Jul. 11th, 2018

izard: (Default)
Aaaah! Another mistery, I am wasting a second work day on. As always, after I find the root cause, this should be the most stupid and embarrassing error I made.

Here it is: I am testing a linux kernel scheduler patch I developed that improves performance of customer's benchmark. It works, but sometimes it breaks another benchmark, hence there is a bug. OK, trying to reproduce the bug.

I boot original kernel - no bug. I boot modified kernel - it takes a while to reproduce a bug. I reboot, booting a modified kernel again - the bug is instantly reproducible. But there is no state maintained between reboots in the linux kernel!!!

Profile

izard: (Default)
izard

July 2025

S M T W T F S
  12345
67 8 91011 12
13141516171819
20212223242526
27 28293031  

Most Popular Tags

Page Summary

Style Credit

Expand Cut Tags

No cut tags
Page generated Aug. 13th, 2025 06:21 pm
Powered by Dreamwidth Studios