[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lmi-commits] [lmi] master d3d0a775 13/13: Record speed measurements
From: |
Greg Chicares |
Subject: |
[lmi-commits] [lmi] master d3d0a775 13/13: Record speed measurements |
Date: |
Mon, 21 Nov 2022 17:29:19 -0500 (EST) |
branch: master
commit d3d0a7757cccfe9b27443830b09635a6b6d56bb9
Author: Gregory W. Chicares <gchicares@sbcglobal.net>
Commit: Gregory W. Chicares <gchicares@sbcglobal.net>
Record speed measurements
Compared to commit b830f9a0: gcc speed is substantially the same for
both architectures, as expected because gcc builds haven't changed; and
clang is perhaps one percent faster for ill reg and three percent faster
for finra, which would suggest that its C++ standard library performs
slightly better in these particular tests.
---
Speed_clang_x86_64-pc-linux-gnu | 12 ++++++------
Speed_gcc_x86_64-pc-linux-gnu | 12 ++++++------
Speed_gcc_x86_64-w64-mingw32 | 12 ++++++------
3 files changed, 18 insertions(+), 18 deletions(-)
diff --git a/Speed_clang_x86_64-pc-linux-gnu b/Speed_clang_x86_64-pc-linux-gnu
index d5df79a2..1ca2f17c 100644
--- a/Speed_clang_x86_64-pc-linux-gnu
+++ b/Speed_clang_x86_64-pc-linux-gnu
@@ -1,7 +1,7 @@
Test speed:
- naic, no solve : 1.435e-02 s mean; 14049 us least of 70 runs
- naic, specamt solve : 2.962e-02 s mean; 28860 us least of 34 runs
- naic, ee prem solve : 2.621e-02 s mean; 25673 us least of 39 runs
- finra, no solve : 5.405e-03 s mean; 5269 us least of 100 runs
- finra, specamt solve: 1.943e-02 s mean; 19042 us least of 52 runs
- finra, ee prem solve: 1.720e-02 s mean; 16920 us least of 59 runs
+ naic, no solve : 1.496e-02 s mean; 13939 us least of 67 runs
+ naic, specamt solve : 2.892e-02 s mean; 28629 us least of 35 runs
+ naic, ee prem solve : 2.553e-02 s mean; 25179 us least of 40 runs
+ finra, no solve : 5.210e-03 s mean; 5087 us least of 100 runs
+ finra, specamt solve: 1.880e-02 s mean; 18506 us least of 54 runs
+ finra, ee prem solve: 1.676e-02 s mean; 16428 us least of 60 runs
diff --git a/Speed_gcc_x86_64-pc-linux-gnu b/Speed_gcc_x86_64-pc-linux-gnu
index de9f1592..6d1e3076 100644
--- a/Speed_gcc_x86_64-pc-linux-gnu
+++ b/Speed_gcc_x86_64-pc-linux-gnu
@@ -1,7 +1,7 @@
Test speed:
- naic, no solve : 1.313e-02 s mean; 12783 us least of 77 runs
- naic, specamt solve : 3.233e-02 s mean; 31723 us least of 31 runs
- naic, ee prem solve : 2.846e-02 s mean; 27760 us least of 36 runs
- finra, no solve : 5.096e-03 s mean; 4910 us least of 100 runs
- finra, specamt solve: 2.314e-02 s mean; 22790 us least of 44 runs
- finra, ee prem solve: 2.037e-02 s mean; 20046 us least of 50 runs
+ naic, no solve : 1.295e-02 s mean; 12778 us least of 78 runs
+ naic, specamt solve : 3.203e-02 s mean; 31643 us least of 32 runs
+ naic, ee prem solve : 2.799e-02 s mean; 27708 us least of 36 runs
+ finra, no solve : 5.028e-03 s mean; 4862 us least of 100 runs
+ finra, specamt solve: 2.290e-02 s mean; 22725 us least of 44 runs
+ finra, ee prem solve: 2.021e-02 s mean; 19933 us least of 50 runs
diff --git a/Speed_gcc_x86_64-w64-mingw32 b/Speed_gcc_x86_64-w64-mingw32
index 1ab70fe5..e8619a93 100644
--- a/Speed_gcc_x86_64-w64-mingw32
+++ b/Speed_gcc_x86_64-w64-mingw32
@@ -1,7 +1,7 @@
Test speed:
- naic, no solve : 2.229e-02 s mean; 21606 us least of 45 runs
- naic, specamt solve : 4.147e-02 s mean; 40859 us least of 25 runs
- naic, ee prem solve : 3.713e-02 s mean; 36468 us least of 27 runs
- finra, no solve : 1.054e-02 s mean; 10168 us least of 95 runs
- finra, specamt solve: 3.005e-02 s mean; 28402 us least of 34 runs
- finra, ee prem solve: 2.616e-02 s mean; 25661 us least of 39 runs
+ naic, no solve : 2.213e-02 s mean; 20995 us least of 46 runs
+ naic, specamt solve : 4.184e-02 s mean; 41330 us least of 24 runs
+ naic, ee prem solve : 3.733e-02 s mean; 37024 us least of 27 runs
+ finra, no solve : 1.053e-02 s mean; 10256 us least of 95 runs
+ finra, specamt solve: 2.839e-02 s mean; 28117 us least of 36 runs
+ finra, ee prem solve: 2.572e-02 s mean; 25400 us least of 39 runs
- [lmi-commits] [lmi] master 7c47d71b 03/13: Perform speed test for all toolchains, (continued)
- [lmi-commits] [lmi] master 7c47d71b 03/13: Perform speed test for all toolchains, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 55f92d87 02/13: Record gcc version twelve pc-linux-gnu speed for commit 5a4fcaed7, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master c6528973 06/13: Specify all '*FLAGS' in one place, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 93e400db 10/13: Avoid literal '***' in a 'set -vx' script, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 8cb31fd7 01/13: Record gcc version eleven pc-linux-gnu speed for commit 5a4fcaed7, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 1a7a1e59 12/13: Use libc++ rather than libstdc++ with clang, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master b830f9a0 04/13: Record speed measurements, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 4b6b2b5c 05/13: Split long lines that will soon become longer, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master d6c7ec7a 08/13: Build and install libraries for all toolchains, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master 1e6d03ee 09/13: Print log messages before and after building, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master d3d0a775 13/13: Record speed measurements,
Greg Chicares <=
- [lmi-commits] [lmi] master cc85ea5a 11/13: Build libraries with specified toolchain, Greg Chicares, 2022/11/21
- [lmi-commits] [lmi] master a60c106c 07/13: Use clang's preprocessor, Greg Chicares, 2022/11/21