Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Ben Avison <bavison@riscosopen.org>
To: ffmpeg-devel@ffmpeg.org
Cc: Ben Avison <bavison@riscosopen.org>
Subject: [FFmpeg-devel] [PATCH 04/10] avcodec/vc1: Introduce fast path for unescaping bitstream buffer
Date: Fri, 25 Mar 2022 18:52:51 +0000
Message-ID: <20220325185257.513933-5-bavison@riscosopen.org> (raw)
In-Reply-To: <20220325185257.513933-1-bavison@riscosopen.org>

Includes a checkasm test.

Signed-off-by: Ben Avison <bavison@riscosopen.org>
---
 libavcodec/vc1dec.c     | 20 +++++++-------
 libavcodec/vc1dsp.c     |  2 ++
 libavcodec/vc1dsp.h     |  3 +++
 tests/checkasm/vc1dsp.c | 59 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 74 insertions(+), 10 deletions(-)

diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c
index 1c92b9d401..6a30b5b664 100644
--- a/libavcodec/vc1dec.c
+++ b/libavcodec/vc1dec.c
@@ -490,7 +490,7 @@ static av_cold int vc1_decode_init(AVCodecContext *avctx)
             size = next - start - 4;
             if (size <= 0)
                 continue;
-            buf2_size = vc1_unescape_buffer(start + 4, size, buf2);
+            buf2_size = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2);
             init_get_bits(&gb, buf2, buf2_size * 8);
             switch (AV_RB32(start)) {
             case VC1_CODE_SEQHDR:
@@ -680,7 +680,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                 case VC1_CODE_FRAME:
                     if (avctx->hwaccel)
                         buf_start = start;
-                    buf_size2 = vc1_unescape_buffer(start + 4, size, buf2);
+                    buf_size2 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2);
                     break;
                 case VC1_CODE_FIELD: {
                     int buf_size3;
@@ -697,8 +697,8 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                         ret = AVERROR(ENOMEM);
                         goto err;
                     }
-                    buf_size3 = vc1_unescape_buffer(start + 4, size,
-                                                    slices[n_slices].buf);
+                    buf_size3 = v->vc1dsp.vc1_unescape_buffer(start + 4, size,
+                                                              slices[n_slices].buf);
                     init_get_bits(&slices[n_slices].gb, slices[n_slices].buf,
                                   buf_size3 << 3);
                     slices[n_slices].mby_start = avctx->coded_height + 31 >> 5;
@@ -709,7 +709,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                     break;
                 }
                 case VC1_CODE_ENTRYPOINT: /* it should be before frame data */
-                    buf_size2 = vc1_unescape_buffer(start + 4, size, buf2);
+                    buf_size2 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2);
                     init_get_bits(&s->gb, buf2, buf_size2 * 8);
                     ff_vc1_decode_entry_point(avctx, v, &s->gb);
                     break;
@@ -726,8 +726,8 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                         ret = AVERROR(ENOMEM);
                         goto err;
                     }
-                    buf_size3 = vc1_unescape_buffer(start + 4, size,
-                                                    slices[n_slices].buf);
+                    buf_size3 = v->vc1dsp.vc1_unescape_buffer(start + 4, size,
+                                                              slices[n_slices].buf);
                     init_get_bits(&slices[n_slices].gb, slices[n_slices].buf,
                                   buf_size3 << 3);
                     slices[n_slices].mby_start = get_bits(&slices[n_slices].gb, 9);
@@ -761,7 +761,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                     ret = AVERROR(ENOMEM);
                     goto err;
                 }
-                buf_size3 = vc1_unescape_buffer(divider + 4, buf + buf_size - divider - 4, slices[n_slices].buf);
+                buf_size3 = v->vc1dsp.vc1_unescape_buffer(divider + 4, buf + buf_size - divider - 4, slices[n_slices].buf);
                 init_get_bits(&slices[n_slices].gb, slices[n_slices].buf,
                               buf_size3 << 3);
                 slices[n_slices].mby_start = s->mb_height + 1 >> 1;
@@ -770,9 +770,9 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
                 n_slices1 = n_slices - 1;
                 n_slices++;
             }
-            buf_size2 = vc1_unescape_buffer(buf, divider - buf, buf2);
+            buf_size2 = v->vc1dsp.vc1_unescape_buffer(buf, divider - buf, buf2);
         } else {
-            buf_size2 = vc1_unescape_buffer(buf, buf_size, buf2);
+            buf_size2 = v->vc1dsp.vc1_unescape_buffer(buf, buf_size, buf2);
         }
         init_get_bits(&s->gb, buf2, buf_size2*8);
     } else{
diff --git a/libavcodec/vc1dsp.c b/libavcodec/vc1dsp.c
index a29b91bf3d..11d493f002 100644
--- a/libavcodec/vc1dsp.c
+++ b/libavcodec/vc1dsp.c
@@ -34,6 +34,7 @@
 #include "rnd_avg.h"
 #include "vc1dsp.h"
 #include "startcode.h"
+#include "vc1_common.h"
 
 /* Apply overlap transform to horizontal edge */
 static void vc1_v_overlap_c(uint8_t *src, int stride)
@@ -1030,6 +1031,7 @@ av_cold void ff_vc1dsp_init(VC1DSPContext *dsp)
 #endif /* CONFIG_WMV3IMAGE_DECODER || CONFIG_VC1IMAGE_DECODER */
 
     dsp->startcode_find_candidate = ff_startcode_find_candidate_c;
+    dsp->vc1_unescape_buffer      = vc1_unescape_buffer;
 
     if (ARCH_AARCH64)
         ff_vc1dsp_init_aarch64(dsp);
diff --git a/libavcodec/vc1dsp.h b/libavcodec/vc1dsp.h
index c6443acb20..8be1198071 100644
--- a/libavcodec/vc1dsp.h
+++ b/libavcodec/vc1dsp.h
@@ -80,6 +80,9 @@ typedef struct VC1DSPContext {
      * one or more further zero bytes and a one byte.
      */
     int (*startcode_find_candidate)(const uint8_t *buf, int size);
+
+    /* Copy a buffer, removing startcode emulation escape bytes as we go */
+    int (*vc1_unescape_buffer)(const uint8_t *src, int size, uint8_t *dst);
 } VC1DSPContext;
 
 void ff_vc1dsp_init(VC1DSPContext* c);
diff --git a/tests/checkasm/vc1dsp.c b/tests/checkasm/vc1dsp.c
index 0823ccad31..0ab5892403 100644
--- a/tests/checkasm/vc1dsp.c
+++ b/tests/checkasm/vc1dsp.c
@@ -286,6 +286,20 @@ static matrix *generate_inverse_quantized_transform_coefficients(size_t width, s
         }                                                                   \
     } while (0)
 
+#define TEST_UNESCAPE                                                                                   \
+    do {                                                                                            \
+        for (int count = 100; count > 0; --count) {                                                 \
+            escaped_offset = rnd() & 7;                                                             \
+            unescaped_offset = rnd() & 7;                                                           \
+            escaped_len = (1u << (rnd() % 8) + 3) - (rnd() & 7);                                    \
+            RANDOMIZE_BUFFER8(unescaped, UNESCAPE_BUF_SIZE);                                        \
+            len0 = call_ref(escaped0 + escaped_offset, escaped_len, unescaped0 + unescaped_offset); \
+            len1 = call_new(escaped1 + escaped_offset, escaped_len, unescaped1 + unescaped_offset); \
+            if (len0 != len1 || memcmp(unescaped0, unescaped1, len0))                               \
+                fail();                                                                             \
+        }                                                                                           \
+    } while (0)
+
 void checkasm_check_vc1dsp(void)
 {
     /* Inverse transform input coefficients are stored in a 16-bit buffer
@@ -309,6 +323,14 @@ void checkasm_check_vc1dsp(void)
     LOCAL_ALIGNED_4(uint8_t, filter_buf0, [24 * 24]);
     LOCAL_ALIGNED_4(uint8_t, filter_buf1, [24 * 24]);
 
+    /* This appears to be a typical length of buffer in use */
+#define LOG2_UNESCAPE_BUF_SIZE 17
+#define UNESCAPE_BUF_SIZE (1u<<LOG2_UNESCAPE_BUF_SIZE)
+    LOCAL_ALIGNED_8(uint8_t, escaped0, [UNESCAPE_BUF_SIZE]);
+    LOCAL_ALIGNED_8(uint8_t, escaped1, [UNESCAPE_BUF_SIZE]);
+    LOCAL_ALIGNED_8(uint8_t, unescaped0, [UNESCAPE_BUF_SIZE]);
+    LOCAL_ALIGNED_8(uint8_t, unescaped1, [UNESCAPE_BUF_SIZE]);
+
     VC1DSPContext h;
 
     ff_vc1dsp_init(&h);
@@ -349,4 +371,41 @@ void checkasm_check_vc1dsp(void)
     CHECK_LOOP_FILTER(vc1_h_loop_filter16);
 
     report("loop_filter");
+
+    if (check_func(h.vc1_unescape_buffer, "vc1dsp.vc1_unescape_buffer")) {
+        int len0, len1, escaped_offset, unescaped_offset, escaped_len;
+        declare_func_emms(AV_CPU_FLAG_MMX, int, const uint8_t *, int, uint8_t *);
+
+        /* Test data which consists of escapes sequences packed as tightly as possible */
+        for (int x = 0; x < UNESCAPE_BUF_SIZE; ++x)
+            escaped1[x] = escaped0[x] = 3 * (x % 3 == 0);
+        TEST_UNESCAPE;
+
+        /* Test random data */
+        RANDOMIZE_BUFFER8(escaped, UNESCAPE_BUF_SIZE);
+        TEST_UNESCAPE;
+
+        /* Test data with escape sequences at random intervals */
+        for (int x = 0; x <= UNESCAPE_BUF_SIZE - 4;) {
+            int gap, gap_msb;
+            escaped1[x+0] = escaped0[x+0] = 0;
+            escaped1[x+1] = escaped0[x+1] = 0;
+            escaped1[x+2] = escaped0[x+2] = 3;
+            escaped1[x+3] = escaped0[x+3] = rnd() & 3;
+            gap_msb = 2u << (rnd() % 8);
+            gap = (rnd() &~ -gap_msb) | gap_msb;
+            x += gap;
+        }
+        TEST_UNESCAPE;
+
+        /* Test data which is known to contain no escape sequences */
+        memset(escaped0, 0xFF, UNESCAPE_BUF_SIZE);
+        memset(escaped1, 0xFF, UNESCAPE_BUF_SIZE);
+        TEST_UNESCAPE;
+
+        /* Benchmark the no-escape-sequences case */
+        bench_new(escaped1, UNESCAPE_BUF_SIZE, unescaped1);
+    }
+
+    report("unescape_buffer");
 }
-- 
2.25.1

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

  parent reply	other threads:[~2022-03-25 18:54 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-17 18:58 [FFmpeg-devel] [PATCH 0/6] avcodec/vc1: Arm optimisations Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 1/6] avcodec/vc1: Arm 64-bit NEON deblocking filter fast paths Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 2/6] avcodec/vc1: Arm 32-bit " Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 3/6] avcodec/vc1: Arm 64-bit NEON inverse transform " Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 4/6] avcodec/idctdsp: Arm 64-bit NEON block add and clamp " Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 5/6] avcodec/blockdsp: Arm 64-bit NEON block clear " Ben Avison
2022-03-17 18:58 ` [FFmpeg-devel] [PATCH 6/6] avcodec/vc1: Introduce fast path for unescaping bitstream buffer Ben Avison
2022-03-18 19:10   ` Andreas Rheinhardt
2022-03-21 15:51     ` Ben Avison
2022-03-21 20:44       ` Martin Storsjö
2022-03-19 23:06 ` [FFmpeg-devel] [PATCH 0/6] avcodec/vc1: Arm optimisations Martin Storsjö
2022-03-19 23:07   ` Martin Storsjö
2022-03-21 17:37   ` Ben Avison
2022-03-21 22:29     ` Martin Storsjö
2022-03-25 18:52 ` [FFmpeg-devel] [PATCH v2 00/10] " Ben Avison
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 01/10] checkasm: Add vc1dsp in-loop deblocking filter tests Ben Avison
2022-03-25 22:53     ` Martin Storsjö
2022-03-28 18:28       ` Ben Avison
2022-03-29 11:47         ` Martin Storsjö
2022-03-29 12:24     ` Martin Storsjö
2022-03-29 12:43     ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 02/10] checkasm: Add vc1dsp inverse transform tests Ben Avison
2022-03-29 12:41     ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 03/10] checkasm: Add idctdsp add/put-pixels-clamped tests Ben Avison
2022-03-29 13:13     ` Martin Storsjö
2022-03-29 19:56       ` Martin Storsjö
2022-03-29 20:22       ` Ben Avison
2022-03-29 20:30         ` Martin Storsjö
2022-03-25 18:52   ` Ben Avison [this message]
2022-03-29 20:37     ` [FFmpeg-devel] [PATCH 04/10] avcodec/vc1: Introduce fast path for unescaping bitstream buffer Martin Storsjö
2022-03-31 13:58       ` Ben Avison
2022-03-31 14:07         ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 05/10] avcodec/vc1: Arm 64-bit NEON deblocking filter fast paths Ben Avison
2022-03-30 12:35     ` Martin Storsjö
2022-03-31 15:15       ` Ben Avison
2022-03-31 21:21         ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 06/10] avcodec/vc1: Arm 32-bit " Ben Avison
2022-03-25 19:27     ` Lynne
2022-03-25 19:49       ` Martin Storsjö
2022-03-25 19:55         ` Lynne
2022-03-30 12:37     ` Martin Storsjö
2022-03-30 13:03     ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 07/10] avcodec/vc1: Arm 64-bit NEON inverse transform " Ben Avison
2022-03-30 13:49     ` Martin Storsjö
2022-03-30 14:01       ` Martin Storsjö
2022-03-31 15:37       ` Ben Avison
2022-03-31 21:32         ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 08/10] avcodec/idctdsp: Arm 64-bit NEON block add and clamp " Ben Avison
2022-03-30 14:14     ` Martin Storsjö
2022-03-31 16:47       ` Ben Avison
2022-03-31 21:42         ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 09/10] avcodec/vc1: Arm 64-bit NEON unescape fast path Ben Avison
2022-03-30 14:35     ` Martin Storsjö
2022-03-25 18:52   ` [FFmpeg-devel] [PATCH 10/10] avcodec/vc1: Arm 32-bit " Ben Avison
2022-03-30 14:35     ` Martin Storsjö

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=20220325185257.513933-5-bavison@riscosopen.org \
    --to=bavison@riscosopen.org \
    --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