Bekijk de volledige versie : Activating Afterburner (125Mbps) on the recent wl-500g units
I'M NOT RESPONSIBLE TO ANY DAMAGE CAUSED TO YOUR UNIT BY DOING THIS! THIS IS SOLELY YOUR OWN RISK!
Any reprint of this information SHOULD include the link to this page and PRIOR agreement
So, finally here is an instruction on how activate afterburner (speedboster) feature on the recent wl-500g/wl-300g/wl-hdd units (this should probably also possible for other broadcom based units, including linksys wrt54g). Thanks are going to Antiloop for performing required testing. :)
"Upgradable" units
Why I'm talking about recent units? The reason is simple. All Broadcom 802.11g cards are based on bcm4306 chip, but these chips are not identical to each other. Each chip has a so called revision number. There are at least two important numbers: chip revision and core revision. Afterburner could be enabled only for certain revisions, namely with core revision greater than 4. Most of the old wl-500g are equiped with bcm4306 with a core revision 4, so they are NOT "upgradable".
How to determine your core revision
There several ways for doing this. The simplest is the direct query using wl utility. Just execute
wl revinfo
You will get back and output which looks like this:
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x2
corerev: 0x4
boardid: 0x120f
boardvendor: 0x14e4
boardrev: 0x15
driverrev: 0x1030200
ucoderev: 0x1180015
bus: 0x1
The output above gives you corerev 0x4 - i.e. 4 - this board is not "upgradable".
"Upgradable" board should give you somthing like this:
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x120f
boardvendor: 0x14e4
boardrev: 0x15
driverrev: 0x1030200
ucoderev: 0x1180015
bus: 0x1
Note, that corerev is 5, and chiprev is 3. This board is "upgradable".
Another way for determining your board revision is to use
cat /proc/pci
command.
This command returns bunch of data, including bcm4306 info
Bus 1, device 2, function 0:
Class 0280: PCI device 14e4:4320 (rev 3).
IRQ 6.
Non-prefetchable 32 bit memory at 0x40002000 [0x40003fff].
Please note, that there is no core revision output, but rev 3 is stands to chiprev 3 which is corresponds to corerev 5, which is "upgradable".
"Upgrading" it
First you need enable Afterburner for board itself (in fact, just to say driver, that board is enabled for afterburner). To do this you will need to modify boardflags which are stored along with things like MAC address, vendor/product id, etc... in board srom (non-volotile memory). These flags are 16-bit value, stored at the byte-offset 114 in srom, afterburner enable is a bit resulting in little-endian value of 0x200. By default srom is programmed with boardflags=0xf, you need to set it to 0x20f using these commands:
wsrom eth2 114 $((0x20f))
wsrom eth2 116 $((0xffff))
First command performs required change, while the second workarounds ASUS bug in the wsrom command. :)
Finally you will need to enable afterburner mode for the firmware using these commands:
nvram set wl_afterburner=auto
nvram set wl0_afterburner=auto
nvram commit
Reboot your unit, it should come back with afterburner.
Antiloop
16-10-2004, 21:22
okay now the (current) bad news about 125mbps wireless networking
I have performed this test with help of Oleg, on a WL-HDD bought in augustus i'm pretty sure ALL WL-HDD's have the same revision
for testing on the client i used a Linksys WMP54GS (PCI Speedbooster) card
test setup with NO encryption:
client connects at 125mbps with wl-hdd
test setup with 128bit WEP:
client does NOT connect, but does connect using linksys util but at 54mbps
test setup with WPA-PSK aes/tkip:
client does NOT connect, but does connect using linksys util but at 54mbps
this seems to be due the 'old' wireless drivers in the wl-hdd/wl500g firmware
also read something about it at http://www.sveasoft.com/modules/phpBB2/viewtopic.php?t=4221&highlight=afterburner
i've seen more topics there click here to access their forum: http://www.sveasoft.com/modules/phpBB2/
JOCKYW2001
16-10-2004, 22:15
Bus 1, device 2, function 0:
Class 0280: PCI device 14e4:4320 (rev 3).
IRQ 6.
Non-prefetchable 32 bit memory at 0x40002000 [0x40003fff].
Please note, that there is no core revision output, but rev 3 is stands to chiprev 3 which is corresponds to corerev 5, which is "upgradable".
This is exactly what I see on my WL-HDD. Seems it's also an "afterburner".
(wl revinfo cannot be used, revinfo is an unknown wl param on the WL-HDD)
So, finally here is an instruction on how activate afterburner (speedboster) feature on the recent wl-500g/wl-300g/wl-hdd units.
