Age | Commit message (Collapse) | Author |
|
This is reincarnation of http://reviews.llvm.org/D17648 with the bug fix pointed out by Adhemerval (zatrazz).
Currently ThreadState holds both logical state (required for race-detection algorithm, user-visible)
and physical state (various caches, most notably malloc cache). Move physical state in a new
Process entity. Besides just being the right thing from abstraction point of view, this solves several
problems:
Cache everything on P level in Go. Currently we cache on a mix of goroutine and OS thread levels.
This unnecessary increases memory consumption.
Properly handle free operations in Go. Frees are issue by GC which don't have goroutine context.
As the result we could not do anything more than just clearing shadow. For example, we leaked
sync objects and heap block descriptors.
This will allow to get rid of libc malloc in Go (now we have Processor context for internal allocator cache).
This in turn will allow to get rid of dependency on libc entirely.
Potentially we can make Processor per-CPU in C++ mode instead of per-thread, which will
reduce resource consumption.
The distinction between Thread and Processor is currently used only by Go, C++ creates Processor per OS thread,
which is equivalent to the current scheme.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@267678 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Broke aarch64 and darwin bots.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@262046 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Currently ThreadState holds both logical state (required for race-detection algorithm, user-visible)
and physical state (various caches, most notably malloc cache). Move physical state in a new
Process entity. Besides just being the right thing from abstraction point of view, this solves several
problems:
1. Cache everything on P level in Go. Currently we cache on a mix of goroutine and OS thread levels.
This unnecessary increases memory consumption.
2. Properly handle free operations in Go. Frees are issue by GC which don't have goroutine context.
As the result we could not do anything more than just clearing shadow. For example, we leaked
sync objects and heap block descriptors.
3. This will allow to get rid of libc malloc in Go (now we have Processor context for internal allocator cache).
This in turn will allow to get rid of dependency on libc entirely.
4. Potentially we can make Processor per-CPU in C++ mode instead of per-thread, which will
reduce resource consumption.
The distinction between Thread and Processor is currently used only by Go, C++ creates Processor per OS thread,
which is equivalent to the current scheme.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@262037 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@258202 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
On OS X, for weak function (that user can override by providing their own implementation in the main binary), we need extern `"C" SANITIZER_INTERFACE_ATTRIBUTE SANITIZER_WEAK_ATTRIBUTE NOINLINE`.
Fixes a broken test case on OS X, java_symbolization.cc, which uses a weak function __tsan_symbolize_external.
Differential Revision: http://reviews.llvm.org/D14907
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@254298 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
On OS X, GCD worker threads are created without a call to pthread_create. We need to properly register these threads with ThreadCreate and ThreadStart. This patch uses a libpthread API (`pthread_introspection_hook_install`) to get notifications about new threads and about threads that are about to be destroyed.
Differential Revision: http://reviews.llvm.org/D14328
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@252049 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Embed UBSan runtime into TSan and MSan runtimes in the same as we do
in ASan. Extend UBSan test suite to also run tests for these
combinations.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@235954 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Provide defaults for TSAN_COLLECT_STATS and TSAN_NO_HISTORY.
Replace #ifdef directives with #if. This fixes a bug introduced
in r229112, where building TSan runtime with -DTSAN_COLLECT_STATS=0
would still enable stats collection and reporting.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@229581 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Go does not have freed memory.
Reduces per-goroutine overhead from 455K to 356K.
https://code.google.com/p/thread-sanitizer/issues/detail?id=89
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@229113 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Issue 89: Uses a lot of memory for each goroutine
https://code.google.com/p/thread-sanitizer/issues/detail?id=89
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@229112 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
TSAN_SHADOW_COUNT is defined to 4 in all environments.
Other values of TSAN_SHADOW_COUNT were never tested and
were broken by recent changes to shadow mapping.
Remove it as there is no reason to fix nor maintain it.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@226466 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@225111 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@223732 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Summary:
This change removes `__tsan::StackTrace` class. There are
now three alternatives:
# Lightweight `__sanitizer::StackTrace`, which doesn't own a buffer
of PCs. It is used in functions that need stack traces in read-only
mode, and helps to prevent unnecessary allocations/copies (e.g.
for StackTraces fetched from StackDepot).
# `__sanitizer::BufferedStackTrace`, which stores buffer of PCs in
a constant array. It is used in TraceHeader (non-Go version)
# `__tsan::VarSizeStackTrace`, which owns buffer of PCs, dynamically
allocated via TSan internal allocator.
Test Plan: compiler-rt test suite
Reviewers: dvyukov, kcc
Reviewed By: kcc
Subscribers: llvm-commits, kcc
Differential Revision: http://reviews.llvm.org/D6004
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@221194 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 storage (MetaMap) is based on direct shadow (instead of a hashmap + per-block lists).
This solves a number of problems:
- eliminates quadratic behaviour in SyncTab::GetAndLock (https://code.google.com/p/thread-sanitizer/issues/detail?id=26)
- eliminates contention in SyncTab
- eliminates contention in internal allocator during allocation of sync objects
- removes a bunch of ad-hoc code in java interface
- reduces java shadow from 2x to 1/2x
- allows to memorize heap block meta info for Java and Go
- allows to cleanup sync object meta info for Go
- which in turn enabled deadlock detector for Go
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@209810 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
The mode is enabled with -DTSAN_NO_HISTORY=1 flag.
Intended mostly for research purposes (how fast can it go w/o history).
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@208878 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
|
|
print thread creation stack
and stacks where ignores were enabled.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@195836 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
This allows to increase max shadow stack size to 64K,
and reliably catch shadow stack overflows instead of silently
corrupting memory.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@192797 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Factor out code to be reused in LSan. Also switch from linked list to vector.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@184957 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@178159 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@177154 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@174163 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@173932 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
symbols)
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@173917 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
With this change reports say what mutexes the threads hold around the racy memory accesses.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@169493 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@169259 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@168789 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
everywhere now)
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@168059 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@167575 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@167464 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@167453 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@163326 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
time (by not memorizing full stacks in traces)
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@163322 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@161953 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@160288 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@159754 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@159437 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@159430 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@159294 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
This improves signal-/fork-safety of instrumented programs.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158988 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
implementations of functions. Move strchr to sanitizer_libc.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158517 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158450 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158260 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158202 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158200 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
internal allocations from TSan runtime call InternalAlloc from common runtime
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158148 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
its own CheckFailed function.
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158075 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@158056 91177308-0d34-0410-b5e6-96231b3b80d8
|