On Thu, Jul 09, 2015 at 10:08:08AM +0100, Ceraolo Spurio, Daniele wrote: > Hi, > > I'm hitting a NULL pointer dereference when I enable the > i915_context_free tracepoint (call trace attached). This is caused > by the fact that the trace tries to access ctx->file_priv, which > however may have already been deleted (even if the pointer is != > NULL). I've used that trace extensively back when I've submitted it > a few months ago without ecountering this issue, but it doesn't seem > that there ever was a guarantee that the file_priv would be valid at > ctx free time, so I'll put the blame on my original commit that > introduced the trace: > > commit 198c974d7e80a5135fc4a2e69a07ba3e64122f8a > Author: Daniele Ceraolo Spurio <daniele.ceraolospurio@xxxxxxxxx> > Date: Mon Nov 10 13:44:31 2014 +0000 > > drm/i915: Add tracepoints to track a vm during its lifetime > > The trace doesn't specifically need the file_priv pointer, it just > uses it to reach the drm_device pointer to get the device index. To > fix the issue I've looked for another way to get the drm_device > pointer from the ctx, but I couldn't find any that is valid for > every GEN. Modifying the trace to add an extra parameter is out of > the question for 2 reasons: > - The only variable available in i915_gem_context_free is the ctx ptr > - Modifying a tracepoint could break ABI > > Ideas, anyone? http://patchwork.freedesktop.org/patch/48529/ -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
