linux-kernel - Re: [PATCH] drm/mediatek: Fix XRGB format breakage for blend

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <fe26e581f11b992aaa358141070a36360a97e192.camel@mediatek.com>
Date: Mon, 7 Oct 2024 06:09:31 +0000
From: Jason-JH Lin (林睿祥) <Jason-JH.Lin@...iatek.com>
To: CK Hu (胡俊光) <ck.hu@...iatek.com>,
	"me@...mthiede.com" <me@...mthiede.com>, "yassine.oudjana@...il.com"
	<yassine.oudjana@...il.com>, "chunkuang.hu@...nel.org"
	<chunkuang.hu@...nel.org>, AngeloGioacchino Del Regno
	<angelogioacchino.delregno@...labora.com>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-mediatek@...ts.infradead.org" <linux-mediatek@...ts.infradead.org>,
	Singo Chang (張興國) <Singo.Chang@...iatek.com>,
	Shawn Sung (宋孝謙) <Shawn.Sung@...iatek.com>,
	Nancy Lin (林欣螢) <Nancy.Lin@...iatek.com>,
	"dri-devel@...ts.freedesktop.org" <dri-devel@...ts.freedesktop.org>,
	Project_Global_Chrome_Upstream_Group
	<Project_Global_Chrome_Upstream_Group@...iatek.com>,
	"linux-arm-kernel@...ts.infradead.org" <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH] drm/mediatek: Fix XRGB format breakage for blend_modes
 unsupported SoCs

Hi CK,

On Mon, 2024-10-07 at 05:57 +0000, CK Hu (胡俊光) wrote:
> Hi, Jason:
> 
> On Sat, 2024-10-05 at 17:52 +0800, Jason-JH.Lin wrote:
> > If the constant alpha always set, the SoCs that is not supported
> > the
> > ignore alpha bit will still use constant alpha. That will break the
> > original constant alpha setting of XRGB foramt for blend_modes
> > unsupported SoCs, such as MT8173.
> > 
> > Make the constatnt alpha only set when setting has_alpha or
> > supported
> > blend_modes SoC to fix the downgrade issue.
> 
> I would like this patch to be cherry-picked to kernel 6.11 easily, so
> let this patch base on 6.12-rc1.

OK, I'll move this patch in front of this series:
[1] Fix degradation problem of alpha blending series
- 
https://patchwork.kernel.org/project/linux-mediatek/list/?series=893634
and then send v9 to let you cherry-pick it to kernel-6.11 easily.

> 
> > 
> > Fixes: bc46eb5d5d77 ("drm/mediatek: Support DRM plane alpha in
> > OVL")
> > Signed-off-by: Jason-JH.Lin <jason-jh.lin@...iatek.com>
> > ---
> >  drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 28 +++++++++++++++----
> > ------
> >  1 file changed, 17 insertions(+), 11 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > index 703cd3f6c6d3..de3addd89442 100644
> > --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > @@ -494,19 +494,25 @@ void mtk_ovl_layer_config(struct device *dev,
> > unsigned int idx,
> >  
> >  	con = mtk_ovl_fmt_convert(ovl, state);
> >  	if (state->base.fb) {
> > -		con |= OVL_CON_AEN;
> > -		con |= state->base.alpha & OVL_CON_ALPHA;
> > +		/*
> > +		 * For blend_modes supported SoCs, always set constant
> > alpha.
> > +		 * For blend_modes unsupported SoCs, set constant alpha
> > when has_alpha is set.
> > +		 */
> > +		if (ovl->data->blend_modes || state->base.fb->format-
> > >has_alpha) {
> > +			con |= OVL_CON_AEN;
> > +			con |= state->base.alpha & OVL_CON_ALPHA;
> 
> MT8173 does not support ignore pixel alpha, so OVL_CON_AEN would be
> disabled for XRGB.
> And this is not related to plane alpha, so plane alpha setting should
> be moved out of here.

OK, that make sense.
But I would like to ask Yassine and Adam to test this change again.

> 
> > +		}
> > +
> > +		/*
> > +		 * Although the alpha channel can be ignored, CONST_BLD
> > must be enabled
> > +		 * for XRGB format, otherwise OVL will still read the
> > value from memory.
> > +		 * For RGB888 related formats, whether CONST_BLD is
> > enabled or not won't
> > +		 * affect the result. Therefore we use !has_alpha as
> > the condition.
> > +		 */
> > +		if (blend_mode == DRM_MODE_BLEND_PIXEL_NONE || !state-
> > >base.fb->format->has_alpha)
> > +			ignore_pixel_alpha = OVL_CONST_BLEND;
> 
> This modification is not related to the bug fix, so separate to
> another patch.

OK, I'll separate it.

Regards,
Jason-JH.Lin

> 
> Regards,
> CK