Skip to contentSkip to conversation

rendercopy/gen8: Also emit 3DSTATE_WM_DEPTH_STENCIL.

PATCH In review Kenneth Graunke 3 participants 3 messages Jun 3, 2014 – Jun 4, 2014 Library code

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 rendercopy/gen8: Also emit 3DSTATE_WM_DEPTH_STENCIL.

rendercopy was failing to emit 3DSTATE_WM_DEPTH_STENCIL, which is a new packet on Broadwell. Mesa emits this packet.

This appears to fix various tests on a fresh boot, when Mesa has never run.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78890
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78891
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78935
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78936
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78937
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78938

---
 lib/gen8_render.h     | 1 +
 lib/rendercopy_gen8.c | 4 ++++
 2 files changed, 5 insertions(+)
lib/gen8_render.h +1 -0
index fffc100..0eec80c 100644
@@ -48,6 +48,7 @@
 						GEN6_3D(3, 0, 0x21)
 #define GEN8_3DSTATE_PS_BLEND			GEN6_3D(3, 0, 0x4d)
 # define GEN8_PS_BLEND_HAS_WRITEABLE_RT			(1 << 30)
+#define GEN8_3DSTATE_WM_DEPTH_STENCIL		GEN6_3D(3, 0, 0x4e)
 #define GEN8_3DSTATE_PS_EXTRA			GEN6_3D(3,0, 0x4f)
 # define GEN8_PSX_PIXEL_SHADER_VALID			(1 << 31)
 # define GEN8_PSX_ATTRIBUTE_ENABLE			(1 << 8)
lib/rendercopy_gen8.c +4 -0
index 6f5a698..e7567d2 100644
@@ -816,6 +816,10 @@ gen8_emit_ps(struct intel_batchbuffer *batch, uint32_t kernel) {
 
 static void
 gen8_emit_depth(struct intel_batchbuffer *batch) {
+	OUT_BATCH(GEN8_3DSTATE_WM_DEPTH_STENCIL | (3 - 2));
+	OUT_BATCH(0);
+	OUT_BATCH(0);
+
 	OUT_BATCH(GEN7_3DSTATE_DEPTH_BUFFER | (8-2));
 	OUT_BATCH(0);
 	OUT_BATCH(0);
Signature · 1 line
1.9.1
Reviewed-by on v1 1/1
#1 lore raw
Quoted 6 lines · Kenneth Graunke

rendercopy was failing to emit 3DSTATE_WM_DEPTH_STENCIL, which is a new packet on Broadwell. Mesa emits this packet.

This appears to fix various tests on a fresh boot, when Mesa has never run.

Quoted 8 lines

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78890
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78891
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78935
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78936
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78937
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78938

I sat with Ken all day today (after a few days of debugging myself) and it does what it purport to do, and fixes a bunch of tests.

FWIW: I don't think we need to update the null state in the kernel. It's actually kind of nice for finding userspace bugs.

Signature · 1 line
Ben Widawsky, Intel Open Source Technology Center
Tested-by Guo Jinxian
#2 lore raw
Quoted 4 lines · Ben Widawsky
Quoted 3 lines · Kenneth Graunke

rendercopy was failing to emit 3DSTATE_WM_DEPTH_STENCIL, which is a new packet on Broadwell. Mesa emits this packet.

Quoted 8 lines
Quoted 6 lines

This appears to fix various tests on a fresh boot, when Mesa has never run.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78890
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78891
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78935

From this bug,

Quoted 20 lines
Quoted 5 lines

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78936
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78937
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78938

I sat with Ken all day today (after a few days of debugging myself) and it does what it purport to do, and fixes a bunch of tests.

FWIW: I don't think we need to update the null state in the kernel. It's actually kind of nice for finding userspace bugs.

Signature · 5 lines
Ben Widawsky, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 1 line
Jani Nikula, Intel Open Source Technology Center