* [PATCH 0/2] Consider joiner calculation for panel fitting
@ 2024-12-12 14:33 Nemesa Garg
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
` (2 more replies)
0 siblings, 3 replies; 12+ messages in thread
From: Nemesa Garg @ 2024-12-12 14:33 UTC (permalink / raw)
To: intel-gfx, intel-xe; +Cc: Nemesa Garg
There is an issue when pch_pfit and joiner gets enabled together.
Compute final pfit dst after pipe_src width is adjusted for joiner.
Nemesa Garg (2):
drm/i915/display: After joiner compute pfit_dst
drm/i915/display: Initialize pipe_src in compute stage
drivers/gpu/drm/i915/display/intel_display.c | 58 ++++++++++++++++++--
1 file changed, 54 insertions(+), 4 deletions(-)
--
2.25.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-12 14:33 [PATCH 0/2] Consider joiner calculation for panel fitting Nemesa Garg
@ 2024-12-12 14:33 ` Nemesa Garg
2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-19 16:02 ` Ville Syrjälä
2024-12-12 14:33 ` [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage Nemesa Garg
2024-12-12 15:52 ` ✗ i915.CI.BAT: failure for Consider joiner calculation for panel fitting (rev6) Patchwork
2 siblings, 2 replies; 12+ messages in thread
From: Nemesa Garg @ 2024-12-12 14:33 UTC (permalink / raw)
To: intel-gfx, intel-xe; +Cc: Nemesa Garg
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=y, Size: 2873 bytes --]
In panel fitter/pipe scaler scenario the pch_pfit configuration
currently takes place before accounting for pipe_src width for
joiner. This causes issue when pch_pfit and joiner get enabled
together.So once pipe src is computed adjust the pfit_dst.
It can be done by computing per pipe output area first and then
and then find the intersection of above area with pfit_dst and
then adjust the coordinates.
Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
---
drivers/gpu/drm/i915/display/intel_display.c | 41 ++++++++++++++++++++
1 file changed, 41 insertions(+)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 21319f753a34..7be2ea11b8b0 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
return 0;
}
+/*
+ * The x-coordinate for Primary should be calculated in such a way
+ * that it remains consistent whether the pipes are joined or not.
+ * This means we need to consider the full width of the display even
+ * when the pipes are joined. The x-coordinate for secondaries is 0
+ * because it starts at the leftmost point of its own display area,
+ * ensuring that the framebuffer is centered within Pipe B’s portion
+ * of the overall display.
+ */
+static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
+ struct intel_crtc_state *crtc_state)
+{
+ struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
+ struct drm_rect area;
+
+ if (!crtc_state->pch_pfit.enabled)
+ return 0;
+
+ drm_rect_init(&area, 0, 0,
+ mode->crtc_hdisplay,
+ mode->crtc_vdisplay);
+
+ if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
+ return -EINVAL;
+
+ drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
+
+ return 0;
+}
+
static int intel_crtc_compute_pipe_mode(struct intel_crtc_state *crtc_state)
{
struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
@@ -2645,6 +2675,10 @@ static int intel_crtc_compute_config(struct intel_atomic_state *state,
if (ret)
return ret;
+ ret = intel_crtc_compute_pfit(state, crtc_state);
+ if (ret)
+ return ret;
+
intel_crtc_compute_pixel_rate(crtc_state);
if (crtc_state->has_pch_encoder)
@@ -4865,6 +4899,13 @@ copy_joiner_crtc_state_modeset(struct intel_atomic_state *state,
drm_dp_tunnel_ref_get(primary_crtc_state->dp_tunnel_ref.tunnel,
&secondary_crtc_state->dp_tunnel_ref);
+ if (secondary_crtc_state->pch_pfit.enabled) {
+ struct drm_rect *dst = &secondary_crtc_state->pch_pfit.dst;
+ int y = dst->y1;
+
+ drm_rect_translate_to(dst, 0, y);
+ }
+
copy_joiner_crtc_state_nomodeset(state, secondary_crtc);
secondary_crtc_state->uapi.mode_changed = primary_crtc_state->uapi.mode_changed;
--
2.25.1
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
2024-12-12 14:33 [PATCH 0/2] Consider joiner calculation for panel fitting Nemesa Garg
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
@ 2024-12-12 14:33 ` Nemesa Garg
2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-12 15:52 ` ✗ i915.CI.BAT: failure for Consider joiner calculation for panel fitting (rev6) Patchwork
2 siblings, 1 reply; 12+ messages in thread
From: Nemesa Garg @ 2024-12-12 14:33 UTC (permalink / raw)
To: intel-gfx, intel-xe; +Cc: Nemesa Garg
Add early pipe src initialization for gmch and later
initialize the pipe src during compute_pipe_src.
Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
---
drivers/gpu/drm/i915/display/intel_display.c | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 7be2ea11b8b0..50693b6ca58c 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2534,6 +2534,13 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
{
struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+ int pipe_src_w, pipe_src_h;
+
+ drm_mode_get_hv_timing(&crtc_state->hw.mode,
+ &pipe_src_w, &pipe_src_h);
+
+ drm_rect_init(&crtc_state->pipe_src, 0, 0,
+ pipe_src_w, pipe_src_h);
intel_joiner_compute_pipe_src(crtc_state);
@@ -5016,10 +5023,12 @@ intel_modeset_pipe_config(struct intel_atomic_state *state,
* computation to clearly distinguish it from the adjusted mode, which
* can be changed by the connectors in the below retry loop.
*/
- drm_mode_get_hv_timing(&crtc_state->hw.mode,
- &pipe_src_w, &pipe_src_h);
- drm_rect_init(&crtc_state->pipe_src, 0, 0,
- pipe_src_w, pipe_src_h);
+ if (HAS_GMCH(i915)) {
+ drm_mode_get_hv_timing(&crtc_state->hw.mode,
+ &pipe_src_w, &pipe_src_h);
+ drm_rect_init(&crtc_state->pipe_src, 0, 0,
+ pipe_src_w, pipe_src_h);
+ }
for_each_new_connector_in_state(&state->base, connector, connector_state, i) {
struct intel_encoder *encoder =
--
2.25.1
^ permalink raw reply related [flat|nested] 12+ messages in thread
* ✗ i915.CI.BAT: failure for Consider joiner calculation for panel fitting (rev6)
2024-12-12 14:33 [PATCH 0/2] Consider joiner calculation for panel fitting Nemesa Garg
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
2024-12-12 14:33 ` [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage Nemesa Garg
@ 2024-12-12 15:52 ` Patchwork
2 siblings, 0 replies; 12+ messages in thread
From: Patchwork @ 2024-12-12 15:52 UTC (permalink / raw)
To: Nemesa Garg; +Cc: intel-gfx
== Series Details ==
Series: Consider joiner calculation for panel fitting (rev6)
URL : https://patchwork.freedesktop.org/series/136561/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_15831 -> Patchwork_136561v6
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with Patchwork_136561v6 absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_136561v6, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
to document this new failure mode, which will reduce false positives in CI.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/index.html
Participating hosts (47 -> 45)
------------------------------
Missing (2): bat-adlp-6 fi-snb-2520m
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_136561v6:
### IGT changes ###
#### Possible regressions ####
* igt@kms_flip@basic-flip-vs-wf_vblank@c-dp1:
- bat-apl-1: [PASS][1] -> [DMESG-WARN][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-apl-1/igt@kms_flip@basic-flip-vs-wf_vblank@c-dp1.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-apl-1/igt@kms_flip@basic-flip-vs-wf_vblank@c-dp1.html
Known issues
------------
Here are the changes found in Patchwork_136561v6 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@debugfs_test@basic-hwmon:
- fi-pnv-d510: NOTRUN -> [SKIP][3] +3 other tests skip
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/fi-pnv-d510/igt@debugfs_test@basic-hwmon.html
* igt@dmabuf@all-tests:
- bat-apl-1: [PASS][4] -> [INCOMPLETE][5] ([i915#12904]) +1 other test incomplete
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-apl-1/igt@dmabuf@all-tests.html
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-apl-1/igt@dmabuf@all-tests.html
* igt@gem_exec_gttfill@basic:
- fi-pnv-d510: NOTRUN -> [ABORT][6] ([i915#13169])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/fi-pnv-d510/igt@gem_exec_gttfill@basic.html
* igt@i915_selftest@live:
- bat-twl-1: [PASS][7] -> [ABORT][8] ([i915#12435] / [i915#12919])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-twl-1/igt@i915_selftest@live.html
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-twl-1/igt@i915_selftest@live.html
* igt@i915_selftest@live@requests:
- bat-twl-1: [PASS][9] -> [ABORT][10] ([i915#12919])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-twl-1/igt@i915_selftest@live@requests.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-twl-1/igt@i915_selftest@live@requests.html
* igt@kms_flip@basic-flip-vs-wf_vblank:
- bat-apl-1: [PASS][11] -> [DMESG-WARN][12] ([i915#1982])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-apl-1/igt@kms_flip@basic-flip-vs-wf_vblank.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-apl-1/igt@kms_flip@basic-flip-vs-wf_vblank.html
#### Possible fixes ####
* igt@i915_module_load@load:
- fi-pnv-d510: [ABORT][13] ([i915#13203]) -> [PASS][14]
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/fi-pnv-d510/igt@i915_module_load@load.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/fi-pnv-d510/igt@i915_module_load@load.html
* igt@i915_pm_rpm@module-reload:
- bat-dg1-7: [FAIL][15] ([i915#12903]) -> [PASS][16]
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-dg1-7/igt@i915_pm_rpm@module-reload.html
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-dg1-7/igt@i915_pm_rpm@module-reload.html
- bat-rpls-4: [FAIL][17] ([i915#12903]) -> [PASS][18]
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-rpls-4/igt@i915_pm_rpm@module-reload.html
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-rpls-4/igt@i915_pm_rpm@module-reload.html
* igt@i915_selftest@live:
- fi-hsw-4770: [DMESG-WARN][19] ([i915#12806]) -> [PASS][20] +1 other test pass
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/fi-hsw-4770/igt@i915_selftest@live.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/fi-hsw-4770/igt@i915_selftest@live.html
- bat-arlh-3: [INCOMPLETE][21] ([i915#12435] / [i915#13194]) -> [PASS][22]
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-arlh-3/igt@i915_selftest@live.html
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-arlh-3/igt@i915_selftest@live.html
* igt@i915_selftest@live@execlists:
- bat-arlh-3: [INCOMPLETE][23] -> [PASS][24]
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-arlh-3/igt@i915_selftest@live@execlists.html
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-arlh-3/igt@i915_selftest@live@execlists.html
* igt@i915_selftest@live@mman:
- bat-jsl-3: [INCOMPLETE][25] ([i915#13241]) -> [PASS][26] +1 other test pass
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-jsl-3/igt@i915_selftest@live@mman.html
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-jsl-3/igt@i915_selftest@live@mman.html
* igt@i915_selftest@live@workarounds:
- bat-arls-5: [ABORT][27] ([i915#12061]) -> [PASS][28]
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-arls-5/igt@i915_selftest@live@workarounds.html
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-arls-5/igt@i915_selftest@live@workarounds.html
* igt@kms_cursor_legacy@basic-flip-before-cursor-legacy:
- bat-apl-1: [DMESG-WARN][29] -> [PASS][30]
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_15831/bat-apl-1/igt@kms_cursor_legacy@basic-flip-before-cursor-legacy.html
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/bat-apl-1/igt@kms_cursor_legacy@basic-flip-before-cursor-legacy.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[i915#12061]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12061
