From: "Rémi Denis-Courmont" <remi@remlab.net>
To: ffmpeg-devel@ffmpeg.org
Subject: [FFmpeg-devel] [PATCH 2/3] lavu/riscv: add ff_rv_vlen_least()
Date: Fri, 10 May 2024 19:39:42 +0300
Message-ID: <20240510163943.109471-2-remi@remlab.net> (raw)
In-Reply-To: <20240510163943.109471-1-remi@remlab.net>
This inline function checks that the vector length is at least a given
value. With this, most run-time VLEN checks can be optimised away.
---
libavutil/riscv/cpu.h | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/libavutil/riscv/cpu.h b/libavutil/riscv/cpu.h
index 56035f8556..af1440f626 100644
--- a/libavutil/riscv/cpu.h
+++ b/libavutil/riscv/cpu.h
@@ -22,6 +22,7 @@
#define AVUTIL_RISCV_CPU_H
#include "config.h"
+#include <stdbool.h>
#include <stddef.h>
#include "libavutil/cpu.h"
@@ -42,4 +43,24 @@ static inline size_t ff_get_rv_vlenb(void)
return vlenb;
}
#endif
+
+/**
+ * Checks that the vector bit-size is at least the given value.
+ * This is potentially undefined behaviour if vectors are not implemented.
+ */
+static inline bool ff_rv_vlen_least(unsigned int bits)
+{
+#ifdef __riscv_v_min_vlen
+ if (bits <= __riscv_min_vlen)
+ return true;
+#else
+ /*
+ * Vector lengths smaller than 128 bits are only possible in embedded cases
+ * and cannot be run-time detected, so we can assume 128 bits at least.
+ */
+ if (bits <= 128)
+ return true;
+#endif
+ return bits <= (8 * ff_get_rv_vlenb());
+}
#endif
--
2.43.0
_______________________________________________
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:[~2024-05-10 16:40 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-10 16:39 [FFmpeg-devel] [PATCH 1/3] lavc/vp9dsp: fix indentation Rémi Denis-Courmont
2024-05-10 16:39 ` Rémi Denis-Courmont [this message]
2024-05-10 16:39 ` [FFmpeg-devel] [PATCH 3/3] lavc/riscv: use ff_rv_vlen_least() Rémi Denis-Courmont
2024-05-11 10:11 ` [FFmpeg-devel] [PATCH 1/3] lavc/vp9dsp: fix indentation flow gg
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=20240510163943.109471-2-remi@remlab.net \
--to=remi@remlab.net \
--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