[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lmi-commits] [lmi] master eb70cdc 7/7: Prefer not to pass std::vector b
From: |
Greg Chicares |
Subject: |
[lmi-commits] [lmi] master eb70cdc 7/7: Prefer not to pass std::vector by value |
Date: |
Sat, 16 Jan 2021 04:07:20 -0500 (EST) |
branch: master
commit eb70cdc84c20ee2f48a151d5f37913198cf7dbaf
Author: Gregory W. Chicares <gchicares@sbcglobal.net>
Commit: Gregory W. Chicares <gchicares@sbcglobal.net>
Prefer not to pass std::vector by value
These changes might improve speed by one percent or so, but an effect
that small cannot be measured reliably.
---
Speed_gcc_i686-w64-mingw32 | 12 ++++++------
Speed_gcc_x86_64-pc-linux-gnu | 12 ++++++------
Speed_gcc_x86_64-w64-mingw32 | 12 ++++++------
dbo_rules.cpp | 2 +-
dbo_rules.hpp | 2 +-
report_table_test.cpp | 2 +-
round_to.hpp | 5 +++--
7 files changed, 24 insertions(+), 23 deletions(-)
diff --git a/Speed_gcc_i686-w64-mingw32 b/Speed_gcc_i686-w64-mingw32
index 6fc22f6..6b0f9cb 100644
--- a/Speed_gcc_i686-w64-mingw32
+++ b/Speed_gcc_i686-w64-mingw32
@@ -1,7 +1,7 @@
Test speed:
- naic, no solve : 5.307e-02 s mean; 52248 us least of 19 runs
- naic, specamt solve : 9.451e-02 s mean; 92868 us least of 11 runs
- naic, ee prem solve : 8.641e-02 s mean; 85247 us least of 12 runs
- finra, no solve : 2.213e-02 s mean; 21762 us least of 46 runs
- finra, specamt solve: 5.981e-02 s mean; 59198 us least of 17 runs
- finra, ee prem solve: 5.614e-02 s mean; 55063 us least of 18 runs
+ naic, no solve : 5.456e-02 s mean; 51931 us least of 19 runs
+ naic, specamt solve : 9.267e-02 s mean; 92286 us least of 11 runs
+ naic, ee prem solve : 8.516e-02 s mean; 84830 us least of 12 runs
+ finra, no solve : 2.183e-02 s mean; 21703 us least of 46 runs
+ finra, specamt solve: 5.913e-02 s mean; 58925 us least of 17 runs
+ finra, ee prem solve: 5.501e-02 s mean; 54808 us least of 19 runs
diff --git a/Speed_gcc_x86_64-pc-linux-gnu b/Speed_gcc_x86_64-pc-linux-gnu
index c345d58..dad77b7 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 : 1.965e-02 s mean; 18957 us least of 51 runs
- naic, specamt solve : 3.346e-02 s mean; 32667 us least of 30 runs
- naic, ee prem solve : 3.081e-02 s mean; 30138 us least of 33 runs
- finra, no solve : 7.876e-03 s mean; 7616 us least of 100 runs
- finra, specamt solve: 2.083e-02 s mean; 20463 us least of 48 runs
- finra, ee prem solve: 1.949e-02 s mean; 18879 us least of 52 runs
+ naic, no solve : 1.918e-02 s mean; 18753 us least of 53 runs
+ naic, specamt solve : 3.312e-02 s mean; 32434 us least of 31 runs
+ naic, ee prem solve : 3.046e-02 s mean; 30031 us least of 33 runs
+ finra, no solve : 7.629e-03 s mean; 7377 us least of 100 runs
+ finra, specamt solve: 2.056e-02 s mean; 20093 us least of 49 runs
+ finra, ee prem solve: 1.913e-02 s mean; 18751 us least of 53 runs
diff --git a/Speed_gcc_x86_64-w64-mingw32 b/Speed_gcc_x86_64-w64-mingw32
index 00ccc0f..139c622 100644
--- a/Speed_gcc_x86_64-w64-mingw32
+++ b/Speed_gcc_x86_64-w64-mingw32
@@ -1,7 +1,7 @@
Test speed:
- naic, no solve : 2.779e-02 s mean; 26362 us least of 36 runs
- naic, specamt solve : 4.214e-02 s mean; 41570 us least of 24 runs
- naic, ee prem solve : 3.924e-02 s mean; 38670 us least of 26 runs
- finra, no solve : 1.489e-02 s mean; 14589 us least of 68 runs
- finra, specamt solve: 2.901e-02 s mean; 28609 us least of 35 runs
- finra, ee prem solve: 2.750e-02 s mean; 27122 us least of 37 runs
+ naic, no solve : 2.674e-02 s mean; 26388 us least of 38 runs
+ naic, specamt solve : 4.206e-02 s mean; 41541 us least of 24 runs
+ naic, ee prem solve : 3.916e-02 s mean; 38679 us least of 26 runs
+ finra, no solve : 1.476e-02 s mean; 14484 us least of 68 runs
+ finra, specamt solve: 2.882e-02 s mean; 28440 us least of 35 runs
+ finra, ee prem solve: 2.728e-02 s mean; 26912 us least of 37 runs
diff --git a/dbo_rules.cpp b/dbo_rules.cpp
index 84be0b2..0fa52f0 100644
--- a/dbo_rules.cpp
+++ b/dbo_rules.cpp
@@ -96,7 +96,7 @@ bool dbo_transition_is_allowed(mce_dbopt from, mce_dbopt to)
// The return value is fairly useless, and may later be removed.
-bool dbo_sequence_is_allowed(std::vector<mce_dbopt> v)
+bool dbo_sequence_is_allowed(std::vector<mce_dbopt> const& v)
{
if(0 == lmi::ssize(v))
{
diff --git a/dbo_rules.hpp b/dbo_rules.hpp
index 205767d..4206326 100644
--- a/dbo_rules.hpp
+++ b/dbo_rules.hpp
@@ -32,6 +32,6 @@
bool dbo_at_issue_is_allowed(mce_dbopt);
bool dbo_transition_is_allowed(mce_dbopt from, mce_dbopt to);
-bool dbo_sequence_is_allowed(std::vector<mce_dbopt>);
+bool dbo_sequence_is_allowed(std::vector<mce_dbopt> const&);
#endif // dbo_rules_hpp
diff --git a/report_table_test.cpp b/report_table_test.cpp
index b10b3c8..3beadbb 100644
--- a/report_table_test.cpp
+++ b/report_table_test.cpp
@@ -69,7 +69,7 @@ std::vector<table_column_info> bloat
return v;
}
-int sum(std::vector<int> z)
+int sum(std::vector<int> const& z)
{
return std::accumulate(z.begin(), z.end(), 0);
}
diff --git a/round_to.hpp b/round_to.hpp
index 2338e24..4d551a7 100644
--- a/round_to.hpp
+++ b/round_to.hpp
@@ -264,7 +264,7 @@ class round_to
bool operator==(round_to const&) const;
RealType operator()(RealType r) const;
- std::vector<RealType> operator()(std::vector<RealType> r) const;
+ std::vector<RealType> operator()(std::vector<RealType> const& r) const;
int decimals() const;
rounding_style style() const;
@@ -361,7 +361,8 @@ inline RealType round_to<RealType>::operator()(RealType r)
const
}
template<typename RealType>
-inline std::vector<RealType>
round_to<RealType>::operator()(std::vector<RealType> r) const
+inline std::vector<RealType> round_to<RealType>::operator()
+ (std::vector<RealType> const& r) const
{
std::vector<RealType> z;
z.reserve(r.size());
- [lmi-commits] [lmi] master updated (f4619be -> eb70cdc), Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master 400f752 2/7: Do away with unused return value #1, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master a73a260 6/7: Do away with unused return value #5, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master 997d4d0 5/7: Do away with unused return value #4, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master 8ee9eaf 1/7: Don't name unused template parameters, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master 78d6d35 4/7: Do away with unused return value #3, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master bad394a 3/7: Do away with unused return value #2, Greg Chicares, 2021/01/16
- [lmi-commits] [lmi] master eb70cdc 7/7: Prefer not to pass std::vector by value,
Greg Chicares <=