[Bug 105720] screen corruption using SNA and TearFree on Intel GeminiLake

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=105720

--- Comment #31 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to Clinton Taylor from comment #30)
> (In reply to Chris Wilson from comment #28)
> > Make that DamageRegionAppend(&sna->front->drawable,
> > &sna->mode.shadow_region); for a quieter compiler.
> 
> This makes a big difference to the green display (patch from comment #26) on
> the rotated display and a little difference on the normal display.
> 
> I have also not replicated the issue yet on either display with
> DamageRegionAppend() added.

It's the "either display" that keeps me worried. If it was just the rotated
one, the next step would be

diff --git a/src/sna/sna_display.c b/src/sna/sna_display.c
index 7fccb508..1c090946 100644
--- a/src/sna/sna_display.c
+++ b/src/sna/sna_display.c
@@ -9191,6 +9191,9 @@ void sna_mode_redisplay(struct sna *sna)
                                        RegionUnion(&damage, &damage,
&sna_crtc->crtc_damage);
                                sna_crtc->crtc_damage = new_damage;

+                               damage.extents = crtc->bounds;
+                               damage.data = NULL;
+
                                sna_crtc_redisplay(crtc, &damage, bo);
                                kgem_bo_submit(&sna->kgem, bo);
                                __kgem_bo_clear_dirty(bo);

to force the entire per-crtc bo to be regenerate on each update.

Oh well if DamageRegionAppend survives, at least we have a w/a. TBD if it's a
happy w/a.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20180330/18b66734/attachment.html>