From: "Govindapillai, Vinod" <vinod.govindapillai@intel.com>
To: "igt-dev@lists.freedesktop.org" <igt-dev@lists.freedesktop.org>,
"Hogander, Jouni" <jouni.hogander@intel.com>
Cc: "Saarinen, Jani" <jani.saarinen@intel.com>,
"B, Jeevan" <jeevan.b@intel.com>
Subject: Re: [PATCH i-g-t v1] lib/i915/fbc: Disable FBC in case of selective update in xe2lpd
Date: Tue, 25 Mar 2025 08:09:42 +0000 [thread overview]
Message-ID: <de52bc11da8be17771026c5be6e5708d8400dd46.camel@intel.com> (raw)
In-Reply-To: <851cac0a1f7a049560b8e2b8b411ad07fe90cfb3.camel@intel.com>
On Tue, 2025-03-25 at 06:23 +0000, Hogander, Jouni wrote:
> On Thu, 2025-03-20 at 13:09 +0200, Vinod Govindapillai wrote:
> > In xe2lpd onwards, FBC and PSR2 selective update should be
> > selectively enabled based on the dirty rect threshold. Display
> > driver lacks such a logic at the moment and instead disables
> > FBC in case selective update is on. So replicate the same
> > in related IGT cases as well.
>
> As a name intel_fbc_psr_combo_supported is not specifically referring
> PSR2. It is generally PSR. Also checked usage of
> intel_fbc_psr_combo_supported and it seems to be used in PSR1-FBC
> testcases as well. This will skip those tests as well. I don't think we
> want that.
Ack! Need to fix it! As that was replaced some previous inline gfx version check, wonder it had been
skipping the FBC + PSR1 cases all these times!
BR
Vinod
>
> BR,
>
> Jouni Högander
>
> >
> > Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
> > ---
> > lib/i915/intel_fbc.c | 15 ++++++++++-----
> > 1 file changed, 10 insertions(+), 5 deletions(-)
> >
> > diff --git a/lib/i915/intel_fbc.c b/lib/i915/intel_fbc.c
> > index 3971e4920..b0b4278ad 100644
> > --- a/lib/i915/intel_fbc.c
> > +++ b/lib/i915/intel_fbc.c
> > @@ -167,11 +167,16 @@ bool intel_fbc_plane_size_supported(int fd,
> > uint32_t width, uint32_t height)
> > */
> > bool intel_fbc_psr_combo_supported(int device)
> > {
> > - int ver = intel_display_ver(intel_get_drm_devid(device));
> > -
> > - /* In Xe3 FBC PSR combo not supported because of FBC dirty
> > rect */
> > - if (ver >= 20 && ver < 30)
> > - return true;
> > + /*
> > + * Because of some random underruns, FBC is disabled when
> > selective
> > + * update is enabled for xe2lpd onwards. The display driver
> > should
> > + * implement a selection logic between FBC and PSR2 / PR
> > selective
> > + * update from xe2lpd onwards. Until that is implemented,
> > keep
> > + * FBC disabled if PSR2 selective update is on.
> > + *
> > + * Though right now this function is empty, keep this for
> > the future
> > + * display version based variations for FBC and PSR
> > + */
> >
> > return false;
> > }
>
prev parent reply other threads:[~2025-03-25 8:09 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-20 11:09 [PATCH i-g-t v1] lib/i915/fbc: Disable FBC in case of selective update in xe2lpd Vinod Govindapillai
2025-03-20 11:31 ` ✓ Xe.CI.BAT: success for " Patchwork
2025-03-20 11:53 ` ✓ i915.CI.BAT: " Patchwork
2025-03-20 12:14 ` ✗ Xe.CI.Full: failure " Patchwork
2025-03-20 13:36 ` ✗ i915.CI.Full: " Patchwork
2025-03-25 6:23 ` [PATCH i-g-t v1] " Hogander, Jouni
2025-03-25 8:09 ` Govindapillai, Vinod [this message]
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=de52bc11da8be17771026c5be6e5708d8400dd46.camel@intel.com \
--to=vinod.govindapillai@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=jani.saarinen@intel.com \
--cc=jeevan.b@intel.com \
--cc=jouni.hogander@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