From: "Sharma, Swati2" <swati2.sharma@intel.com>
To: Kunal Joshi <kunal1.joshi@intel.com>, <igt-dev@lists.freedesktop.org>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Subject: Re: [PATCH i-g-t 5/7] tests/intel/kms_dp_linktrain_fallback: add dsc-fallback test
Date: Fri, 14 Feb 2025 12:52:11 +0530 [thread overview]
Message-ID: <e9cf7cac-adf2-4769-b38c-cd6dc7e0547c@intel.com> (raw)
In-Reply-To: <20250211180542.1430676-6-kunal1.joshi@intel.com>
Hi Kunal,
On 11-02-2025 11:35 pm, Kunal Joshi wrote:
> Add new dsc-fallback test to check whether DSC
> automatically kicks in while falling back to a
> lower rate/lane.
>
> v2: fix description (Swati)
> fix indentation (Swati)
> remove extra line (Swati)
> make commit description clear (Ankit)
> split refactor and dsc-fallback test (Ankit)
> switch to igt_debug (Ankit)
>
> v3: use macro instead of magic number (Ankit)
> fix test order (Ankit)
> new line before return (Ankit)
>
> Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com>
> Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
> ---
> tests/intel/kms_dp_linktrain_fallback.c | 162 +++++++++++++++++++++++-
> tests/meson.build | 4 +-
> 2 files changed, 162 insertions(+), 4 deletions(-)
>
> diff --git a/tests/intel/kms_dp_linktrain_fallback.c b/tests/intel/kms_dp_linktrain_fallback.c
> index c6e9fba53..d26e25bba 100644
> --- a/tests/intel/kms_dp_linktrain_fallback.c
> +++ b/tests/intel/kms_dp_linktrain_fallback.c
> @@ -17,10 +17,14 @@
> #include "igt_sysfs.h"
> #include "igt.h"
> #include "kms_mst_helper.h"
> +#include "kms_dsc_helper.h"
>
> /**
> * SUBTEST: dp-fallback
> * Description: Test fallback on DP connectors
> + *
> + * SUBTEST: dsc-fallback
> + * Description: Test fallback to DSC when BW isn't sufficient
Please add functionality here as dsc.
> */
>
> #define RETRAIN_COUNT 1
> @@ -46,7 +50,7 @@ typedef struct {
>
> typedef int (*condition_check_fn)(int drm_fd, igt_output_t *output);
>
> -IGT_TEST_DESCRIPTION("Test link training fallback");
> +IGT_TEST_DESCRIPTION("Test link-training / dsc fallback");
>
> static bool setup_mst_outputs(data_t *data, igt_output_t *mst_output[],
> int *output_count)
> @@ -424,7 +428,7 @@ static void test_fallback(data_t *data, bool is_mst)
> }
> }
>
> -static bool run_test(data_t *data)
> +static bool run_lt_fallback_test(data_t *data)
> {
> bool ran = false;
> igt_output_t *output;
> @@ -462,6 +466,153 @@ static bool run_test(data_t *data)
> return ran;
> }
>
> +static void test_dsc_sst_fallback(data_t *data)
> +{
> + bool non_dsc_mode_found = false;
> + bool dsc_fallback_successful = false;
> + int ret;
> + struct udev_monitor *mon;
> + drmModeModeInfo *mode_to_check;
> + igt_output_t *outputs[IGT_MAX_PIPES];
> + int output_count = 0;
> +
> + igt_info("Checking DSC fallback on %s\n", igt_output_name(data->output));
> + data->pipe = PIPE_A;
> +
> + igt_display_reset(&data->display);
> + igt_reset_link_params(data->drm_fd, data->output);
> + igt_force_link_retrain(data->drm_fd, data->output, RETRAIN_COUNT);
> +
> + /* Find a mode that doesn't require DSC initially */
> + for_each_connector_mode(data->output) {
> + data->mode = &data->output->config.connector->modes[j__];
> + igt_create_color_fb(data->drm_fd, data->mode->hdisplay,
> + data->mode->vdisplay, DRM_FORMAT_XRGB8888,
> + DRM_FORMAT_MOD_LINEAR, 0.0, 1.0, 0.0,
> + &data->fb);
> + igt_output_override_mode(data->output, data->mode);
> + igt_output_set_pipe(data->output, data->pipe);
> + data->primary = igt_output_get_plane_type(data->output,
> + DRM_PLANE_TYPE_PRIMARY);
> + igt_plane_set_fb(data->primary, &data->fb);
> +
> + ret = igt_display_try_commit_atomic(&data->display,
> + DRM_MODE_ATOMIC_TEST_ONLY |
> + DRM_MODE_ATOMIC_ALLOW_MODESET,
> + NULL);
> + if (ret != 0) {
> + igt_debug("Skipping mode %dx%d@%d on %s\n",
> + data->mode->hdisplay, data->mode->vdisplay,
> + data->mode->vrefresh,
> + igt_output_name(data->output));
> + continue;
> + }
> + igt_display_commit2(&data->display, COMMIT_ATOMIC);
> +
> + if (!igt_is_dsc_enabled(data->drm_fd,
> + data->output->name)) {
> + drmModeModeInfo *non_dsc_mode
> + = igt_output_get_mode(data->output);
> + igt_info("Found mode %dx%d@%d %s that doesn't need DSC with link rate %d and lane count %d\n",
> + non_dsc_mode->hdisplay, non_dsc_mode->vdisplay,
> + non_dsc_mode->vrefresh, non_dsc_mode->name,
> + igt_get_current_link_rate(data->drm_fd, data->output),
> + igt_get_current_lane_count(data->drm_fd, data->output));
> + non_dsc_mode_found = true;
> + break;
> + }
> + }
> + igt_require_f(non_dsc_mode_found,
> + "No non-DSC mode found on %s\n",
> + igt_output_name(data->output));
> +
> + /* Repeatedly force link failure until DSC is required (or link is disabled) */
> + while (!igt_get_dp_link_retrain_disabled(data->drm_fd, data->output)) {
> + mon = igt_watch_uevents();
> +
> + igt_assert_f(force_failure_and_wait(data, data->output,
> + LT_FAILURE_REDUCED_CAPS,
> + RETRAIN_COUNT, 1.0, 20.0),
> + "Forcing DSC fallback timed out\n");
> +
> + if (igt_get_dp_link_retrain_disabled(data->drm_fd,
> + data->output)) {
> + igt_reset_connectors();
> + igt_flush_uevents(mon);
> + return;
> + }
> +
> + igt_assert_f(wait_for_hotplug_and_check_bad(data->drm_fd,
> + data,
> + data->output,
> + mon,
> + 20.0),
> + "Didn't get hotplug or link-status=BAD for DSC\n");
> + igt_flush_uevents(mon);
> +
> + outputs[output_count++] = data->output;
> + set_connector_link_status_good(data, outputs, &output_count);
> + igt_display_commit2(&data->display, COMMIT_ATOMIC);
> +
> + mode_to_check = igt_output_get_mode(data->output);
> +
> + if (igt_is_dsc_enabled(data->drm_fd, data->output->name)) {
> + igt_info("mode %dx%d@%d now requires DSC with link rate %d and lane count %d\n",
> + mode_to_check->hdisplay, mode_to_check->vdisplay,
> + mode_to_check->vrefresh,
> + igt_get_current_link_rate(data->drm_fd, data->output),
> + igt_get_current_lane_count(data->drm_fd, data->output));
> + igt_info("DSC fallback successful on %s\n",
> + igt_output_name(data->output));
> + dsc_fallback_successful = true;
> + break;
> + } else {
> + igt_info("mode %dx%d@%d still doesn't require DSC\n",
> + mode_to_check->hdisplay, mode_to_check->vdisplay,
> + mode_to_check->vrefresh);
> + }
> + }
> + igt_assert_f(dsc_fallback_successful, "DSC fallback unsuccessful\n");
> +}
> +
> +static bool run_dsc_sst_fallaback_test(data_t *data)
> +{
> + bool ran = false;
> + igt_output_t *output;
> +
> + if (!is_dsc_supported_by_source(data->drm_fd)) {
> + igt_info("DSC not supported by source.\n");
> + return ran;
> + }
> +
> + for_each_connected_output(&data->display, output) {
> + data->output = output;
> +
> + if (!igt_has_force_link_training_failure_debugfs(data->drm_fd,
> + data->output)) {
> + igt_info("Output %s doesn't support forcing link training.\n",
> + igt_output_name(data->output));
> + continue;
> + }
> +
> + if (output->config.connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) {
> + igt_info("Skipping output %s as it's not DP\n", output->name);
> + continue;
> + }
> +
> + if (!is_dsc_supported_by_sink(data->drm_fd, data->output)) {
> + igt_info("Skipping output %s as DSC not supported by sink\n",
> + igt_output_name(data->output));
> + continue;
> + }
> +
> + ran = true;
> + test_dsc_sst_fallback(data);
> + }
> +
> + return ran;
> +}
> +
> igt_main
> {
> data_t data = {};
> @@ -486,10 +637,15 @@ igt_main
> }
>
> igt_subtest("dp-fallback") {
> - igt_require_f(run_test(&data),
> + igt_require_f(run_lt_fallback_test(&data),
> "Skipping test as no output found or none supports fallback\n");
> }
>
> + igt_subtest("dsc-fallback") {
> + igt_require_f(run_dsc_sst_fallaback_test(&data),
> + "Skipping test: DSC fallback conditions not met.\n");
Make this skip statement as consistent to above. Remove : and full stop
Something like
"Skipping test as DSC fallback..."
> + }
> +
> igt_fixture {
> igt_remove_fb(data.drm_fd, &data.fb);
> igt_display_fini(&data.display);
> diff --git a/tests/meson.build b/tests/meson.build
> index a6f6ad560..95892762e 100644
> --- a/tests/meson.build
> +++ b/tests/meson.build
> @@ -366,7 +366,9 @@ extra_sources = {
> 'kms_chamelium_edid': [ join_paths ('chamelium', 'kms_chamelium_helper.c') ],
> 'kms_chamelium_frames': [ join_paths ('chamelium', 'kms_chamelium_helper.c') ],
> 'kms_chamelium_hpd': [ join_paths ('chamelium', 'kms_chamelium_helper.c') ],
> - 'kms_dp_linktrain_fallback': [ join_paths ('intel', 'kms_mst_helper.c') ],
> + 'kms_dp_linktrain_fallback': [
> + join_paths ('intel', 'kms_mst_helper.c'),
> + join_paths ('intel', 'kms_dsc_helper.c') ],
> 'kms_dsc': [ join_paths ('intel', 'kms_dsc_helper.c') ],
> 'kms_psr2_sf': [ join_paths ('intel', 'kms_dsc_helper.c') ],
> }
next prev parent reply other threads:[~2025-02-14 7:23 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-11 18:05 [PATCH i-g-t 0/7] add dsc-fallback test Kunal Joshi
2025-02-11 18:05 ` [PATCH i-g-t 1/7] tests/intel/kms_dp_linktrain_fallback: refactor find_mst_outputs Kunal Joshi
2025-02-13 5:00 ` Nautiyal, Ankit K
2025-02-14 7:08 ` Sharma, Swati2
2025-02-11 18:05 ` [PATCH i-g-t 2/7] tests/intel/kms_mst_helper: add helper for MST-related functions Kunal Joshi
2025-02-13 5:44 ` Nautiyal, Ankit K
2025-02-14 7:13 ` Sharma, Swati2
2025-02-11 18:05 ` [PATCH i-g-t 3/7] tests/intel/kms_dp_linktrain_fallback: refactor dp-fallback subtest Kunal Joshi
2025-02-11 18:05 ` [PATCH i-g-t 4/7] tests/intel/kms_dp_linktrain_fallback: fix typo Kunal Joshi
2025-02-13 5:48 ` Nautiyal, Ankit K
2025-02-11 18:05 ` [PATCH i-g-t 5/7] tests/intel/kms_dp_linktrain_fallback: add dsc-fallback test Kunal Joshi
2025-02-14 7:22 ` Sharma, Swati2 [this message]
2025-02-14 7:28 ` Sharma, Swati2
2025-02-11 18:05 ` [PATCH i-g-t 6/7] tests/intel/kms_dp_linktrain_fallback: disable ignore long hpd Kunal Joshi
2025-02-11 18:05 ` [PATCH i-g-t 7/7] HAX: Do not merge Kunal Joshi
2025-02-11 20:09 ` ✗ Xe.CI.BAT: failure for add dsc-fallback test (rev4) Patchwork
2025-02-11 20:37 ` ✗ i915.CI.BAT: " Patchwork
2025-02-12 3:53 ` ✗ Xe.CI.Full: " Patchwork
2025-02-13 5:53 ` [PATCH i-g-t 0/7] add dsc-fallback test Nautiyal, Ankit K
2025-02-16 11:05 ` Joshi, Kunal1
2025-02-14 6:48 ` ✓ i915.CI.BAT: success for add dsc-fallback test (rev4) Patchwork
2025-02-14 9:39 ` ✗ i915.CI.Full: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e9cf7cac-adf2-4769-b38c-cd6dc7e0547c@intel.com \
--to=swati2.sharma@intel.com \
--cc=ankit.k.nautiyal@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=kunal1.joshi@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.