Page MenuHomePhabricator

Have a clean(er) shutdown when detaching from processes

Authored by zturner on Sep 18 2015, 1:10 PM.



Log file pre-patch:

1442531634.442000000 [0448/5520]: DoDetach called for process 25769804660 while state = 2673276.  Detaching...
1442531634.442000000 [0448/5520]: StopDebugging('false') called (inferior=2972).
1442531634.442000000 [0448/5520]: StopDebugging waiting for detach from process 2972 to complete.
1442531639.442000000 [0448/5520]: error: StopDebugging WaitForSingleObject(0x0000033C, 5000) returned 258
1442531639.442000000 [0448/5520]: Process::SetPrivateState (detached)

Log file post-patch:

1442606676.750000000 [4a20/574c]: DoDetach called for process 25769804636 while state = 2673276.  Detaching...
1442606676.750000000 [4a20/4c5c]: Process::lldb_private::Process::HandlePrivateEvent (pid = 19724) suppressing state running (old state running): should_broadcast == false
1442606676.750000000 [4a20/574c]: StopDebugging('false') called (inferior=19724).
1442606676.750000000 [4a20/4c5c]: Process::lldb_private::Process::WaitForEventsPrivate (timeout = 00000000, event_sp)...
1442606676.750000000 [4a20/574c]: StopDebugging waiting for detach from process 19724 to complete.
1442606676.750000000 [4a20/4a78]: Breakpoint exception is cue to detach from process 0x4d0c
1442606676.750000000 [4a20/4a78]: WaitForDebugEvent loop completed, exiting.
1442606676.750000000 [4a20/574c]: StopDebugging detach from process 19724 completed successfully.

Diff Detail

Event Timeline

zturner updated this revision to Diff 35127.Sep 18 2015, 1:10 PM
zturner retitled this revision from to Have a clean(er) shutdown when detaching from processes.
zturner updated this object.
zturner added a reviewer: amccarth.
zturner updated this object.
zturner added a subscriber: lldb-commits.
amccarth accepted this revision.Sep 18 2015, 1:36 PM
amccarth edited edge metadata.
amccarth added inline comments.

Open brace goes on the next line. (Which appears to have originally been my mistake. Sorry.)


This is fine, but I'm growing concerned about the growing spaghetti-ness. It's too bad we can't push this logic down into HandleExceptionEvent, in order to keep the cases in this deeply-nested switch statement easy to understand.

This revision is now accepted and ready to land.Sep 18 2015, 1:36 PM

I should be able to move it into HandleExceptionEvent

zturner closed this revision.Oct 15 2015, 1:52 PM