The next patch will make it possible for a generic PM domain to have
multiple parents (i.e. multiple PM domains it depends on). To
prepare for that change it is necessary to change pm_genpd_poweron()
so that it doesn't jump to the start label after running itself
recursively for the parent domain. For this purpose, introduce a new
PM domain status value GPD_STATE_WAIT_PARENT that will be set by
pm_genpd_poweron() before calling itself recursively for the parent
domain and modify the code in drivers/base/power/domain.c so that
the GPD_STATE_WAIT_PARENT status is guaranteed to be preserved during
the execution of pm_genpd_poweron() for the parent.
This change also causes pm_genpd_add_subdomain() and
pm_genpd_remove_subdomain() to wait for started pm_genpd_poweron() to
complete and allows pm_genpd_runtime_resume() to avoid dropping the
lock after powering on the PM domain.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Currently, pm_genpd_poweron() and pm_genpd_poweroff() need to take
the parent PM domain's lock in order to modify the parent's counter
of active subdomains in a nonracy way. This causes the locking to be
considerably complex and in fact is not necessary, because the
subdomain counters may be implemented as atomic fields and they
won't have to be modified under a lock.
Replace the unsigned in sd_count field in struct generic_pm_domain
by an atomic_t one and modify the code in drivers/base/power/domain.c
to take this change into account.
This patch doesn't change the locking yet, that is going to be done
in a separate subsequent patch.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: check size of FUSE_NOTIFY_INVAL_ENTRY message
fuse: mark pages accessed when written to
fuse: delete dead .write_begin and .write_end aops
fuse: fix flock
fuse: fix non-ANSI void function notation
* 'for-linus' of git://git.kernel.dk/linux-block: (23 commits)
Revert "cfq: Remove special treatment for metadata rqs."
block: fix flush machinery for stacking drivers with differring flush flags
block: improve rq_affinity placement
blktrace: add FLUSH/FUA support
Move some REQ flags to the common bio/request area
allow blk_flush_policy to return REQ_FSEQ_DATA independent of *FLUSH
xen/blkback: Make description more obvious.
cfq-iosched: Add documentation about idling
block: Make rq_affinity = 1 work as expected
block: swim3: fix unterminated of_device_id table
block/genhd.c: remove useless cast in diskstats_show()
drivers/cdrom/cdrom.c: relax check on dvd manufacturer value
drivers/block/drbd/drbd_nl.c: use bitmap_parse instead of __bitmap_parse
bsg-lib: add module.h include
cfq-iosched: Reduce linked group count upon group destruction
blk-throttle: correctly determine sync bio
loop: fix deadlock when sysfs and LOOP_CLR_FD race against each other
loop: add BLK_DEV_LOOP_MIN_COUNT=%i to allow distros 0 pre-allocated loop devices
loop: add management interface for on-demand device allocation
loop: replace linked list of allocated devices with an idr index
...
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
PCI: OF: Don't crash when bridge parent is NULL.
PCI: export pcie_bus_configure_settings symbol
PCI: code and comments cleanup
PCI: make cardbus-bridge resources optional
PCI: make SRIOV resources optional
PCI : ability to relocate assigned pci-resources
PCI: honor child buses add_size in hot plug configuration
PCI: Set PCI-E Max Payload Size on fabric
This allows the cast in lowmem_page_address (introduced as a warning
fixup to 33dd4e0ec9 "mm: make some struct page's const") to be
removed.
Propagate const'ness to page_to_section() as well since it is required
by __page_to_pfn.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Followup to 33dd4e0ec9 "mm: make some struct page's const" which missed the
HASHED_PAGE_VIRTUAL case.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
rtc: Limit RTC PIE frequency
rtc: Fix hrtimer deadlock
rtc: Handle errors correctly in rtc_irq_set_state()
Fixup trivial conflicts in drivers/rtc/interface.c due to slightly
trivially versions of the same patch coming in two different ways.
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
irq: Track the owner of irq descriptor
irq: Always set IRQF_ONESHOT if no primary handler is specified
genirq: Fix wrong bit operation
Commit ae1b153962, block: reimplement
FLUSH/FUA to support merge, introduced a performance regression when
running any sort of fsyncing workload using dm-multipath and certain
storage (in our case, an HP EVA). The test I ran was fs_mark, and it
dropped from ~800 files/sec on ext4 to ~100 files/sec. It turns out
that dm-multipath always advertised flush+fua support, and passed
commands on down the stack, where those flags used to get stripped off.
The above commit changed that behavior:
static inline struct request *__elv_next_request(struct request_queue *q)
{
struct request *rq;
while (1) {
- while (!list_empty(&q->queue_head)) {
+ if (!list_empty(&q->queue_head)) {
rq = list_entry_rq(q->queue_head.next);
- if (!(rq->cmd_flags & (REQ_FLUSH | REQ_FUA)) ||
- (rq->cmd_flags & REQ_FLUSH_SEQ))
- return rq;
- rq = blk_do_flush(q, rq);
- if (rq)
- return rq;
+ return rq;
}
Note that previously, a command would come in here, have
REQ_FLUSH|REQ_FUA set, and then get handed off to blk_do_flush:
struct request *blk_do_flush(struct request_queue *q, struct request *rq)
{
unsigned int fflags = q->flush_flags; /* may change, cache it */
bool has_flush = fflags & REQ_FLUSH, has_fua = fflags & REQ_FUA;
bool do_preflush = has_flush && (rq->cmd_flags & REQ_FLUSH);
bool do_postflush = has_flush && !has_fua && (rq->cmd_flags &
REQ_FUA);
unsigned skip = 0;
...
if (blk_rq_sectors(rq) && !do_preflush && !do_postflush) {
rq->cmd_flags &= ~REQ_FLUSH;
if (!has_fua)
rq->cmd_flags &= ~REQ_FUA;
return rq;
}
So, the flush machinery was bypassed in such cases (q->flush_flags == 0
&& rq->cmd_flags & (REQ_FLUSH|REQ_FUA)).
Now, however, we don't get into the flush machinery at all. Instead,
__elv_next_request just hands a request with flush and fua bits set to
the scsi_request_fn, even if the underlying request_queue does not
support flush or fua.
The agreed upon approach is to fix the flush machinery to allow
stacking. While this isn't used in practice (since there is only one
request-based dm target, and that target will now reflect the flush
flags of the underlying device), it does future-proof the solution, and
make it function as designed.
In order to make this work, I had to add a field to the struct request,
inside the flush structure (to store the original req->end_io). Shaohua
had suggested overloading the union with rb_node and completion_data,
but the completion data is used by device mapper and can also be used by
other drivers. So, I didn't see a way around the additional field.
I tested this patch on an HP EVA with both ext4 and xfs, and it recovers
the lost performance. Comments and other testers, as always, are
appreciated.
Cheers,
Jeff
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Function genpd_queue_power_off_work() is not defined for
CONFIG_PM_RUNTIME, so pm_genpd_poweroff_unused() causes a build
error to happen in that case. Fix the problem by making
pm_genpd_poweroff_unused() depend on CONFIG_PM_RUNTIME too.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (44 commits)
e1000e: increase driver version number
e1000e: alternate MAC address update
e1000e: do not disable receiver on 82574/82583
e1000e: alternate MAC address does not work on device id 0x1060
PCnet: Fix section mismatch
bnx2x: disable dcb on 578xx since not supported yet
bnx2x: properly clean indirect addresses
bnx2x: prevent race between undi_unload and load flows
bnx2x: fix select_queue when FCoE is disabled
bnx2x: init FCOE FP only once
ipv4: some rt_iif -> rt_route_iif conversions
net/bridge/netfilter/ebtables.c: use available error handling code
net/netlabel/netlabel_kapi.c: add missing cleanup code
net/irda: sh_sir: tidyup compile warning
net/irda: sh_sir: add missing header
net/irda: sh_irda: add missing header
slcan: ldisc generated skbs are received in softirq context
scm: Capture the full credentials of the scm sender
tcp: initialize variable ecn_ok in syncookies path
drivers/net/wireless/wl1251: add missing kfree
...
The patch http://lkml.org/lkml/2003/7/13/226 introduced an RLIMIT_NPROC
check in set_user() to check for NPROC exceeding via setuid() and
similar functions.
Before the check there was a possibility to greatly exceed the allowed
number of processes by an unprivileged user if the program relied on
rlimit only. But the check created new security threat: many poorly
written programs simply don't check setuid() return code and believe it
cannot fail if executed with root privileges. So, the check is removed
in this patch because of too often privilege escalations related to
buggy programs.
The NPROC can still be enforced in the common code flow of daemons
spawning user processes. Most of daemons do fork()+setuid()+execve().
The check introduced in execve() (1) enforces the same limit as in
setuid() and (2) doesn't create similar security issues.
Neil Brown suggested to track what specific process has exceeded the
limit by setting PF_NPROC_EXCEEDED process flag. With the change only
this process would fail on execve(), and other processes' execve()
behaviour is not changed.
Solar Designer suggested to re-check whether NPROC limit is still
exceeded at the moment of execve(). If the process was sleeping for
days between set*uid() and execve(), and the NPROC counter step down
under the limit, the defered execve() failure because NPROC limit was
exceeded days ago would be unexpected. If the limit is not exceeded
anymore, we clear the flag on successful calls to execve() and fork().
The flag is also cleared on successful calls to set_user() as the limit
was exceeded for the previous user, not the current one.
Similar check was introduced in -ow patches (without the process flag).
v3 - clear PF_NPROC_EXCEEDED on successful calls to set_user().
Reviewed-by: James Morris <jmorris@namei.org>
Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add FLUSH/FUA support to blktrace. As FLUSH precedes WRITE and/or
FUA follows WRITE, use the same 'F' flag for both cases and
distinguish them by their (relative) position. The end results
look like (other flags might be shown also):
- WRITE: W
- WRITE_FLUSH: FW
- WRITE_FUA: WF
- WRITE_FLUSH_FUA: FWF
Note that we reuse TC_BARRIER due to lack of bit space of act_mask
so that the older versions of blktrace tools will report flush
requests as barriers from now on.
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
REQ_SECURE, REQ_FLUSH and REQ_FUA may all be set on a bio as well as
on a request, so relocate them to the shared part of the enum.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
The patch adds empty function of_get_property for non-dt build, so that
drivers migrating to dt can save some '#ifdef CONFIG_OF'.
This also fixes the current Tegra compile problem in linux-next.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
In commit 2efaca927f ("mm/futex: fix futex writes on archs with SW
tracking of dirty & young") we forgot about MMU=n. This patch fixes
that.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: David Howells <dhowells@redhat.com>
Link: http://lkml.kernel.org/r/1311761831.24752.413.camel@twins
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Avoid annoying warnings from these functions ("discards qualifiers")
because they assign 'current_cred()' to a non-const pointer.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 3295514841 ("fix rcu annotations noise in cred.h") accidentally
dropped the const of current->cred inside current_cred() by the
insertion of a cast to deal with an RCU annotation loss warning from
sparce.
Use an appropriate RCU wrapper instead so as not to lose the const.
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit a9ff4f87 "fuse: support BSD locking semantics" overlooked a
number of issues with supporing flock locks over existing POSIX
locking infrastructure:
- it's not backward compatible, passing flock(2) calls to userspace
unconditionally (if userspace sets FUSE_POSIX_LOCKS)
- it doesn't cater for the fact that flock locks are automatically
unlocked on file release
- it doesn't take into account the fact that flock exclusive locks
(write locks) don't need an fd opened for write.
The last one invalidates the original premise of the patch that flock
locks can be emulated with POSIX locks.
This patch fixes the first two issues. The last one needs to be fixed
in userspace if the filesystem assumed that a write lock will happen
only on a file operned for write (as in the case of the current fuse
library).
Reported-by: Sebastian Pipping <webmaster@hartwork.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Currently userland will barf when including linux/netlink.h unless it
precisely includes sys/socket.h first. The issue is where the
definition of "sa_family_t" comes from.
We've been back and forth on how to fix this issue in the past, see:
http://thread.gmane.org/gmane.linux.debian.devel.bugs.general/622621http://thread.gmane.org/gmane.linux.network/143380
Ben Hutchings suggested we take a hint from how we handle the
sockaddr_storage type. First we define a "__kernel_sa_family_t"
to linux/socket.h that is always defined.
Then if __KERNEL__ is defined, we also define "sa_family_t" as
equal to "__kernel_sa_family_t".
Then in places like linux/netlink.h we use __kernel_sa_family_t
in user visible datastructures.
Reported-by: Michel Machado <michel@digirati.com.br>
Signed-off-by: David S. Miller <davem@davemloft.net>
task->cred is declared as __rcu, and access to other tasks' ->cred is,
indeed, protected. Access to current->cred does not need rcu_dereference()
at all, since only the task itself can change its ->cred. sparse, of
course, has no way of knowing that...
Add force-cast in current_cred(), make current_fsuid() et.al. use it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.open-osd.org/linux-open-osd:
ore: Make ore its own module
exofs: Rename raid engine from exofs/ios.c => ore
exofs: ios: Move to a per inode components & device-table
exofs: Move exofs specific osd operations out of ios.c
exofs: Add offset/length to exofs_get_io_state
exofs: Fix truncate for the raid-groups case
exofs: Small cleanup of exofs_fill_super
exofs: BUG: Avoid sbi realloc
exofs: Remove pnfs-osd private definitions
nfs_xdr: Move nfs4_string definition out of #ifdef CONFIG_NFS_V4
The inode structure layout is largely random, and some of the vfs paths
really do care. The path lookup in particular is already quite D$
intensive, and profiles show that accessing the 'inode->i_op->xyz'
fields is quite costly.
We already optimized the dcache to not unnecessarily load the d_op
structure for members that are often NULL using the DCACHE_OP_xyz bits
in dentry->d_flags, and this does something very similar for the inode
ops that are used during pathname lookup.
It also re-orders the fields so that the fields accessed by 'stat' are
together at the beginning of the inode structure, and roughly in the
order accessed.
The effect of this seems to be in the 1-2% range for an empty kernel
"make -j" run (which is fairly kernel-intensive, mostly in filename
lookup), so it's visible. The numbers are fairly noisy, though, and
likely depend a lot on exact microarchitecture. So there's more tuning
to be done.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Gcc tends to generate better code with small integers, including the
DCACHE_xyz flag tests - so move the common ones to be first in the list.
Also just remove the unused DCACHE_INOTIFY_PARENT_WATCHED and
DCACHE_AUTOFS_PENDING values, their users no longer exists in the source
tree.
And add a "unlikely()" to the DCACHE_OP_COMPARE test, since we want the
common case to be a nice straight-line fall-through.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
net: Compute protocol sequence numbers and fragment IDs using MD5.
crypto: Move md5_transform to lib/md5.c
Computers have become a lot faster since we compromised on the
partial MD4 hash which we use currently for performance reasons.
MD5 is a much safer choice, and is inline with both RFC1948 and
other ISS generators (OpenBSD, Solaris, etc.)
Furthermore, only having 24-bits of the sequence number be truly
unpredictable is a very serious limitation. So the periodic
regeneration and 8-bit counter have been removed. We compute and
use a full 32-bit sequence number.
For ipv6, DCCP was found to use a 32-bit truncated initial sequence
number (it needs 43-bits) and that is fixed here as well.
Reported-by: Dan Kaminsky <dan@doxpara.com>
Tested-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
For ChromiumOS, we use SHA-1 to verify the integrity of the root
filesystem. The speed of the kernel sha-1 implementation has a major
impact on our boot performance.
To improve boot performance, we investigated using the heavily optimized
sha-1 implementation used in git. With the git sha-1 implementation, we
see a 11.7% improvement in boot time.
10 reboots, remove slowest/fastest.
Before:
Mean: 6.58 seconds Stdev: 0.14
After (with git sha-1, this patch):
Mean: 5.89 seconds Stdev: 0.07
The other cool thing about the git SHA-1 implementation is that it only
needs 64 bytes of stack for the workspace while the original kernel
implementation needed 320 bytes.
Signed-off-by: Mandeep Singh Baines <msb@chromium.org>
Cc: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
Cc: Nicolas Pitre <nico@cam.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: David S. Miller <davem@davemloft.net>
Cc: linux-crypto@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
RCUify freeing acls, let check_acl() go ahead in RCU mode if acl is cached
get rid of boilerplate switches in posix_acl.h
fix block device fallout from ->fsync() changes
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx:
dmaengine: use DEFINE_IDR for static initialization
ioat: fix xor_idx_to_desc
Avoid section type conflict in dma/ioat/dma_v3.c
ioat: Adding PCI IDs for IOAT devices on SandyBridge platforms
exofs file system wants to use pnfs_osd_xdr.h file instead of
redefining pnfs-objects types in it's private "pnfs.h" headr.
Before we do the switch we must make sure pnfs_osd_xdr.h is
compilable also under NFS versions smaller than 4.1. Since now
it is needed regardless of version, by the exofs code.
nfs4_string is not the only nfs4 type out in the global scope.
Ack-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
This reverts commit 750f463a74.
of_alias_* still needs work to be generalized for 'promtree' dt
platforms, and to no implicitly create entries for available ids.
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
* 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6:
cpuidle: stop depending on pm_idle
x86 idle: move mwait_idle_with_hints() to where it is used
cpuidle: replace xen access to x86 pm_idle and default_idle
cpuidle: create bootparam "cpuidle.off=1"
mrst_pmu: driver for Intel Moorestown Power Management Unit
* 'apei-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
ACPI, APEI, EINJ Param support is disabled by default
APEI GHES: 32-bit buildfix
ACPI: APEI build fix
ACPI, APEI, GHES: Add hardware memory error recovery support
HWPoison: add memory_failure_queue()
ACPI, APEI, GHES, Error records content based throttle
ACPI, APEI, GHES, printk support for recoverable error via NMI
lib, Make gen_pool memory allocator lockless
lib, Add lock-less NULL terminated single list
Add Kconfig option ARCH_HAVE_NMI_SAFE_CMPXCHG
ACPI, APEI, Add WHEA _OSC support
ACPI, APEI, Add APEI bit support in generic _OSC call
ACPI, APEI, GHES, Support disable GHES at boot time
ACPI, APEI, GHES, Prevent GHES to be built as module
ACPI, APEI, Use apei_exec_run_optional in APEI EINJ and ERST
ACPI, APEI, Add apei_exec_run_optional
ACPI, APEI, GHES, Do not ratelimit fatal error printk before panic
ACPI, APEI, ERST, Fix erst-dbg long record reading issue
ACPI, APEI, ERST, Prevent erst_dbg from loading if ERST is disabled
We have already acknowledged that swapoff of a tmpfs file is slower than
it was before conversion to the generic radix_tree: a little slower
there will be acceptable, if the hotter paths are faster.
But it was a shock to find swapoff of a 500MB file 20 times slower on my
laptop, taking 10 minutes; and at that rate it significantly slows down
my testing.
Now, most of that turned out to be overhead from PROVE_LOCKING and
PROVE_RCU: without those it was only 4 times slower than before; and
more realistic tests on other machines don't fare as badly.
I've tried a number of things to improve it, including tagging the swap
entries, then doing lookup by tag: I'd expected that to halve the time,
but in practice it's erratic, and often counter-productive.
The only change I've so far found to make a consistent improvement, is
to short-circuit the way we go back and forth, gang lookup packing
entries into the array supplied, then shmem scanning that array for the
target entry. Scanning in place doubles the speed, so it's now only
twice as slow as before (or three times slower when the PROVEs are on).
So, add radix_tree_locate_item() as an expedient, once-off,
single-caller hack to do the lookup directly in place. #ifdef it on
CONFIG_SHMEM and CONFIG_SWAP, as much to document its limited
applicability as save space in other configurations. And, sadly,
#include sched.h for cond_resched().
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
But we've not yet removed the old swp_entry_t i_direct[16] from
shmem_inode_info. That's because it was still being shared with the
inline symlink. Remove it now (saving 64 or 128 bytes from shmem inode
size), and use kmemdup() for short symlinks, say, those up to 128 bytes.
I wonder why mpol_free_shared_policy() is done in shmem_destroy_inode()
rather than shmem_evict_inode(), where we usually do such freeing? I
guess it doesn't matter, and I'm not into NUMA mpol testing right now.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove mem_cgroup_shmem_charge_fallback(): it was only required when we
had to move swappage to filecache with GFP_NOWAIT.
Remove the GFP_NOWAIT special case from mem_cgroup_cache_charge(), by
moving its call out from shmem_add_to_page_cache() to two of thats three
callers. But leave it doing mem_cgroup_uncharge_cache_page() on error:
although asymmetrical, it's easier for all 3 callers to handle.
These two changes would also be appropriate if anyone were to start
using shmem_read_mapping_page_gfp() with GFP_NOWAIT.
Remove mem_cgroup_get_shmem_target(): mc_handle_file_pte() can test
radix_tree_exceptional_entry() to get what it needs for itself.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
While it's at its least, make a number of boring nitpicky cleanups to
shmem.c, mostly for consistency of variable naming. Things like "swap"
instead of "entry", "pgoff_t index" instead of "unsigned long idx".
And since everything else here is prefixed "shmem_", better change
init_tmpfs() to shmem_init().
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The maximum size of a shmem/tmpfs file has been limited by the maximum
size of its triple-indirect swap vector. With 4kB page size, maximum
filesize was just over 2TB on a 32-bit kernel, but sadly one eighth of
that on a 64-bit kernel. (With 8kB page size, maximum filesize was just
over 4TB on a 64-bit kernel, but 16TB on a 32-bit kernel,
MAX_LFS_FILESIZE being then more restrictive than swap vector layout.)
It's a shame that tmpfs should be more restrictive than ramfs, and this
limitation has now been noticed. Add another level to the swap vector?
No, it became obscure and hard to maintain, once I complicated it to
make use of highmem pages nine years ago: better choose another way.
Surely, if 2.4 had had the radix tree pagecache introduced in 2.5, then
tmpfs would never have invented its own peculiar radix tree: we would
have fitted swap entries into the common radix tree instead, in much the
same way as we fit swap entries into page tables.
And why should each file have a separate radix tree for its pages and
for its swap entries? The swap entries are required precisely where and
when the pages are not. We want to put them together in a single radix
tree: which can then avoid much of the locking which was needed to
prevent them from being exchanged underneath us.
This also avoids the waste of memory devoted to swap vectors, first in
the shmem_inode itself, then at least two more pages once a file grew
beyond 16 data pages (pages accounted by df and du, but not by memcg).
Allocated upfront, to avoid allocation when under swapping pressure, but
pure waste when CONFIG_SWAP is not set - I have never spattered around
the ifdefs to prevent that, preferring this move to sharing the common
radix tree instead.
There are three downsides to sharing the radix tree. One, that it binds
tmpfs more tightly to the rest of mm, either requiring knowledge of swap
entries in radix tree there, or duplication of its code here in shmem.c.
I believe that the simplications and memory savings (and probable higher
performance, not yet measured) justify that.
Two, that on HIGHMEM systems with SWAP enabled, it's the lowmem radix
nodes that cannot be freed under memory pressure - whereas before it was
the less precious highmem swap vector pages that could not be freed.
I'm hoping that 64-bit has now been accessible for long enough, that the
highmem argument has grown much less persuasive.
Three, that swapoff is slower than it used to be on tmpfs files, since
it's using a simple generic mechanism not tailored to it: I find this
noticeable, and shall want to improve, but maybe nobody else will
notice.
So... now remove most of the old swap vector code from shmem.c. But,
for the moment, keep the simple i_direct vector of 16 pages, with simple
accessors shmem_put_swap() and shmem_get_swap(), as a toy implementation
to help mark where swap needs to be handled in subsequent patches.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If swap entries are to be stored along with struct page pointers in a
radix tree, they need to be distinguished as exceptional entries.
Most of the handling of swap entries in radix tree will be contained in
shmem.c, but a few functions in filemap.c's common code need to check
for their appearance: find_get_page(), find_lock_page(),
find_get_pages() and find_get_pages_contig().
So as not to slow their fast paths, tuck those checks inside the
existing checks for unlikely radix_tree_deref_slot(); except for
find_lock_page(), where it is an added test. And make it a BUG in
find_get_pages_tag(), which is not applied to tmpfs files.
A part of the reason for eliminating shmem_readpage() earlier, was to
minimize the places where common code would need to allow for swap
entries.
The swp_entry_t known to swapfile.c must be massaged into a slightly
different form when stored in the radix tree, just as it gets massaged
into a pte_t when stored in page tables.
In an i386 kernel this limits its information (type and page offset) to
30 bits: given 32 "types" of swapfile and 4kB pagesize, that's a maximum
swapfile size of 128GB. Which is less than the 512GB we previously
allowed with X86_PAE (where the swap entry can occupy the entire upper
32 bits of a pte_t), but not a new limitation on 32-bit without PAE; and
there's not a new limitation on 64-bit (where swap filesize is already
limited to 16TB by a 32-bit page offset). Thirty areas of 128GB is
probably still enough swap for a 64GB 32-bit machine.
Provide swp_to_radix_entry() and radix_to_swp_entry() conversions, and
enforce filesize limit in read_swap_header(), just as for ptes.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
A patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its
peculiar swap vector, instead keeping a file's swap entries in the same
radix tree as its struct page pointers: thus saving memory, and
simplifying its code and locking.
This patch:
The radix_tree is used by several subsystems for different purposes. A
major use is to store the struct page pointers of a file's pagecache for
memory management. But what if mm wanted to store something other than
page pointers there too?
The low bit of a radix_tree entry is already used to denote an indirect
pointer, for internal use, and the unlikely radix_tree_deref_retry()
case.
Define the next bit as denoting an exceptional entry, and supply inline
functions radix_tree_exception() to return non-0 in either unlikely
case, and radix_tree_exceptional_entry() to return non-0 in the second
case.
If a subsystem already uses radix_tree with that bit set, no problem: it
does not affect internal workings at all, but is defined for the
convenience of those storing well-aligned pointers in the radix_tree.
The radix_tree_gang_lookups have an implicit assumption that the caller
can deduce the offset of each entry returned e.g. by the page->index of
a struct page. But that may not be feasible for some kinds of item to
be stored there.
radix_tree_gang_lookup_slot() allow for an optional indices argument,
output array in which to return those offsets. The same could be added
to other radix_tree_gang_lookups, but for now keep it to the only one
for which we need it.
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
