From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexandre Courbot Subject: Re: [PATCH] media: venus: fix reported size of 0-length buffers Date: Wed, 14 Nov 2018 13:12:46 +0900 Message-ID: References: <20181113093048.236201-1-acourbot@chromium.org> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Nicolas Dufresne Cc: Stanimir Varbanov , Mauro Carvalho Chehab , Linux Media Mailing List , linux-arm-msm@vger.kernel.org, LKML List-Id: linux-arm-msm@vger.kernel.org On Wed, Nov 14, 2018 at 3:54 AM Nicolas Dufresne wro= te: > > > > Le mar. 13 nov. 2018 04 h 30, Alexandre Courbot a= =C3=A9crit : >> >> The last buffer is often signaled by an empty buffer with the >> V4L2_BUF_FLAG_LAST flag set. Such buffers were returned with the >> bytesused field set to the full size of the OPB, which leads >> user-space to believe that the buffer actually contains useful data. Fix >> this by passing the number of bytes reported used by the firmware. > > > That means the driver does not know on time which one is last. Why not ju= st returned EPIPE to userspace on DQBUF and ovoid this useless roundtrip ? Sorry, I don't understand what you mean. EPIPE is supposed to be returned after a buffer with V4L2_BUF_FLAG_LAST is made available for dequeue. This patch amends the code that prepares this LAST-flagged buffer. How could we avoid a roundtrip in this case? > >> >> Signed-off-by: Alexandre Courbot >> --- >> drivers/media/platform/qcom/venus/vdec.c | 4 +--- >> 1 file changed, 1 insertion(+), 3 deletions(-) >> >> diff --git a/drivers/media/platform/qcom/venus/vdec.c b/drivers/media/pl= atform/qcom/venus/vdec.c >> index 189ec975c6bb..282de21cf2e1 100644 >> --- a/drivers/media/platform/qcom/venus/vdec.c >> +++ b/drivers/media/platform/qcom/venus/vdec.c >> @@ -885,10 +885,8 @@ static void vdec_buf_done(struct venus_inst *inst, = unsigned int buf_type, >> vbuf->field =3D V4L2_FIELD_NONE; >> >> if (type =3D=3D V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE) { >> - unsigned int opb_sz =3D venus_helper_get_opb_size(inst); >> - >> vb =3D &vbuf->vb2_buf; >> - vb2_set_plane_payload(vb, 0, bytesused ? : opb_sz); >> + vb2_set_plane_payload(vb, 0, bytesused); >> vb->planes[0].data_offset =3D data_offset; >> vb->timestamp =3D timestamp_us * NSEC_PER_USEC; >> vbuf->sequence =3D inst->sequence_cap++; >> -- >> 2.19.1.930.g4563a0d9d0-goog >>