Skip to contentSkip to conversation

SDMA out-of-bounds write access of tiled surface

DISCUSSION Mads 2 participants 5 messages Aug 5, 2016 – Aug 25, 2016 27 messages not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): CAAxE2A60Rf3m764QyWmzwFiiuQAiGMdbmqyNE+pKxwBJJok5YA@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): CADnq5_M=MZUHCeLB4=Xtykzn-Lzfqw8gbdvi=-tMbwkpoBZmHw@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): CAAxE2A5=7NYLJWJtcqiY0=eWffhRfHwSvjjXFuJOR4fW=A3hHA@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): CADnq5_NDv+tMvD1m16WGzLTuvNFoKfXVPiFdapQi7wQyisK5BQ@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): CAAxE2A6_1PHyzjwhmCb5V=9yjG+c2BwRWvbM2Q7VnOoU2w3wWA@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

Mads

author
#23 lore raw

Just to update, R600_DEBUG=nodma is still needed when updating to xorg-server 1.18.4, plasma 5.7.3 and linux 4.7.0, and libdrm, mesa and llvm built from live sources from sometime this week.

Should this info be in some bugzilla somewhere?

- Mads

Quoted 20 lines · Nicolai Hähnle

The culprit SDMA buffer was actually submitted by the X server, I believe.

Nicolai

Quoted 16 lines

On 11.07.2016 23:52, Mads wrote: Yes, also with 5.7 the corruption appears if I start /usr/bin/konsole or /usr/bin/dolphin (and I guess it will be apparent with more
applications, if you want I could test some more). Although, I have no
trouble using chromium or steam, playing all sorts of games. As you can
see if you look through the thread, it might be reason to believe that
plasmashell itself might be the culprit sending the data that corrupts
the screen.

- Mads

Quoted 6 lines · Marek Olšák

Can you see any corruption or is there just the message in dmesg? (It might just be a page directory fetch and not a memory access)

Marek

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): [email protected]

#25 lore raw

FYI, I've disabled SDMA texture copying for Carrizo in Mesa. The VM faults should no longer occur.

Marek

Quoted 4 lines · Mads

Just to update, R600_DEBUG=nodma is still needed when updating to xorg-server 1.18.4, plasma 5.7.3 and linux 4.7.0, and libdrm, mesa and llvm built from live sources from sometime this week.

Quoted 28 lines

Should this info be in some bugzilla somewhere?

- Mads

Quoted 21 lines · Nicolai Hähnle

The culprit SDMA buffer was actually submitted by the X server, I believe.

Nicolai

Quoted 17 lines

On 11.07.2016 23:52, Mads wrote: Yes, also with 5.7 the corruption appears if I start /usr/bin/konsole or
/usr/bin/dolphin (and I guess it will be apparent with more applications, if you want I could test some more). Although, I have no trouble using chromium or steam, playing all sorts of games. As you can see if you look through the thread, it might be reason to believe that plasmashell itself might be the culprit sending the data that corrupts the screen.

- Mads

Quoted 6 lines · Marek Olšák

Can you see any corruption or is there just the message in dmesg? (It might just be a page directory fetch and not a memory access)

Marek

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): CAAxE2A7ojOJj_8VxPgQBExCGp0iNNoYdxskRwp_Y-XGnMo7isA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

author
#27 lore raw

Thank you for notifying :)

Just curious, is there a way to enable them in the same way as disabling them is done with R600_DEBUG=nodma? Because, while desktop applications gets buggy, games and such does not trigger any artifacts, so normally I disable R600_DEBUG=nodma when gaming (as I've assumed disabling dma must lead to less performance).

- Mads

Quoted 5 lines · Marek Olšák

FYI, I've disabled SDMA texture copying for Carrizo in Mesa. The VM faults should no longer occur.

Marek

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): [email protected]

#29 lore raw
Quoted 7 lines · Mads

Thank you for notifying :)

Just curious, is there a way to enable them in the same way as disabling them is done with R600_DEBUG=nodma? Because, while desktop applications gets buggy, games and such does not trigger any artifacts, so normally I disable R600_DEBUG=nodma when gaming (as I've assumed disabling dma must lead to less performance).

No, there is no way to enable it. Generally, SDMA isn't important for performance and isn't invoked very often, because it's incompatible with delta-color compression (DCC).

Marek
_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): CAAxE2A4ihSCWvXHA9LpYhm=heuSKTsytGizc2cTSU+Svsz8WaQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

author
#31 lore raw
Quoted 10 lines · Marek Olšák
Quoted 7 lines · Mads

Thank you for notifying :)

Just curious, is there a way to enable them in the same way as disabling them is done with R600_DEBUG=nodma? Because, while desktop applications gets buggy, games and such does not trigger any artifacts, so normally I disable R600_DEBUG=nodma when gaming (as I've assumed disabling dma must lead to less performance).

Quoted 5 lines

No, there is no way to enable it. Generally, SDMA isn't important for performance and isn't invoked very often, because it's incompatible with delta-color compression (DCC).

Marek

Ah, cool, I guess that also explains why I never saw any artifacting in any games.

Thank you for replying!

- Mads
_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx