<html> <head> <meta content="text/html; charset=iso-8859-15" http-equiv="Content-Type"> </head> <body text="#000000" bgcolor="#FFFFFF"> <div class="moz-cite-prefix">Am 20.07.2016 um 06:21 schrieb Zhang, Boyuan: </div> <blockquote cite="mid:CY4PR12MB1127076C0EAC7F6C4FC3D5C987080@CY4PR12MB1127.namprd12.prod.outlook.com" type="cite"> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-15"> <meta name="Generator" content="Microsoft Exchange Server">  <style></style> <meta content="text/html; charset=UTF-8"> <style type="text/css" style="">  </style> <div dir="ltr"> <div id="x_divtagdefaultwrapper" style="font-size:12pt; color:#000000; background-color:#FFFFFF; font-family:Calibri,Arial,Helvetica,sans-serif"> >> - context->decoder->begin_frame(context->decoder, context->target, &context->desc.base); >> + if (context->decoder->entrypoint != PIPE_VIDEO_ENTRYPOINT_ENCODE) >> + context->decoder->begin_frame(context->decoder, context->target, &context->desc.base); >Why do we do so here? Could we avoid that? >I would rather like to keep the begin_frame()/end_frame() handling as it is. >Christian. This is on purpose. Based on my testing, application will call begin_frame first, then call PictureParameter/SequenceParameter/... to pass us all picture related parameters. However, some of those values are actually required by begin_picture call in radeon_vce. So we have to delay the call until we receive all the parameters that needed. Same applies to encode_bitstream call. That's why I delay both calls to end_frame where we get all necessary values. </div> </div> </blockquote> We can keep it like this for now, but I would prefer that we clean this up and change the radeon_vce so that it matches the begin/encode/end calls from VA-API. We should probably work on this together with the performance improvements. Regards, Christian. <blockquote cite="mid:CY4PR12MB1127076C0EAC7F6C4FC3D5C987080@CY4PR12MB1127.namprd12.prod.outlook.com" type="cite"> <div dir="ltr"> <div id="x_divtagdefaultwrapper" style="font-size:12pt; color:#000000; background-color:#FFFFFF; font-family:Calibri,Arial,Helvetica,sans-serif"> Regards, Boyuan </div> <hr tabindex="-1" style="display:inline-block; width:98%"> <div id="x_divRplyFwdMsg" dir="ltr">From: Christian König <a class="moz-txt-link-rfc2396E" href="mailto:deathsimple@vodafone.de"><deathsimple@vodafone.de></a> Sent: July 19, 2016 4:55:43 AM To: Zhang, Boyuan; <a class="moz-txt-link-abbreviated" href="mailto:mesa-dev@lists.freedesktop.org">mesa-dev@lists.freedesktop.org</a> Cc: <a class="moz-txt-link-abbreviated" href="mailto:adf.lists@gmail.com">adf.lists@gmail.com</a> Subject: Re: [PATCH 09/12] st/va: add functions for VAAPI encode <div> </div> </div> </div> <div class="PlainText">Am 19.07.2016 um 00:43 schrieb Boyuan Zhang: > Add necessary functions/changes for VAAPI encoding to buffer and picture. These changes will allow driver to handle all Vaapi encode related operations. This patch doesn't change the Vaapi decode behaviour. > > Signed-off-by: Boyuan Zhang <a class="moz-txt-link-rfc2396E" href="mailto:boyuan.zhang@amd.com"><boyuan.zhang@amd.com></a> > --- > src/gallium/state_trackers/va/buffer.c | 6 + > src/gallium/state_trackers/va/picture.c | 169 ++++++++++++++++++++++++++++- > src/gallium/state_trackers/va/va_private.h | 3 + > 3 files changed, 176 insertions(+), 2 deletions(-) > > diff --git a/src/gallium/state_trackers/va/buffer.c b/src/gallium/state_trackers/va/buffer.c > index 7d3167b..dfcebbe 100644 > --- a/src/gallium/state_trackers/va/buffer.c > +++ b/src/gallium/state_trackers/va/buffer.c > @@ -133,6 +133,12 @@ vlVaMapBuffer(VADriverContextP ctx, VABufferID buf_id, void **pbuff) > if (!buf->derived_surface.transfer || !*pbuff) > return VA_STATUS_ERROR_INVALID_BUFFER; > > + if (buf->type == VAEncCodedBufferType) { > + ((VACodedBufferSegment*)buf->data)->buf = *pbuff; > + ((VACodedBufferSegment*)buf->data)->size = buf->coded_size; > + ((VACodedBufferSegment*)buf->data)->next = NULL; > + *pbuff = buf->data; > + } > } else { > pipe_mutex_unlock(drv->mutex); > *pbuff = buf->data; > diff --git a/src/gallium/state_trackers/va/picture.c b/src/gallium/state_trackers/va/picture.c > index 89ac024..4793194 100644 > --- a/src/gallium/state_trackers/va/picture.c > +++ b/src/gallium/state_trackers/va/picture.c > @@ -78,7 +78,8 @@ vlVaBeginPicture(VADriverContextP ctx, VAContextID context_id, VASurfaceID rende > return VA_STATUS_SUCCESS; > } > > - context->decoder->begin_frame(context->decoder, context->target, &context->desc.base); > + if (context->decoder->entrypoint != PIPE_VIDEO_ENTRYPOINT_ENCODE) > + context->decoder->begin_frame(context->decoder, context->target, &context->desc.base); Why do we do so here? Could we avoid that? I would rather like to keep the begin_frame()/end_frame() handling as it is. Christian. > > return VA_STATUS_SUCCESS; > } > @@ -278,6 +279,139 @@ handleVASliceDataBufferType(vlVaContext *context, vlVaBuffer *buf) > num_buffers, (const void * const*)buffers, sizes); > } > > +static VAStatus > +handleVAEncMiscParameterTypeRateControl(vlVaContext *context, VAEncMiscParameterBuffer *misc) > +{ > + VAEncMiscParameterRateControl *rc = (VAEncMiscParameterRateControl *)misc->data; > + if (context->desc.h264enc.rate_ctrl.rate_ctrl_method == > + PIPE_H264_ENC_RATE_CONTROL_METHOD_CONSTANT) > + context->desc.h264enc.rate_ctrl.target_bitrate = rc->bits_per_second; > + else > + context->desc.h264enc.rate_ctrl.target_bitrate = rc->bits_per_second * rc->target_percentage; > + context->desc.h264enc.rate_ctrl.peak_bitrate = rc->bits_per_second; > + if (context->desc.h264enc.rate_ctrl.target_bitrate < 2000000) > + context->desc.h264enc.rate_ctrl.vbv_buffer_size = MIN2((context->desc.h264enc.rate_ctrl.target_bitrate * 2.75), 2000000); > + else > + context->desc.h264enc.rate_ctrl.vbv_buffer_size = context->desc.h264enc.rate_ctrl.target_bitrate; > + context->desc.h264enc.rate_ctrl.target_bits_picture = > + context->desc.h264enc.rate_ctrl.target_bitrate / context->desc.h264enc.rate_ctrl.frame_rate_num; > + context->desc.h264enc.rate_ctrl.peak_bits_picture_integer = > + context->desc.h264enc.rate_ctrl.peak_bitrate / context->desc.h264enc.rate_ctrl.frame_rate_num; > + context->desc.h264enc.rate_ctrl.peak_bits_picture_fraction = 0; > + > + return VA_STATUS_SUCCESS; > +} > + > +static VAStatus > +handleVAEncSequenceParameterBufferType(vlVaDriver *drv, vlVaContext *context, vlVaBuffer *buf) > +{ > + VAEncSequenceParameterBufferH264 *h264 = (VAEncSequenceParameterBufferH264 *)buf->data; > + if (!context->decoder) { > + context->templat.max_references = h264->max_num_ref_frames; > + context->templat.level = h264->level_idc; > + context->decoder = drv->pipe->create_video_codec(drv->pipe, &context->templat); > + if (!context->decoder) > + return VA_STATUS_ERROR_ALLOCATION_FAILED; > + } > + context->desc.h264enc.gop_size = h264->intra_idr_period; > + context->desc.h264enc.rate_ctrl.frame_rate_num = h264->time_scale / 2; > + context->desc.h264enc.rate_ctrl.frame_rate_den = 1; > + return VA_STATUS_SUCCESS; > +} > + > +static VAStatus > +handleVAEncMiscParameterBufferType(vlVaContext *context, vlVaBuffer *buf) > +{ > + VAStatus vaStatus = VA_STATUS_SUCCESS; > + VAEncMiscParameterBuffer *misc; > + misc = buf->data; > + > + switch (misc->type) { > + case VAEncMiscParameterTypeRateControl: > + vaStatus = handleVAEncMiscParameterTypeRateControl(context, misc); > + break; > + > + default: > + break; > + } > + > + return vaStatus; > +} > + > +static VAStatus > +handleVAEncPictureParameterBufferType(vlVaDriver *drv, vlVaContext *context, vlVaBuffer *buf) > +{ > + VAEncPictureParameterBufferH264 *h264; > + vlVaBuffer *coded_buf; > + > + h264 = buf->data; > + context->desc.h264enc.frame_num = h264->frame_num; > + context->desc.h264enc.not_referenced = false; > + context->desc.h264enc.is_idr = (h264->pic_fields.bits.idr_pic_flag == 1); > + context->desc.h264enc.pic_order_cnt = h264->CurrPic.TopFieldOrderCnt / 2; > + if (context->desc.h264enc.is_idr) > + context->desc.h264enc.i_remain = 1; > + else > + context->desc.h264enc.i_remain = 0; > + > + context->desc.h264enc.p_remain = context->desc.h264enc.gop_size - context->desc.h264enc.gop_cnt - context->desc.h264enc.i_remain; > + > + coded_buf = handle_table_get(drv->htab, h264->coded_buf); > + if (!coded_buf->derived_surface.resource) > + coded_buf->derived_surface.resource = pipe_buffer_create(drv->pipe->screen, PIPE_BIND_VERTEX_BUFFER, > + PIPE_USAGE_STREAM, coded_buf->size); > + context->coded_buf = coded_buf; > + > + context->desc.h264enc.frame_idx[h264->CurrPic.picture_id] = h264->frame_num; > + if (context->desc.h264enc.is_idr) > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_IDR; > + else > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_P; > + > + context->desc.h264enc.frame_num_cnt++; > + context->desc.h264enc.gop_cnt++; > + if (context->desc.h264enc.gop_cnt == context->desc.h264enc.gop_size) > + context->desc.h264enc.gop_cnt = 0; > + > + return VA_STATUS_SUCCESS; > +} > + > +static VAStatus > +handleVAEncSliceParameterBufferType(vlVaDriver *drv, vlVaContext *context, vlVaBuffer *buf) > +{ > + VAEncSliceParameterBufferH264 *h264; > + > + h264 = buf->data; > + context->desc.h264enc.ref_idx_l0 = VA_INVALID_ID; > + context->desc.h264enc.ref_idx_l1 = VA_INVALID_ID; > + > + for (int i = 0; i < 32; i++) { > + if (h264->RefPicList0[i].picture_id != VA_INVALID_ID) { > + if (context->desc.h264enc.ref_idx_l0 == VA_INVALID_ID) > + context->desc.h264enc.ref_idx_l0 = context->desc.h264enc.frame_idx[h264->RefPicList0[i].picture_id]; > + } > + if (h264->RefPicList1[i].picture_id != VA_INVALID_ID && h264->slice_type == 1) { > + if (context->desc.h264enc.ref_idx_l1 == VA_INVALID_ID) > + context->desc.h264enc.ref_idx_l1 = context->desc.h264enc.frame_idx[h264->RefPicList1[i].picture_id]; > + } > + } > + > + if (h264->slice_type == 1) > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_B; > + else if (h264->slice_type == 0) > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_P; > + else if (h264->slice_type == 2) { > + if (context->desc.h264enc.is_idr){ > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_IDR; > + context->desc.h264enc.idr_pic_id++; > + } else > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_I; > + } else > + context->desc.h264enc.picture_type = PIPE_H264_ENC_PICTURE_TYPE_SKIP; > + > + return VA_STATUS_SUCCESS; > +} > + > VAStatus > vlVaRenderPicture(VADriverContextP ctx, VAContextID context_id, VABufferID *buffers, int num_buffers) > { > @@ -328,6 +462,22 @@ vlVaRenderPicture(VADriverContextP ctx, VAContextID context_id, VABufferID *buff > vaStatus = vlVaHandleVAProcPipelineParameterBufferType(drv, context, buf); > break; > > + case VAEncSequenceParameterBufferType: > + vaStatus = handleVAEncSequenceParameterBufferType(drv, context, buf); > + break; > + > + case VAEncMiscParameterBufferType: > + vaStatus = handleVAEncMiscParameterBufferType(context, buf); > + break; > + > + case VAEncPictureParameterBufferType: > + vaStatus = handleVAEncPictureParameterBufferType(drv, context, buf); > + break; > + > + case VAEncSliceParameterBufferType: > + vaStatus = handleVAEncSliceParameterBufferType(drv, context, buf); > + break; > + > default: > break; > } > @@ -342,6 +492,9 @@ vlVaEndPicture(VADriverContextP ctx, VAContextID context_id) > { > vlVaDriver *drv; > vlVaContext *context; > + vlVaBuffer *coded_buf; > + unsigned int coded_size; > + void *feedback; > > if (!ctx) > return VA_STATUS_ERROR_INVALID_CONTEXT; > @@ -365,7 +518,19 @@ vlVaEndPicture(VADriverContextP ctx, VAContextID context_id) > } > > context->mpeg4.frame_num++; > - context->decoder->end_frame(context->decoder, context->target, &context->desc.base); > + > + if (context->decoder->entrypoint == PIPE_VIDEO_ENTRYPOINT_ENCODE) { > + coded_buf = context->coded_buf; > + context->decoder->begin_frame(context->decoder, context->target, &context->desc.base); > + context->decoder->encode_bitstream(context->decoder, context->target, > + coded_buf->derived_surface.resource, &feedback); > + context->decoder->end_frame(context->decoder, context->target, &context->desc.base); > + context->decoder->flush(context->decoder); > + context->decoder->get_feedback(context->decoder, feedback, &coded_size); > + coded_buf->coded_size = coded_size; > + } > + else > + context->decoder->end_frame(context->decoder, context->target, &context->desc.base); > > return VA_STATUS_SUCCESS; > } > diff --git a/src/gallium/state_trackers/va/va_private.h b/src/gallium/state_trackers/va/va_private.h > index ad9010a..6d3ac38 100644 > --- a/src/gallium/state_trackers/va/va_private.h > +++ b/src/gallium/state_trackers/va/va_private.h > @@ -229,6 +229,7 @@ typedef struct { > struct pipe_vc1_picture_desc vc1; > struct pipe_h264_picture_desc h264; > struct pipe_h265_picture_desc h265; > + struct pipe_h264_enc_picture_desc h264enc; > } desc; > > struct { > @@ -241,6 +242,7 @@ typedef struct { > } mpeg4; > > struct vl_deint_filter *deint; > + struct vlVaBuffer *coded_buf; > } vlVaContext; > > typedef struct { > @@ -260,6 +262,7 @@ typedef struct { > } derived_surface; > unsigned int export_refcount; > VABufferInfo export_state; > + unsigned int coded_size; > } vlVaBuffer; > > typedef struct { </div> </blockquote> </body> </html>