From: Jani Nikula <jani.nikula@intel.com>
To: Sowmiya S <sowmiya.s@intel.com>, igt-dev@lists.freedesktop.org
Cc: swati2.sharma@intel.com, kunal1.joshi@intel.com,
Sowmiya S <sowmiya.s@intel.com>
Subject: Re: [PATCH i-g-t v1 1/2] lib/igt_kms: Add MST bandwidth-fitting support and helpers
Date: Tue, 21 Apr 2026 15:09:03 +0300 [thread overview]
Message-ID: <c34cb162b9ed31f40153a375348f52d7d2ce8b5f@intel.com> (raw)
In-Reply-To: <20260421061114.2262809-2-sowmiya.s@intel.com>
On Tue, 21 Apr 2026, Sowmiya S <sowmiya.s@intel.com> wrote:
> Add helpers to handle shared-link DP bandwidth fitting for
> MST/daisy-chain configurations.
>
> On MST hubs/daisy-chains multiple streams share one DP link
> and a joint atomic commit can be rejected unless modes are
> chosen to fit the link BW. This will centralizes MST detection
> and fitting so tests can request fitting only when required.
>
> Signed-off-by: Sowmiya S <sowmiya.s@intel.com>
> ---
> lib/igt_kms.c | 33 +++++++++++++++++++++++++++++++++
> lib/igt_kms.h | 3 +++
> 2 files changed, 36 insertions(+)
>
> diff --git a/lib/igt_kms.c b/lib/igt_kms.c
> index 2d758b63c..0c7965954 100644
> --- a/lib/igt_kms.c
> +++ b/lib/igt_kms.c
> @@ -2744,6 +2744,7 @@ static void igt_output_reset(igt_output_t *output)
> {
> output->pending_crtc = NULL;
> output->use_override_mode = false;
> + output->bw_fit_applied = false;
> memset(&output->override_mode, 0, sizeof(output->override_mode));
>
> igt_output_set_prop_value(output, IGT_CONNECTOR_CRTC_ID, 0);
> @@ -3227,6 +3228,15 @@ void igt_display_require(igt_display_t *display, int drm_fd)
>
> igt_display_reset_outputs(display);
>
> + /* Detect MST outputs after they are fully initialized */
> + for (i = 0; i < display->n_outputs; i++) {
for_each_output()
> + if (display->outputs[i].config.connector &&
> + igt_check_output_is_dp_mst(&display->outputs[i])) {
> + display->needs_bw_fit = true;
So that really means "has DP MST output"? Why not name it according to
what it actually *is*, not what it's used for? Or add a function to
check for that instead of caching it and dealing with all the troubles
of caching.
> + break;
> + }
> + }
> +
> out:
> LOG_UNINDENT(display);
>
> @@ -5372,6 +5382,7 @@ bool __override_all_active_output_modes_to_fit_bw(igt_display_t *display,
> int ret;
>
> igt_output_override_mode(output, mode);
> + output->bw_fit_applied = true;
>
> if (__override_all_active_output_modes_to_fit_bw(display, outputs, n_outputs, base + 1))
> return true;
> @@ -5441,6 +5452,9 @@ bool igt_fit_modes_in_bw(igt_display_t *display)
> {
> int ret;
>
> + for (int i = 0; i < display->n_outputs; i++)
> + display->outputs[i].bw_fit_applied = false;
for_each_output()
> +
> if (display->is_atomic)
> ret = igt_display_try_commit_atomic(display,
> DRM_MODE_ATOMIC_TEST_ONLY |
> @@ -5461,6 +5475,25 @@ bool igt_fit_modes_in_bw(igt_display_t *display)
> return true;
> }
>
> +void igt_display_fit_bw(igt_display_t *display)
I probably haven't always documented every function I've added, but
things like igt_crtc_for_crtc_index() are pretty obvious.
I have no idea what this actually does.
> +{
> + int active_mst = 0;
> +
> + if (!display->needs_bw_fit)
> + return;
> + for (int i = 0; i < display->n_outputs; i++) {
for_each_output()
> + igt_output_t *output = &display->outputs[i];
> +
> + if (output->pending_crtc &&
> + igt_check_output_is_dp_mst(output))
> + active_mst++;
> + }
> + if (active_mst < 2)
> + return;
> + igt_require_f(igt_fit_modes_in_bw(display),
> + "No valid mode combination fits available MST link BW\n");
> +}
> +
> /**
> * igt_crtc_refresh:
> * @crtc: CRTC to refresh
> diff --git a/lib/igt_kms.h b/lib/igt_kms.h
> index fcbb6a5ad..5c60e7bb4 100644
> --- a/lib/igt_kms.h
> +++ b/lib/igt_kms.h
> @@ -506,6 +506,7 @@ typedef struct {
>
> /* bitmask of changed properties */
> uint64_t changed;
> + bool bw_fit_applied;
>
> uint32_t props[IGT_NUM_CONNECTOR_PROPS];
> uint64_t values[IGT_NUM_CONNECTOR_PROPS];
> @@ -536,6 +537,7 @@ struct _igt_display {
> uint64_t *modifiers;
> uint32_t *formats;
> int format_mod_count;
> + bool needs_bw_fit;
> };
>
> typedef struct {
> @@ -1286,6 +1288,7 @@ drmModePropertyBlobRes *igt_get_writeback_formats_blob(igt_output_t *output);
> uint64_t igt_get_writeback_fb_id(igt_output_t *output);
> void igt_detach_crtc(igt_display_t *display, igt_output_t *output);
> void igt_get_and_wait_out_fence(igt_output_t *output);
> +void igt_display_fit_bw(igt_display_t *display);
>
> igt_colorop_t *igt_find_colorop(igt_display_t *display, uint32_t id);
--
Jani Nikula, Intel
next prev parent reply other threads:[~2026-04-21 12:09 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-21 6:11 [PATCH i-g-t v1 0/2] Fix 2-display test for MST shared-link BW Sowmiya S
2026-04-21 6:11 ` [PATCH i-g-t v1 1/2] lib/igt_kms: Add MST bandwidth-fitting support and helpers Sowmiya S
2026-04-21 12:09 ` Jani Nikula [this message]
2026-04-27 9:03 ` S, Sowmiya
2026-04-21 6:11 ` [PATCH i-g-t v1 2/2] tests/kms_plane_multiple: Fix 2-display test for MST shared-link BW Sowmiya S
2026-04-21 21:03 ` ✓ i915.CI.BAT: success for " Patchwork
2026-04-21 21:55 ` ✓ Xe.CI.BAT: " Patchwork
2026-04-22 2:02 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-04-22 4:34 ` ✗ 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=c34cb162b9ed31f40153a375348f52d7d2ce8b5f@intel.com \
--to=jani.nikula@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=kunal1.joshi@intel.com \
--cc=sowmiya.s@intel.com \
--cc=swati2.sharma@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