Leo Yan [Tue, 15 Sep 2015 10:56:45 +0000 (18:56 +0800)]
sched/fair: Remove unnecessary parameter for group_classify()
The group_classify() function does not use the "env" parameter, so remove it.
Also unify code to always use group_classify() to calculate group's
load type.
Signed-off-by: Leo Yan <leo.yan@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> Link: http://lkml.kernel.org/r/1442314605-14838-1-git-send-email-leo.yan@linaro.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
sched/numa: Limit the amount of virtual memory scanned in task_numa_work()
Currently task_numa_work() scans up to numa_balancing_scan_size_mb worth
of memory per invocation, but only counts memory areas that have at
least one PTE that is still present and not marked for numa hint faulting.
It will skip over arbitarily large amounts of memory that are either
unused, full of swap ptes, or full of PTEs that were already marked
for NUMA hint faults but have not been faulted on yet.
This can cause excessive amounts of CPU use, due to there being
essentially no upper limit on the scan rate of very large processes
that are not yet in a phase where they are actively accessing old
memory pages (eg. they are still initializing their data).
Avoid that problem by placing an upper limit on the amount of virtual
memory that task_numa_work() scans in each invocation. This can be a
higher limit than "pages", to ensure the task still skips over unused
areas fairly quickly.
While we are here, also fix the "nr_pte_updates" logic, so it only
counts page ranges with ptes in them.
Reported-by: Andrea Arcangeli <aarcange@redhat.com> Reported-by: Jan Stancek <jstancek@redhat.com> Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Mel Gorman <mgorman@suse.de> 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/20150911090027.4a7987bd@annuminas.surriel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
Henrik Austad [Wed, 9 Sep 2015 15:00:41 +0000 (17:00 +0200)]
sched/core: Make policy-testing consistent
Most of the policy-tests are done via the <class>_policy() helpers with
the notable exception of idle. A new wrapper for valid_policy() has also
been added to improve readability in set_load_weight().
This commit does not change the logical behavior of the scheduler core.
Signed-off-by: Henrik Austad <henrik@austad.us> 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: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1441810841-4756-1-git-send-email-henrik@austad.us Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge tag 'platform-drivers-x86-v4.3-2' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86
Pull x86 platform driver fixes from Darren Hart:
"Fix an issue introduced by the previous major toshiba rework. Add a
quirk. Workaround a few platform specific firmware items. One
cleanup to wmi I inadvertently dropped from a previous pull request.
Details:
hp-wmi:
- limit hotkey enable
toshiba_acpi:
- Fix hotkeys registration on some toshiba models
- Fix USB Sleep and Music always disabled
wmi:
- Remove private %pUL implementation
asus-nb-wmi:
- Add wapf=4 quirk for X456UA/X456UF"
* tag 'platform-drivers-x86-v4.3-2' of git://git.infradead.org/users/dvhart/linux-platform-drivers-x86:
hp-wmi: limit hotkey enable
toshiba_acpi: Fix hotkeys registration on some toshiba models
toshiba_acpi: Fix USB Sleep and Music always disabled
wmi: Remove private %pUL implementation
asus-nb-wmi: Add wapf=4 quirk for X456UA/X456UF
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
revert "mm: make sure all file VMAs have ->vm_ops set"
MAINTAINERS: update LTP mailing list
userfaultfd: add missing mmput() in error path
lib/string_helpers.c: fix infinite loop in string_get_size()
alpha: lib: export __delay
alpha: io: define ioremap_uc
kasan: fix last shadow judgement in memory_is_poisoned_16()
zram: fix possible use after free in zcomp_create()
Andrew Morton [Thu, 17 Sep 2015 23:02:00 +0000 (16:02 -0700)]
revert "mm: make sure all file VMAs have ->vm_ops set"
Revert commit 6dc296e7df4c "mm: make sure all file VMAs have ->vm_ops
set".
Will Deacon reports that it "causes some mmap regressions in LTP, which
appears to use a MAP_PRIVATE mmap of /dev/zero as a way to get anonymous
pages in some of its tests (specifically mmap10 [1])".
William Shuman reports Oracle crashes.
So revert the patch while we work out what to do.
Reported-by: William Shuman <wshuman3@gmail.com> Reported-by: Will Deacon <will.deacon@arm.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lib/string_helpers.c: fix infinite loop in string_get_size()
Some string_get_size() calls (e.g.:
string_get_size(1, 512, STRING_UNITS_10, ..., ...)
string_get_size(15, 64, STRING_UNITS_10, ..., ...)
) result in an infinite loop. The problem is that if size is equal to
divisor[units]/blk_size and is smaller than divisor[units] we'll end
up with size == 0 when we start doing sf_cap calculations:
For string_get_size(1, 512, STRING_UNITS_10, ..., ...) case:
...
remainder = do_div(size, divisor[units]); -> size is 0, remainder is 1
remainder *= blk_size; -> remainder is 512
...
size *= blk_size; -> size is still 0
size += remainder / divisor[units]; -> size is still 0
The caller causing the issue is sd_read_capacity(), the problem was
noticed on Hyper-V, such weird size was reported by host when scanning
collides with device removal. This is probably a separate issue worth
fixing, this patch is intended to prevent the library routine from
infinite looping.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Acked-by: James Bottomley <JBottomley@Odin.com> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Cc: "K. Y. Srinivasan" <kys@microsoft.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__delay was not exported as a result while building with allmodconfig we
were getting build error of undefined symbol. __delay is being used by:
drivers/net/phy/mdio-octeon.c
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
ioremap_uc was not defined and as a result while building with
allmodconfig were getting build error of: implicit declaration of
function 'ioremap_uc'.
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kasan: fix last shadow judgement in memory_is_poisoned_16()
The shadow which correspond 16 bytes memory may span 2 or 3 bytes. If
the memory is aligned on 8, then the shadow takes only 2 bytes. So we
check "shadow_first_bytes" is enough, and need not to call
"memory_is_poisoned_1(addr + 15);". But the code "if
(likely(!last_byte))" is wrong judgement.
e.g. addr=0, so last_byte = 15 & KASAN_SHADOW_MASK = 7, then the code
will continue to call "memory_is_poisoned_1(addr + 15);"
Signed-off-by: Xishi Qiu <qiuxishi@huawei.com> Acked-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: Andrey Konovalov <adech.fo@gmail.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Michal Marek <mmarek@suse.cz> Cc: <zhongjiang@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Luis Henriques [Thu, 17 Sep 2015 23:01:40 +0000 (16:01 -0700)]
zram: fix possible use after free in zcomp_create()
zcomp_create() verifies the success of zcomp_strm_{multi,single}_create()
through comp->stream, which can potentially be pointing to memory that
was freed if these functions returned an error.
While at it, replace a 'ERR_PTR(-ENOMEM)' by a more generic
'ERR_PTR(error)' as in the future zcomp_strm_{multi,siggle}_create()
could return other error codes. Function documentation updated
accordingly.
Fixes: beca3ec71fe5 ("zram: add multi stream functionality") Signed-off-by: Luis Henriques <luis.henriques@canonical.com> Acked-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Acked-by: Minchan Kim <minchan@kernel.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Do not write initialize magic on systems that do not have
feature query 0xb. Fixes Bug #82451.
Redefine FEATURE_QUERY to align with 0xb and FEATURE2 with 0xd
for code clearity.
Add a new test function, hp_wmi_bios_2008_later() & simplify
hp_wmi_bios_2009_later(), which fixes a bug in cases where
an improper value is returned. Probably also fixes Bug #69131.
Add missing __init tag.
Signed-off-by: Kyle Evans <kvans32@gmail.com> Cc: stable@vger.kernel.org Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
Pull Ceph fixes from Sage Weil:
"These are both fixes to the new and improved keepalive2 behavior"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
libceph: advertise support for keepalive2
libceph: don't access invalid memory in keepalive2 path
Merge tag 'for-v4.3-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply
Pull power supply fixes from Sebastian Reichel:
"twl4030-charger fixes"
* tag 'for-v4.3-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply:
twl4030_charger: fix another compile error
Revert "twl4030_charger: correctly handle -EPROBE_DEFER from devm_usb_get_phy_by_node"
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"This addresses some problems with filesystem writeback due to the
recently merged hardware DBM patches, which caused us to treat some
read-only pages as dirty.
There are also some other, less significant fixes that are described
in the summary below:
A mixture of fixes for regressions introduced during the merge window,
some longer standing problems that we spotted and a couple of hardware
errata. The main changes are:
- Fix fallout from the h/w DBM patches, causing filesystem writeback
issues on both v8 and v8.1 CPUs
- Workaround for Cortex-A53 erratum #843419 in the module loader
- Fix for long-standing issue with compat big-endian signal handlers
using the saved floating point state"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: errata: add module build workaround for erratum #843419
arm64: compat: fix vfp save/restore across signal handlers in big-endian
arm64: cpu hotplug: ensure we mask out CPU_TASKS_FROZEN in notifiers
arm64: head.S: initialise mdcr_el2 in el2_setup
arm64: enable generic idle loop
arm64: pgtable: use a single bit for PTE_WRITE regardless of DBM
arm64: Fix pte_modify() to preserve the hardware dirty information
arm64: Fix the pte_hw_dirty() check when AF/DBM is enabled
arm64: dma-mapping: check whether cma area is initialized or not
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
- misc fixes all around the map
- block non-root vm86(old) if mmap_min_addr != 0
- two small debuggability improvements
- removal of obsolete paravirt op
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/platform: Fix Geode LX timekeeping in the generic x86 build
x86/apic: Serialize LVTT and TSC_DEADLINE writes
x86/ioapic: Force affinity setting in setup_ioapic_dest()
x86/paravirt: Remove the unused pv_time_ops::get_tsc_khz method
x86/ldt: Fix small LDT allocation for Xen
x86/vm86: Fix the misleading CONFIG_VM86 Kconfig help text
x86/cpu: Print family/model/stepping in hex
x86/vm86: Block non-root vm86(old) if mmap_min_addr != 0
x86/alternatives: Make optimize_nops() interrupt safe and synced
x86/mm/srat: Print non-volatile flag in SRAT
x86/cpufeatures: Enable cpuid for Intel SHA extensions
Merge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer fixes from Ingo Molnar:
"A fix for an abs()/abs64() bug that caused too slow NTP convergence on
32-bit kernels, plus a removal of an obsolete clockevents driver
facility after all users got converted during the merge window"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
clockevents: Remove unused set_mode() callback
time: Fix timekeeping_freqadjust()'s incorrect use of abs() instead of abs64()
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo MOlnar:
"Mostly tooling fixes, but also two x86 PMU driver fixes"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf tests: Fix software clock events test setting maps
perf tests: Fix task exit test setting maps
perf evlist: Fix create_syswide_maps() not propagating maps
perf evlist: Fix add() not propagating maps
perf evlist: Factor out a function to propagate maps for a single evsel
perf evlist: Make create_maps() use set_maps()
perf evlist: Make set_maps() more resilient
perf evsel: Add own_cpus member
perf evlist: Fix missing thread_map__put in propagate_maps()
perf evlist: Fix splice_list_tail() not setting evlist
perf evlist: Add has_user_cpus member
perf evlist: Remove redundant validation from propagate_maps()
perf evlist: Simplify set_maps() logic
perf evlist: Simplify propagate_maps() logic
perf top: Fix segfault pressing -> with no hist entries
perf header: Fixup reading of HEADER_NRCPUS feature
perf/x86/intel: Fix constraint access
perf/x86/intel/bts: Set event->hw.itrace_started in pmu::start to match the new logic
perf tools: Fix use of wrong event when processing exit events
perf tools: Fix parse_events_add_pmu caller
We are the client, but advertise keepalive2 anyway - for consistency,
if nothing else. In the future the server might want to know whether
its clients support keepalive2.
wraps ceph_ts into a kvec and adds it to con->out_kvec array, yet
ceph_ts becomes invalid on return from prepare_write_keepalive(). As
a result, we send out bogus keepalive2 stamps. Fix this by encoding
into a ceph_timespec member, similar to how acks are read and written.
Merge branch 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking fixes from Ingo Molnar:
"Spinlock performance regression fix, plus documentation fixes"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/static_keys: Fix up the static keys documentation
locking/qspinlock/x86: Only emit the test-and-set fallback when building guest support
locking/qspinlock/x86: Fix performance regression under unaccelerated VMs
locking/static_keys: Fix a silly typo
Will Deacon [Tue, 17 Mar 2015 12:15:02 +0000 (12:15 +0000)]
arm64: errata: add module build workaround for erratum #843419
Cortex-A53 processors <= r0p4 are affected by erratum #843419 which can
lead to a memory access using an incorrect address in certain sequences
headed by an ADRP instruction.
There is a linker fix to generate veneers for ADRP instructions, but
this doesn't work for kernel modules which are built as unlinked ELF
objects.
This patch adds a new config option for the erratum which, when enabled,
builds kernel modules with the mcmodel=large flag. This uses absolute
addressing for all kernel symbols, thereby removing the use of ADRP as
a PC-relative form of addressing. The ADRP relocs are removed from the
module loader so that we fail to load any potentially affected modules.
Cc: <stable@vger.kernel.org> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
Will Deacon [Tue, 15 Sep 2015 11:07:06 +0000 (12:07 +0100)]
arm64: compat: fix vfp save/restore across signal handlers in big-endian
When saving/restoring the VFP registers from a compat (AArch32)
signal frame, we rely on the compat registers forming a prefix of the
native register file and therefore make use of copy_{to,from}_user to
transfer between the native fpsimd_state and the compat_vfp_sigframe.
Unfortunately, this doesn't work so well in a big-endian environment.
Our fpsimd save/restore code operates directly on 128-bit quantities
(Q registers) whereas the compat_vfp_sigframe represents the registers
as an array of 64-bit (D) registers. The architecture packs the compat D
registers into the Q registers, with the least significant bytes holding
the lower register. Consequently, we need to swap the 64-bit halves when
converting between these two representations on a big-endian machine.
This patch replaces the __copy_{to,from}_user invocations in our
compat VFP signal handling code with explicit __put_user loops that
operate on 64-bit values and swap them accordingly.
Cc: <stable@vger.kernel.org> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Pull rdma driver move from Doug Ledford:
"This is a move only, no functional changes.
I tried to get it in prior to the rc1 release, but we were waiting on
IBM to get back to us that they were OK with the deprecation and
eventual removal of this driver. That OK didn't materialize until
last week, so integration and testing time pushed us beyond the rc1
release.
Summary:
- Move ehca driver to staging/rdma and schedule for deletion"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma:
IB/ehca: Deprecate driver, move to staging, schedule deletion
Merge tag 'hwmon-for-linus-v4.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon fixes from Guenter Roeck:
"Two patches for the nct6775 driver: add support for NCT6793D, and fix
swapped registers"
* tag 'hwmon-for-linus-v4.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (nct6775) Add support for NCT6793D
hwmon: (nct6775) Swap STEP_UP_TIME and STEP_DOWN_TIME registers for most chips
Merge tag 'pinctrl-v4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
Pull pin control fixes from Linus Walleij:
"This is a first set of pin control fixes for the v4.3 series. Nothing
special to say, business as usual.
- Some IS_ERR() fixes from Julia Lawall. I always wanted the
compiler to catch these but error pointers by nailing them as an
err pointer intrinsic type or something seem to be a "no can do".
In any case, cocinelle is obviously up to the task, better than
bugs staying around.
- Better error handling for NULL GPIO chips.
- Fix a compile error from the big irq desc refactoring. I'm
surprised the fallout wasn't bigger than this"
* tag 'pinctrl-v4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: samsung: s3c24xx: fix syntax error
pinctrl: core: Warn about NULL gpio_chip in pinctrl_ready_for_gpio_range()
pinctrl: join lines that can be a single line within 80 columns
pinctrl: digicolor: convert null test to IS_ERR test
pinctrl: qcom: ssbi: convert null test to IS_ERR test
Merge tag 'gpio-v4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
"This is the first round of GPIO fixes for v4.3. Quite a lot of
patches, but the influx of new stuff in the merge window was equally
big, so I'm not surprised.
- Return value checks and thus nicer errorpath for two drivers.
- Make GPIO_RCAR arch neutral.
- Propagate errors from GPIO chip ->get() vtable call. It turned out
these can actually fail sometimes, especially on slowpath
controllers doing I2C traffic and similar.
- Update documentation to be in sync with the massive changes in the
v4.3 merge window, phew.
- Handle deferred probe properly in the OMAP driver.
