Excerpt from a Linux-wireless discussion (this is Linux kernel developer's list for wireless interfaces)
They asked me to do some tests: the major one was MMIO tracing the driver:
> You need to enable MMIO tracing support in your kernel config, and
> recompile your kernel (assuming it is not enabled already - distros
> usually disable the entire tracing framework by default). Then:
> -Mount debugfs to /debug. (Usually it is mounted by default, but some
> distros like to mount it @ /sys/kernel/debug instead - remount it to
> /debug.)
> -"echo mmiotrace > /debug/tracing/current_tracer"
> -"cat /debug/tracing/trace_pipe > trace.txt &" - at this point, tracing begins.
> -Load r8187b and connect.
> -"echo "Driver loaded!" > /debug/tracing/trace_marker"
> -"echo nop
- as seno said here on the list in 14-05-2010 17:20:02
It installs a module called r8187b, witch I managed to use ONCE, after
it installed, then after it disconnected it would not connect again, and
after I rebooted there was no more wi-fi interface (!!!) ifconfig and
iwconfig confirm that my wi-fi card (wlan0) is nowhere to be found
5) Rate seems weird on this card. If I define the rate with
"iwconfig wlan0 rate 11M" it gives the best download speed, even if I
rase this to 24M or 54M I get at least a 40% difference, same if I lower
the rate.
Maybe what Hin-Tak Leung (2010-04-03) said about the Minstrel algorithm?
Could we try to install both Minstrel and older algorithms side by side
and make a comparison? from what I read here:
Excerpt from a Linux-wireless discussion (this is Linux kernel developer's list for wireless interfaces)
They asked me to do some tests: the major one was MMIO tracing the driver:
> You need to enable MMIO tracing support in your kernel config, and tracing/ current_ tracer" tracing/ trace_pipe > trace.txt &" - at this point, tracing begins. tracing/ trace_marker"
> recompile your kernel (assuming it is not enabled already - distros
> usually disable the entire tracing framework by default). Then:
> -Mount debugfs to /debug. (Usually it is mounted by default, but some
> distros like to mount it @ /sys/kernel/debug instead - remount it to
> /debug.)
> -"echo mmiotrace > /debug/
> -"cat /debug/
> -Load r8187b and connect.
> -"echo "Driver loaded!" > /debug/
> -"echo nop
Ok here are some news:
1) The rtl8187b problem is a know bug in Ubuntu since forever - bug /bugs.launchpad .net/ubuntu/ +source/ linux/+ bug/215802
#215802 https:/
2) Tryied the Realtek driver found in: ubuntuforums. org/showthread. php?t=1446894 linux_26. 1056.1112. 2009.release. tar.gz )
http://
(rtl8187B_
- as seno said here on the list in 14-05-2010 17:20:02
It installs a module called r8187b, witch I managed to use ONCE, after
it installed, then after it disconnected it would not connect again, and
after I rebooted there was no more wi-fi interface (!!!) ifconfig and
iwconfig confirm that my wi-fi card (wlan0) is nowhere to be found
5) Rate seems weird on this card. If I define the rate with
"iwconfig wlan0 rate 11M" it gives the best download speed, even if I
rase this to 24M or 54M I get at least a 40% difference, same if I lower
the rate.
Maybe what Hin-Tak Leung (2010-04-03) said about the Minstrel algorithm?
Could we try to install both Minstrel and older algorithms side by side
and make a comparison? from what I read here:
http:// wireless. kernel. org/en/ developers/ Documentation/ mac80211/ RateControl/ minstrel
"Further, it means that the 1 Mbps rate (which typically has a high
probability of successful transmission) will not be used in preference
to the 11 Mbps rate."
But my rate is always a defaulted 1Mb/s on bootup. I put it to 11M
manually and the connection improves dramatically
4) Ran the mmiotrace on the rtl8187 driver, it all seems gibberish to
me ;)
one while it was at 1m from the router giving me top speed (same as
Win7): trace1.txt:
VERSION 20070824
PCIDEV 0000 10390671 0 d0000000 0 0 0 0 0 0 8000000 0 0 0 0 0 0
agpgart-sis
PCIDEV 0008 10390003 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PCIDEV 0010 10390968 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PCIDEV 0015 10395513 10 1f0 3f6 170 376 1081 0 0 8 0 8 0 10 0 0 pata_sis
PCIDEV 0018 10397001 14 d8404000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ohci_hcd
PCIDEV 0019 10397001 15 d8405000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ohci_hcd
PCIDEV 001b 10397002 16 d8406000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ehci_hcd
PCIDEV 0020 10390191 13 d8407000 1001 0 0 0 0 0 80 80 0 0 0 0 0 sis190
PCIDEV 0028 10391183 11 10c9 10bd 10c1 10b9 10a1 0 0 8 4 8 4 10 0 0
sata_sis
PCIDEV 0030 1039000a 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0038 1039000a 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0078 10397502 12 d8400000 0 0 0 0 0 0 4000 0 0 0 0 0 0 HDA Intel
PCIDEV 00f8 10390004 1a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0100 10396351 10 c0000008 d8000000 9001 0 0 0 2 10000000 20000 80
0 0 0 0
PCIDEV 0300 197b2382 10 d8100000 0 0 0 0 0 d8500000 100 0 0 0 0 0 10000
sdhci-pci
PCIDEV 0302 197b2381 10 d8100400 0 0 0 0 0 0 100 0 0 0 0 0 0
PCIDEV 0303 197b2383 10 d8100800 0 0 0 0 0 0 100 0 0 0 0 0 0 jmb38x_ms
** then when it was about 10m - failed to connect (Windows 7 connects
normally) trace2.txt
VERSION 20070824
PCIDEV 0000 10390671 0 d0000000 0 0 0 0 0 0 8000000 0 0 0 0 0 0
agpgart-sis
PCIDEV 0008 10390003 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PCIDEV 0010 10390968 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PCIDEV 0015 10395513 10 1f0 3f6 170 376 1081 0 0 8 0 8 0 10 0 0 pata_sis
PCIDEV 0018 10397001 14 d8404000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ohci_hcd
PCIDEV 0019 10397001 15 d8405000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ohci_hcd
PCIDEV 001b 10397002 16 d8406000 0 0 0 0 0 0 1000 0 0 0 0 0 0 ehci_hcd
PCIDEV 0020 10390191 13 d8407000 1001 0 0 0 0 0 80 80 0 0 0 0 0 sis190
PCIDEV 0028 10391183 11 10c9 10bd 10c1 10b9 10a1 0 0 8 4 8 4 10 0 0
sata_sis
PCIDEV 0030 1039000a 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0038 1039000a 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0078 10397502 12 d8400000 0 0 0 0 0 0 4000 0 0 0 0 0 0 HDA Intel
PCIDEV 00f8 10390004 1a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pcieport
PCIDEV 0100 10396351 10 c0000008 d8000000 9001 0 0 0 2 10000000 20000 80
0 0 0 0
PCIDEV 0300 197b2382 10 d8100000 0 0 0 0 0 d8500000 100 0 0 0 0 0 10000
sdhci-pci
PCIDEV 0302 197b2381 10 d8100400 0 0 0 0 0 0 100 0 0 0 0 0 0
PCIDEV 0303 197b2383 10 d8100800 0 0 0 0 0 0 100 0 0 0 0 0 0 jmb38x_ms
*** All the steps asked by Gábor were done.
Thanks all you guys.
Rogerio