- 13 Dec, 2016 22 commits
-
-
Greg Thelen authored
Reading /proc/slabinfo or monitoring slabtop(1) can become very expensive if there are many slab caches and if there are very lengthy per-node partial and/or free lists. Commit 07a63c41 ("mm/slab: improve performance of gathering slabinfo stats") addressed the per-node full lists which showed a significant improvement when no objects were freed. This patch has the same motivation and optimizes the remainder of the usecases where there are very lengthy partial and free lists. This patch maintains per-node active_slabs (full and partial) and free_slabs rather than iterating the lists at runtime when reading /proc/slabinfo. When allocating 100GB of slab from a test cache where every slab page is on the partial list, reading /proc/slabinfo (includes all other slab caches on the system) takes ~247ms on average with 48 samples. As a result of this patch, the same read takes ~0.856ms on average. [rientjes@google.com: changelog] Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1611081505240.13403@chino.kir.corp.google.comSigned-off-by:
Greg Thelen <gthelen@google.com> Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Thomas Garnier authored
Verify that kmem_create_cache flags are not allocator specific. It is done before removing flags that are not available with the current configuration. The current kmem_cache_create removes incorrect flags but do not validate the callers are using them right. This change will ensure that callers are not trying to create caches with flags that won't be used because allocator specific. Link: http://lkml.kernel.org/r/1478553075-120242-2-git-send-email-thgarnie@google.comSigned-off-by:
Thomas Garnier <thgarnie@google.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by:
-
Arnd Bergmann authored
The slub allocator gives us some incorrect warnings when CONFIG_PROFILE_ANNOTATED_BRANCHES is set, as the unlikely() macro prevents it from seeing that the return code matches what it was before: mm/slub.c: In function `kmem_cache_free_bulk': mm/slub.c:262:23: error: `df.s' may be used uninitialized in this function [-Werror=maybe-uninitialized] mm/slub.c:2943:3: error: `df.cnt' may be used uninitialized in this function [-Werror=maybe-uninitialized] mm/slub.c:2933:4470: error: `df.freelist' may be used uninitialized in this function [-Werror=maybe-uninitialized] mm/slub.c:2943:3: error: `df.tail' may be used uninitialized in this function [-Werror=maybe-uninitialized] I have not been able to come up with a perfect way for dealing with this, the three options I see are: - add a bogus initialization, which would increase the runtime overhead - replace unlikely() with unlikely_notrace() - remove the unlikely() annotation completely I checked the object code for a typical x86 configuration and the last two cases produce the same result, so I went for the last one, which is the simplest. Link: http://lkml.kernel.org/r/20161024155704.3114445-1-arnd@arndb.deSigned-off-by:
Arnd Bergmann <arnd@arndb.de> Acked-by:
Jesper Dangaard Brouer <brouer@redhat.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Laura Abbott <labbott@fedoraproject.org> Cc: Alexander Potapenko <glider@google.com> Signed-off-by:
-
Vladimir Davydov authored
synchronize_sched() is a heavy operation and calling it per each cache owned by a memory cgroup being destroyed may take quite some time. What is worse, it's currently called under the slab_mutex, stalling all works doing cache creation/destruction. Actually, there isn't much point in calling synchronize_sched() for each cache - it's enough to call it just once - after setting cpu_partial for all caches and before shrinking them. This way, we can also move it out of the slab_mutex, which we have to hold for iterating over the slab cache list. Link: https://bugzilla.kernel.org/show_bug.cgi?id=172991 Link: http://lkml.kernel.org/r/0a10d71ecae3db00fb4421bcd3f82bcc911f4be4.1475329751.git.vdavydov.dev@gmail.comSigned-off-by:
Vladimir Davydov <vdavydov.dev@gmail.com> Reported-by:
Doug Smythies <dsmythies@telus.net> Acked-by:
Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by:
-
Vladimir Davydov authored
Creating a lot of cgroups at the same time might stall all worker threads with kmem cache creation works, because kmem cache creation is done with the slab_mutex held. The problem was amplified by commits 801faf0d ("mm/slab: lockless decision to grow cache") in case of SLAB and 81ae6d03 ("mm/slub.c: replace kick_all_cpus_sync() with synchronize_sched() in kmem_cache_shrink()") in case of SLUB, which increased the maximal time the slab_mutex can be held. To prevent that from happening, let's use a special ordered single threaded workqueue for kmem cache creation. This shouldn't introduce any functional changes regarding how kmem caches are created, as the work function holds the global slab_mutex during its whole runtime anyway, making it impossible to run more than one work at a time. By using a single threaded workqueue, we just avoid creating a thread per each work. Ordering is required to avoid a situation when a cgroup's work is put off indefinitely because there are other cgroups to serve, in other words to guarantee fairness. Link: https://bugzilla.kernel.org/show_bug.cgi?id=172981 Link: http://lkml.kernel.org/r/20161004131417.GC1862@esperanzaSigned-off-by:
Michal Hocko <mhocko@suse.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by:
-
Deepa Dinamani authored
CURRENT_TIME is not y2038 safe. Use y2038 safe ktime_get_real_seconds() here for timestamps. struct heartbeat_block's hb_seq and deletetion time are already 64 bits wide and accommodate times beyond y2038. Also use y2038 safe ktime_get_real_ts64() for on disk inode timestamps. These are also wide enough to accommodate time64_t. Link: http://lkml.kernel.org/r/1475365298-29236-1-git-send-email-deepa.kernel@gmail.comSigned-off-by:
Deepa Dinamani <deepa.kernel@gmail.com> Reviewed-by:
-
Deepa Dinamani authored
struct timespec is not y2038 safe. Use time64_t which is y2038 safe to represent orphan scan times. time64_t is sufficient here as only the seconds delta times are relevant. Also use appropriate time functions that return time in time64_t format. Time functions now return monotonic time instead of real time as only delta scan times are relevant and these values are not persistent across reboots. The format string for the debug print is still using long as this is only the time elapsed since the last scan and long is sufficient to represent this value. Link: http://lkml.kernel.org/r/1475365138-20567-1-git-send-email-deepa.kernel@gmail.comSigned-off-by:
-
Ashish Samant authored
In ocfs2_lock_refcount_tree, if ocfs2_read_refcount_block() returns an error, we do ocfs2_refcount_tree_put twice (once in ocfs2_unlock_refcount_tree and once outside it), thereby reducing the refcount of the refcount tree twice, but we dont delete the tree in this case. This will make refcnt of the tree = 0 and the ocfs2_refcount_tree_put will eventually call ocfs2_mark_lockres_freeing, setting OCFS2_LOCK_FREEING for the refcount_tree->rf_lockres. The error returned by ocfs2_read_refcount_block is propagated all the way back and for next iteration of write, ocfs2_lock_refcount_tree gets the same tree back from ocfs2_get_refcount_tree because we havent deleted the tree. Now we have the same tree, but OCFS2_LOCK_FREEING is set for rf_lockres and eventually, when _ocfs2_lock_refcount_tree is called in this iteration, BUG_ON( __ocfs2_cluster_lock:1395 ERROR: Cluster lock called on freeing lockres T00000000000000000386019775b08d! flags 0x81) is triggerred. Call stack: (loop16,11155,0):ocfs2_lock_refcount_tree:482 ERROR: status = -5 (loop16,11155,0):ocfs2_refcount_cow_hunk:3497 ERROR: status = -5 (loop16,11155,0):ocfs2_refcount_cow:3560 ERROR: status = -5 (loop16,11155,0):ocfs2_prepare_inode_for_refcount:2111 ERROR: status = -5 (loop16,11155,0):ocfs2_prepare_inode_for_write:2190 ERROR: status = -5 (loop16,11155,0):ocfs2_file_write_iter:2331 ERROR: status = -5 (loop16,11155,0):__ocfs2_cluster_lock:1395 ERROR: bug expression: lockres->l_flags & OCFS2_LOCK_FREEING (loop16,11155,0):__ocfs2_cluster_lock:1395 ERROR: Cluster lock called on freeing lockres T00000000000000000386019775b08d! flags 0x81 kernel BUG at fs/ocfs2/dlmglue.c:1395! invalid opcode: 0000 [#1] SMP CPU 0 Modules linked in: tun ocfs2 jbd2 xen_blkback xen_netback xen_gntdev .. sd_mod crc_t10dif ext3 jbd mbcache RIP: __ocfs2_cluster_lock+0x31c/0x740 [ocfs2] RSP: e02b:ffff88017c0138a0 EFLAGS: 00010086 Process loop16 (pid: 11155, threadinfo ffff88017c010000, task ffff8801b5374300) Call Trace: ocfs2_refcount_lock+0xae/0x130 [ocfs2] __ocfs2_lock_refcount_tree+0x29/0xe0 [ocfs2] ocfs2_lock_refcount_tree+0xdd/0x320 [ocfs2] ocfs2_refcount_cow_hunk+0x1cb/0x440 [ocfs2] ocfs2_refcount_cow+0xa9/0x1d0 [ocfs2] ocfs2_prepare_inode_for_refcount+0x115/0x200 [ocfs2] ocfs2_prepare_inode_for_write+0x33b/0x470 [ocfs2] ocfs2_file_write_iter+0x220/0x8c0 [ocfs2] aio_write_iter+0x2e/0x30 Fix this by avoiding the second call to ocfs2_refcount_tree_put() Link: http://lkml.kernel.org/r/1473984404-32011-1-git-send-email-ashish.samant@oracle.comSigned-off-by:Ashish Samant <ashish.samant@oracle.com> Reviewed-by:
Eric Ren <zren@suse.com> Cc: Mark Fasheh <mfasheh@versity.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Signed-off-by:
-
piaojun authored
'page' parameter in ocfs2_write_end_nolock() is never used. Link: http://lkml.kernel.org/r/582FD91A.5000902@huawei.comSigned-off-by:
Jun Piao <piaojun@huawei.com> Reviewed-by:
Joseph Qi <jiangqi903@gmail.com> Cc: Mark Fasheh <mfasheh@versity.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Signed-off-by:
-
piaojun authored
When 'dispatch_assert' is set, 'response' must be DLM_MASTER_RESP_YES, and 'res' won't be null, so execution can't reach these two branch. Link: http://lkml.kernel.org/r/58174C91.3040004@huawei.comSigned-off-by:
-
Guozhonghua authored
The variable `set_maybe' is redundant when the mle has been found in the map. So it is ok to set the node_idx into mle's maybe_map directly. Link: http://lkml.kernel.org/r/71604351584F6A4EBAE558C676F37CA4A3D490DD@H3CMLB12-EX.srv.huawei-3com.comSigned-off-by:
Guozhonghua <guozhonghua@h3c.com> Reviewed-by:
Mark Fasheh <mfasheh@versity.com> Reviewed-by:
-
piaojun authored
The value of 'stage' must be between 1 and 2, so the switch can't reach the default case. Link: http://lkml.kernel.org/r/57FB5EB2.7050002@huawei.comSigned-off-by:
-
Sudip Mukherjee authored
If request_irq() fails it passes the error to the caller. The caller now checks it and jumps to the common error path on failure. Link: http://lkml.kernel.org/r/1474237304-897-3-git-send-email-sudipm.mukherjee@gmail.comSigned-off-by:
Sudip Mukherjee <sudip.mukherjee@codethink.co.uk> Signed-off-by:
-
Sudip Mukherjee authored
Use a common error path for the failure. Link: http://lkml.kernel.org/r/1474237304-897-2-git-send-email-sudipm.mukherjee@gmail.comSigned-off-by:
-
Sudip Mukherjee authored
While building m32r allmodconfig we were getting warning: drivers/pcmcia/m32r_pcc.c:331:2: warning: ignoring return value of 'request_irq', declared with attribute warn_unused_result request_irq() can fail and we should always be checking the result from it. Check the result and return it to the caller. Link: http://lkml.kernel.org/r/1474237304-897-1-git-send-email-sudipm.mukherjee@gmail.comSigned-off-by:
-
Sudip Mukherjee authored
While building m32r defconfig we got warnings: arch/m32r/platforms/m32700ut/setup.c:249:24: warning: 'm32700ut_lcdpld_irq_type' defined but not used [-Wunused-variable] m32700ut_lcdpld_irq_type is only used when CONFIG_USB is enabled. Modify the code to declare the related variables and functions only when CONFIG_USB is enabled. Link: http://lkml.kernel.org/r/1479244406-7507-1-git-send-email-sudipm.mukherjee@gmail.comSigned-off-by:
-
Sudip Mukherjee authored
Some builds of m32r were failing as it tried to build few drivers which needed dma but m32r is not having dma support. Objections were raised when it was tried to make those drivers depend on HAS_DMA. So the next best thing is to add dma support to m32r. dma_noop is a very simple dma with 1:1 memory mapping. Link: http://lkml.kernel.org/r/1475949198-31623-1-git-send-email-sudipm.mukherjee@gmail.comSigned-off-by:
-
Sam Protsenko authored
When SUBARCH is "omap1" or "omap2", plat-omap/ directory must be indexed. Handle this special case properly. While at it, check if mach- directory exists at all. Link: http://lkml.kernel.org/r/20161202122148.15001-1-joe.skb7@gmail.comSigned-off-by:
Sam Protsenko <semen.protsenko@linaro.org> Cc: Michal Marek <mmarek@suse.com> Signed-off-by:
-
Alexey Dobriyan authored
Every often used regex is better be compiled in Python. Speedup is about ~9.8% (whee!) $ perf stat -r 16 taskset -c 15 ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux >/dev/null 7.091202853 seconds time elapsed ( +- 0.15% ) +re.compile 6.397564973 seconds time elapsed ( +- 0.34% ) Link: http://lkml.kernel.org/r/20161119004417.GB1200@avx2Signed-off-by:Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Alexey Dobriyan authored
readlines() conses whole list before doing anything which is slower for big object files. Use per line iterator. Speed up is ~2% on "allyesconfig" type of kernel. $ perf stat -r 16 taskset -c 15 ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux >/dev/null ... Before: 7.247708646 seconds time elapsed ( +- 0.28% ) After: 7.091202853 seconds time elapsed ( +- 0.15% ) Link: http://lkml.kernel.org/r/20161119004143.GA1200@avx2Signed-off-by: -
Stanislav Kinsburskiy authored
This limitation came with the reason to remove "another way for malicious code to obscure a compromised program and masquerade as a benign process" by allowing "security-concious program can use this prctl once during its early initialization to ensure the prctl cannot later be abused for this purpose": http://marc.info/?l=linux-kernel&m=133160684517468&w=2 This explanation doesn't look sufficient. The only thing "exe" link is indicating is the file, used to execve, which is basically nothing and not reliable immediately after process has returned from execve system call. Moreover, to use this feture, all the mappings to previous exe file have to be unmapped and all the new exe file permissions must be satisfied. Which means, that changing exe link is very similar to calling execve on the binary. The need to remove this limitations comes from migration of NFS mount point, which is not accessible during restore and replaced by other file system. Because of this exe link has to be changed twice. [akpm@linux-foundation.org: fix up comment] Link: http://lkml.kernel.org/r/20160927153755.9337.69650.stgit@localhost.localdomainSigned-off-by:Stanislav Kinsburskiy <skinsbursky@virtuozzo.com> Acked-by:
Oleg Nesterov <oleg@redhat.com> Acked-by:
Cyrill Gorcunov <gorcunov@openvz.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Kees Cook <keescook@chromium.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: John Stultz <john.stultz@linaro.org> Cc: Matt Helsley <matthltc@us.ibm.com> Cc: Pavel Emelyanov <xemul@virtuozzo.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by:
-
Nicolas Iooss authored
When commit fbae2d44 ("kthread: add kthread_create_worker*()") introduced some kthread_create_...() functions which were taking printf-like parametter, it introduced __printf attributes to some functions (e.g. kthread_create_worker()). Nevertheless some new functions were forgotten (they have been detected thanks to -Wmissing-format-attribute warning flag). Add the missing __printf attributes to the newly-introduced functions in order to detect formatting issues at build-time with -Wformat flag. Link: http://lkml.kernel.org/r/20161126193543.22672-1-nicolas.iooss_linux@m4x.orgSigned-off-by:
Nicolas Iooss <nicolas.iooss_linux@m4x.org> Reviewed-by:
Petr Mladek <pmladek@suse.com> Signed-off-by:
-
- 12 Dec, 2016 18 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull x86 RAS updates from Ingo Molnar: "The main changes in this development cycle were: - more AMD northbridge support work, mostly in preparation for Fam17h CPUs (Yazen Ghannam, Borislav Petkov) - cleanups/refactorings and fixes (Borislav Petkov, Tony Luck, Yinghai Lu)" * 'ras-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/mce: Include the PPIN in MCE records when available x86/mce/AMD: Add system physical address translation for AMD Fam17h x86/amd_nb: Add SMN and Indirect Data Fabric access for AMD Fam17h x86/amd_nb: Add Fam17h Data Fabric as "Northbridge" x86/amd_nb: Make all exports EXPORT_SYMBOL_GPL x86/amd_nb: Make amd_northbridges internal to amd_nb.c x86/mce/AMD: Reset Threshold Limit after logging error x86/mce/AMD: Fix HWID_MCATYPE calculation by grouping arguments x86/MCE: Correct TSC timestamping of error records x86/RAS: Hide SMCA bank names x86/RAS: Rename smca_bank_names to smca_names x86/RAS: Simplify SMCA HWID descriptor struct x86/RAS: Simplify SMCA bank descriptor struct x86/MCE: Dump MCE to dmesg if no consumers x86/RAS: Add TSC timestamp to the injected MCE x86/MCE: Do not look at panic_on_oops in the severity grading -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull hotplug API fix from Ingo Molnar: "Late breaking fix from the v4.9 cycle: fix a hotplug register/ unregister notifier API asymmetry bug that can cause kernel warnings (and worse) with certain Kconfig combinations" * 'smp-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: hotplug: Make register and unregister notifier API symmetric
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull scheduler updates from Ingo Molnar: "The main scheduler changes in this cycle were: - support Intel Turbo Boost Max Technology 3.0 (TBM3) by introducig a notion of 'better cores', which the scheduler will prefer to schedule single threaded workloads on. (Tim Chen, Srinivas Pandruvada) - enhance the handling of asymmetric capacity CPUs further (Morten Rasmussen) - improve/fix load handling when moving tasks between task groups (Vincent Guittot) - simplify and clean up the cputime code (Stanislaw Gruszka) - improve mass fork()ed task spread a.k.a. hackbench speedup (Vincent Guittot) - make struct kthread kmalloc()ed and related fixes (Oleg Nesterov) - add uaccess atomicity debugging (when using access_ok() in the wrong context), under CONFIG_DEBUG_ATOMIC_SLEEP=y (Peter Zijlstra) - implement various fixes, cleanups and other enhancements (Daniel Bristot de Oliveira, Martin Schwidefsky, Rafael J. Wysocki)" * 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (41 commits) sched/core: Use load_avg for selecting idlest group sched/core: Fix find_idlest_group() for fork kthread: Don't abuse kthread_create_on_cpu() in __kthread_create_worker() kthread: Don't use to_live_kthread() in kthread_[un]park() kthread: Don't use to_live_kthread() in kthread_stop() Revert "kthread: Pin the stack via try_get_task_stack()/put_task_stack() in to_live_kthread() function" kthread: Make struct kthread kmalloc'ed x86/uaccess, sched/preempt: Verify access_ok() context sched/x86: Make CONFIG_SCHED_MC_PRIO=y easier to enable sched/x86: Change CONFIG_SCHED_ITMT to CONFIG_SCHED_MC_PRIO x86/sched: Use #include <linux/mutex.h> instead of #include <asm/mutex.h> cpufreq/intel_pstate: Use CPPC to get max performance acpi/bus: Set _OSC for diverse core support acpi/bus: Enable HWP CPPC objects x86/sched: Add SD_ASYM_PACKING flags to x86 ITMT CPU x86/sysctl: Add sysctl for ITMT scheduling feature x86: Enable Intel Turbo Boost Max Technology 3.0 x86/topology: Define x86's arch_update_cpu_topology sched: Extend scheduler's asym packing sched/fair: Clean up the tunable parameter definitions ... -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull perf updates from Ingo Molnar: "This update is pretty big and almost exclusively includes tooling changes, because v4.9's LTS status forced to completion most of the pending kernel side hardware enablement work and because we tried to freeze core perf work a bit to give a time window for the fuzzing efforts. The diff is large mostly due to the JSON hardware event tables added for Intel and Power8 CPUs. This was a popular feature request from people working close to hardware and from the HPC community. Tree size is big because this added the CPU event tables for over a decade of Intel CPUs. Future changes for a CPU vendor alrady support should be much smaller, as events for new models are added. The new events are listed in 'perf list', for the CPU model the tool is running on. If you find an interesting event it can be used as-is: $ perf stat -a -e l2_lines_out.pf_clean sleep 1 