maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

Fix group stop signal handling issue on Linux. 

This patch addresses a bug where in a multi-threaded program a new
signal from the inferior may be received before all group-stop
messages from an earlier signal have been handled.

Patch by Andrew MacPherson

llvm-svn: 200226
This commit is contained in:
Todd Fiala 2014-01-27 17:03:57 +00:00
parent 6a62ebdd0f
commit 1b0539c7f6
1 changed files with 14 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
lldb/source/Plugins/Process/Linux/ProcessMonitor.cpp
View File

@ -1798,14 +1798,23 @@ ProcessMonitor::StopThread(lldb::tid_t tid)
int ptrace_err;
if (!GetSignalInfo(wait_pid, &info, ptrace_err))
{
if (log)
// another signal causing a StopAllThreads may have been received
// before wait_pid's group-stop was processed, handle it now
if (ptrace_err == EINVAL)
{
log->Printf ("ProcessMonitor::%s() GetSignalInfo failed.", __FUNCTION__);
assert(WIFSTOPPED(status) && WSTOPSIG(status) == SIGSTOP);
// This would be a particularly interesting case
if (ptrace_err == EINVAL)
log->Printf ("ProcessMonitor::%s() in group-stop", __FUNCTION__);
if (log)
log->Printf ("ProcessMonitor::%s() resuming from group-stop", __FUNCTION__);
// inferior process is in 'group-stop', so deliver SIGSTOP signal
if (!Resume(wait_pid, SIGSTOP)) {
assert(0 && "SIGSTOP delivery failed while in 'group-stop' state");
}
continue;
}
if (log)
log->Printf ("ProcessMonitor::%s() GetSignalInfo failed.", __FUNCTION__);
return false;
}