Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Javier Martinez Canillas <javierm@redhat.com>
To:     Pierre Asselin <pa@panix.com>
Cc:     Jocelyn Falempe <jfalempe@redhat.com>,
        Pierre Asselin <pa@panix.com>,
        Daniel Vetter <daniel.vetter@ffwll.ch>,
        linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
        Hans de Goede <hdegoede@redhat.com>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Ard Biesheuvel <ardb@kernel.org>
Subject: Re: [PATCH] firmware/sysfb: Fix wrong stride when bits-per-pixel is
 calculated
In-Reply-To: <2e07f818ccdff7023a060e732d7c4ef6.squirrel@mail.panix.com>
References: <20230412150225.3757223-1-javierm@redhat.com>
 <2e07f818ccdff7023a060e732d7c4ef6.squirrel@mail.panix.com>
Date:   Wed, 12 Apr 2023 19:31:31 +0200
Message-ID: <87jzyhror0.fsf@minerva.mail-host-address-is-not-set>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

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

>> 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: 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
>>
>> [patch elided]
>
> NOOOOOO !  The 1024x768x32 screen is all garbled.
> (gfxpayload=keep, gfxpayload=1024x768x32 or vga=0x318).
>

That's suprising... I tested the patch with vga=ask and picked
1024x768x15, 1024x768x16, 1024x768x24 and 1024x768x32. For all
cases the bits-per-pixel and line length values were correct.

But I don't have a machine with legacy BIOS so I testee using QEMU and
SeaBIOS.

> The other modes work as before (but the dmesg has less information;
> I'll investigate.)
>

Interesting. So you don't have the simplefb output that you had before ?

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat