Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Naladala, Ramanaidu" <Ramanaidu.naladala@intel.com>
To: Lee Shawn C <shawn.c.lee@intel.com>, <igt-dev@lists.freedesktop.org>
Subject: Re: [v5] tests/kms_vrr: add support for full range refresh rate testing
Date: Wed, 13 May 2026 19:59:06 +0530	[thread overview]
Message-ID: <bc911537-0987-4054-a4ae-74ce36679766@intel.com> (raw)
In-Reply-To: <20260513061657.980046-1-shawn.c.lee@intel.com>

[-- Attachment #1: Type: text/plain, Size: 8529 bytes --]

Hi Lee Shawn,

With the recent kernel changes, the midpoint approach is no longer 
necessary. It would be better to first revert commit |862eb1762| 
(/tests/kms_vrr: Start virtual RR test above midpoint to avoid mode 
fallback/), and then layer your changes on top of that baseline.

On 5/13/2026 11:46 AM, Lee Shawn C wrote:
> Currently, kms_vrr tests fixed step increments within the VRR range.
> To better validate hardware stability and potential issues across the
> entire spectrum, this patch adds a 'full-range' testing mode.
>
> When the full-range flag is enabled, the test will:
> 1. Iterate from the maximum to the minimum refresh rate (stepping by 1).
> 2. For standard testing, maintain the existing behavior of incrementing
>     from minimum to maximum using the defined step size.
>
> v2: Fix build failure by adding documentation for the new subtest.
> v3: Remove unnecessary '!!' operator when assigning to bool.
> v4: Add full-range (top-down/bottom-up) subtests to verify all supported
>      RR within VRR range.
> v5: Fix build failure for the new subtest.
>
> Cc: Naladala Ramanaidu<ramanaidu.naladala@intel.com>
> Signed-off-by: Lee Shawn C<shawn.c.lee@intel.com>
> ---
>   tests/kms_vrr.c | 118 ++++++++++++++++++++++++++++++++++++++++--------
>   1 file changed, 100 insertions(+), 18 deletions(-)
>
> diff --git a/tests/kms_vrr.c b/tests/kms_vrr.c
> index 6043d40f1d74..bf2fab9ce23f 100644
> --- a/tests/kms_vrr.c
> +++ b/tests/kms_vrr.c
> @@ -65,6 +65,11 @@
>    * SUBTEST: seamless-rr-switch-virtual
>    * Description: Test to create a Virtual Mode in VRR range and switch to it
>    * 		without a full modeset.
> + * SUBTEST: seamless-rr-switch-virtual-top-down
> + * Description: Vrr seamless refresh rate switch for virtual modes using full range
> + *
> + * SUBTEST: seamless-rr-switch-virtual-bottom-up
> + * Description: Vrr seamless refresh rate switch for virtual modes using full range
>    *
>    * SUBTEST: lobf
>    * Description: Test to validate link-off between active frames in non-psr
> @@ -94,11 +99,13 @@ enum {
>   	TEST_SEAMLESS_VRR = 1 << 4,
>   	TEST_SEAMLESS_DRRS = 1 << 5,
>   	TEST_SEAMLESS_VIRTUAL_RR = 1 << 6,
> -	TEST_FASTSET = 1 << 7,
> -	TEST_MAXMIN = 1 << 8,
> -	TEST_LINK_OFF = 1 << 10,
> -	TEST_NEGATIVE = 1 << 11,
> -	TEST_FORCE_RR = 1 << 12,
> +	TEST_SEAMLESS_VIRTUAL_RR_TOP_DOWN = 1 << 7,
> +	TEST_SEAMLESS_VIRTUAL_RR_BOTTOM_UP = 1 << 8,
> +	TEST_FASTSET = 1 << 9,
> +	TEST_MAXMIN = 1 << 10,
> +	TEST_LINK_OFF = 1 << 11,
> +	TEST_NEGATIVE = 1 << 12,
> +	TEST_FORCE_RR = 1 << 13,
>   };
>   
>   enum {
> @@ -235,13 +242,58 @@ low_rr_mode_with_same_res(igt_output_t *output, unsigned int vrr_min)
>   	return mode;
>   }
>   
> -static void
> -virtual_rr_vrr_range_mode(drmModeModeInfo *mode, float virtual_refresh_rate)
> +static bool
> +virtual_rr_vrr_range_mode(data_t *data, drmModeModeInfo *mode, float virtual_refresh_rate)
>   {
>   	uint64_t clock_hz = mode->clock * 1000;
> +	uint16_t virtual_vtotal;
> +
> +	virtual_vtotal = clock_hz / (mode->htotal * virtual_refresh_rate);
> +	if (is_intel_device(data->drm_fd) && virtual_vtotal > 8192) {
> +		igt_info("VTotal (%d) already exceed 8192 at %fHz\n", virtual_vtotal, virtual_refresh_rate);
> +		return false;
> +	}
>   
> -	mode->vtotal = clock_hz / (mode->htotal * virtual_refresh_rate);
> +	mode->vsync_start = virtual_vtotal - (mode->vtotal - mode->vsync_start);
> +	mode->vsync_end   = virtual_vtotal - (mode->vtotal - mode->vsync_end);
> +	mode->vtotal = virtual_vtotal;
>   	mode->vrefresh = virtual_refresh_rate;
> +
> +	return true;
> +}
> +
> +static drmModeModeInfo*
> +get_selected_fixed_mode(igt_output_t *output, bool lowest)
> +{
> +	drmModeConnectorPtr connector = output->config.connector;
> +	drmModeModeInfo *selected_mode = NULL;
> +
> +	for (int i = 0; i < connector->count_modes; i++) {
> +		drmModeModeInfo *m = &connector->modes[i];
> +
> +		if (m->type & (DRM_MODE_TYPE_PREFERRED | DRM_MODE_TYPE_DRIVER)) {
> +			if (!selected_mode) {
> +				selected_mode = m;
> +				continue;
> +			}
> +
> +			if (lowest) {
> +				if (m->vrefresh < selected_mode->vrefresh)
> +					selected_mode = m;
> +			} else {
> +				if (m->vrefresh > selected_mode->vrefresh)
> +					selected_mode = m;
> +			}
> +		}
> +	}
> +
> +	if (selected_mode) {
> +		igt_info("Found Target Fixed Mode: \"%s\" %dHz (vtotal: %d)\n",
> +			 selected_mode->name, selected_mode->vrefresh, selected_mode->vtotal);
> +		return selected_mode;
> +	}
> +
> +	return NULL;
>   }
>   
>   /* Read min and max vrr range from the connector debugfs. */
> @@ -757,19 +809,25 @@ test_seamless_virtual_rr_basic(data_t *data, igt_crtc_t *crtc,
>   	unsigned int vrefresh;
>   	uint64_t rate[] = {0};
>   	uint32_t step_size;
> -	drmModeModeInfo virtual_mode;
> +	drmModeModeInfo *virtual_mode = NULL;
> +	int start, end, step;
> +	bool need_low_rr = flags & TEST_SEAMLESS_VIRTUAL_RR_BOTTOM_UP;
>   
> -	igt_info("Use HIGH_RR Mode as default\n");
> -	kmstest_dump_mode(&data->switch_modes[HIGH_RR_MODE]);
> +	virtual_mode = get_selected_fixed_mode(output, need_low_rr);
> +	if (!virtual_mode)
> +		return;
> +
> +	igt_info("Use %s_RR Mode as default\n", need_low_rr ? "LOW" : "HIGH");
> +	kmstest_dump_mode(virtual_mode);
>   
>   	prepare_test(data, output, crtc);
> -	rate[0] = igt_kms_frame_time_from_vrefresh(data->switch_modes[HIGH_RR_MODE].vrefresh);
> +	rate[0] = igt_kms_frame_time_from_vrefresh(virtual_mode->vrefresh);
>   
>   	/*
>   	 * Sink with DRR and VRR can be in downclock mode so
>   	 * switch to highest refresh rate mode.
>   	 */
> -	igt_output_override_mode(output, &data->switch_modes[HIGH_RR_MODE]);
> +	igt_output_override_mode(output, virtual_mode);
>   	igt_assert(igt_display_try_commit_atomic(&data->display, DRM_MODE_PAGE_FLIP_EVENT, NULL) == 0);
>   
>   	result = flip_and_measure(data, output, rate, 1, TEST_DURATION_NS);
> @@ -784,7 +842,7 @@ test_seamless_virtual_rr_basic(data_t *data, igt_crtc_t *crtc,
>   	step_size = (data->range.max - data->range.min) / 5;
>   
>   	/* Switch to Virtual RR */
> -	virtual_mode = *igt_output_get_mode(output);
> +	virtual_mode = igt_output_get_mode(output);
>   
>   	/*
>   	 * Start virtual RR testing from above the midpoint of the VRR range when multiple
> @@ -795,13 +853,29 @@ test_seamless_virtual_rr_basic(data_t *data, igt_crtc_t *crtc,
>   		   (((data->range.max + data->range.min) / 2) + step_size) :
>   		   data->range.min + step_size;
>   
> -	for ( ; vrefresh < data->range.max; vrefresh += step_size) {
> -		virtual_rr_vrr_range_mode(&virtual_mode, vrefresh);
> +	if (flags & TEST_SEAMLESS_VIRTUAL_RR_TOP_DOWN) {
> +		start = data->range.max;
> +		end = data->range.min;
> +		step = -1;
> +	} else if (flags & TEST_SEAMLESS_VIRTUAL_RR_BOTTOM_UP) {
> +		start = data->range.min;
> +		end = data->range.max;
> +		step = 1;
> +	} else {
> +		start = data->range.min;
> +		end = data->range.max;
> +		step = step_size;
> +	}
> +
> +	for (vrefresh = start; (step > 0) ? (vrefresh <= end) : (vrefresh >= end); vrefresh += step) {
> +		if (!virtual_rr_vrr_range_mode(data, virtual_mode, vrefresh))
> +			continue;
>   
>   		igt_info("Requesting Virtual Mode with Refresh Rate (%u Hz): \n", vrefresh);
> -		kmstest_dump_mode(&virtual_mode);
> +		kmstest_dump_mode(virtual_mode);
>   
> -		igt_output_override_mode(output, &virtual_mode);
> +		igt_output_override_mode(output, virtual_mode);
> +		igt_assert(igt_display_try_commit_atomic(&data->display, DRM_MODE_ATOMIC_TEST_ONLY, NULL) == 0);
>   		igt_assert(igt_display_try_commit_atomic(&data->display, 0, NULL) == 0);
>   
>   		rate[0] = igt_kms_frame_time_from_vrefresh(vrefresh);
> @@ -1105,6 +1179,14 @@ int igt_main_args("drs:", long_opts, help_str, opt_handler, &data)
>   		igt_subtest_with_dynamic("seamless-rr-switch-virtual")
>   			run_vrr_test(&data, test_seamless_virtual_rr_basic, TEST_SEAMLESS_VIRTUAL_RR);
>   
> +		igt_describe("Test to switch to all virtual modes in VRR range without modeset.");
> +		igt_subtest_with_dynamic("seamless-rr-switch-virtual-top-down")
> +			run_vrr_test(&data, test_seamless_virtual_rr_basic, TEST_SEAMLESS_VIRTUAL_RR_TOP_DOWN);
> +
> +		igt_describe("Test to switch to all virtual modes in VRR range without modeset.");
> +		igt_subtest_with_dynamic("seamless-rr-switch-virtual-bottom-up")
> +			run_vrr_test(&data, test_seamless_virtual_rr_basic, TEST_SEAMLESS_VIRTUAL_RR_BOTTOM_UP);
> +
>   		igt_describe("Test to validate the link-off between active frames in "
>   			     "non-PSR operation.");
>   		igt_subtest_with_dynamic("lobf") {

[-- Attachment #2: Type: text/html, Size: 9288 bytes --]

  reply	other threads:[~2026-05-13 14:29 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-27 12:09 [PATCH i-g-t] tests/kms_vrr: add support for full range refresh rate testing Lee Shawn C
2026-04-28  6:12 ` [v2] " Lee Shawn C
2026-04-28  7:10   ` Jani Nikula
2026-04-28  7:30     ` Lee, Shawn C
2026-04-28  7:29 ` ✓ i915.CI.BAT: success for tests/kms_vrr: add support for full range refresh rate testing (rev2) Patchwork
2026-04-28  7:34 ` [v3] tests/kms_vrr: add support for full range refresh rate testing Lee Shawn C
2026-04-28  7:43 ` ✓ Xe.CI.BAT: success for tests/kms_vrr: add support for full range refresh rate testing (rev2) Patchwork
2026-04-28  8:49 ` ✓ i915.CI.BAT: success for tests/kms_vrr: add support for full range refresh rate testing (rev3) Patchwork
2026-04-28  9:18 ` ✓ Xe.CI.BAT: " Patchwork
2026-04-28 13:53 ` ✗ i915.CI.Full: failure for tests/kms_vrr: add support for full range refresh rate testing (rev2) Patchwork
2026-04-28 14:45 ` ✗ Xe.CI.FULL: " Patchwork
2026-04-28 15:00 ` ✗ i915.CI.Full: failure for tests/kms_vrr: add support for full range refresh rate testing (rev3) Patchwork
2026-04-28 16:43 ` ✓ Xe.CI.FULL: success " Patchwork
2026-05-07 13:53 ` [v4] tests/kms_vrr: add support for full range refresh rate testing Lee Shawn C
2026-05-13  6:16 ` [v5] " Lee Shawn C
2026-05-13 14:29   ` Naladala, Ramanaidu [this message]
2026-05-13 11:04 ` ✗ LGCI.VerificationFailed: failure for tests/kms_vrr: add support for full range refresh rate testing (rev5) Patchwork
2026-05-13 13:22 ` ✓ Xe.CI.BAT: success " Patchwork
2026-05-13 13:33 ` ✓ i915.CI.BAT: " Patchwork
2026-05-14 10:33 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-05-14 12:50 ` ✗ i915.CI.Full: " 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=bc911537-0987-4054-a4ae-74ce36679766@intel.com \
    --to=ramanaidu.naladala@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=shawn.c.lee@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox