Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Ville Syrjala <ville.syrjala@linux.intel.com>
To: igt-dev@lists.freedesktop.org
Subject: [PATCH i-g-t 06/37] lib/rendercopy: Always setup clear color for TGL
Date: Wed,  3 Jul 2024 02:27:46 +0300	[thread overview]
Message-ID: <20240702232817.31147-7-ville.syrjala@linux.intel.com> (raw)
In-Reply-To: <20240702232817.31147-1-ville.syrjala@linux.intel.com>

From: Ville Syrjälä <ville.syrjala@linux.intel.com>

TGL apparently automagically converts regular output to fast
clears when the output matches the configured clear color.
And if we don't enable the clear color packet at all then we
just get some rainbow gibberish on all black parts of the
output.

To avoid always set up the clear color packet when using
using a non-clear color modifier. We'll just stick a bunch
of NaNs into the clear value so it'll never match any
legitimate output, and thus automagic fast clear should not
happen.

TODO: Hide this better inside rendercopy_gen9.c without
      requring extra allocation in the FB BO
TODO: Figure out if other platforms need this sort stuff

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
 lib/igt_fb.c          | 24 ++++++++++++++++++++++++
 lib/intel_bufops.h    |  1 +
 lib/rendercopy_gen9.c | 14 +++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/lib/igt_fb.c b/lib/igt_fb.c
index 9d7808ba8312..857ffb6547a3 100644
--- a/lib/igt_fb.c
+++ b/lib/igt_fb.c
@@ -963,6 +963,16 @@ void igt_calc_fb_size(struct igt_fb *fb)
 		size += calc_plane_size(fb, plane);
 	}
 
+	/*
+	 * We always need a clear color on TGL, make some extra
+	 * room for one it if it's not explicit in the modifier.
+	 *
+	 * TODO: probably better to allocate this as part of the
+	 * batch instead so the fb size doesn't need to change...
+	 */
+	if (fb->modifier == I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS)
+		size = ALIGN(size + 64, 64);
+
 	if (is_xe_device(fb->fd))
 		size = ALIGN(size, xe_get_default_alignment(fb->fd));
 
@@ -2661,6 +2671,20 @@ igt_fb_create_intel_buf(int fd, struct buf_ops *bops,
 	if (fb->modifier == I915_FORMAT_MOD_4_TILED_DG2_RC_CCS_CC)
 		buf->cc.offset = fb->offsets[1];
 
+	/*
+	 * TGL appears to do automagic fast clear when rendering
+	 * black and the clear color isn't specified, or when the
+	 * output matches the specified clear color. Force a
+	 * non-sensical clear color to prevent it from doing this
+	 * when using a non-clear color modifier.
+	 *
+	 * TODO: figure out if other platforms are affected...
+	 */
+	if (fb->modifier == I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS) {
+		buf->cc.disable = true;
+		buf->cc.offset = fb->size - 64;
+	}
+
 	return buf;
 }
 
diff --git a/lib/intel_bufops.h b/lib/intel_bufops.h
index 84e71d41a2c2..06e72ba4ba93 100644
--- a/lib/intel_bufops.h
+++ b/lib/intel_bufops.h
@@ -38,6 +38,7 @@ struct intel_buf {
 	} ccs[2];
 	struct {
 		uint32_t offset;
+		bool disable;
 	} cc;
 	struct {
 		uint64_t offset;
diff --git a/lib/rendercopy_gen9.c b/lib/rendercopy_gen9.c
index 5ee4c89f5cdb..f0757a8e6b78 100644
--- a/lib/rendercopy_gen9.c
+++ b/lib/rendercopy_gen9.c
@@ -1133,15 +1133,23 @@ void _gen9_render_op(struct intel_bb *ibb,
 
 	gen12_emit_aux_pgtable_state(ibb, aux_pgtable_state, true);
 
-	if (fast_clear) {
+	if (fast_clear || dst->cc.disable) {
 		for (int i = 0; i < 4; i++) {
 			intel_bb_out(ibb, MI_STORE_DWORD_IMM_GEN4);
 			intel_bb_emit_reloc(ibb, dst->handle,
 					    I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
                                             dst->cc.offset + i*sizeof(float),
 					    dst->addr.offset);
-			intel_bb_out(ibb, *(uint32_t*)&clear_color[i]);
-               }
+			if (fast_clear) {
+				intel_bb_out(ibb, *(uint32_t*)&clear_color[i]);
+			} else {
+				/*
+				 * Emit NaN so it'll match nothing and thus prevent
+				 * TGL from doing its automagic fast clear tricks.
+				 */
+				intel_bb_out(ibb, 0xffffffff);
+			}
+		}
        }
 
 
-- 
2.44.2


  parent reply	other threads:[~2024-07-02 23:28 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-02 23:27 [PATCH i-g-t 00/37] Intel CCS + 10bpc/fp16 stuff Ville Syrjala
2024-07-02 23:27 ` [PATCH i-g-t 01/37] lib/intel_aux_pgtable: Library to add support for RGB16161616_64B format Ville Syrjala
2024-08-27 15:16   ` Juha-Pekka Heikkila
2024-08-28 13:03     ` Ville Syrjälä
2024-07-02 23:27 ` [PATCH i-g-t 02/37] lib/rendercopy: Add deltas to all surface relocs Ville Syrjala
2024-07-02 23:27 ` [PATCH i-g-t 03/37] tests/kms_big_fb: Use igt_fb_create_intel_buf() Ville Syrjala
2024-07-02 23:27 ` [PATCH i-g-t 04/37] tests/kms_frontbuffer_tracking: Use igt_create_fb() Ville Syrjala
2024-08-27 15:17   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 05/37] lib/igt_fb: Make igt_calc_fb_size() somewhat usable Ville Syrjala
2024-07-02 23:27 ` Ville Syrjala [this message]
2024-07-02 23:27 ` [PATCH i-g-t 07/37] lib/rendercopy: Don't skip clearcolor on flat CCS Ville Syrjala
2024-08-27 15:16   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 08/37] lib/rendercopy: Fix fastclear scaling Ville Syrjala
2024-08-27 15:17   ` Juha-Pekka Heikkila
2024-08-29 15:11     ` Ville Syrjälä
2024-07-02 23:27 ` [PATCH i-g-t 09/37] lib/rendercopy: Extract gen4_surface_format() Ville Syrjala
2024-08-27 15:22   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 10/37] lib/rendercopy: Extract {dg2, lnl}_compression_format() Ville Syrjala
2024-08-27 15:24   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 11/37] lib/rendercopy: Add specific support for 2:10:10:10 formats Ville Syrjala
2024-08-27 16:03   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 12/37] lib/rendercopy: Use the proper compression format for 10bpc on dg2/lnl+ Ville Syrjala
2024-08-27 16:04   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 13/37] lib/rendercopy: Use the proper compression format for 16bpc " Ville Syrjala
2024-08-27 16:25   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 14/37] lib/igt_fb: Extract is_gen12_rc_ccs_cc_modifier() Ville Syrjala
2024-08-27 16:26   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 15/37] lib/igt_fb: Extract ccs_needs_enginecopy() Ville Syrjala
2024-08-27 16:34   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 16/37] lib/igt_fb: Require enginecopy for clear color Ville Syrjala
2024-08-27 16:37   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 17/37] lib/igt_fb: Expose igt_fb_is_ccs_modifier() Ville Syrjala
2024-08-27 16:43   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 18/37] lib/igt_fb: Expose igt_fb_is_gen12_rc_ccs_cc_modifier() Ville Syrjala
2024-08-27 16:45   ` Juha-Pekka Heikkila
2024-07-02 23:27 ` [PATCH i-g-t 19/37] lib/igt_fb: Expose igt_fb_is_gen12_mc_ccs_modifier() Ville Syrjala