- Get rid of surplus MODULE_ALIAS() from sx150x"
* tag 'gpio-v4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: omap: Fix GPIO numbering for deferred probe
Documentation: gpio: Explain that <function>-gpio is also supported
gpio: omap: Fix gpiochip_add() handling for deferred probe
gpio: sx150x: Remove unnecessary MODULE_ALIAS()
Documentation: gpio: board: describe the con_id parameter
Documentation: gpio: board: add flags parameter to gpiod_get*() functions
gpio: Propagate errors from chip->get()
gpio: rcar: GPIO_RCAR doesn't relate to ARM
gpio: mxs: need to check return value of irq_alloc_generic_chip
gpio: mxc: need to check return value of irq_alloc_generic_chip
Pull crypto fixes from Herbert Xu:
"This fixes the following issues:
- The selftest overreads the IV test vector.
- Fix potential infinite loop in sunxi-ss driver.
- Fix powerpc build failure when VMX is set without VSX"
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: testmgr - don't copy from source IV too much
crypto: sunxi-ss - Fix a possible driver hang with ciphers
crypto: vmx - VMX crypto should depend on CONFIG_VSX
David Woodhouse [Wed, 16 Sep 2015 13:10:03 +0000 (14:10 +0100)]
x86/platform: Fix Geode LX timekeeping in the generic x86 build
In 2007, commit 07190a08eef36 ("Mark TSC on GeodeLX reliable")
bypassed verification of the TSC on Geode LX. However, this code
(now in the check_system_tsc_reliable() function in
arch/x86/kernel/tsc.c) was only present if CONFIG_MGEODE_LX was
set.
OpenWRT has recently started building its generic Geode target
for Geode GX, not LX, to include support for additional
platforms. This broke the timekeeping on LX-based devices,
because the TSC wasn't marked as reliable:
https://dev.openwrt.org/ticket/20531
By adding a runtime check on is_geode_lx(), we can also include
the fix if CONFIG_MGEODEGX1 or CONFIG_X86_GENERIC are set, thus
fixing the problem.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com> Cc: Andres Salomon <dilinger@queued.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Marcelo Tosatti <marcelo@kvack.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1442409003.131189.87.camel@infradead.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent
Pull perf/urgent fixes from Arnaldo Carvalho de Melo:
- Fix segfault pressing -> in 'perf top' with no hist entries. (Wang Nan)
E.g:
perf top -e page-faults --pid 11400 # 11400 generates no page-fault
- Fix propagation of thread and cpu maps, that got broken when doing incomplete
changes to better support events with a PMU cpu mask, leading to Intel PT to
fail with an error like:
$ perf record -e intel_pt//u uname
Error: The sys_perf_event_open() syscall returned with
22 (Invalid argument) for event (sched:sched_switch).
Because intel_pt adds that sched:sched_switch evsel to the evlist after the
thread/cpu maps were propagated to the evsels, fix it. (Adrian Hunter)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org>
David Woodhouse [Tue, 15 Sep 2015 15:03:36 +0000 (16:03 +0100)]
modsign: Fix GPL/OpenSSL licence incompatibility
The GPL does not permit us to link against the OpenSSL library. Use
LGPL for sign-file and extract-file instead.
[ The whole "openssl isn't compatible with gpl" is really just
fear-mongering, but there's no reason not to make modsign LGPL, so
nobody cares. - Linus ]
Reported-by: Julian Andres Klode <jak@jak-linux.org> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com> Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: Julian Andres Klode <jak@jak-linux.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Will Deacon [Wed, 2 Sep 2015 17:49:28 +0000 (18:49 +0100)]
arm64: head.S: initialise mdcr_el2 in el2_setup
When entering the kernel at EL2, we fail to initialise the MDCR_EL2
register which controls debug access and PMU capabilities at EL1.
This patch ensures that the register is initialised so that all traps
are disabled and all the PMU counters are available to the host. When a
guest is scheduled, KVM takes care to configure trapping appropriately.
Cc: <stable@vger.kernel.org> Acked-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
Adrian Hunter [Tue, 8 Sep 2015 07:59:02 +0000 (10:59 +0300)]
perf tests: Fix software clock events test setting maps
The test titled "Test software clock events have valid period values"
was setting cpu/thread maps directly. Make it use the proper function
perf_evlist__set_maps() especially now that it also propagates the maps.
Adrian Hunter [Tue, 8 Sep 2015 07:59:01 +0000 (10:59 +0300)]
perf tests: Fix task exit test setting maps
The test titled "Test number of exit event of a simple workload" was
setting cpu/thread maps directly. Make it use the proper function
perf_evlist__set_maps() especially now that it also propagates the maps.
Adrian Hunter [Tue, 8 Sep 2015 07:58:59 +0000 (10:58 +0300)]
perf evlist: Fix add() not propagating maps
If evsels are added after maps are created, then they won't have any
maps propagated to them. Fix that.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@intel.com> Link: http://lkml.kernel.org/r/1441699142-18905-12-git-send-email-adrian.hunter@intel.com
[ Moved the moving of propagate_maps() to the patch before, so that this
one does _just_ the one lile fix calling in add()] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adrian Hunter [Tue, 8 Sep 2015 07:58:58 +0000 (10:58 +0300)]
perf evlist: Factor out a function to propagate maps for a single evsel
Subsequent fixes will need a function that just propagates maps for a
single evsel so factor it out.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@intel.com> Link: http://lkml.kernel.org/r/1441699142-18905-11-git-send-email-adrian.hunter@intel.com
[ Moved them to before perf_evlist__add() to avoid having to move it in the next patch ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adrian Hunter [Tue, 8 Sep 2015 07:58:56 +0000 (10:58 +0300)]
perf evlist: Make set_maps() more resilient
Make perf_evlist__set_maps() more resilient by allowing for the
possibility that one or another of the maps isn't being changed and
therefore should not be "put".
Adrian Hunter [Tue, 8 Sep 2015 07:58:55 +0000 (10:58 +0300)]
perf evsel: Add own_cpus member
perf_evlist__propagate_maps() cannot easily tell if an evsel has its own
cpu map. To make that simpler, keep a copy of the PMU cpu map and
adjust the propagation logic accordingly.
Adrian Hunter [Tue, 8 Sep 2015 07:58:54 +0000 (10:58 +0300)]
perf evlist: Fix missing thread_map__put in propagate_maps()
perf_evlist__propagate_maps() incorrectly assumes evsel->threads is NULL
before reassigning it, but it won't be NULL when perf_evlist__set_maps()
is used to set different (or NULL) maps. Thus thread_map__put must be
used, which works even if evsel->threads is NULL.
Adrian Hunter [Tue, 8 Sep 2015 07:58:53 +0000 (10:58 +0300)]
perf evlist: Fix splice_list_tail() not setting evlist
Commit d49e46950772 ("perf evsel: Add a backpointer to the evlist a
evsel is in") updated perf_evlist__add() but not
perf_evlist__splice_list_tail().
This illustrates that it is better if perf_evlist__splice_list_tail()
calls perf_evlist__add() instead of duplicating the logic, so do that.
This will also simplify a subsequent fix for propagating maps.
Adrian Hunter [Tue, 8 Sep 2015 07:58:52 +0000 (10:58 +0300)]
perf evlist: Add has_user_cpus member
Subsequent patches will need to call perf_evlist__propagate_maps without
reference to a "target". Add evlist->has_user_cpus to record whether
the user has specified which cpus to target (and therefore whether that
list of cpus should override the default settings for a selected event
i.e. the cpu maps should be propagated)
Adrian Hunter [Tue, 8 Sep 2015 07:58:51 +0000 (10:58 +0300)]
perf evlist: Remove redundant validation from propagate_maps()
The validation checks that the values that were just assigned, got
assigned i.e. the error can't ever happen. Subsequent patches will call
this code in places where errors are not being returned. Changing those
code paths to return this non-existent error is counter-productive, so
just remove it.
That in turn results in perf_evlist__set_maps not needing to return an
error, but callers aren't checking it either, so remove that too.
Adrian Hunter [Tue, 8 Sep 2015 07:58:49 +0000 (10:58 +0300)]
perf evlist: Simplify propagate_maps() logic
If evsel->cpus is to be reassigned then the current value must be "put",
which works even if it is NULL. Simplify the current logic by moving
the "put" next to the assignment.
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"A couple build fixes for drivers introduced in the merge window and a
handful of patches to add more critical clocks on rockchip SoCs that
are affected by newly introduced gpio clock handling"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: rockchip: Add pclk_peri to critical clocks on RK3066/RK3188
clk: rockchip: add pclk_cpu to the list of rk3188 critical clocks
clk: rockchip: handle critical clocks after registering all clocks
clk: Hi6220: separately build stub clock driver
clk: h8s2678: Fix compile error
Merge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6
Pull CIFS fixes from Steve French:
"Two small cifs fixes"
* 'for-next' of git://git.samba.org/sfrench/cifs-2.6:
[CIFS] mount option sec=none not displayed properly in /proc/mounts
CIFS: fix type confusion in copy offload ioctl
Merge branch 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm
Pull ARM fixes from Russell King:
"A number of fixes for the merge window, fixing a number of cases
missed when testing the uaccess code, particularly cases which only
show up with certain compiler versions"
* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
ARM: 8431/1: fix alignement of __bug_table section entries
arm/xen: Enable user access to the kernel before issuing a privcmd call
ARM: domains: add memory dependencies to get_domain/set_domain
ARM: domains: thread_info.h no longer needs asm/domains.h
ARM: uaccess: fix undefined instruction on ARMv7M/noMMU
ARM: uaccess: remove unneeded uaccess_save_and_disable macro
ARM: swpan: fix nwfpe for uaccess changes
ARM: 8429/1: disable GCC SRA optimization
The bug resides in perf_evsel__hists_browse() that, in the above
circumstance browser->selection can be NULL, but code after
skip_annotation doesn't consider it.
This patch fix it by checking browser->selection before fetching
browser->selection->map.
Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Zefan Li <lizefan@huawei.com> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1442226235-117265-1-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Shaohua Li [Thu, 30 Jul 2015 23:24:43 +0000 (16:24 -0700)]
x86/apic: Serialize LVTT and TSC_DEADLINE writes
The APIC LVTT register is MMIO mapped but the TSC_DEADLINE register is an
MSR. The write to the TSC_DEADLINE MSR is not serializing, so it's not
guaranteed that the write to LVTT has reached the APIC before the
TSC_DEADLINE MSR is written. In such a case the write to the MSR is
ignored and as a consequence the local timer interrupt never fires.
The SDM decribes this issue for xAPIC and x2APIC modes. The
serialization methods recommended by the SDM differ.
xAPIC:
"1. Memory-mapped write to LVT Timer Register, setting bits 18:17 to 10b.
2. WRMSR to the IA32_TSC_DEADLINE MSR a value much larger than current time-stamp counter.
3. If RDMSR of the IA32_TSC_DEADLINE MSR returns zero, go to step 2.
4. WRMSR to the IA32_TSC_DEADLINE MSR the desired deadline."
x2APIC:
"To allow for efficient access to the APIC registers in x2APIC mode,
the serializing semantics of WRMSR are relaxed when writing to the
APIC registers. Thus, system software should not use 'WRMSR to APIC
registers in x2APIC mode' as a serializing instruction. Read and write
accesses to the APIC registers will occur in program order. A WRMSR to
an APIC register may complete before all preceding stores are globally
visible; software can prevent this by inserting a serializing
instruction, an SFENCE, or an MFENCE before the WRMSR."
The xAPIC method is to just wait for the memory mapped write to hit
the LVTT by checking whether the MSR write has reached the hardware.
There is no reason why a proper MFENCE after the memory mapped write would
not do the same. Andi Kleen confirmed that MFENCE is sufficient for the
xAPIC case as well.
Issue MFENCE before writing to the TSC_DEADLINE MSR. This can be done
unconditionally as all CPUs which have TSC_DEADLINE also have MFENCE
support.
[ tglx: Massaged the changelog ]
Signed-off-by: Shaohua Li <shli@fb.com> Reviewed-by: Ingo Molnar <mingo@kernel.org> Cc: <Kernel-team@fb.com> Cc: <lenb@kernel.org> Cc: <fenghua.yu@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: stable@vger.kernel.org #v3.7+ Link: http://lkml.kernel.org/r/20150909041352.GA2059853@devbig257.prn2.facebook.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Mon, 14 Sep 2015 10:00:55 +0000 (12:00 +0200)]
x86/ioapic: Force affinity setting in setup_ioapic_dest()
The recent ioapic cleanups changed the affinity setting in
setup_ioapic_dest() from a direct write to the hardware to the delayed
affinity setup via irq_set_affinity().
That results in a warning from chained_irq_exit():
WARNING: CPU: 0 PID: 5 at kernel/irq/migration.c:32 irq_move_masked_irq
[<ffffffff810a0a88>] irq_move_masked_irq+0xb8/0xc0
[<ffffffff8103c161>] ioapic_ack_level+0x111/0x130
[<ffffffff812bbfe8>] intel_gpio_irq_handler+0x148/0x1c0
The reason is that irq_set_affinity() does not write directly to the
hardware. It marks the affinity setting as pending and executes it
from the next interrupt. The chained handler infrastructure does not
take the irq descriptor lock for performance reasons because such a
chained interrupt is not visible to any interfaces. So the delayed
affinity setting triggers the warning in irq_move_masked_irq().
Restore the old behaviour by calling the set_affinity function of the
ioapic chip in setup_ioapic_dest(). This is safe as none of the
interrupts can be on the fly at this point.
Fixes: aa5cb97f14a2 'x86/irq: Remove x86_io_apic_ops.set_affinity and related interfaces' Reported-and-tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Jiang Liu <jiang.liu@linux.intel.com> Cc: jarkko.nikula@linux.intel.com
Will Deacon [Fri, 11 Sep 2015 17:22:02 +0000 (18:22 +0100)]
arm64: pgtable: use a single bit for PTE_WRITE regardless of DBM
Depending on CONFIG_ARM64_HW_AFDBM, we use either bit 57 or 51 of the
pte to represent PTE_WRITE. Given that bit 51 is reserved prior to
ARMv8.1, we can just use that bit regardless of the config option. That
also matches what happens if a kernel configured with ARM64_HW_AFDBM=y
is run on a CPU without the DBM functionality.
arm64: Fix pte_modify() to preserve the hardware dirty information
The pte_modify() function with hardware AF/DBM enabled must transfer the
hardware dirty information to the software PTE_DIRTY bit. However, it
was setting this bit in newprot and the mask does not cover such bit.
This patch sets PTE_DIRTY on the original pte which will be preserved in
the returned value.
Fixes: 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits") Cc: Julien Grall <julien.grall@citrix.com> Tested-by: Julien Grall <julien.grall@citrix.com> Tested-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
arm64: Fix the pte_hw_dirty() check when AF/DBM is enabled
Commit 2f4b829c625e ("arm64: Add support for hardware updates of the
access and dirty pte bits") introduced support for handling hardware
updates of the access flag and dirty status. The PTE is automatically
dirtied in hardware (if supported) by clearing the PTE_RDONLY bit when
the PTE_DBM/PTE_WRITE bit is set. The pte_hw_dirty() macro was added to
detect a hardware dirtied pte. The pte_dirty() macro checks for both
software PTE_DIRTY and pte_hw_dirty().
Functions like pte_modify() clear the PTE_RDONLY bit since it is meant
to be set in set_pte_at() when written to memory. In such cases,
pte_hw_dirty() would return true even though such pte is clean. This
patch changes pte_hw_dirty() to test the PTE_DBM/PTE_WRITE bit together
with PTE_RDONLY.
Fixes: 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits") Reported-by: Julien Grall <julien.grall@citrix.com> Tested-by: Julien Grall <julien.grall@citrix.com> Tested-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
arm64: dma-mapping: check whether cma area is initialized or not
If CMA is turned on and CMA size is set to zero, kernel should
behave as if CMA was not enabled at compile time.
Every dma allocation should check existence of cma area
before requesting memory.
Arm has done this by commit e464ef16c4f0 ("arm: dma-mapping: add
checking cma area initialized"), also do this for arm64.
Jan Beulich [Wed, 2 Sep 2015 15:45:58 +0000 (09:45 -0600)]
x86/ldt: Fix small LDT allocation for Xen
While the following commit:
37868fe113 ("x86/ldt: Make modify_ldt synchronous")
added a nice comment explaining that Xen needs page-aligned
whole page chunks for guest descriptor tables, it then
nevertheless used kzalloc() on the small size path.
