +++ b/arch/x86/realmode/init.c @@ -4,6 +4,7 @@ #include #include #include +#include struct real_mode_header *real_mode_header; u32 *trampoline_cr4_features; @@ -84,7 +85,7 @@ void __init setup_real_mode(void) trampoline_header->start = (u64) secondary_startup_64; trampoline_cr4_features = &trampoline_header->cr4; - *trampoline_cr4_features = __read_cr4(); + *trampoline_cr4_features = mmu_cr4_features; trampoline_pgd = (u64 *) __va(real_mode_header->trampoline_pgd); trampoline_pgd[0] = trampoline_pgd_entry.pgd; From d0de0f685db7faf2ae4597a39a59996dd84e18c7 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 10 Aug 2016 02:29:15 -0700 Subject: [PATCH 191/235] x86/boot: Defer setup_real_mode() to early_initcall time There's no need to run setup_real_mode() as early as we run it. Defer it to the same early_initcall that sets up the page permissions for the real mode code. This should be a code size reduction. More importantly, it give us a longer window in which we can allocate the real mode trampoline. Signed-off-by: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mario Limonciello Cc: Matt Fleming Cc: Matthew Garrett Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/fd62f0da4f79357695e9bf3e365623736b05f119.1470821230.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/include/asm/realmode.h | 1 - arch/x86/kernel/setup.c | 2 -- arch/x86/realmode/init.c | 15 ++++++++++++--- 3 files changed, 12 insertions(+), 6 deletions(-) diff --git a/arch/x86/include/asm/realmode.h b/arch/x86/include/asm/realmode.h index 9c6b890d5e7a..8d6777724ba4 100644 --- a/arch/x86/include/asm/realmode.h +++ b/arch/x86/include/asm/realmode.h @@ -59,6 +59,5 @@ extern unsigned char secondary_startup_64[]; #endif void reserve_real_mode(void); -void setup_real_mode(void); #endif /* _ARCH_X86_REALMODE_H */ diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 95c8c9c7dd6d..0fa60f5f5a16 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1141,8 +1141,6 @@ void __init setup_arch(char **cmdline_p) /* A CPU has %cr4 if and only if it has CPUID. */ mmu_cr4_features = __read_cr4(); - setup_real_mode(); - memblock_set_current_limit(get_max_mapped()); /* diff --git a/arch/x86/realmode/init.c b/arch/x86/realmode/init.c index c5bdc4e473e7..747b71e8f547 100644 --- a/arch/x86/realmode/init.c +++ b/arch/x86/realmode/init.c @@ -30,7 +30,7 @@ void __init reserve_real_mode(void) base, (unsigned long long)mem, size); } -void __init setup_real_mode(void) +static void __init setup_real_mode(void) { u16 real_mode_seg; const u32 *rel; @@ -101,7 +101,7 @@ void __init setup_real_mode(void) * need to mark it executable at do_pre_smp_initcalls() at least, * thus run it as a early_initcall(). */ -static int __init set_real_mode_permissions(void) +static void __init set_real_mode_permissions(void) { unsigned char *base = (unsigned char *) real_mode_header; size_t size = PAGE_ALIGN(real_mode_blob_end - real_mode_blob); @@ -120,7 +120,16 @@ static int __init set_real_mode_permissions(void) set_memory_nx((unsigned long) base, size >> PAGE_SHIFT); set_memory_ro((unsigned long) base, ro_size >> PAGE_SHIFT); set_memory_x((unsigned long) text_start, text_size >> PAGE_SHIFT); +} + +static int __init init_real_mode(void) +{ + if (!real_mode_header) + panic("Real mode trampoline was not allocated"); + + setup_real_mode(); + set_real_mode_permissions(); return 0; } -early_initcall(set_real_mode_permissions); +early_initcall(init_real_mode); From 5ff3e2c3c3eebe13967d81ad1f23b9468fefea81 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 10 Aug 2016 02:29:16 -0700 Subject: [PATCH 192/235] x86/boot: Rework reserve_real_mode() to allow multiple tries If reserve_real_mode() fails, panicing immediately means we're doomed. Make it safe to try more than once to allocate the trampoline: - Degrade a failure from panic() to pr_info(). (If we make it to setup_real_mode() without reserving the trampoline, we'll panic them.) - Factor out helpers so that platform code can supply a specific address to try. - Warn if reserve_real_mode() is called after we're done with the memblock allocator. If that were to happen, we would behave unpredictably. Signed-off-by: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mario Limonciello Cc: Matt Fleming Cc: Matthew Garrett Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/876e383038f3e9971aa72fd20a4f5da05f9d193d.1470821230.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/include/asm/realmode.h | 9 +++++++++ arch/x86/realmode/init.c | 35 ++++++++++++++++++++++----------- 2 files changed, 33 insertions(+), 11 deletions(-) diff --git a/arch/x86/include/asm/realmode.h b/arch/x86/include/asm/realmode.h index 8d6777724ba4..b2988c0ed829 100644 --- a/arch/x86/include/asm/realmode.h +++ b/arch/x86/include/asm/realmode.h @@ -58,6 +58,15 @@ extern unsigned char boot_gdt[]; extern unsigned char secondary_startup_64[]; #endif +static inline size_t real_mode_size_needed(void) +{ + if (real_mode_header) + return 0; /* already allocated. */ + + return ALIGN(real_mode_blob_end - real_mode_blob, PAGE_SIZE); +} + +void set_real_mode_mem(phys_addr_t mem, size_t size); void reserve_real_mode(void); #endif /* _ARCH_X86_REALMODE_H */ diff --git a/arch/x86/realmode/init.c b/arch/x86/realmode/init.c index 747b71e8f547..5db706f14111 100644 --- a/arch/x86/realmode/init.c +++ b/arch/x86/realmode/init.c @@ -1,4 +1,5 @@ #include +#include #include #include @@ -12,24 +13,36 @@ u32 *trampoline_cr4_features; /* Hold the pgd entry used on booting additional CPUs */ pgd_t trampoline_pgd_entry; -void __init reserve_real_mode(void) +void __init set_real_mode_mem(phys_addr_t mem, size_t size) { - phys_addr_t mem; - unsigned char *base; - size_t size = PAGE_ALIGN(real_mode_blob_end - real_mode_blob); + void *base = __va(mem); - /* Has to be under 1M so we can execute real-mode AP code. */ - mem = memblock_find_in_range(0, 1<<20, size, PAGE_SIZE); - if (!mem) - panic("Cannot allocate trampoline\n"); - - base = __va(mem); - memblock_reserve(mem, size); real_mode_header = (struct real_mode_header *) base; printk(KERN_DEBUG "Base memory trampoline at [%p] %llx size %zu\n", base, (unsigned long long)mem, size); } +void __init reserve_real_mode(void) +{ + phys_addr_t mem; + size_t size = real_mode_size_needed(); + + if (!size) + return; + + WARN_ON(slab_is_available()); + + /* Has to be under 1M so we can execute real-mode AP code. */ + mem = memblock_find_in_range(0, 1<<20, size, PAGE_SIZE); + if (!mem) { + pr_info("No sub-1M memory is available for the trampoline\n"); + return; + } + + memblock_reserve(mem, size); + set_real_mode_mem(mem, size); +} + static void __init setup_real_mode(void) { u16 real_mode_seg; From 5bc653b7318217c54244a14f248f1f07abe0a865 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 10 Aug 2016 02:29:17 -0700 Subject: [PATCH 193/235] x86/efi: Allocate a trampoline if needed in efi_free_boot_services() On my Dell XPS 13 9350 with firmware 1.4.4 and SGX on, if I boot Fedora 24's grub2-efi off a hard disk, my first 1MB of RAM looks like: efi: mem00: [Runtime Data |RUN| | | | | | | |WB|WT|WC|UC] range=[0x0000000000000000-0x0000000000000fff] (0MB) efi: mem01: [Boot Data | | | | | | | | |WB|WT|WC|UC] range=[0x0000000000001000-0x0000000000027fff] (0MB) efi: mem02: [Loader Data | | | | | | | | |WB|WT|WC|UC] range=[0x0000000000028000-0x0000000000029fff] (0MB) efi: mem03: [Reserved | | | | | | | | |WB|WT|WC|UC] range=[0x000000000002a000-0x000000000002bfff] (0MB) efi: mem04: [Runtime Data |RUN| | | | | | | |WB|WT|WC|UC] range=[0x000000000002c000-0x000000000002cfff] (0MB) efi: mem05: [Loader Data | | | | | | | | |WB|WT|WC|UC] range=[0x000000000002d000-0x000000000002dfff] (0MB) efi: mem06: [Conventional Memory| | | | | | | | |WB|WT|WC|UC] range=[0x000000000002e000-0x0000000000057fff] (0MB) efi: mem07: [Reserved | | | | | | | | |WB|WT|WC|UC] range=[0x0000000000058000-0x0000000000058fff] (0MB) efi: mem08: [Conventional Memory| | | | | | | | |WB|WT|WC|UC] range=[0x0000000000059000-0x000000000009ffff] (0MB) My EBDA is at 0x2c000, which blocks off everything from 0x2c000 and up, and my trampoline is 0x6000 bytes (6 pages), so it doesn't fit in the loader data range at 0x28000. Without this patch, it panics due to a failure to allocate the trampoline. With this patch, it works: [ +0.001744] Base memory trampoline at [ffff880000001000] 1000 size 24576 Signed-off-by: Andy Lutomirski Reviewed-by: Matt Fleming Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mario Limonciello Cc: Matt Fleming Cc: Matthew Garrett Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/998c77b3bf709f3dfed85cb30701ed1a5d8a438b.1470821230.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/platform/efi/quirks.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/arch/x86/platform/efi/quirks.c b/arch/x86/platform/efi/quirks.c index 4480c06cade7..89d1146f5a6f 100644 --- a/arch/x86/platform/efi/quirks.c +++ b/arch/x86/platform/efi/quirks.c @@ -254,6 +254,7 @@ void __init efi_free_boot_services(void) for_each_efi_memory_desc(md) { unsigned long long start = md->phys_addr; unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; + size_t rm_size; if (md->type != EFI_BOOT_SERVICES_CODE && md->type != EFI_BOOT_SERVICES_DATA) @@ -263,6 +264,26 @@ void __init efi_free_boot_services(void) if (md->attribute & EFI_MEMORY_RUNTIME) continue; + /* + * Nasty quirk: if all sub-1MB memory is used for boot + * services, we can get here without having allocated the + * real mode trampoline. It's too late to hand boot services + * memory back to the memblock allocator, so instead + * try to manually allocate the trampoline if needed. + * + * I've seen this on a Dell XPS 13 9350 with firmware + * 1.4.4 with SGX enabled booting Linux via Fedora 24's + * grub2-efi on a hard disk. (And no, I don't know why + * this happened, but Linux should still try to boot rather + * panicing early.) + */ + rm_size = real_mode_size_needed(); + if (rm_size && (start + rm_size) < (1<<20) && size >= rm_size) { + set_real_mode_mem(start, rm_size); + start += rm_size; + size -= rm_size; + } + free_bootmem_late(start, size); } From f72075c9eda8a43aeea2f9dbb8d187afd4a76f0b Mon Sep 17 00:00:00 2001 From: Alex Thorlton Date: Thu, 11 Aug 2016 11:41:59 +0100 Subject: [PATCH 194/235] x86/platform/uv: Skip UV runtime services mapping in the efi_runtime_disabled case This problem has actually been in the UV code for a while, but we didn't catch it until recently, because we had been relying on EFI_OLD_MEMMAP to allow our systems to boot for a period of time. We noticed the issue when trying to kexec a recent community kernel, where we hit this NULL pointer dereference in efi_sync_low_kernel_mappings(): [ 0.337515] BUG: unable to handle kernel NULL pointer dereference at 0000000000000880 [ 0.346276] IP: [] efi_sync_low_kernel_mappings+0x5d/0x1b0 The problem doesn't show up with EFI_OLD_MEMMAP because we skip the chunk of setup_efi_state() that sets the efi_loader_signature for the kexec'd kernel. When the kexec'd kernel boots, it won't set EFI_BOOT in setup_arch, so we completely avoid the bug. We always kexec with noefi on the command line, so this shouldn't be an issue, but since we're not actually checking for efi_runtime_disabled in uv_bios_init(), we end up trying to do EFI runtime callbacks when we shouldn't be. This patch just adds a check for efi_runtime_disabled in uv_bios_init() so that we don't map in uv_systab when runtime_disabled == true. Signed-off-by: Alex Thorlton Signed-off-by: Matt Fleming Cc: # v4.7 Cc: Andy Lutomirski Cc: Ard Biesheuvel Cc: Borislav Petkov Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Mike Travis Cc: Peter Zijlstra Cc: Russ Anderson Cc: Thomas Gleixner Cc: linux-efi@vger.kernel.org Link: http://lkml.kernel.org/r/1470912120-22831-2-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- arch/x86/platform/uv/bios_uv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/platform/uv/bios_uv.c b/arch/x86/platform/uv/bios_uv.c index 66b2166ea4a1..0df8a0370d32 100644 --- a/arch/x86/platform/uv/bios_uv.c +++ b/arch/x86/platform/uv/bios_uv.c @@ -187,7 +187,8 @@ EXPORT_SYMBOL_GPL(uv_bios_set_legacy_vga_target); void uv_bios_init(void) { uv_systab = NULL; - if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) || !efi.uv_systab) { + if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) || + !efi.uv_systab || efi_runtime_disabled()) { pr_crit("UV: UVsystab: missing\n"); return; } From 6862e6ad95e984991a6ceec592cf67831658f928 Mon Sep 17 00:00:00 2001 From: Austin Christ Date: Thu, 11 Aug 2016 11:42:00 +0100 Subject: [PATCH 195/235] efi/capsule: Allocate whole capsule into virtual memory According to UEFI 2.6 section 7.5.3, the capsule should be in contiguous virtual memory and firmware may consume the capsule immediately. To correctly implement this functionality, the kernel driver needs to vmap the entire capsule at the time it is made available to firmware. The virtual allocation of the capsule update has been changed from kmap, which was only allocating the first page of the update, to vmap, and allocates the entire data payload. Signed-off-by: Austin Christ Signed-off-by: Matt Fleming Reviewed-by: Matt Fleming Reviewed-by: Lee, Chun-Yi Cc: # v4.7 Cc: Andy Lutomirski Cc: Ard Biesheuvel Cc: Borislav Petkov Cc: Brian Gerst Cc: Bryan O'Donoghue Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Kweh