On Thu, Sep 04, 2014 at 08:03:26AM +0200, Daniel Vetter wrote: > On Wed, Sep 3, 2014 at 10:26 PM, Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> wrote: > >> @@ -1385,7 +1372,7 @@ out: > >> /* If this -EIO is due to a gpu hang, give the reset code a > >> * chance to clean up the mess. Otherwise return the proper > >> * SIGBUS. */ > >> - if (!atomic_read(&dev_priv->gpu_error.wedged)) > >> + if (i915_terminally_wedged(&dev_priv->gpu_error)) > >> return VM_FAULT_SIGBUS; > > > > (i915_gem_fault()) > > > > This if() is backwards. > > It should keep the logic mostly unchanged. But I guess returning > SIGBUS if the gpu died for real isn't too friendly? No, the patch inverted the logic. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
