User-mode Linux kernel port / Patches / #21 SYSEMU Performance Patch for reducing context switches

SYSEMU Performance Patch for reducing context switches

#21 SYSEMU Performance Patch for reducing context switches

Status: open

Owner: nobody

Labels: None

Priority: 5

Updated: 2004-05-20

Created: 2004-05-20

Creator: Anonymous

Private: No

this patch reduces the context switch rate on your host
system and gives a performance benefit for uml.

The patch is similar to skas: it needs the UML _AND_
the HOST to be patched.

some tests for comparison

"GETPID LOOP" (worst case-/no "real world"-scenario)
------- getpid.c -------------
#include <stdio.h>

int main(int argc, char** argv)
{
int i;
int loop = atol(argv[1]);

for (i = 0; i < loop; i++)
{
getpid();
}
}
-----------------------------

time ./getpid 10000000

native speed on HOST:
----------------------
real 0m7.920s
user 0m3.990s
sys 0m3.940s

UML w/o sysemu:
-----------------
real 6m16.956s
user 0m5.250s
sys 0m58.462s

real 6m17.126s
user 0m4.846s
sys 1m1.731s

real 6m16.461s
user 0m5.077s
sys 0m59.365s

UML with sysemu:
-----------------
real 3m55.052s
user 0m5.962s
sys 0m46.385s

real 3m56.964s
user 0m5.846s
sys 0m42.635s

real 3m54.179s
user 0m5.981s
sys 0m47.154s
---------------------------------------

KERNEL COMPILE:
-----------------

UML w/o sysemu:
-----------------
time make dep > /dev/null

real 4m51.669s
user 0m37.462s
sys 0m51.846s

time make bzImage modules > /dev/null

real 13m35.457s
user 4m18.885s
sys 1m40.481s

UML with sysemu:
-----------------
time make dep > /dev/null

real 3m52.486s
user 0m37.981s
sys 0m50.808s

time make bzImage modules > /dev/null

real 13m5.980s
user 4m18.096s
sys 1m40.115s

(Tests have been made by Laurent with 2.4 Kernel)

This patch needs testing - please post further
performance comparisons (some graphs?) !
:)

Discussion