Keith Packard schrieb: > Here's a patch which switches from using pipestat to using iir to see > vblank interrupts. I'm wondering what this will do on machines where > people are seeing interrupts get stuck on (and then disabled by the > kernel). > > If you're having any IRQ issues, please give this a try and let me know > what you discover.
With this patch, my machine behaves essentially the same as without. After some time the irq gets stuck. Adding the 1000 loop delay at the end of the for loop(see bug 18609) helped, meaning i am still waiting for the irq to get stuck. So far for the facts ;-). My current random theory is, the irq line itself gets active a short time (a few 100ns) after the reason for the irq can be seen in IIR. When the IIR gets cleared before the irq line is activated, the irq line will stay active indefinitely(stuck) until reset(or similar). It should be possible to test this by setting up at least one irq source and then clearing the IIR as fast as possible. IRQ-handler should probably just acknowledge the irq to the kernel in that case. I myself am not proficient enough with the kernel/xorg-driver stuff to test that. I don't have problems with applying patches and compiling+running software, and ssh into my machine if the display hardware is locked up. Regards, Pierre ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ -- _______________________________________________ Dri-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dri-devel
