[lldb] [Process/FreeBSDRemote] Fix attaching via lldb-server

Fix two bugs that caused attaching to a process in a pre-connected
lldb-server to fail.  These are:

1. Prematurely reporting status in NativeProcessFreeBSD::Attach().
   The SetState() call defaulted to notify the process, and LLGS tried
   to send the stopped packet before the process instance was assigned
   to it.  While at it, add an assert for that in LLGS.

2. Duplicate call to ReinitializeThreads() (via SetupTrace()) that
   overwrote the stopped status in threads.  Now SetupTrace() is called
   directly by NativeProcessFreeBSD::Attach() (not the Factory) in place
   of ReinitializeThreads().

This fixes at least commands/process/attach/TestProcessAttach.py
and python_api/hello_world/TestHelloWorld.py.

Differential Revision: https://reviews.llvm.org/D90525
This commit is contained in:
Michał Górny 2020-10-31 10:30:13 +01:00
parent f893b29397
commit 8e6bcbb417
2 changed files with 6 additions and 3 deletions

View File

@ -698,8 +698,9 @@ Status NativeProcessFreeBSD::Attach() {
0)
return Status(errno, eErrorTypePOSIX);
/* Initialize threads */
status = ReinitializeThreads();
// Initialize threads and tracing status
// NB: this needs to be called before we set thread state
status = SetupTrace();
if (status.Fail())
return status;
@ -707,7 +708,8 @@ Status NativeProcessFreeBSD::Attach() {
static_cast<NativeThreadFreeBSD &>(*thread).SetStoppedBySignal(SIGSTOP);
// Let our process instance know the thread has stopped.
SetState(StateType::eStateStopped);
SetCurrentThreadID(m_threads.front()->GetID());
SetState(StateType::eStateStopped, false);
return Status();
}

View File

@ -1728,6 +1728,7 @@ GDBRemoteCommunicationServerLLGS::SendStopReasonForState(
case eStateSuspended:
case eStateStopped:
case eStateCrashed: {
assert(m_debugged_process_up != nullptr);
lldb::tid_t tid = m_debugged_process_up->GetCurrentThreadID();
// Make sure we set the current thread so g and p packets return the data
// the gdb will expect.