Hock Leong Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-efi@vger.kernel.org Link: http://lkml.kernel.org/r/1470912120-22831-3-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- drivers/firmware/efi/capsule-loader.c | 8 +++++--- drivers/firmware/efi/capsule.c | 6 +++--- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/firmware/efi/capsule-loader.c b/drivers/firmware/efi/capsule-loader.c index c99c24bc79b0..9ae6c116c474 100644 --- a/drivers/firmware/efi/capsule-loader.c +++ b/drivers/firmware/efi/capsule-loader.c @@ -16,6 +16,7 @@ #include #include #include +#include #define NO_FURTHER_WRITE_ACTION -1 @@ -108,14 +109,15 @@ static ssize_t efi_capsule_submit_update(struct capsule_info *cap_info) int ret; void *cap_hdr_temp; - cap_hdr_temp = kmap(cap_info->pages[0]); + cap_hdr_temp = vmap(cap_info->pages, cap_info->index, + VM_MAP, PAGE_KERNEL); if (!cap_hdr_temp) { - pr_debug("%s: kmap() failed\n", __func__); + pr_debug("%s: vmap() failed\n", __func__); return -EFAULT; } ret = efi_capsule_update(cap_hdr_temp, cap_info->pages); - kunmap(cap_info->pages[0]); + vunmap(cap_hdr_temp); if (ret) { pr_err("%s: efi_capsule_update() failed\n", __func__); return ret; diff --git a/drivers/firmware/efi/capsule.c b/drivers/firmware/efi/capsule.c index 53b9fd2293ee..6eedff45e6d7 100644 --- a/drivers/firmware/efi/capsule.c +++ b/drivers/firmware/efi/capsule.c @@ -190,9 +190,9 @@ efi_capsule_update_locked(efi_capsule_header_t *capsule, * map the capsule described by @capsule with its data in @pages and * send it to the firmware via the UpdateCapsule() runtime service. * - * @capsule must be a virtual mapping of the first page in @pages - * (@pages[0]) in the kernel address space. That is, a - * capsule_header_t that describes the entire contents of the capsule + * @capsule must be a virtual mapping of the complete capsule update in the + * kernel address space, as the capsule can be consumed immediately. + * A capsule_header_t that describes the entire contents of the capsule * must be at the start of the first data page. * * Even though this function will validate that the firmware supports From 26f2c75cd2cf10a6120ef02ca9a94db77cc9c8e0 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 11 Aug 2016 14:58:24 +0200 Subject: [PATCH 196/235] sched/cputime: Fix omitted ticks passed in parameter Commit: f9bcf1e0e014 ("sched/cputime: Fix steal time accounting") ... fixes a leak on steal time accounting but forgets to account the ticks passed in parameters, assuming there is only one to take into account. Let's consider that parameter back. Signed-off-by: Frederic Weisbecker Acked-by: Wanpeng Li Cc: Linus Torvalds Cc: Paolo Bonzini Cc: Peter Zijlstra Cc: Radim Cc: Rik van Riel Cc: Thomas Gleixner Cc: Wanpeng Li Cc: linux-tip-commits@vger.kernel.org Link: http://lkml.kernel.org/r/20160811125822.GB4214@lerouge Signed-off-by: Ingo Molnar --- kernel/sched/cputime.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 8b9bcc5a58fa..9858266fb0b3 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -509,12 +509,13 @@ void account_process_tick(struct task_struct *p, int user_tick) void account_idle_ticks(unsigned long ticks) { cputime_t cputime, steal; + if (sched_clock_irqtime) { irqtime_account_idle_ticks(ticks); return; } - cputime = cputime_one_jiffy; + cputime = jiffies_to_cputime(ticks); steal = steal_account_process_time(cputime); if (steal >= cputime) From d52c0569bab4edc888832df44dc7ac28517134f6 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Thu, 11 Aug 2016 16:08:35 +0200 Subject: [PATCH 197/235] x86/apic/x2apic, smp/hotplug: Don't use before alloc in x2apic_cluster_probe() I made a mistake while converting the driver to the hotplug state machine and as a result x2apic_cluster_probe() was accessing cpus_in_cluster before allocating it. This patch fixes it by setting the cpumask after the allocation the memory succeeded. While at it, I marked two functions static which are only used within this file. Reported-by: Laura Abbott Signed-off-by: Sebastian Andrzej Siewior Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 6b2c28471de5 ("x86/x2apic: Convert to CPU hotplug state machine") Link: http://lkml.kernel.org/r/1470924515-9444-1-git-send-email-bigeasy@linutronix.de Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/x2apic_cluster.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c index 6368fa69d2af..54f35d988025 100644 --- a/arch/x86/kernel/apic/x2apic_cluster.c +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -155,7 +155,7 @@ static void init_x2apic_ldr(void) /* * At CPU state changes, update the x2apic cluster sibling info. */ -int x2apic_prepare_cpu(unsigned int cpu) +static int x2apic_prepare_cpu(unsigned int cpu) { if (!zalloc_cpumask_var(&per_cpu(cpus_in_cluster, cpu), GFP_KERNEL)) return -ENOMEM; @@ -168,7 +168,7 @@ int x2apic_prepare_cpu(unsigned int cpu) return 0; } -int x2apic_dead_cpu(unsigned int this_cpu) +static int x2apic_dead_cpu(unsigned int this_cpu) { int cpu; @@ -186,13 +186,18 @@ int x2apic_dead_cpu(unsigned int this_cpu) static int x2apic_cluster_probe(void) { int cpu = smp_processor_id(); + int ret; if (!x2apic_mode) return 0; + ret = cpuhp_setup_state(CPUHP_X2APIC_PREPARE, "X2APIC_PREPARE", + x2apic_prepare_cpu, x2apic_dead_cpu); + if (ret < 0) { + pr_err("Failed to register X2APIC_PREPARE\n"); + return 0; + } cpumask_set_cpu(cpu, per_cpu(cpus_in_cluster, cpu)); - cpuhp_setup_state(CPUHP_X2APIC_PREPARE, "X2APIC_PREPARE", - x2apic_prepare_cpu, x2apic_dead_cpu); return 1; } From c21377f8366c95440d533edbe47d070f662c62ef Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Thu, 11 Aug 2016 09:35:57 -0600 Subject: [PATCH 198/235] nvme: Suspend all queues before deletion When nvme_delete_queue fails in the first pass of the nvme_disable_io_queues() loop, we return early, failing to suspend all of the IO queues. Later, on the nvme_pci_disable path, this causes us to disable MSI without actually having freed all the IRQs, which triggers the BUG_ON in free_msi_irqs(), as show below. This patch refactors nvme_disable_io_queues to suspend all queues before start submitting delete queue commands. This way, we ensure that we have at least returned every IRQ before continuing with the removal path. [ 487.529200] kernel BUG at ../drivers/pci/msi.c:368! cpu 0x46: Vector: 700 (Program Check) at [c0000078c5b83650] pc: c000000000627a50: free_msi_irqs+0x90/0x200 lr: c000000000627a40: free_msi_irqs+0x80/0x200 sp: c0000078c5b838d0 msr: 9000000100029033 current = 0xc0000078c5b40000 paca = 0xc000000002bd7600 softe: 0 irq_happened: 0x01 pid = 1376, comm = kworker/70:1H kernel BUG at ../drivers/pci/msi.c:368! Linux version 4.7.0.mainline+ (root@iod76) (gcc version 5.3.1 20160413 (Ubuntu/IBM 5.3.1-14ubuntu2.1) ) #104 SMP Fri Jul 29 09:20:17 CDT 2016 enter ? for help [c0000078c5b83920] d0000000363b0cd8 nvme_dev_disable+0x208/0x4f0 [nvme] [c0000078c5b83a10] d0000000363b12a4 nvme_timeout+0xe4/0x250 [nvme] [c0000078c5b83ad0] c0000000005690e4 blk_mq_rq_timed_out+0x64/0x110 [c0000078c5b83b40] c00000000056c930 bt_for_each+0x160/0x170 [c0000078c5b83bb0] c00000000056d928 blk_mq_queue_tag_busy_iter+0x78/0x110 [c0000078c5b83c00] c0000000005675d8 blk_mq_timeout_work+0xd8/0x1b0 [c0000078c5b83c50] c0000000000e8cf0 process_one_work+0x1e0/0x590 [c0000078c5b83ce0] c0000000000e9148 worker_thread+0xa8/0x660 [c0000078c5b83d80] c0000000000f2090 kthread+0x110/0x130 [c0000078c5b83e30] c0000000000095f0 ret_from_kernel_thread+0x5c/0x6c Signed-off-by: Gabriel Krisman Bertazi Cc: Brian King Cc: Keith Busch Cc: linux-nvme@lists.infradead.org Signed-off-by: Jens Axboe --- drivers/nvme/host/pci.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index d7c33f9361aa..8dcf5a960951 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1543,15 +1543,10 @@ static void nvme_disable_io_queues(struct nvme_dev *dev) reinit_completion(&dev->ioq_wait); retry: timeout = ADMIN_TIMEOUT; - for (; i > 0; i--) { - struct nvme_queue *nvmeq = dev->queues[i]; - - if (!pass) - nvme_suspend_queue(nvmeq); - if (nvme_delete_queue(nvmeq, opcode)) + for (; i > 0; i--, sent++) + if (nvme_delete_queue(dev->queues[i], opcode)) break; - ++sent; - } + while (sent--) { timeout = wait_for_completion_io_timeout(&dev->ioq_wait, timeout); if (timeout == 0) @@ -1693,11 +1688,12 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown) nvme_stop_queues(&dev->ctrl); csts = readl(dev->bar + NVME_REG_CSTS); } + + for (i = dev->queue_count - 1; i > 0; i--) + nvme_suspend_queue(dev->queues[i]); + if (csts & NVME_CSTS_CFS || !(csts & NVME_CSTS_RDY)) { - for (i = dev->queue_count - 1; i >= 0; i--) { - struct nvme_queue *nvmeq = dev->queues[i]; - nvme_suspend_queue(nvmeq); - } + nvme_suspend_queue(dev->queues[0]); } else { nvme_disable_io_queues(dev); nvme_disable_admin_queue(dev, shutdown); From 005411ea7ee776a56b1e0120a31c65efdee5cab1 Mon Sep 17 00:00:00 2001 From: Joe Lawrence Date: Tue, 9 Aug 2016 14:01:30 -0400 Subject: [PATCH 199/235] doc: update block/queue-sysfs.txt entries Add descriptions for dax, io_poll, and write_same_max_bytes files. Signed-off-by: Joe Lawrence Signed-off-by: Jens Axboe --- Documentation/block/queue-sysfs.txt | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/Documentation/block/queue-sysfs.txt b/Documentation/block/queue-sysfs.txt index d515d58962b9..2a3904030dea 100644 --- a/Documentation/block/queue-sysfs.txt +++ b/Documentation/block/queue-sysfs.txt @@ -14,6 +14,12 @@ add_random (RW) This file allows to turn off the disk entropy contribution. Default value of this file is '1'(on). +dax (RO) +-------- +This file indicates whether the device supports Direct Access (DAX), +used by CPU-addressable storage to bypass the pagecache. It shows '1' +if true, '0' if not. + discard_granularity (RO) ----------------------- This shows the size of internal allocation of the device in bytes, if @@ -46,6 +52,12 @@ hw_sector_size (RO) ------------------- This is the hardware sector size of the device, in bytes. +io_poll (RW) +------------ +When read, this file shows the total number of block IO polls and how +many returned success. Writing '0' to this file will disable polling +for this device. Writing any non-zero value will enable this feature. + iostats (RW) ------------- This file is used to control (on/off) the iostats accounting of the @@ -151,5 +163,11 @@ device state. This means that it might not be safe to toggle the setting from "write back" to "write through", since that will also eliminate cache flushes issued by the kernel. +write_same_max_bytes (RO) +------------------------- +This is the number of bytes the device can write in a single write-same +command. A value of '0' means write-same is not supported by this +device. + Jens Axboe , February 2009 From 1ea049b2de5d803374fdbf43add23c8d1c518e7b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 11 Aug 2016 10:15:56 +0200 Subject: [PATCH 200/235] bvec: avoid variable shadowing warning Due to the (indirect) nesting of min(..., min(...)), sparse will show a variable shadowing warning whenever bvec.h is included. Avoid that by assigning the inner min() to a temporary variable first. Signed-off-by: Johannes Berg Signed-off-by: Jens Axboe --- include/linux/bvec.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/linux/bvec.h b/include/linux/bvec.h index 701b64a3b7c5..89b65b82d98f 100644 --- a/include/linux/bvec.h +++ b/include/linux/bvec.h @@ -74,7 +74,8 @@ static inline void bvec_iter_advance(const struct bio_vec *bv, "Attempted to advance past end of bvec iter\n"); while (bytes) { - unsigned len = min(bytes, bvec_iter_len(bv, *iter)); + unsigned iter_len = bvec_iter_len(bv, *iter); + unsigned len = min(bytes, iter_len); bytes -= len; iter->bi_size -= len; From 502aa0a5be633e6558574ebcf63b65afdfbfcd7a Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 10 Aug 2016 14:46:27 -0400 Subject: [PATCH 201/235] nfsd: fix dentry refcounting on create b44061d0b9 introduced a dentry ref counting bug. Previously we were grabbing one ref to dchild in nfsd_create(), but with the creation of nfsd_create_locked() we have a ref for dchild from the lookup in nfsd_create(), and then another ref in nfsd_create_locked(). The ref from the lookup in nfsd_create() is never dropped and results in dentries still in use at unmount. Signed-off-by: Josef Bacik Fixes: b44061d0b9 "nfsd: reorganize nfsd_create" Reported-by: kernel test robot Reviewed-by: Jeff Layton Acked-by: Al Viro Signed-off-by: J. Bruce Fields --- fs/nfsd/vfs.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ba944123167b..ff476e654b8f 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1252,10 +1252,13 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dchild)) return nfserrno(host_err); err = fh_compose(resfhp, fhp->fh_export, dchild, fhp); - if (err) { - dput(dchild); + /* + * We unconditionally drop our ref to dchild as fh_compose will have + * already grabbed its own ref for it. + */ + dput(dchild); + if (err) return err; - } return nfsd_create_locked(rqstp, fhp, fname, flen, iap, type, rdev, resfhp); } From ad05711cec12131e1277ce749a99d08ecf233aa7 Mon Sep 17 00:00:00 2001 From: "David A. Long" Date: Wed, 10 Aug 2016 16:44:51 -0400 Subject: [PATCH 202/235] arm64: Remove stack duplicating code from jprobes Because the arm64 calling standard allows stacked function arguments to be anywhere in the stack frame, do not attempt to duplicate the stack frame for jprobes handler functions. Documentation changes to describe this issue have been broken out into a separate patch in order to simultaneously address them in other architecture(s). Signed-off-by: David A. Long Acked-by: Masami Hiramatsu Acked-by: Marc Zyngier Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/kprobes.h | 2 -- arch/arm64/kernel/probes/kprobes.c | 31 +++++------------------------- 2 files changed, 5 insertions(+), 28 deletions(-) diff --git a/arch/arm64/include/asm/kprobes.h b/arch/arm64/include/asm/kprobes.h index 61b49150dfa3..1737aecfcc5e 100644 --- a/arch/arm64/include/asm/kprobes.h +++ b/arch/arm64/include/asm/kprobes.h @@ -22,7 +22,6 @@ #define __ARCH_WANT_KPROBES_INSN_SLOT #define MAX_INSN_SIZE 1 -#define MAX_STACK_SIZE 128 #define flush_insn_slot(p) do { } while (0) #define kretprobe_blacklist_size 0 @@ -47,7 +46,6 @@ struct kprobe_ctlblk { struct prev_kprobe prev_kprobe; struct kprobe_step_ctx ss_ctx; struct pt_regs jprobe_saved_regs; - char jprobes_stack[MAX_STACK_SIZE]; }; void arch_remove_kprobe(struct kprobe *); diff --git a/arch/arm64/kernel/probes/kprobes.c b/arch/arm64/kernel/probes/kprobes.c index bf9768588288..c6b0f40620d8 100644 --- a/arch/arm64/kernel/probes/kprobes.c +++ b/arch/arm64/kernel/probes/kprobes.c @@ -41,18 +41,6 @@ DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); static void __kprobes post_kprobe_handler(struct kprobe_ctlblk *, struct pt_regs *); -static inline unsigned long min_stack_size(unsigned long addr) -{ - unsigned long size; - - if (on_irq_stack(addr, raw_smp_processor_id())) - size = IRQ_STACK_PTR(raw_smp_processor_id()) - addr; - else - size = (unsigned long)current_thread_info() + THREAD_START_SP - addr; - - return min(size, FIELD_SIZEOF(struct kprobe_ctlblk, jprobes_stack)); -} - static void __kprobes arch_prepare_ss_slot(struct kprobe *p) { /* prepare insn slot */ @@ -489,20 +477,15 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) { struct jprobe *jp = container_of(p, struct jprobe, kp); struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); - long stack_ptr = kernel_stack_pointer(regs); kcb->jprobe_saved_regs = *regs; /* - * As Linus pointed out, gcc assumes that the callee - * owns the argument space and could overwrite it, e.g. - * tailcall optimization. So, to be absolutely safe - * we also save and restore enough stack bytes to cover - * the argument area. + * Since we can't be sure where in the stack frame "stacked" + * pass-by-value arguments are stored we just don't try to + * duplicate any of the stack. Do not use jprobes on functions that + * use more than 64 bytes (after padding each to an 8 byte boundary) + * of arguments, or pass individual arguments larger than 16 bytes. */ - kasan_disable_current(); - memcpy(kcb->jprobes_stack, (void *)stack_ptr, - min_stack_size(stack_ptr)); - kasan_enable_current(); instruction_pointer_set(regs, (unsigned long) jp->entry); preempt_disable(); @@ -554,10 +537,6 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs) } unpause_graph_tracing(); *regs = kcb->jprobe_saved_regs; - kasan_disable_current(); - memcpy((void *)stack_addr, kcb->jprobes_stack, - min_stack_size(stack_addr)); - kasan_enable_current(); preempt_enable_no_resched(); return 1; } From 42691398be08bd1fe99326911a0aa31f2c041d53 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2016 10:37:30 -0400 Subject: [PATCH 203/235] nfsd: Fix race between FREE_STATEID and LOCK When running LTP's nfslock01 test, the Linux client can send a LOCK and a FREE_STATEID request at the same time. The outcome is: Frame 324 R OPEN stateid [2,O] Frame 115004 C LOCK lockowner_is_new stateid [2,O] offset 672000 len 64 Frame 115008 R LOCK stateid [1,L] Frame 115012 C WRITE stateid [0,L] offset 672000 len 64 Frame 115016 R WRITE NFS4_OK Frame 115019 C LOCKU stateid [1,L] offset 672000 len 64 Frame 115022 R LOCKU NFS4_OK Frame 115025 C FREE_STATEID stateid [2,L] Frame 115026 C LOCK lockowner_is_new stateid [2,O] offset 672128 len 64 Frame 115029 R FREE_STATEID NFS4_OK Frame 115030 R LOCK stateid [3,L] Frame 115034 C WRITE stateid [0,L] offset 672128 len 64 Frame 115038 R WRITE NFS4ERR_BAD_STATEID In other words, the server returns stateid L in a successful LOCK reply, but it has already released it. Subsequent uses of stateid L fail. To address this, protect the generation check in nfsd4_free_stateid with the st_mutex. This should guarantee that only one of two outcomes occurs: either LOCK returns a fresh valid stateid, or FREE_STATEID returns NFS4ERR_LOCKS_HELD. Reported-by: Alexey Kodanev Fix-suggested-by: Jeff Layton Signed-off-by: Chuck Lever Tested-by: Alexey Kodanev Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4state.c | 40 ++++++++++++++++++++++++++++------------ 1 file changed, 28 insertions(+), 12 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 8410ca275db1..0edc1822f144 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -4903,6 +4903,32 @@ nfsd4_test_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return nfs_ok; } +static __be32 +nfsd4_free_lock_stateid(stateid_t *stateid, struct nfs4_stid *s) +{ + struct nfs4_ol_stateid *stp = openlockstateid(s); + __be32 ret; + + mutex_lock(&stp->st_mutex); + + ret = check_stateid_generation(stateid, &s->sc_stateid, 1); + if (ret) + goto out; + + ret = nfserr_locks_held; + if (check_for_locks(stp->st_stid.sc_file, + lockowner(stp->st_stateowner))) + goto out; + + release_lock_stateid(stp); + ret = nfs_ok; + +out: + mutex_unlock(&stp->st_mutex); + nfs4_put_stid(s); + return ret; +} + __be32 nfsd4_free_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_free_stateid *free_stateid) @@ -4910,7 +4936,6 @@ nfsd4_free_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stateid_t *stateid = &free_stateid->fr_stateid; struct nfs4_stid *s; struct nfs4_delegation *dp; - struct nfs4_ol_stateid *stp; struct nfs4_client *cl = cstate->session->se_client; __be32 ret = nfserr_bad_stateid; @@ -4929,18 +4954,9 @@ nfsd4_free_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, ret = nfserr_locks_held; break; case NFS4_LOCK_STID: - ret = check_stateid_generation(stateid, &s->sc_stateid, 1); - if (ret) - break; - stp = openlockstateid(s); - ret = nfserr_locks_held; - if (check_for_locks(stp->st_stid.sc_file, - lockowner(stp->st_stateowner))) - break; - WARN_ON(!unhash_lock_stateid(stp)); + atomic_inc(&s->sc_count); spin_unlock(&cl->cl_lock); - nfs4_put_stid(s); - ret = nfs_ok; + ret = nfsd4_free_lock_stateid(stateid, s); goto out; case NFS4_REVOKED_DELEG_STID: dp = delegstateid(s); From c1470b33bb6e18cddd361fef339ef225b8339fe7 Mon Sep 17 00:00:00 2001 From: zhong jiang Date: Thu, 11 Aug 2016 15:32:55 -0700 Subject: [PATCH 204/235] mm/hugetlb: fix incorrect hugepages count during mem hotplug When memory hotplug operates, free hugepages will be freed if the movable node is offline. Therefore, /proc/sys/vm/nr_hugepages will be incorrect. Fix it by reducing max_huge_pages when the node is offlined. n-horiguchi@ah.jp.nec.com said: : dissolve_free_huge_page intends to break a hugepage into buddy, and the : destination hugepage is supposed to be allocated from the pool of the : destination node, so the system-wide pool size is reduced. So adding : h->max_huge_pages-- makes sense to me. Link: http://lkml.kernel.org/r/1470624546-902-1-git-send-email-zhongjiang@huawei.com Signed-off-by: zhong jiang Cc: Mike Kravetz Acked-by: Naoya Horiguchi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index b9aa1b0b38b0..87e11d8ad536 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1448,6 +1448,7 @@ static void dissolve_free_huge_page(struct page *page) list_del(&page->lru); h->free_huge_pages--; h->free_huge_pages_node[nid]--; + h->max_huge_pages--; update_and_free_page(h, page); } spin_unlock(&hugetlb_lock); From 2f95ff90b99600f53df4a0aa652322d349d67957 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 11 Aug 2016 15:32:57 -0700 Subject: [PATCH 205/235] proc, meminfo: use correct helpers for calculating LRU sizes in meminfo meminfo_proc_show() and si_mem_available() are using the wrong helpers for calculating the size of the LRUs. The user-visible impact is that there appears to be an abnormally high number of unevictable pages. Link: http://lkml.kernel.org/r/20160805105805.GR2799@techsingularity.net Signed-off-by: Mel Gorman Cc: Dave Chinner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/meminfo.c | 2 +- mm/page_alloc.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index 09e18fdf61e5..b9a8c813e5e6 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -46,7 +46,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v) cached = 0; for (lru = LRU_BASE; lru < NR_LRU_LISTS; lru++) - pages[lru] = global_page_state(NR_LRU_BASE + lru); + pages[lru] = global_node_page_state(NR_LRU_BASE + lru); available = si_mem_available(); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ab2c0ff8c2e6..3fbe73a6fe4b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4060,7 +4060,7 @@ long si_mem_available(void) int lru; for (lru = LRU_BASE; lru < NR_LRU_LISTS; lru++) - pages[lru] = global_page_state(NR_LRU_BASE + lru); + pages[lru] = global_node_page_state(NR_LRU_BASE + lru); for_each_zone(zone) wmark_low += zone->watermark[WMARK_LOW]; From 1f47b61fb4077936465dcde872a4e5cc4fe708da Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Thu, 11 Aug 2016 15:33:00 -0700 Subject: [PATCH 206/235] mm: memcontrol: fix swap counter leak on swapout from offline cgroup An offline memory cgroup might have anonymous memory or shmem left charged to it and no swap. Since only swap entries pin the id of an offline cgroup, such a cgroup will have no id and so an attempt to swapout its anon/shmem will not store memory cgroup info in the swap cgroup map. As a result, memcg->swap or memcg->memsw will never get uncharged from it and any of its ascendants. Fix this by always charging swapout to the first ancestor cgroup that hasn't released its id yet. [hannes@cmpxchg.org: add comment to mem_cgroup_swapout] [vdavydov@virtuozzo.com: use WARN_ON_ONCE() in mem_cgroup_id_get_online()] Link: http://lkml.kernel.org/r/20160803123445.GJ13263@esperanza Fixes: 73f576c04b941 ("mm: memcontrol: fix cgroup creation failure after many small jobs") Link: http://lkml.kernel.org/r/5336daa5c9a32e776067773d9da655d2dc126491.1470219853.git.vdavydov@virtuozzo.com Signed-off-by: Vladimir Davydov Acked-by: Johannes Weiner Acked-by: Michal Hocko Cc: [3.19+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 48 ++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 40 insertions(+), 8 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index e74d7080ec9e..791b00ca4e8b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4082,6 +4082,24 @@ static void mem_cgroup_id_get(struct mem_cgroup *memcg) atomic_inc(&memcg->id.ref); } +static struct mem_cgroup *mem_cgroup_id_get_online(struct mem_cgroup *memcg) +{ + while (!atomic_inc_not_zero(&memcg->id.ref)) { + /* + * The root cgroup cannot be destroyed, so it's refcount must + * always be >= 1. + */ + if (WARN_ON_ONCE(memcg == root_mem_cgroup)) { + VM_BUG_ON(1); + break; + } + memcg = parent_mem_cgroup(memcg); + if (!memcg) + memcg = root_mem_cgroup; + } + return memcg; +} + static void mem_cgroup_id_put(struct mem_cgroup *memcg) { if (atomic_dec_and_test(&memcg->id.ref)) { @@ -5800,7 +5818,7 @@ subsys_initcall(mem_cgroup_init); */ void mem_cgroup_swapout(struct page *page, swp_entry_t entry) { - struct mem_cgroup *memcg; + struct mem_cgroup *memcg, *swap_memcg; unsigned short oldid; VM_BUG_ON_PAGE(PageLRU(page), page); @@ -5815,16 +5833,27 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry) if (!memcg) return; - mem_cgroup_id_get(memcg); - oldid = swap_cgroup_record(entry, mem_cgroup_id(memcg)); + /* + * In case the memcg owning these pages has been offlined and doesn't + * have an ID allocated to it anymore, charge the closest online + * ancestor for the swap instead and transfer the memory+swap charge. + */ + swap_memcg = mem_cgroup_id_get_online(memcg); + oldid = swap_cgroup_record(entry, mem_cgroup_id(swap_memcg)); VM_BUG_ON_PAGE(oldid, page); - mem_cgroup_swap_statistics(memcg, true); + mem_cgroup_swap_statistics(swap_memcg, true); page->mem_cgroup = NULL; if (!mem_cgroup_is_root(memcg)) page_counter_uncharge(&memcg->memory, 1); + if (memcg != swap_memcg) { + if (!mem_cgroup_is_root(swap_memcg)) + page_counter_charge(&swap_memcg->memsw, 1); + page_counter_uncharge(&memcg->memsw, 1); + } + /* * Interrupts should be disabled here because the caller holds the * mapping->tree_lock lock which is taken with interrupts-off. It is @@ -5863,11 +5892,14 @@ int mem_cgroup_try_charge_swap(struct page *page, swp_entry_t entry) if (!memcg) return 0; - if (!mem_cgroup_is_root(memcg) && - !page_counter_try_charge(&memcg->swap, 1, &counter)) - return -ENOMEM; + memcg = mem_cgroup_id_get_online(memcg); + + if (!mem_cgroup_is_root(memcg) && + !page_counter_try_charge(&memcg->swap, 1, &counter)) { + mem_cgroup_id_put(memcg); + return -ENOMEM; + } - mem_cgroup_id_get(memcg); oldid = swap_cgroup_record(entry, mem_cgroup_id(memcg)); VM_BUG_ON_PAGE(oldid, page); mem_cgroup_swap_statistics(memcg, true); From 615d66c37c755c49ce022c9e5ac0875d27d2603d Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Thu, 11 Aug 2016 15:33:03 -0700 Subject: [PATCH 207/235] mm: memcontrol: fix memcg id ref counter on swap charge move Since commit 73f576c04b94 ("mm: memcontrol: fix cgroup creation failure after many small jobs") swap entries do not pin memcg->css.refcnt directly. Instead, they pin memcg->id.ref. So we should adjust the reference counters accordingly when moving swap charges between cgroups. Fixes: 73f576c04b941 ("mm: memcontrol: fix cgroup creation failure after many small jobs") Link: http://lkml.kernel.org/r/9ce297c64954a42dc90b543bc76106c4a94f07e8.1470219853.git.vdavydov@virtuozzo.com Signed-off-by: Vladimir Davydov Acked-by: Michal Hocko Acked-by: Johannes Weiner Cc: [3.19+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 791b00ca4e8b..2ff0289ad061 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4077,9 +4077,9 @@ static struct cftype mem_cgroup_legacy_files[] = { static DEFINE_IDR(mem_cgroup_idr); -static void mem_cgroup_id_get(struct mem_cgroup *memcg) +static void mem_cgroup_id_get_many(struct mem_cgroup *memcg, unsigned int n) { - atomic_inc(&memcg->id.ref); + atomic_add(n, &memcg->id.ref); } static struct mem_cgroup *mem_cgroup_id_get_online(struct mem_cgroup *memcg) @@ -4100,9 +4100,9 @@ static struct mem_cgroup *mem_cgroup_id_get_online(struct mem_cgroup *memcg) return memcg; } -static void mem_cgroup_id_put(struct mem_cgroup *memcg) +static void mem_cgroup_id_put_many(struct mem_cgroup *memcg, unsigned int n) { - if (atomic_dec_and_test(&memcg->id.ref)) { + if (atomic_sub_and_test(n, &memcg->id.ref)) { idr_remove(&mem_cgroup_idr, memcg->id.id); memcg->id.id = 0; @@ -4111,6 +4111,16 @@ static void mem_cgroup_id_put(struct mem_cgroup *memcg) } } +static inline void mem_cgroup_id_get(struct mem_cgroup *memcg) +{ + mem_cgroup_id_get_many(memcg, 1); +} + +static inline void mem_cgroup_id_put(struct mem_cgroup *memcg) +{ + mem_cgroup_id_put_many(memcg, 1); +} + /** * mem_cgroup_from_id - look up a memcg from a memcg id * @id: the memcg id to look up @@ -4745,6 +4755,8 @@ static void __mem_cgroup_clear_mc(void) if (!mem_cgroup_is_root(mc.from)) page_counter_uncharge(&mc.from->memsw, mc.moved_swap); + mem_cgroup_id_put_many(mc.from, mc.moved_swap); + /* * we charged both to->memory and to->memsw, so we * should uncharge to->memory. @@ -4752,9 +4764,9 @@ static void __mem_cgroup_clear_mc(void) if (!mem_cgroup_is_root(mc.to)) page_counter_uncharge(&mc.to->memory, mc.moved_swap); - css_put_many(&mc.from->css, mc.moved_swap); + mem_cgroup_id_get_many(mc.to, mc.moved_swap); + css_put_many(&mc.to->css, mc.moved_swap); - /* we've already done css_get(mc.to) */ mc.moved_swap = 0; } memcg_oom_recover(from); From bcbf0d566b6e59a6e873bfe415cc415111a819e2 Mon Sep 17 00:00:00 2001 From: Alexander Potapenko Date: Thu, 11 Aug 2016 15:33:06 -0700 Subject: [PATCH 208/235] kasan: remove the unnecessary WARN_ONCE from quarantine.c It's quite unlikely that the user will so little memory that the per-CPU quarantines won't fit into the given fraction of the available memory. Even in that case he won't be able to do anything with the information given in the warning. Link: http://lkml.kernel.org/r/1470929182-101413-1-git-send-email-glider@google.com Signed-off-by: Alexander Potapenko Acked-by: Andrey Ryabinin Cc: Dmitry Vyukov Cc: Andrey Konovalov Cc: Christoph Lameter Cc: Joonsoo Kim Cc: Kuthonuzo Luruo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/kasan/quarantine.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/mm/kasan/quarantine.c b/mm/kasan/quarantine.c index b6728a33a4ac..baabaad4a4aa 100644 --- a/mm/kasan/quarantine.c +++ b/mm/kasan/quarantine.c @@ -217,11 +217,8 @@ void quarantine_reduce(void) new_quarantine_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) / QUARANTINE_FRACTION; percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus(); - if (WARN_ONCE(new_quarantine_size < percpu_quarantines, - "Too little memory, disabling global KASAN quarantine.\n")) - new_quarantine_size = 0; - else - new_quarantine_size -= percpu_quarantines; + new_quarantine_size = (new_quarantine_size < percpu_quarantines) ? + 0 : new_quarantine_size - percpu_quarantines; WRITE_ONCE(quarantine_size, new_quarantine_size); last = global_quarantine.head; From f33e6f0671b3ba81acef4d7c078af86afcc855c4 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 11 Aug 2016 15:33:09 -0700 Subject: [PATCH 209/235] mm, oom: fix uninitialized ret in task_will_free_mem() mm/oom_kill.c: In function `task_will_free_mem': mm/oom_kill.c:767: warning: `ret' may be used uninitialized in this function If __task_will_free_mem() is never called inside the for_each_process() loop, ret will not be initialized. Fixes: 1af8bb43269563e4 ("mm, oom: fortify task_will_free_mem()") Link: http://lkml.kernel.org/r/1470255599-24841-1-git-send-email-geert@linux-m68k.org Signed-off-by: Geert Uytterhoeven Acked-by: Tetsuo Handa Acked-by: Michal Hocko Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/oom_kill.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 7d0a275df822..d53a9aa00977 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -764,7 +764,7 @@ bool task_will_free_mem(struct task_struct *task) { struct mm_struct *mm = task->mm; struct task_struct *p; - bool ret; + bool ret = true; /* * Skip tasks without mm because it might have passed its exit_mm and From 5830169f47269f78f6624bd70165eb571270da82 Mon Sep 17 00:00:00 2001 From: Reza Arbab Date: Thu, 11 Aug 2016 15:33:12 -0700 Subject: [PATCH 210/235] mm/memory_hotplug.c: initialize per_cpu_nodestats for hotadded pgdats The following oops occurs after a pgdat is hotadded: Unable to handle kernel paging request for data at address 0x00c30001 Faulting instruction address: 0xc00000000022f8f4 Oops: Kernel access of bad area, sig: 11 [#1] SMP NR_CPUS=2048 NUMA pSeries Modules linked in: ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter nls_utf8 isofs sg virtio_balloon uio_pdrv_genirq uio ip_tables xfs libcrc32c sr_mod cdrom sd_mod virtio_net ibmvscsi scsi_transport_srp virtio_pci virtio_ring virtio dm_mirror dm_region_hash dm_log dm_mod CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 4.8.0-rc1-device #110 task: c000000000ef3080 task.stack: c000000000f6c000 NIP: c00000000022f8f4 LR: c00000000022f948 CTR: 0000000000000000 REGS: c000000000f6fa50 TRAP: 0300 Tainted: G W (4.8.0-rc1-device) MSR: 800000010280b033 CR: 84002028 XER: 20000000 CFAR: d000000001d2013c DAR: 0000000000c30001 DSISR: 40000000 SOFTE: 0 NIP refresh_cpu_vm_stats+0x1a4/0x2f0 LR refresh_cpu_vm_stats+0x1f8/0x2f0 Call Trace: refresh_cpu_vm_stats+0x1f8/0x2f0 (unreliable) Add per_cpu_nodestats initialization to the hotplug codepath. Link: http://lkml.kernel.org/r/1470931473-7090-1-git-send-email-arbab@linux.vnet.ibm.com Signed-off-by: Reza Arbab Cc: Mel Gorman Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 3894b65b1555..41266dc29f33 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -1219,6 +1219,7 @@ static pg_data_t __ref *hotadd_new_pgdat(int nid, u64 start) /* init node's zones as empty zones, we don't have any present pages.*/ free_area_init_node(nid, zones_size, start_pfn, zholes_size); + pgdat->per_cpu_nodestats = alloc_percpu(struct per_cpu_nodestat); /* * The node we allocated has no zone fallback lists. For avoiding @@ -1249,6 +1250,7 @@ static pg_data_t __ref *hotadd_new_pgdat(int nid, u64 start) static void rollback_node_hotadd(int nid, pg_data_t *pgdat) { arch_refresh_nodedata(nid, NULL); + free_percpu(pgdat->per_cpu_nodestats); arch_free_nodedata(pgdat); return; } From 68187872c76a96ed4db7bfb064272591f02e208b Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Thu, 11 Aug 2016 17:45:21 +0200 Subject: [PATCH 211/235] uprobes/x86: Fix RIP-relative handling of EVEX-encoded instructions Since instruction decoder now supports EVEX-encoded instructions, two fixes are needed to correctly handle them in uprobes. Extended bits for MODRM.rm field need to be sanitized just like we do it for VEX3, to avoid encoding wrong register for register-relative access. EVEX has _two_ extended bits: b and x. Theoretically, EVEX.x should be ignored by the CPU (since GPRs go only up to 15, not 31), but let's be paranoid here: proper encoding for register-relative access should have EVEX.x = 1. Secondly, we should fetch vex.vvvv for EVEX too. This is now super easy because instruction decoder populates vex_prefix.bytes[2] for all flavors of (e)vex encodings, even for VEX2. Signed-off-by: Denys Vlasenko Acked-by: Masami Hiramatsu Acked-by: Srikar Dronamraju Cc: Alexander Shishkin Cc: Andy Lutomirski Cc: Arnaldo Carvalho de Melo Cc: Borislav Petkov Cc: Brian Gerst Cc: H. Peter Anvin Cc: Jim Keniston Cc: Jiri Olsa Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: linux-kernel@vger.kernel.org Cc: # v4.1+ Fixes: 8a764a875fe3 ("x86/asm/decoder: Create artificial 3rd byte for 2-byte VEX") Link: http://lkml.kernel.org/r/20160811154521.20469-1-dvlasenk@redhat.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/uprobes.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/arch/x86/kernel/uprobes.c b/arch/x86/kernel/uprobes.c index 6c1ff31d99ff..495c776de4b4 100644 --- a/arch/x86/kernel/uprobes.c +++ b/arch/x86/kernel/uprobes.c @@ -357,20 +357,22 @@ static void riprel_analyze(struct arch_uprobe *auprobe, struct insn *insn) *cursor &= 0xfe; } /* - * Similar treatment for VEX3 prefix. - * TODO: add XOP/EVEX treatment when insn decoder supports them + * Similar treatment for VEX3/EVEX prefix. + * TODO: add XOP treatment when insn decoder supports them */ - if (insn->vex_prefix.nbytes == 3) { + if (insn->vex_prefix.nbytes >= 3) { /* * vex2: c5 rvvvvLpp (has no b bit) * vex3/xop: c4/8f rxbmmmmm wvvvvLpp * evex: 62 rxbR00mm wvvvv1pp zllBVaaa - * (evex will need setting of both b and x since - * in non-sib encoding evex.x is 4th bit of MODRM.rm) - * Setting VEX3.b (setting because it has inverted meaning): + * Setting VEX3.b (setting because it has inverted meaning). + * Setting EVEX.x since (in non-SIB encoding) EVEX.x + * is the 4th bit of MODRM.rm, and needs the same treatment. + * For VEX3-encoded insns, VEX3.x value has no effect in + * non-SIB encoding, the change is superfluous but harmless. */ cursor = auprobe->insn + insn_offset_vex_prefix(insn) + 1; - *cursor |= 0x20; + *cursor |= 0x60; } /* @@ -415,12 +417,10 @@ static void riprel_analyze(struct arch_uprobe *auprobe, struct insn *insn) reg = MODRM_REG(insn); /* Fetch modrm.reg */ reg2 = 0xff; /* Fetch vex.vvvv */ - if (insn->vex_prefix.nbytes == 2) - reg2 = insn->vex_prefix.bytes[1]; - else if (insn->vex_prefix.nbytes == 3) + if (insn->vex_prefix.nbytes) reg2 = insn->vex_prefix.bytes[2]; /* - * TODO: add XOP, EXEV vvvv reading. + * TODO: add XOP vvvv reading. * * vex.vvvv field is in bits 6-3, bits are inverted. * But in 32-bit mode, high-order bit may be ignored. From 10e9e7bd598f9a66a11a22514c68c13c41fc821b Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Thu, 11 Aug 2016 07:30:20 -0700 Subject: [PATCH 212/235] perf/x86/intel/uncore: Fix uncore num_counters Some uncore boxes' num_counters value for Haswell server and Broadwell server are not correct (too large, off by one). This issue was found by comparing the code with the document. Although there is no bug report from users yet, accessing non-existent counters is dangerous and the behavior is undefined: it may cause miscounting or even crashes. This patch makes them consistent with the uncore document. Reported-by: Lukasz Odzioba Signed-off-by: Kan Liang Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: Link: http://lkml.kernel.org/r/1470925820-59847-1-git-send-email-kan.liang@intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/uncore_snbep.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index 824e54086e07..8aee83bcf71f 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -2626,7 +2626,7 @@ void hswep_uncore_cpu_init(void) static struct intel_uncore_type hswep_uncore_ha = { .name = "ha", - .num_counters = 5, + .num_counters = 4, .num_boxes = 2, .perf_ctr_bits = 48, SNBEP_UNCORE_PCI_COMMON_INIT(), @@ -2645,7 +2645,7 @@ static struct uncore_event_desc hswep_uncore_imc_events[] = { static struct intel_uncore_type hswep_uncore_imc = { .name = "imc", - .num_counters = 5, + .num_counters = 4, .num_boxes = 8, .perf_ctr_bits = 48, .fixed_ctr_bits = 48, @@ -2691,7 +2691,7 @@ static struct intel_uncore_type hswep_uncore_irp = { static struct intel_uncore_type hswep_uncore_qpi = { .name = "qpi", - .num_counters = 5, + .num_counters = 4, .num_boxes = 3, .perf_ctr_bits = 48, .perf_ctr = SNBEP_PCI_PMON_CTR0, @@ -2773,7 +2773,7 @@ static struct event_constraint hswep_uncore_r3qpi_constraints[] = { static struct intel_uncore_type hswep_uncore_r3qpi = { .name = "r3qpi", - .num_counters = 4, + .num_counters = 3, .num_boxes = 3, .perf_ctr_bits = 44, .constraints = hswep_uncore_r3qpi_constraints, @@ -2972,7 +2972,7 @@ static struct intel_uncore_type bdx_uncore_ha = { static struct intel_uncore_type bdx_uncore_imc = { .name = "imc", - .num_counters = 5, + .num_counters = 4, .num_boxes = 8, .perf_ctr_bits = 48, .fixed_ctr_bits = 48, From 95f3be798472f63b495ca4712af005ea5ac7aa47 Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Thu, 11 Aug 2016 07:31:14 -0700 Subject: [PATCH 213/235] perf/x86/intel/uncore: Add enable_box for client MSR uncore There are bug reports about miscounting uncore counters on some client machines like Sandybridge, Broadwell and Skylake. It is very likely to be observed on idle systems. This issue is caused by a hardware issue. PERF_GLOBAL_CTL could be cleared after Package C7, and nothing will be count. The related errata (HSD 158) could be found in: www.intel.com/content/dam/www/public/us/en/documents/specification-updates/4th-gen-core-family-desktop-specification-update.pdf This patch tries to work around this issue by re-enabling PERF_GLOBAL_CTL in ->enable_box(). The workaround does not cover all cases. It helps for new events after returning from C7. But it cannot prevent C7, it will still miscount if a counter is already active. There is no drawback in leaving it enabled, so it does not need disable_box() here. Signed-off-by: Kan Liang Cc: Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Link: http://lkml.kernel.org/r/1470925874-59943-1-git-send-email-kan.liang@intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/uncore_snb.