For your information, "recent" would be quite a relative term. I bought my WL-500g (online in Germany) on 2004-06-22, and it appears to be upgradeable, given this result output of "wl revinfo":
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x120f
boardvendor: 0x1043
boardrev: 0x15
driverrev: 0x1030200
ucoderev: 0x1180015
bus: 0x1
I have yet to try this hack (with 1.8.1.7-1 custom), as I currently do not use the WLAN portion of the unit at all.
My wl500g's radio also has a corerev of 5. Unfortunately I don't have any other "Speedbooster" device, so I can't test if it works.
vendorid: 14e4
deviceid: 4320
radiorev: 2205017f
chipnum: 4306
chiprev: 3
corerev: 5
boardid: 120f
boardvendor: 1043
boardrev: 15
driverrev: 332150a
ucoderev: 0
bus: 1
By the way, does anyone know why my boardvendor is 1043 ("Asustek Computer Inc") while Oleg's one is 14e4 ("Broadcom Corp")?
This is exactly what I see on my WL-HDD. Seems it's also an "afterburner".
(wl revinfo cannot be used, revinfo is an unknown wl param on the WL-HDD)
Wl-hdd owners can flash wl-300g 1.8.x.x firmware to use an updated wl utility.
My wl500g's radio also has a corerev of 5. Unfortunately I don't have any other "Speedbooster" device, so I can't test if it works.
Well in fact it's also possible to "upgrade" mini-pci and pci cards, which are installed in PCs.
By the way, does anyone know why my boardvendor is 1043 ("Asustek Computer Inc") while Oleg's one is 14e4 ("Broadcom Corp")?
This output is from wl-hdd, it has different boardid/boardvendor and this values does not affects afterburner. In fact you can set it to whatever you want, it's also stored in the srom. use
wl srom
to examine srom contents.
Well in fact it's also possible to "upgrade" mini-pci and pci cards, which are installed in PCs.
How? Where to find more info on this specifically for WL-138g?
currently running firmware 1.6.5.3, wl utility does not seem to be present there or is not reachable from Web System Command page(Main_AdmStatus_Content.asp)
I cannot search the forum for "wl", where to get more info on this utility?
Antiloop
18-10-2004, 01:13
How? Where to find more info on this specifically for WL-138g?
currently running firmware 1.6.5.3, wl utility does not seem to be present there or is not reachable from Web System Command page(Main_AdmStatus_Content.asp)
I cannot search the forum for "wl", where to get more info on this utility?
wl-138g is NOT a broadcom powered pci card, so it won't work
only broadcom stuff is affected
Please note: Any reprint of this information SHOULD include the link to this page and PRIOR agreement
A tool to enable afterburner easily without using wsrom
This message has an attachement containg source code and precompiled binaries for mipsel and x86 architectures, which can be used to enable afterburner. The requirements are: linux and at least corerev 5 broadcom 4320 card.
This is proof of concept program, which can damage your card SROM if used incorrectly. So, use on your own risk.
Perform these steps to enable afterburner:
verify, that the program can access your SROM correctly by running
b43_srom 14e44320
This should produce number of lines and a "SROM CRC ok" message
check, that word at offset 0039 is 000f, if so re-run it using
b43_srom 14e44320 39 20f y
This would took some time (several seconds) to commit your changes
It's also possible to run the tool on the PCs equipped with mini-pci or PCI cards utilizing broadcom 4320 chip, althought I've not tried it. Use on your own risk.
Rob ter Veer
20-10-2004, 22:06
I've got two wl500g's bridged together with WDS (1.8.1.7-2a) working quite good.
Is it possible with this hack to increase WDS speed also, so that overall speed is upped? And also does afterburner work with an 100g pcmcia card?
Is it possible with this hack to increase WDS speed also, so that overall speed is upped?
I've no idea. At the moment WDS does not work with afterburner enabled.
And also does afterburner work with an 100g pcmcia card?
This depends on the revision of the card you have. Theoritically it could be also enabled, but this requires special handling for pcmcia. Unfortunatly I do not own such card to play with it.
After reading the information shown on this thread I logged into my wrt54g v1.1 openwrt unit and issued the command:
# wl revinfo
vendorid: 14e4
deviceid: 4320
radiorev: 2205017f
chipnum: 4306
chiprev: 3
corerev: 5
boardid: 1603
boardvendor: 16a5
boardrev: 45
driverrev: 332150a
ucoderev: f500cb
bus: 1
As you can see it seems like a potential afterburner-upgradable unit due to corerev and chiprev values.
I quickly downloaded the srom.mipsel binary and issue the first command:
# ./b43_srom.mips 14e44320
Found pci device 14e44320 at /proc/bus/pci/01/08.0@0x40002000
Enabling device...
