Mirrors/perf-little-book
1
0
Fork 0
mirror of https://github.com/NanXiao/perf-little-book.git synced 2026-07-17 16:35:13 +00:00
Code Issues Projects Releases Wiki Activity

Merge pull request #1 from justanotherdot/topic/fix-typo

Browse source 
Fix typo on DEFINE in sample code
This commit is contained in:
Nan Xiao 2020-09-02 10:35:04 +08:00 • committed by GitHub
commit c25de25edf
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

12
posts/record-event-counts-per-thread.md
View file

@ -1,22 +1,22 @@
# Record event counts per thread # Record event counts per thread
`perf record` can be used to record event counts for every thread when using `-s` option. Check following multi-thread program: `perf record` can be used to record event counts for every thread when using `-s` option. Check following multi-thread program:
``` ```
$ cat sum.c $ cat sum.c
#include <omp.h> #include <omp.h>
#define N 100000000 #define N 100000000
#define THRAED_NUM 8 #define THREAD_NUM 8
int values[N]; int values[N];
int main(void) int main(void)
{ {
int sum[THRAED_NUM]; int sum[THREAD_NUM];
#pragma omp parallel for #pragma omp parallel for
for (int i = 0; i < THRAED_NUM; i++) for (int i = 0; i < THREAD_NUM; i++)
{ {
int local_sum; int local_sum;
for (int j = 0; j < N; j++) for (int j = 0; j < N; j++)
@ -30,7 +30,7 @@ int main(void)
} }
``` ```
Build and use `perf record` to note down event counts for every thread: Build and use `perf record` to note down event counts for every thread:
``` ```
# gcc -fopenmp -g sum.c -o sum # gcc -fopenmp -g sum.c -o sum
@ -39,7 +39,7 @@ Build and use `perf record` to note down event counts for every thread:
[ perf record: Captured and wrote 0.404 MB perf.data (8757 samples) ] [ perf record: Captured and wrote 0.404 MB perf.data (8757 samples) ]
``` ```
Use `perf report` to analyze it (`-T` option means displaying per-thread event counts): Use `perf report` to analyze it (`-T` option means displaying per-thread event counts):
``` ```
# perf report -T # perf report -T