On Tue, Aug 18, 2009 at 6:31 PM, Thomas Hellström<tho...@shipmail.org> wrote: > Jesse Barnes wrote: >> >> Anyway, to me this discussion is more of a "future directions" one than >> a blocker for this particular patchset. AFAICT the only thing that >> needs fixing with this patch is my lock confusion (struct_mutex vs >> mode_config). Or would you like something substantial changed with >> these bits before they land? >> >> > > I think as long as there are efficient ways to short-circuit this > implementation if the command submission mechanism does a better job, I > think it's OK. I had a quick look in the new dri2 code and it looks like the > driver can trick dri2 into thinking there are no outstanding swaps, thus > avoiding the event reads? What's important is that the IOCTL interface gets > right.
Let me follow up on this tomorrow, I've used up my out-going mail quota for today. > What remains to be fixed are the mutex issue and possibly the poll issue. I'll resend the patch with the mutex issue fixed. The poll issue is not a problem as long as we don't enable DRI1 and DRI2 at the same time. > I also think without looking to closely that the drm_read() function doesn't > do return values properly (see ldd3 and man 2 read for return values for the > various blocking modes). In particular, wait_event_interruptible() may > return -ERESTARTSYS which should never be returned to user-space. Returning -ERESTARTSYS from read is the expected behaviour in case you get a signal as far as I know and is handled by glibc. It's used all over the kernel. > It also looks like the new ioctl argument lengths are not 64 bit aligned? Not sure what you mean... that the size of the ioctl struct isn't a multiple of 64 bits? That's not a requirement. cheers, Kristian ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july -- _______________________________________________ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
