Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Niklas Haas <ffmpeg@haasn.xyz>
To: ffmpeg-devel@ffmpeg.org
Cc: Niklas Haas <git@haasn.dev>
Subject: [FFmpeg-devel] [PATCH 12/17] tests/checkasm: increase number of runs in between measurements
Date: Sat, 26 Apr 2025 19:41:16 +0200
Message-ID: <20250426175603.726924-13-ffmpeg@haasn.xyz> (raw)
In-Reply-To: <20250426175603.726924-1-ffmpeg@haasn.xyz>

From: Niklas Haas <git@haasn.dev>

Sometimes, when measuring very small functions, rdtsc is not accurate enough
to get a reliable measurement. This increases the number of runs inside the
inner loop from 4 to 32, which should help a lot. Less important when using
the more precise linux-perf API, but still useful.

There should be no user-visible change since the number of runs is adjusted
to keep the total time spent measuring the same.
---
 tests/checkasm/checkasm.c |  2 +-
 tests/checkasm/checkasm.h | 24 +++++++++++++++++++-----
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
index 412b8b2cd1..87b75ec36c 100644
--- a/tests/checkasm/checkasm.c
+++ b/tests/checkasm/checkasm.c
@@ -624,7 +624,7 @@ static inline double avg_cycles_per_call(const CheckasmPerf *const p)
     if (p->iterations) {
         const double cycles = (double)(10 * p->cycles) / p->iterations - state.nop_time;
         if (cycles > 0.0)
-            return cycles / 4.0; /* 4 calls per iteration */
+            return cycles / 32.0; /* 32 calls per iteration */
     }
     return 0.0;
 }
diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
index ad239fb2a4..215d64e076 100644
--- a/tests/checkasm/checkasm.h
+++ b/tests/checkasm/checkasm.h
@@ -340,6 +340,22 @@ typedef struct CheckasmPerf {
 #define PERF_STOP(t)  t = AV_READ_TIME() - t
 #endif
 
+#define CALL4(...)\
+    do {\
+        tfunc(__VA_ARGS__); \
+        tfunc(__VA_ARGS__); \
+        tfunc(__VA_ARGS__); \
+        tfunc(__VA_ARGS__); \
+    } while (0)
+
+#define CALL16(...)\
+    do {\
+        CALL4(__VA_ARGS__); \
+        CALL4(__VA_ARGS__); \
+        CALL4(__VA_ARGS__); \
+        CALL4(__VA_ARGS__); \
+    } while (0)
+
 /* Benchmark the function */
 #define bench_new(...)\
     do {\
@@ -350,14 +366,12 @@ typedef struct CheckasmPerf {
             uint64_t tsum = 0;\
             uint64_t ti, tcount = 0;\
             uint64_t t = 0; \
-            const uint64_t truns = bench_runs;\
+            const uint64_t truns = FFMAX(bench_runs >> 3, 1);\
             checkasm_set_signal_handler_state(1);\
             for (ti = 0; ti < truns; ti++) {\
                 PERF_START(t);\
-                tfunc(__VA_ARGS__);\
-                tfunc(__VA_ARGS__);\
-                tfunc(__VA_ARGS__);\
-                tfunc(__VA_ARGS__);\
+                CALL16(__VA_ARGS__);\
+                CALL16(__VA_ARGS__);\
                 PERF_STOP(t);\
                 if (t*tcount <= tsum*4 && ti > 0) {\
                     tsum += t;\
-- 
2.49.0

_______________________________________________
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".

  parent reply	other threads:[~2025-04-26 17:57 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-26 17:41 [FFmpeg-devel] [PATCH 00/17] swscale v2: new framework [RFC] Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 01/17] tests/swscale: improve colorization of speedup Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 02/17] swscale/graph: expose ff_sws_graph_add_pass Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 03/17] swscale/graph: make noop loop more robust Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 04/17] swscale/graph: move vshift() and shift_img() to shared header Niklas Haas
2025-05-16 15:41   ` Ramiro Polla
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 05/17] swscale/graph: prefer bools to ints Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 06/17] doc: add swscale rewrite design document Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 07/17] swscale: add SWS_EXPERIMENTAL flag Niklas Haas
2025-05-08 11:37   ` Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 08/17] swscale/ops: introduce new low level framework Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 09/17] swscale/ops_chain: add internal abstraction for kernel linking Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 10/17] swscale/ops_backend: add reference backend basend on C templates Niklas Haas
2025-05-02 15:06   ` Michael Niedermayer
2025-05-08 12:24     ` Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 11/17] swscale/x86: add SIMD backend Niklas Haas
2025-04-29 13:00   ` Michael Niedermayer
2025-04-30 16:24     ` Niklas Haas
2025-04-26 17:41 ` Niklas Haas [this message]
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 13/17] tests/checkasm: add checkasm_check_float Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 14/17] tests/checkasm: add checkasm tests for swscale ops Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 15/17] swscale/format: rename legacy format conversion table Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 16/17] swscale/format: add new format decode/encode logic Niklas Haas
2025-05-02 14:10   ` Michael Niedermayer
2025-05-02 14:36     ` Niklas Haas
2025-04-26 17:41 ` [FFmpeg-devel] [PATCH 17/17] swscale/graph: allow experimental use of new format handler Niklas Haas
2025-04-26 22:22 ` [FFmpeg-devel] [PATCH 00/17] swscale v2: new framework [RFC] Niklas Haas
2025-05-02 17:51 ` Niklas Haas
2025-05-16 11:09 ` Niklas Haas
2025-05-16 14:32   ` Ramiro Polla
2025-05-16 14:39     ` Niklas Haas
2025-05-16 15:44       ` Ramiro Polla

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=20250426175603.726924-13-ffmpeg@haasn.xyz \
    --to=ffmpeg@haasn.xyz \
    --cc=ffmpeg-devel@ffmpeg.org \
    --cc=git@haasn.dev \
    /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