This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug string/21331] strcpy/strncpy suffering performance downgrade with short string copy from glibc2.15
- From: "mousuanming at huawei dot com" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sourceware dot org
- Date: Thu, 30 Mar 2017 06:57:55 +0000
- Subject: [Bug string/21331] strcpy/strncpy suffering performance downgrade with short string copy from glibc2.15
- Auto-submitted: auto-generated
- References: <bug-21331-131@http.sourceware.org/bugzilla/>
https://sourceware.org/bugzilla/show_bug.cgi?id=21331
--- Comment #1 from Mou <mousuanming at huawei dot com> ---
Update the libMicro test result.
The orginal test result:
[root@localhost bin-x86_64]# taskset -c 4-7 ./strcpy -E -C 200 -L -S -W -N
strcpy_10 -s 10 -I 5
Running: strcpy_10# ./strcpy -E -C 200 -L -S -W -N strcpy_10 -s 10 -I
5
prc thr usecs/call samples errors cnt/samp size
strcpy_10 1 1 0.00590 200 0 20000 10
#
# STATISTICS usecs/call (raw) usecs/call (outliers removed)
# min 0.00580 0.00580
# max 0.00719 0.00605
# mean 0.00590 0.00590
# median 0.00590 0.00590
# stddev 0.00011 0.00005
# standard error 0.00001 0.00000
# 99% confidence level 0.00002 0.00001
# skew 8.87776 0.43456
# kurtosis 104.59876 0.01184
# time correlation -0.00000 0.00000
#
# elasped time 0.02455
# number of samples 200
# number of outliers 2
# getnsecs overhead 0
#
# DISTRIBUTION
# counts usecs/call means
# 190 0.00000 |******************************** 0.00589
#
# 10 > 95% |* 0.00602
#
# mean of 95% 0.00589
# 95th %ile 0.00600
for 0.05098 seconds
And the ‘SourceStringAlignmentLess32’ ignored test result:
[root@localhost bin-x86_64]# taskset -c 4-7 ./strcpy -E -C 200 -L -S -W -N
strcpy_10 -s 10 -I 5
Running: strcpy_10# ./strcpy -E -C 200 -L -S -W -N strcpy_10 -s 10 -I
5
prc thr usecs/call samples errors cnt/samp size
strcpy_10 1 1 0.00490 199 0 20000 10
#
# STATISTICS usecs/call (raw) usecs/call (outliers removed)
# min 0.00480 0.00480
# max 0.00535 0.00506
# mean 0.00491 0.00491
# median 0.00490 0.00490
# stddev 0.00006 0.00005
# standard error 0.00000 0.00000
# 99% confidence level 0.00001 0.00001
# skew 2.53638 0.92894
# kurtosis 12.13214 0.80037
# time correlation -0.00000 0.00000
#
# elasped time 0.02056
# number of samples 199
# number of outliers 3
# getnsecs overhead 0
#
# DISTRIBUTION
# counts usecs/call means
# 189 0.00000 |******************************** 0.00490
#
# 10 > 95% |* 0.00504
#
# mean of 95% 0.00490
# 95th %ile 0.00500
for 0.02911 seconds
--
You are receiving this mail because:
You are on the CC list for the bug.