From patchwork Sat May 21 14:07:40 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Aman Karmani X-Patchwork-Id: 35876 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a21:9992:b0:82:461d:f3b with SMTP id ve18csp1081809pzb; Sat, 21 May 2022 07:08:17 -0700 (PDT) X-Google-Smtp-Source: ABdhPJxDShxpkOYW8xF0xlbrUyWI+3rO8t1oFKNvF0nMSdLRv48I2FpJhf4akAduRjYDQs1O4vNk X-Received: by 2002:a17:907:6d06:b0:6fa:8499:66ce with SMTP id sa6-20020a1709076d0600b006fa849966cemr13172062ejc.718.1653142096764; Sat, 21 May 2022 07:08:16 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1653142096; cv=none; d=google.com; s=arc-20160816; b=KZ8IPPPop/+b2UxIOgsihaFRr0R04d1I5bbF166Aaz79ff4x1hXYT3X6q/a+z52KaR 0SNguPnhZZ3ENd7Uh5d8gnFOgo2Fbt9w9AiyOv6ziECLF59j7H7wxlYp2MUD8yp1E4jW lUBWw8iAli4X2QpiiX2rUz+qxFgVtP+nfsr3U0IDmokygh7MQfIIyPauMKcwl3kYDG3m 9PDNiarSzd9dTD1z6HpvC8CeMGD9PL35fA4YIOveDuljnxu346K8qmSorGeKyEjd+7Mz jpJumdCKHLqtrJB/7NtbzynvSjBPD8utZblvUE++X/qLl5ar9qht04rjucux2yG7gO/q iNvw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:to:mime-version:fcc:date:references :in-reply-to:message-id:from:dkim-signature:delivered-to; bh=uqY240/thdRswmNC5lRyiPwAarze5IRq59soV5HyfNc=; b=eEvfC/7PDiuorHwfDYyJOqTETNr4VdzFQLPrImC2IhP37gAyL+hGK+6oIJm0vc8WyG wgkfZTY2xEhOOe7Q3o6jSu9YyX+mYdgbB3JtnyxeCxfB9aGYkOw03QXjG4aly5S9OI0e KwcbEYNqObWgCbinM0KHAIyb2GPRXR7lsymW3MXl8oMKKt2u3sw1Xix3kQRGdYOzBZ7X k5yFVqAD3jQ7C21ijFR3vzvuxIuih/SLBjEwukegKnriDiKf5qEFOnKT2dFHnsyjR1Rh 4l1q05bFbEfyo0Wx6NqNsif+C6jVgSX2stwsA3L236+iLha66DASmkOYZRvOKG/EuCQw FhgA== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@gmail.com header.s=20210112 header.b=YoXsSbEZ; spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org; dmarc=fail (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id y5-20020a056402440500b00427ad71530fsi14429090eda.528.2022.05.21.07.08.16; Sat, 21 May 2022 07:08:16 -0700 (PDT) Received-SPF: pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) client-ip=79.124.17.100; Authentication-Results: mx.google.com; dkim=neutral (body hash did not verify) header.i=@gmail.com header.s=20210112 header.b=YoXsSbEZ; spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org; dmarc=fail (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 99DD168B475; Sat, 21 May 2022 17:07:55 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-pj1-f54.google.com (mail-pj1-f54.google.com [209.85.216.54]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 0E1AD68A2F6 for ; Sat, 21 May 2022 17:07:48 +0300 (EEST) Received: by mail-pj1-f54.google.com with SMTP id l14so10255475pjk.2 for ; Sat, 21 May 2022 07:07:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:message-id:in-reply-to:references:date:subject:fcc :content-transfer-encoding:mime-version:to:cc; bh=u2gEFs/dHnEVna8RkQ8KbNasCai+Mdvz7MO5iZtNvzc=; b=YoXsSbEZv+q51mIMzE99HZgR/xP634caG8ErGrhDKiIS0tSOBB8YRHJR4WRhG6T21/ D+/ObABc+IPt1rKWFQ1a2ZboYlT8+LfLhoFP9aSYyxcJMPGzs0phILJzJ5FiyGQ2BrDD CULstTQzkJo8XYnrV4+MCuO+hvLOTwtQfIkVutXzKBBjgXgLZkUsnqlLN/PvZ7bJL8+9 BfLolS9z/UhpAVI7zr8iMS4qraAprtKWpJaHR508sLZceq8fOpULTfa0jZ3AQnG6OrxM 6RW4et0CCSfsVl2ajvWdCYaOSxWRxlxblcDA1p3A8Ipv6VSmCiqRSqQz+LK+qMflbRz4 5Kig== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:message-id:in-reply-to:references:date :subject:fcc:content-transfer-encoding:mime-version:to:cc; bh=u2gEFs/dHnEVna8RkQ8KbNasCai+Mdvz7MO5iZtNvzc=; b=ZZNDDn3ITmdsJwdNvB1cRvzCg5b13YOS6ozh9t8LDk9lig/AfCVB0dCif4ZZ+l+NXP