Chromium Code Reviews Chromium Code Reviews
Issue 10806096:
Add base/md5* to ThreadSanitizer ignore list. (Closed)
Base URL: svn://chrome-svn/chrome/trunk/src| OLD | NEW |
|---|---|
| 1 # This file lists the functions, object files and source files | 1 # This file lists the functions, object files and source files |
| 2 # which should be ignored (i.e. not instrumented) by ThreadSanitizer. | 2 # which should be ignored (i.e. not instrumented) by ThreadSanitizer. |
| 3 # See http://code.google.com/p/data-race-test/wiki/ThreadSanitizerIgnores. | 3 # See http://code.google.com/p/data-race-test/wiki/ThreadSanitizerIgnores. |
| 4 | 4 |
| 5 # ignore these libraries | 5 # ignore these libraries |
| 6 obj:*/libfreetype* | 6 obj:*/libfreetype* |
| 7 obj:*/libdbus* | 7 obj:*/libdbus* |
| 8 | 8 |
| 9 # we ignore the whole NSS library for now since | 9 # we ignore the whole NSS library for now since |
| 10 # its instrumentation is very slow. | 10 # its instrumentation is very slow. |
| (...skipping 47 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
| 58 src:*base/synchronization/lock* | 58 src:*base/synchronization/lock* |
| 59 src:*base/synchronization/waitable_event* | 59 src:*base/synchronization/waitable_event* |
| 60 | 60 |
| 61 # Don't instrument code dealing with atomics (base::subtle) | 61 # Don't instrument code dealing with atomics (base::subtle) |
| 62 fun:*base*subtle*Release_Store* | 62 fun:*base*subtle*Release_Store* |
| 63 fun:*base*subtle*NoBarrier_CompareAndSwap* | 63 fun:*base*subtle*NoBarrier_CompareAndSwap* |
| 64 fun:*base*subtle*NoBarrier_Load* | 64 fun:*base*subtle*NoBarrier_Load* |
| 65 # Keep some mangling so we don't match NoBarrier_AtomicIncrement | 65 # Keep some mangling so we don't match NoBarrier_AtomicIncrement |
| 66 fun:*base*subtle23Barrier_AtomicIncrement* | 66 fun:*base*subtle23Barrier_AtomicIncrement* |
| 67 | 67 |
| 68 # MD5 computations are very slow due since sums are computed by | |
| 69 # repeatedly calling tiny functions. | |
|
Timur Iskhodzhanov
2012/07/26 15:53:01
Please add "... which are unlikely to race with an
scherkus (not reviewing)
2012/07/26 18:02:48
Done.
| |
| 70 src:*base/md5* | |
| 71 | |
| 68 # Don't instrument tcmalloc | 72 # Don't instrument tcmalloc |
| 69 src:*/tcmalloc/* | 73 src:*/tcmalloc/* |
| 70 | 74 |
| 71 # This function is heavy in net_unittests | 75 # This function is heavy in net_unittests |
| 72 fun_r:*disk_cache*BackendImpl*CheckAllEntries* | 76 fun_r:*disk_cache*BackendImpl*CheckAllEntries* |
| 73 | 77 |
| 74 # V8 is a hot-spot under ThreadSanitizer. | 78 # V8 is a hot-spot under ThreadSanitizer. |
| 75 # Lots of tiny functions there... | 79 # Lots of tiny functions there... |
| 76 # TODO(timurrrr): | 80 # TODO(timurrrr): |
| 77 # Can we miss data races on V8 objects due to non thread-safe API calls | 81 # Can we miss data races on V8 objects due to non thread-safe API calls |
| (...skipping 75 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
| 153 fun:__gen_tempname | 157 fun:__gen_tempname |
| 154 | 158 |
| 155 # The sqlite cache is racing against a few different stacktraces, | 159 # The sqlite cache is racing against a few different stacktraces, |
| 156 # so let's ignore it recursively. See http://crbug.com/84094 | 160 # so let's ignore it recursively. See http://crbug.com/84094 |
| 157 fun_r:pcache1Fetch | 161 fun_r:pcache1Fetch |
| 158 | 162 |
| 159 # "Suppress" a data race in TraceLog::GetCategory which has | 163 # "Suppress" a data race in TraceLog::GetCategory which has |
| 160 # fun:MessageLoop::RunTask at the top of the "current" stack which we don't want | 164 # fun:MessageLoop::RunTask at the top of the "current" stack which we don't want |
| 161 # to suppress. See http://crbug.com/98926 | 165 # to suppress. See http://crbug.com/98926 |
| 162 fun:*base*debug*TraceLog*GetCategoryInternal* | 166 fun:*base*debug*TraceLog*GetCategoryInternal* |
| OLD | NEW |
