The dst pointer was being advanced by the clip width, not the full line
stride, resulting in corruption. The clip offset was also calculated
incorrectly.
Cc: [email protected]
Signed-off-by: Hector Martin <[email protected]>
---
drivers/gpu/drm/drm_format_helper.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
index e676921422b8..12bc6b45e95b 100644
--- a/drivers/gpu/drm/drm_format_helper.c
+++ b/drivers/gpu/drm/drm_format_helper.c
@@ -366,12 +366,12 @@ void drm_fb_xrgb8888_to_rgb888_dstclip(void __iomem *dst, unsigned int dst_pitch
return;
vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32));
- dst += clip_offset(clip, dst_pitch, sizeof(u16));
+ dst += clip_offset(clip, dst_pitch, 3);
for (y = 0; y < lines; y++) {
drm_fb_xrgb8888_to_rgb888_line(dbuf, vaddr, linepixels);
memcpy_toio(dst, dbuf, dst_len);
vaddr += fb->pitches[0];
- dst += dst_len;
+ dst += dst_pitch;
}
kfree(dbuf);
--
2.33.0
Hi
Am 17.11.21 um 15:22 schrieb Hector Martin:
> The dst pointer was being advanced by the clip width, not the full line
> stride, resulting in corruption. The clip offset was also calculated
> incorrectly.
>
> Cc: [email protected]
> Signed-off-by: Hector Martin <[email protected]>
Thanks for your patch, but you're probably on the wrong branch. We
rewrote this code recently and fixed the issue in drm-misc-next. [1][2]
If anything, this patch could go into stable. If anyone wants to merge
it there, then
Acked-by: Thomas Zimmermann <[email protected]>
Best regards
Thomas
[1]
https://lore.kernel.org/dri-devel/[email protected]/
[2]
https://cgit.freedesktop.org/drm/drm-misc/commit/?id=53bc2098d2b6ccff25fe13f9345cbb5c0ef34a99
> ---
> drivers/gpu/drm/drm_format_helper.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
> index e676921422b8..12bc6b45e95b 100644
> --- a/drivers/gpu/drm/drm_format_helper.c
> +++ b/drivers/gpu/drm/drm_format_helper.c
> @@ -366,12 +366,12 @@ void drm_fb_xrgb8888_to_rgb888_dstclip(void __iomem *dst, unsigned int dst_pitch
> return;
>
> vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32));
> - dst += clip_offset(clip, dst_pitch, sizeof(u16));
> + dst += clip_offset(clip, dst_pitch, 3);
> for (y = 0; y < lines; y++) {
> drm_fb_xrgb8888_to_rgb888_line(dbuf, vaddr, linepixels);
> memcpy_toio(dst, dbuf, dst_len);
> vaddr += fb->pitches[0];
> - dst += dst_len;
> + dst += dst_pitch;
> }
>
> kfree(dbuf);
>
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Ivo Totev
On 17/11/2021 23.56, Thomas Zimmermann wrote:
> Hi
>
> Am 17.11.21 um 15:22 schrieb Hector Martin:
>> The dst pointer was being advanced by the clip width, not the full line
>> stride, resulting in corruption. The clip offset was also calculated
>> incorrectly.
>>
>> Cc: [email protected]
>> Signed-off-by: Hector Martin <[email protected]>
>
> Thanks for your patch, but you're probably on the wrong branch. We
> rewrote this code recently and fixed the issue in drm-misc-next. [1][2]
Oops. I was on linux-next as of Nov 1. Looks like I missed it by a week!
Sounds like I'm going to have to rebase/rewrite the other series I just
sent too...
--
Hector Martin ([email protected])
Public Key: https://mrcn.st/pub