[PATCH 0/6] drm: ci: fixes

[PATCH 0/6] drm: ci: fixes

[Vignesh Raman]

The patch series contains improvements, enabling new ci jobs which
enables testing for Mediatek MT8173, Qualcomm APQ 8016 and VirtIO GPU,
fixing issues with the ci jobs and updating the expectation files.
This series is intended for drm branch topic/drm-ci.

Vignesh Raman (6):
  drm: ci: igt_runner: remove todo
  drm: ci: Force db410c to host mode
  drm: ci: virtio: update ci variables
  drm: ci: Enable configs to fix mt8173 boot hang issue
  drm: ci: Update xfails
  drm: ci: Remove rules

 drivers/gpu/drm/ci/arm64.config                  |  2 ++
 drivers/gpu/drm/ci/build.sh                      |  4 ++++
 drivers/gpu/drm/ci/gitlab-ci.yml                 |  2 +-
 drivers/gpu/drm/ci/igt_runner.sh                 |  1 -
 drivers/gpu/drm/ci/test.yml                      | 14 ++++----------
 .../gpu/drm/ci/xfails/amdgpu-stoney-fails.txt    |  1 -
 drivers/gpu/drm/ci/xfails/i915-cml-fails.txt     |  1 -
 drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt    |  2 ++
 drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt    |  1 +
 .../gpu/drm/ci/xfails/mediatek-mt8173-fails.txt  |  2 --
 .../gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt | 16 ++++++++++++++++
 drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt |  2 ++
 .../gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt |  1 +
 .../gpu/drm/ci/xfails/rockchip-rk3399-fails.txt  |  4 ++--
 .../gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt |  3 +++
 15 files changed, 38 insertions(+), 18 deletions(-)

-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 1/6] drm: ci: igt_runner: remove todo

[Vignesh Raman]

/sys/kernel/debug/dri/*/state exist for every atomic KMS driver.
We do not test non-atomic drivers, so remove the todo.

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/gpu/drm/ci/igt_runner.sh | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh
index 2bb759165063..5bf130ac57c9 100755
--- a/drivers/gpu/drm/ci/igt_runner.sh
+++ b/drivers/gpu/drm/ci/igt_runner.sh
@@ -15,7 +15,6 @@ cat /sys/kernel/debug/device_component/*
 '
 
 # Dump drm state to confirm that kernel was able to find a connected display:
-# TODO this path might not exist for all drivers.. maybe run modetest instead?
 set +e
 cat /sys/kernel/debug/dri/*/state
 set -e
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 2/6] drm: ci: Force db410c to host mode

[Vignesh Raman]

Force db410c to host mode to fix network issue which results in failure
to mount root fs via NFS.
See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8

Since this fix is not sent upstream, add it to build.sh script
before building the kernel and dts. Better approach would be
to use devicetree overlays.

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/gpu/drm/ci/build.sh | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
index 7b014287a041..c39834bd6bd7 100644
--- a/drivers/gpu/drm/ci/build.sh
+++ b/drivers/gpu/drm/ci/build.sh
@@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
     fi
 fi
 
+# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
+# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
+sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
+
 for opt in $ENABLE_KCONFIGS; do
   echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
 done
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 3/6] drm: ci: virtio: update ci variables

[Vignesh Raman]

Update ci variables to fix the below error,
ERROR - Igt error: malloc(): corrupted top size
ERROR - Igt error: Received signal SIGABRT.
ERROR - Igt error: Stack trace:
ERROR - Igt error:  #0 [fatal_sig_handler+0x17b]

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/gpu/drm/ci/test.yml | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml
index 6473cddaa7a9..d85add39f425 100644
--- a/drivers/gpu/drm/ci/test.yml
+++ b/drivers/gpu/drm/ci/test.yml
@@ -316,8 +316,11 @@ virtio_gpu:none:
   stage: virtio-gpu
   variables:
     CROSVM_GALLIUM_DRIVER: llvmpipe
-    DRIVER_NAME: virtio_gpu
+    DRIVER_NAME: virtio
     GPU_VERSION: none
+    CROSVM_MEMORY: 12288
+    CROSVM_CPU: $FDO_CI_CONCURRENT
+    CROSVM_GPU_ARGS: "vulkan=true,gles=false,backend=virglrenderer,egl=true,surfaceless=true"
   extends:
     - .test-gl
   tags:
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 4/6] drm: ci: Enable configs to fix mt8173 boot hang issue

[Vignesh Raman]

Enable regulator
Enable MT6397 RTC driver

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/gpu/drm/ci/arm64.config | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/ci/arm64.config b/drivers/gpu/drm/ci/arm64.config
index 817e18ddfd4f..ea7a6cceff40 100644
--- a/drivers/gpu/drm/ci/arm64.config
+++ b/drivers/gpu/drm/ci/arm64.config
@@ -184,6 +184,8 @@ CONFIG_HW_RANDOM_MTK=y
 CONFIG_MTK_DEVAPC=y
 CONFIG_PWM_MTK_DISP=y
 CONFIG_MTK_CMDQ=y
+CONFIG_REGULATOR_DA9211=y
+CONFIG_RTC_DRV_MT6397=y
 
 # For nouveau.  Note that DRM must be a module so that it's loaded after NFS is up to provide the firmware.
 CONFIG_ARCH_TEGRA=y
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 5/6] drm: ci: Update xfails

[Vignesh Raman]

Update amdgpu-stoney-fails, mediatek-mt8173-flakes,
mediatek-mt8173-fails, rockchip-rk3399-fails, rockchip-rk3399-flakes,
rockchip-rk3288-flakes, i915-cml-fails, i915-cml-flakes,
msm-apq8016-flakes files.

Add tests that fail sometimes into the *-flakes file and tests
that are failing into the *-fails file.

Signed-off-by: Helen Koike <helen.koike@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 .../gpu/drm/ci/xfails/amdgpu-stoney-fails.txt    |  1 -
 drivers/gpu/drm/ci/xfails/i915-cml-fails.txt     |  1 -
 drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt    |  2 ++
 drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt    |  1 +
 .../gpu/drm/ci/xfails/mediatek-mt8173-fails.txt  |  2 --
 .../gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt | 16 ++++++++++++++++
 drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt |  2 ++
 .../gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt |  1 +
 .../gpu/drm/ci/xfails/rockchip-rk3399-fails.txt  |  4 ++--
 .../gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt |  3 +++
 10 files changed, 27 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
index bd9392536e7c..58bfded8a3fc 100644
--- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
@@ -1,7 +1,6 @@
 kms_addfb_basic@bad-pitch-65536,Fail
 kms_addfb_basic@bo-too-small,Fail
 kms_async_flips@invalid-async-flip,Fail
-kms_atomic@plane-immutable-zpos,Fail
 kms_atomic_transition@plane-toggle-modeset-transition,Fail
 kms_bw@linear-tiling-1-displays-2560x1440p,Fail
 kms_bw@linear-tiling-1-displays-3840x2160p,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
index 6139b410e767..5f513c638beb 100644
--- a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
@@ -1,4 +1,3 @@
-kms_color@ctm-0-25,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt
index 0514a7b3fdb0..f06f1a5b16f9 100644
--- a/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt
@@ -7,6 +7,8 @@ kms_bw@linear-tiling-3-displays-3840x2160p
 kms_bw@linear-tiling-4-displays-1920x1080p
 kms_bw@linear-tiling-4-displays-2560x1440p
 kms_bw@linear-tiling-4-displays-3840x2160p
+kms_color@ctm-0-25
+kms_cursor_legacy@torture-move
 kms_draw_crc@draw-method-xrgb8888-render-xtiled
 kms_flip@flip-vs-suspend
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling
diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt
index fc41d13a2d56..3aee1f11ee90 100644
--- a/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt
@@ -8,6 +8,7 @@ kms_bw@linear-tiling-3-displays-3840x2160p
 kms_bw@linear-tiling-4-displays-1920x1080p
 kms_bw@linear-tiling-4-displays-2560x1440p
 kms_bw@linear-tiling-4-displays-3840x2160p
+kms_cursor_legacy@torture-bo
 kms_flip@blocking-wf_vblank
 kms_flip@wf_vblank-ts-check
 kms_flip@wf_vblank-ts-check-interruptible
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
index 671916067dba..c8e64bbfd480 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
@@ -1,5 +1,4 @@
 kms_3d,Fail
-kms_addfb_basic@addfb25-bad-modifier,Fail
 kms_bw@linear-tiling-1-displays-1920x1080p,Fail
 kms_bw@linear-tiling-1-displays-2560x1440p,Fail
 kms_bw@linear-tiling-1-displays-3840x2160p,Fail
@@ -11,7 +10,6 @@ kms_bw@linear-tiling-3-displays-2560x1440p,Fail
 kms_bw@linear-tiling-3-displays-3840x2160p,Fail
 kms_color@pipe-A-invalid-gamma-lut-sizes,Fail
 kms_color@pipe-B-invalid-gamma-lut-sizes,Fail
-kms_force_connector_basic@force-connector-state,Fail
 kms_force_connector_basic@force-edid,Fail
 kms_force_connector_basic@force-load-detect,Fail
 kms_force_connector_basic@prune-stale-modes,Fail
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
index e69de29bb2d1..9ed6722df2c2 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
@@ -0,0 +1,16 @@
+core_setmaster_vs_auth
+kms_addfb_basic@addfb25-bad-modifier
+kms_color@invalid-gamma-lut-sizes
+kms_cursor_legacy@cursor-vs-flip-atomic
+kms_cursor_legacy@cursor-vs-flip-legacy
+kms_force_connector_basic@force-connector-state
+kms_hdmi_inject@inject-4k
+kms_plane_scaling@plane-scaler-with-pixel-format-unity-scaling
+kms_plane_scaling@plane-upscale-with-modifiers-20x20
+kms_plane_scaling@plane-upscale-with-pixel-format-20x20
+kms_plane_scaling@plane-upscale-with-rotation-20x20
+kms_plane_scaling@planes-downscale-factor-0-25-upscale-20x20
+kms_plane_scaling@planes-downscale-factor-0-5-upscale-20x20
+kms_plane_scaling@planes-downscale-factor-0-75-upscale-20x20
+kms_prop_blob@invalid-set-prop
+kms_prop_blob@invalid-set-prop-any
diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt
index 0e3b60d3fade..cd48a8c1d569 100644
--- a/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt
@@ -1,3 +1,5 @@
+kms_cursor_legacy@forked-bo
+kms_cursor_legacy@forked-move
 kms_force_connector_basic@force-connector-state
 kms_force_connector_basic@force-edid
 kms_force_connector_basic@force-load-detect
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt
index 45c54c75c899..081fe6f8d488 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt
@@ -1,6 +1,7 @@
 kms_addfb_basic@addfb25-bad-modifier
 kms_cursor_crc@.*
 kms_flip@basic-flip-vs-wf_vblank
+kms_flip@blocking-wf_vblank
 kms_invalid_mode@int-max-clock,Crash
 kms_pipe_crc_basic@.*
 kms_properties@connector-properties-atomic,Crash
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
index 6db08ba6b008..029a3be3164d 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
@@ -2,14 +2,13 @@ kms_color@legacy-gamma,Fail
 kms_color@pipe-A-legacy-gamma,Fail
 kms_color@pipe-B-legacy-gamma,Fail
 kms_flip@basic-flip-vs-wf_vblank,Fail
-kms_flip@blocking-wf_vblank,Fail
 kms_flip@dpms-vs-vblank-race,Fail
 kms_flip@flip-vs-absolute-wf_vblank,Fail
 kms_flip@flip-vs-absolute-wf_vblank-interruptible,Fail
 kms_flip@flip-vs-blocking-wf-vblank,Fail
 kms_flip@flip-vs-panning,Fail
 kms_flip@flip-vs-panning-interruptible,Fail
-kms_flip@flip-vs-wf_vblank-interruptible,Fail
+kms_flip@modeset-vs-vblank-race,Fail
 kms_flip@plain-flip-fb-recreate,Fail
 kms_flip@plain-flip-fb-recreate-interruptible,Fail
 kms_flip@plain-flip-ts-check,Fail
@@ -35,3 +34,4 @@ kms_plane_multiple@atomic-pipe-B-tiling-none,Fail
 kms_plane_multiple@tiling-none,Fail
 kms_prime@basic-crc,Fail
 kms_rmfb@close-fd,Fail
+kms_universal_plane@universal-plane-pipe-B-functional,Fail
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
index 4c0539b4beaf..869426a02cbd 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
@@ -1,7 +1,10 @@
 