Device memory mapped ok
SROM CRC ok
0000 4001 0000 1603 16a5 4320 8000 0002 0002
0008 1000 1800 0000 0000 ffff ffff ffff ffff
0010 ffff ffff ffff ffff ffff ffff ffff ffff
0018 ffff ffff ffff ffff ffff ffff ffff ffff
0020 ffff ffff ffff ffff 000c 419d 7221 ffff
0028 ffff ffff ffff ffff ffff ffff 3545 111c
0030 fbcc fe84 ffff ffff 0040 ffff ffff ffff
0038 003e 000d ffff 0000 0000 0000 0000 6b01
The value at offset 0039 is not 000f so I did not issue the second command because I was afraid of bricking the card.
Any hints? Maybe wrt54g v1.1 units are also afterburner capable... that would be superb for our backbone here in Valencia (Spain).
Thanks in advance :-)
You can find a driver for the wl100g on the asus website. Apparently they have
added Support for Afterburner: "only be run on a 4306 C0 card and only on a board with the appropriate SPROM settings (boardflags)"
http://www.asus.com.tw/support/download/selectftp.aspx?l1_id=11&l2_id=1&l3_id=0&m_id=2&f_name=Eng_2201.zip~zaqwedc
Unfortunately I cannot try it because my WL500G is an "older type" and I really do not know whicht type my wl100G is.
Hi all,
I tried to run Olegs procedure to active
in order to activate afterburner. My WL100g
supports Afterburner and I'm curious to see
if it works.
However, I couldn't run wsrom command.
It's unknown at my unit.
I use Olegs customized firmware 1.7.5.6-5.
Thanks in advance for any help.
Regards
Nik
However, I couldn't run wsrom command.
It's unknown at my unit.
I use Olegs customized firmware 1.7.5.6-5.
Upgrade to 1.8.1.7-2a
After reading the information shown on this thread I logged into my wrt54g v1.1 openwrt unit and issued the command:
# wl revinfo
vendorid: 14e4
deviceid: 4320
radiorev: 2205017f
chipnum: 4306
chiprev: 3
corerev: 5
boardid: 1603
boardvendor: 16a5
boardrev: 45
driverrev: 332150a
ucoderev: f500cb
bus: 1
As you can see it seems like a potential afterburner-upgradable unit due to corerev and chiprev values.
I quickly downloaded the srom.mipsel binary and issue the first command:
# ./b43_srom.mips 14e44320
Found pci device 14e44320 at /proc/bus/pci/01/08.0@0x40002000
Enabling device...
Device memory mapped ok
SROM CRC ok
0000 4001 0000 1603 16a5 4320 8000 0002 0002
0008 1000 1800 0000 0000 ffff ffff ffff ffff
0010 ffff ffff ffff ffff ffff ffff ffff ffff
0018 ffff ffff ffff ffff ffff ffff ffff ffff
0020 ffff ffff ffff ffff 000c 419d 7221 ffff
0028 ffff ffff ffff ffff ffff ffff 3545 111c
0030 fbcc fe84 ffff ffff 0040 ffff ffff ffff
0038 003e 000d ffff 0000 0000 0000 0000 6b01
The value at offset 0039 is not 000f so I did not issue the second command because I was afraid of bricking the card.
Any hints? Maybe wrt54g v1.1 units are also afterburner capable... that would be superb for our backbone here in Valencia (Spain).
Thanks in advance :-)
Your unit seems to be upgradable. Use 20d instead of 20f.
Also, you can write 20f - this should not damage your board (in fact b43_srom can be used to recover partially corrupted srom), while should enable other features.
@OpenWrt:/tmp# ./b43_srom.mips 14e44320
Found pci device 14e44320 at /proc/bus/pci/00/01.0@0x18001000
Enabling device...
PCI: Enabling device /proc/bus/pci/00/01.0 (0004 -> 0006)
Device memory mapped ok
Bus error
This is a WRT54G v2. Core and chiprev is the same as the WRT54GS.
@OpenWrt:/tmp# wl revinfo
vendorid: 14e4
deviceid: 4320
radiorev: 2205017f
chipnum: 4712
chiprev: 1
corerev: 7
boardid: 101
boardvendor: 14e4
boardrev: 10
driverrev: 332150a
ucoderev: f500cb
bus: 0
Do you've wl.o module loaded at the time of running utility? Also, can you please post
cat /proc/pci
cat /proc/bus/pci/devices
output?
Seems start address was picked incorrectly - this is due to the fact, that it's not the real bcm4306, but bcm4712 integrated.
