Support H264 4K on MT8192
Subsystems (3)
v4 cover Support H264 4K on MT8192
Add MT8192 H264 venc driver and support H264 4K encoding on MT8192.
Change notes:
v4: rebase on Linux 5.13-rc1
Alexandre Courbot (1):
media: mtk-vcodec: venc: remove redundant code
Irui Wang (5):
dt-bindings: media: mtk-vcodec: Add dma-ranges property
media: mtk-vcodec: Support 4GB~8GB range iova space for venc
dt-bindings: media: mtk-vcodec: Add binding for MT8192 VENC
media: mtk-vcodec: Add MT8192 H264 venc driver
media: mtk-vcodec: Support MT8192 H264 4K encoding
.../bindings/media/mediatek-vcodec.txt | 3 + .../platform/mtk-vcodec/mtk_vcodec_drv.h | 1 + .../platform/mtk-vcodec/mtk_vcodec_enc.c | 89 +++++++++++-------- .../platform/mtk-vcodec/mtk_vcodec_enc_drv.c | 17 ++++ .../platform/mtk-vcodec/venc/venc_h264_if.c | 4 + 5 files changed, 78 insertions(+), 36 deletions(-)
Signature · 1 line
2.18.0
PATCH v4 1/6 media: mtk-vcodec: venc: remove redundant code
From: Alexandre Courbot <[email protected]>
vidioc_try_fmt() does clamp height and width when called on the OUTPUT queue, so clamping them prior to calling this function is redundant. Set the queue's parameters after calling vidioc_try_fmt() so we can use the values it computed.
--- .../media/platform/mtk-vcodec/mtk_vcodec_enc.c | 16 ++++------------ 1 file changed, 4 insertions(+), 12 deletions(-)
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +4 -12
@@ -443,7 +443,6 @@ static int vidioc_venc_s_fmt_out(struct file *file, void *priv,struct mtk_q_data *q_data;int ret, i;const struct mtk_video_fmt *fmt; -struct v4l2_pix_format_mplane *pix_fmt_mp = &f->fmt.pix_mp; vq =v4l2_m2m_get_vq (ctx->m2m_ctx, f->type);if (!vq) {
@@ -468,20 +467,13 @@ static int vidioc_venc_s_fmt_out(struct file *file, void *priv, f->fmt.pix.pixelformat = fmt->fourcc; } - pix_fmt_mp->height =clamp (pix_fmt_mp->height, -MTK_VENC_MIN_H , -MTK_VENC_MAX_H ); - pix_fmt_mp->width =clamp (pix_fmt_mp->width, -MTK_VENC_MIN_W , -MTK_VENC_MAX_W ); - - q_data->visible_width = f->fmt.pix_mp.width; - q_data->visible_height = f->fmt.pix_mp.height; - q_data->fmt = fmt; - ret =vidioc_try_fmt (f, q_data->fmt); + ret =vidioc_try_fmt (f, fmt);if (ret)return ret; + q_data->fmt = fmt; + q_data->visible_width = f->fmt.pix_mp.width; + q_data->visible_height = f->fmt.pix_mp.height; q_data->coded_width = f->fmt.pix_mp.width; q_data->coded_height = f->fmt.pix_mp.height;
Signature · 1 line
2.18.0
Re: [PATCH v4,1/6] media: mtk-vcodec: venc: remove redundant code
Quoted 9 lines · Irui Wang
From: Alexandre Courbot <[email protected]>
vidioc_try_fmt()does clamp height and width when called on the OUTPUT queue, so clamping them prior to calling this function is redundant. Set the queue's parameters after callingvidioc_try_fmt()so we can use the values it computed.Signed-off-by Alexandre Courbot <[email protected]>
From the guideline[1]:
Notably, the last Signed-off-by: must always be that of the developer submitting the patch
You should provide your signed-off.
PATCH v4 2/6 dt-bindings: media: mtk-vcodec: Add dma-ranges property
In reply to Irui WangThe mt8192 iommu support 0~16GB iova. We separate it to four banks:
0~4G; 4G~8G; 8G~12G; 12G~16G.
The "dma-ranges" could be used to adjust the bank we locate. If we don't set this property. The default range always is 0~4G.
Here we don't have actual bus/parent concept here. And the iova requirement is for our HW. Thus put the property in our node.
--- Documentation/devicetree/bindings/media/mediatek-vcodec.txt | 2 ++ 1 file changed, 2 insertions(+)
Documentation/devicetree/bindings/media/mediatek-vcodec.txt +2 -0
@@ -22,6 +22,8 @@ Required properties: - iommus : should point to the respective IOMMU block with master port as argument, see Documentation/devicetree/bindings/iommu/mediatek,iommu.yaml for details. +- dma-ranges : describes how the physical address space of the IOMMU maps + to memory. One of the two following nodes: - mediatek,vpu : the node of the video processor unit, if using VPU. - mediatek,scp : the node of the SCP unit, if using SCP.
