Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: Ville Syrjala <ville.syrjala@linux.intel.com>,
	<intel-gfx@lists.freedesktop.org>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH v3 7/8] drm/i915/vrr: Check that the push send bit is clear after delayed vblank
Date: Tue, 11 Feb 2025 12:38:47 +0530	[thread overview]
Message-ID: <6e7cb456-7866-4063-8ad0-e7e1c2bcd72d@intel.com> (raw)
In-Reply-To: <20250210160711.24010-1-ville.syrjala@linux.intel.com>


On 2/10/2025 9:37 PM, Ville Syrjala wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> Since we don't do mailbox updates the push send bit
> should alwyas clear by the time the delay vblank fires
> and the flip completes. Check for that to make sure we
> haven't screwed up the sequencing/vblank evasion/etc.
>
> On the DSB path we should be able to guarantee this
> since we don't have to deal with any scheduler latencies
> and whatnot. I suppose unexpected DMA/memory latencies
> might be the only thing that might trip us up here.
>
> For the MMIO path we do always have a non-zero chance
> that vblank evasion fails (since we can't really guarantee
> anything about the scheduling behaviour). That could trip
> up this check, but that seems fine since we already print
> errors for other types of vblank evasion failures.
>
> Should the CPU vblank evasion actually fail, then the push
> send bit can still be set when the next commit happens. But
> both the DSB and MMIO paths should handle that situation
> gracefully.
>
> v2: Only check once instead of polling for two scanlines
>      since we should now be guaranteed to be past the
>      delayed vblank.
>      Also check in the MMIO path for good measure
> v3: Skip the push send check when VRR is enabled.
>      With joiner the secondary pipe's DSBs doen't have access
>      to the transcoder registers, and so doing this check
>      there triggers a reponse timeout error on the DSB. VRR
>      is not currently allowed when using joiner, so this will
>      prevent the bogus register access.
>
> Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com> #v1
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>   drivers/gpu/drm/i915/display/intel_color.c   |  1 +
>   drivers/gpu/drm/i915/display/intel_display.c |  4 +++
>   drivers/gpu/drm/i915/display/intel_vrr.c     | 34 ++++++++++++++++++++
>   drivers/gpu/drm/i915/display/intel_vrr.h     |  2 ++
>   4 files changed, 41 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
> index 4d8f6509cac4..cfe14162231d 100644
> --- a/drivers/gpu/drm/i915/display/intel_color.c
> +++ b/drivers/gpu/drm/i915/display/intel_color.c
> @@ -1991,6 +1991,7 @@ void intel_color_prepare_commit(struct intel_atomic_state *state,
>   	if (crtc_state->use_dsb) {
>   		intel_vrr_send_push(crtc_state->dsb_color_vblank, crtc_state);
>   		intel_dsb_wait_vblank_delay(state, crtc_state->dsb_color_vblank);
> +		intel_vrr_check_push_sent(crtc_state->dsb_color_vblank, crtc_state);
>   		intel_dsb_interrupt(crtc_state->dsb_color_vblank);
>   	}
>   
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 0790b2a4583e..34434071a415 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -7737,6 +7737,7 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state,
>   
>   			intel_vrr_send_push(new_crtc_state->dsb_commit, new_crtc_state);
>   			intel_dsb_wait_vblank_delay(state, new_crtc_state->dsb_commit);
> +			intel_vrr_check_push_sent(new_crtc_state->dsb_commit, new_crtc_state);
>   			intel_dsb_interrupt(new_crtc_state->dsb_commit);
>   		}
>   	}
> @@ -7886,6 +7887,9 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
>   			intel_crtc_disable_flip_done(state, crtc);
>   
>   		intel_atomic_dsb_wait_commit(new_crtc_state);
> +
> +		if (!state->base.legacy_cursor_update && !new_crtc_state->use_dsb)
> +			intel_vrr_check_push_sent(NULL, new_crtc_state);
>   	}
>   
>   	/*
> diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
> index adb51609d0a3..cac49319026d 100644
> --- a/drivers/gpu/drm/i915/display/intel_vrr.c
> +++ b/drivers/gpu/drm/i915/display/intel_vrr.c
> @@ -416,6 +416,40 @@ void intel_vrr_send_push(struct intel_dsb *dsb,
>   		intel_dsb_nonpost_end(dsb);
>   }
>   
> +void intel_vrr_check_push_sent(struct intel_dsb *dsb,
> +			       const struct intel_crtc_state *crtc_state)
> +{
> +	struct intel_display *display = to_intel_display(crtc_state);
> +	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> +	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
> +
> +	if (!crtc_state->vrr.enable)

I think you mean:

if (crtc_state->vrr.enable) return;

Regards,

Ankit

> +		return;
> +
> +	/*
> +	 * Make sure the push send bit has cleared. This should
> +	 * already be the case as long as the caller makes sure
> +	 * this is called after the delayed vblank has occurred.
> +	 */
> +	if (dsb) {
> +		int wait_us, count;
> +
> +		wait_us = 2;
> +		count = 1;
> +
> +		/*
> +		 * If the bit hasn't cleared the DSB will
> +		 * raise the poll error interrupt.
> +		 */
> +		intel_dsb_poll(dsb, TRANS_PUSH(display, cpu_transcoder),
> +			       TRANS_PUSH_SEND, 0, wait_us, count);
> +	} else {
> +		if (intel_vrr_is_push_sent(crtc_state))
> +			drm_err(display->drm, "[CRTC:%d:%s] VRR push send still pending\n",
> +				crtc->base.base.id, crtc->base.name);
> +	}
> +}
> +
>   bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state)
>   {
>   	struct intel_display *display = to_intel_display(crtc_state);
> diff --git a/drivers/gpu/drm/i915/display/intel_vrr.h b/drivers/gpu/drm/i915/display/intel_vrr.h
> index 899cbf40f880..514822577e8a 100644
> --- a/drivers/gpu/drm/i915/display/intel_vrr.h
> +++ b/drivers/gpu/drm/i915/display/intel_vrr.h
> @@ -25,6 +25,8 @@ void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state)
>   void intel_vrr_enable(const struct intel_crtc_state *crtc_state);
>   void intel_vrr_send_push(struct intel_dsb *dsb,
>   			 const struct intel_crtc_state *crtc_state);
> +void intel_vrr_check_push_sent(struct intel_dsb *dsb,
> +			       const struct intel_crtc_state *crtc_state);
>   bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state);
>   void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state);
>   void intel_vrr_get_config(struct intel_crtc_state *crtc_state);

  reply	other threads:[~2025-02-11  7:09 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-07 22:31 [PATCH v2 0/8] drm/i915/vrr: Fix DSB+VRR usage for PTL+ Ville Syrjala
2025-02-07 22:31 ` [PATCH v2 1/8] drm/i915/dsb: Move the +1 usec adjustment into dsb_wait_usec() Ville Syrjala
2025-02-11  8:58   ` Nautiyal, Ankit K
2025-02-07 22:31 ` [PATCH v2 2/8] drm/i915/vrr: Don't send push for legacy cursor updates Ville Syrjala
2025-02-11  9:01   ` Nautiyal, Ankit K
2025-02-07 22:31 ` [PATCH v2 3/8] drm/i915/vrr: Account for TRANS_PUSH delay Ville Syrjala
2025-02-11  9:03   ` Nautiyal, Ankit K
2025-02-07 22:31 ` [PATCH v2 4/8] drm/i915/dsb: Compute use_dsb earlier Ville Syrjala
2025-02-11  9:39   ` Nautiyal, Ankit K
2025-02-07 22:31 ` [PATCH v2 5/8] drm/i915/dsb: Introduce intel_dsb_poll() Ville Syrjala
2025-02-07 22:31 ` [PATCH v2 6/8] drm/i915/vrr: Reorder the DSB "wait for safe window" vs. TRANS_PUSH Ville Syrjala
2025-02-07 22:31 ` [PATCH v2 7/8] drm/i915/vrr: Check that the push send bit is clear after delayed vblank Ville Syrjala
2025-02-10 16:07   ` [PATCH v3 " Ville Syrjala
2025-02-11  7:08     ` Nautiyal, Ankit K [this message]
2025-02-11 17:38       ` Ville Syrjälä
2025-02-12 13:09         ` Nautiyal, Ankit K
2025-02-12 13:23           ` Nautiyal, Ankit K
2025-02-12 16:45           ` Ville Syrjälä
2025-02-07 22:31 ` [PATCH v2 8/8] drm/i915/dsb: Decode DSB error interrupts Ville Syrjala
2025-02-11  8:54   ` Nautiyal, Ankit K
2025-02-08  1:19 ` ✓ CI.Patch_applied: success for drm/i915/vrr: Fix DSB+VRR usage for PTL+ (rev2) Patchwork
2025-02-08  1:19 ` ✓ CI.checkpatch: " Patchwork
2025-02-08  1:20 ` ✓ CI.KUnit: " Patchwork
2025-02-08  1:37 ` ✓ CI.Build: " Patchwork
2025-02-08  1:39 ` ✓ CI.Hooks: " Patchwork
2025-02-08  1:41 ` ✗ CI.checksparse: warning " Patchwork
2025-02-08  2:00 ` ✓ Xe.CI.BAT: success " Patchwork
2025-02-08 20:19 ` ✗ Xe.CI.Full: failure " Patchwork
2025-02-10 16:17 ` ✓ CI.Patch_applied: success for drm/i915/vrr: Fix DSB+VRR usage for PTL+ (rev3) Patchwork
2025-02-10 16:18 ` ✗ CI.checkpatch: warning " Patchwork
2025-02-10 16:19 ` ✓ CI.KUnit: success " Patchwork
2025-02-10 16:35 ` ✓ CI.Build: " Patchwork
2025-02-10 16:38 ` ✓ CI.Hooks: " Patchwork
2025-02-10 16:39 ` ✗ CI.checksparse: warning " Patchwork
2025-02-11  6:30 ` ✓ Xe.CI.BAT: success " Patchwork
2025-02-12  9:49 ` ✗ Xe.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=6e7cb456-7866-4063-8ad0-e7e1c2bcd72d@intel.com \
    --to=ankit.k.nautiyal@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=ville.syrjala@linux.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