On Wed, May 27, 2015 at 07:12:02PM +0100, Tomas Elf wrote: > On 22/05/2015 18:05, Mika Kuoppala wrote: > >During review of dynamic page tables series, I was able > >to hit a lite restore bug with execlists. I assume that > >due to incorrect pd, the batch run out of legit address space > >and into the scratch page area. The ACTHD was increasing > >due to scratch being all zeroes (MI_NOOPs). And as gen8 > >address space is quite large, the hangcheck happily waited > >for a long long time, keeping the process effectively stuck. > > > >According to Chris Wilson any modern gpu will grind to halt > >if it encounters commands of all ones. This seemed to do the > >trick and hang was declared promptly when the gpu wandered into > >the scratch land. > > > >v2: Use 0xffff00ff pattern (Chris) > > Just for my own benefit: > > 1. Is there any particular reason for this pattern rather than 0xffffffff? It is more obvious when userspace reads from the page and copies it into its own data structures or surfaces. See below, if this does impact userspace we should probably revert this patch anyway. > 2. Someone please correct me if I'm wrong here but at least based on > my own experiences with gen9 submitting batch buffers filled with > bad instructions (0xffffffff) to the GPU does not hang it. I'm > guessing that is because there's allegedly a hardware security > parser that MI_NOOPs out invalid instructions during execution. If > that's the case here then I guess we might have to come up with > something else for gen9+ if we want to induce engine hangs once the > execution reaches the scratch page? It's not a problem, there will be a GPU hang eventually (in theory at least). Mika is just trying to shortcircuit that by causing an immediate hang. > On the other hand, on gen9+ page faulting is supposedly not broken > anymore so maybe we don't need the scratch page to begin with there > so maybe it's all moot at that point? Again, if I'm making no sense > here feel free to set things straight, I'm very curious about how > all of this is supposed to work. Generating a pagefault for invalid access is an ABI change and requires opt-in (we have discussed context flags in the past). The most obvious example is the CS prefetch, which we have to prevent generating faults by providing guard pages (on older chipsets at least). But as we have been historically lax on allowing userspace to access invalid pages, we have to assume that userspace has been taking advantage of that. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
