On Fri, 27 Apr 2012 21:17:02 +0200, Daniel Vetter <daniel at ffwll.ch> wrote: > On Fri, Apr 27, 2012 at 11:49:16AM -0700, Eric Anholt wrote: > > On Fri, 27 Apr 2012 15:17:47 +0200, Daniel Vetter <daniel.vetter at ffwll.ch> wrote: > > > It's busted. Compared to ilk, snb and ivb where I've run the i-g-t > > > hangman test in a loop for a few hours, my gm45 never managed to reset > > > the gpu. And Chris Wilson confirmed on irc that he's never seen a case > > > where the gen4 gpu reset code actually helped. > > > > I remember originally when the hangcheck stuff landed, on a 965gm > > running compiz, dragging a window with a bad shader around and seeing a > > stutter as it reset the hardware successfully before crashing soon after > > (then turning off the hanging plugin and continuing to use the system). > > I'm confused - so hangcheck once indeed managed to reset a i965gm gpu or > never? As far as I can remember, at one point it was reliably resetting. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20120427/672f6b4c/attachment.pgp>