Performance counter stats for 'system wide': 7,860,403 l2_lines_out.pf_clean 1.000624918 seconds time elapsed The event lists can be searched the usual 'perf list' fashion for (case insensitive) substrings as well: $ perf list l2_lines_out List of pre-defined events (to be used in -e): cache: l2_lines_out.demand_clean [Clean L2 cache lines evicted by demand] l2_lines_out.demand_dirty [Dirty L2 cache lines evicted by demand] l2_lines_out.dirty_all [Dirty L2 cache lines filling the L2] l2_lines_out.pf_clean [Clean L2 cache lines evicted by L2 prefetch] l2_lines_out.pf_dirty [Dirty L2 cache lines evicted by L2 prefetch] etc. There's a few high level categories as well that can be listed: 'cache', 'floating point', 'frontend', 'memory', 'pipeline', 'virtual memory'. Existing generic events and workflows should work as-is. The only kernel side change is a late breaking fix for an older regression, related to Intel BTS, LBR and PT feature interaction. On the tooling side there are three new tools / major features: - The new 'perf c2c' tool provides means for Shared Data C2C/HITM analysis. This allows you to track down cacheline contention. The tool is based on x86's load latency and precise store facility events provided by Intel CPUs. It was tested by Joe Mario and has proven to be useful, finding some cacheline contentions. Joe also wrote a blog about c2c tool with examples: https://joemario.github.io/blog/2016/09/01/c2c-blog/ excerpt of the content on this site: At a high level, “perf c2c” will show you: * The cachelines where false sharing was detected. * The readers and writers to those cachelines, and the offsets where those accesses occurred. * The pid, tid, instruction addr, function name, binary object name for those readers and writers. * The source file and line number for each reader and writer. * The average load latency for the loads to those cachelines. * Which numa nodes the samples a cacheline came from and which CPUs were involved. Using perf c2c is similar to using the Linux perf tool today. First collect data with “perf c2c record”, then generate a report output with “perf c2c report” There one finds extensive details on using the tool, with tips on reducing the volume of samples while still capturing enough to do its job. (Dick Fowles, Joe Mario, Don Zickus, Jiri Olsa) - The new 'perf sched timehist' tool provides tailored analysis of scheduling events. Example usage: perf sched record -- sleep 1 perf sched timehist By default it shows the individual schedule events, including the wait time (time between sched-out and next sched-in events for the task), the task scheduling delay (time between wakeup and actually running) and run time for the task: time cpu task name wait time sch delay run time [tid/pid] (msec) (msec) (msec) -------- ------ ---------------- --------- --------- -------- 1.874569 [0011] gcc[31949] 0.014 0.000 1.148 1.874591 [0010] gcc[31951] 0.000 0.000 0.024 1.874603 [0010] migration/10[59] 3.350 0.004 0.011 1.874604 [0011] <idle> 1.148 0.000 0.035 1.874723 [0005] <idle> 0.016 0.000 1.383 1.874746 [0005] gcc[31949] 0.153 0.078 0.022 ... Times are in msec.usec. (David Ahern, Namhyung Kim) - Add CPU vendor hardware event tables: Add JSON files with vendor event naming for Intel and Power8 processors, allowing users of tools like oprofile to keep using the event names they are used to, as well as people reading vendor documentation, where such naming is used. (Andi Kleen, Sukadev Bhattiprolu) You should see all the new events with 'perf list' and you should be able to search them, for example 'perf list miss' will list all the myriads of miss events. Other tooling features added were: - Cross-arch annotation support: o Improve ARM support in the annotation code, affecting 'perf annotate', 'perf report' and live annotation in 'perf top' (Kim Phillips) o Initial support for PowerPC in the annotation code (Ravi Bangoria) o Support AArch64 in the 'annotate' code, native/local and cross-arch/remote (Kim Phillips) - Allow considering just events in a given time interval, via the '--time start.s.ms,end.s.ms' command line, added to 'perf kmem', 'perf report', 'perf sched timehist' and 'perf script' (David Ahern) - Add option to stop printing a callchain at one of a given group of symbol names (David Ahern) - Track memory freed in 'perf kmem stat' (David Ahern) - Allow querying and setting .perfconfig variables (Taeung Song) - Show branch information in callchains (predicted, TSX aborts, loop iteractions, etc) (Jin Yao) - Dynamicly change verbosity level by pressing 'V' in the 'perf top/report' hists TUI browser (Alexis Berlemont) - Implement 'perf trace --delay' in the same fashion as in 'perf record --delay', to skip sampling workload initialization events (Alexis Berlemont) - Make vendor named events case insensitive in 'perf list', i.e. 'perf list LONGEST_LAT' works just the same as 'perf list longest_lat' (Andi Kleen) - Add unwinding support for jitdump (Stefano Sanfilippo) Tooling infrastructure changes: - Support linking perf with clang and LLVM libraries, initially statically, but this limitation will be lifted and shared libraries, when available, will be preferred to the static build, that should, as with other features, be enabled explicitly (Wang Nan) - Add initial support (and perf test entry) for tooling hooks, starting with 'record_start' and 'record_end', that will have as its initial user the eBPF infrastructure, where perf_ prefixed functions will be JITed and run when such hooks are called (Wang Nan) - Implement assorted libbpf improvements (Wang Nan)" ... and lots of other changes, features, cleanups and refactorings I did not list, see the shortlog and the git log for details" * 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (220 commits) perf/x86: Fix exclusion of BTS and LBR for Goldmont perf tools: Explicitly document that --children is enabled by default perf sched timehist: Cleanup idle_max_cpu handling perf sched timehist: Handle zero sample->tid properly perf callchain: Introduce callchain_cursor__copy() perf sched: Cleanup option processing perf sched timehist: Improve error message when analyzing wrong file perf tools: Move perf build related variables under non fixdep leg perf tools: Force fixdep compilation at the start of the build perf tools: Move PERF-VERSION-FILE target into rules area perf build: Check LLVM version in feature check perf annotate: Show raw form for jump instruction with indirect target perf tools: Add non config targets perf tools: Cleanup build directory before each test perf tools: Move python/perf.so target into rules area perf tools: Move install-gtk target into rules area tools build: Move tabs to spaces where suitable tools build: Make the .cmd file more readable perf clang: Compile BPF script using builtin clang support perf clang: Support compile IR to BPF object and add testcase ... -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull mm/PAT cleanup from Ingo Molnar: "A single cleanup for a generic interface that was originally introduced for PAT" * 'mm-pat-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/pat, mm: Make track_pfn_insert() return void
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull locking updates from Ingo Molnar: "The tree got pretty big in this development cycle, but the net effect is pretty good: 115 files changed, 673 insertions(+), 1522 deletions(-) The main changes were: - Rework and generalize the mutex code to remove per arch mutex primitives. (Peter Zijlstra) - Add vCPU preemption support: add an interface to query the preemption status of vCPUs and use it in locking primitives - this optimizes paravirt performance. (Pan Xinhui, Juergen Gross, Christian Borntraeger) - Introduce cpu_relax_yield() and remov cpu_relax_lowlatency() to clean up and improve the s390 lock yielding machinery and its core kernel impact. (Christian Borntraeger) - Micro-optimize mutexes some more. (Waiman Long) - Reluctantly add the to-be-deprecated mutex_trylock_recursive() interface on a temporary basis, to give the DRM code more time to get rid of its locking hacks. Any other users will be NAK-ed on sight. (We turned off the deprecation warning for the time being to not pollute the build log.) (Peter Zijlstra) - Improve the rtmutex code a bit, in light of recent long lived bugs/races. (Thomas Gleixner) - Misc fixes, cleanups" * 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (36 commits) x86/paravirt: Fix bool return type for PVOP_CALL() x86/paravirt: Fix native_patch() locking/ww_mutex: Use relaxed atomics locking/rtmutex: Explain locking rules for rt_mutex_proxy_unlock()/init_proxy_locked() locking/rtmutex: Get rid of RT_MUTEX_OWNER_MASKALL x86/paravirt: Optimize native pv_lock_ops.vcpu_is_preempted() locking/mutex: Break out of expensive busy-loop on {mutex,rwsem}_spin_on_owner() when owner vCPU is preempted locking/osq: Break out of spin-wait busy waiting loop for a preempted vCPU in osq_lock() Documentation/virtual/kvm: Support the vCPU preemption check x86/xen: Support the vCPU preemption check x86/kvm: Support the vCPU preemption check x86/kvm: Support the vCPU preemption check kvm: Introduce kvm_write_guest_offset_cached() locking/core, x86/paravirt: Implement vcpu_is_preempted(cpu) for KVM and Xen guests locking/spinlocks, s390: Implement vcpu_is_preempted(cpu) locking/core, powerpc: Implement vcpu_is_preempted(cpu) sched/core: Introduce the vcpu_is_preempted(cpu) interface sched/wake_q: Rename WAKE_Q to DEFINE_WAKE_Q locking/core: Provide common cpu_relax_yield() definition locking/mutex: Don't mark mutex_trylock_recursive() as deprecated, temporarily ... -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull EFI updates from Ingo Molnar: "The main changes in this development cycle were: - Implement EFI dev path parser and other changes to fully support thunderbolt devices on Apple Macbooks (Lukas Wunner) - Add RNG seeding via the EFI stub, on ARM/arm64 (Ard Biesheuvel) - Expose EFI framebuffer configuration to user-space, to improve tooling (Peter Jones) - Misc fixes and cleanups (Ivan Hu, Wei Yongjun, Yisheng Xie, Dan Carpenter, Roy Franz)" * 'efi-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: efi/libstub: Make efi_random_alloc() allocate below 4 GB on 32-bit thunderbolt: Compile on x86 only thunderbolt, efi: Fix Kconfig dependencies harder thunderbolt, efi: Fix Kconfig dependencies thunderbolt: Use Device ROM retrieved from EFI x86/efi: Retrieve and assign Apple device properties efi: Allow bitness-agnostic protocol calls efi: Add device path parser efi/arm*/libstub: Invoke EFI_RNG_PROTOCOL to seed the UEFI RNG table efi/libstub: Add random.c to ARM build efi: Add support for seeding the RNG from a UEFI config table MAINTAINERS: Add ARM and arm64 EFI specific files to EFI subsystem efi/libstub: Fix allocation size calculations efi/efivar_ssdt_load: Don't return success on allocation failure efifb: Show framebuffer layout as device attributes efi/efi_test: Use memdup_user() as a cleanup efi/efi_test: Fix uninitialized variable 'rv' efi/efi_test: Fix uninitialized variable 'datasize' efi/arm*: Fix efi_init() error handling efi: Remove unused include of <linux/version.h> -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull SMP bootup updates from Ingo Molnar: "Three changes to unify/standardize some of the bootup message printing in kernel/smp.c between architectures" * 'core-smp-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: kernel/smp: Tell the user we're bringing up secondary CPUs kernel/smp: Make the SMP boot message common on all arches kernel/smp: Define pr_fmt() for smp.c
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull RCU updates from Ingo Molnar: "The main RCU changes in this development cycle were: - Miscellaneous fixes, including a change to call_rcu()'s rcu_head alignment check. - Security-motivated list consistency checks, which are disabled by default behind DEBUG_LIST. - Torture-test updates. - Documentation updates, yet again just simple changes" * 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: torture: Prevent jitter from delaying build-only runs torture: Remove obsolete files from rcutorture .gitignore rcu: Don't kick unless grace period or request rcu: Make expedited grace periods recheck dyntick idle state torture: Trace long read-side delays rcu: RCU_TRACE enables event tracing as well as debugfs rcu: Remove obsolete comment from __call_rcu() rcu: Remove obsolete rcu_check_callbacks() header comment rcu: Tighten up __call_rcu() rcu_head alignment check Documentation/RCU: Fix minor typo documentation: Present updated RCU guarantee bug: Avoid Kconfig warning for BUG_ON_DATA_CORRUPTION lib/Kconfig.debug: Fix typo in select statement lkdtm: Add tests for struct list corruption bug: Provide toggle for BUG on data corruption list: Split list_del() debug checking into separate function rculist: Consolidate DEBUG_LIST for list_add_rcu() list: Split list_add() debug checking into separate function -
git://git.kernel.org/pub/scm/linux/kernel/git/jesper/crisLinus Torvalds authored
Pull CRIS updates from Jesper Nilsson: "Three patches for minor issues" * tag 'cris-for-4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jesper/cris: cris: No need to append -O2 and $(LINUXINCLUDE) tty: serial: make crisv10 explicitly non-modular cris: Only build flash rescue image if CONFIG_ETRAX_AXISFLASHMAP is selected
-
git://github.com/openrisc/linuxLinus Torvalds authored
Pull Openrisc updates from Stafford Horne: - changes to MAINTAINER for openrisc - probably biggest actual change is the move to memblock from bootmem - ... plus several bug and build fixes * tag 'openrisc-for-linus' of git://github.com/openrisc/linux: openrisc: prevent VGA console, fix builds openrisc: include l.swa in check for write data pagefault openrisc: Updates after openrisc.net has been lost openrisc: Consolidate setup to use memblock instead of bootmem openrisc: remove the redundant of_platform_populate openrisc: add NR_CPUS Kconfig default value openrisc: Support both old (or32) and new (or1k) toolchain openrisc: Add thread-local storage (TLS) support openrisc: restore all regs on rt_sigreturn openrisc: fix PTRS_PER_PGD define
-
git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68kLinus Torvalds authored
Pull m68k updates from Geert Uytterhoeven: "Use seq_puts() for fixed strings" * tag 'm68k-for-v4.10-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k: m68k/atari: Use seq_puts() in atari_get_hardware_list() m68k/amiga: Use seq_puts() in amiga_get_hardware_list()
-
git://git.kernel.org/pub/scm/linux/kernel/git/egtvedt/linux-avr32Linus Torvalds authored
Pull AVR32 updates from Hans-Christian Noren Egtvedt. * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/egtvedt/linux-avr32: avr32: wire up pkey syscalls AVR32-pio: Replace two seq_printf() calls by seq_puts() in pio_bank_show() AVR32-pio: Use seq_putc() in pio_bank_show() AVR32-clock: Combine nine seq_printf() calls into one call in clk_show() AVR32-clock: Use seq_putc() in two functions
-
git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommuLinus Torvalds authored
Pull m68knommu updates from Greg Ungerer: "There are two sets of changes in this pull. The largest is the addition of the ColdFire platform side i2c support (the IO addressing, setup and clock definitions). The i2c hardware module itself is driven by the kernels existing iMX i2c driver. The other change is the addition of support for the Amcore board" * 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu: m68knommu: AMCORE board, add iMX i2c support m68k: add Sysam AMCORE open board support m68knommu: platform support for i2c devices on ColdFire SoC
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparcLinus Torvalds authored
Pull sparc updates from David Miller: "Just a bunch of small cleanups and fixes here, and support for user probes from Allen Pais" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc: sparc: fix a building error reported by kbuild sparc64: fix typo in pgd_clear() sparc64: restore irq in error paths in iommu sparc: leon: Fix a retry loop in leon_init_timers() sparc64: make string buffers large enough sparc64: move dereference after check for NULL sparc: kernel: use builtin_platform_driver sparc64:Support User Probes for sparc
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-nextLinus Torvalds authored
Pull networking updates from David Miller: 1) Platform regulatory domain support for ath10k, from Bartosz Markowski. 2) Centralize min/max MTU checking, thus removing tons of duplicated code all of the the various drivers. From Jarod Wilson. 3) Support ingress actions in act_mirred, from Shmulik Ladkani. 4) Improve device adjacency tracking, from David Ahern. 5) Add support for LED triggers on PHY link state changes, from Zach Brown. 6) Improve UDP socket memory accounting, from Paolo Abeni. 7) Set SK_MEM_QUANTUM to a fixed size of 4096, instead of PAGE_SIZE. From Eric Dumazet. 8) Collapse TCP SKBs at retransmit time even if the right side SKB has frags. Also from Eric Dumazet. 9) Add IP_RECVFRAGSIZE and IPV6_RECVFRAGSIZE cmsgs, from Willem de Bruijn. 10) Support routing by UID, from Lorenzo Colitti. 11) Handle L3 domain binding (ie. VRF) for RAW sockets, from David Ahern. 12) tcp_get_info() can run lockless, from Eric Dumazet. 13) 4-tuple UDP hashing in SFC driver, from Edward Cree. 14) Avoid reorders in GRO code, from Eric Dumazet. 15) IPV6 Segment Routing support, from David Lebrun. 16) Support MPLS push and pop for L3 packets in openvswitch, from Jiri Benc. 17) Add LRU datastructure support for BPF, Martin KaFai Lau. 18) VF support in liquidio driver, from Raghu Vatsavayi. 19) Multiqueue support in alx driver, from Tobias Regnery. 20) Networking cgroup BPF support, from Daniel Mack. 21) TCP chronograph measurements, from Francis Yan. 22) XDP support for qed driver, from Yuval Mintz. 23) BPF based lwtunnels, from Thomas Graf. 24) Consistent FIB dumping to offloading drivers, from Ido Schimmel. 25) Many optimizations for UDP under high load, from Eric Dumazet. * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1522 commits) netfilter: nft_counter: rework atomic dump and reset e1000: use disable_hardirq() for e1000_netpoll() i40e: don't truncate match_method assignment net: ethernet: ti: netcp: add support of cpts net: phy: phy drivers should not set SUPPORTED_[Asym_]Pause net: l2tp: ppp: change PPPOL2TP_MSG_* => L2TP_MSG_* net: l2tp: deprecate PPPOL2TP_MSG_* in favour of L2TP_MSG_* net: l2tp: export debug flags to UAPI net: ethernet: stmmac: remove private tx queue lock net: ethernet: sxgbe: remove private tx queue lock net: bridge: shorten ageing time on topology change net: bridge: add helper to set topology change net: bridge: add helper to offload ageing time net: nicvf: use new api ethtool_{get|set}_link_ksettings net: ethernet: ti: cpsw: sync rates for channels in dual emac mode net: ethernet: ti: cpsw: re-split res only when speed is changed net: ethernet: ti: cpsw: combine budget and weight split and check net: ethernet: ti: cpsw: don't start queue twice net: ethernet: ti: cpsw: use same macros to get active slave net: mvneta: select GENERIC_ALLOCATOR ... -
Randy Dunlap authored
OpenRISC does not support VGA console, so prevent that kconfig symbol from being enabled for OpenRISC, thus fixing these build errors: drivers/built-in.o: In function `vgacon_save_screen': vgacon.c:(.text+0x20e0): undefined reference to `screen_info' vgacon.c:(.text+0x20e8): undefined reference to `screen_info' drivers/built-in.o: In function `vgacon_init': vgacon.c:(.text+0x284c): undefined reference to `screen_info' vgacon.c:(.text+0x2850): undefined reference to `screen_info' drivers/built-in.o: In function `vgacon_startup': vgacon.c:(.text+0x28d8): undefined reference to `screen_info' drivers/built-in.o:vgacon.c:(.text+0x28f0): more undefined references to `screen_info' follow Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Reported-by:
kbuild test robot <fengguang.wu@intel.com> Cc: Chen Gang <gang.chen@asianux.com> Cc: Jonas Bonn <jonas@southpole.se> Signed-off-by:
Stafford Horne <shorne@gmail.com>
-
Stefan Kristiansson authored
During page fault handling we check the last instruction to understand if the fault was for a read or for a write. By default we fall back to read. New instructions were added to the openrisc 1.1 spec for an atomic load/store pair (l.lwa/l.swa). This patch adds the opcode for l.swa (0x33) allowing it to be treated as a write operation. Signed-off-by:
Stefan Kristiansson <stefan.kristiansson@saunalahti.fi> [shorne@gmail.com: expanded a bit on the comment] Signed-off-by:
-
