From: Vladimir D. <vo...@mi...> - 2002-02-27 08:01:21
|
On Tue, 26 Feb 2002, Peter Surda wrote: > On Tue, Feb 26, 2002 at 04:32:25PM +0100, Michel D=E4nzer wrote: > > > > > Hmm.. Michel, Peter, is it possible that "poll" function in DRM d= river is > > > > > screwed up ? Though I thought that texture transfer goes through = an ioctl.. > > > > It does, and Peter says the cycles are wasted in user space, or wha= t are > > > > you getting at? > > > Michel, correct me if I am wrong, but I thought the cycles will be co= unted > > > as "system" time only if we call schedule when inside the kernel, rig= ht ? > > Maybe, I don't know the hairy details. > > > > Anyway, I forgot to mention the main point: Peter observes the same > > effect with and without DRI so it can't be related to the DRM. > Yes indeed, it seems to "eat" rougly the same amount of CPU time regardle= ss > of whether DRI is enabled or not. The time wasted is roughly equivalent t= o the > time it takes to transfer the data (memcpy or x*r128blittexture). The odd > thing is that as I said a wisely placed usleep in r128_video.c "fixes" it= =2E > "fixes" meaning the time wasted is dramatically reduced, although it make= s the > video less fluent and judder is more visible. > > And yes, it occurs in userspace, in contrast to e.g. load caused by memcp= y > which seems to occur in "system". > > Executing the PutImage on r128 takes a long time, about 11ms on DVD-sized > picture. Could it be that something is waiting for PutImage to complete o= r > vice versa? Can X execute a driver function and something else simultaneo= usly? > And what change was done about in November that caused this? > > Mit freundlichen Gr=FC=DFen > > Peter Surda (Shurdeek) <shu...@pa...>, ICQ 10236103, +4365= 05122023 Peter, could you run the following test: allocate 1 meg of ram in a new pPriv struct (in SetupImageVideo) and change the code to do plain memcpy to that memory. I am wondering what the CPU usage will be. Vladimir Dergachev > > -- > Hello, this is Bill Gates and I pronounce Monopoly, er, Windows as Wind= ows. > |