| 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
| ||
|
Message-ID: <9890110452673e7ed4122bab7bb26f99df081f7f.camel@mediatek.com>
Date: Tue, 19 Dec 2023 05:59:39 +0000
From: Shawn Sung (宋孝謙) <Shawn.Sung@...iatek.com>
To: CK Hu (胡俊光) <ck.hu@...iatek.com>,
"chunkuang.hu@...nel.org" <chunkuang.hu@...nel.org>,
"angelogioacchino.delregno@...labora.com"
<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>,
"daniel@...ll.ch" <daniel@...ll.ch>, "p.zabel@...gutronix.de"
<p.zabel@...gutronix.de>, "seanpaul@...omium.org" <seanpaul@...omium.org>,
"dri-devel@...ts.freedesktop.org" <dri-devel@...ts.freedesktop.org>,
"airlied@...il.com" <airlied@...il.com>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>, "matthias.bgg@...il.com"
<matthias.bgg@...il.com>
Subject: Re: [PATCH v4 1/1] drm/mediatek: Fix errors when reporting rotation
capability
Hi CK,
On Tue, 2023-12-19 at 05:24 +0000, CK Hu (胡俊光) wrote:
> Hi, Hsiao-chien:
>
> On Tue, 2023-12-19 at 03:45 +0000, Shawn Sung (宋孝謙) wrote:
> > Hi CK,
> >
> > On Tue, 2023-12-19 at 02:35 +0000, CK Hu (胡俊光) wrote:
> > > Hi, Hsiao-chien:
> > >
> > > On Fri, 2023-11-24 at 18:00 +0800, Hsiao Chien Sung wrote:
> > > > Create rotation property according to the hardware capability.
> > > > Since currently OVL of all chips support same rotation,
> > > > no need to define it in the driver data.
> > > >
> > > > Fixes: 84d805753983 ("drm/mediatek: Support reflect-y plane
> > > > rotation")
> > > >
> > > > Reviewed-by: AngeloGioacchino Del Regno <
> > > > angelogioacchino.delregno@...labora.com>
> > > > Signed-off-by: Hsiao Chien Sung <shawn.sung@...iatek.com>
> > > > ---
> > > > drivers/gpu/drm/mediatek/mtk_disp_drv.h | 1 +
> > > > drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 18 ++++++----
> > > > ----
> > > > ----
> > > > .../gpu/drm/mediatek/mtk_disp_ovl_adaptor.c | 9 +++++++++
> > > > drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c | 1 +
> > > > drivers/gpu/drm/mediatek/mtk_drm_plane.c | 2 +-
> > > > 5 files changed, 18 insertions(+), 13 deletions(-)
> > > >
> > > > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_drv.h
> > > > b/drivers/gpu/drm/mediatek/mtk_disp_drv.h
> > > > index 4d6e8b667bc3..c5afeb7c5527 100644
> > > > --- a/drivers/gpu/drm/mediatek/mtk_disp_drv.h
> > > > +++ b/drivers/gpu/drm/mediatek/mtk_disp_drv.h
> > > > @@ -127,6 +127,7 @@ void
> > > > mtk_ovl_adaptor_register_vblank_cb(struct
> > > > device *dev, void (*vblank_cb)(vo
> > > > void mtk_ovl_adaptor_unregister_vblank_cb(struct device *dev);
> > > > void mtk_ovl_adaptor_enable_vblank(struct device *dev);
> > > > void mtk_ovl_adaptor_disable_vblank(struct device *dev);
> > > > +unsigned int mtk_ovl_adaptor_supported_rotations(struct device
> > > > *dev);
> > > > void mtk_ovl_adaptor_start(struct device *dev);
> > > > void mtk_ovl_adaptor_stop(struct device *dev);
> > > > unsigned int mtk_ovl_adaptor_layer_nr(struct device *dev);
> > > > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > > > b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > > > index ecc38932fd44..319bbfde5cf9 100644
> > > > --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > > > +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > > > @@ -415,6 +415,10 @@ unsigned int mtk_ovl_layer_nr(struct
> > > > device
> > > > *dev)
> > > >
> > > > unsigned int mtk_ovl_supported_rotations(struct device *dev)
> > > > {
> > > > + /*
> > > > + * although currently OVL can only do reflection,
> > > > + * reflect x + reflect y = rotate 180
> > > > + */
> > > > return DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_180 |
> > > > DRM_MODE_REFLECT_X | DRM_MODE_REFLECT_Y;
> > > > }
> > > > @@ -423,27 +427,17 @@ int mtk_ovl_layer_check(struct device
> > > > *dev,
> > > > unsigned int idx,
> > > > struct mtk_plane_state *mtk_state)
> > > > {
> > > > struct drm_plane_state *state = &mtk_state->base;
> > > > - unsigned int rotation = 0;
> > > >
> > > > - rotation = drm_rotation_simplify(state->rotation,
> > > > - DRM_MODE_ROTATE_0 |
> > > > - DRM_MODE_REFLECT_X |
> > > > - DRM_MODE_REFLECT_Y);
> > > > - rotation &= ~DRM_MODE_ROTATE_0;
> > > > -
> > > > - /* We can only do reflection, not rotation */
> > > > - if ((rotation & DRM_MODE_ROTATE_MASK) != 0)
> > > > + if (state->rotation &
> > > > ~mtk_ovl_supported_rotations(dev))
> > > > return -EINVAL;
> > >
> > > The commit message of this patch is "Create rotation property
> > > according
> > > to the hardware capability". I think this modification is not
> > > related
> > > to create property, so separate this to another patch.
