* [FFmpeg-devel] [PATCH 1/2] lavu/tx: add missing prints for the type of dctI/dstI
@ 2023-09-06 4:19 Lynne
[not found] ` <Ndchl_G--3-9@lynne.ee-NdchpZh----9>
0 siblings, 1 reply; 3+ messages in thread
From: Lynne @ 2023-09-06 4:19 UTC (permalink / raw)
To: Ffmpeg Devel
[-- Attachment #1: Type: text/plain, Size: 81 bytes --]
Forgot to add the new types to the print, currently
they fallback to "unknown".
[-- Attachment #2: 0001-lavu-tx-add-missing-prints-for-the-type-of-dctI-dstI.patch --]
[-- Type: text/x-diff, Size: 1423 bytes --]
From 439f1e881313ef63ecb8e8e264f4847d6612619d Mon Sep 17 00:00:00 2001
From: Lynne <dev@lynne.ee>
Date: Sun, 3 Sep 2023 16:47:53 +0200
Subject: [PATCH 1/2] lavu/tx: add missing prints for the type of dctI/dstI
---
libavutil/tx.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/libavutil/tx.c b/libavutil/tx.c
index e9826e6107..24b2015b44 100644
--- a/libavutil/tx.c
+++ b/libavutil/tx.c
@@ -578,12 +578,18 @@ static void print_type(AVBPrint *bp, enum AVTXType type)
type == AV_TX_FLOAT_FFT ? "fft_float" :
type == AV_TX_FLOAT_MDCT ? "mdct_float" :
type == AV_TX_FLOAT_RDFT ? "rdft_float" :
+ type == AV_TX_FLOAT_DCT_I ? "dctI_float" :
+ type == AV_TX_FLOAT_DST_I ? "dstI_float" :
type == AV_TX_DOUBLE_FFT ? "fft_double" :
type == AV_TX_DOUBLE_MDCT ? "mdct_double" :
type == AV_TX_DOUBLE_RDFT ? "rdft_double" :
+ type == AV_TX_DOUBLE_DCT_I ? "dctI_double" :
+ type == AV_TX_DOUBLE_DST_I ? "dstI_double" :
type == AV_TX_INT32_FFT ? "fft_int32" :
type == AV_TX_INT32_MDCT ? "mdct_int32" :
type == AV_TX_INT32_RDFT ? "rdft_int32" :
+ type == AV_TX_INT32_DCT_I ? "dctI_int32" :
+ type == AV_TX_INT32_DST_I ? "dstI_int32" :
"unknown");
}
--
2.40.1
[-- Attachment #3: Type: text/plain, Size: 251 bytes --]
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 3+ messages in thread
* [FFmpeg-devel] [PATCH 2/2] vulkan_decode: convert max level from vulkan to av for comparisons
[not found] ` <Ndchl_G--3-9@lynne.ee-NdchpZh----9>
@ 2023-09-06 4:21 ` Lynne
[not found] ` <NdciLba--3-9@lynne.ee-NdciOcd----9>
1 sibling, 0 replies; 3+ messages in thread
From: Lynne @ 2023-09-06 4:21 UTC (permalink / raw)
To: FFmpeg development discussions and patches
[-- Attachment #1: Type: text/plain, Size: 558 bytes --]
The spec finally clarified the meaning of the field to:
> maxLevelIdc is a StdVideoH264LevelIdc value specifying the maximum H.264 level supported by the profile, where enum constant STD_VIDEO_H264_LEVEL_IDC_<major>_<minor> identifies H.264 level <major>.<minor> as defined in section A.3 of the ITU-T H.264 Specification.1
Annoyingly, we have to convert it to an IDC value to compare it.
Mesa commit:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24649
Only mesa currently uses the IDC value, all other drivers use the enum.
Patch attached.
[-- Attachment #2: 0002-vulkan_decode-convert-max-level-from-vulkan-to-av-fo.patch --]
[-- Type: text/x-diff, Size: 4021 bytes --]
From eea4b14917083ad1a8881e2793bdc045869d5b15 Mon Sep 17 00:00:00 2001
From: Lynne <dev@lynne.ee>
Date: Wed, 6 Sep 2023 06:15:32 +0200
Subject: [PATCH 2/2] vulkan_decode: convert max level from vulkan to av for
comparisons
---
libavcodec/vulkan_decode.c | 4 ++--
libavcodec/vulkan_video.c | 45 ++++++++++++++++++++++++++++++++++++++
libavcodec/vulkan_video.h | 7 ++++++
3 files changed, 54 insertions(+), 2 deletions(-)
diff --git a/libavcodec/vulkan_decode.c b/libavcodec/vulkan_decode.c
index 1b2ff0045d..d97abe8896 100644
--- a/libavcodec/vulkan_decode.c
+++ b/libavcodec/vulkan_decode.c
@@ -837,8 +837,8 @@ static int vulkan_decode_get_profile(AVCodecContext *avctx, AVBufferRef *frames_
return AVERROR_EXTERNAL;
}
- max_level = avctx->codec_id == AV_CODEC_ID_H264 ? h264_caps.maxLevelIdc :
- avctx->codec_id == AV_CODEC_ID_H265 ? h265_caps.maxLevelIdc :
+ max_level = avctx->codec_id == AV_CODEC_ID_H264 ? ff_vk_h264_level_to_av(h264_caps.maxLevelIdc) :
+ avctx->codec_id == AV_CODEC_ID_H265 ? ff_vk_h265_level_to_av(h265_caps.maxLevelIdc) :
avctx->codec_id == AV_CODEC_ID_AV1 ? av1_caps.maxLevelIdc :
0;
diff --git a/libavcodec/vulkan_video.c b/libavcodec/vulkan_video.c
index 9a363aab02..236aa124bb 100644
--- a/libavcodec/vulkan_video.c
+++ b/libavcodec/vulkan_video.c
@@ -154,6 +154,51 @@ VkVideoComponentBitDepthFlagBitsKHR ff_vk_depth_from_av_depth(int depth)
return VK_VIDEO_COMPONENT_BIT_DEPTH_INVALID_KHR;
}
+int ff_vk_h264_level_to_av(StdVideoH264LevelIdc level)
+{
+ switch (level) {
+ case STD_VIDEO_H264_LEVEL_IDC_1_0: return 10;
+ case STD_VIDEO_H264_LEVEL_IDC_1_1: return 11;
+ case STD_VIDEO_H264_LEVEL_IDC_1_2: return 12;
+ case STD_VIDEO_H264_LEVEL_IDC_1_3: return 13;
+ case STD_VIDEO_H264_LEVEL_IDC_2_0: return 20;
+ case STD_VIDEO_H264_LEVEL_IDC_2_1: return 21;
+ case STD_VIDEO_H264_LEVEL_IDC_2_2: return 22;
+ case STD_VIDEO_H264_LEVEL_IDC_3_0: return 30;
+ case STD_VIDEO_H264_LEVEL_IDC_3_1: return 31;
+ case STD_VIDEO_H264_LEVEL_IDC_3_2: return 32;
+ case STD_VIDEO_H264_LEVEL_IDC_4_0: return 40;
+ case STD_VIDEO_H264_LEVEL_IDC_4_1: return 41;
+ case STD_VIDEO_H264_LEVEL_IDC_4_2: return 42;
+ case STD_VIDEO_H264_LEVEL_IDC_5_0: return 50;
+ case STD_VIDEO_H264_LEVEL_IDC_5_1: return 51;
+ case STD_VIDEO_H264_LEVEL_IDC_5_2: return 52;
+ case STD_VIDEO_H264_LEVEL_IDC_6_0: return 60;
+ case STD_VIDEO_H264_LEVEL_IDC_6_1: return 61;
+ default:
+ case STD_VIDEO_H264_LEVEL_IDC_6_2: return 62;
+ }
+}
+
+int ff_vk_h265_level_to_av(StdVideoH265LevelIdc level)
+{
+ switch (level) {
+ case STD_VIDEO_H265_LEVEL_IDC_1_0: return 10;
+ case STD_VIDEO_H265_LEVEL_IDC_2_0: return 20;
+ case STD_VIDEO_H265_LEVEL_IDC_2_1: return 21;
+ case STD_VIDEO_H265_LEVEL_IDC_3_0: return 30;
+ case STD_VIDEO_H265_LEVEL_IDC_3_1: return 31;
+ case STD_VIDEO_H265_LEVEL_IDC_4_0: return 40;
+ case STD_VIDEO_H265_LEVEL_IDC_4_1: return 41;
+ case STD_VIDEO_H265_LEVEL_IDC_5_0: return 50;
+ case STD_VIDEO_H265_LEVEL_IDC_5_1: return 51;
+ case STD_VIDEO_H265_LEVEL_IDC_6_0: return 60;
+ case STD_VIDEO_H265_LEVEL_IDC_6_1: return 61;
+ default:
+ case STD_VIDEO_H265_LEVEL_IDC_6_2: return 62;
+ }
+}
+
static void free_data_buf(void *opaque, uint8_t *data)
{
FFVulkanContext *ctx = opaque;
diff --git a/libavcodec/vulkan_video.h b/libavcodec/vulkan_video.h
index 183ce89bf0..b28e3fe0bd 100644
--- a/libavcodec/vulkan_video.h
+++ b/libavcodec/vulkan_video.h
@@ -71,6 +71,13 @@ VkVideoChromaSubsamplingFlagBitsKHR ff_vk_subsampling_from_av_desc(const AVPixFm
*/
VkVideoComponentBitDepthFlagBitsKHR ff_vk_depth_from_av_depth(int depth);
+
+/**
+ * Convert level from Vulkan to AV.
+ */
+int ff_vk_h264_level_to_av(StdVideoH264LevelIdc level);
+int ff_vk_h265_level_to_av(StdVideoH265LevelIdc level);
+
typedef struct FFVkVideoBuffer {
FFVkBuffer buf;
uint8_t *mem;
--
2.40.1
[-- Attachment #3: Type: text/plain, Size: 251 bytes --]
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [FFmpeg-devel] [PATCH 2/2] vulkan_decode: convert max level from vulkan to av for comparisons
[not found] ` <NdciLba--3-9@lynne.ee-NdciOcd----9>
@ 2023-09-08 4:59 ` Lynne
0 siblings, 0 replies; 3+ messages in thread
From: Lynne @ 2023-09-08 4:59 UTC (permalink / raw)
To: FFmpeg development discussions and patches
Sep 6, 2023, 06:21 by dev@lynne.ee:
> The spec finally clarified the meaning of the field to:
>
>> maxLevelIdc is a StdVideoH264LevelIdc value specifying the maximum H.264 level supported by the profile, where enum constant STD_VIDEO_H264_LEVEL_IDC_<major>_<minor> identifies H.264 level <major>.<minor> as defined in section A.3 of the ITU-T H.264 Specification.1
>>
>
> Annoyingly, we have to convert it to an IDC value to compare it.
> Mesa commit:
> https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24649
> Only mesa currently uses the IDC value, all other drivers use the enum.
>
> Patch attached.
>
Pushed.
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-09-08 4:59 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-09-06 4:19 [FFmpeg-devel] [PATCH 1/2] lavu/tx: add missing prints for the type of dctI/dstI Lynne
[not found] ` <Ndchl_G--3-9@lynne.ee-NdchpZh----9>
2023-09-06 4:21 ` [FFmpeg-devel] [PATCH 2/2] vulkan_decode: convert max level from vulkan to av for comparisons Lynne
[not found] ` <NdciLba--3-9@lynne.ee-NdciOcd----9>
2023-09-08 4:59 ` Lynne
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
This inbox may be cloned and mirrored by anyone:
git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
ffmpegdev@gitmailbox.com
public-inbox-index ffmpegdev
Example config snippet for mirrors.
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git