[i915#12435]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12435
[i915#12806]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12806
[i915#12903]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12903
[i915#12904]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12904
[i915#12919]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12919
[i915#13169]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13169
[i915#13194]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13194
[i915#13203]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13203
[i915#13241]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13241
[i915#1982]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/1982
Build changes
-------------
* Linux: CI_DRM_15831 -> Patchwork_136561v6
CI-20190529: 20190529
CI_DRM_15831: f574af90d83d2d7bd4c99cdb79764dea5bf8b20c @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_8150: 7812065f4aebab1629b570bd78ef71e09480b359 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_136561v6: f574af90d83d2d7bd4c99cdb79764dea5bf8b20c @ git://anongit.freedesktop.org/gfx-ci/linux
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_136561v6/index.html
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
@ 2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-19 15:13 ` Garg, Nemesa
2024-12-19 16:02 ` Ville Syrjälä
1 sibling, 1 reply; 12+ messages in thread
From: Cavitt, Jonathan @ 2024-12-12 15:54 UTC (permalink / raw)
To: Garg, Nemesa, intel-gfx@lists.freedesktop.org,
intel-xe@lists.freedesktop.org
Cc: Garg, Nemesa, Cavitt, Jonathan
-----Original Message-----
From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of Nemesa Garg
Sent: Thursday, December 12, 2024 6:33 AM
To: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
Cc: Garg, Nemesa <nemesa.garg@intel.com>
Subject: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
>
> In panel fitter/pipe scaler scenario the pch_pfit configuration
> currently takes place before accounting for pipe_src width for
> joiner. This causes issue when pch_pfit and joiner get enabled
> together.So once pipe src is computed adjust the pfit_dst.
> It can be done by computing per pipe output area first and then
> and then find the intersection of above area with pfit_dst and
> then adjust the coordinates.
Maybe reword the commit message as such:
"""
In the panel fitter/pipe scaler scenario, the pch_pfit configuration
currently takes place before accounting for pipe_src width for
joiner. This causes issues when pch_pfit and joiner get enabled
together. So, once pipe_src is computed, adjust the pfit_dst.
This can be done by first computing per pipe output area, then
finding the intersection of above area with pfit_dst before
finally adjusting the coordinates.
"""
The above is just a suggested full revision. The following
is all that's strictly necessary to fix:
s/together.So/together. So/
s/area first and then and then/area first and then/
>
> Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> ---
> drivers/gpu/drm/i915/display/intel_display.c | 41 ++++++++++++++++++++
> 1 file changed, 41 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 21319f753a34..7be2ea11b8b0 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
> return 0;
> }
>
> +/*
> + * The x-coordinate for Primary should be calculated in such a way
> + * that it remains consistent whether the pipes are joined or not.
> + * This means we need to consider the full width of the display even
> + * when the pipes are joined. The x-coordinate for secondaries is 0
> + * because it starts at the leftmost point of its own display area,
> + * ensuring that the framebuffer is centered within Pipe B’s portion
Pipe B’s portion of the overall display? This is probably just a corruption
in the email due to my mail viewer of choice, but if it's not, this needs to be
fixed before pushing.
> + * of the overall display.
> + */
> +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> + struct intel_crtc_state *crtc_state)
> +{
> + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
> + struct drm_rect area;
> +
> + if (!crtc_state->pch_pfit.enabled)
> + return 0;
> +
> + drm_rect_init(&area, 0, 0,
> + mode->crtc_hdisplay,
> + mode->crtc_vdisplay);
> +
> + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> + return -EINVAL;
> +
> + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
> +
> + return 0;
> +}
> +
> static int intel_crtc_compute_pipe_mode(struct intel_crtc_state *crtc_state)
> {
> struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> @@ -2645,6 +2675,10 @@ static int intel_crtc_compute_config(struct intel_atomic_state *state,
> if (ret)
> return ret;
>
> + ret = intel_crtc_compute_pfit(state, crtc_state);
> + if (ret)
> + return ret;
> +
> intel_crtc_compute_pixel_rate(crtc_state);
>
> if (crtc_state->has_pch_encoder)
> @@ -4865,6 +4899,13 @@ copy_joiner_crtc_state_modeset(struct intel_atomic_state *state,
> drm_dp_tunnel_ref_get(primary_crtc_state->dp_tunnel_ref.tunnel,
> &secondary_crtc_state->dp_tunnel_ref);
>
> + if (secondary_crtc_state->pch_pfit.enabled) {
> + struct drm_rect *dst = &secondary_crtc_state->pch_pfit.dst;
> + int y = dst->y1;
> +
> + drm_rect_translate_to(dst, 0, y);
> + }
> +
Aside from the above minor grammatical issues:
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
-Jonathan Cavitt
> copy_joiner_crtc_state_nomodeset(state, secondary_crtc);
>
> secondary_crtc_state->uapi.mode_changed = primary_crtc_state->uapi.mode_changed;
> --
> 2.25.1
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
2024-12-12 14:33 ` [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage Nemesa Garg
@ 2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-19 15:15 ` Garg, Nemesa
0 siblings, 1 reply; 12+ messages in thread
From: Cavitt, Jonathan @ 2024-12-12 15:54 UTC (permalink / raw)
To: Garg, Nemesa, intel-gfx@lists.freedesktop.org,
intel-xe@lists.freedesktop.org
Cc: Garg, Nemesa, Cavitt, Jonathan
-----Original Message-----
From: Intel-xe <intel-xe-bounces@lists.freedesktop.org> On Behalf Of Nemesa Garg
Sent: Thursday, December 12, 2024 6:33 AM
To: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
Cc: Garg, Nemesa <nemesa.garg@intel.com>
Subject: [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
>
> Add early pipe src initialization for gmch and later
> initialize the pipe src during compute_pipe_src.
>
> Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
LGTM.
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
-Jonathan Cavitt
> ---
> drivers/gpu/drm/i915/display/intel_display.c | 17 +++++++++++++----
> 1 file changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 7be2ea11b8b0..50693b6ca58c 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -2534,6 +2534,13 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
> {
> struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> struct drm_i915_private *i915 = to_i915(crtc->base.dev);
> + int pipe_src_w, pipe_src_h;
> +
> + drm_mode_get_hv_timing(&crtc_state->hw.mode,
> + &pipe_src_w, &pipe_src_h);
> +
> + drm_rect_init(&crtc_state->pipe_src, 0, 0,
> + pipe_src_w, pipe_src_h);
>
> intel_joiner_compute_pipe_src(crtc_state);
>
> @@ -5016,10 +5023,12 @@ intel_modeset_pipe_config(struct intel_atomic_state *state,
> * computation to clearly distinguish it from the adjusted mode, which
> * can be changed by the connectors in the below retry loop.
> */
> - drm_mode_get_hv_timing(&crtc_state->hw.mode,
> - &pipe_src_w, &pipe_src_h);
> - drm_rect_init(&crtc_state->pipe_src, 0, 0,
> - pipe_src_w, pipe_src_h);
> + if (HAS_GMCH(i915)) {
> + drm_mode_get_hv_timing(&crtc_state->hw.mode,
> + &pipe_src_w, &pipe_src_h);
> + drm_rect_init(&crtc_state->pipe_src, 0, 0,
> + pipe_src_w, pipe_src_h);
> + }
>
> for_each_new_connector_in_state(&state->base, connector, connector_state, i) {
> struct intel_encoder *encoder =
> --
> 2.25.1
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-12 15:54 ` Cavitt, Jonathan
@ 2024-12-19 15:13 ` Garg, Nemesa
0 siblings, 0 replies; 12+ messages in thread
From: Garg, Nemesa @ 2024-12-19 15:13 UTC (permalink / raw)
To: Cavitt, Jonathan, intel-gfx@lists.freedesktop.org,
intel-xe@lists.freedesktop.org, Ville Syrjala
> -----Original Message-----
> From: Cavitt, Jonathan <jonathan.cavitt@intel.com>
> Sent: Thursday, December 12, 2024 9:24 PM
> To: Garg, Nemesa <nemesa.garg@intel.com>; intel-gfx@lists.freedesktop.org;
> intel-xe@lists.freedesktop.org
> Cc: Garg, Nemesa <nemesa.garg@intel.com>; Cavitt, Jonathan
> <jonathan.cavitt@intel.com>
> Subject: RE: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
>
> -----Original Message-----
> From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of
> Nemesa Garg
> Sent: Thursday, December 12, 2024 6:33 AM
> To: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
> Cc: Garg, Nemesa <nemesa.garg@intel.com>
> Subject: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
> >
> > In panel fitter/pipe scaler scenario the pch_pfit configuration
> > currently takes place before accounting for pipe_src width for joiner.
> > This causes issue when pch_pfit and joiner get enabled together.So
> > once pipe src is computed adjust the pfit_dst.
> > It can be done by computing per pipe output area first and then and
> > then find the intersection of above area with pfit_dst and then adjust
> > the coordinates.
>
> Maybe reword the commit message as such:
>
> """
> In the panel fitter/pipe scaler scenario, the pch_pfit configuration currently
> takes place before accounting for pipe_src width for joiner. This causes issues
> when pch_pfit and joiner get enabled together. So, once pipe_src is
> computed, adjust the pfit_dst.
> This can be done by first computing per pipe output area, then finding the
> intersection of above area with pfit_dst before finally adjusting the
> coordinates.
> """
>
> The above is just a suggested full revision. The following is all that's strictly
> necessary to fix:
> s/together.So/together. So/
> s/area first and then and then/area first and then/
>
> >
> > Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> > ---
> > drivers/gpu/drm/i915/display/intel_display.c | 41
> > ++++++++++++++++++++
> > 1 file changed, 41 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> > b/drivers/gpu/drm/i915/display/intel_display.c
> > index 21319f753a34..7be2ea11b8b0 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct
> intel_crtc_state *crtc_state)
> > return 0;
> > }
> >
> > +/*
> > + * The x-coordinate for Primary should be calculated in such a way
> > + * that it remains consistent whether the pipes are joined or not.
> > + * This means we need to consider the full width of the display even
> > + * when the pipes are joined. The x-coordinate for secondaries is 0
> > + * because it starts at the leftmost point of its own display area,
> > + * ensuring that the framebuffer is centered within Pipe B’s
> > +portion
>
> Pipe B’s portion of the overall display? This is probably just a corruption
> in the email due to my mail viewer of choice, but if it's not, this needs to be
> fixed before pushing.
>
> > + * of the overall display.
> > + */
> > +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> > + struct intel_crtc_state *crtc_state) {
> > + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
> > + struct drm_rect area;
> > +
> > + if (!crtc_state->pch_pfit.enabled)
> > + return 0;
> > +
> > + drm_rect_init(&area, 0, 0,
> > + mode->crtc_hdisplay,
> > + mode->crtc_vdisplay);
> > +
> > + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> > + return -EINVAL;
> > +
> > + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
> > +
> > + return 0;
> > +}
> > +
> > static int intel_crtc_compute_pipe_mode(struct intel_crtc_state
> > *crtc_state) {
> > struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> > @@ -2645,6 +2675,10 @@ static int intel_crtc_compute_config(struct
> intel_atomic_state *state,
> > if (ret)
> > return ret;
> >
> > + ret = intel_crtc_compute_pfit(state, crtc_state);
> > + if (ret)
> > + return ret;
> > +
> > intel_crtc_compute_pixel_rate(crtc_state);
> >
> > if (crtc_state->has_pch_encoder)
> > @@ -4865,6 +4899,13 @@ copy_joiner_crtc_state_modeset(struct
> intel_atomic_state *state,
> > drm_dp_tunnel_ref_get(primary_crtc_state-
> >dp_tunnel_ref.tunnel,
> > &secondary_crtc_state->dp_tunnel_ref);
> >
> > + if (secondary_crtc_state->pch_pfit.enabled) {
> > + struct drm_rect *dst = &secondary_crtc_state->pch_pfit.dst;
> > + int y = dst->y1;
> > +
> > + drm_rect_translate_to(dst, 0, y);
> > + }
> > +
>
> Aside from the above minor grammatical issues:
> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> -Jonathan Cavitt
>
Thanks for review. I'll edit the commit message.
Regards,
Nemesa
> > copy_joiner_crtc_state_nomodeset(state, secondary_crtc);
> >
> > secondary_crtc_state->uapi.mode_changed =
> > primary_crtc_state->uapi.mode_changed;
> > --
> > 2.25.1
> >
> >
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
2024-12-12 15:54 ` Cavitt, Jonathan
@ 2024-12-19 15:15 ` Garg, Nemesa
0 siblings, 0 replies; 12+ messages in thread
From: Garg, Nemesa @ 2024-12-19 15:15 UTC (permalink / raw)
To: Cavitt, Jonathan, intel-gfx@lists.freedesktop.org,
intel-xe@lists.freedesktop.org, Syrjala, Ville
> -----Original Message-----
> From: Cavitt, Jonathan <jonathan.cavitt@intel.com>
> Sent: Thursday, December 12, 2024 9:25 PM
> To: Garg, Nemesa <nemesa.garg@intel.com>; intel-gfx@lists.freedesktop.org;
> intel-xe@lists.freedesktop.org
> Cc: Garg, Nemesa <nemesa.garg@intel.com>; Cavitt, Jonathan
> <jonathan.cavitt@intel.com>
> Subject: RE: [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
>
> -----Original Message-----
> From: Intel-xe <intel-xe-bounces@lists.freedesktop.org> On Behalf Of
> Nemesa Garg
> Sent: Thursday, December 12, 2024 6:33 AM
> To: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
> Cc: Garg, Nemesa <nemesa.garg@intel.com>
> Subject: [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage
> >
> > Add early pipe src initialization for gmch and later initialize the
> > pipe src during compute_pipe_src.
> >
> > Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
>
> LGTM.
> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> -Jonathan Cavitt
Hi Ville,
Are these patches good for merge?
Regards,
Nemesa
>
> > ---
> > drivers/gpu/drm/i915/display/intel_display.c | 17 +++++++++++++----
> > 1 file changed, 13 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> > b/drivers/gpu/drm/i915/display/intel_display.c
> > index 7be2ea11b8b0..50693b6ca58c 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > @@ -2534,6 +2534,13 @@ static int intel_crtc_compute_pipe_src(struct
> > intel_crtc_state *crtc_state) {
> > struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> > struct drm_i915_private *i915 = to_i915(crtc->base.dev);
> > + int pipe_src_w, pipe_src_h;
> > +
> > + drm_mode_get_hv_timing(&crtc_state->hw.mode,
> > + &pipe_src_w, &pipe_src_h);
> > +
> > + drm_rect_init(&crtc_state->pipe_src, 0, 0,
> > + pipe_src_w, pipe_src_h);
> >
> > intel_joiner_compute_pipe_src(crtc_state);
> >
> > @@ -5016,10 +5023,12 @@ intel_modeset_pipe_config(struct
> intel_atomic_state *state,
> > * computation to clearly distinguish it from the adjusted mode, which
> > * can be changed by the connectors in the below retry loop.
> > */
> > - drm_mode_get_hv_timing(&crtc_state->hw.mode,
> > - &pipe_src_w, &pipe_src_h);
> > - drm_rect_init(&crtc_state->pipe_src, 0, 0,
> > - pipe_src_w, pipe_src_h);
> > + if (HAS_GMCH(i915)) {
> > + drm_mode_get_hv_timing(&crtc_state->hw.mode,
> > + &pipe_src_w, &pipe_src_h);
> > + drm_rect_init(&crtc_state->pipe_src, 0, 0,
> > + pipe_src_w, pipe_src_h);
> > + }
> >
> > for_each_new_connector_in_state(&state->base, connector,
> connector_state, i) {
> > struct intel_encoder *encoder =
> > --
> > 2.25.1
> >
> >
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
2024-12-12 15:54 ` Cavitt, Jonathan
@ 2024-12-19 16:02 ` Ville Syrjälä
2024-12-20 6:07 ` Garg, Nemesa
2024-12-20 9:04 ` Ville Syrjälä
1 sibling, 2 replies; 12+ messages in thread
From: Ville Syrjälä @ 2024-12-19 16:02 UTC (permalink / raw)
To: Nemesa Garg; +Cc: intel-gfx, intel-xe
On Thu, Dec 12, 2024 at 08:03:28PM +0530, Nemesa Garg wrote:
> In panel fitter/pipe scaler scenario the pch_pfit configuration
> currently takes place before accounting for pipe_src width for
> joiner. This causes issue when pch_pfit and joiner get enabled
> together.So once pipe src is computed adjust the pfit_dst.
