This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
在 星期四 15 十二月 2005 07:32,Martin Hunt 写道: > OK, now this is much different than what Sun got. There is certainly no > munmap() problem here. Without detail about Sun's emulation environment > and what version of linux they are running, I cannot even guess what the > problem is. Maybe one clue is this odd comment "[...]what I've heard my > colleague, Bryan, refer to as the "top problem": your system is slow, so > you run top. What's the top process? Top!" The result, Because some system call (such sys_select, sys_open, device relative system call) are may yield cpu for some reason, than go schedule, when this system call return ,the time calculate like your function may bigger than the real time. So we must minus the yield schedule time. I write one to calculate each process's system call time and user function time for comparison, But there are some bug, Hope for correct. Attachment 1: cmpcall.stp Attachment 2: result -Wei Kong
Attachment:
cmpcall.stp
Description: Text document
Attachment:
result
Description: Text document
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |