Hi, On Thu, Sep 24, 2015 at 03:53:07PM -0700, Matt Roper wrote: > Just pull the info out of the plane state structure rather than staging > it in an additional structure. > > v2: Add 'visible' condition to sprites_scaled so that we don't limit the > WM level when the sprite isn't enabled. (Ville) It looks like this patch - specifically the visible condition in ilk_compute_cur_wm - causes screen flicker when moving the cursor from one screen to another one in a multi-screen setup. My hardware is a Thinkpad x201s (http://www.thinkwiki.org/wiki/Category:X201s) 00:02.0 VGA compatible controller [0300]: Intel Corporation Core Processor Integrated Graphics Controller [8086:0046] (rev 02), CPU is an i7-620LM. The screen flickering is the one the coursor is leaving. In most cases, parts of the screen just blank for a very short moment. In some rare cases, the screen even stays blanked until I move the cursor back to that screen. No stability issues were observed, this seems to be a purely cosmetic issue. I bisected the issue to 43d59eda1f69631c267e06ab6b94ed3c14f1f6d1. As I knew the flickering was cursor-related, I tried the following minimal patch, which actually fixed the symptom, when applied to 4.4-rc8: diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index f091ad1..1ef0c54 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -1791,7 +1791,7 @@ static uint32_t ilk_compute_cur_wm(const struct intel_crtc_state *cstate, { int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0; - if (!cstate->base.active || !pstate->visible) + if (!cstate->base.active) return 0; return ilk_wm_method2(ilk_pipe_pixel_rate(cstate), I have no idea at all if this is actually fixing anything or if it's just hiding the real bug. All I can say is that the flickering doesn't occur any longer. Jan _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
