sounds good :)Quote:
Originally Posted by wpte
btw., TARPIT module not loaded by default, so if you want to use it, you have to issue "insmod ipt_TARPIT.o" somewhere (in post-boot for example).
Printable View
sounds good :)Quote:
Originally Posted by wpte
btw., TARPIT module not loaded by default, so if you want to use it, you have to issue "insmod ipt_TARPIT.o" somewhere (in post-boot for example).
Yeh I do that in the beginning of post-firewall:)Quote:
Originally Posted by lly
Sorry, forgot to take a look to attached post-firewall.txt. IMHO, better to load modules in post-boot, since post-firewall may be called many times due to IP changes etc.Quote:
Originally Posted by wpte
k will do that;)Quote:
Originally Posted by lly
I found these messages a few times in the log btw:
After a bit of googeling I found out this is a bug?Quote:
Apr 4 19:39:37 kernel: Warning: DQ5 raised while erase operation was in progress, but erase completed OK
Apr 4 19:41:25 dnsmasq[51]: DHCPREQUEST(br0) ****
Apr 4 19:41:25 dnsmasq[51]: DHCPACK(br0) ****
Apr 4 19:47:34 ntp client: Synchronizing time with time.nist.gov ...
Apr 4 19:54:44 kernel: Warning: DQ5 raised while erase operation was in progress
It is "induced" bug. Unfortunately, I can't reproduce it myself.Quote:
Originally Posted by wpte
Got another one:
I had about 100 of these errors (all the same) within a single second:pQuote:
Apr 10 20:52:01 kernel: __alloc_pages: 0-order allocation failed (gfp=0x20/0)
and nothing I can see in the log that started it... the last few messages where about DHCP leases.
btw... do I need to post up errors that possibly concern the firmware here?
just wondering, don't want to spam you guys;)
wpte:
It means that memory was ceased. At all... You can play with tmps limits if /tmp if full, or watch which process eats memory.
About errors - if you suspect that it is firmware bug, of course, yes! I hope, you will never post messages like::DQuote:
It is a critical bug! Repair it immediately!PHP Code:[admin@WL-0018F398D447 root]$ wine
-sh: wine: not found
So you mean that the flash memory became full?Quote:
Originally Posted by lly
like in a way of an exploding log file by example?
haha, no I won't post those things up:D:p
OMG NO WINDOWZE ON MY ROUTER!:)
It appears that the patch mentioned here http://home.in.tum.de/~pustka/WL-500...soundcard.html, which was added by Oleg in 1.9.2.7-7b, has been omitted.
Please could it be added to the new firmware along with the same treatment for another Creative based USB sound card (the Toshiba Mediacenter) which has code 041e:3048 and suffers from being muted by default.
Keep up the good work :)
First of all, we base on newer kernel, which already contains most of this patch, except mute control.Quote:
Originally Posted by sebw
Second, in 2.6 kernel I found totally different workaround for Audigy 2 NX - see snd_usb_audigy2nx_boot_quirk() function.
Since I don't own this card - can you answer which way is right? Is patch from your link is enough?
If you means unmute patch, yes. But, answer to my first question first, please.Quote:
Please could it be added to the new firmware along with the same treatment for another Creative based USB sound card (the Toshiba Mediacenter) which has code 041e:3048 and suffers from being muted by default.
lly
Thanks very much for you reply.
I have downloaded "linux-2.4.37.tar.bz2" and searched for the snd_usb_audigy2nx_boot_quirk() function without success. I have also looked through the various patches in http://code.google.com/p/wl500g/source/browse/. Where can I find the function?
The only change I made to the audio.o module to fix my muting problem, in Oleg's 1.9.2.7-10 F/W, was to change the 0x3020 reference to 0x3048 and force the USB port to 1.0 by puttingin my post-mount. I was hoping that I might be able to avoid this second step by using your newer firmwareCode:rmmod ehci-hcd
In case it is of interest, the product I am using is a Toshiba Multimedia Center: http://www.toshibadirect.com/td/b2c/...ltimediacenterQuote:
* kernel 2.4.37.1
o fix usbaudio driver work via USB 2.0 hub
This function present in 2.6 kernel branch only! I look at 2.6.27 sound/usb/usbaudio.c. Since, we have to backport much of code from current kernel branch and I'm not familiar with Audigy 2NX, I asked you about function above.Quote:
Originally Posted by sebw
Hope yes, but I can't exclude some other not realized features in our old kernel...Quote:
The only change I made to the audio.o module to fix my muting problem, in Oleg's 1.9.2.7-10 F/W, was to change the 0x3020 reference to 0x3048 and force the USB port to 1.0 by puttingin my post-mount. I was hoping that I might be able to avoid this second step by using your newer firmwareCode:rmmod ehci-hcd
P.S. Can you build firmware from SVN yourself?
I am a bit of a noob, but I shall try to follow the "Building of the custom firmware" instructions tonight on a Ubuntu machine.Quote:
Originally Posted by lly
In the morning:
I appear to have successfully created: WL500gx-1.9.2.7-d-r255.trx (3,649,536 bytes):)
lly:
Is it posible to use on this fw bluetooth with pan(http://en.wikipedia.org/wiki/Personal_area_network) ?
I want to share internet connection to my bluetooth phone (sony ericsson).
Strange question :confused: Bluetooth modules are present, bluez already in optware...Quote:
Originally Posted by Lesiuk
Nobody wrote step-by-step guide? Sorry, I know this topic only in russian Bluetooth â wl500gP
Hi lly
Many thanks for including my sound card (041e:3048) in the r256 unmute patch.
I shall try it out in the next day or two.
I do not know Russian language but it looks like somebody did it. Nice.Quote:
Originally Posted by lly
I will try it.
Thanks.
Btw, you can try http://transtale.google.com to read it from russian to your nativeQuote:
Originally Posted by Lesiuk
Just to confirm that I now have my sound card running unmuted across USB 2.0 using 1.9.2.7-d-r256 on my WL500gx.Quote:
Originally Posted by sebw
Many thanks.
hey lly
would it be possible to set the default regulatory mode in the wireless advanced tab (for wl-500w routers) to 802.11h?
since it's something all the intel cards use and most other wireless cards to get a faster speed than 54mbit.
802.11h is also newer plus it's used in draft-N 2.0 (with the 5GHz band):
http://en.wikipedia.org/wiki/IEEE_802.11h
http://en.wikipedia.org/wiki/IEEE_802.11d
to set it to H I think most people will stick with oleg because they won't notice any difference with original firmware on wireless speeds (since most of them don't look to advanced settings).
thanks:p
we will try to do that.Quote:
Originally Posted by wpte
I am having trouble using my webcam which worked under 1.9.2.7-10 F/W.
I am using Palantir.
I have modprobed the pwc and pwcx modules but get the following:The client shows no picture and top shows no CPU load for Palantir.Code:dmesg output...
Linux video capture interface: v1.00
pwc Philips webcam module version 9.0.2 loaded.
pwc Supports Philips PCA645/646, PCVC675/680/690, PCVC720[40]/730/740/750 & PCVC830/840.
pwc Also supports the Askey VC010, various Logitech Quickcams, Samsung MPC-C10 and MPC-C30,
pwc the Creative WebCam 5 & Pro Ex, SOTEC Afina Eye and Visionite VCS-UC300 and VCS-UM100.
usb.c: registered new driver Philips webcam
pwc Logitech QuickCam 4000 Pro USB webcam detected.
pwc Registered as /dev/video0.
pwc Philips webcam decompressor routines version 9.0-BETA-2
pwc Supports all cameras supported by the main module (pwc).
pwc Adding decompressor for model 645.
pwc Adding decompressor for model 646.
pwc Adding decompressor for model 675.
pwc Adding decompressor for model 680.
pwc Adding decompressor for model 690.
pwc Adding decompressor for model 720.
pwc Adding decompressor for model 730.
pwc Adding decompressor for model 740.
pwc Adding decompressor for model 750.
Palantir ouput....
$ Apr 24 11:47:31 [config] Ignoring deprecated `Audio' directive at line 67
Apr 24 11:47:31 [main] -- palantir 2.7 starting --
Apr 24 11:47:31 [main] Definitions for 1 devices found
Apr 24 11:47:31 [main] No serial port specified
Apr 24 11:47:31 [main] No named pipe specified
Apr 24 11:47:31 [video] Card: Logitech QuickCam Pro 4000 (/dev/v4l/video0)
Apr 24 11:47:31 [video] Capabilities: 1 (capture )
Apr 24 11:47:31 [video] Size: (160x120)-(640x480)
Apr 24 11:47:31 [video] Channel no. 0 ('Webcam') tuners: 0, flags: 0, type: 2, norm: 10937
Apr 24 11:47:31 [video] Unmuted audio input 0
Apr 24 11:47:31 [video] Audio input 0 does not support volume control
Apr 24 11:47:31 [video] RGB palette unavailable.
Apr 24 11:47:31 [video] trying YUV420P...
Apr 24 11:47:31 [video] brightness: 32768 hue: 65535 colour: 32768
Apr 24 11:47:31 [video] contrast: 32768 whiteness: 32768
Apr 24 11:47:31 [video] depth: (24) palette: 15
Apr 24 11:47:31 [video] Actual capture box is 320x240 (requested 320x240)
Apr 24 11:47:31 [video] Camera has no pan/tilt control
Apr 24 11:47:31 [video] Frame rate set to 15 f/s
Apr 24 11:47:31 [video] Video buffer size: 921600, frames: 2
Apr 24 11:47:31 [video] Allocating yuv->rgb buffer (230400 bytes)
Apr 24 11:47:31 [main] Starting device thread
Apr 24 11:47:31 [audio] Setting up audio device /dev/sound/dsp1
Apr 24 11:47:31 [audio] Audio sample rate set to 8000 Hz
Apr 24 11:47:31 [audio] Audio input buffering: 4 fragments of 256 bytes each
Apr 24 11:47:31 [main] Listening on 0.0.0.0 port 3000
Apr 24 11:47:31 [main] Listening on 0.0.0.0 port 3001
Apr 24 11:47:31 [main] Control queue timeout set to 30 s
Palantir output following client connection....
Apr 24 12:09:32 [main] 192.168.0.1 - #0: connection accepted, spawning server.
Apr 24 12:09:32 [main] 1 client(s) connected
Apr 24 12:09:32 [main] Capture thread started
Apr 24 12:09:32 [http] 192.168.0.1 - video server started
Apr 24 12:09:32 [http] 192.168.0.1 - requested URL `/'
Apr 24 12:09:32 [http] 192.168.0.1 - User-Agent: pclient/ack 2.4 [1] (Win32)
Apr 24 12:09:32 [http] 192.168.0.1 - streaming mode: continuous, closed loop
back to dmesg output....
pwc Failed to set LED on/off time.
pwc set_video_mode(176x144 @ 10, palette 15).
pwc decode_size = 1.
pwc Using alternate setting 1.
hcd.c: use explicit queuing not urb->next
pwc isoc_init() submit_urb 0 failed with error -22
hcd.c: use explicit queuing not urb->next
pwc isoc_init() submit_urb 1 failed with error -22
pwc set_video_mode(320x240 @ 15, palette 15).
pwc decode_size = 3.
pwc Using alternate setting 4.
hcd.c: use explicit queuing not urb->next
pwc isoc_init() submit_urb 0 failed with error -22
hcd.c: use explicit queuing not urb->next
pwc isoc_init() submit_urb 1 failed with error -22
I presume the problem is to do with the error -22 which occurs in the pwc-if.c module.
Any suggestions or help would be appreciated.
This is due to updated kernel code.Quote:
Originally Posted by sebw
Updated:
I was wrong - pwc 10.0.12-rc1 isn't compatible with 2.4 kernels :( [ At least, urb->next don't used in it. ]
Problems similar to this, present in most old webcam drivers :( For example - https://dev.openwrt.org/ticket/532
We will try to fix this issue, but I can't you promise exact date when it happens.
Should I enter this pwc issue here: http://code.google.com/p/wl500g/issues/list
As you wish. People from russian forum discover this bug a few weeks ago. I didn't put this link here since most of people in this thread don't understand russian language.Quote:
Originally Posted by sebw
Unfortunately, webcam developers switched to 2.6 trunk several years ago, so we have to backport much things from new kernels, and this job is not so easy...
confirmed. only our free time, desire, ability, volunteers and/or donations are required ;)Quote:
Originally Posted by lly
Why is it that we stay with kernel 2.4?Quote:
Originally Posted by lly
when I read some openWRT stuff I found out the only thing is the wireless driver is still in kernel 2.4 and not 2.6
isn't it easier to port the wireless driver to 2.6 then?
I guess there are also other things that are not compatible:o
isn't. broadcom wireless driver is closed sourced and shipped in binnaries only for 2.4 kernels yetQuote:
Originally Posted by wpte
I hate closed software for linux:(Quote:
Originally Posted by theMIROn
they probably didn't update it ever since they released it as well I guess...
lets see if they allow decompiling in the ual:D
or... maybe send a letter:p
I'm on this:cool:
As I know, you are not the first, who want ask Broadcom about wl driver sources (OpenWRT team, Oleg). Unfortunately, result is negative :mad:Quote:
Originally Posted by wpte
At the other hand, there're good news upcoming 'bout 2.6 kernel and broadcom native wl driver
http://wl500g.info/showpost.php?p=136257&postcount=789
New drivers probably can omit support for old cards :(Quote:
Originally Posted by theMIROn
Yep, it's possible... We'll see, holding our fingers crossed.Quote:
Originally Posted by lly
AFAIK 802.11h disables connectivity for Intel 2200BG and similar wireless cards (by setting the spectrum management flag to ON), so it's probably a good idea to keep regulatory mode "off" by default.Quote:
Originally Posted by wpte
Alternatively wlconf can be patched to keep spectrum management off even if regulatory is set to "h" - but I don't know if this change may have any other implications...
well yes, spectrum management is one of the key good features of 802.11h... less intefierenceQuote:
Originally Posted by teddy_bear
but by default the regulatory mode is set to 802.11d, wich isn't that great imo
I didn't have any problems with B/G cards with 802.11h tho:confused:
Not sure... I don't have any Intel 2200BG card to test it myself, so this info is mostly based on this OpenWRT ticket, and this Tomato thread. Also, Oleg recommended to set mode to "d" when he releazed 1.9.2.7-8 - specifically to "solve some problems with Intel WiFi cards".Quote:
Originally Posted by wpte
But what I know for a fact, is that the Atheros wl card in my notebook works better when regulatory mode is "off" - otherwise it keeps reconnecting all the time...
I'd vote for "off" as a default value, since according to the above and what I could google out, it causes less problems with various wifi clients than "d" or "h" modes. Anyway - not a big deal, the option is still available and can be set to "h" if that works better for you :).
+1, completely agreeQuote:
Originally Posted by teddy_bear
a bit offtopic, but:
I completed my own firmware compiles successfully thanks to the wiki on googlecode:D
Now can I use "export PATH=$PATH:/opt/brcm/hndtools-mipsel-uclibc/bin" for other projects to cross compile?
since I'm trying to get the master make file to work... but the openwrt svn is offline so it wont complete:(
Yes, you may. But toolchain from wl500g.googlecode.com is incompatible with optware since it uses different uClibc (0.9.29 vs. 0.9.28.2).Quote:
Originally Posted by wpte
but it uses the uclibc in oleg firmware right?Quote:
Originally Posted by lly
or won't I be able to run it on the router?:confused:
btw, I get as build system when I do a configure: i686 architecture... that's not right I guess? :P