2024-08-27 16:46   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 20/37] lib/igt_fb: Adjust how we pick the blitter compression format Ville Syrjala
2024-08-27 16:50   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 21/37] lib/igt_fb: Add DRM_FORMAT_XRGB2101010 compression format for the Ville Syrjala
2024-08-27 16:52   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 22/37] lib/igt_fb: Add 16bpc compression format for the blitter Ville Syrjala
2024-08-27 16:53   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 23/37] lib/igt_fb: Fix planar block copy Ville Syrjala
2024-08-27 16:55   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 24/37] lib/igt_fb: Fix blitter compression format handling Ville Syrjala
2024-08-27 17:00   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 25/37] lib/igt_fb: Try to fix block copy media compression handling Ville Syrjala
2024-08-27 17:04   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 26/37] lib/igt_fb: Assert that we have no clear color when using the bltter Ville Syrjala
2024-08-27 17:09   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 27/37] lib/vebox: Add support for fp16 RGB formats Ville Syrjala
2024-08-27 17:20   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 28/37] lib/vebox: Add 10bpc support Ville Syrjala
2024-08-27 17:39   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 29/37] lib/igt_fb: Treat 2:10:10:10 properly Ville Syrjala
2024-08-27 17:41   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 30/37] tests/kms_plane: Extract skip_format_mod() Ville Syrjala
2024-08-27 17:44   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 31/37] tests/kms_plane: Skip 10bpc formats with media compression Ville Syrjala
2024-08-27 17:46   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 32/37] tests/kms_ccs: Reuse igt_fb_is_gen12_rc_ccs_cc_modifier() Ville Syrjala
2024-08-27 17:47   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 33/37] tests/kms_ccs: Correctly check clear color for 10bpc formats Ville Syrjala
2024-08-27 17:48   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 34/37] tests/kms_ccs: Correctly check clear color for fp16 formats Ville Syrjala
2024-07-02 23:28 ` [PATCH i-g-t 35/37] tests/kms_ccs: Skip 10bpc formats with media compression Ville Syrjala
2024-07-02 23:28 ` [PATCH i-g-t 36/37] tests/kms_ccs: Skip testing on identical plane types Ville Syrjala
2024-08-27 17:50   ` Juha-Pekka Heikkila
2024-07-02 23:28 ` [PATCH i-g-t 37/37] tests/kms_ccs: Provide a hint as to what we're testing Ville Syrjala
2024-08-27 17:51   ` Juha-Pekka Heikkila
2024-07-03 12:55 ` ✓ CI.xeBAT: success for Intel CCS + 10bpc/fp16 stuff Patchwork
2024-07-03 13:10 ` ✗ Fi.CI.BAT: failure " Patchwork
2024-07-03 13:17   ` Ville Syrjälä
2024-07-03 15:31 ` ✓ CI.xeFULL: success " Patchwork
2024-07-05 16:38 ` ✓ Fi.CI.BAT: success for Intel CCS + 10bpc/fp16 stuff (rev2) Patchwork
2024-07-05 16:43 ` ✓ CI.xeBAT: " Patchwork
2024-07-05 20:04 ` ✓ CI.xeFULL: " Patchwork
2024-07-06 19:07 ` ✗ Fi.CI.IGT: failure " Patchwork
2024-07-13 13:29 ` ✓ CI.xeBAT: success for Intel CCS + 10bpc/fp16 stuff (rev3) Patchwork
2024-07-13 13:39 ` ✓ Fi.CI.BAT: " Patchwork
2024-07-13 14:22 ` ✗ CI.xeFULL: failure " Patchwork
2024-07-15  7:46 ` ✓ Fi.CI.IGT: success " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240702232817.31147-7-ville.syrjala@linux.intel.com \
    --to=ville.syrjala@linux.intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox