* [PATCH] fix AMDGPU failure with periodic signal
@ 2025-11-02 21:38 Mikulas Patocka
0 siblings, 0 replies; only message in thread
From: Mikulas Patocka @ 2025-11-02 21:38 UTC (permalink / raw)
To: Alex Deucher, Christian König, Andrew Morton,
David Hildenbrand
Cc: amd-gfx, linux-mm
Hi
I have a program that registers a periodic timer that sends SIGALRM with
1ms interval. If I compile a dummy OpenCL kernel while the signal is
triggered, I get these messages:
"amdgpu: init_user_pages: Failed to register MMU notifier: -4"
Apparently, there needs to be fatal_signal_pending.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org
---
mm/vma.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
Index: linux-2.6/mm/vma.c
===================================================================
--- linux-2.6.orig/mm/vma.c 2025-10-13 21:42:48.000000000 +0200
+++ linux-2.6/mm/vma.c 2025-11-02 19:54:04.000000000 +0100
@@ -2156,7 +2156,7 @@ static void vm_lock_mapping(struct mm_st
* mm_take_all_locks() and mm_drop_all_locks are expensive operations
* that may have to take thousand of locks.
*
- * mm_take_all_locks() can fail if it's interrupted by signals.
+ * mm_take_all_locks() can fail if it's interrupted by fatal signals.
*/
int mm_take_all_locks(struct mm_struct *mm)
{
@@ -2175,14 +2175,14 @@ int mm_take_all_locks(struct mm_struct *
* is reached.
*/
for_each_vma(vmi, vma) {
- if (signal_pending(current))
+ if (fatal_signal_pending(current))
goto out_unlock;
vma_start_write(vma);
}
vma_iter_init(&vmi, mm, 0);
for_each_vma(vmi, vma) {
- if (signal_pending(current))
+ if (fatal_signal_pending(current))
goto out_unlock;
if (vma->vm_file && vma->vm_file->f_mapping &&
is_vm_hugetlb_page(vma))
@@ -2191,7 +2191,7 @@ int mm_take_all_locks(struct mm_struct *
vma_iter_init(&vmi, mm, 0);
for_each_vma(vmi, vma) {
- if (signal_pending(current))
+ if (fatal_signal_pending(current))
goto out_unlock;
if (vma->vm_file && vma->vm_file->f_mapping &&
!is_vm_hugetlb_page(vma))
@@ -2200,7 +2200,7 @@ int mm_take_all_locks(struct mm_struct *
vma_iter_init(&vmi, mm, 0);
for_each_vma(vmi, vma) {
- if (signal_pending(current))
+ if (fatal_signal_pending(current))
goto out_unlock;
if (vma->anon_vma)
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma)
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2025-11-03 8:02 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-11-02 21:38 [PATCH] fix AMDGPU failure with periodic signal Mikulas Patocka
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox