Skip to content

Fix SIGUSR1/SIGUSR2 handler clobbering breaking JVM processes #160

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
charford wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix SIGUSR1/SIGUSR2 handler clobbering breaking JVM processes #160

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 20 additions & 2 deletions src/multiprocess/multiprocess_memory_limit.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
#include <unistd.h>
#include <time.h>
#include <signal.h>
#include <string.h>

#include <assert.h>
#include <cuda.h>
Expand Down Expand Up @@ -73,12 +74,21 @@ void set_current_gpu_status(int status){
}
}

static struct sigaction old_sigusr1_sa;
static struct sigaction old_sigusr2_sa;

void sig_restore_stub(int signo){
set_current_gpu_status(1);
// Chain to previous handler so JVM (and other runtimes) can process SIGUSR1
if (old_sigusr1_sa.sa_handler != SIG_DFL && old_sigusr1_sa.sa_handler != SIG_IGN)
old_sigusr1_sa.sa_handler(signo);
}

void sig_swap_stub(int signo){
set_current_gpu_status(2);
// Chain to previous handler so JVM (and other runtimes) can process SIGUSR2
if (old_sigusr2_sa.sa_handler != SIG_DFL && old_sigusr2_sa.sa_handler != SIG_IGN)
old_sigusr2_sa.sa_handler(signo);
}


Expand Down Expand Up @@ -689,8 +699,16 @@ void init_proc_slot_withlock() {
if (proc_num >= SHARED_REGION_MAX_PROCESS_NUM) {
exit_withlock(-1);
}
signal(SIGUSR2,sig_swap_stub);
signal(SIGUSR1,sig_restore_stub);
struct sigaction sa;
memset(&sa, 0, sizeof(sa));
sigemptyset(&sa.sa_mask);
sa.sa_flags = SA_RESTART;

sa.sa_handler = sig_swap_stub;
sigaction(SIGUSR2, &sa, &old_sigusr2_sa);

sa.sa_handler = sig_restore_stub;
sigaction(SIGUSR1, &sa, &old_sigusr1_sa);

// If, by any means a pid of itself is found in region->process, then it is probably caused by crashloop
// we need to reset it.
Expand Down