Skip to content

gourryinverse/interruptor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Usage

edit: This was all due to a cpu erratum, derp derp!

This driver produces unexpected results on E5-2650, i7-6700.

Ubuntu with linux kernels 4.6.4, 4.8.0, and 4.10.x. Most testing done on Ubuntu 4.6.4.

Expected: Always 0

E5-2650 Result: 0, 1, 2, ..., up to 19 and occasionally higher

i7-6700 Result: 0, 118, 119

E5-1650 Result: 0

When correct, the driver will not produce any messages in dmesg.

When the branch count is > 0, it will produce a message "interruptor: %d" where %d is the non-zero value detected.

Building

make

Inserting the driver

sudo insmod interruptor.ko

Generating incorrect results

Launch tester and rtcspikers for each logical cpu.

sudo su

./tester &

./tester &

./tester &

./tester &

./tester &

./tester &

./tester &

./tester &

./rtcspiker &

./rtcspiker &

./rtcspiker &

./rtcspiker &

./rtcspiker &

./rtcspiker &

./rtcspiker &

./rtcspiker &

Results

Expected: Always 0, no output in dmesg.

E5-2650: common 1s and 2s - dmesg output "interruptor: 1"

i7-6700: occasional 118s and 119s - dmesg output "interruptor: 118"

About

Branch Counting driver producing errors

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published