From: James Almer <jamrial@gmail.com>
To: ffmpeg-devel@ffmpeg.org
Subject: Re: [FFmpeg-devel] [PATCH 3/3] lavu/hwcontext: clarify behavior on av_hwframe_map() failure
Date: Wed, 19 Jan 2022 10:45:12 -0300
Message-ID: <51796511-a27c-3318-9d97-bb8ce4c972a0@gmail.com> (raw)
In-Reply-To: <20220119134040.774-3-anton@khirnov.net>
On 1/19/2022 10:40 AM, Anton Khirnov wrote:
> Clear anything that av_hwframe_map() might have done to the destination
> frame, but leave caller-provided fields unchanged.
> ---
> libavutil/hwcontext.c | 23 +++++++++++++++++++++--
> libavutil/hwcontext.h | 4 ++++
> 2 files changed, 25 insertions(+), 2 deletions(-)
>
> diff --git a/libavutil/hwcontext.c b/libavutil/hwcontext.c
> index 31c7840dba..ae33da1262 100644
> --- a/libavutil/hwcontext.c
> +++ b/libavutil/hwcontext.c
> @@ -18,6 +18,7 @@
>
> #include "config.h"
>
> +#include "avassert.h"
> #include "buffer.h"
> #include "common.h"
> #include "hwcontext.h"
> @@ -788,6 +789,8 @@ fail:
>
> int av_hwframe_map(AVFrame *dst, const AVFrame *src, int flags)
> {
> + AVBufferRef *orig_dst_frames = dst->hw_frames_ctx;
> + enum AVPixelFormat orig_dst_fmt = dst->format;
> AVHWFramesContext *src_frames, *dst_frames;
> HWMapDescriptor *hwmap;
> int ret;
> @@ -825,7 +828,7 @@ int av_hwframe_map(AVFrame *dst, const AVFrame *src, int flags)
> ret = src_frames->internal->hw_type->map_from(src_frames,
> dst, src, flags);
> if (ret != AVERROR(ENOSYS))
> - return ret;
> + goto fail;
What if ret is 0? Is the stuff in the fail label meant to run on success?
> }
> }
>
> @@ -837,11 +840,27 @@ int av_hwframe_map(AVFrame *dst, const AVFrame *src, int flags)
> ret = dst_frames->internal->hw_type->map_to(dst_frames,
> dst, src, flags);
> if (ret != AVERROR(ENOSYS))
> - return ret;
> + goto fail;
> }
> }
>
> return AVERROR(ENOSYS);
> +
> +fail:
> + // if the caller provided dst frames context, it should be preserved
> + // by this function
> + av_assert0(orig_dst_frames == NULL ||
> + orig_dst_frames == dst->hw_frames_ctx);
> +
> + // preserve user-provided dst frame fields, but clean
> + // anything we might have set
> + dst->hw_frames_ctx = NULL;
> + av_frame_unref(dst);
> +
> + dst->hw_frames_ctx = orig_dst_frames;
> + dst->format = orig_dst_fmt;
> +
> + return ret;
> }
>
> int av_hwframe_ctx_create_derived(AVBufferRef **derived_frame_ctx,
> diff --git a/libavutil/hwcontext.h b/libavutil/hwcontext.h
> index 04d19d89c2..c18b7e1e8b 100644
> --- a/libavutil/hwcontext.h
> +++ b/libavutil/hwcontext.h
> @@ -571,6 +571,10 @@ enum {
> * possible with the given arguments and hwframe setup, while other return
> * values indicate that it failed somehow.
> *
> + * On failure, the destination frame will be left blank, except for the
> + * hw_frames_ctx/format fields thay may have been set by the caller - those will
> + * be preserved as they were.
> + *
> * @param dst Destination frame, to contain the mapping.
> * @param src Source frame, to be mapped.
> * @param flags Some combination of AV_HWFRAME_MAP_* flags.
_______________________________________________
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".
next prev parent reply other threads:[~2022-01-19 13:45 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-19 13:40 [FFmpeg-devel] [PATCH 1/3] lavu/hwcontext_opencl: clear dangling pointers on map failure Anton Khirnov
2022-01-19 13:40 ` [FFmpeg-devel] [PATCH 2/3] lavu/hwcontext_vulkan: " Anton Khirnov
2022-01-19 13:40 ` [FFmpeg-devel] [PATCH 3/3] lavu/hwcontext: clarify behavior on av_hwframe_map() failure Anton Khirnov
2022-01-19 13:45 ` James Almer [this message]
2022-01-19 13:47 ` Anton Khirnov
2022-01-19 14:01 ` [FFmpeg-devel] [PATCH] " Anton Khirnov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=51796511-a27c-3318-9d97-bb8ce4c972a0@gmail.com \
--to=jamrial@gmail.com \
--cc=ffmpeg-devel@ffmpeg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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