**** BEGIN LOGGING AT Fri Nov 14 02:59:57 2008
Nov 14 03:02:15 <gramulhaozin> hey guys
Nov 14 03:02:27 <gramulhaozin> If I put my package Makefile over packages/net
Nov 14 03:02:35 <gramulhaozin> What do I have to Update to be listed on the menuconfig
Nov 14 03:02:52 <gramulhaozin> make package/symlinks didn't do the trick
Nov 14 03:07:31 <russell_> hmm.  brokenness on kmod-hostap?
Nov 14 03:14:02 <russell_> hostap-pci doesn't show up in menuconfig despite CONFIG_PCI_SUPPORT=y
Nov 14 03:18:09 <gramulhaozin> did you check the package requirements ?
Nov 14 03:18:19 <russell_> yeah
Nov 14 03:18:22 <gramulhaozin> look at the makefile, maybe it reuqires mor than PCI SUPPORT
Nov 14 03:18:36 <russell_> that and kmod-hostap, which is selected
Nov 14 03:21:46 <russell_>   DEPENDS:=@PCI_SUPPORT kmod-hostap
Nov 14 03:22:28 <russell_> in package/hostap-driver/Makefile
Nov 14 03:54:49 <gramulhaozin> hey
Nov 14 03:55:12 <gramulhaozin> russell--: what about the platform ?
Nov 14 04:59:27 <johnrw> Would Intel have left out pci init routines for via chips?
Nov 14 04:59:55 <johnrw> Makes sense...
Nov 14 05:01:00 <johnrw> Especially if Intel can say Via didn't stick to the spec...
Nov 14 05:01:51 <johnrw> Not that the net is not full of people with problems using Via Usb chips
Nov 14 05:03:16 <johnrw> and the RedBoot world is unique... in problems with via mixing with intel stuff.
Nov 14 05:30:05 <johnrw> I am beginning to think your lines are hardcoded to the 00:6:0 00:7:0 slots. Because in your Redboot lspci, they were on those slots... and mine only get there after the correction in Jungo... but then they run fine. It's when mine are up on 00:0f:0 and 00:10:0 that they cause problems. I know a kernel via quirk said if something was less than 0xf linus would make adjustments... and if...
Nov 14 05:30:07 <johnrw> ...above... it would assume the hardware knew what it was doing by being there, or something close
Nov 14 05:37:15 <johnrw> ok, here it is in pci quirks, quirk_via_acpi line 643 in my 2.6.26 kernel
Nov 14 05:39:19 <johnrw> and below that... man quirks is full of via stuff.
Nov 14 07:52:58 <CIA-8> hcg * r13200 /trunk/toolchain/uClibc/patches/160-add_openat_and_friends.patch: Add support for openat system call - this must be extended for the rest of these system calls
Nov 14 08:16:40 <russell_> thepeople: question about hostap.sh and it's /etc/config/wireless
Nov 14 08:16:58 <russell_> its
Nov 14 08:20:23 <thepeople> ask away
Nov 14 08:20:59 <russell_> it's weird
Nov 14 08:21:13 <russell_> not like what gets created for madwifi
Nov 14 08:21:26 <russell_> there is only a wifi-device section
Nov 14 08:21:38 <russell_> not a wifi-iface section
Nov 14 08:21:48 <russell_> is there a reason for that?
Nov 14 08:23:14 <russell_> for hostap.sh -> http://www.pastebin.ca/1255982
Nov 14 08:23:38 <russell_> if it worked, i would care so much ;-)
Nov 14 08:24:08 <russell_> but now that i'm going to need to fix it, i could use some direction
Nov 14 08:24:51 <thepeople> russell_: that looks like a bug of sorts in the prism configuration script
Nov 14 08:24:59 <russell_> also, the /lib/wifi/hostap.sh has some reference to 'atheros' that look wrong
Nov 14 08:25:09 <russell_> references
Nov 14 08:27:39 <russell_> thepeople: can you put me in touch with Acinonyx?
Nov 14 08:28:23 <thepeople> russell_: I normally catch him on irc
Nov 14 08:28:32 <russell_> okay, i'll watch out for him
Nov 14 08:28:39 <thepeople> I think there was a issue with the order of the interfaces
Nov 14 08:28:51 <thepeople> but I don't know as I don't have any prism cards
Nov 14 08:28:59 <russell_> i'll fiddle, see if i can come up with something better
Nov 14 08:30:01 <thepeople> ok
Nov 14 08:57:37 <CIA-8> juhosg * r13201 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ (ag71xx.h ag71xx_main.c ag71xx_mdio.c ag71xx_phy.c): [ar71xx] define some bits of the ethernet controller's registers
Nov 14 09:30:43 <russell_> Acinonyx: how come hostap.sh's /etc/config/wireless looks so different from (for example) madwifi.sh's?
Nov 14 09:44:10 <Acinonyx> hostap doesnt support vaps
Nov 14 09:44:24 <Acinonyx> device and interface is the same
Nov 14 09:44:39 <Acinonyx> thats why there is only a device config section
Nov 14 09:45:15 <russell_> what is with the references to "atheros" in hostap.sh?
Nov 14 09:45:30 <Acinonyx> there are references to atheros?
Nov 14 09:45:31 <russell_>  [ "$(config_get "$dev" type)" = "atheros" ] && wifi down "$dev"
Nov 14 09:45:36 <Acinonyx> aaaah
Nov 14 09:45:37 <Acinonyx> yes
Nov 14 09:46:24 <Acinonyx> IIRC this is to download firmwares to prism cards
Nov 14 09:46:36 <russell_> yeah, looks like it.
Nov 14 09:46:56 <russell_> it wasn't working for me, which got me to looking at it.
Nov 14 09:47:10 <russell_> not the firmware part, the whole driver
Nov 14 09:47:56 <Acinonyx> the driver needs some update
Nov 14 09:47:57 <russell_> the card isn't seeing a madwifi network nearby, maybe a b vs g thing.
Nov 14 09:48:02 <Acinonyx> it doens work with .27 kernel
Nov 14 09:48:26 <russell_> when i substituted "ANY" for the essid, it started working
Nov 14 09:48:35 <russell_> (in station mode)
Nov 14 09:49:12 <Acinonyx> did you find the cause of this?
Nov 14 09:49:27 <russell_> not yet
Nov 14 09:50:19 <Acinonyx> what are the PRI and STA firmwares on prism?
Nov 14 09:50:20 <russell_> the non-working ap is on my roof, the working on is in my basement
Nov 14 09:50:31 <Acinonyx> i mean what version
Nov 14 09:50:44 <russell_> NICID: id=0x8013 v1.0.0 (PRISM II (2.5) Mini-PCI (SST parallel flash))
Nov 14 09:50:44 <russell_> PRIID: id=0x0015 v1.1.1
Nov 14 09:50:44 <russell_> STAID: id=0x001f v1.8.0 (station firmware)
Nov 14 09:51:04 <Acinonyx> i think STA 1.7.4 is more stable
Nov 14 09:51:33 <russell_> oh, and i can't set the channel with iwconfig
Nov 14 09:51:34 <Acinonyx> NICID: id=0x8013 v1.0.0 (PRISM II (2.5) Mini-PCI (SST parallel flash))
Nov 14 09:51:34 <Acinonyx> PRIID: id=0x0015 v1.1.4
Nov 14 09:51:34 <Acinonyx> STAID: id=0x001f v1.7.4 (station firmware)
Nov 14 09:51:39 <Acinonyx> these are mine
Nov 14 09:52:35 <Acinonyx> what is the version of your kernel?
Nov 14 09:52:37 <russell_> you know what's up with iwconfig wlan0 channel 0"
Nov 14 09:52:52 <russell_> 2.6.25.20
Nov 14 09:53:02 <russell_> maybe channel 0 never worked
Nov 14 09:53:09 <Acinonyx> try channel auto
Nov 14 09:53:10 <russell_> on hostap (does on madwifi)
Nov 14 09:53:19 <russell_> nope
Nov 14 09:53:23 <Acinonyx> I don't know
Nov 14 09:53:33 <russell_> Error for wireless request "Set Frequency" (8B04) :
Nov 14 09:53:33 <russell_>     SET failed on device wifi0 ; Invalid argument.
Nov 14 09:53:53 <Acinonyx> this should work in managed mode
Nov 14 09:53:59 <russell_> you'd think
Nov 14 09:54:44 <Acinonyx> it's been a long time
Nov 14 09:54:56 <Acinonyx> i don't remember much of hostap
Nov 14 09:55:35 <russell_> yeah, i'm just working on a friends soekris+prism setup
Nov 14 09:55:47 <Acinonyx> i think you can only set channel in master mode
Nov 14 09:56:12 <Acinonyx> and managed is always auto regerdless the setting
Nov 14 09:56:16 <russell_> with madwifi, i find sometimes kismet sticks the channel and channel 0 unsticks it.
Nov 14 09:56:27 <russell_> that makes sense
Nov 14 09:56:58 <russell_> should "mode" use the standard values used for other drivers?
Nov 14 09:57:07 <russell_> e.g. "sta" instead of "managed"
Nov 14 09:57:17 <russell_> as far as the config file goes
Nov 14 09:57:53 <Acinonyx> you think is misleading?
Nov 14 09:58:05 <russell_> just different (or so it seems to me)
Nov 14 09:58:40 <Acinonyx> maybe
Nov 14 09:58:48 <Acinonyx> there is also another difference
Nov 14 09:58:58 <Acinonyx> with mac policy
Nov 14 09:59:08 <russell_> e.g. http://wiki.openwrt.org/OpenWrtDocs/KamikazeConfiguration#head-5925c40dcc0fae0849dc007b29504868170a7fe
Nov 14 09:59:40 <Acinonyx> "allow" option means excactly the opposite from what you'd think
Nov 14 09:59:50 <Acinonyx> :P
Nov 14 10:00:19 <russell_> fantastic! ;-)
Nov 14 10:00:39 <Acinonyx> it allows only the mac listed
Nov 14 10:01:13 <Acinonyx> this is how Jouni Malinen originally documented it
Nov 14 10:01:15 <Acinonyx> so i kept it
Nov 14 10:01:20 <russell_> oh, yeah, i got that though.
Nov 14 10:01:30 <russell_> intuitively, it refers to the list
Nov 14 10:01:44 <russell_> *those* should be allowed
Nov 14 10:02:00 <russell_> or *those* should be denied
Nov 14 10:03:53 <Acinonyx> also txpower is NOT in dB
Nov 14 10:04:10 <Acinonyx> it refers to tx attenuation
Nov 14 10:04:23 <russell_> heh.  iwconfig reports it as dbm
Nov 14 10:04:53 <Acinonyx> the range is -128 to 127
Nov 14 10:05:09 <russell_> how do you convert?
Nov 14 10:05:21 <Acinonyx> 127 is maximum attenutation (minimum txpower)
Nov 14 10:05:31 <Acinonyx> -128 is minimum attentuation (maximum txpower)
Nov 14 10:05:37 <Acinonyx> tottaly mesleading!
Nov 14 10:05:40 <Acinonyx> lol
Nov 14 10:05:42 <russell_> yeah
Nov 14 10:05:56 <russell_> but good for interference in the neighborhood!
Nov 14 10:06:13 <Acinonyx> you don't! each card is calibrated differently
Nov 14 10:06:20 <russell_> nice
Nov 14 10:07:04 <russell_> maybe i ought to just toss this card over the fence and make my friend buy something modern
Nov 14 10:07:45 <Acinonyx> i have a netgear ma311
Nov 14 10:08:04 <Acinonyx> it goes where atheros can't
Nov 14 10:08:07 <Acinonyx> :P
Nov 14 10:08:23 <russell_> how so?
Nov 14 10:08:23 <Acinonyx> for some reason it is extremely noise immune
Nov 14 10:08:28 <russell_> oh.
Nov 14 10:08:30 <russell_> nice
Nov 14 10:08:43 <Acinonyx> maybe because it has a hyperheterodyne receiver
Nov 14 10:09:13 <Acinonyx> (i spelled that incorrectly)
Nov 14 10:09:23 <Acinonyx> superheterodyne
Nov 14 10:09:45 <russell_> yeah, i got the idea
Nov 14 10:11:28 <Acinonyx> if you have too much noise in your area i think you should keep it
Nov 14 10:12:52 <russell_> and not give it back to my friend? ;-)
Nov 14 10:13:03 <Acinonyx> you plural
Nov 14 10:13:04 <Acinonyx> :P
Nov 14 10:13:37 <russell_> he's a poor student, so he wasn't inclined to go for another radio anyway.
Nov 14 10:13:53 <russell_> although he has lustful feelings for 802.11a
Nov 14 10:14:34 <Acinonyx> heh\
Nov 14 10:34:21 <Weedy> i want superheterodyne on minipci-e
Nov 14 10:53:51 <Weedy> war
Nov 14 10:53:54 <Weedy> wat*
Nov 14 10:54:07 <Weedy> you finaly get a new cert and is a wildcard?
Nov 14 10:55:47 <Weedy> no one every taught you about SubjectAltName?
Nov 14 11:20:13 <Acinonyx> bb all
Nov 14 11:52:41 <pensador> im having problems with uclib 0.9.29, how can i change it to 0.9.28.3?
Nov 14 16:21:33 <CIA-8> hcg * r13202 /trunk/target/linux/ps3/ (4 files in 3 dirs): Added support for 2.6.27 kernel
Nov 14 17:01:42 <CIA-8> nbd * r13203 /trunk/package/uci/Makefile: uci: upgrade to 0.6.4 - adds extended syntax to the cli and fixes a lua binding problem
Nov 14 17:03:14 <CIA-8> nbd * r13204 /branches/8.09/package/uci/Makefile: merge r13203 to 8.09
Nov 14 17:37:57 <DagMoller> AR7Port on WRTP54G: mii-tool always show eth with link status down. It is a AR7Port issue or is only related on WRTP54G?
Nov 14 17:51:35 <normunds> is ther a patch for np25g
Nov 14 18:06:33 <DagMoller> normunds, it is the answer for my question?
Nov 14 18:08:15 <normunds> its a separate question
Nov 14 22:37:37 <CIA-8> lars * r13205 /trunk/package/e2fsprogs/Makefile: Put libext2fs into its own package.
Nov 14 23:09:49 <CIA-8> florian * r13206 /trunk/package/mac80211/ (3 files in 3 dirs): Build p54 driver from compat-wireless
Nov 14 23:35:06 <CIA-8> nbd * r13207 /trunk/package/madwifi/patches/ (3 files): madwifi: add iwpriv for setting the maximum number of associated clients (ap mode)
Nov 14 23:35:19 <CIA-8> nbd * r13208 /trunk/package/madwifi/ (files/lib/wifi/madwifi.sh patches/388-apsta_fix.patch): madwifi: ap+sta race conditions
Nov 15 00:50:47 <russell_> i wonder why iwconfig of hostap-driver'd interface is showing me an extra character at the end of the ssid
Nov 15 01:04:41 <russell_> why is kmod-b43 dependent on TARGET_brcm47xx ?
Nov 15 01:54:48 <russell_> hmm.  kismet_server on hostap driver generates lots of these: SKB BUG: Invalid truesize (432) len=259, sizeof(sk_buff)=176
Nov 15 01:55:09 <russell_> probably not an openwrt problem though
**** ENDING LOGGING AT Sat Nov 15 02:59:57 2008