dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Javier Martinez Canillas <javierm@redhat.com>
To: Pierre Asselin <pa@panix.com>
Cc: Pierre Asselin <pa@panix.com>,
	mairacanal@riseup.net, dri-devel@lists.freedesktop.org,
	tzimmermann@suse.de, jose.exposito89@gmail.com
Subject: Re: [PATCH v3 01/13] firmware/sysfb: Fix EFI/VESA format selection
Date: Wed, 12 Apr 2023 13:22:37 +0200	[thread overview]
Message-ID: <87sfd5s5tu.fsf@minerva.mail-host-address-is-not-set> (raw)
In-Reply-To: <dfb4f25ca8dfb0d81d778d6315f104ad.squirrel@mail.panix.com>

"Pierre Asselin" <pa@panix.com> writes:

>> Can you please share you grub config file? It seems that is set to
>> GRUB_GFXMODE=1024x768x32 but the actual mode is set to 1024x768x24 ?
>
> Okay, but you'll be sorry...  The gfxmode is set to "keep" in all the
> entries.  https://www.panix.com/~pa/linux-6.3-simplefb/grub.cfg .
>
> The "TEST" entry was used to bisect.  The "PRE-TEST" was to set things
> up, to receive the bzImages compiled on a faster machine. Now I boot
> the "Linux 6.3.0-rc5-x86".
>
>
>> That is, it fails when the picked format is DRM_FORMAT_RGB8888 but
>> works when is DRM_FORMAT_XRGB888. I can't spot any error in Thomas'
>> patch so I wonder if the problem is with what grub is passing to the
>> kernel.
>>
>> The mentioned vga=0x318 workaround that you mentioned makes the mode
>> passed to match the selected DRM_FORMAT_RGB888 which I guess is why
>> that worked for you.
>
> All right, I did a series of reboots, editing the grub command line
> to change the "gfxpayload=" grub option or the "vga=" kernel option.
> In each case I captured the output of
>   "dmesg | egrep -i 'fbcon|console|fb0|frameb|simple|vga|vesa'
>
> https://www.panix.com/~pa/linux-6.3-simplefb/selected-modes
>
> (D'oh.  My script printed "vga=vga=" twice when that option was set.
> Good enough as is.)
>
> Note the difference in linelength= between the bad and good r8g8b8.
> Does it mean anything ?
>  (bad)> format=r8g8b8, mode=1024x768x24, linelength=4096
> (good)> format=r8g8b8, mode=1024x768x24, linelength=3072
>

Ah! That's a good data point and I believe that found a possible issue in
the sysfb format selection logic. Can you please try the following patch?

From 55b5375c528b4128350dfa2126277049f8821349 Mon Sep 17 00:00:00 2001
From: Javier Martinez Canillas <javierm@redhat.com>
Date: Wed, 12 Apr 2023 13:20:48 +0200
Subject: [PATCH] firmware/sysfb: Fix wrong stride when bits-per-pixel is
 calculated

The commit f35cd3fa7729 ("firmware/sysfb: Fix EFI/VESA format selection")
fixed format selection by calculating the bits-per-pixel instead of just
using the reported color depth.

But unfortunately this broke some modes because the stride is always set
to the reported line length (in bytes), which could not match the actual
stride if the calculated bits-per-pixel doesn't match the reported depth.

Fixes: commit f35cd3fa7729 ("firmware/sysfb: Fix EFI/VESA format selection")
Reported-by: Pierre Asselin <pa@panix.com>
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/firmware/sysfb_simplefb.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/firmware/sysfb_simplefb.c b/drivers/firmware/sysfb_simplefb.c
index 82c64cb9f531..5dc23e57089f 100644
--- a/drivers/firmware/sysfb_simplefb.c
+++ b/drivers/firmware/sysfb_simplefb.c
@@ -28,7 +28,7 @@ __init bool sysfb_parse_mode(const struct screen_info *si,
 			     struct simplefb_platform_data *mode)
 {
 	__u8 type;
-	u32 bits_per_pixel;
+	u32 bits_per_pixel, stride;
 	unsigned int i;
 
 	type = si->orig_video_isVGA;
@@ -54,14 +54,19 @@ __init bool sysfb_parse_mode(const struct screen_info *si,
 	 * bits_per_pixel here and ignore lfb_depth. In the loop below,
 	 * ignore simplefb formats with alpha bits, as EFI and VESA
 	 * don't specify alpha channels.
+	 *
+	 * If a calculated bits_per_pixel is used instead of lfb_depth,
+	 * then also ignore lfb_linelength and calculate the stride.
 	 */
 	if (si->lfb_depth > 8) {
 		bits_per_pixel = max(max3(si->red_size + si->red_pos,
 					  si->green_size + si->green_pos,
 					  si->blue_size + si->blue_pos),
 				     si->rsvd_size + si->rsvd_pos);
+		stride = DIV_ROUND_UP(si->lfb_width * bits_per_pixel, 8);
 	} else {
 		bits_per_pixel = si->lfb_depth;
+		stride = si->lfb_linelength;
 	}
 
 	for (i = 0; i < ARRAY_SIZE(formats); ++i) {
@@ -80,7 +85,7 @@ __init bool sysfb_parse_mode(const struct screen_info *si,
 			mode->format = f->name;
 			mode->width = si->lfb_width;
 			mode->height = si->lfb_height;
-			mode->stride = si->lfb_linelength;
+			mode->stride = stride;
 			return true;
 		}
 	}

base-commit: fd35174e13f98f9232c4aa66689816731d34ca28
-- 

Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat


  reply	other threads:[~2023-04-12 11:22 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-02 11:29 [PATCH v3 00/13] drm: Fix color-format selection in fbdev emulation Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 01/13] firmware/sysfb: Fix EFI/VESA format selection Thomas Zimmermann
2023-04-06 15:45   ` Pierre Asselin
2023-04-08 11:26     ` Linux regression tracking #adding (Thorsten Leemhuis)
2023-04-16 12:06       ` Linux regression tracking #update (Thorsten Leemhuis)
2023-04-08 16:10     ` Pierre Asselin
2023-04-11 15:56       ` Javier Martinez Canillas
2023-04-11 19:39         ` Pierre Asselin
2023-04-12 11:22           ` Javier Martinez Canillas [this message]
2023-04-17  7:34             ` Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 02/13] drm/format-helper: Comment on RGB888 byte order Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 03/13] drm/format-helper: Fix test-input format conversion Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 04/13] drm/format-helper: Store RGB565 in little-endian order Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 05/13] drm/format-helper: Type fixes in format-helper tests Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 06/13] drm/format-helper: Flip src/dst-format branches in blit helper Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 07/13] drm/format-helper: Add conversion from XRGB8888 to ARGB8888 Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 08/13] drm/format-helper: Add conversion from XRGB8888 to ARGB2101010 Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 09/13] drm/format-helper: Add conversion from XRGB8888 to 15-bit RGB555 formats Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 10/13] drm/fh-helper: Split fbdev single-probe helper Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 11/13] drm/fb-helper: Fix single-probe color-format selection Thomas Zimmermann
2023-01-03 21:18   ` Maíra Canal
2023-01-04  8:14     ` Thomas Zimmermann
2023-01-04 10:34       ` Maíra Canal
2023-05-12 13:20   ` Linus Walleij
2023-05-12 14:11     ` Thomas Zimmermann
2023-05-15  8:01       ` Linus Walleij
2023-05-15  8:17         ` Thomas Zimmermann
2023-05-15  8:59           ` Linus Walleij
2023-05-15  9:26             ` Thomas Zimmermann
2023-05-15  9:30               ` Linus Walleij
2023-05-15  8:16       ` Daniel Vetter
2023-05-15  8:42         ` Thomas Zimmermann
2023-05-14 12:10     ` Linux regression tracking #adding (Thorsten Leemhuis)
2023-05-26 12:22       ` Linux regression tracking #update (Thorsten Leemhuis)
2023-01-02 11:29 ` [PATCH v3 12/13] drm/format-helper: Simplify drm_fb_build_fourcc_list() Thomas Zimmermann
2023-01-02 11:29 ` [PATCH v3 13/13] drm/format-helper: Remove unnecessary conversion helpers Thomas Zimmermann

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=87sfd5s5tu.fsf@minerva.mail-host-address-is-not-set \
    --to=javierm@redhat.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=jose.exposito89@gmail.com \
    --cc=mairacanal@riseup.net \
    --cc=pa@panix.com \
    --cc=tzimmermann@suse.de \
    /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