public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
To: Tomasz Figa <tfiga@chromium.org>
Cc: linux-media@vger.kernel.org, linux-kernel@vger.kernel.org,
	Hans Verkuil <hverkuil-cisco@xs4all.nl>,
	Marek Szyprowski <m.szyprowski@samsung.com>
Subject: Re: [PATCH 6/8] rkvdec: Use vb2_find_buffer
Date: Fri, 8 Jul 2022 08:21:23 -0300	[thread overview]
Message-ID: <YsgTM1zYMDHa2HJd@eze-laptop> (raw)
In-Reply-To: <CAAFQd5BSDnZ6MyXfcRWNM9f9WF82sjPX3F1ZjYjcOuz3b1fPkw@mail.gmail.com>

Hi Tomasz,

On Fri, Jul 08, 2022 at 01:40:53PM +0900, Tomasz Figa wrote:
> Hi Ezequiel,
> 
> On Thu, Jul 7, 2022 at 3:27 AM Ezequiel Garcia
> <ezequiel@vanguardiasur.com.ar> wrote:
> >
> > Use the newly introduced vb2_find_buffer API to get a vb2_buffer
> > given a buffer timestamp.
> >
> > Signed-off-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
> > ---
> >  drivers/staging/media/rkvdec/rkvdec-h264.c | 41 ++++++++--------------
> >  drivers/staging/media/rkvdec/rkvdec-vp9.c  | 10 +++---
> >  2 files changed, 19 insertions(+), 32 deletions(-)
> >
> > diff --git a/drivers/staging/media/rkvdec/rkvdec-h264.c b/drivers/staging/media/rkvdec/rkvdec-h264.c
> > index 2992fb87cf72..4af5a831bde0 100644
> > --- a/drivers/staging/media/rkvdec/rkvdec-h264.c
> > +++ b/drivers/staging/media/rkvdec/rkvdec-h264.c
> > @@ -109,7 +109,7 @@ 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 vb2_buffer *ref_buf[V4L2_H264_NUM_DPB_ENTRIES];
> 
> How do we guarantee that those pointers remain valid through the
> lifetime of this structure?
> 

The rkvdec_h264_run is populated in .device_run, and used to program
the hardware for each decode job.

So these videobuf2 buffer won't outlive a given decode job. The vb2
queue can't be released (so buffers can't be released) while
a job is runnning (i.e. the driver is in a "streaming" state).

We should be good, right?

Thanks for the review,
Ezequiel

> Best regards,
> Tomasz
> 
> >  };
> >
> >  struct rkvdec_h264_ctx {
> > @@ -742,17 +742,16 @@ static void lookup_ref_buf_idx(struct rkvdec_ctx *ctx,
> >                 struct v4l2_m2m_ctx *m2m_ctx = ctx->fh.m2m_ctx;
> >                 const struct v4l2_h264_dpb_entry *dpb = run->decode_params->dpb;
> >                 struct vb2_queue *cap_q = &m2m_ctx->cap_q_ctx.q;
> > -               int buf_idx = -1;
> > +               struct vb2_buffer *buf = NULL;
> >
> >                 if (dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE) {
> > -                       buf_idx = vb2_find_timestamp(cap_q,
> > -                                                    dpb[i].reference_ts, 0);
> > -                       if (buf_idx < 0)
> > +                       buf = vb2_find_buffer(cap_q, dpb[i].reference_ts);
> > +                       if (!buf)
> >                                 pr_debug("No buffer for reference_ts %llu",
> >                                          dpb[i].reference_ts);
> >                 }
> >
> > -               run->ref_buf_idx[i] = buf_idx;
> > +               run->ref_buf[i] = buf;
> >         }
> >  }
> >
> > @@ -805,7 +804,7 @@ static void assemble_hw_rps(struct rkvdec_ctx *ctx,
> >                         if (WARN_ON(ref->index >= ARRAY_SIZE(dec_params->dpb)))
> >                                 continue;
> >
> > -                       dpb_valid = run->ref_buf_idx[ref->index] >= 0;
> > +                       dpb_valid = run->ref_buf[ref->index] != NULL;
> >                         bottom = ref->fields == V4L2_H264_BOTTOM_FIELD_REF;
> >
> >                         set_ps_field(hw_rps, DPB_INFO(i, j),
> > @@ -881,24 +880,6 @@ static const u32 poc_reg_tbl_bottom_field[16] = {
> >         RKVDEC_REG_H264_POC_REFER2(1)
> >  };
> >
> > -static struct vb2_buffer *
> > -get_ref_buf(struct rkvdec_ctx *ctx, struct rkvdec_h264_run *run,
> > -           unsigned int dpb_idx)
> > -{
> > -       struct v4l2_m2m_ctx *m2m_ctx = ctx->fh.m2m_ctx;
> > -       struct vb2_queue *cap_q = &m2m_ctx->cap_q_ctx.q;
> > -       int buf_idx = run->ref_buf_idx[dpb_idx];
> > -
> > -       /*
> > -        * If a DPB entry is unused or invalid, address of current destination
> > -        * buffer is returned.
> > -        */
> > -       if (buf_idx < 0)
> > -               return &run->base.bufs.dst->vb2_buf;
> > -
> > -       return vb2_get_buffer(cap_q, buf_idx);
> > -}
> > -
> >  static void config_registers(struct rkvdec_ctx *ctx,
> >                              struct rkvdec_h264_run *run)
> >  {
> > @@ -971,8 +952,14 @@ static void config_registers(struct rkvdec_ctx *ctx,
> >
> >         /* config ref pic address & poc */
> >         for (i = 0; i < ARRAY_SIZE(dec_params->dpb); i++) {
> > -               struct vb2_buffer *vb_buf = get_ref_buf(ctx, run, i);
> > -
> > +               struct vb2_buffer *vb_buf = run->ref_buf[i];
> > +
> > +               /*
> > +                * If a DPB entry is unused or invalid, address of current destination
> > +                * buffer is returned.
> > +                */
> > +               if (!vb_buf)
> > +                       vb_buf = &dst_buf->vb2_buf;
> >                 refer_addr = vb2_dma_contig_plane_dma_addr(vb_buf, 0);
> >
> >                 if (dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE)
> > diff --git a/drivers/staging/media/rkvdec/rkvdec-vp9.c b/drivers/staging/media/rkvdec/rkvdec-vp9.c
> > index c2f42e76be10..d8c1c0db15c7 100644
> > --- a/drivers/staging/media/rkvdec/rkvdec-vp9.c
> > +++ b/drivers/staging/media/rkvdec/rkvdec-vp9.c
> > @@ -383,17 +383,17 @@ get_ref_buf(struct rkvdec_ctx *ctx, struct vb2_v4l2_buffer *dst, u64 timestamp)
> >  {
> >         struct v4l2_m2m_ctx *m2m_ctx = ctx->fh.m2m_ctx;
> >         struct vb2_queue *cap_q = &m2m_ctx->cap_q_ctx.q;
> > -       int buf_idx;
> > +       struct vb2_buffer *buf;
> >
> >         /*
> >          * If a ref is unused or invalid, address of current destination
> >          * buffer is returned.
> >          */
> > -       buf_idx = vb2_find_timestamp(cap_q, timestamp, 0);
> > -       if (buf_idx < 0)
> > -               return vb2_to_rkvdec_decoded_buf(&dst->vb2_buf);
> > +       buf = vb2_find_buffer(cap_q, timestamp);
> > +       if (!buf)
> > +               buf = &dst->vb2_buf;
> >
> > -       return vb2_to_rkvdec_decoded_buf(vb2_get_buffer(cap_q, buf_idx));
> > +       return vb2_to_rkvdec_decoded_buf(buf);
> >  }
> >
> >  static dma_addr_t get_mv_base_addr(struct rkvdec_decoded_buffer *buf)
> > --
> > 2.34.3
> >

  reply	other threads:[~2022-07-08 11:21 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-06 18:26 [PATCH 0/8] videobuf2: Replace vb2_find_timestamp() with vb2_find_buffer() Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 1/8] videobuf2: Introduce vb2_find_buffer() Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 2/8] mediatek: vcodec: Use vb2_find_buffer Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 3/8] tegra-vde: " Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 4/8] vicodec: " Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 5/8] hantro: " Ezequiel Garcia
2022-07-06 18:26 ` [PATCH 6/8] rkvdec: " Ezequiel Garcia
2022-07-08  4:40   ` Tomasz Figa
2022-07-08 11:21     ` Ezequiel Garcia [this message]
2022-07-08 11:45       ` Tomasz Figa
2022-07-06 18:26 ` [PATCH 7/8] cedrus: " Ezequiel Garcia
2022-07-11 18:33   ` Jernej Škrabec
2022-07-06 18:26 ` [PATCH 8/8] videobuf2: Remove vb2_find_timestamp() Ezequiel Garcia
2022-07-08  4:45   ` Tomasz Figa
2022-07-08 11:15     ` Ezequiel Garcia
2022-07-08  4:47 ` [PATCH 0/8] videobuf2: Replace vb2_find_timestamp() with vb2_find_buffer() Tomasz Figa
2022-07-08 11:47   ` Tomasz Figa

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YsgTM1zYMDHa2HJd@eze-laptop \
    --to=ezequiel@vanguardiasur.com.ar \
    --cc=hverkuil-cisco@xs4all.nl \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=m.szyprowski@samsung.com \
    --cc=tfiga@chromium.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox