Age | Commit message (Collapse) | Author |
|
The stats are too dependent on implementation
and won't be relevant in future.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@307786 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
There are several problems with the current annotations (AnnotateRWLockCreate and friends):
- they don't fully support deadlock detection (we need a hook _before_ mutex lock)
- they don't support insertion of random artificial delays to perturb execution (again we need a hook _before_ mutex lock)
- they don't support setting extended mutex attributes like read/write reentrancy (only "linker init" was bolted on)
- they don't support setting mutex attributes if a mutex don't have a "constructor" (e.g. static, Java, Go mutexes)
- they don't ignore synchronization inside of lock/unlock operations which leads to slowdown and false negatives
The new annotations solve of the above problems. See tsan_interface.h for the interface specification and comments.
Reviewed in https://reviews.llvm.org/D31093
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@298809 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Fixes crash reported in:
https://bugs.chromium.org/p/v8/issues/detail?id=4995
The problem is that we don't have a processor in a free interceptor
during thread exit.
The crash was introduced by introduction of Processors.
However, previously we silently leaked memory which wasn't any better.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@268782 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Race deduplication code proved to be a performance bottleneck in the past if suppressions/annotations are used, or just some races left unaddressed. And we still get user complaints about this:
https://groups.google.com/forum/#!topic/thread-sanitizer/hB0WyiTI4e4
ReportRace already has several layers of caching for racy pcs/addresses to make deduplication faster. However, ReportRace still takes a global mutex (ThreadRegistry and ReportMutex) during deduplication and also calls mmap/munmap (which take process-wide semaphore in kernel), this makes deduplication non-scalable.
This patch moves race deduplication outside of global mutexes and also removes all mmap/munmap calls.
As the result, race_stress.cc with 100 threads and 10000 iterations become 30x faster:
before:
real 0m21.673s
user 0m5.932s
sys 0m34.885s
after:
real 0m0.720s
user 0m23.646s
sys 0m1.254s
http://reviews.llvm.org/D12554
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@246758 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
The optimization is two-fold:
First, the algorithm now uses SSE instructions to
handle all 4 shadow slots at once. This makes processing
faster.
Second, if shadow contains the same access, we do not
store the event into trace. This increases effective
trace size, that is, tsan can remember up to 10x more
previous memory accesses.
Perofrmance impact:
Before:
[ OK ] DISABLED_BENCH.Mop8Read (2461 ms)
[ OK ] DISABLED_BENCH.Mop8Write (1836 ms)
After:
[ OK ] DISABLED_BENCH.Mop8Read (1204 ms)
[ OK ] DISABLED_BENCH.Mop8Write (976 ms)
But this measures only fast-path.
On large real applications the speedup is ~20%.
Trace size impact:
On app1:
Memory accesses : 1163265870
Including same : 791312905 (68%)
on app2:
Memory accesses : 166875345
Including same : 150449689 (90%)
90% of filtered events means that trace size is effectively 10x larger.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@209897 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
the new optimizations break when thread ids gets reused (clocks go backwards)
add the necessary tests as well
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@206035 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Make vector clock operations O(1) for several important classes of use cases.
See comments for details.
Below are stats from a large server app, 77% of all clock operations are handled as O(1).
Clock acquire : 25983645
empty clock : 6288080
fast from release-store : 14917504
contains my tid : 4515743
repeated (fast) : 2141428
full (slow) : 2636633
acquired something : 1426863
Clock release : 2544216
resize : 6241
fast1 : 197693
fast2 : 1016293
fast3 : 2007
full (slow) : 1797488
was acquired : 709227
clear tail : 1
last overflow : 0
Clock release store : 3446946
resize : 200516
fast : 469265
slow : 2977681
clear tail : 0
Clock acquire-release : 820028
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@204656 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
than that, but that's a start.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@201861 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
They seems to be unused, but cause maintenance pain.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@200308 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Intercept and sanitize arguments passed to printf functions in ASan and TSan
(don't do this in MSan for now). The checks are controlled by runtime flag
(off by default for now).
Patch http://llvm-reviews.chandlerc.com/D2480 by Yuri Gribov!
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@199729 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Patch by Alexander Taran.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@196098 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@195918 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@195917 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@195839 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193903 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193730 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193655 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193645 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193602 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193415 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
The function is deprecated.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193409 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193405 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193396 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193348 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193241 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@193158 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@192965 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@192959 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
The annotations are AnnotateIgnoreSyncBegin/End,
may be useful to ignore some infrastructure synchronization
that introduces lots of false negatives.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@192355 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191898 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
LibIgnore allows to ignore all interceptors called from a particular set
of dynamic libraries. LibIgnore remembers all "called_from_lib" suppressions
from the provided SuppressionContext; finds code ranges for the libraries;
and checks whether the provided PC value belongs to the code ranges.
Also make malloc and friends interceptors use SCOPED_INTERCEPTOR_RAW instead of
SCOPED_TSAN_INTERCEPTOR, because if they are called from an ignored lib,
then must call our internal allocator instead of libc malloc.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191897 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191516 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
sigwait
sigwaitinfo
sigtimedwait
sigemptyset
sigfillset
sigpending
sigprocmask
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191374 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191305 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191153 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
versioned symbols can not be linked into dynamic library w/o linker script
also simplifies code as side effect
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@191056 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@189767 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@188177 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@188167 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187982 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187978 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Re-applying with a more reliable test case.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187876 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
This reverts commit r187788.
The test case is unreliable (as the test may be run in a situation in
which it has no affinity with cpu0). This can be recommitted with a more
reliable test - possibly using CPU_COUNT != 0 instead (I wasn't entirely
sure that a process was guaranteed to have at least one affinity, though
it seems reasonable, or I'd have made the change myself).
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187841 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187788 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@187412 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Intercepting it makes it process pending signal before return.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@186400 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Handle realpath(path, NULL) form.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@185921 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@185626 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
bcopy
strtoimax, strtoumax
mbstowcs, mbsrtowcs, mbsnrtowcs
wcstombs, wcsrtombs, wcsnrtombs
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@185624 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Move getcwd to common interceptors.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@185424 91177308-0d34-0410-b5e6-96231b3b80d8
|