From: sashiko-bot@kernel.org
To: "Arun R Murthy" <arun.r.murthy@intel.com>,
intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
Subject: Re: [PATCH v4 6/7] drm/i915/display: Program and enable the DP symbol based timing generator
Date: Tue, 03 Nov 2026 10:23:10 +0000 [thread overview]
Message-ID: <sashiko-outbox-195038@kernel.org> (raw)
In-Reply-To: <20261103-vlk-64036-dp-symbol-tg-v4-6-6cf8a588b54c@intel.com>
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/i915/display: invalid CMTG MMIO write in intel_set_transcoder_timings()
--
commit d5e98872efc6064e978436ee34bbcfdf86c29006
Author: Arun R Murthy <arun.r.murthy@intel.com>
drm/i915/display: Program and enable the DP symbol based timing generator
This commit programs and enables the DP link layer symbol based timing
generator, including the 128-bit Data M/N registers, HACT/HTOTAL/CHUNK_SIZE
timing registers, and TRANSCONF configuration and state readout.
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 15117b715e1d5..d450b34b8035a 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
[ ... ]
> @@ -2936,6 +2996,17 @@ void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_state,
> intel_de_write(display, DP_MIN_HBLANK_CTL(transcoder),
> crtc_state->min_hblank);
> }
> +
> + if (crtc_state->dp_symbol_tg.enable) {
> + const struct intel_dp_symbol_tg *tg = &crtc_state->dp_symbol_tg;
> +
> + intel_de_write(display, TRANS_LLSYM_HACT(display, transcoder),
> + TRANS_LLSYM_HACT_VAL(tg->hactive - 1));
> + intel_de_write(display, TRANS_LLSYM_HTOTAL(display, transcoder),
> + TRANS_LLSYM_HTOTAL_VAL(tg->htotal - 1));
> + intel_de_write(display, TRANS_LLSYM_CHUNK_SIZE(display, transcoder),
> + TRANS_LLSYM_CHUNK_SIZE_VAL(tg->chunk_size));
> + }
> }
[Severity: Medium]
Can this write to non-existent CMTG MMIO registers when
intel_set_transcoder_timings() is called for TRANSCODER_CMTG0 or
TRANSCODER_CMTG1?
During modeset and DC state restore on display version 35,
hsw_configure_cpu_transcoder() and intel_cmtg_restore() call
intel_cmtg_set_timings() -> intel_set_transcoder_timings(crtc_state,
cmtg_transcoder) when cpu_transcoder is TRANSCODER_A or TRANSCODER_B on an
eDP output:
drivers/gpu/drm/i915/display/intel_cmtg.c:intel_cmtg_set_timings() {
enum transcoder cmtg_transcoder = to_cmtg_transcoder(crtc_state->cpu_transcoder);
...
if (type == LRR)
intel_set_transcoder_timings_lrr(crtc_state, cmtg_transcoder);
else
intel_set_transcoder_timings(crtc_state, cmtg_transcoder);
}
When an eDP panel operates at a 128b/132b UHBR link rate without MSO
splitter, intel_dp_compute_symbol_tg() sets crtc_state->dp_symbol_tg.enable
to true.
Just above this block in intel_set_transcoder_timings(), the
DP_MIN_HBLANK_CTL write checks that transcoder is not a CMTG transcoder:
if (DISPLAY_VER(display) >= 30 &&
transcoder != TRANSCODER_CMTG0 &&
transcoder != TRANSCODER_CMTG1) {
intel_de_write(display, DP_MIN_HBLANK_CTL(transcoder),
crtc_state->min_hblank);
}
Without a similar check on transcoder here, _MMIO_TRANS2() translates
TRANS_LLSYM_HACT, TRANS_LLSYM_HTOTAL, and TRANS_LLSYM_CHUNK_SIZE using the
CMTG base offset and writes to unclaimed register offsets 0x6f0dc..0x6f0e4
(or 0x6f1dc..0x6f1e4).
Should this block also check transcoder != TRANSCODER_CMTG0 &&
transcoder != TRANSCODER_CMTG1 before writing the TRANS_LLSYM_* registers?
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261103-vlk-64036-dp-symbol-tg-v4-0-6cf8a588b54c@intel.com?part=6
next prev parent reply other threads:[~2026-11-03 10:23 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-03 9:58 [PATCH v4 0/7] drm/i915/dp: Add support for the DP symbol based timing generator Arun R Murthy
2026-11-03 9:58 ` [PATCH v4 1/7] drm/i915/dp: Factor out the DSC slice count and link bpp lookup Arun R Murthy
2026-11-03 9:58 ` [PATCH v4 2/7] drm/i915/display: Add the DP link layer symbol timing generator registers Arun R Murthy
2026-11-03 9:58 ` [PATCH v4 3/7] drm/i915/display: Add the DP symbol timing generator crtc state Arun R Murthy
2026-11-03 9:58 ` [PATCH v4 4/7] drm/i915/display: Add HAS_DP_SYMBOL_TG() Arun R Murthy
2026-11-03 9:58 ` [PATCH v4 5/7] drm/i915/dp: Compute the DP symbol based timing generator parameters Arun R Murthy
2026-11-05 4:32 ` Kandpal, Suraj
2026-11-03 9:58 ` [PATCH v4 6/7] drm/i915/display: Program and enable the DP symbol based timing generator Arun R Murthy
2026-11-03 10:23 ` sashiko-bot [this message]
2026-11-03 9:58 ` [PATCH v4 7/7] drm/i915/dp: Don't limit the Link M/N ratio with the symbol " Arun R Murthy
2026-11-04 1:30 ` Kandpal, Suraj
2026-11-03 11:21 ` ✓ i915.CI.BAT: success for drm/i915/dp: Add support for the DP symbol based timing generator (rev3) Patchwork
2026-11-03 14:30 ` ✗ i915.CI.Full: failure " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2026-11-04 3:24 [PATCH v4 0/7] drm/i915/dp: Add support for the DP symbol based timing generator Arun R Murthy
2026-11-04 3:24 ` [PATCH v4 6/7] drm/i915/display: Program and enable " Arun R Murthy
2026-11-05 4:47 ` Kandpal, Suraj
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=sashiko-outbox-195038@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=arun.r.murthy@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=sashiko-reviews@lists.linux.dev \
/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