|
|
#1 | |
|
Registered User
|
I'm getting hangs on glXWaitVideoSyncSGI on a new GTX 520 card. If I have two GPUs, one works properly, but the 520 hangs after 10 or so iterations of my render loop.
My loop is book-ended by the very standard: glXGetVideoSyncSGI (&retraceCount); glXWaitVideoSyncSGI (2, (retraceCount + 1) % 2, &retraceCount); One screen hangs on glXWaitVideoSyncSGI. It does support the extension, and the first try returns no error. Other info - it's a screen in a window on a twinview X screen (not filling the screen). Both screens are Nvidia cards (275 & 520), both are vsync blanking. Each GPU is one XScreen, on one thread, with it's own context and X display connection. This code has worked for 2-3 years. I'm on kernel 3.1.4, I think. It sounds a lot like the bug fixed in 185.18.31: "· Fixed a bug that caused glXGetVideoSyncSGI, glXWaitVideoSyncSGI, and glXGetRefreshRateSGI to operate on the wrong screen when there are multiple X screens." Is there an alternate way to wait for the start of the next frame (apart from glFinish). Bruce |
|
|
|
|
![]() |
| Thread Tools | |
|
|