Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: James Almer <jamrial@gmail.com>
To: ffmpeg-devel@ffmpeg.org
Subject: Re: [FFmpeg-devel] [PATCH] avcodec/nvdec: don't free NVDECContext->bitstream
Date: Tue, 6 Feb 2024 20:05:31 -0300
Message-ID: <723507be-77aa-4223-8f8d-e83882426636@gmail.com> (raw)
In-Reply-To: <b09e4172-0c77-4bf1-b483-58d1ddf60055@rothenpieler.org>

On 2/6/2024 8:00 PM, Timo Rothenpieler wrote:
> On 06.02.2024 23:57, James Almer wrote:
>> Ensure all hwaccels that allocate a buffer use 
>> NVDECContext->bitstream_internal.
>>
>> Signed-off-by: James Almer <jamrial@gmail.com>
>> ---
>>   libavcodec/nvdec.c      | 2 +-
>>   libavcodec/nvdec_h264.c | 4 ++--
>>   libavcodec/nvdec_hevc.c | 4 ++--
>>   3 files changed, 5 insertions(+), 5 deletions(-)
>>
>> diff --git a/libavcodec/nvdec.c b/libavcodec/nvdec.c
>> index 27be644356..d13b790632 100644
>> --- a/libavcodec/nvdec.c
>> +++ b/libavcodec/nvdec.c
>> @@ -259,8 +259,8 @@ int ff_nvdec_decode_uninit(AVCodecContext *avctx)
>>   {
>>       NVDECContext *ctx = avctx->internal->hwaccel_priv_data;
>> -    av_freep(&ctx->bitstream);
>>       av_freep(&ctx->bitstream_internal);
>> +    ctx->bitstream           = NULL;
>>       ctx->bitstream_len       = 0;
>>       ctx->bitstream_allocated = 0;
>> diff --git a/libavcodec/nvdec_h264.c b/libavcodec/nvdec_h264.c
>> index f022619b64..8c72d5f4f7 100644
>> --- a/libavcodec/nvdec_h264.c
>> +++ b/libavcodec/nvdec_h264.c
>> @@ -138,11 +138,11 @@ static int 
>> nvdec_h264_decode_slice(AVCodecContext *avctx, const uint8_t *buffer,
>>       const H264SliceContext *sl = &h->slice_ctx[0];
>>       void *tmp;
>> -    tmp = av_fast_realloc(ctx->bitstream, &ctx->bitstream_allocated,
>> +    tmp = av_fast_realloc(ctx->bitstream_internal, 
>> &ctx->bitstream_allocated,
>>                             ctx->bitstream_len + size + 3);
>>       if (!tmp)
>>           return AVERROR(ENOMEM);
>> -    ctx->bitstream = tmp;
>> +    ctx->bitstream = ctx->bitstream_internal = tmp;
>>       tmp = av_fast_realloc(ctx->slice_offsets, 
>> &ctx->slice_offsets_allocated,
>>                             (ctx->nb_slices + 1) * 
>> sizeof(*ctx->slice_offsets));
>> diff --git a/libavcodec/nvdec_hevc.c b/libavcodec/nvdec_hevc.c
>> index b83d5edcf9..25319a1328 100644
>> --- a/libavcodec/nvdec_hevc.c
>> +++ b/libavcodec/nvdec_hevc.c
>> @@ -274,11 +274,11 @@ static int 
>> nvdec_hevc_decode_slice(AVCodecContext *avctx, const uint8_t *buffer,
>>       NVDECContext *ctx = avctx->internal->hwaccel_priv_data;
>>       void *tmp;
>> -    tmp = av_fast_realloc(ctx->bitstream, &ctx->bitstream_allocated,
>> +    tmp = av_fast_realloc(ctx->bitstream_internal, 
>> &ctx->bitstream_allocated,
>>                             ctx->bitstream_len + size + 3);
>>       if (!tmp)
>>           return AVERROR(ENOMEM);
>> -    ctx->bitstream = tmp;
>> +    ctx->bitstream = ctx->bitstream_internal = tmp;
>>       tmp = av_fast_realloc(ctx->slice_offsets, 
>> &ctx->slice_offsets_allocated,
>>                             (ctx->nb_slices + 1) * 
>> sizeof(*ctx->slice_offsets));
> 
> bitstream_internal was added for the av1 slice reconstructions.
> Probably for the exact reason to avoid this confusion.
> 
> Looking at this some more... With this patch in place, there should 
> never be a situation in which we want to free/freep ctx->bitstream?
> So the freep() on it should probably be removed alongside, and replace 
> with just setting it to NULL.

That's what i did above, in the libavcodec/nvdec.c chunk.
_______________________________________________
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".

  parent reply	other threads:[~2024-02-06 23:05 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-06 21:26 [FFmpeg-devel] [PATCH] avcodec/nvdec: don't attempt to free NVDECContext->bitstream on simple decode_slice hwaccels James Almer
2024-02-06 22:14 ` Andreas Rheinhardt
2024-02-06 22:57   ` [FFmpeg-devel] [PATCH] avcodec/nvdec: don't free NVDECContext->bitstream James Almer
2024-02-06 23:00     ` Timo Rothenpieler
2024-02-06 23:02       ` Timo Rothenpieler
2024-02-06 23:05       ` James Almer [this message]
2024-02-06 23:06         ` Timo Rothenpieler
2024-02-06 22:58   ` [FFmpeg-devel] [PATCH] avcodec/nvdec: don't attempt to free NVDECContext->bitstream on simple decode_slice hwaccels James Almer

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=723507be-77aa-4223-8f8d-e83882426636@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