Magic-Mirror/gecko-dev
1
0
Fork 0
mirror of https://github.com/mozilla/gecko-dev.git synced 2024-11-30 08:12:05 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Bug 1718333 - Remove unnecessary AssertLinkThread assertions, r=handyman a=pascalc

Browse Source 
After looking through the methods which have this assertion, I couldn't
find any examples of places where not having a specific "link thread"
sequence would cause any issues. I think these assertions can and should
be removed.

The main change required by this was to remove the `!NS_IsMainThread()`
assertion from the SchedulerGroup listener. Due to how callbacks work,
it would be possible for a vsync message to be detected by a
MessageChannel from the main thread if it was sent before the channel
was bound. I don't believe that this change should cause any issues.

Differential Revision: https://phabricator.services.mozilla.com/D119348
This commit is contained in:
Nika Layzell 2021-07-15 21:09:55 +00:00
parent 5fec9b6832
commit 6475099976
3 changed files with 16 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
ipc/glue/MessageChannel.cpp
View File

@ -709,8 +709,7 @@ bool MessageChannel::CanSend() const {
}
void MessageChannel::Clear() {
// Don't clear mWorkerThread; we use it in AssertLinkThread() and
// AssertWorkerThread().
// Don't clear mWorkerThread; we use it in AssertWorkerThread().
//
// Also don't clear mListener. If we clear it, then sending a message
// through this channel after it's Clear()'ed can cause this process to
@ -1042,7 +1041,6 @@ class CancelMessage : public IPC::Message {
};
bool MessageChannel::MaybeInterceptSpecialIOMessage(const Message& aMsg) {
AssertLinkThread();
mMonitor->AssertCurrentThreadOwns();
if (MSG_ROUTING_NONE == aMsg.routing_id()) {
@ -1124,10 +1122,11 @@ bool MessageChannel::ShouldDeferMessage(const Message& aMsg) {
}
void MessageChannel::OnMessageReceivedFromLink(Message&& aMsg) {
AssertLinkThread();
mMonitor->AssertCurrentThreadOwns();
if (MaybeInterceptSpecialIOMessage(aMsg)) return;
if (MaybeInterceptSpecialIOMessage(aMsg)) {
return;
}
mListener->OnChannelReceivedMessage(aMsg);
@ -1204,8 +1203,8 @@ void MessageChannel::OnMessageReceivedFromLink(Message&& aMsg) {
// before returning.
bool shouldPostTask = !shouldWakeUp || wakeUpSyncSend;
IPC_LOG("Receive on link thread; seqno=%d, xid=%d, shouldWakeUp=%d",
aMsg.seqno(), aMsg.transaction_id(), shouldWakeUp);
IPC_LOG("Receive from link; seqno=%d, xid=%d, shouldWakeUp=%d", aMsg.seqno(),
aMsg.transaction_id(), shouldWakeUp);
if (reuseTask) {
return;
@ -2257,7 +2256,7 @@ bool MessageChannel::WaitResponse(bool aWaitTimedOut) {
#ifndef OS_WIN
bool MessageChannel::WaitForSyncNotify(bool /* aHandleWindowsMessages */) {
# ifdef DEBUG
// WARNING: We don't release the lock here. We can't because the link thread
// WARNING: We don't release the lock here. We can't because the link
// could signal at this time and we would miss it. Instead we require
// ArtificialTimeout() to be extremely simple.
if (mListener->ArtificialTimeout()) {
@ -2425,7 +2424,6 @@ bool MessageChannel::MaybeHandleError(Result code, const Message& aMsg,
}
void MessageChannel::OnChannelErrorFromLink() {
AssertLinkThread();
mMonitor->AssertCurrentThreadOwns();
IPC_LOG("OnChannelErrorFromLink");

18
ipc/glue/MessageChannel.h
View File

@ -553,24 +553,6 @@ class MessageChannel : HasResultCodes {
"not on worker thread!");
}
// The "link" thread is either the I/O thread (ProcessLink), the other
// actor's work thread (ThreadLink), or the worker thread (same-thread
// channels).
void AssertLinkThread() const {
if (mIsSameThreadChannel) {
// If we're a same-thread channel, we have to be on our worker
// thread.
AssertWorkerThread();
return;
}
// If we aren't a same-thread channel, our "link" thread is _not_ our
// worker thread!
MOZ_ASSERT(mWorkerThread, "Channel hasn't been opened yet");
MOZ_RELEASE_ASSERT(mWorkerThread && !mWorkerThread->IsOnCurrentThread(),
"on worker thread but should not be!");
}
private:
class MessageTask : public CancelableRunnable,
public LinkedListElement<RefPtr<MessageTask>>,

17
xpcom/threads/SchedulerGroup.cpp
View File

@ -38,20 +38,21 @@ nsresult SchedulerGroup::UnlabeledDispatch(
/* static */
void SchedulerGroup::MarkVsyncReceived() {
if (gEarliestUnprocessedVsync) {
// If we've seen a vsync already, but haven't handled it, keep the
// older one.
return;
}
MOZ_ASSERT(!NS_IsMainThread());
// May be called on any thread when a vsync is received and scheduled to be
// processed. This may occur on the main thread due to queued messages when
// the channel is connected.
bool inconsistent = false;
TimeStamp creation = TimeStamp::ProcessCreation(&inconsistent);
if (inconsistent) {
return;
}
gEarliestUnprocessedVsync = (TimeStamp::Now() - creation).ToMicroseconds();
// Attempt to set gEarliestUnprocessedVsync to our new value. If we've seen a
// vsync already, but haven't handled it, the `compareExchange` will fail and
// the static won't be updated.
uint64_t unprocessedVsync =
uint64_t((TimeStamp::Now() - creation).ToMicroseconds());
gEarliestUnprocessedVsync.compareExchange(0, unprocessedVsync);
}
/* static */