> It can be done by computing per pipe output area first and then
> and then find the intersection of above area with pfit_dst and
> then adjust the coordinates.
>
> Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> ---
> drivers/gpu/drm/i915/display/intel_display.c | 41 ++++++++++++++++++++
> 1 file changed, 41 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 21319f753a34..7be2ea11b8b0 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
> return 0;
> }
>
> +/*
> + * The x-coordinate for Primary should be calculated in such a way
> + * that it remains consistent whether the pipes are joined or not.
> + * This means we need to consider the full width of the display even
> + * when the pipes are joined. The x-coordinate for secondaries is 0
> + * because it starts at the leftmost point of its own display area,
> + * ensuring that the framebuffer is centered within Pipe B’s portion
> + * of the overall display.
> + */
> +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> + struct intel_crtc_state *crtc_state)
> +{
> + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
const struct drm_display_mode *pipe_mode = ...
> + struct drm_rect area;
> +
> + if (!crtc_state->pch_pfit.enabled)
> + return 0;
> +
> + drm_rect_init(&area, 0, 0,
This needs a proper x offset to match the pipe's positon in
the overall screen layout (can be determined similarly to
intel_joiner_adjust_pipe_src(), except using crtc_hdisplay
as the width instead of the pipe_src width).
> + mode->crtc_hdisplay,
> + mode->crtc_vdisplay);
> +
> + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> + return -EINVAL;
> +
> + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
And this needs to remove the same offset we added originally.
> +
> + return 0;
> +}
> +
> static int intel_crtc_compute_pipe_mode(struct intel_crtc_state *crtc_state)
> {
> struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> @@ -2645,6 +2675,10 @@ static int intel_crtc_compute_config(struct intel_atomic_state *state,
> if (ret)
> return ret;
>
> + ret = intel_crtc_compute_pfit(state, crtc_state);
> + if (ret)
> + return ret;
> +
> intel_crtc_compute_pixel_rate(crtc_state);
>
> if (crtc_state->has_pch_encoder)
> @@ -4865,6 +4899,13 @@ copy_joiner_crtc_state_modeset(struct intel_atomic_state *state,
> drm_dp_tunnel_ref_get(primary_crtc_state->dp_tunnel_ref.tunnel,
> &secondary_crtc_state->dp_tunnel_ref);
>
> + if (secondary_crtc_state->pch_pfit.enabled) {
> + struct drm_rect *dst = &secondary_crtc_state->pch_pfit.dst;
> + int y = dst->y1;
> +
> + drm_rect_translate_to(dst, 0, y);
> + }
> +
> copy_joiner_crtc_state_nomodeset(state, secondary_crtc);
>
> secondary_crtc_state->uapi.mode_changed = primary_crtc_state->uapi.mode_changed;
> --
> 2.25.1
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-19 16:02 ` Ville Syrjälä
@ 2024-12-20 6:07 ` Garg, Nemesa
2024-12-20 9:04 ` Ville Syrjälä
1 sibling, 0 replies; 12+ messages in thread
From: Garg, Nemesa @ 2024-12-20 6:07 UTC (permalink / raw)
To: Ville Syrjälä
Cc: intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
> -----Original Message-----
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Sent: Thursday, December 19, 2024 9:32 PM
> To: Garg, Nemesa <nemesa.garg@intel.com>
> Cc: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
> Subject: Re: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
>
> On Thu, Dec 12, 2024 at 08:03:28PM +0530, Nemesa Garg wrote:
> > In panel fitter/pipe scaler scenario the pch_pfit configuration
> > currently takes place before accounting for pipe_src width for joiner.
> > This causes issue when pch_pfit and joiner get enabled together.So
> > once pipe src is computed adjust the pfit_dst.
> > It can be done by computing per pipe output area first and then and
> > then find the intersection of above area with pfit_dst and then adjust
> > the coordinates.
> >
> > Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> > ---
> > drivers/gpu/drm/i915/display/intel_display.c | 41
> > ++++++++++++++++++++
> > 1 file changed, 41 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> > b/drivers/gpu/drm/i915/display/intel_display.c
> > index 21319f753a34..7be2ea11b8b0 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct
> intel_crtc_state *crtc_state)
> > return 0;
> > }
> >
> > +/*
> > + * The x-coordinate for Primary should be calculated in such a way
> > + * that it remains consistent whether the pipes are joined or not.
> > + * This means we need to consider the full width of the display even
> > + * when the pipes are joined. The x-coordinate for secondaries is 0
> > + * because it starts at the leftmost point of its own display area,
> > + * ensuring that the framebuffer is centered within Pipe B’s portion
> > + * of the overall display.
> > + */
> > +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> > + struct intel_crtc_state *crtc_state) {
> > + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
>
> const struct drm_display_mode *pipe_mode = ...
>
> > + struct drm_rect area;
> > +
> > + if (!crtc_state->pch_pfit.enabled)
> > + return 0;
> > +
> > + drm_rect_init(&area, 0, 0,
>
> This needs a proper x offset to match the pipe's positon in the overall screen
> layout (can be determined similarly to intel_joiner_adjust_pipe_src(), except
> using crtc_hdisplay as the width instead of the pipe_src width).
>
Thank you for the suggestion. While implementing this code the problem was intel_crtc_compute_config()
(intel_crtc_compute_pfit is getting called from this) is getting called for primary pipe and not for secondary pipe so then I thought of finding a intersection point for primary pipe and then same thing can be translated for secondary pipe in copy_joiner_crtc_state_modeset(), but yes this will work for 2 pipes only, not for 4 pipes.
Regards,
Nemesa
> > + mode->crtc_hdisplay,
> > + mode->crtc_vdisplay);
> > +
> > + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> > + return -EINVAL;
> > +
> > + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
>
> And this needs to remove the same offset we added originally.
>
> > +
> > + return 0;
> > +}
> > +
> > static int intel_crtc_compute_pipe_mode(struct intel_crtc_state
> > *crtc_state) {
> > struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> > @@ -2645,6 +2675,10 @@ static int intel_crtc_compute_config(struct
> intel_atomic_state *state,
> > if (ret)
> > return ret;
> >
> > + ret = intel_crtc_compute_pfit(state, crtc_state);
> > + if (ret)
> > + return ret;
> > +
> > intel_crtc_compute_pixel_rate(crtc_state);
> >
> > if (crtc_state->has_pch_encoder)
> > @@ -4865,6 +4899,13 @@ copy_joiner_crtc_state_modeset(struct
> intel_atomic_state *state,
> > drm_dp_tunnel_ref_get(primary_crtc_state-
> >dp_tunnel_ref.tunnel,
> > &secondary_crtc_state->dp_tunnel_ref);
> >
> > + if (secondary_crtc_state->pch_pfit.enabled) {
> > + struct drm_rect *dst = &secondary_crtc_state->pch_pfit.dst;
> > + int y = dst->y1;
> > +
> > + drm_rect_translate_to(dst, 0, y);
> > + }
> > +
> > copy_joiner_crtc_state_nomodeset(state, secondary_crtc);
> >
> > secondary_crtc_state->uapi.mode_changed =
> > primary_crtc_state->uapi.mode_changed;
> > --
> > 2.25.1
>
> --
> Ville Syrjälä
> Intel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-19 16:02 ` Ville Syrjälä
2024-12-20 6:07 ` Garg, Nemesa
@ 2024-12-20 9:04 ` Ville Syrjälä
2025-01-29 12:54 ` Garg, Nemesa
1 sibling, 1 reply; 12+ messages in thread
From: Ville Syrjälä @ 2024-12-20 9:04 UTC (permalink / raw)
To: Nemesa Garg; +Cc: intel-gfx, intel-xe
On Thu, Dec 19, 2024 at 06:02:25PM +0200, Ville Syrjälä wrote:
> On Thu, Dec 12, 2024 at 08:03:28PM +0530, Nemesa Garg wrote:
> > In panel fitter/pipe scaler scenario the pch_pfit configuration
> > currently takes place before accounting for pipe_src width for
> > joiner. This causes issue when pch_pfit and joiner get enabled
> > together.So once pipe src is computed adjust the pfit_dst.
> > It can be done by computing per pipe output area first and then
> > and then find the intersection of above area with pfit_dst and
> > then adjust the coordinates.
> >
> > Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> > ---
> > drivers/gpu/drm/i915/display/intel_display.c | 41 ++++++++++++++++++++
> > 1 file changed, 41 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> > index 21319f753a34..7be2ea11b8b0 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
> > return 0;
> > }
> >
> > +/*
> > + * The x-coordinate for Primary should be calculated in such a way
> > + * that it remains consistent whether the pipes are joined or not.
> > + * This means we need to consider the full width of the display even
> > + * when the pipes are joined. The x-coordinate for secondaries is 0
> > + * because it starts at the leftmost point of its own display area,
> > + * ensuring that the framebuffer is centered within Pipe B’s portion
> > + * of the overall display.
> > + */
> > +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> > + struct intel_crtc_state *crtc_state)
> > +{
> > + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
>
> const struct drm_display_mode *pipe_mode = ...
>
> > + struct drm_rect area;
> > +
> > + if (!crtc_state->pch_pfit.enabled)
> > + return 0;
> > +
> > + drm_rect_init(&area, 0, 0,
>
> This needs a proper x offset to match the pipe's positon in
> the overall screen layout (can be determined similarly to
> intel_joiner_adjust_pipe_src(), except using crtc_hdisplay
> as the width instead of the pipe_src width).
>
> > + mode->crtc_hdisplay,
> > + mode->crtc_vdisplay);
> > +
> > + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> > + return -EINVAL;
> > +
> > + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
>
> And this needs to remove the same offset we added originally.
Hmm. Now that I looked at this a bit I think we need a lot
more than this to make things actually work correctly.
I think what we need is roughly:
- move pipesrc/pfit.dst/pixel_rate calculations out from
intel_crtc_compute_config() into some new function (could
call it something like intel_crtc_compute_config_final(())
which needs to be called after the joiner state copy has been
performed (ie. where we now call intel_joiner_adjust_pipe_src()
in intel_atomic_check()).
- redesign the pipesrc readout to iterate through all the joined
pipes to accumulate the x offset for the pipes, since not all
the joined pipes necessarily have the same pipesrc width
- reuse the same for the uapi.mode hdisplay/vdisplay fixup
- steal the scaled clipping code from
intel_atomic_plane_check_clipping() to properly calculate
both pipesrc and pfit.dst
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
2024-12-20 9:04 ` Ville Syrjälä
@ 2025-01-29 12:54 ` Garg, Nemesa
0 siblings, 0 replies; 12+ messages in thread
From: Garg, Nemesa @ 2025-01-29 12:54 UTC (permalink / raw)
To: Ville Syrjälä
Cc: intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
> -----Original Message-----
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Sent: Friday, December 20, 2024 2:35 PM
> To: Garg, Nemesa <nemesa.garg@intel.com>
> Cc: intel-gfx@lists.freedesktop.org; intel-xe@lists.freedesktop.org
> Subject: Re: [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst
>
> On Thu, Dec 19, 2024 at 06:02:25PM +0200, Ville Syrjälä wrote:
> > On Thu, Dec 12, 2024 at 08:03:28PM +0530, Nemesa Garg wrote:
> > > In panel fitter/pipe scaler scenario the pch_pfit configuration
> > > currently takes place before accounting for pipe_src width for
> > > joiner. This causes issue when pch_pfit and joiner get enabled
> > > together.So once pipe src is computed adjust the pfit_dst.
> > > It can be done by computing per pipe output area first and then and
> > > then find the intersection of above area with pfit_dst and then
> > > adjust the coordinates.
> > >
> > > Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
> > > ---
> > > drivers/gpu/drm/i915/display/intel_display.c | 41
> > > ++++++++++++++++++++
> > > 1 file changed, 41 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> > > b/drivers/gpu/drm/i915/display/intel_display.c
> > > index 21319f753a34..7be2ea11b8b0 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > > @@ -2563,6 +2563,36 @@ static int intel_crtc_compute_pipe_src(struct
> intel_crtc_state *crtc_state)
> > > return 0;
> > > }
> > >
> > > +/*
> > > + * The x-coordinate for Primary should be calculated in such a way
> > > + * that it remains consistent whether the pipes are joined or not.
> > > + * This means we need to consider the full width of the display
> > > +even
> > > + * when the pipes are joined. The x-coordinate for secondaries is 0
> > > + * because it starts at the leftmost point of its own display area,
> > > + * ensuring that the framebuffer is centered within Pipe B’s
> > > +portion
> > > + * of the overall display.
> > > + */
> > > +static int intel_crtc_compute_pfit(struct intel_atomic_state *state,
> > > + struct intel_crtc_state *crtc_state) {
> > > + struct drm_display_mode *mode = &crtc_state->hw.pipe_mode;
> >
> > const struct drm_display_mode *pipe_mode = ...
> >
> > > + struct drm_rect area;
> > > +
> > > + if (!crtc_state->pch_pfit.enabled)
> > > + return 0;
> > > +
> > > + drm_rect_init(&area, 0, 0,
> >
> > This needs a proper x offset to match the pipe's positon in the
> > overall screen layout (can be determined similarly to
> > intel_joiner_adjust_pipe_src(), except using crtc_hdisplay as the
> > width instead of the pipe_src width).
> >
> > > + mode->crtc_hdisplay,
> > > + mode->crtc_vdisplay);
> > > +
> > > + if (!drm_rect_intersect(&crtc_state->pch_pfit.dst, &area))
> > > + return -EINVAL;
> > > +
> > > + drm_rect_translate(&crtc_state->pch_pfit.dst, -area.x1, -area.y1);
> >
> > And this needs to remove the same offset we added originally.
>
> Hmm. Now that I looked at this a bit I think we need a lot more than this to
> make things actually work correctly.
>
> I think what we need is roughly:
> - move pipesrc/pfit.dst/pixel_rate calculations out from
> intel_crtc_compute_config() into some new function (could
> call it something like intel_crtc_compute_config_final(())
> which needs to be called after the joiner state copy has been
> performed (ie. where we now call intel_joiner_adjust_pipe_src()
> in intel_atomic_check()).
> - redesign the pipesrc readout to iterate through all the joined
> pipes to accumulate the x offset for the pipes, since not all
> the joined pipes necessarily have the same pipesrc width
> - reuse the same for the uapi.mode hdisplay/vdisplay fixup
> - steal the scaled clipping code from
> intel_atomic_plane_check_clipping() to properly calculate
> both pipesrc and pfit.dst
>
Hi Ville,
Sorry for late reply. Agree it needs more work and I got your points. I will work on this and will send the new revision.
Thanks and Regards,
Nemesa
> --
> Ville Syrjälä
> Intel
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2025-01-29 12:54 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-12-12 14:33 [PATCH 0/2] Consider joiner calculation for panel fitting Nemesa Garg
2024-12-12 14:33 ` [PATCH 1/2] drm/i915/display: After joiner compute pfit_dst Nemesa Garg
2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-19 15:13 ` Garg, Nemesa
2024-12-19 16:02 ` Ville Syrjälä
2024-12-20 6:07 ` Garg, Nemesa
2024-12-20 9:04 ` Ville Syrjälä
2025-01-29 12:54 ` Garg, Nemesa
2024-12-12 14:33 ` [PATCH 2/2] drm/i915/display: Initialize pipe_src in compute stage Nemesa Garg
2024-12-12 15:54 ` Cavitt, Jonathan
2024-12-19 15:15 ` Garg, Nemesa
2024-12-12 15:52 ` ✗ i915.CI.BAT: failure for Consider joiner calculation for panel fitting (rev6) Patchwork
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).