Signature · 1 line
2.18.0
Re: [PATCH v4,2/6] dt-bindings: media: mtk-vcodec: Add dma-ranges property
Quoted 6 lines · Irui Wang
The mt8192 iommu support 0~16GB iova. We separate it to four banks:
0~4G; 4G~8G; 8G~12G; 12G~16G.The "dma-ranges" could be used to adjust the bank we locate. If we don't set this property. The default range always is 0~4G.
Quoted 8 lines
Here we don't have actual bus/parent concept here. And the iova requirement is for our HW. Thus put the property in our node.
Signed-off-by Irui Wang <[email protected]>--- Documentation/devicetree/bindings/media/mediatek-vcodec.txt | 2 ++ 1 file changed, 2 insertions(+)
PATCH v4 3/6 media: mtk-vcodec: Support 4GB~8GB range iova space for venc
In reply to Irui WangUse the dma_set_mask_and_coherent helper to set venc DMA bit mask to support 4GB~8GB range iova space.
--- drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +3 -0
@@ -361,6 +361,9 @@ static int mtk_vcodec_probe(struct platform_device *pdev) goto err_event_workq; } +if (of_get_property (pdev->dev.of_node,"dma-ranges" ,NULL )) +dma_set_mask_and_coherent (&pdev->dev,DMA_BIT_MASK (34 )); + ret =video_register_device (vfd_enc,VFL_TYPE_VIDEO ,1 );if (ret) {mtk_v4l2_err ("Failed to register video device" );
Signature · 1 line
2.18.0
Re: [PATCH v4,3/6] media: mtk-vcodec: Support 4GB~8GB range iova space for venc
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static int mtk_vcodec_probe(struct platform_device *pdev) · line 366
Quoted 3 lines · Irui Wang
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +3
-0+if (of_get_property (pdev->dev.of_node,"dma-ranges" ,NULL )) +dma_set_mask_and_coherent (&pdev->dev,DMA_BIT_MASK (34 )); +
From your previous dt-bindings patch[1], it has 4 banks.
Quoted 1 line
0~4G; 4G~8G; 8G~12G; 12G~16G.
But, the code treats it as a boolean. I.e. 0~4GB if non-existent; otherwise, 4GB~8GB.
[1]: https://patchwork.linuxtv.org/project/linux-media/patch/[email protected]/
Re: [PATCH v4,3/6] media: mtk-vcodec: Support 4GB~8GB range iova space for venc
Quoted 10 lines · Tzung-Bi Shih
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static int mtk_vcodec_probe(struct platform_device *pdev) · line 366
Quoted 3 lines · Irui Wang
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +3
-0+if (of_get_property (pdev->dev.of_node,"dma-ranges" ,NULL )) +dma_set_mask_and_coherent (&pdev->dev,DMA_BIT_MASK (34 )); +From your previous dt-bindings patch[1], it has 4 banks.
Quoted 1 line
0~4G; 4G~8G; 8G~12G; 12G~16G.
But, the code treats it as a boolean. I.e. 0~4GB if non-existent; otherwise, 4GB~8GB.
Here DMA_BIT_MASK(34) means dma address support 34bits(16GB).
The code is ok for me. Only the commit message should not say 4G-8G. it could be something like:
Use the dma_set_mask_and_coherent helper to set venc DMA bit mask to support 34bits iova space(16GB) that the mt8192 iommu HW support. Whole the iova range separate to 0~4G/4G~8G/8G~12G/12G~16G, Regarding which iova range VENC actually locate, it depend on the dma-ranges property of the venc dtsi node.
Quoted 7 lines
[1]: https://patchwork.linuxtv.org/project/linux-media/patch/[email protected]/
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
PATCH v4 4/6 dt-bindings: media: mtk-vcodec: Add binding for MT8192 VENC
In reply to Irui WangUpdates binding document for mt8192 encoder driver.
--- Documentation/devicetree/bindings/media/mediatek-vcodec.txt | 1 + 1 file changed, 1 insertion(+)
Documentation/devicetree/bindings/media/mediatek-vcodec.txt +1 -0
@@ -9,6 +9,7 @@ Required properties: "mediatek,mt8173-vcodec-enc" for mt8173 avc encoder. "mediatek,mt8183-vcodec-enc" for MT8183 encoder. "mediatek,mt8173-vcodec-dec" for MT8173 decoder. + "mediatek,mt8192-vcodec-enc" for MT8192 encoder. - reg : Physical base address of the video codec registers and length of memory mapped region. - interrupts : interrupt number to the cpu.
Signature · 1 line
2.18.0
PATCH v4 5/6 media: mtk-vcodec: Add MT8192 H264 venc driver
In reply to Irui WangAdd MT8192 venc driver's compatible and device private data.
--- drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h | 1 + .../media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+)
drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h +1 -0
@@ -302,6 +302,7 @@ struct mtk_vcodec_ctx {enum mtk_chip {MTK_MT8173 ,MTK_MT8183 , +MTK_MT8192 , };/**
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +14 -0
@@ -425,12 +425,26 @@ static const struct mtk_vcodec_enc_pdata mt8183_pdata = { .core_id = VENC_SYS, }; +static const struct mtk_vcodec_enc_pdata mt8192_pdata = { + .chip =MTK_MT8192 , + .uses_ext = true, +/* MT8192 supports the same capture formats as MT8183*/ + .capture_formats = mtk_video_formats_capture_mt8183, + .num_capture_formats =ARRAY_SIZE (mtk_video_formats_capture_mt8183), +/* MT8192 supports the same output formats as MT8173 */ + .output_formats = mtk_video_formats_output_mt8173, + .num_output_formats =ARRAY_SIZE (mtk_video_formats_output_mt8173), + .min_bitrate =64 , + .max_bitrate =100000000 , + .core_id =VENC_SYS , +};static const struct of_device_id mtk_vcodec_enc_match[] = { {.compatible ="mediatek,mt8173-vcodec-enc" , .data = &mt8173_avc_pdata}, {.compatible ="mediatek,mt8173-vcodec-enc-vp8" , .data = &mt8173_vp8_pdata}, {.compatible ="mediatek,mt8183-vcodec-enc" , .data = &mt8183_pdata}, + {.compatible ="mediatek,mt8192-vcodec-enc" , .data = &mt8192_pdata}, {}, };MODULE_DEVICE_TABLE (of, mtk_vcodec_enc_match);
Signature · 1 line
2.18.0
Re: [PATCH v4,5/6] media: mtk-vcodec: Add MT8192 H264 venc driver
Quoted 4 lines · Irui Wang
Add MT8192 venc driver's compatible and device private data.
Signed-off-by Irui Wang <[email protected]>
Re: [PATCH v4,5/6] media: mtk-vcodec: Add MT8192 H264 venc driver
In reply to Irui WangHi
Quoted 25 lines · Irui Wang
Add MT8192 venc driver's compatible and device private data.
Signed-off-by Irui Wang <[email protected]>--- drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h | 1 + .../media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+)drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h +1
-0@@ -302,6 +302,7 @@ struct mtk_vcodec_ctx {enum mtk_chip {MTK_MT8173 ,MTK_MT8183 , +MTK_MT8192 , };/** drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +0
-0
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static const struct mtk_vcodec_enc_pdata mt8183_pdata = { · line 431
Quoted 7 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +4
-0@@ -425,12 +425,26 @@ static const struct mtk_vcodec_enc_pdata mt8183_pdata = { .core_id = VENC_SYS, }; +static const struct mtk_vcodec_enc_pdata mt8192_pdata = { + .chip =MTK_MT8192 , + .uses_ext = true, +/* MT8192 supports the same capture formats as MT8183*/
missing space before '*/'
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static const struct mtk_vcodec_enc_pdata mt8183_pdata = { · line 436
Quoted 5 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +5
-0+ .capture_formats = mtk_video_formats_capture_mt8183, + .num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_mt8183), +/* MT8192 supports the same output formats as MT8173 */ + .output_formats = mtk_video_formats_output_mt8173, + .num_output_formats = ARRAY_SIZE(mtk_video_formats_output_mt8173),
If all three models mt8173, mt8183 and mt8192 support the same formats then we should probably not name the structs according to one of the models, so maybe:
s/mtk_video_formats_capture_mt8183/mtk_video_formats_capture/ s/mtk_video_formats_output_mt8173/mtk_video_formats_output/
or some other naming
Thanks,
Dafna
Quoted 15 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +5
-0+ .min_bitrate = 64, + .max_bitrate = 100000000, + .core_id = VENC_SYS, +};static const struct of_device_id mtk_vcodec_enc_match[] = { {.compatible ="mediatek,mt8173-vcodec-enc" , .data = &mt8173_avc_pdata}, {.compatible ="mediatek,mt8173-vcodec-enc-vp8" , .data = &mt8173_vp8_pdata}, {.compatible ="mediatek,mt8183-vcodec-enc" , .data = &mt8183_pdata}, + {.compatible ="mediatek,mt8192-vcodec-enc" , .data = &mt8192_pdata}, {}, };MODULE_DEVICE_TABLE (of, mtk_vcodec_enc_match);
Re: [PATCH v4,5/6] media: mtk-vcodec: Add MT8192 H264 venc driver
Hi,
Quoted 47 lines · Dafna Hirschfeld
Hi
Quoted 25 lines · Irui Wang
Add MT8192 venc driver's compatible and device private data.
Signed-off-by Irui Wang <[email protected]>--- drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h | 1 + .../media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+)drivers/media/platform/mtk-vcodec/mtk_vcodec_drv.h +1
-0@@ -302,6 +302,7 @@ struct mtk_vcodec_ctx {enum mtk_chip {MTK_MT8173 ,MTK_MT8183 , +MTK_MT8192 , };/** drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +0
-0drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static const struct mtk_vcodec_enc_pdata mt8183_pdata = { · line 431
Quoted 7 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +4
-0@@ -425,12 +425,26 @@ static const struct mtk_vcodec_enc_pdata mt8183_pdata = { .core_id = VENC_SYS, }; +static const struct mtk_vcodec_enc_pdata mt8192_pdata = { + .chip =MTK_MT8192 , + .uses_ext = true, +/* MT8192 supports the same capture formats as MT8183*/ missing space before '*/'
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c static const struct mtk_vcodec_enc_pdata mt8183_pdata = { · line 436
Quoted 5 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +5
-0+ .capture_formats = mtk_video_formats_capture_mt8183, + .num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_mt8183), +/* MT8192 supports the same output formats as MT8173 */ + .output_formats = mtk_video_formats_output_mt8173, + .num_output_formats = ARRAY_SIZE(mtk_video_formats_output_mt8173),If all three models mt8173, mt8183 and mt8192 support the same formats then we should probably not name the structs according to one of the models, so maybe:
Quoted 7 lines
s/mtk_video_formats_capture_mt8183/mtk_video_formats_capture/ s/mtk_video_formats_output_mt8173/mtk_video_formats_output/
or some other naming
Thanks,
Dafna
Thanks for your review and comments, we will rename these in another patchset, this patchset just for MT8192 now.
Quoted 17 lines
Quoted 17 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c +5
-0+ .min_bitrate = 64, + .max_bitrate = 100000000, + .core_id = VENC_SYS, +};static const struct of_device_id mtk_vcodec_enc_match[] = { {.compatible ="mediatek,mt8173-vcodec-enc" , .data = &mt8173_avc_pdata}, {.compatible ="mediatek,mt8173-vcodec-enc-vp8" , .data = &mt8173_vp8_pdata}, {.compatible ="mediatek,mt8183-vcodec-enc" , .data = &mt8183_pdata}, + {.compatible ="mediatek,mt8192-vcodec-enc" , .data = &mt8192_pdata}, {}, };MODULE_DEVICE_TABLE (of, mtk_vcodec_enc_match);
PATCH v4 6/6 media: mtk-vcodec: Support MT8192 H264 4K encoding
In reply to Irui WangMT8192 H264 support 4k(3840x2176) and Level 5.1 encoding, add related path according to enc_capability.
--- .../platform/mtk-vcodec/mtk_vcodec_enc.c | 75 ++++++++++++------- .../platform/mtk-vcodec/venc/venc_h264_if.c | 4 + 2 files changed, 54 insertions(+), 25 deletions(-)
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +50 -25
@@ -19,23 +19,32 @@#define MTK_VENC_MIN_W 160U#define MTK_VENC_MIN_H 128U -#define MTK_VENC_MAX_W 1920U -#define MTK_VENC_MAX_H 1088U +#define MTK_VENC_HD_MAX_W 1920U +#define MTK_VENC_HD_MAX_H 1088U +#define MTK_VENC_4K_MAX_W 3840U +#define MTK_VENC_4K_MAX_H 2176U +#define DFT_CFG_WIDTH MTK_VENC_MIN_W#define DFT_CFG_HEIGHT MTK_VENC_MIN_H#define MTK_MAX_CTRLS_HINT 20#define MTK_DEFAULT_FRAMERATE_NUM 1001#define MTK_DEFAULT_FRAMERATE_DENOM 30000 +#define MTK_VENC_4K_CAPABILITY_ENABLE BIT(0)static void mtk_venc_worker (struct work_struct *work); -static const struct v4l2_frmsize_stepwise mtk_venc_framesizes = { -MTK_VENC_MIN_W ,MTK_VENC_MAX_W ,16 , -MTK_VENC_MIN_H ,MTK_VENC_MAX_H ,16 , +static const struct v4l2_frmsize_stepwise mtk_venc_hd_framesizes = { +MTK_VENC_MIN_W ,MTK_VENC_HD_MAX_W ,16 , +MTK_VENC_MIN_H ,MTK_VENC_HD_MAX_H ,16 , +}; + +static const struct v4l2_frmsize_stepwise mtk_venc_4k_framesizes = { +MTK_VENC_MIN_W ,MTK_VENC_4K_MAX_W ,16 , +MTK_VENC_MIN_H ,MTK_VENC_4K_MAX_H ,16 , }; -#define NUM_SUPPORTED_FRAMESIZE ARRAY_SIZE(mtk_venc_framesizes) +#define NUM_SUPPORTED_FRAMESIZE ARRAY_SIZE(mtk_venc_hd_framesizes)static int vidioc_venc_s_ctrl (struct v4l2_ctrl *ctrl) {
@@ -151,17 +160,20 @@ static int vidioc_enum_framesizes(struct file *file, void *fh, struct v4l2_frmsizeenum *fsize) {const struct mtk_video_fmt *fmt; +struct mtk_vcodec_ctx *ctx =fh_to_ctx (fh);if (fsize->index !=0 )return -EINVAL ; fmt =mtk_venc_find_format (fsize->pixel_format, -fh_to_ctx (fh)->dev->venc_pdata); + ctx->dev->venc_pdata);if (!fmt)return -EINVAL ; fsize->type =V4L2_FRMSIZE_TYPE_STEPWISE ; - fsize->stepwise = mtk_venc_framesizes; + fsize->stepwise = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? + mtk_venc_4k_framesizes : mtk_venc_hd_framesizes;return 0 ; }
@@ -248,7 +260,7 @@ static struct mtk_q_data *mtk_venc_get_q_data(struct mtk_vcodec_ctx *ctx,/* V4L2 specification suggests the driver corrects the format struct if any of * the dimensions is unsupported */ -static int vidioc_try_fmt (struct v4l2_format *f, +static int vidioc_try_fmt (struct mtk_vcodec_ctx *ctx,struct v4l2_format *f,const struct mtk_video_fmt *fmt) {struct v4l2_pix_format_mplane *pix_fmt_mp = &f->fmt.pix_mp;
@@ -260,13 +272,22 @@ static int vidioc_try_fmt(struct v4l2_format *f, pix_fmt_mp->plane_fmt[0 ].bytesperline =0 ; }else if (f->type ==V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE ) {int tmp_w, tmp_h; +unsigned int max_width, max_height; + +if (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) { + max_width =MTK_VENC_4K_MAX_W ; + max_height =MTK_VENC_4K_MAX_H ; + }else { + max_width =MTK_VENC_HD_MAX_W ; + max_height =MTK_VENC_HD_MAX_H ; + } pix_fmt_mp->height =clamp (pix_fmt_mp->height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ); + max_height); pix_fmt_mp->width =clamp (pix_fmt_mp->width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ); + max_width);/* find next closer width align 16, heign align 32, size align * 64 rectangle
@@ -275,16 +296,16 @@ static int vidioc_try_fmt(struct v4l2_format *f, tmp_h = pix_fmt_mp->height;v4l_bound_align_image (&pix_fmt_mp->width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ,4 , + max_width,4 , &pix_fmt_mp->height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ,5 ,6 ); + max_height,5 ,6 );if (pix_fmt_mp->width < tmp_w && - (pix_fmt_mp->width +16 ) <=MTK_VENC_MAX_W ) + (pix_fmt_mp->width +16 ) <= max_width) pix_fmt_mp->width +=16 ;if (pix_fmt_mp->height < tmp_h && - (pix_fmt_mp->height +32 ) <=MTK_VENC_MAX_H ) + (pix_fmt_mp->height +32 ) <= max_height) pix_fmt_mp->height +=32 ; mtk_v4l2_debug(0,
@@ -405,7 +426,7 @@ static int vidioc_venc_s_fmt_cap(struct file *file, void *priv, } q_data->fmt = fmt; - ret =vidioc_try_fmt (f, q_data->fmt); + ret =vidioc_try_fmt (ctx, f, q_data->fmt);if (ret)return ret;
@@ -467,7 +488,7 @@ static int vidioc_venc_s_fmt_out(struct file *file, void *priv, f->fmt.pix.pixelformat = fmt->fourcc; } - ret =vidioc_try_fmt (f, fmt); + ret =vidioc_try_fmt (ctx, f, fmt);if (ret)return ret;
@@ -545,7 +566,7 @@ static int vidioc_try_fmt_vid_cap_mplane(struct file *file, void *priv, f->fmt.pix_mp.quantization = ctx->quantization; f->fmt.pix_mp.xfer_func = ctx->xfer_func; -return vidioc_try_fmt (f, fmt); +return vidioc_try_fmt (ctx, f, fmt); } static int vidioc_try_fmt_vid_out_mplane(struct file *file, void *priv,
@@ -567,7 +588,7 @@ static int vidioc_try_fmt_vid_out_mplane(struct file *file, void *priv, f->fmt.pix_mp.xfer_func =V4L2_XFER_FUNC_DEFAULT ; } -return vidioc_try_fmt (f, fmt); +return vidioc_try_fmt (ctx, f, fmt); } static int vidioc_venc_g_selection(struct file *file, void *priv,
@@ -1171,16 +1192,16 @@ void mtk_vcodec_enc_set_default_params(struct mtk_vcodec_ctx *ctx)v4l_bound_align_image (&q_data->coded_width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ,4 , +MTK_VENC_HD_MAX_W ,4 , &q_data->coded_height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ,5 ,6 ); +MTK_VENC_HD_MAX_H ,5 ,6 );if (q_data->coded_width <DFT_CFG_WIDTH && - (q_data->coded_width +16 ) <=MTK_VENC_MAX_W ) + (q_data->coded_width +16 ) <=MTK_VENC_HD_MAX_W ) q_data->coded_width +=16 ;if (q_data->coded_height <DFT_CFG_HEIGHT && - (q_data->coded_height +32 ) <=MTK_VENC_MAX_H ) + (q_data->coded_height +32 ) <=MTK_VENC_HD_MAX_H ) q_data->coded_height +=32 ; q_data->sizeimage[0] =
@@ -1210,6 +1231,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) {const struct v4l2_ctrl_ops *ops = &mtk_vcodec_enc_ctrl_ops;struct v4l2_ctrl_handler *handler = &ctx->ctrl_hdl; +const u8 h264_max_level = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? +V4L2_MPEG_VIDEO_H264_LEVEL_5_1 :V4L2_MPEG_VIDEO_H264_LEVEL_4_2 ;v4l2_ctrl_handler_init (handler,MTK_MAX_CTRLS_HINT );
@@ -1240,8 +1264,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx)V4L2_MPEG_VIDEO_H264_PROFILE_HIGH , 0, V4L2_MPEG_VIDEO_H264_PROFILE_HIGH);v4l2_ctrl_new_std_menu (handler, ops,V4L2_CID_MPEG_VIDEO_H264_LEVEL , -V4L2_MPEG_VIDEO_H264_LEVEL_4_2 , -0 ,V4L2_MPEG_VIDEO_H264_LEVEL_4_0 ); + h264_max_level, +0 ,V4L2_MPEG_VIDEO_H264_LEVEL_4_0 ); +if (handler->error) {mtk_v4l2_err ("Init control handler fail %d" , handler->error);
drivers/media/platform/mtk-vcodec/venc/venc_h264_if.c +4 -0
@@ -215,6 +215,10 @@ static unsigned int h264_get_level(struct venc_h264_inst *inst,return 41 ;case V4L2_MPEG_VIDEO_H264_LEVEL_4_2 :return 42 ; +case V4L2_MPEG_VIDEO_H264_LEVEL_5_0 : +return 50 ; +case V4L2_MPEG_VIDEO_H264_LEVEL_5_1 : +return 51 ;default :mtk_vcodec_debug (inst,"unsupported level %d" , level);return 31 ;
Signature · 1 line
2.18.0
Re: [PATCH v4,6/6] media: mtk-vcodec: Support MT8192 H264 4K encoding
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c static int vidioc_enum_framesizes(struct file *file, void *fh, · line 176
Quoted 5 lines · Irui Wang
fsize->type = V4L2_FRMSIZE_TYPE_STEPWISE;drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +3
-1- fsize->stepwise = mtk_venc_framesizes; + fsize->stepwise = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? + mtk_venc_4k_framesizes : mtk_venc_hd_framesizes;
Use a normal if .. else .. is more readable.
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) · line 1236
Quoted 7 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +3
-0@@ -1210,6 +1231,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) {const struct v4l2_ctrl_ops *ops = &mtk_vcodec_enc_ctrl_ops;struct v4l2_ctrl_handler *handler = &ctx->ctrl_hdl; +const u8 h264_max_level = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? +V4L2_MPEG_VIDEO_H264_LEVEL_5_1 :V4L2_MPEG_VIDEO_H264_LEVEL_4_2 ;
Again, I would like it to be a normal if .. else ..
Re: [PATCH v4,6/6] media: mtk-vcodec: Support MT8192 H264 4K encoding
Quoted 8 lines · Tzung-Bi Shih
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c static int vidioc_enum_framesizes(struct file *file, void *fh, · line 176
Quoted 5 lines · Irui Wang
fsize->type = V4L2_FRMSIZE_TYPE_STEPWISE;drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +3
-1- fsize->stepwise = mtk_venc_framesizes; + fsize->stepwise = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? + mtk_venc_4k_framesizes : mtk_venc_hd_framesizes;Use a normal if .. else .. is more readable.
if (ctx->dev->enc_capability & MTK_VENC_4K_CAPABILITY_ENABLE)
fsize->stepwise = mtk_venc_4k_framesizes; else fsize->stepwise = mtk_venc_hd_framesizes;
Quoted 9 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) · line 1236
Quoted 8 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +3
-0@@ -1210,6 +1231,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) {const struct v4l2_ctrl_ops *ops = &mtk_vcodec_enc_ctrl_ops;struct v4l2_ctrl_handler *handler = &ctx->ctrl_hdl; +const u8 h264_max_level = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? +V4L2_MPEG_VIDEO_H264_LEVEL_5_1 :V4L2_MPEG_VIDEO_H264_LEVEL_4_2 ;Again, I would like it to be a normal if .. else ..
u8 h264_max_level;
if (ctx->dev->enc_capability & MTK_VENC_4K_CAPABILITY_ENABLE)
h264_max_level = V4L2_MPEG_VIDEO_H264_LEVEL_5_1; else h264_max_level = V4L2_MPEG_VIDEO_H264_LEVEL_4_2;
Patch V5 will be sent with these if there is no more comments.
Re: [PATCH v4,6/6] media: mtk-vcodec: Support MT8192 H264 4K encoding
In reply to Irui WangHi,
Quoted 39 lines · Irui Wang
MT8192 H264 support 4k(3840x2176) and Level 5.1 encoding, add related path according to
enc_capability.Signed-off-by Irui Wang <[email protected]>--- .../platform/mtk-vcodec/mtk_vcodec_enc.c | 75 ++++++++++++------- .../platform/mtk-vcodec/venc/venc_h264_if.c | 4 + 2 files changed, 54 insertions(+), 25 deletions(-)drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +8
-5@@ -19,23 +19,32 @@#define MTK_VENC_MIN_W 160U#define MTK_VENC_MIN_H 128U -#define MTK_VENC_MAX_W 1920U -#define MTK_VENC_MAX_H 1088U +#define MTK_VENC_HD_MAX_W 1920U +#define MTK_VENC_HD_MAX_H 1088U +#define MTK_VENC_4K_MAX_W 3840U +#define MTK_VENC_4K_MAX_H 2176U +#define DFT_CFG_WIDTH MTK_VENC_MIN_W#define DFT_CFG_HEIGHT MTK_VENC_MIN_H#define MTK_MAX_CTRLS_HINT 20#define MTK_DEFAULT_FRAMERATE_NUM 1001#define MTK_DEFAULT_FRAMERATE_DENOM 30000 +#define MTK_VENC_4K_CAPABILITY_ENABLE BIT(0)static void mtk_venc_worker (struct work_struct *work); -static const struct v4l2_frmsize_stepwise mtk_venc_framesizes = { -MTK_VENC_MIN_W ,MTK_VENC_MAX_W ,16 , -MTK_VENC_MIN_H ,MTK_VENC_MAX_H ,16 , +static const struct v4l2_frmsize_stepwise mtk_venc_hd_framesizes = { +MTK_VENC_MIN_W ,MTK_VENC_HD_MAX_W ,16 ,
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c line 47
Quoted 10 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +7
-1+ MTK_VENC_MIN_H, MTK_VENC_HD_MAX_H, 16, +}; + +static const struct v4l2_frmsize_stepwise mtk_venc_4k_framesizes = { +MTK_VENC_MIN_W ,MTK_VENC_4K_MAX_W ,16 , +MTK_VENC_MIN_H ,MTK_VENC_4K_MAX_H ,16 , }; -#define NUM_SUPPORTED_FRAMESIZE ARRAY_SIZE(mtk_venc_framesizes) +#define NUM_SUPPORTED_FRAMESIZE ARRAY_SIZE(mtk_venc_hd_framesizes)
This define is never used so it can be removed.
Thanks,
Dafna
Quoted 166 lines
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +35
-19static int vidioc_venc_s_ctrl (struct v4l2_ctrl *ctrl) {@@ -151,17 +160,20 @@ static int vidioc_enum_framesizes(struct file *file, void *fh, struct v4l2_frmsizeenum *fsize) {const struct mtk_video_fmt *fmt; +struct mtk_vcodec_ctx *ctx =fh_to_ctx (fh);if (fsize->index !=0 )return -EINVAL ; fmt =mtk_venc_find_format (fsize->pixel_format, -fh_to_ctx (fh)->dev->venc_pdata); + ctx->dev->venc_pdata);if (!fmt)return -EINVAL ; fsize->type =V4L2_FRMSIZE_TYPE_STEPWISE ; - fsize->stepwise = mtk_venc_framesizes; + fsize->stepwise = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? + mtk_venc_4k_framesizes : mtk_venc_hd_framesizes;return 0 ; }@@ -248,7 +260,7 @@ static struct mtk_q_data *mtk_venc_get_q_data(struct mtk_vcodec_ctx *ctx,/* V4L2 specification suggests the driver corrects the format struct if any of * the dimensions is unsupported */ -static int vidioc_try_fmt (struct v4l2_format *f, +static int vidioc_try_fmt (struct mtk_vcodec_ctx *ctx,struct v4l2_format *f,const struct mtk_video_fmt *fmt) {struct v4l2_pix_format_mplane *pix_fmt_mp = &f->fmt.pix_mp;@@ -260,13 +272,22 @@ static int vidioc_try_fmt(struct v4l2_format *f, pix_fmt_mp->plane_fmt[0 ].bytesperline =0 ; }else if (f->type ==V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE ) {int tmp_w, tmp_h; +unsigned int max_width, max_height; + +if (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) { + max_width =MTK_VENC_4K_MAX_W ; + max_height =MTK_VENC_4K_MAX_H ; + }else { + max_width =MTK_VENC_HD_MAX_W ; + max_height =MTK_VENC_HD_MAX_H ; + } pix_fmt_mp->height =clamp (pix_fmt_mp->height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ); + max_height); pix_fmt_mp->width =clamp (pix_fmt_mp->width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ); + max_width);/* find next closer width align 16, heign align 32, size align * 64 rectangle @@ -275,16 +296,16 @@ static int vidioc_try_fmt(struct v4l2_format *f, tmp_h = pix_fmt_mp->height;v4l_bound_align_image (&pix_fmt_mp->width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ,4 , + max_width,4 , &pix_fmt_mp->height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ,5 ,6 ); + max_height,5 ,6 );if (pix_fmt_mp->width < tmp_w && - (pix_fmt_mp->width +16 ) <=MTK_VENC_MAX_W ) + (pix_fmt_mp->width +16 ) <= max_width) pix_fmt_mp->width +=16 ;if (pix_fmt_mp->height < tmp_h && - (pix_fmt_mp->height +32 ) <=MTK_VENC_MAX_H ) + (pix_fmt_mp->height +32 ) <= max_height) pix_fmt_mp->height +=32 ; mtk_v4l2_debug(0,@@ -405,7 +426,7 @@ static int vidioc_venc_s_fmt_cap(struct file *file, void *priv, } q_data->fmt = fmt; - ret =vidioc_try_fmt (f, q_data->fmt); + ret =vidioc_try_fmt (ctx, f, q_data->fmt);if (ret)return ret;@@ -467,7 +488,7 @@ static int vidioc_venc_s_fmt_out(struct file *file, void *priv, f->fmt.pix.pixelformat = fmt->fourcc; } - ret =vidioc_try_fmt (f, fmt); + ret =vidioc_try_fmt (ctx, f, fmt);if (ret)return ret;@@ -545,7 +566,7 @@ static int vidioc_try_fmt_vid_cap_mplane(struct file *file, void *priv, f->fmt.pix_mp.quantization = ctx->quantization; f->fmt.pix_mp.xfer_func = ctx->xfer_func; -return vidioc_try_fmt (f, fmt); +return vidioc_try_fmt (ctx, f, fmt); } static int vidioc_try_fmt_vid_out_mplane(struct file *file, void *priv,@@ -567,7 +588,7 @@ static int vidioc_try_fmt_vid_out_mplane(struct file *file, void *priv, f->fmt.pix_mp.xfer_func =V4L2_XFER_FUNC_DEFAULT ; } -return vidioc_try_fmt (f, fmt); +return vidioc_try_fmt (ctx, f, fmt); } static int vidioc_venc_g_selection(struct file *file, void *priv,@@ -1171,16 +1192,16 @@ void mtk_vcodec_enc_set_default_params(struct mtk_vcodec_ctx *ctx)v4l_bound_align_image (&q_data->coded_width,MTK_VENC_MIN_W , -MTK_VENC_MAX_W ,4 , +MTK_VENC_HD_MAX_W ,4 , &q_data->coded_height,MTK_VENC_MIN_H , -MTK_VENC_MAX_H ,5 ,6 ); +MTK_VENC_HD_MAX_H ,5 ,6 );if (q_data->coded_width <DFT_CFG_WIDTH && - (q_data->coded_width +16 ) <=MTK_VENC_MAX_W ) + (q_data->coded_width +16 ) <=MTK_VENC_HD_MAX_W ) q_data->coded_width +=16 ;if (q_data->coded_height <DFT_CFG_HEIGHT && - (q_data->coded_height +32 ) <=MTK_VENC_MAX_H ) + (q_data->coded_height +32 ) <=MTK_VENC_HD_MAX_H ) q_data->coded_height +=32 ; q_data->sizeimage[0] =@@ -1210,6 +1231,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx) {const struct v4l2_ctrl_ops *ops = &mtk_vcodec_enc_ctrl_ops;struct v4l2_ctrl_handler *handler = &ctx->ctrl_hdl; +const u8 h264_max_level = + (ctx->dev->enc_capability &MTK_VENC_4K_CAPABILITY_ENABLE ) ? +V4L2_MPEG_VIDEO_H264_LEVEL_5_1 :V4L2_MPEG_VIDEO_H264_LEVEL_4_2 ;v4l2_ctrl_handler_init (handler,MTK_MAX_CTRLS_HINT );@@ -1240,8 +1264,9 @@ int mtk_vcodec_enc_ctrls_setup(struct mtk_vcodec_ctx *ctx)V4L2_MPEG_VIDEO_H264_PROFILE_HIGH , 0, V4L2_MPEG_VIDEO_H264_PROFILE_HIGH);v4l2_ctrl_new_std_menu (handler, ops,V4L2_CID_MPEG_VIDEO_H264_LEVEL , -V4L2_MPEG_VIDEO_H264_LEVEL_4_2 , -0 ,V4L2_MPEG_VIDEO_H264_LEVEL_4_0 ); + h264_max_level, +0 ,V4L2_MPEG_VIDEO_H264_LEVEL_4_0 ); +if (handler->error) {mtk_v4l2_err ("Init control handler fail %d" , handler->error);drivers/media/platform/mtk-vcodec/venc/venc_h264_if.c +0
-0
drivers/media/platform/mtk-vcodec/venc/venc_h264_if.c static unsigned int h264_get_level(struct venc_h264_inst *inst, · line 224
Quoted 10 lines
drivers/media/platform/mtk-vcodec/venc/venc_h264_if.c +4
-0@@ -215,6 +215,10 @@ static unsigned int h264_get_level(struct venc_h264_inst *inst,return 41 ;case V4L2_MPEG_VIDEO_H264_LEVEL_4_2 :return 42 ; +case V4L2_MPEG_VIDEO_H264_LEVEL_5_0 : +return 50 ; +case V4L2_MPEG_VIDEO_H264_LEVEL_5_1 : +return 51 ;default :mtk_vcodec_debug (inst,"unsupported level %d" , level);return 31 ;
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek