With the recent syntax extension, Kconfig is now able to evaluate the
compiler / toolchain capability.
However, accumulating flags to 'LD' is not compatible with the way
it works; 'LD' must be passed to Kconfig to call $(ld-option,...)
from Kconfig files. If you tweak 'LD' in arch Makefile depending on
CONFIG_CPU_BIG_ENDIAN, this would end up with circular dependency
between Makefile and Kconfig.
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
These patches for building 32-bit RISC-V kernel.
- Fix the compile errors and warnings on RV32I.
- Fix some incompatible problem on RV32I.
- Add format.h for compatible of print format.
The fixed width integer types format for Elf_Addr will move to
generic header by another patch. For now, there are some warning
about unexpected argument of type on RV32I.
Change in v1:
- Fix some error in v1
- Remove implementation of fixed width integer types format for Elf_Addr.
In riscv_gpr_set, pass regs instead of ®s to user_regset_copyin to fix
gdb segfault.
Signed-off-by: Jim Wilson <jimw@sifive.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
This file has never existed in the upstream kernel, but it's guarded by
an #ifdef that's also never existed in the upstream kernel. As a part
of our interrupt controller refactoring this header is no longer
necessary, but this reference managed to sneak in anyway.
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The DT core will call of_platform_default_populate, so it is not
necessary for arch specific code to call it unless there are custom
match entries, auxdata or parent device. Neither of those apply here, so
remove the call.
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: linux-riscv@lists.infradead.org
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The R_RISCV_ADD32/R_RISCV_SUB32 relocations should add/subtract the
address of the symbol (without overflow check), not its contents.
Signed-off-by: Andreas Schwab <schwab@suse.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
On 32-bit, it need to use __ucmpdi2, otherwise, it can't find the __ucmpdi2
symbol.
Signed-off-by: Zong Li <zong@andestech.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The DMA32 is for 64-bit usage.
Signed-off-by: Zong Li <zong@andestech.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The aarch64linux and aarch64linuxb emulation modes are not supported by
bare-metal toolchains and Linux using them forbids building the kernel
with these toolchains.
Since there is apparently no reason to target these emulation modes, the
more generic elf modes are used instead, allowing to build on bare-metal
toolchains as well as the already-supported ones.
Fixes: 3d6a7b99e3 ("arm64: ensure the kernel is compiled for LP64")
Cc: stable@vger.kernel.org
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Current implementation does not guarantee packed pixel modes working
with every dongle. There are some dongles, which require selecting
the output mode explicitly.
Write proper values to registers in packed_pixel mode, based on how it
is done in vendor's code. Select output color space: RGB
(no packed pixel) or YCBCR422 (packed pixel).
This reverts commit e8b92efa62
("drm/bridge/sii8620: fix display of packed pixel modes in MHL2").
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530204243-6370-3-git-send-email-m.purski@samsung.com
Currently AVI infoframe is sent only in MHL3. However, some MHL2 dongles
need AVI infoframe to work correctly in either packed pixel mode or
non-packed pixel mode.
Send AVI infoframe in set_infoframes() in every case. Create an
infoframe using drm_hdmi_infoframe_from_display_mode() instead of
manually filling each infoframe structure's field.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530204243-6370-2-git-send-email-m.purski@samsung.com
A hooking API was implemented for 4.17 in fa93854f7a followed
by hooks for Thinkpad laptops in 2801b9683f. The Thinkpad
drivers did not support the Thinkpad 13 and the hooking API crashes
on unsupported batteries by altering a list of hooks during unsafe
iteration. Thus, Thinkpad 13 laptops could no longer boot.
Additionally, a lock was kept in place and debugging information was
printed out of order.
Fixes: fa93854f7a (battery: Add the battery hooking API)
Cc: 4.17+ <stable@vger.kernel.org> # 4.17+
Signed-off-by: Jouke Witteveen <j.witteveen@gmail.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The buffer object backing the user fence is reserved using the non-user
fence, i.e., as soon as the non-user fence is signaled, the user fence
buffer object can be moved or even destroyed.
Therefore, emit the user fence first.
Both fences have the same cache invalidation behavior, so this should
have no user-visible effect.
Signed-off-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
If struct page is poisoned, and uninitialized access is detected via
PF_POISONED_CHECK(page) dump_page() is called to output the page. But,
the dump_page() itself accesses struct page to determine how to print
it, and therefore gets into a recursive loop.
For example:
dump_page()
__dump_page()
PageSlab(page)
PF_POISONED_CHECK(page)
VM_BUG_ON_PGFLAGS(PagePoisoned(page), page)
dump_page() recursion loop.
Link: http://lkml.kernel.org/r/20180702180536.2552-1-pasha.tatashin@oracle.com
Fixes: f165b378bb ("mm: uninitialized struct page poisoning sanity checking")
Signed-off-by: Pavel Tatashin <pasha.tatashin@oracle.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The ARM trusted foundations code is currently broken in linux-next when
CONFIG_KCOV_INSTRUMENT_ALL is set:
/tmp/ccHdQsCI.s: Assembler messages:
/tmp/ccHdQsCI.s:37: Error: .err encountered
/tmp/ccHdQsCI.s:38: Error: .err encountered
/tmp/ccHdQsCI.s:39: Error: .err encountered
scripts/Makefile.build:311: recipe for target 'arch/arm/firmware/trusted_foundations.o' failed
I could not find a function attribute that lets me disable
-fsanitize-coverage=trace-pc for just one function, so this turns it off
for the entire file instead.
Link: http://lkml.kernel.org/r/20180529103636.1535457-1-arnd@arndb.de
Fixes: 758517202b ("arm: port KCOV to arm")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Olof Johansson <olof@lixom.net>
Tested-by: Olof Johansson <olof@lixom.net>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There is a special case that the size is "(N << KASAN_SHADOW_SCALE_SHIFT)
Pages plus X", the value of X is [1, KASAN_SHADOW_SCALE_SIZE-1]. The
operation "size >> KASAN_SHADOW_SCALE_SHIFT" will drop X, and the
roundup operation can not retrieve the missed one page. For example:
size=0x28006, PAGE_SIZE=0x1000, KASAN_SHADOW_SCALE_SHIFT=3, we will get
shadow_size=0x5000, but actually we need 6 pages.
shadow_size = round_up(size >> KASAN_SHADOW_SCALE_SHIFT, PAGE_SIZE);
This can lead to a kernel crash when kasan is enabled and the value of
mod->core_layout.size or mod->init_layout.size is like above. Because
the shadow memory of X has not been allocated and mapped.
move_module:
ptr = module_alloc(mod->core_layout.size);
...
memset(ptr, 0, mod->core_layout.size); //crashed
Unable to handle kernel paging request at virtual address ffff0fffff97b000
......
Call trace:
__asan_storeN+0x174/0x1a8
memset+0x24/0x48
layout_and_allocate+0xcd8/0x1800
load_module+0x190/0x23e8
SyS_finit_module+0x148/0x180
Link: http://lkml.kernel.org/r/1529659626-12660-1-git-send-email-thunder.leizhen@huawei.com
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Dmitriy Vyukov <dvyukov@google.com>
Acked-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Hanjun Guo <guohanjun@huawei.com>
Cc: Libin <huawei.libin@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When booting with very large numbers of gigantic (i.e. 1G) pages, the
operations in the loop of gather_bootmem_prealloc, and specifically
prep_compound_gigantic_page, takes a very long time, and can cause a
softlockup if enough pages are requested at boot.
For example booting with 3844 1G pages requires prepping
(set_compound_head, init the count) over 1 billion 4K tail pages, which
takes considerable time.
Add a cond_resched() to the outer loop in gather_bootmem_prealloc() to
prevent this lockup.
Tested: Booted with softlockup_panic=1 hugepagesz=1G hugepages=3844 and
no softlockup is reported, and the hugepages are reported as
successfully setup.
Link: http://lkml.kernel.org/r/20180627214447.260804-1-cannonmatthews@google.com
Signed-off-by: Cannon Matthews <cannonmatthews@google.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Andres Lagar-Cavilla <andreslc@google.com>
Cc: Peter Feiner <pfeiner@google.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use huge_ptep_get() to translate huge ptes to normal ptes so we can
check them with the huge_pte_* functions. Otherwise some architectures
will check the wrong values and will not wait for userspace to bring in
the memory.
Link: http://lkml.kernel.org/r/20180626132421.78084-1-frankja@linux.ibm.com
Fixes: 369cd2121b ("userfaultfd: hugetlbfs: userfaultfd_huge_must_wait for hugepmd ranges")
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
clear_ftrace_function is not used outside of ftrace.c and is not help to
use a function, so nuke it per Steve's suggestion.
Link: http://lkml.kernel.org/r/1517537689-34947-1-git-send-email-xieyisheng1@huawei.com
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Yisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Silence warnings (triggered at W=1) by adding relevant __printf attributes.
CC kernel/trace/trace.o
kernel/trace/trace.c: In function ‘__trace_array_vprintk’:
kernel/trace/trace.c:2979:2: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]
len = vscnprintf(tbuffer, TRACE_BUF_SIZE, fmt, args);
^~~
AR kernel/trace/built-in.o
Link: http://lkml.kernel.org/r/20180308205843.27447-1-malat@debian.org
Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Simplify and optimize the logic in trace_buffer_iter() to use a conditional
operation instead of an if conditional.
Link: http://lkml.kernel.org/r/20180408113631.3947-1-cugyly@163.com
Signed-off-by: yuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The comment in create_filter() states that the passed in filter pointer
(filterp) will either be NULL or contain an error message stating why the
filter failed. But it also expects the filter pointer to point to NULL when
passed in. If it is not, the function create_filter_start() will warn and
return an error message without updating the filter pointer. This is not
what the comment states.
As we always expect the pointer to point to NULL, if it is not, trigger a
WARN_ON(), set it to NULL, and then continue the path as the rest will work
as the comment states. Also update the comment to state it must point to
NULL.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
'err' is used as a NUL-terminated string, but using strncpy() with the length
equal to the buffer size may result in lack of the termination:
kernel/trace/trace_events_hist.c: In function 'hist_err_event':
kernel/trace/trace_events_hist.c:396:3: error: 'strncpy' specified bound 256 equals destination size [-Werror=stringop-truncation]
strncpy(err, var, MAX_FILTER_STR_VAL);
This changes it to use the safer strscpy() instead.
Link: http://lkml.kernel.org/r/20180328140920.2842153-1-arnd@arndb.de
Cc: stable@vger.kernel.org
Fixes: f404da6e1d ("tracing: Add 'last error' error facility for hist triggers")
Acked-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
My networking merge (commit 4e33d7d479: "Pull networking fixes from
David Miller") got the poll() handling conflict wrong for af_smc.
The conflict between my a11e1d432b ("Revert changes to convert to
->poll_mask() and aio IOCB_CMD_POLL") and Ursula Braun's 24ac3a08e6
("net/smc: rebuild nonblocking connect") should have left the call to
sock_poll_wait() in place, just without the socket lock release/retake.
And I really should have realized that. But happily, I at least asked
Ursula to double-check the merge, and she set me right.
This also fixes an incidental whitespace issue nearby that annoyed me
while looking at this.
Pointed-out-by: Ursula Braun <ubraun@linux.ibm.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
- A couple of imx defconfig updates selecting USB ULPI support to fix
a regression seen with USB driver, which is caused by commit
03e6275ae3 ("usb: chipidea: Fix ULPI on imx51").
- A fix on imx51-zii-rdu1 board touchscreen pinctrl setting, which
causes an interrupt storm.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJbO3x0AAoJEFBXWFqHsHzODGkIAIIJFPOWBYHyxAVLAtqpNTt7
+Yggu9LGe7rhspT+Hi2DocKIEj1Z11x49KJCqxwrEHg+7xyax6d78b1iclCHLRlr
S9jxulipzkBs/neebrhh6cyF8sxA7TPUgXtEMdRJwYygUoZhe9hngTO3jv+xoDky
l1l9zH9gAlAQA1cxN6Z4ZWYDkNS0mYK3e1VzaeyCXsq5GJ1Vm5Qw1ooTE5DL3K1g
5B2Hw057DJvcSwOhIVir/MQR8sAWj1sKYdsCouxrO/ThEyofEAtfn22romcPaGPW
WBOuTNrOrMy6O/+ndMBNJBXeGt1tONL2cjLbrpffaXJYq6m/56efsee9u1u18Vg=
=4Agh
-----END PGP SIGNATURE-----
Merge tag 'imx-fixes-4.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux into fixes
i.MX fixes for 4.18, round 2:
- A couple of imx defconfig updates selecting USB ULPI support to fix
a regression seen with USB driver, which is caused by commit
03e6275ae3 ("usb: chipidea: Fix ULPI on imx51").
- A fix on imx51-zii-rdu1 board touchscreen pinctrl setting, which
causes an interrupt storm.
* tag 'imx-fixes-4.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
ARM: imx_v4_v5_defconfig: Select ULPI support
ARM: imx_v6_v7_defconfig: Select ULPI support
ARM: dts: imx51-zii-rdu1: fix touchscreen pinctrl
Signed-off-by: Olof Johansson <olof@lixom.net>
There are no legacy behavior in drivers to consider while attaching a
device to genpd - for the multiple PM domain case.
For that reason, let's instead require the driver to runtime resume the
device, via calling pm_runtime_get_sync() for example, when it needs to
power on the corresponding PM domain.
This allows us to improve the situation during attach. Instead of always
power on the PM domain, which may be unnecessary, let's leave it in its
current state. Additionally, to avoid the PM domain to stay powered on,
let's schedule a power off work.
Fixes: 3c095f32a9 (PM / Domains: Add support for multi PM domains ...)
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This change adds LOOP_SET_BLOCK_SIZE as one of the supported ioctls
in lo_compat_ioctl. It only takes an unsigned long argument, and
in practice a 32-bit value works fine.
Reviewed-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Evan Green <evgreen@chromium.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Select CONFIG_USB_CHIPIDEA_ULPI and CONFIG_USB_ULPI_BUS so that
USB ULPI can be functional on some boards like that use ULPI
interface.
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Select CONFIG_USB_CHIPIDEA_ULPI and CONFIG_USB_ULPI_BUS so that
USB ULPI can be functional on some boards like imx51-babbge.
This fixes a kernel hang in 4.18-rc1 on i.mx51-babbage, caused by commit
03e6275ae3 ("usb: chipidea: Fix ULPI on imx51").
Suggested-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
If the whole object is already pinned by HW for use as scanout, we will
fail to move it to the mappable region and so must resort to using a
partial VMA covering the whole object.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104513
Fixes: aa136d9d72 ("drm/i915: Convert partial ggtt vma to full ggtt if it spans the entire object")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180630090509.469-1-chris@chris-wilson.co.uk
(cherry picked from commit 7e7367d3bc)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
On AMD, the presence of the MSR_SPEC_CTRL feature does not imply that the
SSBD mitigation support should use the SPEC_CTRL MSR. Other features could
have caused the MSR_SPEC_CTRL feature to be set, while a different SSBD
mitigation option is in place.
Update the SSBD support to check for the actual SSBD features that will
use the SPEC_CTRL MSR.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 6ac2f49edb ("x86/bugs: Add AMD's SPEC_CTRL MSR usage")
Link: http://lkml.kernel.org/r/20180702213602.29202.33151.stgit@tlendack-t1.amdoffice.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If either the X86_FEATURE_AMD_SSBD or X86_FEATURE_VIRT_SSBD features are
present, then there is no need to perform the check for the LS_CFG SSBD
mitigation support.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20180702213553.29202.21089.stgit@tlendack-t1.amdoffice.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A previous patch removed OMAP clock aliases that were perceived
to be unnecessary. Unfortunately, it broke the ethernet on the
am3517-evm. This patch enables the MDIO clock and EMAC clock.
Fixes: 0ed266d7ae ("clk: ti: omap3: cleanup unnecessary clock aliases")
Cc: stable@vger.kernel.org #4.16+
Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
On 32-bit kernels, __flush_tlb_all() may have read the CR4 shadow before the
initialization of CR4 shadow in cpu_init().
Fix it by adding an explicit cr4_init_shadow() call into start_secondary()
which is the first function called on non-boot SMP CPUs - ahead of the
__flush_tlb_all() call.
( This is somewhat of a layering violation, but start_secondary() does
CR4 bootstrap in the PCID case anyway. )
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Link: http://lkml.kernel.org/r/b07b6ae9-4b57-4b40-b9bc-50c2c67f1d91@default
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Gaurav reports that commit:
85f1abe001 ("kthread, sched/wait: Fix kthread_parkme() completion issue")
isn't working for him. Because of the following race:
> controller Thread CPUHP Thread
> takedown_cpu
> kthread_park
> kthread_parkme
> Set KTHREAD_SHOULD_PARK
> smpboot_thread_fn
> set Task interruptible
>
>
> wake_up_process
> if (!(p->state & state))
> goto out;
>
> Kthread_parkme
> SET TASK_PARKED
> schedule
> raw_spin_lock(&rq->lock)
> ttwu_remote
> waiting for __task_rq_lock
> context_switch
>
> finish_lock_switch
>
>
>
> Case TASK_PARKED
> kthread_park_complete
>
>
> SET Running
Furthermore, Oleg noticed that the whole scheduler TASK_PARKED
handling is buggered because the TASK_DEAD thing is done with
preemption disabled, the current code can still complete early on
preemption :/
So basically revert that earlier fix and go with a variant of the
alternative mentioned in the commit. Promote TASK_PARKED to special
state to avoid the store-store issue on task->state leading to the
WARN in kthread_unpark() -> __kthread_bind().
But in addition, add wait_task_inactive() to kthread_park() to ensure
the task really is PARKED when we return from kthread_park(). This
avoids the whole kthread still gets migrated nonsense -- although it
would be really good to get this done differently.
Reported-by: Gaurav Kohli <gkohli@codeaurora.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 85f1abe001 ("kthread, sched/wait: Fix kthread_parkme() completion issue")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a cfs_rq is throttled, parent cfs_rq->nr_running is decreased and
everything happens at cfs_rq level. Currently util_est stays unchanged
in such case and it keeps accounting the utilization of throttled tasks.
This can somewhat make sense as we don't dequeue tasks but only throttled
cfs_rq.
If a task of another group is enqueued/dequeued and root cfs_rq becomes
idle during the dequeue, util_est will be cleared whereas it was
accounting util_est of throttled tasks before. So the behavior of util_est
is not always the same regarding throttled tasks and depends of side
activity. Furthermore, util_est will not be updated when the cfs_rq is
unthrottled as everything happens at cfs_rq level. Main results is that
util_est will stay null whereas we now have running tasks. We have to wait
for the next dequeue/enqueue of the previously throttled tasks to get an
up to date util_est.
Remove the assumption that cfs_rq's estimated utilization of a CPU is 0
if there is no running task so the util_est of a task remains until the
latter is dequeued even if its cfs_rq has been throttled.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Patrick Bellasi <patrick.bellasi@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 7f65ea42eb ("sched/fair: Add util_est on top of PELT")
Link: http://lkml.kernel.org/r/1528972380-16268-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When period gets restarted after some idle time, start_cfs_bandwidth()
doesn't update the expiration information, expire_cfs_rq_runtime() will
see cfs_rq->runtime_expires smaller than rq clock and go to the clock
drift logic, wasting needless CPU cycles on the scheduler hot path.
Update the global expiration in start_cfs_bandwidth() to avoid frequent
expire_cfs_rq_runtime() calls once a new period begins.
Signed-off-by: Xunlei Pang <xlpang@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20180620101834.24455-2-xlpang@linux.alibaba.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
I noticed that cgroup task groups constantly get throttled even
if they have low CPU usage, this causes some jitters on the response
time to some of our business containers when enabling CPU quotas.
It's very simple to reproduce:
mkdir /sys/fs/cgroup/cpu/test
cd /sys/fs/cgroup/cpu/test
echo 100000 > cpu.cfs_quota_us
echo $$ > tasks
then repeat:
cat cpu.stat | grep nr_throttled # nr_throttled will increase steadily
After some analysis, we found that cfs_rq::runtime_remaining will
be cleared by expire_cfs_rq_runtime() due to two equal but stale
"cfs_{b|q}->runtime_expires" after period timer is re-armed.
The current condition to judge clock drift in expire_cfs_rq_runtime()
is wrong, the two runtime_expires are actually the same when clock
drift happens, so this condtion can never hit. The orginal design was
correctly done by this commit:
a9cf55b286 ("sched: Expire invalid runtime")
... but was changed to be the current implementation due to its locking bug.
This patch introduces another way, it adds a new field in both structures
cfs_rq and cfs_bandwidth to record the expiration update sequence, and
uses them to figure out if clock drift happens (true if they are equal).
Signed-off-by: Xunlei Pang <xlpang@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 51f2176d74 ("sched/fair: Fix unlocked reads of some cfs_b->quota/period")
Link: http://lkml.kernel.org/r/20180620101834.24455-1-xlpang@linux.alibaba.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With commit:
8f111bc357 ("cpufreq/schedutil: Rewrite CPUFREQ_RT support")
the schedutil governor uses rq->rt.rt_nr_running to detect whether an
RT task is currently running on the CPU and to set frequency to max
if necessary.
cpufreq_update_util() is called in enqueue/dequeue_top_rt_rq() but
rq->rt.rt_nr_running has not been updated yet when dequeue_top_rt_rq() is
called so schedutil still considers that an RT task is running when the
last task is dequeued. The update of rq->rt.rt_nr_running happens later
in dequeue_rt_stack().
In fact, we can take advantage of the sequence that the dequeue then
re-enqueue rt entities when a rt task is enqueued or dequeued;
As a result enqueue_top_rt_rq() is always called when a task is
enqueued or dequeued and also when groups are throttled or unthrottled.
The only place that not use enqueue_top_rt_rq() is when root rt_rq is
throttled.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: efault@gmx.de
Cc: juri.lelli@redhat.com
Cc: patrick.bellasi@arm.com
Cc: viresh.kumar@linaro.org
Fixes: 8f111bc357 ('cpufreq/schedutil: Rewrite CPUFREQ_RT support')
Link: http://lkml.kernel.org/r/1530021202-21695-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
