- 29 Jan, 2016 13 commits
-
-
Peter Zijlstra authored
Now that the perf_event_ctx_lock_nested() call has moved from put_event() into perf_event_release_kernel() the first reason is no longer valid as that can no longer happen. The second reason seems to have been invalidated when Al Viro made fput() unconditionally async in the following commit: 4a9d4b02 ("switch fput to task_work_add") such that munmap()->fput()->release()->perf_release() would no longer happen. Therefore, remove the annotation. This should increase the efficiency of lockdep coverage of perf locking. Suggested-by:
Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
Peter Zijlstra authored
The orphan cleanup workqueue doesn't always catch orphans, for example, if they never schedule after they are orphaned. IOW, the event leak is still very real. It also wouldn't work for kernel counters. Doing it synchonously is a little hairy due to lock inversion issues, but is made to work. Patch based on work by Alexander Shishkin. Suggested-by:
-
Peter Zijlstra authored
There are two concepts of owner wrt an event and they are conflated: - event::owner / event::owner_list, used by prctl(.option = PR_TASK_PERF_EVENTS_{EN,DIS}ABLE). - the 'owner' of the event object, typically the file descriptor. Currently these two concepts are conflated, which gives trouble with scm_rights passing of file descriptors. Passing the event and then closing the creating task would render the event 'orphan' and would have it cleared out. Unlikely what is expectd. This patch untangles these two concepts by using PERF_EVENT_STATE_EXIT to denote the second type. Reported-by:Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by:
-
Peter Zijlstra authored
In preparation to adding more options, convert the boolean argument into a flags word. Signed-off-by:
-
Peter Zijlstra authored
sync_child_event() has outgrown its purpose, it does far too much. Bring it back to its named purpose. Rename __perf_event_exit_task() to perf_event_exit_event() to better reflect what it does and move the event->state assignment under the ctx->lock, like state changes ought to be. Signed-off-by:
-
Peter Zijlstra authored
Use smp_store_release() to clear event->owner and lockless_dereference() to observe it. Further use READ_ONCE() for all lockless reads. This changes perf_remove_from_owner() to leave event->owner cleared. Signed-off-by:
-
Peter Zijlstra authored
We should never attempt to enable a STATE_EXIT event. Signed-off-by:
-
Peter Zijlstra authored
Update the locking order to note that ctx::lock nests inside of child_mutex, as per: perf_ioctl(): ctx::mutex -> perf_event_for_each(): event::child_mutex -> _perf_event_enable(): ctx::lock Signed-off-by: -
Peter Zijlstra authored
There is but a single caller, remove the function - we already have _free_event(), the extra indirection is nonsensical.. Signed-off-by:
-
Alexei Starovoitov authored
Robustify refcounting. Signed-off-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Peter Zijlstra authored
Dan reported: 1229 if (ctx->task == TASK_TOMBSTONE || 1230 !atomic_inc_not_zero(&ctx->refcount)) { 1231 raw_spin_unlock(&ctx->lock); 1232 ctx = NULL; ^^^^^^^^^^ ctx is NULL. 1233 } 1234 1235 WARN_ON_ONCE(ctx->task != task); ^^^^^^^^^^^^^^^^^ The patch adds a NULL dereference. Reported-by:Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Peter Zijlstra authored
Get rid of the 'onln' obfuscation. Signed-off-by:
-
Peter Zijlstra authored
When calling intel_alt_er() with .idx != EXTRA_REG_RSP_* we will not initialize alt_idx and then use this uninitialized value to index an array. When that is not fatal, it can result in an infinite loop in its caller __intel_shared_reg_get_constraints(), with IRQs disabled. Alternative error modes are random memory corruption due to the cpuc->shared_regs->regs[] array overrun, which manifest in either get_constraints or put_constraints doing weird stuff. Only took 6 hours of painful debugging to find this. Neither GCC nor Smatch warnings flagged this bug. Signed-off-by:
-
- 28 Jan, 2016 2 commits
-
-
Peter Zijlstra authored
There is a race between perf_event_exit_task_context() and orphans_remove_work() which results in a use-after-free. We mark ctx->task with TASK_TOMBSTONE to indicate a context is 'dead', under ctx->lock. After which point event_function_call() on any event of that context will NOP A concurrent orphans_remove_work() will only hold ctx->mutex for the list iteration and not serialize against this. Therefore its possible that orphans_remove_work()'s perf_remove_from_context() call will fail, but we'll continue to free the event, with the result of free'd memory still being on lists and everything. Once perf_event_exit_task_context() gets around to acquiring ctx->mutex it too will iterate the event list, encounter the already free'd event and proceed to free it _again_. This fails with the WARN in free_event(). Plug the race by having perf_event_exit_task_context() hold ctx::mutex over the whole tear-down, thereby 'naturally' serializing against all other sites, including the orphan work. Signed-off-by:
-
Peter Zijlstra authored
We should set event->owner before we install the event, otherwise there is a hole where the target task can fork() and we'll not inherit the event because it thinks the event is orphaned. Signed-off-by:
-
- 21 Jan, 2016 16 commits
-
-
Alexander Shishkin authored
We are currently using asynchronous deallocation in the error path in AUX mmap code, which is unnecessary and also presents a problem for users that wish to probe for the biggest possible buffer size they can get: they'll get -EINVAL on all subsequent attemts to allocate a smaller buffer before the asynchronous deallocation callback frees up the pages from the previous unsuccessful attempt. Currently, gdb does that for allocating AUX buffers for Intel PT traces. More specifically, overwrite mode of AUX pmus that don't support hardware sg (some implementations of Intel PT, for instance) is limited to only one contiguous high order allocation for its buffer and there is no way of knowing its size without trying. This patch changes error path freeing to be synchronous as there won't be any contenders for the AUX pages at that point. Reported-by:
Markus Metzger <markus.t.metzger@intel.com> Signed-off-by:
-
Stephane Eranian authored
This patch enables the uncore_imc PMU for Intel SkyLake Desktop processors (Core i7-6700, model 94). It is possible to compute memory read/write bandwidth using: $ perf stat -a -e uncore_imc/data_reads/,uncore_imc/data_writes/ .... Signed-off-by:
Stephane Eranian <eranian@google.com> Signed-off-by:
-
Peter Zijlstra authored
There is a race against perf_event_exit_task() vs event_function_call(),find_get_context(),perf_install_in_context() (iow, everyone). Since there is no permanent marker on a context that its dead, it is quite possible that we access (and even modify) a context after its passed through perf_event_exit_task(). For instance, find_get_context() might find the context still installed, but by the time we get to perf_install_in_context() it might already have passed through perf_event_exit_task() and be considered dead, we will however still add the event to it. Solve this by marking a ctx dead by setting its ctx->task value to -1, it must be !0 so we still know its a (former) task context. Signed-off-by:
-
Peter Zijlstra authored
Try to trigger warnings before races do damage. Signed-off-by:
-
Peter Zijlstra authored
There is one common bug left in all the event_function_call() users, between loading ctx->task and getting to the remote_function(), ctx->task can already have been changed. Therefore we need to double check and retry if ctx->task != current. Insert another trampoline specific to event_function_call() that checks for this and further validates state. This also allows getting rid of the active/inactive functions. Signed-off-by:
-
Peter Zijlstra authored
The perf_remove_from_context() usage in __perf_event_exit_task() is different from the other usages in that this site has already detached and scheduled out the task context. This will stand in the way of stronger assertions checking the (task) context scheduling invariants. Signed-off-by:
-
Peter Zijlstra authored
There is a very nasty problem wrt disabling the perf task scheduling hooks. Currently we {set,clear} ctx->is_active on every __perf_event_task_sched_{in,out}, _however_ this means that if we disable these calls we'll have task contexts with ->is_active set that are not active and 'active' task contexts without ->is_active set. This can result in event_function_call() looping on the ctx->is_active condition basically indefinitely. Resolve this by changing things such that contexts without events do not set ->is_active like we used to. From this invariant it trivially follows that if there are no (task) events, every task ctx is inactive and disabling the context switch hooks is harmless. This leaves two places that need attention (and already had accumulated weird and wonderful hacks to work around, without recognising this actual problem). Namely: - perf_install_in_context() will need to deal with installing events in an inactive context, meaning it cannot rely on ctx-is_active for its IPIs. - perf_remove_from_context() will have to mark a context as inactive when it removes the last event. For specific detail, see the patch/comments. Signed-off-by: -
Peter Zijlstra authored
For no apparent reason and to great confusion the rules for ctx->is_active and cpuctx->task_ctx are different. This means that its not always possible to find all active (task) contexts. Fix this such that if ctx->is_active gets set, we also set (or verify) cpuctx->task_ctx. Signed-off-by:
-
Peter Zijlstra authored
It doesn't make sense to take up-to _4_ references on perf_sched_events() per event, avoid doing this. Signed-off-by:
-
Peter Zijlstra authored
Like perf_enable_on_exec(), perf_event_enable() event scheduling has problems respecting the context hierarchy when trying to schedule events (for example, it will try and add a pinned event without first removing existing flexible events). So simplify it by using the new ctx_resched() call which will DTRT. Signed-off-by:
-
Peter Zijlstra authored
We have a function that does exactly what we want here, use it. This reduces the amount of cpuctx->task_ctx muckery. Signed-off-by:
-
Peter Zijlstra authored
There are two problems with the current perf_enable_on_exec() event scheduling: - the newly enabled events will be immediately scheduled irrespective of their ctx event list order. - there's a hole in the ctx->lock between scheduling the events out and putting them back on. Esp. the latter issue is a real problem because a hole in event scheduling leaves the thing in an observable inconsistent state, confusing things. Fix both issues by first doing the enable iteration and at the end, when there are newly enabled events, reschedule the ctx in one go. Signed-off-by: -
Peter Zijlstra authored
The comment here is horribly out of date, remove it. Signed-off-by:
-
Peter Zijlstra authored
There is a comment that states that perf_event_context_sched_in() will also switch in the cgroup events, I cannot find it does so. Therefore all the resulting logic goes out the window too. Clean that up. Signed-off-by:
-
Peter Zijlstra authored
There appears to be a problem in __perf_event_task_sched_in() wrt cgroup event scheduling. The normal event scheduling order is: CPU pinned Task pinned CPU flexible Task flexible And since perf_cgroup_sched*() only schedules the cpu context, we must call this _before_ adding the task events. Note: double check what happens on the ctx switch optimization where the task ctx isn't scheduled. Signed-off-by:
-
Peter Zijlstra authored
Make various bugs easier to see. Signed-off-by:
-
- 19 Jan, 2016 1 commit
-
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent Pull perf tools improvements and fixes from Arnaldo Carvalho de Melo: User visible bug fixes: - Fix reading of build-id from vDSO (Ben Hutchings) - Fix processing samples for guests, noticed with 'perf kvm', but noticeable as well via other tools, such as 'perf top' (Ravi Bangoria) Build infrastructure: - Add feature-dump target and FEATURES_DUMP make variable, to allow reusing the feature detection results among multiple tools/ living codebases, such as perf and lib/bpf (Jiri Olsa) - 'make -C tools/perf build-test' improvements, making it more paralelizable and allowing building it outside of the source tree, using O= (Wang Nan) Signed-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by:
-
- 15 Jan, 2016 8 commits
-
-
Jiri Olsa authored
Introducing FEATURES_DUMP make variable to provide features detection dump file and bypass the feature detection. The intention is to use this during build tests to skip repeated features detection, like: Get feature dump static build into /tmp/fd file: $ make feature-dump FEATURE_DUMP_COPY=/tmp/fd LDFLAGS=-static BUILD: Doing 'make -j4' parallel build Auto-detecting system features: ... dwarf: [ OFF ] SNIP FEATURE-DUMP file copied into /tmp/fd Use /tmp/fd to build perf: $ make FEATURES_DUMP=/tmp/fd LDFLAGS=-static $ file perf perf: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), statically linked, for ... Suggested-by:Wang Nan <wangnan0@huawei.com> Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Tested-by:
-
Jiri Olsa authored
To provide FEATURE-DUMP into $(FEATURE_DUMP_COPY) if defined, with no further action. Get feature dump of the current build: $ make feature-dump BUILD: Doing 'make -j4' parallel build Auto-detecting system features: ... dwarf: [ on ] FEATURE-DUMP file available in FEATURE-DUMP Get feature dump static build into /tmp/fd file: $ make feature-dump FEATURE_DUMP_COPY=/tmp/fd LDFLAGS=-static BUILD: Doing 'make -j4' parallel build Auto-detecting system features: ... dwarf: [ OFF ] SNIP FEATURE-DUMP file copied into /tmp/fd Suggested-by: -
Wang Nan authored
Kernel makefile only follows an 'O' option passed from command line explicitely. In build-test with 'O' option set, kernel makefile contaminate kernel source directory. Build test also fail if we don't create output directory manually. K_O_OPT is added and passed to kernel makefile if 'O' is passed to build-test. Signed-off-by:
-
Wang Nan authored
If an 'O' is passed to 'make build-test', many 'test -x' and 'test -f' will fail because perf resides in a different directory. Fix this by computing PERF_OUT according to 'O' and test correct output files. For make_kernelsrc and make_kernelsrc_tools, set KBUILD_OUTPUT_DIR instead because the path is different from others ($(O)/perf vs $(O)/tools/perf). Signed-off-by:
-
Wang Nan authored
Unlike tools/perf/Makefile, tools/perf/Makefile.perf obey 'O' option when it is passed through cmdline only, due to code in tools/scripts/Makefile.include: ifneq ($(O),) ifeq ($(origin O), command line) ... ABSOLUTE_O := $(shell cd $(O) ; pwd) OUTPUT := $(ABSOLUTE_O)/$(if $(subdir),$(subdir)/) endif endif This patch passes 'O' to Makefile.perf through cmdline explicitly to make it follow O variable during build-test. 'make clean' should have identical 'O' option with 'make'. If not, config-clean may error. Signed-off-by:
-
Wang Nan authored
'make build-test' is painful because of time consuming. In a full test, all test cases are built twice with tools/perf/Makefile and tools/perf/Makefile.perf. 'Makefile' automatically computes parallel options for make, but 'Makefile.perf' not, so all test cases is built with one job. It is very slow. This patch adds '-j' options to Makefile.perf testing. It computes parallel building options like what tools/perf/Makefile does, and pass '-j' option to Makefile.perf test. Signed-off-by:
-
Ben Hutchings authored
We need to use the long name (the filename) when reading the build-id from a DSO. Using the short name doesn't work for (at least) vDSOs. Signed-off-by:
Ben Hutchings <ben@decadent.org.uk> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20160113172301.GT28542@decadent.org.ukSigned-off-by:
-
Ravi Bangoria authored
While recording guest samples in host using perf kvm record, it will populate unprocessable sample error, though samples will be recorded properly. While generating report using perf kvm report, no samples will be processed and same error will populate. We have seen this behaviour with upstream perf(4.4-rc3) on x86 and ppc64 hardware. Reason behind this failure is, when it tries to fetch machine from rb_tree of machines, it fails. As a part of tracing a bug, we figured out that this code was incorrectly refactored in commit 54245fdc ("perf session: Remove wrappers to machines__find"). This patch will change the functionality such that if it can't fetch machine in first trial, it will create one node of machine and add that to rb_tree. So next time when it tries to fetch same machine from rb_tree, it won't fail. Actually it was the case before refactoring of code in aforementioned commit. This patch is generated from acme perf/core branch. Below I've mention an example that demonstrate the behaviour before and after applying patch. Before applying patch: [Note: One needs to run guest before recording data in host] ravi@ravi-bangoria:~$ ./perf kvm record -a Warning: 5903 unprocessable samples recorded. Do you have a KVM guest running and not using 'perf kvm'? [ perf record: Captured and wrote 1.409 MB perf.data.guest (285 samples) ] ravi@ravi-bangoria:~$ ./perf kvm report --stdio Warning: 5903 unprocessable samples recorded. Do you have a KVM guest running and not using 'perf kvm'? # To display the perf.data header info, please use --header/--header-only options. # # Total Lost Samples: 0 # # Samples: 285 of event 'cycles' # Event count (approx.): 88715406 # # Overhead Command Shared Object Symbol # ........ ....... ............. ...... # # (For a higher level overview, try: perf report --sort comm,dso) # After applying patch: ravi@ravi-bangoria:~$ ./perf kvm record -a [ perf record: Captured and wrote 1.188 MB perf.data.guest (17 samples) ] ravi@ravi-bangoria:~$ ./perf kvm report --stdio # To display the perf.data header info, please use --header/--header-only options. # # Total Lost Samples: 0 # # Samples: 17 of event 'cycles' # Event count (approx.): 700746 # # Overhead Command Shared Object Symbol # ........ ....... ................ ...................... # 34.19% :5758 [unknown] [g] 0xffffffff818682ab 22.79% :5758 [unknown] [g] 0xffffffff812dc7f8 22.79% :5758 [unknown] [g] 0xffffffff818650d0 14.83% :5758 [unknown] [g] 0xffffffff8161a1b6 2.49% :5758 [unknown] [g] 0xffffffff818692bf 0.48% :5758 [unknown] [g] 0xffffffff81869253 0.05% :5758 [unknown] [g] 0xffffffff81869250 Signed-off-by:
Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: stable@vger.kernel.org # v3.19+ Fixes: 54245fdc ("perf session: Remove wrappers to machines__find") Link: http://lkml.kernel.org/r/1449471302-11283-1-git-send-email-ravi.bangoria@linux.vnet.ibm.comSigned-off-by:
-
