From: Michael Niedermayer <michael@niedermayer.cc>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: [FFmpeg-devel] [PATCH 3/4] avcodec/get_bits: Implement get_vlc_multi()
Date: Tue, 24 Oct 2023 17:04:42 +0200
Message-ID: <20231024150443.7438-3-michael@niedermayer.cc> (raw)
In-Reply-To: <20231024150443.7438-1-michael@niedermayer.cc>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
---
libavcodec/get_bits.h | 61 ++++++++++++++++++++++++++++++++++++++++---
1 file changed, 58 insertions(+), 3 deletions(-)
diff --git a/libavcodec/get_bits.h b/libavcodec/get_bits.h
index 86cea00494a..806e4b16f49 100644
--- a/libavcodec/get_bits.h
+++ b/libavcodec/get_bits.h
@@ -668,13 +668,68 @@ static av_always_inline int get_vlc2(GetBitContext *s, const VLCElem *table,
return code;
}
-static inline int get_vlc_multi(GetBitContext *s, uint8_t *dst,
+/**
+ * Parse a vlc / vlc_multi code.
+ * @param bits is the number of bits which will be read at once, must be
+ * identical to nb_bits in vlc_init(), should be known at compile time
+ * @param max_depth is the number of times bits bits must be read to completely
+ * read the longest vlc code
+ * = (max_vlc_length + bits - 1) / bits
+ * @param dst the parsed symbol(s) will be stored here. Up to 8 bytes are written
+ * @returns number of symbols parsed
+ * If the vlc code is invalid and max_depth=1, then no bits will be removed.
+ * If the vlc code is invalid and max_depth>1, then the number of bits removed
+ * is undefined.
+ */
+static inline int get_vlc_multi(GetBitContext *s, uint8_t * restrict dst,
const VLC_MULTI_ELEM *const Jtable,
const VLCElem *const table,
const int bits, const int max_depth)
{
- dst[0] = get_vlc2(s, table, bits, max_depth);
- return 1;
+ int ret, nb_bits, n;
+ unsigned int index;
+
+ OPEN_READER(re, s);
+ UPDATE_CACHE(re, s);
+
+ index = SHOW_UBITS(re, s, bits);
+ n = Jtable[index].len;
+ if (Jtable[index].num) {
+ AV_COPY64U(dst, Jtable[index].val);
+ ret = Jtable[index].num;
+ } else {
+ int code = table[index].sym;
+
+ if (av_builtin_constant_p(bits <= MIN_CACHE_BITS/2) && bits <= MIN_CACHE_BITS/2) {
+ SKIP_BITS(re, s, bits);
+ } else {
+ LAST_SKIP_BITS(re, s, bits);
+ UPDATE_CACHE(re, s);
+ }
+
+ nb_bits = -n;
+
+ index = SHOW_UBITS(re, s, nb_bits) + code;
+ code = table[index].sym;
+ n = table[index].len;
+ if (max_depth > 2 && n < 0) {
+ LAST_SKIP_BITS(re, s, nb_bits);
+ UPDATE_CACHE(re, s);
+
+ nb_bits = -n;
+
+ index = SHOW_UBITS(re, s, nb_bits) + code;
+ code = table[index].sym;
+ n = table[index].len;
+ }
+ dst[0] = code;
+ ret = n > 0;
+ }
+ LAST_SKIP_BITS(re, s, n);
+
+ CLOSE_READER(re, s);
+
+ return ret;
}
static inline int decode012(GetBitContext *gb)
--
2.17.1
_______________________________________________
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".
next prev parent reply other threads:[~2023-10-24 15:05 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-24 15:04 [FFmpeg-devel] [PATCH 1/4] avcodec/magicyuv: Use a compile time constant for vlc_bits Michael Niedermayer
2023-10-24 15:04 ` [FFmpeg-devel] [PATCH 2/4] avcodec/get_bits: Avoid 2nd bitstream read in GET_VLC() if bits are known at build and small Michael Niedermayer
2023-10-27 3:10 ` Andreas Rheinhardt
2023-10-27 18:38 ` Michael Niedermayer
2023-10-30 20:49 ` Andreas Rheinhardt
2023-10-31 0:25 ` Michael Niedermayer
2023-10-24 15:04 ` Michael Niedermayer [this message]
2023-10-24 15:04 ` [FFmpeg-devel] [PATCH 4/4] avcodec/magicyuv: Set UNCHECKED_BITSTREAM_READER Michael Niedermayer
2023-10-26 21:37 ` [FFmpeg-devel] [PATCH 1/4] avcodec/magicyuv: Use a compile time constant for vlc_bits Andreas Rheinhardt
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=20231024150443.7438-3-michael@niedermayer.cc \
--to=michael@niedermayer.cc \
--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