intercept SIGUSR2 in RCCL
add support for intercepting SIGUSR2 in RCCL. This signal will
not terminate the execution of the application, but print the stacktrace
of the process that the signal was sent to instead.
[ROCm/rccl commit: 2b1d5d3bc1]
Dieser Commit ist enthalten in:
1 Ursprung
24d9d1c37a
Commit
a9e0333dba
1 Datei geändert
+5
-1
@@ -58,6 +58,10 @@ void sig_handler(int signum)
|
||||
free (strings);
|
||||
#endif
|
||||
|
||||
if (signum == SIGUSR2) {
|
||||
return;
|
||||
}
|
||||
|
||||
exit (-1);
|
||||
}
|
||||
|
||||
@@ -69,7 +73,7 @@ void RegisterSignalHandlers()
|
||||
{
|
||||
INFO(NCCL_INIT, "Enabling custom signal handler");
|
||||
|
||||
std::vector<int> signalsToCatch = {SIGILL, SIGBUS, SIGFPE, SIGSEGV};
|
||||
std::vector<int> signalsToCatch = {SIGILL, SIGBUS, SIGFPE, SIGSEGV, SIGUSR2};
|
||||
|
||||
for (auto signum : signalsToCatch)
|
||||
{
|
||||
|
||||
In neuem Issue referenzieren
Einen Benutzer sperren