> >
> > Since these modifications are all related to rotation property, or
> > should I change the title and commit message to "Modify roation
> > property for passing IGT"?
>
> OK. But I don't know why do this, so describe more about this.
>
Got it, thanks. Will add more information to the commit message.
> >
> > >
> > > >
> > > > /*
> > > > * TODO: Rotating/reflecting YUV buffers is not
> > > > supported at
> > > > this time.
> > > > * Only RGB[AX] variants are supported.
> > > > */
> > > > - if (state->fb->format->is_yuv && rotation != 0)
> > > > + if (state->fb->format->is_yuv && (state->rotation &
> > > > ~DRM_MODE_ROTATE_0))
> > > > return -EINVAL;
> > > >
> > > > - state->rotation = rotation;
> > > > -
> > > > return 0;
> > > > }
> > > >
> > > > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
> > > > b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
> > > > index 4398db9a6276..273c79d37bef 100644
> > > > --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
> > > > +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
> > > > @@ -383,6 +383,15 @@ void
> > > > mtk_ovl_adaptor_register_vblank_cb(struct
> > > > device *dev, void (*vblank_cb)(vo
> > > > vblank_cb,
> > > > vblank_cb_data);
> > > > }
> > > >
> > > > +unsigned int mtk_ovl_adaptor_supported_rotations(struct device
> > > > *dev)
> > > > +{
> > > > + /*
> > > > + * should still return DRM_MODE_ROTATE_0 if rotation is
> > > > not
> > > > supported,
> > > > + * or IGT will fail.
> > > > + */
> > > > + return DRM_MODE_ROTATE_0;
> > > > +}
> > > > +
> > > > void mtk_ovl_adaptor_unregister_vblank_cb(struct device *dev)
> > > > {
> > > > struct mtk_disp_ovl_adaptor *ovl_adaptor =
> > > > dev_get_drvdata(dev);
> > > > diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > index ffa4868b1222..206dd6f6f99e 100644
> > > > --- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > +++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > @@ -422,6 +422,7 @@ static const struct mtk_ddp_comp_funcs
> > > > ddp_ovl_adaptor = {
> > > > .remove = mtk_ovl_adaptor_remove_comp,
> > > > .get_formats = mtk_ovl_adaptor_get_formats,
> > > > .get_num_formats = mtk_ovl_adaptor_get_num_formats,
> > > > + .supported_rotations =
> > > > mtk_ovl_adaptor_supported_rotations,
> > > > };
> > > >
> > > > static const char * const
> > > > mtk_ddp_comp_stem[MTK_DDP_COMP_TYPE_MAX]
> > > > =
> > > > {
> > > > diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> > > > b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> > > > index e2ec61b69618..894c39a38a58 100644
> > > > --- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> > > > +++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> > > > @@ -344,7 +344,7 @@ int mtk_plane_init(struct drm_device *dev,
> > > > struct
> > > > drm_plane *plane,
> > > > return err;
> > > > }
> > > >
> > > > - if (supported_rotations & ~DRM_MODE_ROTATE_0) {
> > > > + if (supported_rotations) {
> > >
> > > Why need this modification?
> > > Before Sean's patch [1], only support rotate 0 and does not
> > > create
> > > property and it works. Why does it must create property for only
> > > support rotate 0?
> >
> > Yes, as long as the user doesn't check rotation properties before
> > commiting the pitures, there will be no problem. But IGT somehow
> > checked this DRM_MODE_ROTATE_0 flag before executing the tests (not
> > all
> > of them), and leads to failures in these test itmes.
>
> OK, so, when supported_rotations == NULL (mtk_disp_rdma), it should
> also create rotation property. But I'm confused that if IGT assume
> that
> all drm driver should have rotation property, why drm core does not
> add
> this property for all drm driver?
>
After checking the report, only "graphics.IgtKms.kms_flip_unstable" has
such a constraint. However, since IGT version is still keep updating,
there are not guarantees about this behavior.
I did search for "DRM_MODE_ROTATE_0" in the kernel to check if there is
any rule or even suggestion about setting this property, but the flag
is rarely even mentioned.
Maybe we could also report this issue to IGT team instead of changing
the behavior of our driver.
> Regards,
> CK
>
> >
> > >
> > >
> > > [1]
> > >
> >
> >
>
>
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/drivers/gpu/drm/mediatek/mtk_drm_plane.c?h=v6.7-rc6&id=ef87d3e2dd251374c5c9fa3b6502aeff8fe29da9
> > >
> > > Regards,
> > > CK
> > >
> > > > err = drm_plane_create_rotation_property(plane,
> > > > DRM_MO
> > > > DE_ROTAT
> > > > E_0,
> > > > suppor
> > > > ted_rota
> > > > tions);
> > > > --
> > > > 2.39.2
> > > >
> >
> > Thanks,
> > Shawn
Powered by blists - more mailing lists