Log - f003cfe63c9d1fe10c154c4af6ff4378ece65f05 - platform/external/libchrome

f003cfe Use a more compact license header in source files. by license.bot · 16 years ago
fa5f993 Fix the final gcc warnings in base and net. Provide some macros to selectively disable MSVC warnings. by mmentovai@google.com · 16 years ago
f26fd3a TrackedObjects assumes you can use a "TLS slot" of -1 to indicate uninitialized. This isn't true for the pthread_key_t type, which is unsigned on Linux and reportedly a struct on Macs. This change modifies the Slot type to be a struct containing an "initialized" flag. by evanm@google.com · 16 years ago
c685767 Revert. Failing unit tests. by evanm@google.com · 16 years ago
78257d4 TrackedObjects assumes you can use a "TLS slot" of -1 to indicate uninitialized. This isn't true for the pthread_key_t type, which is unsigned on Linux and reportedly a struct on Macs. This change modifies the Slot type to be a struct containing an "initialized" flag. by evanm@google.com · 16 years ago
73f5423 Disable warning 4355 on Windows builds: "'this' : used in base member initializer list". by maruel@google.com · 16 years ago
6393bed reland r1075 w/ tweak to fix test failures by darin@google.com · 16 years ago
e0f380d rollback r1075 to see if it helps resolve test failures by darin@google.com · 16 years ago
f543a7c Eliminate TimerManager::GetCurrentDelay in favor of always referring to the fire time of the next timer. I changed the MessagePump API to refer to a delayed_work_time instead of a delay. by darin@google.com · 16 years ago
009e5c4 ThreadLocalStorage::Set should take care of validating its inputs. No need to by darin@google.com · 16 years ago
12d40bb Define MessagePumpDefault and use it to implement MessageLoop on non-Windows by darin@google.com · 16 years ago
981f355 Take 2 at the new MessageLoop implementation. by darin@google.com · 16 years ago
6ddeb84 rollback portions of r928 to test to see if it impacts perf by darin@google.com · 16 years ago
ff36ccf Try only re-scheduling work when HandleWorkMessage is called. by darin@google.com · 16 years ago
a2ac02b Introduce MessagePump to represent the native message pump used to drive a by darin@google.com · 16 years ago
ee6fa72 by darin@google.com · 16 years ago
f75dfb6 by darin@google.com · 16 years ago
5e4ef42 Make timer.cc portable by factoring its Windows bits into MessageLoop. by darin@google.com · 16 years ago
989d5f8 Fix bug 1317458: Busy wait in IPC::SyncChannel::Send in some cases by mpcomplete@google.com · 16 years ago
4d1ce05 Apparently, our WndProc can get called for other HWNDs :( by darin@google.com · 16 years ago
3810032 just some hopefully non-contentious stuff to get out of the way before doing the real MessageLoop changes. by darin@google.com · 16 years ago
965e534 ObjectWatcher needs to know when the current thread's MessageLoop is being destroyed. This might also be generically useful, so I added a new API on ML to observe when the ML is being destroyed. The notification is sent to observers just prior to ML::current() being modified to return NULL. by darin@google.com · 16 years ago
f722ff7 Rollback message loop test by jar@google.com · 16 years ago
a4796f6 This is a test, and will rollback asap after a build/test cycle starts. by jar@google.com · 16 years ago
d43f619 Call ResetBirthTime before dispatching the user's Task. Also, remove PostSignaledTask since it won't be used. by darin@google.com · 16 years ago
0985629 A new (private) interface is supplied that allows the object watcher (coding in by jar@google.com · 16 years ago
9239e02 Support RanAllPending() rather than RunOnce(), and integrated into ipc_sync_channel. by jar@google.com · 16 years ago
b4d1bff (Re-landing of) Support RunOnce() in message loop. by jar@google.com · 16 years ago
fbaaf69 Rollback 109 by jar@google.com · 16 years ago
7ff36e6 by jar@google.com · 16 years ago
3f4a732 Add base to the repository. by initial.commit · 16 years ago