1 |
On Sat, 2005-07-23 at 16:48 +0200, Simon Strandman wrote: |
2 |
> With the glibc 2.3.5 currently in gentoo I get: |
3 |
> isidor ~ # ./memcpy 2200 1000 1048576 |
4 |
> Memory to memory copy rate = 1291.600098 MBytes / sec. Block size = 1048576. |
5 |
> |
6 |
> But with glibc 2.3.5 + amd64 optimized strings I get: |
7 |
> isidor ~ # ./memcpy 2200 1000 1048576 |
8 |
> Memory to memory copy rate = 2389.321777 MBytes / sec. Block size = 1048576. |
9 |
> |
10 |
> That's an improvement of over 1000mb/s! Suse 9.3 also gives about |
11 |
> 2300mb/s out of the box. |
12 |
|
13 |
Hmm, I don't see much difference here. |
14 |
|
15 |
before: |
16 |
allan@aw ~ $ ./a.out 2420 1000 1048576 |
17 |
Memory to memory copy rate = 1110.198486 MBytes / sec. Block size = |
18 |
1048576. |
19 |
|
20 |
after: |
21 |
aw allan # ./a.out 2420 1000 1048576 |
22 |
Memory to memory copy rate = 1154.922974 MBytes / sec. Block size = |
23 |
1048576. |
24 |
|
25 |
Allan |
26 |
|
27 |
-- |
28 |
gentoo-amd64@g.o mailing list |