| 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: <9208d58f46f5e748daddaa9c96e777985cbbc5c7.camel@mediatek.com>
Date: Sat, 4 Sep 2021 11:00:22 +0800
From: "yunfei.dong@...iatek.com" <yunfei.dong@...iatek.com>
To: Dafna Hirschfeld <dafna.hirschfeld@...labora.com>,
Alexandre Courbot <acourbot@...omium.org>,
Hans Verkuil <hverkuil-cisco@...all.nl>,
"Tzung-Bi Shih" <tzungbi@...omium.org>,
Tiffany Lin <tiffany.lin@...iatek.com>,
Andrew-CT Chen <andrew-ct.chen@...iatek.com>,
Mauro Carvalho Chehab <mchehab@...nel.org>,
Rob Herring <robh+dt@...nel.org>,
Matthias Brugger <matthias.bgg@...il.com>,
Tomasz Figa <tfiga@...gle.com>,
Laurent Pinchart <laurent.pinchart@...asonboard.com>
CC: Daniel Vetter <daniel@...ll.ch>,
dri-devel <dri-devel@...ts.freedesktop.org>,
Hsin-Yi Wang <hsinyi@...omium.org>,
"Fritz Koenig" <frkoenig@...omium.org>,
Irui Wang <irui.wang@...iatek.com>,
<linux-media@...r.kernel.org>, <devicetree@...r.kernel.org>,
<linux-kernel@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
<srv_heupstream@...iatek.com>,
<linux-mediatek@...ts.infradead.org>,
<Project_Global_Chrome_Upstream_Group@...iatek.com>
Subject: Re: [PATCH v6, 15/15] media: mtk-vcodec: Use codec type to separate
different hardware
Hi Danfa,
Thanks for your suggestion.
On Fri, 2021-09-03 at 14:42 +0200, Dafna Hirschfeld wrote:
>
> On 02.09.21 08:05, yunfei.dong@...iatek.com wrote:
> > On Wed, 2021-09-01 at 14:17 +0200, Dafna Hirschfeld wrote:
> > Hi Dafna,
> >
> > Thanks for your suggestion.
> > > Hi
> > >
> > > On 01.09.21 10:32, Yunfei Dong wrote:
> > > > There are just one core thread, in order to separeate different
> > > > hardware, using codec type to separeate it in scp driver.
> > >
> > > this code seems to relate to the vpu driver not the scp driver.
> > > Is there a corresponding code added to the vpu driver that test
> > > the
> > > codec_type?
> > >
> >
> > Vpu is video processor unit, used to connect with micro processor.
> > In mt8173: vdec_vpu_if.c -> mtk_vpu.c -> micro processor
> > In mt8192/mt8183: vdec_vpu_if.c -> mtk_scp.c ->micro processor
> >
> > This init/dec start/dec_end interfaces are the same for vpu and
> > scp.
> > > >
> > > > Signed-off-by: Yunfei Dong <yunfei.dong@...iatek.com>
> > > > ---
> > > > .../media/platform/mtk-vcodec/vdec_ipi_msg.h | 12 ++++---
> > > > .../media/platform/mtk-vcodec/vdec_vpu_if.c | 34
> > > > ++++++++++++++++---
> > > > .../media/platform/mtk-vcodec/vdec_vpu_if.h | 4 +++
> > > > 3 files changed, 41 insertions(+), 9 deletions(-)
> > > >
> > > > diff --git a/drivers/media/platform/mtk-vcodec/vdec_ipi_msg.h
> > > > b/drivers/media/platform/mtk-vcodec/vdec_ipi_msg.h
> > > > index 9d8079c4f976..c488f0c40190 100644
> > > > --- a/drivers/media/platform/mtk-vcodec/vdec_ipi_msg.h
> > > > +++ b/drivers/media/platform/mtk-vcodec/vdec_ipi_msg.h
> > > > @@ -35,6 +35,8 @@ enum vdec_ipi_msgid {
> > > > * @msg_id : vdec_ipi_msgid
> > > > * @vpu_inst_addr : VPU decoder instance address. Used if
> > > > ABI
> > > > version < 2.
> > > > * @inst_id : instance ID. Used if the ABI version >= 2.
> > > > + * @codec_type : Codec fourcc
> > > > + * @reserved : reserved param
> > > > */
> > > > struct vdec_ap_ipi_cmd {
> > > > uint32_t msg_id;
> > > > @@ -42,6 +44,8 @@ struct vdec_ap_ipi_cmd {
> > > > uint32_t vpu_inst_addr;
> > > > uint32_t inst_id;
> > > > };
> > > > + uint32_t codec_type;
> > > > + uint32_t reserved;
> > > > };
> > > >
> > > > /**
> > > > @@ -59,12 +63,12 @@ struct vdec_vpu_ipi_ack {
> > > > /**
> > > > * struct vdec_ap_ipi_init - for AP_IPIMSG_DEC_INIT
> > > > * @msg_id : AP_IPIMSG_DEC_INIT
> > > > - * @reserved : Reserved field
> > > > + * @codec_type : Codec fourcc
> > > > * @ap_inst_addr : AP video decoder instance address
> > > > */
> > > > struct vdec_ap_ipi_init {
> > > > uint32_t msg_id;
> > > > - uint32_t reserved;
> > > > + uint32_t codec_type;
> > > > uint64_t ap_inst_addr;
> > > > };
> > > >
> > > > @@ -77,7 +81,7 @@ struct vdec_ap_ipi_init {
> > > > * H264 decoder [0]:buf_sz [1]:nal_start
> > > > * VP8 decoder [0]:width/height
> > > > * VP9 decoder [0]:profile, [1][2] width/height
> > > > - * @reserved : Reserved field
> > > > + * @codec_type : Codec fourcc
> > > > */
> > > > struct vdec_ap_ipi_dec_start {
> > > > uint32_t msg_id;
> > > > @@ -86,7 +90,7 @@ struct vdec_ap_ipi_dec_start {
> > > > uint32_t inst_id;
> > > > };
> > > > uint32_t data[3];
> > > > - uint32_t reserved;
> > > > + uint32_t codec_type;
> > > > };
> > > >
> > > > /**
> > > > diff --git a/drivers/media/platform/mtk-vcodec/vdec_vpu_if.c
> > > > b/drivers/media/platform/mtk-vcodec/vdec_vpu_if.c
> > > > index bfd8e87dceff..c84fac52fe26 100644
> > > > --- a/drivers/media/platform/mtk-vcodec/vdec_vpu_if.c
> > > > +++ b/drivers/media/platform/mtk-vcodec/vdec_vpu_if.c
> > > > @@ -100,18 +100,29 @@ static void vpu_dec_ipi_handler(void
> > > > *data,
> > > > unsigned int len, void *priv)
> > > >
> > > > static int vcodec_vpu_send_msg(struct vdec_vpu_inst *vpu,
> > > > void
> > > > *msg, int len)
> > > > {
> > > > - int err;
> > > > + int err, id, msgid;
> > > >
> > > > - mtk_vcodec_debug(vpu, "id=%X", *(uint32_t *)msg);
> > > > + msgid = *(uint32_t *)msg;
> > > > + mtk_vcodec_debug(vpu, "id=%X", msgid);
> > > >
> > > > vpu->failure = 0;
> > > > vpu->signaled = 0;
> > > >
> > > > - err = mtk_vcodec_fw_ipi_send(vpu->ctx->dev->fw_handler,
> > > > vpu-
> > > > > id, msg,
> > > >
> > > > + if (vpu->ctx->dev->vdec_pdata->hw_arch ==
> > > > MTK_VDEC_LAT_SINGLE_CORE) {
> > > > + if (msgid == AP_IPIMSG_DEC_CORE ||
> > > > + msgid == AP_IPIMSG_DEC_CORE_END)
> > > > + id = vpu->core_id;
> > > > + else
> > > > + id = vpu->id;
> > > > + } else {
> > > > + id = vpu->id;
> > > > + }
> > > > +
> > > > + err = mtk_vcodec_fw_ipi_send(vpu->ctx->dev->fw_handler,
> > > > id,
> > > > msg,
> > > > len, 2000);
> > >
> > > so
> > > > if (err) {
> > > > mtk_vcodec_err(vpu, "send fail vpu_id=%d
> > > > msg_id=%X
> > > > status=%d",
> > > > - vpu->id, *(uint32_t *)msg, err);
> > > > + id, msgid, err);
> > > > return err;
> > > > }
> > > >
> > > > @@ -131,6 +142,7 @@ static int vcodec_send_ap_ipi(struct
> > > > vdec_vpu_inst *vpu, unsigned int msg_id)
> > > > msg.vpu_inst_addr = vpu->inst_addr;
> > > > else
> > > > msg.inst_id = vpu->inst_id;
> > > > + msg.codec_type = vpu->codec_type;
> > > >
> > > > err = vcodec_vpu_send_msg(vpu, &msg, sizeof(msg));
> > > > mtk_vcodec_debug(vpu, "- id=%X ret=%d", msg_id, err);
> > > > @@ -149,14 +161,25 @@ int vpu_dec_init(struct vdec_vpu_inst
> > > > *vpu)
> > > >
> > > > err = mtk_vcodec_fw_ipi_register(vpu->ctx->dev-
> > > > >fw_handler,
> > > > vpu->id,
> > > > vpu->handler, "vdec",
> > > > NULL);
> > > > - if (err != 0) {
> > > > + if (err) {
> > >
> > > could be nice to send a patch with other such fixes,
> > > anyway it is better to send unrelated fixes in a separate patch
> > >
> >
> > will fix in next patch.
> > > > mtk_vcodec_err(vpu, "vpu_ipi_register fail
> > > > status=%d",
> > > > err);
> > > > return err;
> > > > }
> > > >
> > > > + if (vpu->ctx->dev->vdec_pdata->hw_arch ==
> > > > MTK_VDEC_LAT_SINGLE_CORE) {
> > > > + err = mtk_vcodec_fw_ipi_register(vpu->ctx->dev-
> > > > > fw_handler,
> > > >
> > > > + vpu->core_id, vpu-
> > > > >handler,
> > > > + "vdec", NULL);
> > > > + if (err) {
> > > > + mtk_vcodec_err(vpu, "vpu_ipi_register
> > > > core fail
> > > > status=%d", err);
> > > > + return err;
> > > > + }
> > > > + }
> > > > +
> > > > memset(&msg, 0, sizeof(msg));
> > > > msg.msg_id = AP_IPIMSG_DEC_INIT;
> > > > msg.ap_inst_addr = (unsigned long)vpu;
> > > > + msg.codec_type = vpu->codec_type;
> > > >
> > > > mtk_vcodec_debug(vpu, "vdec_inst=%p", vpu);
> > > >
> > > > @@ -187,6 +210,7 @@ int vpu_dec_start(struct vdec_vpu_inst
> > > > *vpu,
> > > > uint32_t *data, unsigned int len)
> > > >
> > > > for (i = 0; i < len; i++)
> > > > msg.data[i] = data[i];
> > > > + msg.codec_type = vpu->codec_type;
> > >
> > > I don't see where is the vpu->codec_type initialzied
> > >
> >
> > This patch just add interface to support core hardware decode, in
> > next
> > serial patches based on these will used codec type to separate
> > after
> > these base patches are stable.
>
> Then why not send it on the same series?
>
For this patch is the common part, not only one spec(h264/vp9/av1) use
it. And these specs will send in different patch series, in order to
avoid patch dependency, so add this patch in this common series.
> Thanks,
> Dafna
>
Thanks
Yunfei Dong
> > > Thanks,
> > > Dafna
> > >
> >
> > Thanks
> > Yunfei Dong
> > > >
> > > > err = vcodec_vpu_send_msg(vpu, (void *)&msg,
> > > > sizeof(msg));
> > > > mtk_vcodec_debug(vpu, "- ret=%d", err);
> > > > diff --git a/drivers/media/platform/mtk-vcodec/vdec_vpu_if.h
> > > > b/drivers/media/platform/mtk-vcodec/vdec_vpu_if.h
> > > > index ae24b75d1649..802660770a87 100644
> > > > --- a/drivers/media/platform/mtk-vcodec/vdec_vpu_if.h
> > > > +++ b/drivers/media/platform/mtk-vcodec/vdec_vpu_if.h
> > > > @@ -14,6 +14,7 @@ struct mtk_vcodec_ctx;
> > > > /**
> > > > * struct vdec_vpu_inst - VPU instance for video codec
> > > > * @id : ipi msg id for each decoder
> > > > + * @core_id : core id used to separate different hardware
> > > > * @vsi : driver structure allocated by VPU side and
> > > > shared to AP side
> > > > * for control and info share
> > > > * @failure : VPU execution result status, 0: success,
> > > > others: fail
> > > > @@ -26,9 +27,11 @@ struct mtk_vcodec_ctx;
> > > > * @dev : platform device of VPU
> > > > * @wq : wait queue to wait VPU message ack
> > > > * @handler : ipi handler for each decoder
> > > > + * @codec_type : used codec type to separate different
> > > > codecs
> > > > */
> > > > struct vdec_vpu_inst {
> > > > int id;
> > > > + int core_id;
> > > > void *vsi;
> > > > int32_t failure;
> > > > uint32_t inst_addr;
> > > > @@ -38,6 +41,7 @@ struct vdec_vpu_inst {
> > > > struct mtk_vcodec_ctx *ctx;
> > > > wait_queue_head_t wq;
> > > > mtk_vcodec_ipi_handler handler;
> > > > + unsigned int codec_type;
> > > > };
> > > >
> > > > /**
> > > >
Powered by blists - more mailing lists