Hi!
> As the error capture will compress user buffers as directed to by the
> user, it can take an arbitrary amount of time and space. Break up the
> compression loops with a call to cond_resched(), that will allow other
> processes to schedule (avoiding the soft lockups) and also serve as a
> warning should we try to make this loop atomic in the future.
>
> Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
> Cc: stable@xxxxxxxxxxxxxxx
> Reviewed-by: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
This was queued for 4.19-stable, but is very likely wrong.
> @@ -397,6 +399,7 @@ static int compress_page(struct i915_vma_compress *c,
> if (!(wc && i915_memcpy_from_wc(ptr, src, PAGE_SIZE)))
> memcpy(ptr, src, PAGE_SIZE);
> dst->pages[dst->page_count++] = ptr;
> + cond_resched();
>
> return 0;
> }
4.19 compress_page begins with
static int compress_page(struct compress *c,
...
page = __get_free_page(GFP_ATOMIC | __GFP_NOWARN);
and likely may not sleep. That changed with commit
a42f45a2a85998453078, but that one is not present in 4.19..
I believe we don't need this in stable: dumping of error file will not
take so long to trigger softlockup detectors... and if userland access
blocked, we would be able to reschedule, anyway.
Best regards,
Pavel
--
http://www.livejournal.com/~pavelmachek
Attachment:
signature.asc
Description: Digital signature
