Am Samstag, 28. September 2002 22:04 schrieb Felix Kühling: > On Sat, 28 Sep 2002 21:36:43 +0200 > > Dieter Nützel <[EMAIL PROTECTED]> wrote: > > Am Samstag, 28. September 2002 21:20 schrieb Felix Kühling: > > > On Sat, 28 Sep 2002 17:28:01 +0100 > > > > > > Keith Whitwell <[EMAIL PROTECTED]> wrote: > > > > Felix Kühling wrote: > > > > > > [snip] > > > > > > > > Ok, I think it can be done. However, we would react to load changes > > > > > with one frame delay. Before I go on amusing people with another > > > > > naive patch, I'll explain my idea. If you think it's ok, I'll go > > > > > ahead implementing it, this time using radeonContext instead of a > > > > > static variable ;) > > > > > > > > > > radeonContext would contain the irq_seq number and a boolean > > > > > specifying whether there is an IRQ to wait for. If there is none, > > > > > we have to do busy waiting. > > > > > > > > > > If we wait for an IRQ we check first, whether the last frame has > > > > > already been completed. If so, we don't have to wait. > > > > > > > > > > In both cases (busy waiting and IRQ), if we did really wait, we > > > > > emit an IRQ guessing that it will be the same next time. > > > > > radeonContext is updated accordingly. > > > > > > > > > > To avoid switching the waiting scheme all the time we could switch > > > > > to IRQ waiting only if we had to do at least N usleeps. > > > > > > > > This sounds good. I'm looking forward to see if you can implement it > > > > without things getting too ugly. > > > > > > > :) We'll see. > > > : > > > > In the meantime, why don't you send me a tested version of your > > > > earlier patch, modified to keep 'irq_seq' in rmesa. I'll commit it. > > > > > > Here you go ... tested as good as I can. I don't have/know any GL > > > clients that use more than one context at at time. > > > > VTK http://public.kitware.com/VTK/ > > All of my 3D medical VIS apps? ;-) > > > > Have you tried with 2 times running "gears" or the like? > > I can't run it with r200 and current trunk => lookup > > Worked with the tdfx driver for ages (only some texture corruptions due > > to texmem fightings). > > Even one instance of Q3A and UT 436 worked on my dual Athlon MP 1900+ :-) > > Yeah, I tried that. Works like a charm with IRQs as glxgears goes to > sleep volunteerly. 2x glxgears + <choose-your-3d-game> is no problem and > runs smoothly :)
OK, I have to make something clear. I tried it with r200, so I mixed some things up. Sorry! Tried it with my r200 on current trunk, again. With "setenv R200_NO_USLEEPS 1" or without it 2x gears + 1x isosurf works OK. But when I start an additonal instance of ipers, it goes BUMM. UT and Q3A together lookup after some seconds. => only Reboot works > > > The attached patch contains both, the improved WaitForFrameCompletion > > > and IRQ waiting in radeonFinish. > > > > I'll try it with VTK. > > Thanks. You mean on a radeon (not r200), right? NO, r200 ;-) But after your patch is "only" for the "old" Radeon (I) it makes not much sense... -Dieter ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
