From: Zhao Zhili <quinkblack@foxmail.com>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 3/5] avutil/hwcontext_vulkan: Fix leaks when semaphore creation fails
Date: Tue, 20 Feb 2024 20:23:10 +0800
Message-ID: <tencent_AC354D0C203C4DB44FAE5579273338B61607@qq.com> (raw)
In-Reply-To: <tencent_DE407D9AC29F8EAB344E8E70312D939A7308@qq.com>
I found serious memory leaks (dozens of frames depends on video filter) when test vulkan, which may or may not directly related to vulkan. See trac 10873.
> 在 2024年2月20日,下午8:08,Zhao Zhili <quinkblack@foxmail.com> 写道:
>
> From: Zhao Zhili <zhilizhao@tencent.com>
>
> ---
> libavutil/hwcontext_vulkan.c | 30 +++++++++++++++++++-----------
> 1 file changed, 19 insertions(+), 11 deletions(-)
>
> diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
> index a84713e621..c64094f31c 100644
> --- a/libavutil/hwcontext_vulkan.c
> +++ b/libavutil/hwcontext_vulkan.c
> @@ -1807,23 +1807,30 @@ static void vulkan_frame_free(AVHWFramesContext *hwfc, AVVkFrame *f)
> VulkanDevicePriv *p = hwfc->device_ctx->internal->priv;
> FFVulkanFunctions *vk = &p->vkctx.vkfn;
> int nb_images = ff_vk_count_images(f);
> + int nb_sems = 0;
> +
> + while (nb_sems < FF_ARRAY_ELEMS(f->sem) && f->sem[nb_sems])
> + nb_sems++;
> +
> + if (nb_sems) {
> + VkSemaphoreWaitInfo sem_wait = {
> + .sType = VK_STRUCTURE_TYPE_SEMAPHORE_WAIT_INFO,
> + .flags = 0x0,
> + .pSemaphores = f->sem,
> + .pValues = f->sem_value,
> + .semaphoreCount = nb_sems,
> + };
>
> - VkSemaphoreWaitInfo sem_wait = {
> - .sType = VK_STRUCTURE_TYPE_SEMAPHORE_WAIT_INFO,
> - .flags = 0x0,
> - .pSemaphores = f->sem,
> - .pValues = f->sem_value,
> - .semaphoreCount = nb_images,
> - };
> -
> - vk->WaitSemaphores(hwctx->act_dev, &sem_wait, UINT64_MAX);
> + vk->WaitSemaphores(hwctx->act_dev, &sem_wait, UINT64_MAX);
> + }
>
> vulkan_free_internal(f);
>
> for (int i = 0; i < nb_images; i++) {
> vk->DestroyImage(hwctx->act_dev, f->img[i], hwctx->alloc);
> vk->FreeMemory(hwctx->act_dev, f->mem[i], hwctx->alloc);
> - vk->DestroySemaphore(hwctx->act_dev, f->sem[i], hwctx->alloc);
> + if (f->sem[i])
> + vk->DestroySemaphore(hwctx->act_dev, f->sem[i], hwctx->alloc);
> }
>
> av_free(f);
> @@ -2098,7 +2105,8 @@ static int create_frame(AVHWFramesContext *hwfc, AVVkFrame **frame,
> if (ret != VK_SUCCESS) {
> av_log(hwctx, AV_LOG_ERROR, "Failed to create semaphore: %s\n",
> ff_vk_ret2str(ret));
> - return AVERROR_EXTERNAL;
> + err = AVERROR_EXTERNAL;
> + goto fail;
> }
>
> f->queue_family[i] = p->nb_img_qfs > 1 ? VK_QUEUE_FAMILY_IGNORED : p->img_qfs[0];
> --
> 2.34.1
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
_______________________________________________
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:[~2024-02-20 12:23 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20240220120858.317766-1-quinkblack@foxmail.com>
2024-02-20 12:08 ` [FFmpeg-devel] [PATCH 2/5] avutil/hwcontext: Don't assume frames_uninit is reentrant Zhao Zhili
2024-02-20 12:08 ` [FFmpeg-devel] [PATCH 3/5] avutil/hwcontext_vulkan: Fix leaks when semaphore creation fails Zhao Zhili
2024-02-20 12:23 ` Zhao Zhili [this message]
2024-02-20 17:10 ` Lynne
2024-02-20 17:58 ` Zhao Zhili
2024-02-22 14:42 ` [FFmpeg-devel] [PATCH v2 " Zhao Zhili
2024-02-23 15:03 ` Lynne
2024-02-20 12:08 ` [FFmpeg-devel] [PATCH 4/5] avutil/hwcontext_vulkan: Fix leaks in map_from_drm Zhao Zhili
2024-02-20 12:08 ` [FFmpeg-devel] [PATCH 5/5] avutil/vulkan: avoid overreads in ff_vk_count_images Zhao Zhili
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=tencent_AC354D0C203C4DB44FAE5579273338B61607@qq.com \
--to=quinkblack@foxmail.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