Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
MIME-Version: 1.0
References: <20220331193726.289559-1-nicolas.dufresne@collabora.com>
 <20220331193726.289559-14-nicolas.dufresne@collabora.com> <YkgwmwTMa83RysKB@eze-laptop>
 <92001085580ac1005596b34f79f1dadcc4b3f3c9.camel@collabora.com>
In-Reply-To: <92001085580ac1005596b34f79f1dadcc4b3f3c9.camel@collabora.com>
From:   Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Date:   Tue, 5 Apr 2022 12:34:30 -0300
Message-ID: <CAAEAJfA_+ZAcjHUN7Ydi0b=XB0=kFMaaPQC=h_44CNHq9rBsmQ@mail.gmail.com>
Subject: Re: [PATCH v2 13/23] media: rkvdec: h264: Fix dpb_valid implementation
To:     Nicolas Dufresne <nicolas.dufresne@collabora.com>
Cc:     Mauro Carvalho Chehab <mchehab@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Collabora Kernel ML <kernel@collabora.com>,
        Sebastian Fricke <sebastian.fricke@collabora.com>,
        linux-media <linux-media@vger.kernel.org>,
        "open list:ARM/Rockchip SoC..." <linux-rockchip@lists.infradead.org>,
        "open list:STAGING SUBSYSTEM" <linux-staging@lists.linux.dev>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Apr 5, 2022 at 12:11 PM Nicolas Dufresne
<nicolas.dufresne@collabora.com> wrote:
>
> Le samedi 02 avril 2022 =C3=A0 08:16 -0300, Ezequiel Garcia a =C3=A9crit =
:
> > On Thu, Mar 31, 2022 at 03:37:15PM -0400, Nicolas Dufresne wrote:
> > > The ref builder only provided references that are marked as valid in =
the
> > > dpb. Thus the current implementation of dpb_valid would always set th=
e
> > > flag to 1. This is not representing missing frames (this is called
> > > 'non-existing' pictures in the spec). In some context, these non-exis=
ting
> > > pictures still need to occupy a slot in the reference list according =
to
> > > the spec.
> > >
> >
> > Good catch! OOC, did you find this because it was failing a test vector=
?
>
> The effect is complex, so I could not correlate to specific tests. Also, =
what I
> wanted to fix isn't covered by the ITU conformance, its mostly resiliance
> requirement. But this should remove some of the IOMMU fault on broken str=
eams
> and make it less likely to use references that don't exists or aren't set=
 what
> we expect. After this change, the driver was getting more stable, and res=
ults
> was also more reproducible (specially in parallel decode case, which I us=
e to
> speed up testing).
>

Thanks for the details. This sounds like something that could
be added to the commit description itself.

> >
> > > Signed-off-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
> > > Reviewed-by: Sebastian Fricke <sebastian.fricke@collabora.com>
> >
> > Fixes: cd33c830448ba ("media: rkvdec: Add the rkvdec driver")
> > Reviewed-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
>
> Thanks for the review.
>

No problem :)

> >
> > Thanks,
> > Ezequiel
> >
> > > ---
> > >  drivers/staging/media/rkvdec/rkvdec-h264.c | 33 ++++++++++++++++----=
--
> > >  1 file changed, 24 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/drivers/staging/media/rkvdec/rkvdec-h264.c b/drivers/sta=
ging/media/rkvdec/rkvdec-h264.c
> > > index dff89732ddd0..bcde37d72244 100644
> > > --- a/drivers/staging/media/rkvdec/rkvdec-h264.c
> > > +++ b/drivers/staging/media/rkvdec/rkvdec-h264.c
> > > @@ -112,6 +112,7 @@ struct rkvdec_h264_run {
> > >     const struct v4l2_ctrl_h264_sps *sps;
> > >     const struct v4l2_ctrl_h264_pps *pps;
> > >     const struct v4l2_ctrl_h264_scaling_matrix *scaling_matrix;
> > > +   int ref_buf_idx[V4L2_H264_NUM_DPB_ENTRIES];
> > >  };
> > >
> > >  struct rkvdec_h264_ctx {
> > > @@ -725,6 +726,26 @@ static void assemble_hw_pps(struct rkvdec_ctx *c=
tx,
> > >     }
> > >  }
> > >
> > > +static void lookup_ref_buf_idx(struct rkvdec_ctx *ctx,
> > > +                          struct rkvdec_h264_run *run)
> > > +{
> > > +   const struct v4l2_ctrl_h264_decode_params *dec_params =3D run->de=
code_params;
> > > +   u32 i;
> > > +
> > > +   for (i =3D 0; i < ARRAY_SIZE(dec_params->dpb); i++) {
> > > +           struct v4l2_m2m_ctx *m2m_ctx =3D ctx->fh.m2m_ctx;
> > > +           const struct v4l2_h264_dpb_entry *dpb =3D run->decode_par=
ams->dpb;
> > > +           struct vb2_queue *cap_q =3D &m2m_ctx->cap_q_ctx.q;
> > > +           int buf_idx =3D -1;
> > > +
> > > +           if (dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE)
> > > +                   buf_idx =3D vb2_find_timestamp(cap_q,
> > > +                                                dpb[i].reference_ts,=
 0);
> > > +
> > > +           run->ref_buf_idx[i] =3D buf_idx;
> > > +   }
> > > +}
> > > +
> > >  static void assemble_hw_rps(struct rkvdec_ctx *ctx,
> > >                         struct rkvdec_h264_run *run)
> > >  {
> > > @@ -762,7 +783,7 @@ static void assemble_hw_rps(struct rkvdec_ctx *ct=
x,
> > >
> > >     for (j =3D 0; j < RKVDEC_NUM_REFLIST; j++) {
> > >             for (i =3D 0; i < h264_ctx->reflists.num_valid; i++) {
> > > -                   u8 dpb_valid =3D 0;
> > > +                   bool dpb_valid =3D run->ref_buf_idx[i] >=3D 0;
> > >                     u8 idx =3D 0;
> > >
> > >                     switch (j) {
> > > @@ -779,8 +800,6 @@ static void assemble_hw_rps(struct rkvdec_ctx *ct=
x,
> > >
> > >                     if (idx >=3D ARRAY_SIZE(dec_params->dpb))
> > >                             continue;
> > > -                   dpb_valid =3D !!(dpb[idx].flags &
> > > -                                  V4L2_H264_DPB_ENTRY_FLAG_ACTIVE);
> > >
> > >                     set_ps_field(hw_rps, DPB_INFO(i, j),
> > >                                  idx | dpb_valid << 4);
> > > @@ -859,13 +878,8 @@ get_ref_buf(struct rkvdec_ctx *ctx, struct rkvde=
c_h264_run *run,
> > >         unsigned int dpb_idx)
> > >  {
> > >     struct v4l2_m2m_ctx *m2m_ctx =3D ctx->fh.m2m_ctx;
> > > -   const struct v4l2_h264_dpb_entry *dpb =3D run->decode_params->dpb=
;
> > >     struct vb2_queue *cap_q =3D &m2m_ctx->cap_q_ctx.q;
> > > -   int buf_idx =3D -1;
> > > -
> > > -   if (dpb[dpb_idx].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE)
> > > -           buf_idx =3D vb2_find_timestamp(cap_q,
> > > -                                        dpb[dpb_idx].reference_ts, 0=
);
> > > +   int buf_idx =3D run->ref_buf_idx[dpb_idx];
> > >
> > >     /*
> > >      * If a DPB entry is unused or invalid, address of current destin=
ation
> > > @@ -1102,6 +1116,7 @@ static int rkvdec_h264_run(struct rkvdec_ctx *c=
tx)
> > >
> > >     assemble_hw_scaling_list(ctx, &run);
> > >     assemble_hw_pps(ctx, &run);
> > > +   lookup_ref_buf_idx(ctx, &run);
> > >     assemble_hw_rps(ctx, &run);
> > >     config_registers(ctx, &run);
> > >
> > > --
> > > 2.34.1
> > >
>