luLqxEoxTSXmervC9d/Hu5q8FmaiSJLlWMVifyZW24D16Dl+Otf26djBL7ep8BmgYC1O /BrrYdvY2Yf+e/z1CzZ+9UMozMPegWz7/j+yfT6rx5DBUQlGvGfsJXiGGjDbAUF7Ecp4 dl2UqNNfDn0ltRF3ICUvw7w0gvxj949yKhXqvTcMTs7OgMQmPDAFpg6LOiUWf7+JOxNg u8b5vnbzM8UhcWxkjUXKQxqF8l/kPuVY2gyXdbvwdTTzTaKYgYg69b5+Y5X561gEizHA ge6A== X-Gm-Message-State: AOAM532hI+nprGofrepJ3nqEBYMZKPVYlNxAOSvUkjKEwus4p1lPTpJF /s77UR0SVGiKEKDtOwP1tvMc5WbYgYhxpw== X-Received: by 2002:a17:90b:1e4f:b0:1e0:b91:39ef with SMTP id pi15-20020a17090b1e4f00b001e00b9139efmr5177971pjb.51.1653142066264; Sat, 21 May 2022 07:07:46 -0700 (PDT) Received: from [127.0.0.1] (master.gitmailbox.com. [34.83.118.50]) by smtp.gmail.com with ESMTPSA id z124-20020a623382000000b005183cf12184sm3583638pfz.133.2022.05.21.07.07.45 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Sat, 21 May 2022 07:07:45 -0700 (PDT) From: softworkz X-Google-Original-From: softworkz Message-Id: In-Reply-To: References: Date: Sat, 21 May 2022 14:07:40 +0000 Fcc: Sent MIME-Version: 1.0 To: ffmpeg-devel@ffmpeg.org Subject: [FFmpeg-devel] [PATCH v2 2/4] avutils/hwcontext: add derive-device function which searches for existing devices in both directions X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches Cc: Mark Thompson , softworkz Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: uEu8sSTludv2 From: softworkz The test /libavutil/tests/hwdevice checks that when deriving a device from a source device and then deriving back to the type of the source device, the result is matching the original source device, i.e. the derivation mechanism doesn't create a new device in this case. Previously, this test was usually passed, but only due to two different kind of flaws: 1. The test covers only a single level of derivation (and back) It derives device Y from device X and then Y back to the type of X and checks whether the result matches X. What it doesn't check for, are longer chains of derivation like: CUDA1 > OpenCL2 > CUDA3 and then back to OpenCL4 In that case, the second derivation returns the first device (CUDA3 == CUDA1), but when deriving OpenCL4, hwcontext.c was creating a new OpenCL4 context instead of returning OpenCL2, because there was no link from CUDA1 to OpenCL2 (only backwards from OpenCL2 to CUDA1) If the test would check for two levels of derivation, it would have failed. This patch fixes those (yet untested) cases by introducing forward references (derived_device) in addition to the existing back references (source_device). 2. hwcontext_qsv didn't properly set the source_device In case of QSV, hwcontext_qsv creates a source context internally (vaapi, dxva2 or d3d11va) without calling av_hwdevice_ctx_create_derived and without setting source_device. This way, the hwcontext test ran successful, but what practically happened, was that - for example - deriving vaapi from qsv didn't return the original underlying vaapi device and a new one was created instead: Exactly what the test is intended to detect and prevent. It just couldn't do so, because the original device was hidden (= not set as the source_device of the QSV device). This patch properly makes these setting and fixes all derivation scenarios. (at a later stage, /libavutil/tests/hwdevice should be extended to check longer derivation chains as well) Signed-off-by: softworkz --- libavutil/hwcontext.c | 167 +++++++++++++++++++++++++++++++-- libavutil/hwcontext.h | 20 ++++ libavutil/hwcontext_internal.h | 11 +++ libavutil/hwcontext_qsv.c | 11 ++- 4 files changed, 199 insertions(+), 10 deletions(-) diff --git a/libavutil/hwcontext.c b/libavutil/hwcontext.c index ab9ad3703e..43e39cb2d3 100644 --- a/libavutil/hwcontext.c +++ b/libavutil/hwcontext.c @@ -22,12 +22,17 @@ #include "buffer.h" #include "common.h" #include "hwcontext.h" + +#include + +#include "buffer_internal.h" #include "hwcontext_internal.h" #include "imgutils.h" #include "log.h" #include "mem.h" #include "pixdesc.h" #include "pixfmt.h" +#include "thread.h" static const HWContextType * const hw_table[] = { #if CONFIG_CUDA @@ -80,6 +85,84 @@ static const char *const hw_type_names[] = { [AV_HWDEVICE_TYPE_VULKAN] = "vulkan", }; +#define DEVICE_REGISTRY_SIZE 1024 + +static AVMutex mutex; +static int is_mutex_initialized = 0; +static int max_device_reg_id = 1; +static AVBuffer *hw_device_registry[DEVICE_REGISTRY_SIZE]; + +static int register_hw_device(const AVBufferRef *ref) +{ + AVHWDeviceContext *ctx = (AVHWDeviceContext*)ref->data; + const int reg_id = max_device_reg_id; + + if (ctx == NULL) + return AVERROR(EINVAL); + + if (!is_mutex_initialized) { + int ret; + ret = ff_mutex_init(&mutex, NULL); + if (ret) { + av_log(ctx, AV_LOG_ERROR, "hwcontext: mutex initialization failed! Error code: %d\n", ret); + return AVERROR(EINVAL); + } + + is_mutex_initialized = 1; + } + + ff_mutex_lock(&mutex); + + for (int i = 0; i < max_device_reg_id; ++i) { + if (hw_device_registry[i] != NULL && hw_device_registry[i] == ref->buffer) { + ff_mutex_unlock(&mutex); + return i; + } + } + + if (max_device_reg_id >= DEVICE_REGISTRY_SIZE) { + ff_mutex_unlock(&mutex); + av_log(ctx, AV_LOG_ERROR, "Device registry limit (%d) reached. Please check for excessive device creation.", DEVICE_REGISTRY_SIZE); + return AVERROR(ENOMEM); + } + + hw_device_registry[reg_id] = ref->buffer; + max_device_reg_id++; + + ff_mutex_unlock(&mutex); + + return reg_id; +} + +static void unregister_hw_device(const AVHWDeviceContext *ctx) +{ + if (ctx == NULL || !is_mutex_initialized) + return; + + ff_mutex_lock(&mutex); + + hw_device_registry[ctx->internal->registered_device_id] = NULL; + + ff_mutex_unlock(&mutex); +} + +static AVBufferRef *get_registered_hw_device(int registered_id) +{ + if (registered_id <= 0 || registered_id >= max_device_reg_id) + return NULL; + + ff_mutex_lock(&mutex); + + if (hw_device_registry[registered_id] != NULL && hw_device_registry[registered_id]->data != NULL) { + AVBufferRef *ref = av_ref_from_buffer(hw_device_registry[registered_id]); + return ref; + } + + ff_mutex_unlock(&mutex); + + return NULL; +} + enum AVHWDeviceType av_hwdevice_find_type_by_name(const char *name) { int type; @@ -124,6 +207,8 @@ static void hwdevice_ctx_free(void *opaque, uint8_t *data) { AVHWDeviceContext *ctx = (AVHWDeviceContext*)data; + unregister_hw_device(ctx); + /* uninit might still want access the hw context and the user * free() callback might destroy it, so uninit has to be called first */ if (ctx->internal->hw_type->device_uninit) @@ -612,7 +697,7 @@ int av_hwdevice_ctx_create(AVBufferRef **pdevice_ref, enum AVHWDeviceType type, const char *device, AVDictionary *opts, int flags) { AVBufferRef *device_ref = NULL; - AVHWDeviceContext *device_ctx; + AVHWDeviceContext *device_ctx = NULL; int ret = 0; device_ref = av_hwdevice_ctx_alloc(type); @@ -632,22 +717,58 @@ int av_hwdevice_ctx_create(AVBufferRef **pdevice_ref, enum AVHWDeviceType type, if (ret < 0) goto fail; + ret = register_hw_device(device_ref); + if (ret < 0) + goto fail; + ret = av_hwdevice_ctx_init(device_ref); if (ret < 0) goto fail; + device_ctx->internal->registered_device_id = ret; + *pdevice_ref = device_ref; return 0; fail: + unregister_hw_device(device_ctx); av_buffer_unref(&device_ref); *pdevice_ref = NULL; return ret; } -int av_hwdevice_ctx_create_derived_opts(AVBufferRef **dst_ref_ptr, - enum AVHWDeviceType type, - AVBufferRef *src_ref, - AVDictionary *options, int flags) +static AVBufferRef* find_derived_hwdevice_ctx(AVBufferRef *src_ref, enum AVHWDeviceType type) +{ + AVBufferRef *derived_ref; + AVHWDeviceContext *src_ctx; + int i; + + src_ctx = (AVHWDeviceContext*)src_ref->data; + if (src_ctx->type == type) + return src_ref; + + for (i = 0; i < AV_HWDEVICE_TYPE_NB; i++) + if (src_ctx->internal->derived_device_ids[i]) { + AVBufferRef *tmp_ref = get_registered_hw_device(src_ctx->internal->derived_device_ids[i]); + + if (tmp_ref) { + derived_ref = find_derived_hwdevice_ctx(tmp_ref, type); + + if (tmp_ref != derived_ref) + av_buffer_unref(&tmp_ref); + + if (derived_ref) + return derived_ref; + } + } + + return NULL; +} + +static int hwdevice_ctx_create_derived(AVBufferRef **dst_ref_ptr, + enum AVHWDeviceType type, + AVBufferRef *src_ref, + AVDictionary *options, int flags, + int get_existing) { AVBufferRef *dst_ref = NULL, *tmp_ref; AVHWDeviceContext *dst_ctx, *tmp_ctx; @@ -667,6 +788,18 @@ int av_hwdevice_ctx_create_derived_opts(AVBufferRef **dst_ref_ptr, tmp_ref = tmp_ctx->internal->source_device; } + if (get_existing) { + tmp_ref = find_derived_hwdevice_ctx(src_ref, type); + if (tmp_ref) { + dst_ref = av_buffer_ref(tmp_ref); + if (!dst_ref) { + ret = AVERROR(ENOMEM); + goto fail; + } + goto done; + } + } + dst_ref = av_hwdevice_ctx_alloc(type); if (!dst_ref) { ret = AVERROR(ENOMEM); @@ -688,6 +821,9 @@ int av_hwdevice_ctx_create_derived_opts(AVBufferRef **dst_ref_ptr, ret = AVERROR(ENOMEM); goto fail; } + if (!tmp_ctx->internal->derived_device_ids[type]) + tmp_ctx->internal->derived_device_ids[type] = dst_ctx->internal->registered_device_id; + ret = av_hwdevice_ctx_init(dst_ref); if (ret < 0) goto fail; @@ -712,12 +848,29 @@ fail: return ret; } +int av_hwdevice_ctx_create_derived_opts(AVBufferRef **dst_ref_ptr, + enum AVHWDeviceType type, + AVBufferRef *src_ref, + AVDictionary *options, int flags) +{ + return hwdevice_ctx_create_derived(dst_ref_ptr, type, src_ref, + options, flags, 0); +} + +int av_hwdevice_ctx_get_or_create_derived(AVBufferRef **dst_ref_ptr, + enum AVHWDeviceType type, + AVBufferRef *src_ref, int flags) +{ + return hwdevice_ctx_create_derived(dst_ref_ptr, type, src_ref, + NULL, flags, 1); +} + int av_hwdevice_ctx_create_derived(AVBufferRef **dst_ref_ptr, enum AVHWDeviceType type, AVBufferRef *src_ref, int flags) { - return av_hwdevice_ctx_create_derived_opts(dst_ref_ptr, type, src_ref, - NULL, flags); + return hwdevice_ctx_create_derived(dst_ref_ptr, type, src_ref, + NULL, flags, 0); } static void ff_hwframe_unmap(void *opaque, uint8_t *data) diff --git a/libavutil/hwcontext.h b/libavutil/hwcontext.h index c18b7e1e8b..3785811f98 100644 --- a/libavutil/hwcontext.h +++ b/libavutil/hwcontext.h @@ -37,6 +37,7 @@ enum AVHWDeviceType { AV_HWDEVICE_TYPE_OPENCL, AV_HWDEVICE_TYPE_MEDIACODEC, AV_HWDEVICE_TYPE_VULKAN, + AV_HWDEVICE_TYPE_NB, ///< number of hw device types }; typedef struct AVHWDeviceInternal AVHWDeviceInternal; @@ -328,6 +329,25 @@ int av_hwdevice_ctx_create_derived(AVBufferRef **dst_ctx, enum AVHWDeviceType type, AVBufferRef *src_ctx, int flags); +/** + * Create a new device of the specified type from an existing device. + * + * This function performs the same action as av_hwdevice_ctx_create_derived, + * however, if a derived device of the specified type already exists, + * it returns the existing instance. + * + * @param dst_ctx On success, a reference to the newly-created + * AVHWDeviceContext. + * @param type The type of the new device to create. + * @param src_ctx A reference to an existing AVHWDeviceContext which will be + * used to create the new device. + * @param flags Currently unused; should be set to zero. + * @return Zero on success, a negative AVERROR code on failure. + */ +int av_hwdevice_ctx_get_or_create_derived(AVBufferRef **dst_ctx, + enum AVHWDeviceType type, + AVBufferRef *src_ctx, int flags); + /** * Create a new device of the specified type from an existing device. * diff --git a/libavutil/hwcontext_internal.h b/libavutil/hwcontext_internal.h index e6266494ac..adb649cde4 100644 --- a/libavutil/hwcontext_internal.h +++ b/libavutil/hwcontext_internal.h @@ -109,6 +109,17 @@ struct AVHWDeviceInternal { * context it was derived from. */ AVBufferRef *source_device; + + /** + * An array of device registration ids from device contexts which + * were derived from this device. + */ + int derived_device_ids[AV_HWDEVICE_TYPE_NB]; + + /** + * ID under wich the hw context is registered internally. + */ + int registered_device_id; }; struct AVHWFramesInternal { diff --git a/libavutil/hwcontext_qsv.c b/libavutil/hwcontext_qsv.c index 66c0e38955..423b891ec5 100644 --- a/libavutil/hwcontext_qsv.c +++ b/libavutil/hwcontext_qsv.c @@ -309,7 +309,7 @@ static void qsv_frames_uninit(AVHWFramesContext *ctx) av_buffer_unref(&s->child_frames_ref); } -static void qsv_pool_release_dummy(void *opaque, uint8_t *data) +static void qsv_release_dummy(void *opaque, uint8_t *data) { } @@ -322,7 +322,7 @@ static AVBufferRef *qsv_pool_alloc(void *opaque, size_t size) if (s->nb_surfaces_used < hwctx->nb_surfaces) { s->nb_surfaces_used++; return av_buffer_create((uint8_t*)(s->surfaces_internal + s->nb_surfaces_used - 1), - sizeof(*hwctx->surfaces), qsv_pool_release_dummy, NULL, 0); + sizeof(*hwctx->surfaces), qsv_release_dummy, NULL, 0); } return NULL; @@ -1649,8 +1649,13 @@ static int qsv_device_create(AVHWDeviceContext *ctx, const char *device, child_device = (AVHWDeviceContext*)priv->child_device_ctx->data; impl = choose_implementation(device, child_device_type); + ret = qsv_device_derive_from_child(ctx, impl, child_device, 0); + if (ret >= 0) { + ctx->internal->source_device = av_buffer_ref(priv->child_device_ctx); + child_device->internal->derived_device_ids[ctx->type] = ctx->internal->registered_device_id; + } - return qsv_device_derive_from_child(ctx, impl, child_device, 0); + return ret; } const HWContextType ff_hwcontext_type_qsv = {