Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Thomas Zimmermann" <tzimmermann@suse.de>
Cc: dri-devel@lists.freedesktop.org, linux-hyperv@vger.kernel.org,
	intel-gfx@lists.freedesktop.org, imx@lists.linux.dev,
	virtualization@lists.linux.dev, Frank.Li@kernel.org,
	wsa+renesas@sang-engineering.com,
	"Michael S. Tsirkin" <mst@redhat.com>,
	"Eugenio Perez" <eperezma@redhat.com>
Subject: Re: [PATCH 4/4] drm/pixel: Link draw helpers into drm_pixel_helper.ko
Date: Mon, 19 Oct 2026 16:32:11 +0000	[thread overview]
Message-ID: <sashiko-outbox-177908@kernel.org> (raw)
In-Reply-To: <20261019154612.777074-5-tzimmermann@suse.de>

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:
- [High] drm: missing DRM_PIXEL_HELPER dependency in DRM_KMS_HELPER
- [Medium] drm/pixel: missing DRM_PIXEL_HELPER select in DRM_ST7920, DRM_XE, and DRM_KMS_HELPER
- [Low] drm/pixel: stale drm_draw.c reference in MAINTAINERS after rename
- [Low] drm: drm_panic_helper.o omitted when DRM_KUNIT_TEST=y and DRM_PANIC=n
--

commit 8f1149ac31f00f33aec39cba3d1477d5b9456a23
Author: Thomas Zimmermann <tzimmermann@suse.de>

drm/pixel: Link draw helpers into drm_pixel_helper.ko

This commit moves drm_draw.c (renamed to drm_draw_helper.c) from the DRM
core into drm_pixel_helper.ko, removes the DRM_DRAW Kconfig symbol, and
updates DRM_CLIENT_LIB to select DRM_PIXEL_HELPER when DRM_CLIENT_LOG is
enabled.

> diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
> index bf7c599443403..1b16e6101a53a 100644
> --- a/drivers/gpu/drm/Kconfig
> +++ b/drivers/gpu/drm/Kconfig
> @@ -52,15 +52,10 @@ config DRM_KMS_HELPER
>  config DRM_PIXEL_HELPER
>  	tristate
>  	depends on DRM
> -	select DRM_DRAW if DRM_PANIC
>  	select FONT_SUPPORT if DRM_PANIC
>  	help
>  	  Helpers for working with pixel data in DRM.
>  
> -config DRM_DRAW
> -	bool
> -	depends on DRM
> -
>  config DRM_PANIC
>  	bool "Display a user-friendly message when a kernel panic occurs"
>  	depends on DRM && PRINTK

[Severity: High]
With drm_format_helper.o and drm_panic_helper.o moved from drm_kms_helper.ko
into drm_pixel_helper.ko in this series, does DRM_KMS_HELPER in
drivers/gpu/drm/Kconfig also need to select DRM_PIXEL_HELPER?

In drivers/gpu/drm/Makefile, drm_kms_helper.o unconditionally links in
drm_gem_atomic_helper.o and drm_modeset_helper.o.

In drivers/gpu/drm/drm_gem_atomic_helper.c, the shadow plane helpers call
format conversion functions that are now in drm_pixel_helper.ko:

drivers/gpu/drm/drm_gem_atomic_helper.c:__drm_gem_shadow_plane_state_init() {
	__drm_atomic_helper_plane_state_init(&shadow_plane_state->base, plane);
	drm_format_conv_state_init(&shadow_plane_state->fmtcnv_state);
}

Along with drm_format_conv_state_copy() in
__drm_gem_duplicate_shadow_plane_state() and
drm_format_conv_state_release() in __drm_gem_destroy_shadow_plane_state().

Similarly, in drivers/gpu/drm/drm_modeset_helper.c, primary_plane_funcs uses
DRM_PANIC_PLANE_FUNCS:

static const struct drm_plane_funcs primary_plane_funcs = {
	DRM_PLANE_NON_ATOMIC_FUNCS,
	DRM_PANIC_PLANE_FUNCS,
};

When CONFIG_DRM_PANIC=y, DRM_PANIC_PLANE_FUNCS references
drm_plane_helper_display_panic_screen(), which is now defined in
drm_panic_helper.c inside drm_pixel_helper.ko.

Without DRM_KMS_HELPER selecting DRM_PIXEL_HELPER, won't configurations with
CONFIG_DRM_KMS_HELPER enabled and CONFIG_DRM_PIXEL_HELPER=n (or
CONFIG_DRM_KMS_HELPER=y and CONFIG_DRM_PIXEL_HELPER=m) fail to link with
undefined references to drm_format_conv_state_init(),
drm_format_conv_state_copy(), drm_format_conv_state_release(), and
drm_plane_helper_display_panic_screen()?

[Severity: Medium]
Are there also a couple of drivers missing a select of DRM_PIXEL_HELPER
after moving the format and panic helpers into drm_pixel_helper.ko?

1. In drivers/gpu/drm/sitronix/Kconfig, DRM_ST7571 and DRM_ST7586 were
updated to select DRM_PIXEL_HELPER, but DRM_ST7920 was not:

config DRM_ST7920
	tristate "DRM support for Sitronix ST7920 LCD displays"
	depends on DRM && SPI && MMU
	select DRM_GEM_SHMEM_HELPER
	select DRM_KMS_HELPER
	select REGMAP_SPI

However, drivers/gpu/drm/sitronix/st7920.c calls drm_fb_xrgb8888_to_mono()
in st7920_fb_blit_rect() and drm_format_conv_state_reserve() in
st7920_primary_plane_atomic_check(), both of which are in
drm_format_helper.c.

2. In drivers/gpu/drm/xe/Kconfig, neither DRM_XE nor DRM_XE_DISPLAY selects
DRM_PIXEL_HELPER if DRM_PANIC (unlike DRM_I915), even though
drivers/gpu/drm/xe/Makefile compiles i915-display/skl_universal_plane.o
into xe.ko when CONFIG_DRM_XE_DISPLAY=y, which uses DRM_PANIC_PLANE_FUNCS
and references drm_plane_helper_display_panic_screen().

Could these cause link failures when building DRM_ST7920 or DRM_XE without
DRM_PIXEL_HELPER enabled?

> diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
> index 18a5a5baf0293..ab9eb4c212648 100644
> --- a/drivers/gpu/drm/Makefile
> +++ b/drivers/gpu/drm/Makefile
> @@ -91,7 +91,6 @@ drm-$(CONFIG_DRM_PRIVACY_SCREEN) += \
>  	drm_privacy_screen_x86.o
>  drm-$(CONFIG_DRM_ACCEL) += ../../accel/drm_accel.o
>  drm-$(CONFIG_DRM_PANIC) += drm_panic.o
> -drm-$(CONFIG_DRM_DRAW) += drm_draw.o
>  drm-$(CONFIG_DRM_RAS) += drm_ras.o drm_ras_nl.o drm_ras_genl_family.o
>  obj-$(CONFIG_DRM)	+= drm.o
>  
> @@ -100,6 +99,7 @@ obj-$(CONFIG_DRM_PANEL_ORIENTATION_QUIRKS) += drm_panel_orientation_quirks.o
>  obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) += drm_panel_backlight_quirks.o
>  
>  drm_pixel_helper-y := \
> +	drm_draw_helper.o \
>  	drm_format_helper.o \
>  	drm_pixel_helper_common.o
>  drm_pixel_helper-$(CONFIG_DRM_PANIC) += drm_panic_helper.o

[Severity: Low]
What happens when CONFIG_DRM_KUNIT_TEST is enabled while CONFIG_DRM_PANIC=n
(such as with the default drivers/gpu/drm/tests/.kunitconfig)?

In drivers/gpu/drm/tests/Makefile, drm_panic_helper_test.o is built
unconditionally under CONFIG_DRM_KUNIT_TEST, and
drivers/gpu/drm/tests/drm_panic_helper_test.c calls
drm_panic_helper_draw_screen_user(), drm_panic_helper_draw_screen_kmsg(),
drm_panic_helper_set_description(), and
drm_panic_helper_clear_description():

static int drm_panic_helper_draw_screen_user_default(struct drm_scanout_buffer *sb)
{
	return drm_panic_helper_draw_screen_user(sb, 0x00ffffff, 0x00000000);
}

Previously, DRM_KUNIT_TEST selected DRM_PANIC_HELPER, which built
drm_panic_helper.o and selected FONT_SUPPORT even when CONFIG_DRM_PANIC=n.

Now that DRM_KUNIT_TEST in drivers/gpu/drm/Kconfig.debug selects
DRM_PIXEL_HELPER instead of DRM_PANIC_HELPER, and drm_panic_helper.o is only
compiled into drm_pixel_helper when CONFIG_DRM_PANIC is enabled (and
DRM_PIXEL_HELPER only selects FONT_SUPPORT if DRM_PANIC), won't building
DRM_KUNIT_TEST with CONFIG_DRM_PANIC=n fail at link time with undefined
references to the drm_panic_helper_* functions?

[ ... ]

> diff --git a/drivers/gpu/drm/drm_draw.c b/drivers/gpu/drm/drm_draw_helper.c
> similarity index 100%
> rename from drivers/gpu/drm/drm_draw.c
> rename to drivers/gpu/drm/drm_draw_helper.c

[Severity: Low]
Should theDRM PANIC entry in MAINTAINERS also be updated to reflect the
rename from drivers/gpu/drm/drm_draw.c to drivers/gpu/drm/drm_draw_helper.c?

DRM PANIC
M:	Jocelyn Falempe <jfalempe@redhat.com>
M:	Javier Martinez Canillas <javierm@redhat.com>
L:	dri-devel@lists.freedesktop.org
S:	Supported
T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
F:	drivers/gpu/drm/drm_draw.c

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261019154612.777074-1-tzimmermann@suse.de?part=4

  reply	other threads:[~2026-10-19 16:32 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-19 15:32 [PATCH 0/4] drm: Move drawing to pixel-helper module Thomas Zimmermann
2026-10-19 15:32 ` [PATCH 1/4] drm: Add helper module for working with pixel data Thomas Zimmermann
2026-10-19 15:32 ` [PATCH 2/4] drm/pixel: Link format helpers into drm_pixel_helper.ko Thomas Zimmermann
2026-10-19 16:11   ` sashiko-bot
2026-10-19 15:32 ` [PATCH 3/4] drm/pixel: Link panic " Thomas Zimmermann
2026-10-19 16:19   ` sashiko-bot
2026-10-19 16:32     ` Thomas Zimmermann
2026-10-19 15:32 ` [PATCH 4/4] drm/pixel: Link draw " Thomas Zimmermann
2026-10-19 16:32   ` sashiko-bot [this message]
2026-10-19 17:48 ` ✗ Fi.CI.BUILD: failure for drm: Move drawing to pixel-helper module 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-177908@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=Frank.Li@kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=eperezma@redhat.com \
    --cc=imx@lists.linux.dev \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=linux-hyperv@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=tzimmermann@suse.de \
    --cc=virtualization@lists.linux.dev \
    --cc=wsa+renesas@sang-engineering.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