+kms_color@gamma
 kms_cursor_crc@.*
+kms_flip@blocking-wf_vblank
 kms_flip@dpms-vs-vblank-race-interruptible
 kms_flip@flip-vs-expired-vblank
+kms_flip@flip-vs-wf_vblank-interruptible
 kms_flip@modeset-vs-vblank-race-interruptible
 kms_pipe_crc_basic@.*
 kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-A
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

[PATCH 6/6] drm: ci: Remove rules

[Vignesh Raman]

Remove the rules from the following jobs, as the issues noted in the
TODO comments have been resolved. This will ensure that these jobs
are now included and executed as part of the CI/CD pipeline.

msm:apq8016:
TODO: current issue: it is not fiding the NFS root. Fix and remove this rule.

mediatek:mt8173:
TODO: current issue: device is hanging. Fix and remove this rule.

virtio_gpu:none:
TODO: current issue: malloc(): corrupted top size. Fix and remove this rule.

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/gpu/drm/ci/gitlab-ci.yml | 2 +-
 drivers/gpu/drm/ci/test.yml      | 9 ---------
 2 files changed, 1 insertion(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml
index 2c4df53f5dfe..d2aac4404914 100644
--- a/drivers/gpu/drm/ci/gitlab-ci.yml
+++ b/drivers/gpu/drm/ci/gitlab-ci.yml
@@ -248,4 +248,4 @@ sanity:
 
 # Jobs that need to pass before spending hardware resources on further testing
 .required-for-hardware-jobs:
-  needs: []
\ No newline at end of file
+  needs: []
diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml
index d85add39f425..1771af21e2d9 100644
--- a/drivers/gpu/drm/ci/test.yml
+++ b/drivers/gpu/drm/ci/test.yml
@@ -108,9 +108,6 @@ msm:apq8016:
     RUNNER_TAG: google-freedreno-db410c
   script:
     - ./install/bare-metal/fastboot.sh
-  rules:
-    # TODO: current issue: it is not fiding the NFS root. Fix and remove this rule.
-    - when: never
 
 msm:apq8096:
   extends:
@@ -273,9 +270,6 @@ mediatek:mt8173:
     DEVICE_TYPE: mt8173-elm-hana
     GPU_VERSION: mt8173
     RUNNER_TAG: mesa-ci-x86-64-lava-mt8173-elm-hana
-  rules:
-    # TODO: current issue: device is hanging. Fix and remove this rule.
-    - when: never
 
 mediatek:mt8183:
   extends:
@@ -333,6 +327,3 @@ virtio_gpu:none:
     - debian/x86_64_test-gl
     - testing:x86_64
     - igt:x86_64
-  rules:
-    # TODO: current issue: malloc(): corrupted top size. Fix and remove this rule.
-    - when: never
\ No newline at end of file
-- 
2.40.1


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Jani Nikula]

On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> Force db410c to host mode to fix network issue which results in failure
> to mount root fs via NFS.
> See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
>
> Since this fix is not sent upstream, add it to build.sh script
> before building the kernel and dts. Better approach would be
> to use devicetree overlays.
>
> Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> ---
>  drivers/gpu/drm/ci/build.sh | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> index 7b014287a041..c39834bd6bd7 100644
> --- a/drivers/gpu/drm/ci/build.sh
> +++ b/drivers/gpu/drm/ci/build.sh
> @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
>      fi
>  fi
>  
> +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> +

It seems like a really bad idea to me to have the CI build modify the
source tree before building.

The kernel being built will have a dirty git repo, and the localversion
will have -dirty in it.

I think it would be better to do out-of-tree builds and assume the
source is read-only.

>  for opt in $ENABLE_KCONFIGS; do
>    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
>  done

Ditto for the config changes in the context here. Those are files in
git, don't change them.

Shouldn't this use something like 'scripts/config --enable' or
'scripts/config --disable' on the .config file to be used for building
instead?


BR,
Jani.


-- 
Jani Nikula, Intel Open Source Graphics Center

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Helen Mae Koike Fornazier]

Hi Jani, thanks for your comments

On Friday, August 25, 2023 10:56 -03, Jani Nikula <jani.nikula@linux.intel.com> wrote:

> On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > Force db410c to host mode to fix network issue which results in failure
> > to mount root fs via NFS.
> > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> >
> > Since this fix is not sent upstream, add it to build.sh script
> > before building the kernel and dts. Better approach would be
> > to use devicetree overlays.
> >
> > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > ---
> >  drivers/gpu/drm/ci/build.sh | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > index 7b014287a041..c39834bd6bd7 100644
> > --- a/drivers/gpu/drm/ci/build.sh
> > +++ b/drivers/gpu/drm/ci/build.sh
> > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> >      fi
> >  fi
> >  
> > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > +
> 
> It seems like a really bad idea to me to have the CI build modify the
> source tree before building.
> 
> The kernel being built will have a dirty git repo, and the localversion
> will have -dirty in it.

Is it bad?

The other option was to work with device tree overlays (but we still need to spend some time to
see how to fit it all together)

> 
> I think it would be better to do out-of-tree builds and assume the
> source is read-only.

I'm not sure I get what do you call out-of-tree builds.

Another option would be to apply .patch file, or to have another branch
just with fix ups for ci that would be applied in the tree before building.

> 
> >  for opt in $ENABLE_KCONFIGS; do
> >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> >  done
> 
> Ditto for the config changes in the context here. Those are files in
> git, don't change them.

Probably these changes could go directly into drivers/gpu/drm/ci/${KERNEL_ARCH}.config
files, no need to modify them on the fly here

> 
> Shouldn't this use something like 'scripts/config --enable' or
> 'scripts/config --disable' on the .config file to be used for building
> instead?

I wasn't aware about this possibility, looks cleaner indeed.

Regards,
Helen

> 
> 
> BR,
> Jani.
> 
> 
> -- 
> Jani Nikula, Intel Open Source Graphics Center


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Maxime Ripard]

[-- Attachment #1.1: Type: text/plain, Size: 2106 bytes --]

Hi Helen,

On Fri, Aug 25, 2023 at 03:09:04PM +0100, Helen Mae Koike Fornazier wrote:
> Hi Jani, thanks for your comments
> 
> On Friday, August 25, 2023 10:56 -03, Jani Nikula <jani.nikula@linux.intel.com> wrote:
> 
> > On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > > Force db410c to host mode to fix network issue which results in failure
> > > to mount root fs via NFS.
> > > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > >
> > > Since this fix is not sent upstream, add it to build.sh script
> > > before building the kernel and dts. Better approach would be
> > > to use devicetree overlays.
> > >
> > > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > > ---
> > >  drivers/gpu/drm/ci/build.sh | 4 ++++
> > >  1 file changed, 4 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > > index 7b014287a041..c39834bd6bd7 100644
> > > --- a/drivers/gpu/drm/ci/build.sh
> > > +++ b/drivers/gpu/drm/ci/build.sh
> > > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> > >      fi
> > >  fi
> > >  
> > > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > > +
> > 
> > It seems like a really bad idea to me to have the CI build modify the
> > source tree before building.
> > 
> > The kernel being built will have a dirty git repo, and the localversion
> > will have -dirty in it.
> 
> Is it bad?
> 
> The other option was to work with device tree overlays (but we still
> need to spend some time to see how to fit it all together)

That would be much better. libfdt provides an fdtoverlay command to
merge a base device tree with an overlay.

Do that while setting up the DUT and you're done :)

Maxime

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

[-- Attachment #2: Type: text/plain, Size: 167 bytes --]

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Rob Clark]

On Fri, Aug 25, 2023 at 6:56 AM Jani Nikula <jani.nikula@linux.intel.com> wrote:
>
> On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > Force db410c to host mode to fix network issue which results in failure
> > to mount root fs via NFS.
> > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> >
> > Since this fix is not sent upstream, add it to build.sh script
> > before building the kernel and dts. Better approach would be
> > to use devicetree overlays.
> >
> > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > ---
> >  drivers/gpu/drm/ci/build.sh | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > index 7b014287a041..c39834bd6bd7 100644
> > --- a/drivers/gpu/drm/ci/build.sh
> > +++ b/drivers/gpu/drm/ci/build.sh
> > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> >      fi
> >  fi
> >
> > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > +
>
> It seems like a really bad idea to me to have the CI build modify the
> source tree before building.
>
> The kernel being built will have a dirty git repo, and the localversion
> will have -dirty in it.
>
> I think it would be better to do out-of-tree builds and assume the
> source is read-only.

We have the ${target_branch}-external-fixes mechanism to merge
necessary changes before building the kernel for CI.  Which is
necessary for a couple of reasons:

1) patches like this which aren't appropriate upstream but necessary
due to the CI lab setup
2) target branch if often based on an early -rc, and it isn't unheard
of to need some fix for some board or another which isn't appropriate
to land via drm-next

We should use the -external-fixes branch mechanism for patches like this one.

BR,
-R

> >  for opt in $ENABLE_KCONFIGS; do
> >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> >  done
>
> Ditto for the config changes in the context here. Those are files in
> git, don't change them.
>
> Shouldn't this use something like 'scripts/config --enable' or
> 'scripts/config --disable' on the .config file to be used for building
> instead?
>
>
> BR,
> Jani.
>
>
> --
> Jani Nikula, Intel Open Source Graphics Center

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Helen Mae Koike Fornazier]

On Friday, August 25, 2023 11:30 -03, Rob Clark <robdclark@gmail.com> wrote:

> On Fri, Aug 25, 2023 at 6:56 AM Jani Nikula <jani.nikula@linux.intel.com> wrote:
> >
> > On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > > Force db410c to host mode to fix network issue which results in failure
> > > to mount root fs via NFS.
> > > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > >
> > > Since this fix is not sent upstream, add it to build.sh script
> > > before building the kernel and dts. Better approach would be
> > > to use devicetree overlays.
> > >
> > > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > > ---
> > >  drivers/gpu/drm/ci/build.sh | 4 ++++
> > >  1 file changed, 4 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > > index 7b014287a041..c39834bd6bd7 100644
> > > --- a/drivers/gpu/drm/ci/build.sh
> > > +++ b/drivers/gpu/drm/ci/build.sh
> > > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> > >      fi
> > >  fi
> > >
> > > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > > +
> >
> > It seems like a really bad idea to me to have the CI build modify the
> > source tree before building.
> >
> > The kernel being built will have a dirty git repo, and the localversion
> > will have -dirty in it.
> >
> > I think it would be better to do out-of-tree builds and assume the
> > source is read-only.
> 
> We have the ${target_branch}-external-fixes mechanism to merge
> necessary changes before building the kernel for CI.  Which is
> necessary for a couple of reasons:

Should we create an official topic/drm-ci-external-fixes branch ?

Regards,
Helen

> 
> 1) patches like this which aren't appropriate upstream but necessary
> due to the CI lab setup
> 2) target branch if often based on an early -rc, and it isn't unheard
> of to need some fix for some board or another which isn't appropriate
> to land via drm-next
> 
> We should use the -external-fixes branch mechanism for patches like this one.
> 
> BR,
> -R
> 
> > >  for opt in $ENABLE_KCONFIGS; do
> > >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> > >  done
> >
> > Ditto for the config changes in the context here. Those are files in
> > git, don't change them.
> >
> > Shouldn't this use something like 'scripts/config --enable' or
> > 'scripts/config --disable' on the .config file to be used for building
> > instead?
> >
> >
> > BR,
> > Jani.
> >
> >
> > --
> > Jani Nikula, Intel Open Source Graphics Center


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Rob Clark]

On Fri, Aug 25, 2023 at 7:34 AM Helen Mae Koike Fornazier
<helen.koike@collabora.com> wrote:
>
> On Friday, August 25, 2023 11:30 -03, Rob Clark <robdclark@gmail.com> wrote:
>
> > On Fri, Aug 25, 2023 at 6:56 AM Jani Nikula <jani.nikula@linux.intel.com> wrote:
> > >
> > > On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > > > Force db410c to host mode to fix network issue which results in failure
> > > > to mount root fs via NFS.
> > > > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > >
> > > > Since this fix is not sent upstream, add it to build.sh script
> > > > before building the kernel and dts. Better approach would be
> > > > to use devicetree overlays.
> > > >
> > > > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > > > ---
> > > >  drivers/gpu/drm/ci/build.sh | 4 ++++
> > > >  1 file changed, 4 insertions(+)
> > > >
> > > > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > > > index 7b014287a041..c39834bd6bd7 100644
> > > > --- a/drivers/gpu/drm/ci/build.sh
> > > > +++ b/drivers/gpu/drm/ci/build.sh
> > > > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> > > >      fi
> > > >  fi
> > > >
> > > > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > > > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > > > +
> > >
> > > It seems like a really bad idea to me to have the CI build modify the
> > > source tree before building.
> > >
> > > The kernel being built will have a dirty git repo, and the localversion
> > > will have -dirty in it.
> > >
> > > I think it would be better to do out-of-tree builds and assume the
> > > source is read-only.
> >
> > We have the ${target_branch}-external-fixes mechanism to merge
> > necessary changes before building the kernel for CI.  Which is
> > necessary for a couple of reasons:
>
> Should we create an official topic/drm-ci-external-fixes branch ?

Hmm, maybe.. I guess as we expand this to more driver trees, and want
to be able to re-run CI in the drm tree after merges to
drm-next/drm-fixes, we maybe want to have central
drm-next-external-fixes and drm-fixes-external-fixes.  I guess we can
keep those based on drm-next and drm-fixes?  And if there would be
conflicts because, say, ${driver}-next is behind drm-next, then
${driver}-next could be rebased on drm-next?

BR,
-R

> Regards,
> Helen
>
> >
> > 1) patches like this which aren't appropriate upstream but necessary
> > due to the CI lab setup
> > 2) target branch if often based on an early -rc, and it isn't unheard
> > of to need some fix for some board or another which isn't appropriate
> > to land via drm-next
> >
> > We should use the -external-fixes branch mechanism for patches like this one.
> >
> > BR,
> > -R
> >
> > > >  for opt in $ENABLE_KCONFIGS; do
> > > >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> > > >  done
> > >
> > > Ditto for the config changes in the context here. Those are files in
> > > git, don't change them.
> > >
> > > Shouldn't this use something like 'scripts/config --enable' or
> > > 'scripts/config --disable' on the .config file to be used for building
> > > instead?
> > >
> > >
> > > BR,
> > > Jani.
> > >
> > >
> > > --
> > > Jani Nikula, Intel Open Source Graphics Center
>

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Helen Mae Koike Fornazier]

On Friday, August 25, 2023 11:41 -03, Rob Clark <robdclark@gmail.com> wrote:

> On Fri, Aug 25, 2023 at 7:34 AM Helen Mae Koike Fornazier
> <helen.koike@collabora.com> wrote:
> >
> > On Friday, August 25, 2023 11:30 -03, Rob Clark <robdclark@gmail.com> wrote:
> >
> > > On Fri, Aug 25, 2023 at 6:56 AM Jani Nikula <jani.nikula@linux.intel.com> wrote:
> > > >
> > > > On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > > > > Force db410c to host mode to fix network issue which results in failure
> > > > > to mount root fs via NFS.
> > > > > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > > >
> > > > > Since this fix is not sent upstream, add it to build.sh script
> > > > > before building the kernel and dts. Better approach would be
> > > > > to use devicetree overlays.
> > > > >
> > > > > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > > > > ---
> > > > >  drivers/gpu/drm/ci/build.sh | 4 ++++
> > > > >  1 file changed, 4 insertions(+)
> > > > >
> > > > > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > > > > index 7b014287a041..c39834bd6bd7 100644
> > > > > --- a/drivers/gpu/drm/ci/build.sh
> > > > > +++ b/drivers/gpu/drm/ci/build.sh
> > > > > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> > > > >      fi
> > > > >  fi
> > > > >
> > > > > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > > > > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > > > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > > > > +
> > > >
> > > > It seems like a really bad idea to me to have the CI build modify the
> > > > source tree before building.
> > > >
> > > > The kernel being built will have a dirty git repo, and the localversion
> > > > will have -dirty in it.
> > > >
> > > > I think it would be better to do out-of-tree builds and assume the
> > > > source is read-only.
> > >
> > > We have the ${target_branch}-external-fixes mechanism to merge
> > > necessary changes before building the kernel for CI.  Which is
> > > necessary for a couple of reasons:
> >
> > Should we create an official topic/drm-ci-external-fixes branch ?
> 
> Hmm, maybe.. I guess as we expand this to more driver trees, and want
> to be able to re-run CI in the drm tree after merges to
> drm-next/drm-fixes, we maybe want to have central
> drm-next-external-fixes and drm-fixes-external-fixes.  I guess we can
> keep those based on drm-next and drm-fixes?  And if there would be
> conflicts because, say, ${driver}-next is behind drm-next, then
> ${driver}-next could be rebased on drm-next?
> 

tbh this is one of the reasons I would prefer in-code fixes instead of
commits on a -external-fixes branch, since it seems things start to become
complex to manage all different trees for people executing ci tests
on different history points, but I don't oppose going for -external-fixes
either.

Regards,
Helen

> BR,
> -R
> 
> > Regards,
> > Helen
> >
> > >
> > > 1) patches like this which aren't appropriate upstream but necessary
> > > due to the CI lab setup
> > > 2) target branch if often based on an early -rc, and it isn't unheard
> > > of to need some fix for some board or another which isn't appropriate
> > > to land via drm-next
> > >
> > > We should use the -external-fixes branch mechanism for patches like this one.
> > >
> > > BR,
> > > -R
> > >
> > > > >  for opt in $ENABLE_KCONFIGS; do
> > > > >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> > > > >  done
> > > >
> > > > Ditto for the config changes in the context here. Those are files in
> > > > git, don't change them.
> > > >
> > > > Shouldn't this use something like 'scripts/config --enable' or
> > > > 'scripts/config --disable' on the .config file to be used for building
> > > > instead?
> > > >
> > > >
> > > > BR,
> > > > Jani.
> > > >
> > > >
> > > > --
> > > > Jani Nikula, Intel Open Source Graphics Center
> >


_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Jani Nikula]

On Fri, 25 Aug 2023, "Helen Mae Koike Fornazier" <helen.koike@collabora.com> wrote:
> I'm not sure I get what do you call out-of-tree builds.

Using 'make O=dir' (see make help) to separate source and and output
directories.

BR,
Jani.


-- 
Jani Nikula, Intel Open Source Graphics Center

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 2/6] drm: ci: Force db410c to host mode

[Rob Clark]

On Fri, Aug 25, 2023 at 8:06 AM Helen Mae Koike Fornazier
<helen.koike@collabora.com> wrote:
>
> On Friday, August 25, 2023 11:41 -03, Rob Clark <robdclark@gmail.com> wrote:
>
> > On Fri, Aug 25, 2023 at 7:34 AM Helen Mae Koike Fornazier
> > <helen.koike@collabora.com> wrote:
> > >
> > > On Friday, August 25, 2023 11:30 -03, Rob Clark <robdclark@gmail.com> wrote:
> > >
> > > > On Fri, Aug 25, 2023 at 6:56 AM Jani Nikula <jani.nikula@linux.intel.com> wrote:
> > > > >
> > > > > On Fri, 25 Aug 2023, Vignesh Raman <vignesh.raman@collabora.com> wrote:
> > > > > > Force db410c to host mode to fix network issue which results in failure
> > > > > > to mount root fs via NFS.
> > > > > > See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > > > >
> > > > > > Since this fix is not sent upstream, add it to build.sh script
> > > > > > before building the kernel and dts. Better approach would be
> > > > > > to use devicetree overlays.
> > > > > >
> > > > > > Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> > > > > > ---
> > > > > >  drivers/gpu/drm/ci/build.sh | 4 ++++
> > > > > >  1 file changed, 4 insertions(+)
> > > > > >
> > > > > > diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
> > > > > > index 7b014287a041..c39834bd6bd7 100644
> > > > > > --- a/drivers/gpu/drm/ci/build.sh
> > > > > > +++ b/drivers/gpu/drm/ci/build.sh
> > > > > > @@ -70,6 +70,10 @@ if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then
> > > > > >      fi
> > > > > >  fi
> > > > > >
> > > > > > +# Force db410c to host mode to fix network issue which results in failure to mount root fs via NFS.
> > > > > > +# See https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/cb72a629b8c15c80a54dda510743cefd1c4b65b8
> > > > > > +sed -i '/&usb {/,/status = "okay";/s/status = "okay";/&\n\tdr_mode = "host";/' arch/arm64/boot/dts/qcom/apq8016-sbc.dts
> > > > > > +
> > > > >
> > > > > It seems like a really bad idea to me to have the CI build modify the
> > > > > source tree before building.
> > > > >
> > > > > The kernel being built will have a dirty git repo, and the localversion
> > > > > will have -dirty in it.
> > > > >
> > > > > I think it would be better to do out-of-tree builds and assume the
> > > > > source is read-only.
> > > >
> > > > We have the ${target_branch}-external-fixes mechanism to merge
> > > > necessary changes before building the kernel for CI.  Which is
> > > > necessary for a couple of reasons:
> > >
> > > Should we create an official topic/drm-ci-external-fixes branch ?
> >
> > Hmm, maybe.. I guess as we expand this to more driver trees, and want
> > to be able to re-run CI in the drm tree after merges to
> > drm-next/drm-fixes, we maybe want to have central
> > drm-next-external-fixes and drm-fixes-external-fixes.  I guess we can
> > keep those based on drm-next and drm-fixes?  And if there would be
> > conflicts because, say, ${driver}-next is behind drm-next, then
> > ${driver}-next could be rebased on drm-next?
> >
>
> tbh this is one of the reasons I would prefer in-code fixes instead of
> commits on a -external-fixes branch, since it seems things start to become
> complex to manage all different trees for people executing ci tests
> on different history points, but I don't oppose going for -external-fixes
> either.

If by in-code you mean in the same branch that we are running CI on, I
think that will be difficult to do without having force-pushes later
to remove the unrelated patch.

It doesn't happen every release, but sometimes there is an issue
requiring a fix outside of drm, which really shouldn't land via
drm-next.  In some cases even, we might need a temporary hack fix when
a proper solution is still being worked out.  Since the kernel
development is unlike mesa, where we have a single main branch and CI
runs on every change entering that main branch, we are going to have
to deal occasionally with breakage that comes in via another tree that
isn't running CI.  But IME so far with msm-next-external-fixes, it
hasn't been so bad.. I haven't even had to rebase it every kernel
cycle.  (If the fix cherry-picked into -external-fixes from a previous
release cycle already exists in msm-next then the merge of that commit
is a no-op.)

BR,
-R

> Regards,
> Helen
>
> > BR,
> > -R
> >
> > > Regards,
> > > Helen
> > >
> > > >
> > > > 1) patches like this which aren't appropriate upstream but necessary
> > > > due to the CI lab setup
> > > > 2) target branch if often based on an early -rc, and it isn't unheard
> > > > of to need some fix for some board or another which isn't appropriate
> > > > to land via drm-next
> > > >
> > > > We should use the -external-fixes branch mechanism for patches like this one.
> > > >
> > > > BR,
> > > > -R
> > > >
> > > > > >  for opt in $ENABLE_KCONFIGS; do
> > > > > >    echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config
> > > > > >  done
> > > > >
> > > > > Ditto for the config changes in the context here. Those are files in
> > > > > git, don't change them.
> > > > >
> > > > > Shouldn't this use something like 'scripts/config --enable' or
> > > > > 'scripts/config --disable' on the .config file to be used for building
> > > > > instead?
> > > > >
> > > > >
> > > > > BR,
> > > > > Jani.
> > > > >
> > > > >
> > > > > --
> > > > > Jani Nikula, Intel Open Source Graphics Center
> > >
>

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 4/6] drm: ci: Enable configs to fix mt8173 boot hang issue

[AngeloGioacchino Del Regno]

Il 25/08/23 14:24, Vignesh Raman ha scritto:
> Enable regulator
> Enable MT6397 RTC driver
> 
> Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> ---
>   drivers/gpu/drm/ci/arm64.config | 2 ++
>   1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/gpu/drm/ci/arm64.config b/drivers/gpu/drm/ci/arm64.config
> index 817e18ddfd4f..ea7a6cceff40 100644
> --- a/drivers/gpu/drm/ci/arm64.config
> +++ b/drivers/gpu/drm/ci/arm64.config
> @@ -184,6 +184,8 @@ CONFIG_HW_RANDOM_MTK=y
>   CONFIG_MTK_DEVAPC=y
>   CONFIG_PWM_MTK_DISP=y
>   CONFIG_MTK_CMDQ=y
> +CONFIG_REGULATOR_DA9211=y
> +CONFIG_RTC_DRV_MT6397=y

I wonder if it'd be a better idea to simply add those to the defconfig instead as

CONFIG_REGULATOR_DA9211=m
CONFIG_RTC_DRV_MT6397=m

Any opinion on this? Matthias? Anyone else?

Cheers,
Angelo

>   
>   # For nouveau.  Note that DRM must be a module so that it's loaded after NFS is up to provide the firmware.
>   CONFIG_ARCH_TEGRA=y



_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 4/6] drm: ci: Enable configs to fix mt8173 boot hang issue

[Vignesh Raman]

Hi Angelo,

On 07/09/23 16:16, AngeloGioacchino Del Regno wrote:
> Il 25/08/23 14:24, Vignesh Raman ha scritto:
>> Enable regulator
>> Enable MT6397 RTC driver
>>
>> Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
>> ---
>>   drivers/gpu/drm/ci/arm64.config | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/ci/arm64.config 
>> b/drivers/gpu/drm/ci/arm64.config
>> index 817e18ddfd4f..ea7a6cceff40 100644
>> --- a/drivers/gpu/drm/ci/arm64.config
>> +++ b/drivers/gpu/drm/ci/arm64.config
>> @@ -184,6 +184,8 @@ CONFIG_HW_RANDOM_MTK=y
>>   CONFIG_MTK_DEVAPC=y
>>   CONFIG_PWM_MTK_DISP=y
>>   CONFIG_MTK_CMDQ=y
>> +CONFIG_REGULATOR_DA9211=y
>> +CONFIG_RTC_DRV_MT6397=y
> 
> I wonder if it'd be a better idea to simply add those to the defconfig 
> instead as
> 
> CONFIG_REGULATOR_DA9211=m
> CONFIG_RTC_DRV_MT6397=m
> 
> Any opinion on this? Matthias? Anyone else?

CONFIG_RTC_DRV_MT6397=m is already present in defconfig.
We can also add CONFIG_REGULATOR_DA9211=m to defconfig.

Regards,
Vignesh

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH 4/6] drm: ci: Enable configs to fix mt8173 boot hang issue

[David Heidelberg]

[-- Attachment #1.1.1.1: Type: text/plain, Size: 1659 bytes --]

On 08/09/2023 13:23, Vignesh Raman wrote:

> Hi Angelo,
>
> On 07/09/23 16:16, AngeloGioacchino Del Regno wrote:
>> Il 25/08/23 14:24, Vignesh Raman ha scritto:
>>> Enable regulator
>>> Enable MT6397 RTC driver
>>>
>>> Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
>>> ---
>>>   drivers/gpu/drm/ci/arm64.config | 2 ++
>>>   1 file changed, 2 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/ci/arm64.config 
>>> b/drivers/gpu/drm/ci/arm64.config
>>> index 817e18ddfd4f..ea7a6cceff40 100644
>>> --- a/drivers/gpu/drm/ci/arm64.config
>>> +++ b/drivers/gpu/drm/ci/arm64.config
>>> @@ -184,6 +184,8 @@ CONFIG_HW_RANDOM_MTK=y
>>>   CONFIG_MTK_DEVAPC=y
>>>   CONFIG_PWM_MTK_DISP=y
>>>   CONFIG_MTK_CMDQ=y
>>> +CONFIG_REGULATOR_DA9211=y
>>> +CONFIG_RTC_DRV_MT6397=y
>>
>> I wonder if it'd be a better idea to simply add those to the 
>> defconfig instead as
>>
>> CONFIG_REGULATOR_DA9211=m
>> CONFIG_RTC_DRV_MT6397=m
>>
>> Any opinion on this? Matthias? Anyone else?
>
> CONFIG_RTC_DRV_MT6397=m is already present in defconfig.
> We can also add CONFIG_REGULATOR_DA9211=m to defconfig.

We need add config snippet override for CONFIG_REGULATOR_DA9211=y 
because our CI don't have any initrd with modules, we just immediately 
continue boot from NFS, where we load rest of the boot non-essential 
modules.

For generic kernel, I assume CONFIG_REGULATOR_DA9211=m should be added.

David

>
> Regards,
> Vignesh

-- 
David Heidelberg
Consultant Software Engineer

Collabora Ltd.
Platinum Building, St John's Innovation Park, Cambridge CB4 0DS, UK
Registered in England & Wales, no. 5513718


[-- Attachment #1.1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 695 bytes --]

[-- Attachment #1.2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

[-- Attachment #2: Type: text/plain, Size: 167 bytes --]

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic