From: Jani Nikula <jani.nikula@intel.com>
To: "Lisovskiy, Stanislav" <stanislav.lisovskiy@intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH] drm/i915: Fix mistake in intel_modeset_all_pipes condition
Date: Wed, 09 Aug 2023 12:01:25 +0300 [thread overview]
Message-ID: <87o7jgsi6y.fsf@intel.com> (raw)
In-Reply-To: <ZNNS02EMPmgLZDBc@intel.com>
On Wed, 09 Aug 2023, "Lisovskiy, Stanislav" <stanislav.lisovskiy@intel.com> wrote:
> On Wed, Aug 09, 2023 at 11:38:08AM +0300, Jani Nikula wrote:
>> On Wed, 09 Aug 2023, Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> wrote:
>> > It is supposed to be "!intel_crtc_needs_modeset" - otherwise,
>> > we are active exactly vice versa for active pipes: skipping if modeset
>> > is needed and not skipping if not needed.
>>
>> If the crtc *already* needs a full modeset, there's no need to force a
>> modeset on it.
>>
>> BR,
>> Jani.
>
> We have curently some issue with that. There are multiple places from where
> intel_modeset_all_pipes is called. One is that when we do detect that mbus join
> state had changed. All the previous checks indicated that fastset is enough,
> however once we detect mbus join state had changed in skl_watermarks.c we call
> this function there as well and I think it might act in a wrong way then.
>
> So basically this condition checks whether we need to force a modeset, but not
> if we need it, so no crtc's are supposed to escape this?
> Could be then that we just calling that whole function there wrongly.
Simplified, it's an optimization of:
if (crtc_state->uapi.mode_changed)
continue;
crtc_state->uapi.mode_changed = true;
With your change, it becomes:
if (!crtc_state->uapi.mode_changed)
continue;
crtc_state->uapi.mode_changed = true;
and intel_modeset_all_pipes() roughly becomes a no-op.
BR,
Jani.
>
> Stan
>
>>
>> >
>> > Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
>> > ---
>> > drivers/gpu/drm/i915/display/intel_display.c | 2 +-
>> > 1 file changed, 1 insertion(+), 1 deletion(-)
>> >
>> > diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
>> > index 763ab569d8f32..4b1d7034078ca 100644
>> > --- a/drivers/gpu/drm/i915/display/intel_display.c
>> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
>> > @@ -5439,7 +5439,7 @@ int intel_modeset_all_pipes(struct intel_atomic_state *state,
>> > return PTR_ERR(crtc_state);
>> >
>> > if (!crtc_state->hw.active ||
>> > - intel_crtc_needs_modeset(crtc_state))
>> > + !intel_crtc_needs_modeset(crtc_state))
>> > continue;
>> >
>> > drm_dbg_kms(&dev_priv->drm, "[CRTC:%d:%s] Full modeset due to %s\n",
>>
>> --
>> Jani Nikula, Intel Open Source Graphics Center
--
Jani Nikula, Intel Open Source Graphics Center
next prev parent reply other threads:[~2023-08-09 9:01 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-09 8:29 [Intel-gfx] [PATCH] drm/i915: Fix mistake in intel_modeset_all_pipes condition Stanislav Lisovskiy
2023-08-09 8:38 ` Jani Nikula
2023-08-09 8:48 ` Lisovskiy, Stanislav
2023-08-09 9:01 ` Jani Nikula [this message]
2023-08-09 9:07 ` Lisovskiy, Stanislav
2023-08-09 9:31 ` Jani Nikula
2023-08-09 9:50 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for " Patchwork
2023-08-09 9:50 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2023-08-09 10:02 ` [Intel-gfx] ✓ Fi.CI.BAT: success " 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=87o7jgsi6y.fsf@intel.com \
--to=jani.nikula@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=stanislav.lisovskiy@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;
as well as URLs for NNTP newsgroup(s).