As I'm unaware of guarantees for kmalloc(PAGE_SIZE, ) to return
page-aligned memory blocks, I believe this needs to be switched
back to __get_free_page() (or better get_zeroed_page()).
Signed-off-by: Jan Beulich <jbeulich@suse.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: David Vrabel <david.vrabel@citrix.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> 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/55E735D6020000780009F1E6@prv-mh.provo.novell.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent
Pull perf/urgent fix from Arnaldo Carvalho de Melo:
- The values of _SC_NPROCESSORS_CONF and _SC_NPROCESSORS_ONLN (sysconf(3)) were
being read from perf.data files in the inverse order they are written, fix it.
(Arnaldo Carvalho de Melo)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tony Lindgren [Thu, 3 Sep 2015 17:34:30 +0000 (10:34 -0700)]
pinctrl: core: Warn about NULL gpio_chip in pinctrl_ready_for_gpio_range()
If the gpio driver is confused about the numbers for gpio-ranges,
pinctrl_ready_for_gpio_range() may get called with invalid GPIO
causing a NULL pointer exception. Let's instead provide a warning
that allows fixing the problem and return with error.
Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Julia Lawall [Mon, 24 Aug 2015 21:12:27 +0000 (23:12 +0200)]
pinctrl: digicolor: convert null test to IS_ERR test
Since commit 323de9efdf3e ("pinctrl: make pinctrl_register() return proper
error code"), pinctrl_register returns an error code rather than NULL on
failure. Update a driver that was introduced more recently.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@@
expression e,e1,e2;
@@
e = pinctrl_register(...)
... when != e = e1
if (
- e == NULL
+ IS_ERR(e)
) {
...
return
- e2
+ PTR_ERR(e)
;
}
// </smpl>
Julia Lawall [Mon, 24 Aug 2015 21:12:26 +0000 (23:12 +0200)]
pinctrl: qcom: ssbi: convert null test to IS_ERR test
Since commit 323de9efdf3e ("pinctrl: make pinctrl_register() return proper
error code"), pinctrl_register returns an error code rather than NULL on
failure. Update some drivers that were introduced more recently.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@@
expression e,e1,e2;
@@
e = pinctrl_register(...)
... when != e = e1
if (
- e == NULL
+ IS_ERR(e)
) {
...
return
- e2
+ PTR_ERR(e)
;
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tony Lindgren [Thu, 3 Sep 2015 17:31:27 +0000 (10:31 -0700)]
gpio: omap: Fix GPIO numbering for deferred probe
If gpio-omap probe fails with -EPROBE_DEFER, the GPIO numbering
keeps increasing. Only increase the gpio count if gpiochip_add()
was successful as otherwise the numbers will increase for each
probe attempt.
Cc: Javier Martinez Canillas <javier@dowhile0.org> Cc: Kevin Hilman <khilman@deeprootsystems.com> Cc: Santosh Shilimkar <ssantosh@kernel.org> Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Documentation: gpio: Explain that <function>-gpio is also supported
The GPIO documentation mentions that GPIOs are mapped by defining a
<function>-gpios property in the consumer device's node but a -gpio
sufix is also supported after commit:
dd34c37aa3e8 ("gpio: of: Allow -gpio suffix for property names")
Update the documentation to match the implementation.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Documentation: gpio: board: describe the con_id parameter
The con_id parameter has to match the GPIO description and is automatically
extended by the GPIO suffix if not NULL. I had to look into the code to
understand this and properly find the GPIO I've been looking for, so document
this.
Documentation: gpio: board: add flags parameter to gpiod_get*() functions
With commit 39b2bbe3d715 ("gpio: add flags argument to gpiod_get*()
functions") the gpiod_get*() functions got a 'flags' parameter. Reflect
this in the documentation, too.
Bjorn Andersson [Fri, 28 Aug 2015 16:44:18 +0000 (09:44 -0700)]
gpio: Propagate errors from chip->get()
It's possible to have gpio chips hanging off unreliable remote buses
where the get() operation will fail to acquire a readout of the current
gpio state. Propagate these errors to the consumer so that they can
act on, retry or ignore these failing reads, instead of treating them as
the line being held high.
8cd1470("gpio: rcar: Add r8a7795 (R-Car H3) support") added
GPIO support for r8a7795. r8a7795 based on CONFIG_ARM64.
OTOH, GPIO_RCAR driver can be compiled fine on non-ARM.
This patch removed ARM dependency for it.
Peng Fan [Sun, 23 Aug 2015 13:11:53 +0000 (21:11 +0800)]
gpio: mxs: need to check return value of irq_alloc_generic_chip
Need to check return value of irq_alloc_generic_chip, because
it may return NULL.
1. Change mxs_gpio_init_gc return type from void to int.
2. Add a new lable out_irqdomain_remove to remove the irq domain
when mxc_gpio_init_gc fail.
Peng Fan [Sun, 23 Aug 2015 13:11:52 +0000 (21:11 +0800)]
gpio: mxc: need to check return value of irq_alloc_generic_chip
Need to check return value of irq_alloc_generic_chip, because
it may return NULL.
1. Change mxc_gpio_init_gc return type from void to int.
2. Add a new lable out_irqdomain_remove to remove the irq domain
when mxc_gpio_init_gc fail.
perf header: Fixup reading of HEADER_NRCPUS feature
The original patch introducing this header wrote the number of CPUs available
and online in one order and then swapped those values when reading, fix it.
John Stultz [Wed, 9 Sep 2015 23:07:30 +0000 (16:07 -0700)]
time: Fix timekeeping_freqadjust()'s incorrect use of abs() instead of abs64()
The internal clocksteering done for fine-grained error
correction uses a logarithmic approximation, so any time
adjtimex() adjusts the clock steering, timekeeping_freqadjust()
quickly approximates the correct clock frequency over a series
of ticks.
Unfortunately, the logic in timekeeping_freqadjust(), introduced
in commit:
dc491596f639 ("timekeeping: Rework frequency adjustments to work better w/ nohz")
used the abs() function with a s64 error value to calculate the
size of the approximated adjustment to be made.
Per include/linux/kernel.h:
"abs() should not be used for 64-bit types (s64, u64, long long) - use abs64()".
Thus on 32-bit platforms, this resulted in the clocksteering to
take a quite dampended random walk trying to converge on the
proper frequency, which caused the adjustments to be made much
slower then intended (most easily observed when large
adjustments are made).
This patch fixes the issue by using abs64() instead.
Reported-by: Nuno Gonçalves <nunojpg@gmail.com> Tested-by: Nuno Goncalves <nunojpg@gmail.com> Signed-off-by: John Stultz <john.stultz@linaro.org> Cc: <stable@vger.kernel.org> # v3.17+ Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Miroslav Lichvar <mlichvar@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Richard Cochran <richardcochran@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1441840051-20244-1-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Peter Zijlstra [Wed, 9 Sep 2015 07:06:17 +0000 (09:06 +0200)]
sched/fair: Optimize per entity utilization tracking
Currently the load_{sum,avg} and util_{sum,avg} tracking is asymmetric
in that load tracking gets a 2^10 unit from the weight, but util gets
no such factor.
This results in more lost bits for util scaling and asymmetric scaling
rules.
Fix this by removing shifts, such that we gain the 2^10 factor from
scaling. There is no risk of overflowing the u32 as the max value is
now LOAD_AVG_MAX << 10, which is still well below UINT_MAX.
This further entangles the assumption that both LOAD and CAPACITY
shifts are the same (and 10) so put in an assertion for that.
This fixes the math for the LOAD_RESOLUTION != 0 case.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org>
Which is silly; when made unsigned int, GCC recognises this as
pointless ops and fails to emit them (confirmed on 4.9.3 and 5.1.1).
Furthermore, afaict unsigned is actually the correct type for these
fields anyway, as we've explicitly ruled out negative delta's earlier
in this function.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Peter Zijlstra [Mon, 7 Sep 2015 13:05:42 +0000 (15:05 +0200)]
sched/fair: Rename scale() to cap_scale()
Rename scale() to cap_scale() to better reflect its purpose, it is
after all not a general purpose scale function, it has
SCHED_CAPACITY_SHIFT hardcoded in it.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Morten Rasmussen [Fri, 14 Aug 2015 16:23:14 +0000 (17:23 +0100)]
sched/fair: Initialize task load and utilization before placing task on rq
Task load or utilization is not currently considered in
select_task_rq_fair(), but if we want that in the future we should make
sure it is not zero for new tasks.
Dietmar Eggemann [Fri, 14 Aug 2015 16:23:13 +0000 (17:23 +0100)]
sched/fair: Get rid of scaling utilization by capacity_orig
Utilization is currently scaled by capacity_orig, but since we now have
frequency and cpu invariant cfs_rq.avg.util_avg, frequency and cpu scaling
now happens as part of the utilization tracking itself.
So cfs_rq.avg.util_avg should no longer be scaled in cpu_util().
Dietmar Eggemann [Fri, 14 Aug 2015 16:23:12 +0000 (17:23 +0100)]
sched/fair: Name utilization related data and functions consistently
Use the advent of the per-entity load tracking rewrite to streamline the
naming of utilization related data and functions by using
{prefix_}util{_suffix} consistently. Moreover call both signals
({se,cfs}.avg.util_avg) utilization.
Dietmar Eggemann [Fri, 14 Aug 2015 23:04:41 +0000 (00:04 +0100)]
sched/fair: Make utilization tracking CPU scale-invariant
Besides the existing frequency scale-invariance correction factor, apply
CPU scale-invariance correction factor to utilization tracking to
compensate for any differences in compute capacity. This could be due to
micro-architectural differences (i.e. instructions per seconds) between
cpus in HMP systems (e.g. big.LITTLE), and/or differences in the current
maximum frequency supported by individual cpus in SMP systems. In the
existing implementation utilization isn't comparable between cpus as it
is relative to the capacity of each individual CPU.
Each segment of the sched_avg.util_sum geometric series is now scaled
by the CPU performance factor too so the sched_avg.util_avg of each
sched entity will be invariant from the particular CPU of the HMP/SMP
system on which the sched entity is scheduled.
With this patch, the utilization of a CPU stays relative to the max CPU
performance of the fastest CPU in the system.
In contrast to utilization (sched_avg.util_sum), load
(sched_avg.load_sum) should not be scaled by compute capacity. The
utilization metric is based on running time which only makes sense when
cpus are _not_ fully utilized (utilization cannot go beyond 100% even if
more tasks are added), where load is runnable time which isn't limited
by the capacity of the CPU and therefore is a better metric for
overloaded scenarios. If we run two nice-0 busy loops on two cpus with
different compute capacity their load should be similar since their
compute demands are the same. We have to assume that the compute demand
of any task running on a fully utilized CPU (no spare cycles = 100%
utilization) is high and the same no matter of the compute capacity of
its current CPU, hence we shouldn't scale load by CPU capacity.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/55CE7409.1000700@arm.com Signed-off-by: Ingo Molnar <mingo@kernel.org>