If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Announcement
Collapse
No announcement yet.
NVIDIA 367.18 Beta Linux Driver Released With Many Fixes
Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite
NV_robustness_video_memory_purge sounds like it might be made to fix Gnome's bug with the desktop being corrupted on GPU reclocking events. I think they have a workaround in place right now but when I looked into the bug a while ago it seemed to be caused by them not knowing when they need to reload the image after it's cleared.
NV_robustness_video_memory_purge sounds like it might be made to fix Gnome's bug with the desktop being corrupted on GPU reclocking events. I think they have a workaround in place right now but when I looked into the bug a while ago it seemed to be caused by them not knowing when they need to reload the image after it's cleared.
KDE had the same issue (after suspend the screen would be garbage) and implemented a work around to manually flush things - so now it goes garballed for a moment and reset to normal. Would be awesome if this is a real fix for it.
NV_robustness_video_memory_purge sounds like it might be made to fix Gnome's bug with the desktop being corrupted on GPU reclocking events. I think they have a workaround in place right now but when I looked into the bug a while ago it seemed to be caused by them not knowing when they need to reload the image after it's cleared.
That would be nice. Currently I have to issue a gnome shell restart.
Ctrl + F2 I think it is
"r"
Then it restarts the shell and keeps all apps open.
KDE had the same issue (after suspend the screen would be garbage) and implemented a work around to manually flush things - so now it goes garballed for a moment and reset to normal. Would be awesome if this is a real fix for it.
That is not the same issue. KWin corruption was happening due to driver invalidation of vertex buffers, which is afaik legit driver behaviour. KWin guys introduced a hack where if nvidia driver is detected VBO would be recreated after a second in some situations.
Another undocumented change is that it actually compiles against 4.6+ kernels, even current HEAD of Linuses tree (with minor fixes even -next tree).
Previous beta had to have numerous fixes applied via patches since some internal kernel API functions changed.
The kernel flag CONFIG_DEBUG_VM_PGFLAGS exposing a serious bug in NVIDIA drivers hasn't been fixed ostensibly. Luckily Fedora devs backtracked and decided not to enable the flag by default to avoid problems (this feature affects other drivers as well).
Comment