Do you've wl.o module loaded at the time of running utility?
Yes, I have.
Also, can you please post
cat /proc/bus/pci
output?
Seems start address was picked incorrectly.
Do You mean /proc/pci?
PCI devices found:
Bus 0, device 0, function 0:
Class 0501: PCI device 14e4:0800 (rev 1).
IRQ 3.
Non-prefetchable 32 bit memory at 0x18000000 [0x18000fff].
Non-prefetchable 32 bit memory at 0x1fc00000 [0x1fffffff].
Non-prefetchable 32 bit memory at 0x1c000000 [0x1dffffff].
Non-prefetchable 32 bit memory at 0x1a000000 [0x1bffffff].
Bus 0, device 1, function 0:
Class 0280: PCI device 14e4:4320 (rev 1).
IRQ 4.
Non-prefetchable 32 bit memory at 0x18001000 [0x18001fff].
Bus 0, device 2, function 0:
Class 0200: PCI device 14e4:4713 (rev 1).
IRQ 5.
Non-prefetchable 32 bit memory at 0x18002000 [0x18002fff].
Bus 0, device 3, function 0:
Class 0c03: PCI device 14e4:4717 (rev 1).
IRQ 6.
Non-prefetchable 32 bit memory at 0x18003000 [0x18003fff].
Bus 0, device 4, function 0:
Class 0c03: PCI device 14e4:4716 (rev 1).
IRQ 2.
Non-prefetchable 32 bit memory at 0x18004000 [0x18004fff].
Bus 0, device 5, function 0:
Class 0b30: PCI device 14e4:0816 (rev 1).
IRQ 2.
Non-prefetchable 32 bit memory at 0x18005000 [0x18005fff].
Bus 0, device 6, function 0:
Class 0500: PCI device 14e4:080f (rev 1).
IRQ 3.
Non-prefetchable 32 bit memory at 0x18006000 [0x18006fff].
Non-prefetchable 32 bit memory at 0x0 [0x7ffffff].
Non-prefetchable 32 bit memory at 0x10000000 [0x17ffffff].
Non-prefetchable 32 bit memory at 0x80000000 [0x9fffffff].
Josephus
31-10-2004, 11:52
I have an upgradable wl500g unit running openwrt.
After writing sram, and modifying nvram variables i still can't get it working.
root@ittakert:~# nvram get wl_afterburner
auto
root@ittakert:~# wl gmode
54g Mode is 54g Afterburner (6)
root@ittakert:~# wl status
SSID: "huwico@joshome"
Mode: Managed RSSI: 0 dBm noise: 0 dBm Channel: 1
BSSID: XX:XX:XX:XX:XX:XX Capability: ESS
Supported Rates: [ 1(b) 2(b) 5.5(b) 6 9 11(b) 12 18 24 36 48 54 ]
I have an associated b client if that helps.
Antiloop
31-10-2004, 14:18
I have an upgradable wl500g unit running openwrt.
After writing sram, and modifying nvram variables i still can't get it working.
root@ittakert:~# nvram get wl_afterburner
auto
root@ittakert:~# wl gmode
54g Mode is 54g Afterburner (6)
root@ittakert:~# wl status
SSID: "huwico@joshome"
Mode: Managed RSSI: 0 dBm noise: 0 dBm Channel: 1
BSSID: XX:XX:XX:XX:XX:XX Capability: ESS
Supported Rates: [ 1(b) 2(b) 5.5(b) 6 9 11(b) 12 18 24 36 48 54 ]
I have an associated b client if that helps.
i have no idea how openwrt works with this
but linux does not report 125mbps.. use a afterburner capable client to test it..
wl status did not report 125mbps for me either.. allthough it does work
I have an upgradable wl500g unit running openwrt.
After writing sram, and modifying nvram variables i still can't get it working.
I have an associated b client if that helps.
Yes, according to broadcom you should use afterburner capable only devices in this network, otherwise it will switch to normal operation (so, yours 802.11b disables it). Also, in fact, it's the same 802.11g, but with reduced timings, allowing to pass more data in the same time. That's why it will not report new speed.
Yes, I have.
Do You mean /proc/pci?
I mean both to see all requested resources. But /proc/pci output also looks strange - for some reason only 4096 bytes are requested by device, it should be at least 8192 for srom image...
I will try to check sources for the 4712 specific srom mapping. Stay tuned. :)
I mean both to see all requested resources. But /proc/pci output also looks strange - for some reason only 4096 bytes are requested by device, it should be at least 8192 for srom image...
I will try to check sources for the 4712 specific srom mapping. Stay tuned. :)
Don't forget, that the v2 hardware includes the radio itself. This can be a clue here imho :)
Don't forget, that the v2 hardware includes the radio itself. This can be a clue here imho :)
Can you please post an output for the
cat /proc/bus/pci/devices
Also, pci config could unveil things try this one and attach /tmp/1 to the message...
cat /proc/bus/pci/00/01.0 > /tmp/1
Also, this unit seems to have 2 usb controllers... It would be great if you do
cat /proc/bus/pci/00/03.0 > /tmp/3
cat /proc/bus/pci/00/04.0 > /tmp/4
and then attach /tmp/3 & /tmp/4 - I just want to check for their types.
Just found many answers myself - usb - one is host and another is device, as for srom - it's not on the pci bus, it's at the "silicon backplane" bus, so it does not work this way for bcm4712.
Also, broadcom sources does not provide any info on how to access srom at this internal bus, althought there something similar called sflash - flash on the silicon backplane.
cat /proc/bus/pci/devices
@OpenWrt:/# cat /proc/bus/pci/devices
0000 14e40800 3 18000000 1fc00000 1c000000 1a000000 00000000 00000000 00000000 00001000 00400000 02000000 02000000 00000000 00000000 00000800
0008 14e44320 4 18001000 00000000 00000000 00000000 00000000 00000000 00000000 00001000 00000000 00000000 00000000 00000000 00000000 00000800 wl
0010 14e44713 5 18002000 00000000 00000000 00000000 00000000 00000000 00000000 00001000 00000000 00000000 00000000 00000000 00000000 00000800 et
0018 14e44717 6 18003000 00000000 00000000 00000000 00000000 00000000 00000000 00001000 00000000 00000000 00000000 00000000 00000000 00000800
0020 14e44716 2 18004000 00000000 00000000 00000000 00000000 00000000 00000000 00001000 00000000 00000000 00000000 00000000 00000000 00000800
0028 14e40816 2 18005000 00000000 00000000 00000000 00000000 00000000 00000000 00001000 00000000 00000000 00000000 00000000 00000000 00000800
0030 14e4080f 3 18006000 00000000 10000000 80000000 00000000 00000000 00000000 00001000 08000000 08000000 20000000 00000000 00000000 00000800
cat /proc/bus/pci/00/01.0 > /tmp/1
cat /proc/bus/pci/00/03.0 > /tmp/3
cat /proc/bus/pci/00/04.0 > /tmp/4
attached :)
Kaloz, thanks.
I've prepared yet another utility, which uses wl ioctl to access srom. You should be able to read srom, but writability is depends on the wl.o driver version.
So, it's attached (sorry for zip format, that's the only way to attach it to the thread).
This utility requires wl.o driver to be loaded.
Usage: wlsrom <interface> [ <word-offset> <word-value> ]
So, to dump current srom content try running like this:
wlsrom eth2
To upgrade to afterburner run it like this (check for old value at 0x39 before upgrade):
wlsrom eth2 39 20f
Let me know if it works for you.
Well, yet another piece of information.
The utility above does not work with bcm4712 based units (wrt54g v2.0), so probably most have just downloaded it, tried to run and thinked this utility is a fake. Very Happy
Further investigation using the info provided by Kaloz (thanks!) showed, that 4712 does not have a srom mapped to the PCI, instead flags showed that 4306 on the silicon backplane bus. But according to broadcom sources it's not possible to access srom this bus. Shocked
With help of Kaloz, I've discovered, that instead boardflags are stored in the nvram.
For his board boardflags == 0x0188, I've suggested him to turn magic afterburner bit with
nvram set boardflags=0x0388
nvram commit
and reload the wl.o module. And magically, wl.o has enabled afterburner on his wrt54g v2.0:
<Kaloz> @OpenWrt:/# wl gmode 6
<Kaloz> eth1: Invalid argument
<Kaloz> gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner)
<Kaloz> but openwrt uses the afterburner drivers anyway :)
<oleg__> ok, try setting boardflags to 0x0388, commit changes and reload wl.o
<Kaloz> lol
<Kaloz> @OpenWrt:/# wl gmode 6
<Kaloz> @OpenWrt:/# wl gmode
<Kaloz> 54g Mode is 54g Afterburner (6)
<oleg__> cool.
<oleg__> afterburner demystified.
No utility required for bcm4712 devices.
P.S. Note to James of sveasoft - consider sending us a few bucks before adding this to your new cool firmware and authoring this at sveasoft forums. :D
Just discovered my unit supports Afterburner as well, so I'm going for it. But first a question though:
Do the nvram settings survive a firmware up/downgrade or restore to factory settings? Or do you have to set them everytime a new firmware is uploaded?
Just discovered my unit supports Afterburner as well, so I'm going for it. But first a question though:
Do the nvram settings survive a firmware up/downgrade or restore to factory settings? Or do you have to set them everytime a new firmware is uploaded?
wl500g, wlhdd and wrt54g before 2.0 uses separate srom for this, so you need to do this once.
wl500g, wlhdd and wrt54g before 2.0 uses separate srom for this, so you need to do this once.
Perhaps I misunderstand you, but I was mentioning the nvram settings. These are stored in the WL-500g flash. So how about these settings?
Hmm, just did the upgrade using 'b43_srom.mips' which seemed to go well but after reboot I checked for status and it returnes:
[wl500g]$wl gmode
54g Mode is 54g Auto (1)
[wl500g]$ nvram get wl_afterburner
auto
Is this correct?
Perhaps I misunderstand you, but I was mentioning the nvram settings. These are stored in the WL-500g flash. So how about these settings?
ok, now i see. yes, you need to do this after each firmware upgrade.
Hmm, just did the upgrade using 'b43_srom.mips' which seemed to go well but after reboot I checked for status and it returnes:
[wl500g]$wl gmode
54g Mode is 54g Auto (1)
[wl500g]$ nvram get wl_afterburner
auto
Is this correct?
try dumping srom - is it correct? Also, check wl0_afterburner.
You can turn afterburner on using
wl gmode 6
to see if it supported.
try dumping srom - is it correct?
/tmp/harddisk/wlan/b43_srom.mips 14e44320
Found pci device 14e44320 at /proc/bus/pci/01/02.0@0x40002000
Enabling device...
Device memory mapped ok
SROM CRC ok
0000 4001 0000 120f 1043 4320 8000 0002 0002
0008 17f0 1800 0000 0000 ffff ffff ffff ffff
0010 ffff ffff ffff ffff ffff ffff ffff ffff
0018 ffff ffff ffff ffff ffff ffff ffff ffff
0020 ffff ffff ffff ffff 000e a690 7ec8 ffff
0028 ffff ffff ffff ffff ffff ffff 3015 11b0
0030 fb98 fe4f ffff ffff 003c ffff ffff ffff
0038 003e 020f ffff 0000 0000 0000 0000 f501
This seems ok to me.
Also, check wl0_afterburner.
This seems ok too as it returns:
nvram get wl0_afterburner
auto
You can turn afterburner on using
wl gmode 6
to see if it supported.
This returns an error:
eth2: Invalid argument
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner )
But this seems to work, but what does the result mean?:
wl gmode Afterburner
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner )Which seems to be ok, but I still get:
wl gmode
54g Mode is 54g Auto (1)
Post an output for the
wl revinfo
Post an output for the
wl revinfo
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x120f
boardvendor: 0x1043
boardrev: 0x15
driverrev: 0x1030200
ucoderev: 0x1180015
bus: 0x1
Edit: Antiloop asked me to do:
wl cap
ap sta afterburner
which shows that afterburner is enabled I think :)
Thanks Oleg and Antiloop
Edit: Antiloop asked me to do:
wl cap
ap sta afterburner
which shows that afterburner is enabled I think :)
Thanks Oleg and Antiloop
Well, you probably has non-afterburner client associated to your wl500g.
Wesleyrpg
03-11-2004, 01:46
hi, how are you?
i'm a newbie, actually this is my first post so please be nice. I bought a WL-500 the other day and had absoulutely no problems upgrading the WL-500g to afterburner mode.
The only problem is how do i know/test the mini-pci card built into my laptop, to see if it supports this 'afterburner' mode and if so how do i upgrade the mini pci card itself?
The mini-pci card is based on the Intel 2200BG chipset?
Many thanks
Adam
Antiloop
03-11-2004, 08:21
hi, how are you?
i'm a newbie, actually this is my first post so please be nice. I bought a WL-500 the other day and had absoulutely no problems upgrading the WL-500g to afterburner mode.
The only problem is how do i know/test the mini-pci card built into my laptop, to see if it supports this 'afterburner' mode and if so how do i upgrade the mini pci card itself?
The mini-pci card is based on the Intel 2200BG chipset?
Many thanks
Adam
no need to greet eachother (see faq) we greet everytime already by being at the board ;)
anyway the intel card is not 'speedbooster' or 'afterburner' capable, and thus not supporting 125mbit mode.
...
The mini-pci card is based on the Intel 2200BG chipset?
...Afterburner only works with certain broadcom chipsets. It aint gonna work on your Intel card.
Wesleyrpg
03-11-2004, 13:06
i done a bit more research today on the intel 2200BG chipset, and you guys are right, afterburner definately aint going to happen with this card, which brings me to a few more questions.
1. Is there any noticable difference between 54M and 125M?
2. Does anyone know where i can get a mini-pci card that supports this 'afterburner' mode, and also supports two antennas?
Thanks
Spacejam
08-11-2004, 17:03
With help of Kaloz, I've discovered, that instead boardflags are stored in the nvram.
For his board boardflags == 0x0188, I've suggested him to turn magic afterburner bit with
nvram set boardflags=0x0388
nvram commit
and reload the wl.o module. And magically, wl.o has enabled afterburner on his wrt54g v2.0:
Hi Oleg, Just want to know the command to reload wl.o??
Thanks. Oh..I'm using WRT54G v2.0
Spacejam
08-11-2004, 18:28
I have try running the command below...1 and 2 work fine..
3.How to reload wl.o? Any command to reload wl.o?
1. nvram set boardflags=0x0388
2. nvram commit
3. reload wl.o ???
andrewplanet
08-11-2004, 21:58
I just tried this on my v2.0 WRT54G with Sveasoft Alchemy
I first checked my corerev
wl revinfo
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4712
chiprev: 0x1
corerev: 0x7
boardid: 0x101
boardvendor: 0x14e4
boardrev: 0x10
driverrev: 0x33c0d00
ucoderev: 0x1180016
bus: 0x0
Looks good.
Next i set the board flags
nvram set boardflags=0x0388
nvram commit
Okay done.
Now check to see if afterburner is enabled
nvram get wl_afterburner
auto
nvram get wl0_afterburner
autoI reboot the router.
Then do this.
wl gmode
54g Mode is 54g Only (2)then
wl gmode 6
eth1: Invalid argument
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner)
okay try this
wl gmode Afterburner
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner)
Now to check if it was successful
wl gmode
54g Mode is 54g Only (2)damn... I see the same thing as styno did here
wl cap
ap sta afterburner
Any ideas what's wrong?
I have try running the command below...1 and 2 work fine..
3.How to reload wl.o? Any command to reload wl.o?
1. nvram set boardflags=0x0388
2. nvram commit
3. reload wl.o ???
Just reboot your router.
damn... I see the same thing as styno did here
wl cap
ap sta afterburner
Any ideas what's wrong?
Well, afterburner support is activated, but something is preventing is from being used, possible reasons are - none afterburner capable client associated to your AP, WPA mode activated, wds mode.
In fact afterburner is mostly marketing thing at the moment. 54Mbps is more than enough for all broadcom consumer level access points... It's unlikely, that you will get anything higher than 24 Mbps of real bandwidth.
Spacejam
09-11-2004, 10:24
Just reboot your router.
Have tried reboot, rrmod wl and insmod wl... Still the same, there is no option 6.;(
Have tried reboot, rrmod wl and insmod wl... Still the same, there is no option 6.;(
Which problem exactly you've now?
Spacejam
10-11-2004, 17:58
Which problem exactly you've now?
I have tried this
nvram set boardflags=0x0388
nvram commit
rmmod wl
insmod wl
wl gmode 6
Than I got this:
eth1: Invalid argument
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS|Aft erburner)
I even try reboot after nvram set boardflags=0x0388 and nvram commit
still the same there is no gmode 6..
My router is ver2.0 with corerev: 0x7.
"125Mbps" explained by HP.
http://www.hp.com/united-states/wireless/hsm_download.html
Update: As of today (2005-01-17) Asus ships new wl500g units with Afterburner already enabled, so there is no need to change srom:
Output from new wl-500g unit:
[root@wl500g root]$ wl cap
ap sta wet led wme pio 802.11d 802.11h rm cqa afterburner
[root@wl500g root]$ wl revinfo
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x120f
boardvendor: 0x14e4
boardrev: 0x47
driverrev: 0x35a0700
ucoderev: 0x122000a
bus: 0x1
So, to enable afterburner just issue
nvram set wl_afterburner=auto
nvram set wl0_afterburner=auto
nvram commit
jbkerner
30-01-2005, 06:16
I have the same output as Oleg - but like others set gmode 6 doesn`t work for me. Any new ideas?
ciao,
JBK
Don't use wl gmode - it will not work with new firmwares at all (they do not allow switching modes on the fly). Just set the nvram variables and reboot.
Connect with Afterburner (speedbooster) capable device then...
Is it possible to enable Afterburner on this wl500g ? I have already tried to set nvram but without success.
nvram setting:
wl0_afterburner=auto
wl_afterburner=auto
[admin@(none) root]$ wl revinfo
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x120f
boardvendor: 0x14e4
boardrev: 0x47
driverrev: 0x35a0700
ucoderev: 0x122000a
bus: 0x1
[admin@(none) root]$ wl cap
ap sta wet led wme pio 802.11d 802.11h rm cqa
[admin@(none) root]$ wl srdump
srom[000]: 0x4001 0x0000 0x120f 0x14e4 0x4320 0x8000 0x0002 0x0002
srom[008]: 0x17f0 0x1800 0x0000 0x0000 0xffff 0xffff 0xffff 0xffff
srom[016]: 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff
srom[024]: 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff
srom[032]: 0xffff 0xffff 0xffff 0xffff 0x0011 0x2fe3 0x247a 0xffff
srom[040]: 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x3547 0x11b0
srom[048]: 0xfb98 0xfe4f 0xffff 0xffff 0x0035 0xffff 0xffff 0xffff
srom[056]: 0x003e 0x000f 0xffff 0x0000 0x0000 0x0000 0x0000 0x1001
Thank you !
I have found the solution (hopefully). It looks like I have got an old version.
After changing srom with:
wlsrom eth2 39 20f
Afterburner seems to be enabled.
wl cap
ap sta wet led wme pio 802.11d 802.11h rm cqa afterburner
BTW: is the last byte in srom a checksum ?
It also changed
from
0038 003e 000f ffff 0000 0000 0000 0000 1001
to
0038 003e 020f ffff 0000 0000 0000 0000 ee0
Is there a list of the meaning of all bytes in srom ?
Hi there, I recently formatted my laptop, installed the WL-100G with these 3.70 Broadcom drivers (had to do this manually, Windows says there is no driver in the extracted folder containing 3.70) and hey, I HAVE AFTERBURNER!
you can check the settings of your driver in the hardwarelist, on some tab you can edit the features, and enable Afterburner.
If you have previous driver installed, windows will use that instead of 3.70!!! so thats why most users think the wl-100g doesn't have Afterburner.
Caution!
i have an upgradeble wl500g but when doing commands my router dies
2 times
i had to recover it by putting groundcable on pin 9 or 10? dunno yet
afterwards u need a clean (zero) firmware and have to flash it with the recovery tool,afterwards u can flash the normal firmware
this is for all who stuck same as me.
So, what you've done? It appears, that you tried doing nvram set boardflags on the wl500g, which should not be used.
It works on a very old Belkin F5D7230-4 (hw 1222df) ! (with Oleg's 1.9.2.7-6c-pre5 firmware)
wl revinfo:
vendorid: 0x14e4
deviceid: 0x4320
radiorev: 0x22050000
chipnum: 0x4306
chiprev: 0x3
corerev: 0x5
boardid: 0x7000
boardvendor: 0x1799
boardrev: 0x45
driverrev: 0x35a1700
ucoderev: 0x1220098
bus: 0x1
wl cap:
ap sta wet led wme pio 802.11d 802.11h rm cqa afterburner
b43_srom 14e44320:
Found pci device 14e44320 at /proc/bus/pci/01/01.0@0x40002000
Enabling device...
Device memory mapped ok
SROM CRC ok
0000 4001 0000 7000 1799 4320 8000 0002 0002
0008 17f0 1800 0000 0000 ffff ffff ffff ffff
0010 ffff ffff ffff ffff ffff ffff ffff ffff
0018 ffff ffff ffff ffff ffff ffff ffff ffff
0020 ffff ffff ffff ffff 0030 bdf6 6cec ffff
0028 ffff ffff ffff ffff ffff ffff 3045 11b0
0030 fb98 fe4f ffff ffff 004c ffff ffff ffff
0038 003e 0249 ffff 0000 0000 0000 0000 0701
wl gmode:
gmode Set the 54g Mode (LegacyB|Auto||GOnly|BDeferred|Performance|LRS)
nvram show | grep afterburner:
wl0_afterburner=auto
wl_afterburner=auto
I left gmode in 54g Auto (1) and voila, when i connect with my Belkin F5D7011 High Speed Mode pcmcia adapter, linkspeed is @ 125 mbps ! Effective throughput at short distance is 25 mbps, with peaks above 30 mbps, quite a bit better than 54g performance (which maxed out at 20 mbps)
Anyone got any ideas why i can't see the Afterburner option when i execute "wl gmode help" ? (i've tried the srom register at 39 with value 0x020f but that didn't make any difference)
Greetz
Hey Oleg, I don't have a Linuxpc, isn't this also possible to do from a WindowsXP pc?
I tried Telnet but that doesn't work.