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/x86/events/intel/uncore_snb.c b/arch/x86/events/intel/uncore_snb.c index 97a69dbba649..9d35ec0cb8fc 100644 --- a/arch/x86/events/intel/uncore_snb.c +++ b/arch/x86/events/intel/uncore_snb.c @@ -100,6 +100,12 @@ static void snb_uncore_msr_init_box(struct intel_uncore_box *box) } } +static void snb_uncore_msr_enable_box(struct intel_uncore_box *box) +{ + wrmsrl(SNB_UNC_PERF_GLOBAL_CTL, + SNB_UNC_GLOBAL_CTL_EN | SNB_UNC_GLOBAL_CTL_CORE_ALL); +} + static void snb_uncore_msr_exit_box(struct intel_uncore_box *box) { if (box->pmu->pmu_idx == 0) @@ -127,6 +133,7 @@ static struct attribute_group snb_uncore_format_group = { static struct intel_uncore_ops snb_uncore_msr_ops = { .init_box = snb_uncore_msr_init_box, + .enable_box = snb_uncore_msr_enable_box, .exit_box = snb_uncore_msr_exit_box, .disable_event = snb_uncore_msr_disable_event, .enable_event = snb_uncore_msr_enable_event, @@ -192,6 +199,12 @@ static void skl_uncore_msr_init_box(struct intel_uncore_box *box) } } +static void skl_uncore_msr_enable_box(struct intel_uncore_box *box) +{ + wrmsrl(SKL_UNC_PERF_GLOBAL_CTL, + SNB_UNC_GLOBAL_CTL_EN | SKL_UNC_GLOBAL_CTL_CORE_ALL); +} + static void skl_uncore_msr_exit_box(struct intel_uncore_box *box) { if (box->pmu->pmu_idx == 0) @@ -200,6 +213,7 @@ static void skl_uncore_msr_exit_box(struct intel_uncore_box *box) static struct intel_uncore_ops skl_uncore_msr_ops = { .init_box = skl_uncore_msr_init_box, + .enable_box = skl_uncore_msr_enable_box, .exit_box = skl_uncore_msr_exit_box, .disable_event = snb_uncore_msr_disable_event, .enable_event = snb_uncore_msr_enable_event, From 75a4615c95a2de4fa592c1312fb8446e74b4e5eb Mon Sep 17 00:00:00 2001 From: Julius Niedworok Date: Wed, 3 Aug 2016 16:39:54 +0200 Subject: [PATCH 214/235] KVM: s390: set the prefix initially properly When KVM_RUN is triggered on a VCPU without an initial reset, a validity intercept occurs. Setting the prefix will set the KVM_REQ_MMU_RELOAD bit initially, thus preventing the bug. Reviewed-by: David Hildenbrand Acked-by: Cornelia Huck Signed-off-by: Julius Niedworok Signed-off-by: Christian Borntraeger --- arch/s390/kvm/kvm-s390.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 3f3ae4865d57..e63f6ed0a936 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -1672,6 +1672,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) KVM_SYNC_CRS | KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT; + kvm_s390_set_prefix(vcpu, 0); if (test_kvm_facility(vcpu->kvm, 64)) vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB; /* fprs can be synchronized via vrs, even if the guest has no vx. With From aca411a4b17a4aebe14ecdf253373db5b7ee6058 Mon Sep 17 00:00:00 2001 From: Julius Niedworok Date: Wed, 3 Aug 2016 16:39:55 +0200 Subject: [PATCH 215/235] KVM: s390: reset KVM_REQ_MMU_RELOAD if mapping the prefix failed When triggering KVM_RUN without a user memory region being mapped (KVM_SET_USER_MEMORY_REGION) a validity intercept occurs. This could happen, if the user memory region was not mapped initially or if it was unmapped after the vcpu is initialized. The function kvm_s390_handle_requests checks for the KVM_REQ_MMU_RELOAD bit. The check function always clears this bit. If gmap_mprotect_notify returns an error code, the mapping failed, but the KVM_REQ_MMU_RELOAD was not set anymore. So the next time kvm_s390_handle_requests is called, the execution would fall trough the check for KVM_REQ_MMU_RELOAD. The bit needs to be resetted, if gmap_mprotect_notify returns an error code. Resetting the bit with kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu) fixes the bug. Reviewed-by: David Hildenbrand Signed-off-by: Julius Niedworok Signed-off-by: Christian Borntraeger --- arch/s390/kvm/kvm-s390.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index e63f6ed0a936..f142215ed30d 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -2362,8 +2362,10 @@ static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) rc = gmap_mprotect_notify(vcpu->arch.gmap, kvm_s390_get_prefix(vcpu), PAGE_SIZE * 2, PROT_WRITE); - if (rc) + if (rc) { + kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); return rc; + } goto retry; } From 023e9fddc3616b005c3753fc1bb6526388cd7a30 Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Tue, 9 Aug 2016 19:13:00 +0200 Subject: [PATCH 216/235] KVM: PPC: Move xics_debugfs_init out of create MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As we are about to hold the kvm->lock during the create operation on KVM devices, we should move the call to xics_debugfs_init into its own function, since holding a mutex over extended amounts of time might not be a good idea. Introduce an init operation on the kvm_device_ops struct which cannot fail and call this, if configured, after the device has been created. Signed-off-by: Christoffer Dall Reviewed-by: Paolo Bonzini Signed-off-by: Radim Krčmář --- arch/powerpc/kvm/book3s_xics.c | 10 ++++++++-- include/linux/kvm_host.h | 6 ++++++ virt/kvm/kvm_main.c | 3 +++ 3 files changed, 17 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c index a75ba38a2d81..f2def8e45fef 100644 --- a/arch/powerpc/kvm/book3s_xics.c +++ b/arch/powerpc/kvm/book3s_xics.c @@ -1341,8 +1341,6 @@ static int kvmppc_xics_create(struct kvm_device *dev, u32 type) return ret; } - xics_debugfs_init(xics); - #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE if (cpu_has_feature(CPU_FTR_ARCH_206)) { /* Enable real mode support */ @@ -1354,9 +1352,17 @@ static int kvmppc_xics_create(struct kvm_device *dev, u32 type) return 0; } +static void kvmppc_xics_init(struct kvm_device *dev) +{ + struct kvmppc_xics *xics = (struct kvmppc_xics *)dev->private; + + xics_debugfs_init(xics); +} + struct kvm_device_ops kvm_xics_ops = { .name = "kvm-xics", .create = kvmppc_xics_create, + .init = kvmppc_xics_init, .destroy = kvmppc_xics_free, .set_attr = xics_set_attr, .get_attr = xics_get_attr, diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 01e908ac4a39..d3c9b82812c3 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -1115,6 +1115,12 @@ struct kvm_device_ops { const char *name; int (*create)(struct kvm_device *dev, u32 type); + /* + * init is called after create if create is successful and is called + * outside of holding kvm->lock. + */ + void (*init)(struct kvm_device *dev); + /* * Destroy is responsible for freeing dev. * diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index cc081ccfcaa3..ae642452e91a 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -2838,6 +2838,9 @@ static int kvm_ioctl_create_device(struct kvm *kvm, return ret; } + if (ops->init) + ops->init(dev); + ret = anon_inode_getfd(ops->name, &kvm_device_fops, dev, O_RDWR | O_CLOEXEC); if (ret < 0) { ops->destroy(dev); From a28ebea2adc4a2bef5989a5a181ec238f59fbcad Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Tue, 9 Aug 2016 19:13:01 +0200 Subject: [PATCH 217/235] KVM: Protect device ops->create and list_add with kvm->lock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit KVM devices were manipulating list data structures without any form of synchronization, and some implementations of the create operations also suffered from a lack of synchronization. Now when we've split the xics create operation into create and init, we can hold the kvm->lock mutex while calling the create operation and when manipulating the devices list. The error path in the generic code gets slightly ugly because we have to take the mutex again and delete the device from the list, but holding the mutex during anon_inode_getfd or releasing/locking the mutex in the common non-error path seemed wrong. Signed-off-by: Christoffer Dall Reviewed-by: Paolo Bonzini Acked-by: Christian Borntraeger Signed-off-by: Radim Krčmář --- arch/arm/kvm/arm.c | 6 +++++- arch/powerpc/kvm/book3s_xics.c | 2 -- include/linux/kvm_host.h | 6 ++++++ virt/kvm/arm/vgic/vgic-init.c | 17 ++++------------- virt/kvm/kvm_main.c | 13 ++++++++++++- 5 files changed, 27 insertions(+), 17 deletions(-) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index d94bb9093ead..75f130ef6504 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -1009,9 +1009,13 @@ long kvm_arch_vm_ioctl(struct file *filp, switch (ioctl) { case KVM_CREATE_IRQCHIP: { + int ret; if (!vgic_present) return -ENXIO; - return kvm_vgic_create(kvm, KVM_DEV_TYPE_ARM_VGIC_V2); + mutex_lock(&kvm->lock); + ret = kvm_vgic_create(kvm, KVM_DEV_TYPE_ARM_VGIC_V2); + mutex_unlock(&kvm->lock); + return ret; } case KVM_ARM_SET_DEVICE_ADDR: { struct kvm_arm_device_addr dev_addr; diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c index f2def8e45fef..05aa11399a78 100644 --- a/arch/powerpc/kvm/book3s_xics.c +++ b/arch/powerpc/kvm/book3s_xics.c @@ -1329,12 +1329,10 @@ static int kvmppc_xics_create(struct kvm_device *dev, u32 type) xics->kvm = kvm; /* Already there ? */ - mutex_lock(&kvm->lock); if (kvm->arch.xics) ret = -EEXIST; else kvm->arch.xics = xics; - mutex_unlock(&kvm->lock); if (ret) { kfree(xics); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index d3c9b82812c3..9c28b4d4c90b 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -1113,6 +1113,12 @@ struct kvm_device { /* create, destroy, and name are mandatory */ struct kvm_device_ops { const char *name; + + /* + * create is called holding kvm->lock and any operations not suitable + * to do while holding the lock should be deferred to init (see + * below). + */ int (*create)(struct kvm_device *dev, u32 type); /* diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c index fb4b0a79a950..83777c1cbae0 100644 --- a/virt/kvm/arm/vgic/vgic-init.c +++ b/virt/kvm/arm/vgic/vgic-init.c @@ -73,12 +73,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type) int i, vcpu_lock_idx = -1, ret; struct kvm_vcpu *vcpu; - mutex_lock(&kvm->lock); - - if (irqchip_in_kernel(kvm)) { - ret = -EEXIST; - goto out; - } + if (irqchip_in_kernel(kvm)) + return -EEXIST; /* * This function is also called by the KVM_CREATE_IRQCHIP handler, @@ -87,10 +83,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type) * the proper checks already. */ if (type == KVM_DEV_TYPE_ARM_VGIC_V2 && - !kvm_vgic_global_state.can_emulate_gicv2) { - ret = -ENODEV; - goto out; - } + !kvm_vgic_global_state.can_emulate_gicv2) + return -ENODEV; /* * Any time a vcpu is run, vcpu_load is called which tries to grab the @@ -138,9 +132,6 @@ int kvm_vgic_create(struct kvm *kvm, u32 type) vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx); mutex_unlock(&vcpu->mutex); } - -out: - mutex_unlock(&kvm->lock); return ret; } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index ae642452e91a..195078225aa5 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -696,6 +696,11 @@ static void kvm_destroy_devices(struct kvm *kvm) { struct kvm_device *dev, *tmp; + /* + * We do not need to take the kvm->lock here, because nobody else + * has a reference to the struct kvm at this point and therefore + * cannot access the devices list anyhow. + */ list_for_each_entry_safe(dev, tmp, &kvm->devices, vm_node) { list_del(&dev->vm_node); dev->ops->destroy(dev); @@ -2832,11 +2837,15 @@ static int kvm_ioctl_create_device(struct kvm *kvm, dev->ops = ops; dev->kvm = kvm; + mutex_lock(&kvm->lock); ret = ops->create(dev, cd->type); if (ret < 0) { + mutex_unlock(&kvm->lock); kfree(dev); return ret; } + list_add(&dev->vm_node, &kvm->devices); + mutex_unlock(&kvm->lock); if (ops->init) ops->init(dev); @@ -2844,10 +2853,12 @@ static int kvm_ioctl_create_device(struct kvm *kvm, ret = anon_inode_getfd(ops->name, &kvm_device_fops, dev, O_RDWR | O_CLOEXEC); if (ret < 0) { ops->destroy(dev); + mutex_lock(&kvm->lock); + list_del(&dev->vm_node); + mutex_unlock(&kvm->lock); return ret; } - list_add(&dev->vm_node, &kvm->devices); kvm_get_kvm(kvm); cd->fd = ret; return 0; From c604cffa93478f8888bec62b23d6073dad03d43a Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 11 Aug 2016 11:58:12 +0100 Subject: [PATCH 218/235] MIPS: KVM: Fix mapped fault broken commpage handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit kvm_mips_handle_mapped_seg_tlb_fault() appears to map the guest page at virtual address 0 to PFN 0 if the guest has created its own mapping there. The intention is unclear, but it may have been an attempt to protect the zero page from being mapped to anything but the comm page in code paths you wouldn't expect from genuine commpage accesses (guest kernel mode cache instructions on that address, hitting trapping instructions when executing from that address with a coincidental TLB eviction during the KVM handling, and guest user mode accesses to that address). Fix this to check for mappings exactly at KVM_GUEST_COMMPAGE_ADDR (it may not be at address 0 since commit 42aa12e74e91 ("MIPS: KVM: Move commpage so 0x0 is unmapped")), and set the corresponding EntryLo to be interpreted as 0 (invalid). Fixes: 858dd5d45733 ("KVM/MIPS32: MMU/TLB operations for the Guest.") Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: "Radim Krčmář" Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Cc: # 3.10.x- Signed-off-by: Radim Krčmář --- arch/mips/kvm/mmu.c | 45 ++++++++++++++++++++++++++------------------- 1 file changed, 26 insertions(+), 19 deletions(-) diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index 57319ee57c4f..96e0b24cfe5c 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -138,35 +138,42 @@ int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, unsigned long entryhi = 0, entrylo0 = 0, entrylo1 = 0; struct kvm *kvm = vcpu->kvm; kvm_pfn_t pfn0, pfn1; + long tlb_lo[2]; int ret; - if ((tlb->tlb_hi & VPN2_MASK) == 0) { - pfn0 = 0; - pfn1 = 0; - } else { - if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb->tlb_lo[0]) - >> PAGE_SHIFT) < 0) - return -1; + tlb_lo[0] = tlb->tlb_lo[0]; + tlb_lo[1] = tlb->tlb_lo[1]; - if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb->tlb_lo[1]) - >> PAGE_SHIFT) < 0) - return -1; + /* + * The commpage address must not be mapped to anything else if the guest + * TLB contains entries nearby, or commpage accesses will break. + */ + if (!((tlb->tlb_hi ^ KVM_GUEST_COMMPAGE_ADDR) & + VPN2_MASK & (PAGE_MASK << 1))) + tlb_lo[(KVM_GUEST_COMMPAGE_ADDR >> PAGE_SHIFT) & 1] = 0; - pfn0 = kvm->arch.guest_pmap[ - mips3_tlbpfn_to_paddr(tlb->tlb_lo[0]) >> PAGE_SHIFT]; - pfn1 = kvm->arch.guest_pmap[ - mips3_tlbpfn_to_paddr(tlb->tlb_lo[1]) >> PAGE_SHIFT]; - } + if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb_lo[0]) + >> PAGE_SHIFT) < 0) + return -1; + + if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb_lo[1]) + >> PAGE_SHIFT) < 0) + return -1; + + pfn0 = kvm->arch.guest_pmap[ + mips3_tlbpfn_to_paddr(tlb_lo[0]) >> PAGE_SHIFT]; + pfn1 = kvm->arch.guest_pmap[ + mips3_tlbpfn_to_paddr(tlb_lo[1]) >> PAGE_SHIFT]; /* Get attributes from the Guest TLB */ entrylo0 = mips3_paddr_to_tlbpfn(pfn0 << PAGE_SHIFT) | ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - (tlb->tlb_lo[0] & ENTRYLO_D) | - (tlb->tlb_lo[0] & ENTRYLO_V); + (tlb_lo[0] & ENTRYLO_D) | + (tlb_lo[0] & ENTRYLO_V); entrylo1 = mips3_paddr_to_tlbpfn(pfn1 << PAGE_SHIFT) | ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - (tlb->tlb_lo[1] & ENTRYLO_D) | - (tlb->tlb_lo[1] & ENTRYLO_V); + (tlb_lo[1] & ENTRYLO_D) | + (tlb_lo[1] & ENTRYLO_V); kvm_debug("@ %#lx tlb_lo0: 0x%08lx tlb_lo1: 0x%08lx\n", vcpu->arch.pc, tlb->tlb_lo[0], tlb->tlb_lo[1]); From 8985d50382359e5bf118fdbefc859d0dbf6cebc7 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 11 Aug 2016 11:58:13 +0100 Subject: [PATCH 219/235] MIPS: KVM: Add missing gfn range check MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit kvm_mips_handle_mapped_seg_tlb_fault() calculates the guest frame number based on the guest TLB EntryLo values, however it is not range checked to ensure it lies within the guest_pmap. If the physical memory the guest refers to is out of range then dump the guest TLB and emit an internal error. Fixes: 858dd5d45733 ("KVM/MIPS32: MMU/TLB operations for the Guest.") Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: "Radim Krčmář" Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Cc: # 3.10.x- Signed-off-by: Radim Krčmář --- arch/mips/kvm/mmu.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index 96e0b24cfe5c..701c1b64cef8 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -138,6 +138,7 @@ int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, unsigned long entryhi = 0, entrylo0 = 0, entrylo1 = 0; struct kvm *kvm = vcpu->kvm; kvm_pfn_t pfn0, pfn1; + gfn_t gfn0, gfn1; long tlb_lo[2]; int ret; @@ -152,18 +153,24 @@ int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, VPN2_MASK & (PAGE_MASK << 1))) tlb_lo[(KVM_GUEST_COMMPAGE_ADDR >> PAGE_SHIFT) & 1] = 0; - if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb_lo[0]) - >> PAGE_SHIFT) < 0) + gfn0 = mips3_tlbpfn_to_paddr(tlb_lo[0]) >> PAGE_SHIFT; + gfn1 = mips3_tlbpfn_to_paddr(tlb_lo[1]) >> PAGE_SHIFT; + if (gfn0 >= kvm->arch.guest_pmap_npages || + gfn1 >= kvm->arch.guest_pmap_npages) { + kvm_err("%s: Invalid gfn: [%#llx, %#llx], EHi: %#lx\n", + __func__, gfn0, gfn1, tlb->tlb_hi); + kvm_mips_dump_guest_tlbs(vcpu); + return -1; + } + + if (kvm_mips_map_page(kvm, gfn0) < 0) return -1; - if (kvm_mips_map_page(kvm, mips3_tlbpfn_to_paddr(tlb_lo[1]) - >> PAGE_SHIFT) < 0) + if (kvm_mips_map_page(kvm, gfn1) < 0) return -1; - pfn0 = kvm->arch.guest_pmap[ - mips3_tlbpfn_to_paddr(tlb_lo[0]) >> PAGE_SHIFT]; - pfn1 = kvm->arch.guest_pmap[ - mips3_tlbpfn_to_paddr(tlb_lo[1]) >> PAGE_SHIFT]; + pfn0 = kvm->arch.guest_pmap[gfn0]; + pfn1 = kvm->arch.guest_pmap[gfn1]; /* Get attributes from the Guest TLB */ entrylo0 = mips3_paddr_to_tlbpfn(pfn0 << PAGE_SHIFT) | From 0741f52d1b980dbeb290afe67d88fc2928edd8ab Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 11 Aug 2016 11:58:14 +0100 Subject: [PATCH 220/235] MIPS: KVM: Fix gfn range check in kseg0 tlb faults MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Two consecutive gfns are loaded into host TLB, so ensure the range check isn't off by one if guest_pmap_npages is odd. Fixes: 858dd5d45733 ("KVM/MIPS32: MMU/TLB operations for the Guest.") Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: "Radim Krčmář" Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Cc: # 3.10.x- Signed-off-by: Radim Krčmář --- arch/mips/kvm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index 701c1b64cef8..6a8a21859502 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -99,7 +99,7 @@ int kvm_mips_handle_kseg0_tlb_fault(unsigned long badvaddr, } gfn = (KVM_GUEST_CPHYSADDR(badvaddr) >> PAGE_SHIFT); - if (gfn >= kvm->arch.guest_pmap_npages) { + if ((gfn | 1) >= kvm->arch.guest_pmap_npages) { kvm_err("%s: Invalid gfn: %#llx, BadVaddr: %#lx\n", __func__, gfn, badvaddr); kvm_mips_dump_host_tlbs(); From 9b731bcfdec4c159ad2e4312e25d69221709b96a Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 11 Aug 2016 11:58:15 +0100 Subject: [PATCH 221/235] MIPS: KVM: Propagate kseg0/mapped tlb fault errors MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Propagate errors from kvm_mips_handle_kseg0_tlb_fault() and kvm_mips_handle_mapped_seg_tlb_fault(), usually triggering an internal error since they normally indicate the guest accessed bad physical memory or the commpage in an unexpected way. Fixes: 858dd5d45733 ("KVM/MIPS32: MMU/TLB operations for the Guest.") Fixes: e685c689f3a8 ("KVM/MIPS32: Privileged instruction/target branch emulation.") Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: "Radim Krčmář" Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Cc: # 3.10.x- Signed-off-by: Radim Krčmář --- arch/mips/kvm/emulate.c | 35 ++++++++++++++++++++++++++--------- arch/mips/kvm/mmu.c | 12 +++++++++--- 2 files changed, 35 insertions(+), 12 deletions(-) diff --git a/arch/mips/kvm/emulate.c b/arch/mips/kvm/emulate.c index 6eb52b9c9818..e788515f766b 100644 --- a/arch/mips/kvm/emulate.c +++ b/arch/mips/kvm/emulate.c @@ -1642,8 +1642,14 @@ enum emulation_result kvm_mips_emulate_cache(union mips_instruction inst, preempt_disable(); if (KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG0) { - if (kvm_mips_host_tlb_lookup(vcpu, va) < 0) - kvm_mips_handle_kseg0_tlb_fault(va, vcpu); + if (kvm_mips_host_tlb_lookup(vcpu, va) < 0 && + kvm_mips_handle_kseg0_tlb_fault(va, vcpu)) { + kvm_err("%s: handling mapped kseg0 tlb fault for %lx, vcpu: %p, ASID: %#lx\n", + __func__, va, vcpu, read_c0_entryhi()); + er = EMULATE_FAIL; + preempt_enable(); + goto done; + } } else if ((KVM_GUEST_KSEGX(va) < KVM_GUEST_KSEG0) || KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG23) { int index; @@ -1680,12 +1686,18 @@ enum emulation_result kvm_mips_emulate_cache(union mips_instruction inst, run, vcpu); preempt_enable(); goto dont_update_pc; - } else { - /* - * We fault an entry from the guest tlb to the - * shadow host TLB - */ - kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb); + } + /* + * We fault an entry from the guest tlb to the + * shadow host TLB + */ + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb)) { + kvm_err("%s: handling mapped seg tlb fault for %lx, index: %u, vcpu: %p, ASID: %#lx\n", + __func__, va, index, vcpu, + read_c0_entryhi()); + er = EMULATE_FAIL; + preempt_enable(); + goto done; } } } else { @@ -2659,7 +2671,12 @@ enum emulation_result kvm_mips_handle_tlbmiss(u32 cause, * OK we have a Guest TLB entry, now inject it into the * shadow host TLB */ - kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb); + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb)) { + kvm_err("%s: handling mapped seg tlb fault for %lx, index: %u, vcpu: %p, ASID: %#lx\n", + __func__, va, index, vcpu, + read_c0_entryhi()); + er = EMULATE_FAIL; + } } } diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index 6a8a21859502..6cfdcf55572d 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -368,9 +368,15 @@ u32 kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu) local_irq_restore(flags); return KVM_INVALID_INST; } - kvm_mips_handle_mapped_seg_tlb_fault(vcpu, - &vcpu->arch. - guest_tlb[index]); + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, + &vcpu->arch.guest_tlb[index])) { + kvm_err("%s: handling mapped seg tlb fault failed for %p, index: %u, vcpu: %p, ASID: %#lx\n", + __func__, opc, index, vcpu, + read_c0_entryhi()); + kvm_mips_dump_guest_tlbs(vcpu); + local_irq_restore(flags); + return KVM_INVALID_INST; + } inst = *(opc); } local_irq_restore(flags); From 9adeb8e72dbfe976709df01e259ed556ee60e779 Mon Sep 17 00:00:00 2001 From: Laura Abbott Date: Tue, 9 Aug 2016 18:25:26 -0700 Subject: [PATCH 222/235] arm64: Handle el1 synchronous instruction aborts cleanly Executing from a non-executable area gives an ugly message: lkdtm: Performing direct entry EXEC_RODATA lkdtm: attempting ok execution at ffff0000084c0e08 lkdtm: attempting bad execution at ffff000008880700 Bad mode in Synchronous Abort handler detected on CPU2, code 0x8400000e -- IABT (current EL) CPU: 2 PID: 998 Comm: sh Not tainted 4.7.0-rc2+ #13 Hardware name: linux,dummy-virt (DT) task: ffff800077e35780 ti: ffff800077970000 task.ti: ffff800077970000 PC is at lkdtm_rodata_do_nothing+0x0/0x8 LR is at execute_location+0x74/0x88 The 'IABT (current EL)' indicates the error but it's a bit cryptic without knowledge of the ARM ARM. There is also no indication of the specific address which triggered the fault. The increase in kernel page permissions makes hitting this case more likely as well. Handling the case in the vectors gives a much more familiar looking error message: lkdtm: Performing direct entry EXEC_RODATA lkdtm: attempting ok execution at ffff0000084c0840 lkdtm: attempting bad execution at ffff000008880680 Unable to handle kernel paging request at virtual address ffff000008880680 pgd = ffff8000089b2000 [ffff000008880680] *pgd=00000000489b4003, *pud=0000000048904003, *pmd=0000000000000000 Internal error: Oops: 8400000e [#1] PREEMPT SMP Modules linked in: CPU: 1 PID: 997 Comm: sh Not tainted 4.7.0-rc1+ #24 Hardware name: linux,dummy-virt (DT) task: ffff800077f9f080 ti: ffff800008a1c000 task.ti: ffff800008a1c000 PC is at lkdtm_rodata_do_nothing+0x0/0x8 LR is at execute_location+0x74/0x88 Acked-by: Mark Rutland Signed-off-by: Laura Abbott Signed-off-by: Catalin Marinas --- arch/arm64/kernel/entry.S | 7 +++++++ arch/arm64/mm/fault.c | 14 ++++++++++++-- 2 files changed, 19 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index 96e4a2b64cc1..441420ca7d08 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -353,6 +353,8 @@ el1_sync: lsr x24, x1, #ESR_ELx_EC_SHIFT // exception class cmp x24, #ESR_ELx_EC_DABT_CUR // data abort in EL1 b.eq el1_da + cmp x24, #ESR_ELx_EC_IABT_CUR // instruction abort in EL1 + b.eq el1_ia cmp x24, #ESR_ELx_EC_SYS64 // configurable trap b.eq el1_undef cmp x24, #ESR_ELx_EC_SP_ALIGN // stack alignment exception @@ -364,6 +366,11 @@ el1_sync: cmp x24, #ESR_ELx_EC_BREAKPT_CUR // debug exception in EL1 b.ge el1_dbg b el1_inv + +el1_ia: + /* + * Fall through to the Data abort case + */ el1_da: /* * Data abort handling diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index c8beaa0da7df..05d2bd776c69 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -153,6 +153,11 @@ int ptep_set_access_flags(struct vm_area_struct *vma, } #endif +static bool is_el1_instruction_abort(unsigned int esr) +{ + return ESR_ELx_EC(esr) == ESR_ELx_EC_IABT_CUR; +} + /* * The kernel tried to access some page that wasn't present. */ @@ -161,8 +166,9 @@ static void __do_kernel_fault(struct mm_struct *mm, unsigned long addr, { /* * Are we prepared to handle this kernel fault? + * We are almost certainly not prepared to handle instruction faults. */ - if (fixup_exception(regs)) + if (!is_el1_instruction_abort(esr) && fixup_exception(regs)) return; /* @@ -267,7 +273,8 @@ static inline bool is_permission_fault(unsigned int esr) unsigned int ec = ESR_ELx_EC(esr); unsigned int fsc_type = esr & ESR_ELx_FSC_TYPE; - return (ec == ESR_ELx_EC_DABT_CUR && fsc_type == ESR_ELx_FSC_PERM); + return (ec == ESR_ELx_EC_DABT_CUR && fsc_type == ESR_ELx_FSC_PERM) || + (ec == ESR_ELx_EC_IABT_CUR && fsc_type == ESR_ELx_FSC_PERM); } static bool is_el0_instruction_abort(unsigned int esr) @@ -312,6 +319,9 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr, if (regs->orig_addr_limit == KERNEL_DS) die("Accessing user space memory with fs=KERNEL_DS", regs, esr); + if (is_el1_instruction_abort(esr)) + die("Attempting to execute userspace memory", regs, esr); + if (!search_exception_tables(regs->pc)) die("Accessing user space memory outside uaccess.h routines", regs, esr); } From 0194e760f7d2f42adb5e1db31b27a4331dd89c2f Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 11 Aug 2016 14:11:05 +0100 Subject: [PATCH 223/235] arm64: hibernate: avoid potential TLB conflict In create_safe_exec_page we install a set of global mappings in TTBR0, then subsequently invalidate TLBs. While TTBR0 points at the zero page, and the TLBs should be free of stale global entries, we may have stale ASID-tagged entries (e.g. from the EFI runtime services mappings) for the same VAs. Per the ARM ARM these ASID-tagged entries may conflict with newly-allocated global entries, and we must follow a Break-Before-Make approach to avoid issues resulting from this. This patch reworks create_safe_exec_page to invalidate TLBs while the zero page is still in place, ensuring that there are no potential conflicts when the new TTBR0 value is installed. As a single CPU is online while this code executes, we do not need to perform broadcast TLB maintenance, and can call local_flush_tlb_all(), which also subsumes some barriers. The remaining assembly is converted to use write_sysreg() and isb(). Other than this, we safely manipulate TTBRs in the hibernate dance. The code we install as part of the new TTBR0 mapping (the hibernated kernel's swsusp_arch_suspend_exit) installs a zero page into TTBR1, invalidates TLBs, then installs its preferred value. Upon being restored to the middle of swsusp_arch_suspend, the new image will call __cpu_suspend_exit, which will call cpu_uninstall_idmap, installing the zero page in TTBR0 and invalidating all TLB entries. Fixes: 82869ac57b5d ("arm64: kernel: Add support for hibernate/suspend-to-disk") Signed-off-by: Mark Rutland Acked-by: James Morse Tested-by: James Morse Cc: Lorenzo Pieralisi Cc: Will Deacon Cc: # 4.7+ Signed-off-by: Catalin Marinas --- arch/arm64/kernel/hibernate.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/arch/arm64/kernel/hibernate.c b/arch/arm64/kernel/hibernate.c index 21ab5df9fa76..b2e7de8726e8 100644 --- a/arch/arm64/kernel/hibernate.c +++ b/arch/arm64/kernel/hibernate.c @@ -35,6 +35,7 @@ #include #include #include +#include #include /* @@ -217,12 +218,22 @@ static int create_safe_exec_page(void *src_start, size_t length, set_pte(pte, __pte(virt_to_phys((void *)dst) | pgprot_val(PAGE_KERNEL_EXEC))); - /* Load our new page tables */ - asm volatile("msr ttbr0_el1, %0;" - "isb;" - "tlbi vmalle1is;" - "dsb ish;" - "isb" : : "r"(virt_to_phys(pgd))); + /* + * Load our new page tables. A strict BBM approach requires that we + * ensure that TLBs are free of any entries that may overlap with the + * global mappings we are about to install. + * + * For a real hibernate/resume cycle TTBR0 currently points to a zero + * page, but TLBs may contain stale ASID-tagged entries (e.g. for EFI + * runtime services), while for a userspace-driven test_resume cycle it + * points to userspace page tables (and we must point it at a zero page + * ourselves). Elsewhere we only (un)install the idmap with preemption + * disabled, so T0SZ should be as required regardless. + */ + cpu_set_reserved_ttbr0(); + local_flush_tlb_all(); + write_sysreg(virt_to_phys(pgd), ttbr0_el1); + isb(); *phys_dst_addr = virt_to_phys((void *)dst); From dfbca61af0b654990b9af8297ac574a9986d8275 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 11 Aug 2016 14:11:06 +0100 Subject: [PATCH 224/235] arm64: hibernate: handle allocation failures In create_safe_exec_page(), we create a copy of the hibernate exit text, along with some page tables to map this via TTBR0. We then install the new tables in TTBR0. In swsusp_arch_resume() we call create_safe_exec_page() before trying a number of operations which may fail (e.g. copying the linear map page tables). If these fail, we bail out of swsusp_arch_resume() and return an error code, but leave TTBR0 as-is. Subsequently, the core hibernate code will call free_basic_memory_bitmaps(), which will free all of the memory allocations we made, including the page tables installed in TTBR0. Thus, we may have TTBR0 pointing at dangling freed memory for some period of time. If the hibernate attempt was triggered by a user requesting a hibernate test via the reboot syscall, we may return to userspace with the clobbered TTBR0 value. Avoid these issues by reorganising swsusp_arch_resume() such that we have no failure paths after create_safe_exec_page(). We also add a check that the zero page allocation succeeded, matching what we have for other allocations. Fixes: 82869ac57b5d ("arm64: kernel: Add support for hibernate/suspend-to-disk") Signed-off-by: Mark Rutland Acked-by: James Morse Cc: Lorenzo Pieralisi Cc: Will Deacon Cc: # 4.7+ Signed-off-by: Catalin Marinas --- arch/arm64/kernel/hibernate.c | 59 +++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 27 deletions(-) diff --git a/arch/arm64/kernel/hibernate.c b/arch/arm64/kernel/hibernate.c index b2e7de8726e8..65d81f965e74 100644 --- a/arch/arm64/kernel/hibernate.c +++ b/arch/arm64/kernel/hibernate.c @@ -404,6 +404,38 @@ int swsusp_arch_resume(void) void __noreturn (*hibernate_exit)(phys_addr_t, phys_addr_t, void *, void *, phys_addr_t, phys_addr_t); + /* + * Restoring the memory image will overwrite the ttbr1 page tables. + * Create a second copy of just the linear map, and use this when + * restoring. + */ + tmp_pg_dir = (pgd_t *)get_safe_page(GFP_ATOMIC); + if (!tmp_pg_dir) { + pr_err("Failed to allocate memory for temporary page tables."); + rc = -ENOMEM; + goto out; + } + rc = copy_page_tables(tmp_pg_dir, PAGE_OFFSET, 0); + if (rc) + goto out; + + /* + * Since we only copied the linear map, we need to find restore_pblist's + * linear map address. + */ + lm_restore_pblist = LMADDR(restore_pblist); + + /* + * We need a zero page that is zero before & after resume in order to + * to break before make on the ttbr1 page tables. + */ + zero_page = (void *)get_safe_page(GFP_ATOMIC); + if (!zero_page) { + pr_err("Failed to allocate zero page."); + rc = -ENOMEM; + goto out; + } + /* * Locate the exit code in the bottom-but-one page, so that *NULL * still has disastrous affects. @@ -429,27 +461,6 @@ int swsusp_arch_resume(void) */ __flush_dcache_area(hibernate_exit, exit_size); - /* - * Restoring the memory image will overwrite the ttbr1 page tables. - * Create a second copy of just the linear map, and use this when - * restoring. - */ - tmp_pg_dir = (pgd_t *)get_safe_page(GFP_ATOMIC); - if (!tmp_pg_dir) { - pr_err("Failed to allocate memory for temporary page tables."); - rc = -ENOMEM; - goto out; - } - rc = copy_page_tables(tmp_pg_dir, PAGE_OFFSET, 0); - if (rc) - goto out; - - /* - * Since we only copied the linear map, we need to find restore_pblist's - * linear map address. - */ - lm_restore_pblist = LMADDR(restore_pblist); - /* * KASLR will cause the el2 vectors to be in a different location in * the resumed kernel. Load hibernate's temporary copy into el2. @@ -464,12 +475,6 @@ int swsusp_arch_resume(void) __hyp_set_vectors(el2_vectors); } - /* - * We need a zero page that is zero before & after resume in order to - * to break before make on the ttbr1 page tables. - */ - zero_page = (void *)get_safe_page(GFP_ATOMIC); - hibernate_exit(virt_to_phys(tmp_pg_dir), resume_hdr.ttbr1_el1, resume_hdr.reenter_kernel, lm_restore_pblist, resume_hdr.__hyp_stub_vectors, virt_to_phys(zero_page)); From 2323439fd0c844b0e84d16968699c419809d480e Mon Sep 17 00:00:00 2001 From: Riku Voipio Date: Wed, 15 Jun 2016 11:27:59 +0300 Subject: [PATCH 225/235] arm64: defconfig: add options for virtualization and containers Enable options commonly needed by popular virtualization and container applications. Use modules when possible to avoid too much overhead for users not interested. - add namespace and cgroup options needed - add seccomp - optional, but enhances Qemu etc - bridge, nat, veth, macvtap and multicast for routing guests and containers - btfrs and overlayfs modules for container COW backends - while near it, make fuse a module instead of built-in. Generated with make saveconfig and dropping unrelated spurious change hunks while commiting. bloat-o-meter old-vmlinux vmlinux: add/remove: 905/390 grow/shrink: 767/229 up/down: 183513/-94861 (88652) .... Total: Before=10515408, After=10604060, chg +0.84% Signed-off-by: Riku Voipio Signed-off-by: Catalin Marinas --- arch/arm64/configs/defconfig | 52 +++++++++++++++++++++++++++++++----- 1 file changed, 46 insertions(+), 6 deletions(-) diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index 0555b7caaf2c..2016444601f8 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -15,10 +15,14 @@ CONFIG_IKCONFIG_PROC=y CONFIG_LOG_BUF_SHIFT=14 CONFIG_MEMCG=y CONFIG_MEMCG_SWAP=y +CONFIG_BLK_CGROUP=y +CONFIG_CGROUP_PIDS=y CONFIG_CGROUP_HUGETLB=y -# CONFIG_UTS_NS is not set -# CONFIG_IPC_NS is not set -# CONFIG_NET_NS is not set +CONFIG_CPUSETS=y +CONFIG_CGROUP_DEVICE=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_CGROUP_PERF=y +CONFIG_USER_NS=y CONFIG_SCHED_AUTOGROUP=y CONFIG_BLK_DEV_INITRD=y CONFIG_KALLSYMS_ALL=y @@ -71,6 +75,7 @@ CONFIG_PREEMPT=y CONFIG_KSM=y CONFIG_TRANSPARENT_HUGEPAGE=y CONFIG_CMA=y +CONFIG_SECCOMP=y CONFIG_XEN=y CONFIG_KEXEC=y # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set @@ -84,10 +89,37 @@ CONFIG_NET=y CONFIG_PACKET=y CONFIG_UNIX=y CONFIG_INET=y +CONFIG_IP_MULTICAST=y CONFIG_IP_PNP=y CONFIG_IP_PNP_DHCP=y CONFIG_IP_PNP_BOOTP=y -# CONFIG_IPV6 is not set +CONFIG_IPV6=m +CONFIG_NETFILTER=y +CONFIG_NF_CONNTRACK=m +CONFIG_NF_CONNTRACK_EVENTS=y +CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m +CONFIG_NETFILTER_XT_TARGET_LOG=m +CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m +CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m +CONFIG_NF_CONNTRACK_IPV4=m +CONFIG_IP_NF_IPTABLES=m +CONFIG_IP_NF_FILTER=m +CONFIG_IP_NF_TARGET_REJECT=m +CONFIG_IP_NF_NAT=m +CONFIG_IP_NF_TARGET_MASQUERADE=m +CONFIG_IP_NF_MANGLE=m +CONFIG_NF_CONNTRACK_IPV6=m +CONFIG_IP6_NF_IPTABLES=m +CONFIG_IP6_NF_FILTER=m +CONFIG_IP6_NF_TARGET_REJECT=m +CONFIG_IP6_NF_MANGLE=m +CONFIG_IP6_NF_NAT=m +CONFIG_IP6_NF_TARGET_MASQUERADE=m +CONFIG_BRIDGE=m +CONFIG_BRIDGE_VLAN_FILTERING=y +CONFIG_VLAN_8021Q=m +CONFIG_VLAN_8021Q_GVRP=y +CONFIG_VLAN_8021Q_MVRP=y CONFIG_BPF_JIT=y CONFIG_CFG80211=m CONFIG_MAC80211=m @@ -103,6 +135,7 @@ CONFIG_MTD=y CONFIG_MTD_M25P80=y CONFIG_MTD_SPI_NOR=y CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_NBD=m CONFIG_VIRTIO_BLK=y CONFIG_SRAM=y # CONFIG_SCSI_PROC_FS is not set @@ -120,7 +153,10 @@ CONFIG_SATA_SIL24=y CONFIG_PATA_PLATFORM=y CONFIG_PATA_OF_PLATFORM=y CONFIG_NETDEVICES=y +CONFIG_MACVLAN=m +CONFIG_MACVTAP=m CONFIG_TUN=y +CONFIG_VETH=m CONFIG_VIRTIO_NET=y CONFIG_AMD_XGBE=y CONFIG_NET_XGENE=y @@ -350,12 +386,16 @@ CONFIG_EXYNOS_ADC=y CONFIG_PWM_SAMSUNG=y CONFIG_EXT2_FS=y CONFIG_EXT3_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_BTRFS_FS=m +CONFIG_BTRFS_FS_POSIX_ACL=y CONFIG_FANOTIFY=y CONFIG_FANOTIFY_ACCESS_PERMISSIONS=y CONFIG_QUOTA=y CONFIG_AUTOFS4_FS=y -CONFIG_FUSE_FS=y -CONFIG_CUSE=y +CONFIG_FUSE_FS=m +CONFIG_CUSE=m +CONFIG_OVERLAY_FS=m CONFIG_VFAT_FS=y CONFIG_TMPFS=y CONFIG_HUGETLBFS=y From 53fb45d3df6fb64eff6c314b3fa2e279a2496e5b Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Wed, 8 Jun 2016 21:26:23 +0900 Subject: [PATCH 226/235] arm64: defconfig: enable CONFIG_LOCALVERSION_AUTO When CONFIG_LOCALVERSION_AUTO is disabled, the version string is just a tag name (or with a '+' appended if HEAD is not a tagged commit). During the development (and especially when git-bisecting), longer version string would be helpful to identify the commit we are running. This is a default y option, so drop the unset to enable it. Signed-off-by: Masahiro Yamada Signed-off-by: Catalin Marinas --- arch/arm64/configs/defconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index 2016444601f8..eadf4855ad2d 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -1,4 +1,3 @@ -# CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_POSIX_MQUEUE=y CONFIG_AUDIT=y From dd257933fa4b9fea66a1195f8a15111029810abc Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 11 Aug 2016 10:37:39 -0400 Subject: [PATCH 227/235] nfsd: don't return an unhashed lock stateid after taking mutex nfsd4_lock will take the st_mutex before working with the stateid it gets, but between the time when we drop the cl_lock and take the mutex, the stateid could become unhashed (a'la FREE_STATEID). If that happens the lock stateid returned to the client will be forgotten. Fix this by first moving the st_mutex acquisition into lookup_or_create_lock_state. Then, have it check to see if the lock stateid is still hashed after taking the mutex. If it's not, then put the stateid and try the find/create again. Signed-off-by: Jeff Layton Tested-by: Alexey Kodanev Cc: stable@vger.kernel.org # feb9dad5 nfsd: Always lock state exclusively. Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4state.c | 25 ++++++++++++++++++++----- 1 file changed, 20 insertions(+), 5 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 0edc1822f144..a204d7e109d4 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -5523,7 +5523,7 @@ static __be32 lookup_or_create_lock_state(struct nfsd4_compound_state *cstate, struct nfs4_ol_stateid *ost, struct nfsd4_lock *lock, - struct nfs4_ol_stateid **lst, bool *new) + struct nfs4_ol_stateid **plst, bool *new) { __be32 status; struct nfs4_file *fi = ost->st_stid.sc_file; @@ -5531,7 +5531,9 @@ lookup_or_create_lock_state(struct nfsd4_compound_state *cstate, struct nfs4_client *cl = oo->oo_owner.so_client; struct inode *inode = d_inode(cstate->current_fh.fh_dentry); struct nfs4_lockowner *lo; + struct nfs4_ol_stateid *lst; unsigned int strhashval; + bool hashed; lo = find_lockowner_str(cl, &lock->lk_new_owner); if (!lo) { @@ -5547,12 +5549,27 @@ lookup_or_create_lock_state(struct nfsd4_compound_state *cstate, goto out; } - *lst = find_or_create_lock_stateid(lo, fi, inode, ost, new); - if (*lst == NULL) { +retry: + lst = find_or_create_lock_stateid(lo, fi, inode, ost, new); + if (lst == NULL) { status = nfserr_jukebox; goto out; } + + mutex_lock(&lst->st_mutex); + + /* See if it's still hashed to avoid race with FREE_STATEID */ + spin_lock(&cl->cl_lock); + hashed = !list_empty(&lst->st_perfile); + spin_unlock(&cl->cl_lock); + + if (!hashed) { + mutex_unlock(&lst->st_mutex); + nfs4_put_stid(&lst->st_stid); + goto retry; + } status = nfs_ok; + *plst = lst; out: nfs4_put_stateowner(&lo->lo_owner); return status; @@ -5619,8 +5636,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; status = lookup_or_create_lock_state(cstate, open_stp, lock, &lock_stp, &new); - if (status == nfs_ok) - mutex_lock(&lock_stp->st_mutex); } else { status = nfs4_preprocess_seqid_op(cstate, lock->lk_old_lock_seqid, From 62822e2ec4ad091ba31f823f577ef80db52e3c2c Mon Sep 17 00:00:00 2001 From: Thomas Garnier Date: Thu, 11 Aug 2016 14:49:29 -0700 Subject: [PATCH 228/235] PM / hibernate: Restore processor state before using per-CPU variables Restore the processor state before calling any other functions to ensure per-CPU variables can be used with KASLR memory randomization. Tracing functions use per-CPU variables (GS based on x86) and one was called just before restoring the processor state fully. It resulted in a double fault when both the tracing & the exception handler functions tried to use a per-CPU variable. Fixes: bb3632c6101b (PM / sleep: trace events for suspend/resume) Reported-and-tested-by: Borislav Petkov Reported-by: Jiri Kosina Tested-by: Rafael J. Wysocki Tested-by: Jiri Kosina Signed-off-by: Thomas Garnier Acked-by: Pavel Machek Signed-off-by: Rafael J. Wysocki --- kernel/power/hibernate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 0ee1df0a0bd6..61761aa7cc19 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -300,12 +300,12 @@ static int create_image(int platform_mode) save_processor_state(); trace_suspend_resume(TPS("machine_suspend"), PM_EVENT_HIBERNATE, true); error = swsusp_arch_suspend(); + /* Restore control flow magically appears here */ + restore_processor_state(); trace_suspend_resume(TPS("machine_suspend"), PM_EVENT_HIBERNATE, false); if (error) printk(KERN_ERR "PM: Error %d creating hibernation image\n", error); - /* Restore control flow magically appears here */ - restore_processor_state(); if (!in_suspend) events_check_enabled = false; From 783011b13095430262333fd64e17666954064664 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Mon, 21 Mar 2016 04:20:53 -0700 Subject: [PATCH 229/235] unicore32: mm: Add missing parameter to arch_vma_access_permitted MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit unicore32 fails to compile with the following errors. mm/memory.c: In function ‘__handle_mm_fault’: mm/memory.c:3381: error: too many arguments to function ‘arch_vma_access_permitted’ mm/gup.c: In function ‘check_vma_flags’: mm/gup.c:456: error: too many arguments to function ‘arch_vma_access_permitted’ mm/gup.c: In function ‘vma_permits_fault’: mm/gup.c:640: error: too many arguments to function ‘arch_vma_access_permitted’ Fixes: d61172b4b695b ("mm/core, x86/mm/pkeys: Differentiate instruction fetches") Cc: Dave Hansen Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Guenter Roeck Acked-by: Guan Xuetao --- arch/unicore32/include/asm/mmu_context.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/unicore32/include/asm/mmu_context.h b/arch/unicore32/include/asm/mmu_context.h index e35632ef23c7..62dfc644c908 100644 --- a/arch/unicore32/include/asm/mmu_context.h +++ b/arch/unicore32/include/asm/mmu_context.h @@ -98,7 +98,7 @@ static inline void arch_bprm_mm_init(struct mm_struct *mm, } static inline bool arch_vma_access_permitted(struct vm_area_struct *vma, - bool write, bool foreign) + bool write, bool execute, bool foreign) { /* by default, allow everything */ return true; From 2b05980d895e51fc46d3d27c8d1bcc696dddca99 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Wed, 8 Jun 2016 20:11:58 -0700 Subject: [PATCH 230/235] h8300: Add missing include file to asm/io.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit h8300 builds fail with arch/h8300/include/asm/io.h:9:15: error: unknown type name ‘u8’ arch/h8300/include/asm/io.h:15:15: error: unknown type name ‘u16’ arch/h8300/include/asm/io.h:21:15: error: unknown type name ‘u32’ and many related errors. Fixes: 23c82d41bdf4 ("kexec-allow-architectures-to-override-boot-mapping-fix") Cc: Andrew Morton Signed-off-by: Guenter Roeck --- arch/h8300/include/asm/io.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/h8300/include/asm/io.h b/arch/h8300/include/asm/io.h index 2e221c5f0203..f86918aed9e1 100644 --- a/arch/h8300/include/asm/io.h +++ b/arch/h8300/include/asm/io.h @@ -3,6 +3,8 @@ #ifdef __KERNEL__ +#include + /* H8/300 internal I/O functions */ #define __raw_readb __raw_readb From 694d0d0bb2030d2e36df73e2d23d5770511dbc8d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 14 Aug 2016 19:11:36 -0700 Subject: [PATCH 231/235] Linux 4.8-rc2 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 8c504f324154..5c18baad7218 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 8 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Psychotic Stoned Sheep # *DOCUMENTATION* From 1e12c4a9393b75a744aada2c8115434572698bc3 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Thu, 11 Aug 2016 14:19:42 +0100 Subject: [PATCH 232/235] genirq: Correctly configure the trigger on chained interrupts Commit 1e2a7d78499e ("irqdomain: Don't set type when mapping an IRQ") moved the trigger configuration call from the irqdomain mapping to the interrupt being actually requested. This patch failed to handle the case where we configure a chained interrupt, which doesn't get requested through the usual path. In order to solve this, let's call __irq_set_trigger just before starting the cascade interrupt. Special care must be taken to make the flow handler stick, as the .irq_set_type method could have reset it (it doesn't know we're dealing with a chained interrupt). Based on an initial patch by Jon Hunter. Fixes: 1e2a7d78499e ("irqdomain: Don't set type when mapping an IRQ") Reported-by: John Stultz Reported-by: Linus Walleij Tested-by: John Stultz Acked-by: Jon Hunter Signed-off-by: Marc Zyngier --- kernel/irq/chip.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index b4c1bc7c9ca2..637389088b3f 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -820,6 +820,17 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle, desc->name = name; if (handle != handle_bad_irq && is_chained) { + /* + * We're about to start this interrupt immediately, + * hence the need to set the trigger configuration. + * But the .set_type callback may have overridden the + * flow handler, ignoring that we're dealing with a + * chained interrupt. Reset it immediately because we + * do know better. + */ + __irq_set_trigger(desc, irqd_get_trigger_type(&desc->irq_data)); + desc->handle_irq = handle; + irq_settings_set_noprobe(desc); irq_settings_set_norequest(desc); irq_settings_set_nothread(desc); From 059e232089e45b0befc9933d31209c225e08b426 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 9 Aug 2016 07:50:44 +0100 Subject: [PATCH 233/235] irqchip/gic: Allow self-SGIs for SMP on UP configurations On systems where a single CPU is present, the GIC may not support having SGIs delivered to a target list. In that case, we use the self-SGI mechanism to allow the interrupt to be delivered locally. Tested-by: Fabio Estevam Signed-off-by: Marc Zyngier --- drivers/irqchip/irq-gic.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c index c2cab572c511..390fac59c6bc 100644 --- a/drivers/irqchip/irq-gic.c +++ b/drivers/irqchip/irq-gic.c @@ -769,6 +769,13 @@ static void gic_raise_softirq(const struct cpumask *mask, unsigned int irq) int cpu; unsigned long flags, map = 0; + if (unlikely(nr_cpu_ids == 1)) { + /* Only one CPU? let's do a self-IPI... */ + writel_relaxed(2 << 24 | irq, + gic_data_dist_base(&gic_data[0]) + GIC_DIST_SOFTINT); + return; + } + raw_spin_lock_irqsave(&irq_controller_lock, flags); /* Convert our logical CPU mask into a physical one. */ From ccd9432a5c85f35df7b491a1b701560d247466a5 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Wed, 17 Aug 2016 13:49:19 +0100 Subject: [PATCH 234/235] irqchip/gicv3: Remove disabling redistributor and group1 non-secure interrupts As per the GICv3 specification, to power down a processor using GICv3 and allow automatic power-on if an interrupt must be sent to a processor, software must set Enable to zero for all interrupt groups(by writing to GICC_CTLR or ICC_IGRPEN{0,1}_EL1/3 as appropriate. When commit 3708d52fc6bb ("irqchip: gic-v3: Implement CPU PM notifier") was introduced there were no firmware implementations(in particular PSCI) handling this. Linux kernel may not be aware of the CPU power state details and might fail to identify the power states that require quiescing the CPU interface. Even if it can be aware of those details, it can't determine which CPU power state have been triggered at the platform level and how the power control is implemented. This patch make disabling redistributor and group1 non-secure interrupts in the power down path and re-enabling of redistributor in the power-up path conditional. It will be handled in the kernel if and only if the non-secure accesses are permitted to access and modify control registers. It is left to the platform implementation otherwise. Cc: Marc Zyngier Cc: Lorenzo Pieralisi Cc: Thomas Gleixner Cc: Jason Cooper Tested-by: Christopher Covington Signed-off-by: Sudeep Holla Signed-off-by: Marc Zyngier --- drivers/irqchip/irq-gic-v3.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 6fc56c3466b0..ede5672ab34d 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -667,13 +667,20 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val, #endif #ifdef CONFIG_CPU_PM +/* Check whether it's single security state view */ +static bool gic_dist_security_disabled(void) +{ + return readl_relaxed(gic_data.dist_base + GICD_CTLR) & GICD_CTLR_DS; +} + static int gic_cpu_pm_notifier(struct notifier_block *self, unsigned long cmd, void *v) { if (cmd == CPU_PM_EXIT) { - gic_enable_redist(true); + if (gic_dist_security_disabled()) + gic_enable_redist(true); gic_cpu_sys_reg_init(); - } else if (cmd == CPU_PM_ENTER) { + } else if (cmd == CPU_PM_ENTER && gic_dist_security_disabled()) { gic_write_grpen1(0); gic_enable_redist(false); } From 7611da865c1060b2a7c87a15de663a59035747f8 Mon Sep 17 00:00:00 2001 From: David Daney Date: Thu, 18 Aug 2016 15:41:58 -0700 Subject: [PATCH 235/235] irqchip/gicv3-its: Disable the ITS before initializing it When starting a kexec/kdump kernel, the GIC ITS will already have been enabled. According to the ARM Generic Interrupt Controller Architecture Specification (GIC architecture Version 3.0 and version 4.0), writing to GITS_BASER or GITS_CBASER is "UNPREDICTABLE" when the ITS is enabled. On Cavium Thunder systems, this prevents the ITS from being initializing in the kexec/kdump kernel, resulting in failure to register/enable interrupts for all devices. The fix is to disable the ITS if it is not already in the disabled state. This allows the ITS to be properly initialized and then re-enabled in the kexec/kdump kernel. Acked-by: Marc Zyngier Signed-off-by: David Daney Signed-off-by: Marc Zyngier --- drivers/irqchip/irq-gic-v3-its.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 7ceaba81efb4..36b9c28a5c91 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -1545,7 +1545,12 @@ static int its_force_quiescent(void __iomem *base) u32 val; val = readl_relaxed(base + GITS_CTLR); - if (val & GITS_CTLR_QUIESCENT) + /* + * GIC architecture specification requires the ITS to be both + * disabled and quiescent for writes to GITS_BASER or + * GITS_CBASER to not have UNPREDICTABLE results. + */ + if ((val & GITS_CTLR_QUIESCENT) && !(val & GITS_CTLR_ENABLE)) return 0; /* Disable the generation of all interrupts to this ITS */