Bug 430210

Summary: Benchmarks for b43 - BCM4318
Product: [Fedora] Fedora Reporter: Jonathan Steffan <jonathansteffan>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: low    
Version: 8   
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2008-01-25 15:03:19 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Jonathan Steffan 2008-01-25 10:01:56 UTC
Description of problem:
Just some benchmarking data. I'm sick of my card performance sucking.

[jon@damaestrojr ~]$ dd if=/dev/zero of=/home/jon/test.img bs=1024 count=1048576
1048576+0 records in
1048576+0 records out
1073741824 bytes (1.1 GB) copied, 87.1328 s, 12.3 MB/s

Wired: kernel-2.6.23.9-85.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x: 
test.img                                      100% 1024MB  10.8MB/s   01:35    

real    1m38.579s
user    0m18.033s
sys     0m17.340s

Wireless: kernel-2.6.23.9-85.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x: 
test.img                                           1%   10MB  88.5KB/s 3:15:29 ETA
Killed by signal 2.

real    2m7.620s
user    0m0.839s
sys     0m0.291s

Wired: 2.6.23.14-107.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x:
test.img                                      100% 1024MB  10.1MB/s   01:41    

real    1m41.808s
user    0m18.321s
sys     0m17.643s


Wireless: 2.6.23.14-107.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x: 
test.img                                       36%  374MB   1.6MB/s   06:52 ETA
Killed by signal 2.

real    4m7.868s
user    0m14.910s
sys     0m14.681s

Wired: 2.6.23.14-115.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x:
test.img                                      100% 1024MB  10.7MB/s   01:36    

real    1m36.066s
user    0m18.144s
sys     0m18.193s

Wireless: 2.6.23.14-115.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x:
test.img                                        1%   11MB  93.1KB/s 3:05:43 ETA
Killed by signal 2.

real    2m4.585s
user    0m0.619s
sys     0m0.310s

2.6.23.14-107 seems to give the best performance, but 1.6MB/s still sucks.

Comment 1 John W. Linville 2008-01-25 15:03:19 UTC

*** This bug has been marked as a duplicate of 413291 ***

Comment 2 Jonathan Steffan 2008-01-25 19:41:03 UTC
Hmmm.

2.6.23.14-107 is not working with 802.11b APs, but works great with 802.11g APs.

Comment 3 Jonathan Steffan 2008-01-25 19:41:54 UTC
05:02.0 Network controller: Broadcom Corporation BCM4318 [AirForce One 54g]
802.11g Wireless LAN Controller (rev 02)

Comment 4 Jonathan Steffan 2008-02-13 07:28:08 UTC
Wired: 2.6.23.15-137.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.12:
test.img                                      100% 1024MB  10.9MB/s   01:34    

real    1m40.475s
user    0m18.532s
sys     0m17.693s

Wireless: 2.6.23.15-137.fc8
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x:
test.img                                       27%  283MB   1.3MB/s   09:37 ETA
Killed by signal 2.

real    3m42.484s
user    0m10.191s
sys     0m8.585s


Comment 5 Levente Farkas 2008-02-17 18:38:12 UTC
the same speed problem for me with kernel-2.6.23.15-137.fc8

Comment 6 Jonathan Steffan 2008-02-22 07:50:44 UTC
Wireless: 2.6.23.15-137.fc8
This is with using the latest recommended firmware and b43-fwcutter-011
[jon@damaestrojr ~]$ time scp test.img 192.168.0.x:
test.img                                      100% 1000MB   1.4MB/s   11:50    

real    11m57.563s
user    0m33.097s
sys     0m26.788s


Comment 7 Jonathan Steffan 2008-04-10 02:27:53 UTC
Wireless: 2.6.24.4-64.fc8
[jon@damaestrojr ~]$ scp test.img 192.168.0.x:
test.img                                      100% 1000MB   1.9MB/s   08:54

Good work, performance has been improving/staying steady.