https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #12 from PaX Team (pagee...@freemail.hu) ---
(In reply to Christian König from comment #11)
> The issue is that the offset handling should actually be transparent to TTM.
> So mem.start can have any value here which might as well
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #11 from Christian König (deathsim...@vodafone.de) ---
(In reply to PaX Team from comment #9)
> would the following workaround do the job of not triggering the overflow and
> not causing any other logic bugs for our purposes:
Not
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #10 from Nicolai Hähnle (nhaeh...@gmail.com) ---
1. That shouldn't compile (use LONG_MAX).
2. The cur_placement still needs to be updated even in that case.
Guarding only the assignment to bo->offset in this way is a reasonable
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #9 from PaX Team (pagee...@freemail.hu) ---
would the following workaround do the job of not triggering the overflow and
not causing any other logic bugs for our purposes:
--- a/drivers/gpu/drm/ttm/ttm_bo.c 2016-12-13
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #8 from Christian König (deathsim...@vodafone.de) ---
(In reply to Nicolai Hähnle from comment #7)
> Therefore, I'm inclined to say that this is probably not an actual bug.
Yeah, correct. The calculated offset is never used, so the
https://bugzilla.kernel.org/show_bug.cgi?id=194579
Nicolai Hähnle (nhaeh...@gmail.com) changed:
What|Removed |Added
CC||nhaeh...@gmail.com
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #6 from PaX Team (pagee...@freemail.hu) ---
```
(In reply to Christian König from comment #4)
> Now back to topic, what code base is this? Could you post the branch
> you are using somewhere?
it's vanilla 4.9.9 + grsecurity (we don't
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #5 from fin4...@hotmail.com ---
(In reply to Christian König from comment #3)
> Please ignore the incorrect comment by fin4...@hotmail.com.
>
> The upstream kernel the official base amdgpu driver code. Only a few not
> parts which
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #4 from Christian König (deathsim...@vodafone.de) ---
Now back to topic, what code base is this? Could you post the branch you are
using somewhere?
In the upstream kernel the line at ttm_bo.c:388 is just an assignment and can't
cause
https://bugzilla.kernel.org/show_bug.cgi?id=194579
Christian König (deathsim...@vodafone.de) changed:
What|Removed |Added
CC|
https://bugzilla.kernel.org/show_bug.cgi?id=194579
fin4...@hotmail.com changed:
What|Removed |Added
CC||fin4...@hotmail.com
--- Comment #2
https://bugzilla.kernel.org/show_bug.cgi?id=194579
--- Comment #1 from Thore Bödecker (m...@foxxx0.de) ---
Created attachment 254731
--> https://bugzilla.kernel.org/attachment.cgi?id=254731=edit
dmesg including backtrace
--
You are receiving this mail because:
You are watching the assignee of
12 matches
Mail list logo