**** BEGIN LOGGING AT Mon Nov 12 02:59:56 2007
Nov 12 12:42:40 <slapin> ejka, ping
Nov 12 12:42:45 <slapin> hi, all!
Nov 12 13:55:08 <CIA-23> juhosg * r9543 /trunk/target/linux/adm5120/files/drivers/usb/host/ (adm5120-dbg.c adm5120-hcd.c adm5120-hub.c adm5120.h): [adm5120] USB driver cleanups, no it is still not finished
Nov 12 15:19:43 <CIA-23> juhosg * r9544 /trunk/target/linux/adm5120/image/router_le.mk: [adm5120] make build of the images a bit quiet
Nov 12 16:29:11 <CIA-23> nbd * r9545 /trunk/package/base-files/files/lib/network/config.sh: add support for interfaces with ipv6, but no ipv4 (patch from #2695)
Nov 12 16:44:04 <slapin> did anybody have problems with busybox init, when you change commandline in :respawn: line near getty (e.g. to change terminal), do kill -1 1 and it runs new getty process instead of killing off old one?
Nov 12 19:04:01 <_trine> might be interesting to some
Nov 12 19:04:07 <_trine> http://tree.celinuxforum.org/CelfPubWiki/JapanTechnicalJamboree17?action=AttachFile&do=get&target=PUG_88F5182_Buffalo.pdf
Nov 12 19:05:24 <_trine> http://tree.celinuxforum.org/CelfPubWiki/
Nov 12 19:12:52 <xaviour> nbd: Do you know much about PCI?
Nov 12 19:18:08 <nbd> a bit
Nov 12 19:18:17 <xaviour> In "ssb_fixup_pcibridge", the fixup function limits itself to the 5 official bars instead of the DEVICE_COUNT_RESOURCE=12 supported by Linux and used by the other architecture.
Nov 12 19:18:40 <xaviour> This makes that the aec62xx extension ROM is not mapped.
Nov 12 19:18:48 <xaviour> And linux disables the device.
Nov 12 19:19:11 <xaviour> Making it impossible to load the driver without enabling the device manually first.
Nov 12 19:19:36 <xaviour> You did some checkins related to ssb, didn't you?
Nov 12 19:19:46 <nbd> xaviour: i don't know much about how pci is handled on broadcom, but your comments in the ticket make sense. can you try increasing the count and let me know if it helps?
Nov 12 19:20:02 <nbd> the original limit of 6 was how the 2.4 code did it, iirc
Nov 12 19:20:26 <xaviour> Well, 6 is the official value in the PCI specification, it seams.
Nov 12 19:20:32 <xaviour> The rest is vendor defined.
Nov 12 19:20:40 <nbd> ah
Nov 12 19:20:51 <xaviour> AFAIK
Nov 12 19:21:26 <xaviour> I will give it a try one of these evenings.
Nov 12 19:22:06 <xaviour> nbd: You would not happen to be in contact with Michael Buesch?
Nov 12 19:22:17 <nbd> i am
Nov 12 19:22:35 <xaviour> He must know more.
Nov 12 19:22:54 <xaviour> If the change is done in "ssb", it will affect more devices.
Nov 12 19:24:07 <nbd> will talk to him if the change doesn't work
Nov 12 19:24:11 <xaviour> nbd: If you happen to have an opportunity to ask him about DEVICE_COUNT_RESOURCE, please do so. For my part, I will see if this solve the problem on the WL-700gE.
Nov 12 19:24:22 <nbd> ii don't think the fixup affects anything other than 47xx
Nov 12 19:24:54 <xaviour> Better safe than sorry.
Nov 12 19:25:56 <cardone> HI
Nov 12 19:25:58 <nbd> xaviour: actually i'm pretty sure, because nothing other than 47xx has a host mode pcicore
Nov 12 19:26:02 <nbd> cardone: hi
Nov 12 19:26:03 <wrt> cardone: what?
Nov 12 19:26:18 <cardone> what about fritz?!
Nov 12 19:26:26 <cardone> I am seeing it now
Nov 12 19:26:28 <nbd> what about it?
Nov 12 19:26:38 <cardone> it's danube based?
Nov 12 19:26:54 <nbd> which one?
Nov 12 19:27:10 <xaviour> nbd: What about the 63xx. It has a host mode pcicore...
Nov 12 19:27:29 <nbd> 63xx is not ssb based
Nov 12 19:27:31 <cardone> the fritz box
Nov 12 19:27:38 <xaviour> Really?
Nov 12 19:27:39 <nbd> cardone: there's more than one of those
Nov 12 19:27:44 <nbd> cardone: and most of them are based on ar7
Nov 12 19:28:05 <cardone> but there is someone based on infineon?
Nov 12 19:28:06 <nbd> xaviour: it's very different
Nov 12 19:28:23 <nbd> cardone: yes, i've heard of a new model, but i haven't actually seen it yet
Nov 12 19:28:23 <xaviour> Fine then. Then I cannot break anything.
Nov 12 19:28:39 <xaviour> Hopefully, I will try it tomorrow.
Nov 12 19:28:44 <xaviour> Will talk to you then.
Nov 12 19:28:49 <xaviour> Bye
Nov 12 19:28:55 <nbd> ok, bye
Nov 12 19:28:59 <cardone> ehhehhe may be they are working on kernel 2.6 !
Nov 12 19:29:03 <cardone> just like us
Nov 12 19:29:05 <cardone> cool!
Nov 12 21:41:27 <CIA-23> nbd * r9546 /trunk/package/busybox/patches/530-gunzip_src_fd.patch: add busybox gunzip/ipkg fix from #2691 (thanks, biblbroks)
Nov 12 21:52:06 <dhr> trine: that link to celinuxforum is interesting.  Although it is a Marvell document it seems to be released by Buffalo.  I wonder what the box actually is.
Nov 12 21:52:48 <_trine> dhr: its the buffalo Linkstation pro
Nov 12 21:52:59 <dhr> Buffalo produced the Kuro box that was the original LinkStation.
Nov 12 21:53:07 <_trine> #linkstationwiki
Nov 12 21:55:49 <dhr> http://tree.celinuxforum.org/CelfPubWiki/JapanTechnicalJamboree17 is where this was presented.  In Japanese, so I would not have understood.  (I've met the CE Linux folks a few times at the Ottawa Linux Symposium.)
Nov 12 22:30:19 <h3sp4wn> The new kurobox is a linkstation pro
Nov 12 22:35:13 <slapin_nb> ejka, ping
Nov 12 23:10:12 <slapin_nb> is there some mail bot on this channel?
Nov 12 23:10:23 <nbd> mail bot?
Nov 12 23:10:48 <slapin_nb> nbd, yeah something I could use to leave message to ejka when he will be really online
Nov 12 23:11:40 <nbd> no, we don't have such a thing
Nov 12 23:11:52 <slapin_nb> nbd, too bad for me :(
Nov 12 23:20:38 <nabcore> ejka; any joy/news with the dma/eth0 bug?
Nov 13 01:19:51 <nabcore> Actually mananged to catch the "too much work for IRQ" bug in action: https://dev.openwrt.org/ticket/2332#comment:3
Nov 13 01:20:55 <nabcore> It's interesting since after the "Algorithmics/MIPS FPU Emulator v1.5" (the point it starts to spew the error) is usual point where the preinit is started
**** ENDING LOGGING AT Tue Nov 13 02:59:56 2007