Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Dibin Moolakadan Subrahmanian"
	<dibin.moolakadan.subrahmanian@intel.com>,
	intel-xe@lists.freedesktop.org
Subject: Re: [PATCH] [CI] drm/i915/dmc: Test Xe3p_LPD DMC v2.44
Date: Thu, 05 Nov 2026 09:21:53 +0000	[thread overview]
Message-ID: <sashiko-outbox-198758@kernel.org> (raw)
In-Reply-To: <20261105091559.1784255-1-dibin.moolakadan.subrahmanian@intel.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] drm/i915/dmc: hardcoded CI-only firmware path in XE3P_LPD_DMC_PATH
--

commit 6720c799eff7439b665b842fb4acf1ad6594772f
Author: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>

drm/i915/dmc: Test Xe3p_LPD DMC v2.44

This commit replaces the unversioned DMC_PATH(xe3p_lpd) macro in
XE3P_LPD_DMC_PATH with a hardcoded versioned firmware path for CI testing.

> diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c
> index 56e831b2a8262..864b419098ac0 100644
> --- a/drivers/gpu/drm/i915/display/intel_dmc.c
> +++ b/drivers/gpu/drm/i915/display/intel_dmc.c
> @@ -126,7 +126,7 @@ static bool dmc_firmware_param_disabled(struct intel_display *display)
>  #define DISPLAY_VER13_DMC_MAX_FW_SIZE	0x20000
>  #define DISPLAY_VER12_DMC_MAX_FW_SIZE	ICL_DMC_MAX_FW_SIZE
>  
> -#define XE3P_LPD_DMC_PATH		DMC_PATH(xe3p_lpd)
> +#define XE3P_LPD_DMC_PATH		"i915/xe3p_lpd_dmc_ver2_44.bin"

[Severity: High]
As noted in the commit message, this is a temporary CI test patch and is not
intended to be merged.

All platforms from MTL onward use unversioned DMC firmware paths via
DMC_PATH(), as documented above DMC_LEGACY_PATH in
drivers/gpu/drm/i915/display/intel_dmc.c:

/*
 * New DMC additions should not use this. This is used solely to remain
 * compatible with systems that have not yet updated DMC blobs to use
 * unversioned file names.
 */

During driver probe on Xe3p_LPD (DISPLAY_VERx100(display) == 3500),
intel_dmc_init() acquires the POWER_DOMAIN_INIT runtime PM reference via
intel_dmc_runtime_pm_get(), sets dmc->fw_path to XE3P_LPD_DMC_PATH via
dmc_firmware_default(), and schedules dmc_load_work_fn().

Because dmc_fallback_path() only provides a fallback for Alder Lake-P and
returns NULL on Xe3p_LPD, replacing DMC_PATH(xe3p_lpd) with
"i915/xe3p_lpd_dmc_ver2_44.bin" causes request_firmware() in
dmc_load_work_fn() to fail with -ENOENT on systems shipping the standard
unversioned "i915/xe3p_lpd_dmc.bin" linux-firmware blob:

drivers/gpu/drm/i915/display/intel_dmc.c:dmc_load_work_fn() {
	...
	if (err) {
		drm_notice(display->drm,
			   "Failed to load DMC firmware %s (%pe). Disabling runtime power management.\n",
			   dmc->fw_path, ERR_PTR(err));
		drm_notice(display->drm, "DMC firmware homepage: %s",
			   INTEL_DMC_FIRMWARE_URL);
		return;
	}
	...
}

When this happens, dmc_load_work_fn() returns early without calling
intel_dmc_runtime_pm_put(), leaving the DMC unprogrammed and permanently
disabling display runtime power management. It also changes
MODULE_FIRMWARE(XE3P_LPD_DMC_PATH) so initramfs generators look for the
versioned filename instead of "i915/xe3p_lpd_dmc.bin".

Could CI pass the versioned firmware path via the dmc_firmware_path module
parameter (read by dmc_firmware_param()) instead of modifying
XE3P_LPD_DMC_PATH?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261105091559.1784255-1-dibin.moolakadan.subrahmanian@intel.com?part=1

  reply	other threads:[~2026-11-05  9:22 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-05  9:15 [PATCH] [CI] drm/i915/dmc: Test Xe3p_LPD DMC v2.44 Dibin Moolakadan Subrahmanian
2026-11-05  9:21 ` sashiko-bot [this message]
2026-11-05  9:39 ` ✓ CI.KUnit: success for " Patchwork
2026-11-05 10:24 ` ✓ Xe.CI.BAT: " Patchwork
2026-11-06  5:03 ` ✓ Xe.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=sashiko-outbox-198758@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=dibin.moolakadan.subrahmanian@intel.com \
    --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