**** BEGIN LOGGING AT Thu Mar 18 02:59:56 2010
Mar 18 03:46:42 <CIA-39> thepeople * r20261 /trunk/package/firewall/files/firewall.config: allow ping
Mar 18 07:18:14 <[florian]> rtz2: that's better, but now I get this error: "# Error: fwsize != imghdr->imagesz"
Mar 18 08:41:14 <blogic> rtz2: ping
Mar 18 08:41:24 <blogic> rtz2: got any patche si can already look at ?
Mar 18 08:41:34 <blogic> i am starting to implement mips-machine for itfxmips just mow
Mar 18 08:41:44 <blogic> and i can ideally already strat adding amazon while i am at it
Mar 18 08:41:45 <blogic> ?!
Mar 18 09:06:37 <builder> build #28 of ixp4xx is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/28
Mar 18 09:20:45 <CIA-39> blogic * r20262 /trunk/target/linux/ifxmips/files/arch/mips/pci/pci-ifxmips.c: [ifxmips] Calculate PCI BARMASK11 register value dynamicly, signed off by Ithamar R. Adema
Mar 18 10:45:16 <CIA-39> nico * r20263 /trunk/package/madwifi/Makefile: [package] madwifi: remove redundant conditional dep, atheros has PCI support
Mar 18 10:56:03 <CIA-39> blogic * r20264 /trunk/target/linux/ifxmips/files/arch/mips/ (ifxmips/board.c pci/pci-ifxmips.c): [ifxmips] Make it possible for boards to disable PCI REQ signals in the PCI controller, signed off by Ithamar R. Adema
Mar 18 10:56:30 <CIA-39> blogic * r20265 /trunk/target/linux/ifxmips/files/arch/mips/include/asm/mach-ifxmips/gpio.h: [ifxmips] we have a maximum of 32 gpios
Mar 18 11:01:05 <error32> Are there any downsides to using eglibc instead of uclibc?
Mar 18 11:02:32 <[florian]> it's a bit bigger
Mar 18 11:02:47 <[florian]> but it's just a configurable glibc, so you have maximum compatibility with applications
Mar 18 11:03:06 <[florian]> and the linux system call coverage/features is usually better as well
Mar 18 11:17:47 <CIA-39> blogic * r20266 /trunk/package/br2684ctl/files/br2684ctl: [br2684ctl] Allow UCI configuration of ATM QOS parameters per interface, signed off by Ithamar R. Adema
Mar 18 11:20:36 <CIA-39> blogic * r20267 /trunk/target/linux/ifxmips/image/Makefile: [ifxmips] Also build jffs2 images with 128k eraseblock size, signed off by Ithamar R. Adema
Mar 18 11:24:54 <error32> I guess I'll try eglibc then because (if I have diagnosed correctly) my application hangs due to a threading bug(in uclibc) in libsqlite
Mar 18 12:20:39 <CIA-39> markus * r20268 /trunk/target/linux/brcm47xx/Makefile: added feature "pcmcia"
Mar 18 12:43:49 <builder> build #28 of rb532 is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/rb532/builds/28
Mar 18 12:50:11 <rtz2> [florian]: that's pretty bad
Mar 18 12:50:36 <rtz2> [florian]: rewriting the header is probably not avoidable :/
Mar 18 12:51:57 <[florian]> rtz2: I think so
Mar 18 12:52:07 <[florian]> rtz2: now I remember why sn9 did such a complicated flash map driver
Mar 18 12:52:30 <rtz2> [florian]: i will come up with something
Mar 18 12:52:39 <[florian]> rtz2: allright
Mar 18 12:52:48 <[florian]> rtz2: I can test it tonight
Mar 18 12:52:56 <rtz2> [florian]: ok
Mar 18 13:01:35 <CIA-39> blogic * r20269 /trunk/package/ifxmips-dsl-api/patches/200-mei_compat.patch: adds misisng include in ifxmips dsl driver compat layer
Mar 18 13:02:30 <CIA-39> blogic * r20270 /trunk/package/ifxmips-dsl-api/ (Config.in Makefile patches/400-debug-output.patch): [ifxmips-dsl-api] Add option to disable debug output of DSL drivers signed of by Ithamar R. Adema
Mar 18 13:29:53 <error32> well this is rather odd, somehow I cannot select a different version of gcc other than 3.4.6
Mar 18 13:53:24 <CIA-39> blogic * r20271 /trunk/target/linux/ifxmips/files/arch/mips/ifxmips/board.c: [ifxmips] fixes board detection, signed off by Ithamar R. Adema
Mar 18 13:54:24 <CIA-39> blogic * r20272 /trunk/target/linux/ifxmips/patches-2.6.30/140-mtd.patch: [ifxmips] make mtd handel 128k sectors properly, based on patch by Ithamar R. Adema
Mar 18 13:55:18 <CIA-39> blogic * r20273 /trunk/target/linux/ifxmips/patches-2.6.30/120-serial.patch: [ifxmips] Bugfix: do not call request_irq() with interrupts disabled, signed off by Ithamar R. Adema
Mar 18 14:32:52 <CIA-39> mirko * r20274 /trunk/target/linux/xburst/qi_lb60/config-2.6.32: do not use font <FONT_MINI_4x6> for target lb60 by default as it's unreadable
Mar 18 15:14:02 <augurel> hi there, i want to use arp on my Openwrt kamikaze, but it says "/bin/ash: arp: not found"
Mar 18 15:15:10 <xMff> ping cshore
Mar 18 15:16:18 <augurel> when i connect via ssh and call arp, it is working but when i try to execute via C++ code with using popen it dowsn't
Mar 18 15:16:37 <augurel> xMff : what do u mean "ping cshore" ??
Mar 18 15:20:05 <rtz2> augurel: it's a link to busybox
Mar 18 15:20:15 <rtz2> augurel: call /bin/busybox arp
Mar 18 15:21:35 <xMff> augurel: this means I want to talk to cshore
Mar 18 15:21:57 <augurel> i rebooted now arp is working via serial and ssh (not with popen). When i say "/bin/busybox arp" "arp: applet not found"
Mar 18 15:21:58 <rtz2> augurel: or call busybox with arp as argc[0]
Mar 18 15:22:03 <rtz2> argv[0]
Mar 18 15:22:22 <rtz2> augurel: ok, maybe I was wrong
Mar 18 15:22:36 <xMff> rtz2: actually arp is a shell procedure in /etc/profile
Mar 18 15:22:57 <rtz2> this could be an explaination
Mar 18 15:23:02 <xMff> [ -x /sbin/arp ] || arp() { cat /proc/net/arp; }
Mar 18 15:23:43 <augurel> xMff : thank u so much
Mar 18 15:23:46 <xMff> augurel: just prase /proc/net/arp directly, this is the most portable solution
Mar 18 15:24:17 <xMff> and you avoid a fork
Mar 18 15:24:37 <augurel> forking of a process with popen ??
Mar 18 15:24:47 <xMff> yes
Mar 18 15:24:59 <xMff> popen does fork() and exec() behind the scenes
Mar 18 15:25:04 <xMff> and pipe()
Mar 18 15:25:10 <xMff> it is a heavy operation
Mar 18 15:25:49 <augurel> i see
Mar 18 15:50:07 <CIA-39> thepeople * r20275 /trunk/package/dnsmasq/files/dnsmasq.init: fix limit
Mar 18 16:01:42 <CIA-39> thepeople * r20276 /packages/net/xl2tpd/ (11 files in 3 dirs): (log message trimmed)
Mar 18 16:01:42 <CIA-39> this patch updates xl2tpd to 1.2.5. From the changelog:
Mar 18 16:01:42 <CIA-39> * Fix initscript for
Mar 18 16:01:42 <CIA-39> https://bugzilla.redhat.com/show_bug.cgi%3Fid%3D247100
Mar 18 16:01:42 <CIA-39> * Fix for two Windows machines behind the same NAT with the same
Mar 18 16:01:42 <CIA-39> number of l2tp connection attempts since boot [Shinichi Furuso]
Mar 18 16:01:43 <CIA-39> Changes in this patch:
Mar 18 16:46:59 <builder> build #27 of ar7 is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/27
Mar 18 17:22:21 <cshore> pong xMff
Mar 18 17:27:22 <xMff> cshore: wanted to know whether https://dev.openwrt.org/ticket/5441 is still relevant
Mar 18 17:29:28 <cshore> xMff: not sure: I worked around it by doing a redir in/out from /dev/null
Mar 18 17:29:59 <cshore> xMff: I can check later if you want
Mar 18 17:30:19 <xMff> would be nice
Mar 18 19:18:38 <CIA-39> juhosg * r20277 /trunk/target/linux/ar71xx/files/arch/mips/include/asm/mach-ar71xx/ar71xx.h: ar71xx: fix ar724x_pci_wr{,_nf} helpers
Mar 18 19:18:49 <CIA-39> juhosg * r20278 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/setup.c: ar71xx: remove unnecessary includes from setup.c
Mar 18 19:18:50 <CIA-39> juhosg * r20279 /trunk/target/linux/ar71xx/files/arch/mips/ (3 files in 2 dirs): ar71xx: merge AR71XX_IRQ_CPU_{PCI,WMAC} into AR71XX_IRQ_CPU_IP2
Mar 18 19:18:52 <CIA-39> juhosg * r20280 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/irq.c: ar71xx: use set_irq_chained_handler for the PCI IRQs
Mar 18 19:18:56 <CIA-39> juhosg * r20281 /trunk/target/linux/ar71xx/files/arch/mips/ (ar71xx/irq.c pci/pci-ar71xx.c pci/pci-ar724x.c): ar71xx: move PCI intterupt handling code to pci-ar7{1xx,24x}.c
Mar 18 19:19:06 <CIA-39> juhosg * r20282 /trunk/target/linux/ar71xx/files/arch/mips/ (include/asm/mach-ar71xx/ar71xx.h pci/pci-ar724x.c): ar71xx: move ar724x_pci_* helpers to ar724x-pci.c
Mar 18 19:19:06 <CIA-39> juhosg * r20283 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar724x.c: ar71xx: add error handling to ar724x_pcibios_init
Mar 18 19:19:09 <CIA-39> juhosg * r20284 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar724x.c: ar71xx: introduce global ar724x_pci_ctrl_base variable
Mar 18 19:19:12 <CIA-39> juhosg * r20285 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar724x.c: ar71xx: optimize register access in ar724x_pci.c
Mar 18 19:19:15 <CIA-39> juhosg * r20286 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/irq.c: ar71xx: optimize register access in irq.c
Mar 18 19:19:18 <CIA-39> juhosg * r20287 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar71xx.c: ar71xx: optimize register access in ar71xx_pci.c
Mar 18 19:19:21 <CIA-39> juhosg * r20288 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar71xx.c: ar71xx: check returned value of ioremap in ar71xx_pcibios_setup
Mar 18 19:19:23 <CIA-39> juhosg * r20289 /trunk/target/linux/generic-2.6/base-files/init: base-files: show boot state on intramfs images
Mar 18 19:25:37 <lfcorreia> blogic_, hi, you there?
Mar 18 19:53:46 <blogic_> lfcorreia: hi
Mar 18 19:55:00 <lfcorreia> hi
Mar 18 19:55:29 <lfcorreia> I talked today with nbd about supporting a new ralink based boars in owrt
Mar 18 19:55:41 <lfcorreia> he re-directed me to you :)
Mar 18 19:58:13 <lfcorreia> so I bought this SMC WBR14S-N4 which has an RT3050 1T1R 11n wireless
Mar 18 19:58:35 <lfcorreia> and I wanted to know what is the best way to create a new set of file to support it :)
Mar 18 20:10:08 <lfcorreia> blogic_, any idea is welcomed
Mar 18 20:14:36 <lfcorreia> blogic_, the image I created for that other Ralink board also 'runs' in this board
Mar 18 20:14:51 <[florian]> rtz2: do you have a new patch for me to test?
Mar 18 20:15:20 <[florian]> rtz2: would you like to test only the kernel changes to see if that works?
Mar 18 20:18:35 <rtz2> [florian]: 5 min
Mar 18 20:19:16 <rtz2> [florian]: let me make sure, it compiles
Mar 18 20:32:46 <rtz2> [florian]: http://openwrt.pastebin.com/A214fV3C
Mar 18 20:33:18 <rtz2> [florian]: this is the new boards/ar525w.c file
Mar 18 20:33:21 <rtz2> [florian]: this also needs a small kernel patch I didn't include
Mar 18 20:33:55 <rtz2> [florian]: there is a static erase_write function in mtdbock.c
Mar 18 20:34:17 <rtz2> [florian]: you have to remove the static from it and add a declaration for it in mtd.h
Mar 18 20:34:53 <rtz2> [florian]: another possibility would be, to copy this function and the  erase_callback function over to ar525w.x
Mar 18 20:37:29 <rtz2> [florian]: and this is the new makefile snipped: http://openwrt.pastebin.com/4VT4hrUC
Mar 18 20:37:43 <rtz2> [florian]: I can also create a proper patch, if you like this better
Mar 18 20:38:39 <[florian]> rtz2: that's fine, thank you
Mar 18 20:38:41 <[florian]> rtz2: let me test this
Mar 18 20:39:25 <blogic_> lfcorreia: soz was away
Mar 18 20:39:26 <blogic_> ermmmm
Mar 18 20:39:35 <blogic_> is it spi based ?
Mar 18 20:39:40 <[florian]> rtz2: are you using vim?
Mar 18 20:39:42 <blogic_> if so i can merge a spi driver i guess
Mar 18 20:40:20 <rtz2> [florian]: no, kwrite, but in a vm with the cygwin x server running on windows
Mar 18 20:40:25 <rtz2> [florian]: why?
Mar 18 20:40:32 <lfcorreia> blogic_, is the spi question for me (sorry for being dumb)
Mar 18 20:40:40 <[florian]> rtz2: ok, because you got quite a lot of whitespace at each new line actually
Mar 18 20:41:27 <rtz2> [florian]: where for example?
Mar 18 20:41:32 <blogic_> lfcorreia: ok
Mar 18 20:41:38 <blogic_> what kind of flash is it ?
Mar 18 20:41:54 <blogic_> a small chip with 8 or 16 legs or a big one with 30-40 small ones ?
Mar 18 20:42:03 <lfcorreia> blogic_, it's a 4Mb flash chip, standard paralell MX
Mar 18 20:42:05 <blogic_> 30-40 -> NOR we support it
Mar 18 20:42:10 <blogic_> ah cool
Mar 18 20:42:13 <blogic_> that will work
Mar 18 20:42:17 <blogic_> what do you need exactly ?
Mar 18 20:42:23 <[florian]> rtz2: line 64 of the ar525w board file and every new line then
Mar 18 20:42:45 <blogic_> i know those chipsets really well, so ask  ... :)
Mar 18 20:42:53 <lfcorreia> to know how to create owrt board configs
Mar 18 20:43:00 <blogic_> !
Mar 18 20:43:01 <blogic_> ?
Mar 18 20:43:13 <blogic_> you mean a mips-machine thingy ?
Mar 18 20:43:18 <blogic_> so the baord is detected ?
Mar 18 20:43:24 <lfcorreia> blogic_, yes :)
Mar 18 20:43:27 <blogic_> ahhhh :)
Mar 18 20:43:34 <rtz2> [florian]: it autogenerates tabs and doesn't remove them
Mar 18 20:43:38 <blogic_> do you know anything about the board already ?
Mar 18 20:43:46 <rtz2> [florian]: I can remove them manually, if this is a problem
Mar 18 20:43:48 <blogic_> or is this a clean box fromt he shelf ?
Mar 18 20:43:57 <lfcorreia> and same goes for my other rt3052-v24rw-2x2
Mar 18 20:43:58 <[florian]> rtz2: no it's fine, I can fix that up myself
Mar 18 20:44:22 <lfcorreia> blogic_, i've got it open in front of me, it has a 3.3v serial port
Mar 18 20:44:30 <blogic_> oki
Mar 18 20:44:36 <blogic_> 4lan 1 wan ?
Mar 18 20:44:37 <lfcorreia> blogic_, it is fresh from the shelf
Mar 18 20:44:40 <lfcorreia> yup
Mar 18 20:44:43 <blogic_> oki
Mar 18 20:44:45 <lfcorreia> 1T1R wireless
Mar 18 20:44:46 <blogic_> well
Mar 18 20:44:52 <blogic_> use the fonera20n image
Mar 18 20:45:04 <blogic_> and tell me if it works :)
Mar 18 20:45:10 <lfcorreia> it boots it over tftp
Mar 18 20:45:13 <blogic_> in theory the fon20n image is for the ref bord
Mar 18 20:45:16 <blogic_> oki
Mar 18 20:45:19 <blogic_> ethernet works ?
Mar 18 20:45:25 <lfcorreia> but it seems the eth isn't working that well
Mar 18 20:45:35 <lfcorreia> let me check
Mar 18 20:45:39 <blogic_> ok
Mar 18 20:45:43 <blogic_> not well as in ?!
Mar 18 20:45:59 <lfcorreia> again, it's my owrt problems :)
Mar 18 20:46:16 <blogic_> hehe
Mar 18 20:46:21 <lfcorreia> the eth port where I have the eth cable in, got an IP address
Mar 18 20:46:27 <blogic_> ok
Mar 18 20:46:27 <lfcorreia> eth0.2
Mar 18 20:46:35 <lfcorreia> but it won't ping
Mar 18 20:46:40 <blogic_> what ip ?
Mar 18 20:46:43 <blogic_> well
Mar 18 20:46:46 <lfcorreia> wait
Mar 18 20:46:59 <lfcorreia> from the board it resolves names to IP's
Mar 18 20:47:05 <blogic_> ok
Mar 18 20:47:05 <lfcorreia> so DNS config is working
Mar 18 20:47:18 <lfcorreia> aha
Mar 18 20:47:24 <lfcorreia> it pings my router
Mar 18 20:47:28 <lfcorreia> and my server
Mar 18 20:47:31 <blogic_> so, that is the wan port
Mar 18 20:47:59 <lfcorreia> actually it's one of the yellow lan ports
Mar 18 20:47:59 <blogic_> is that also the wan port of the casing ? or are you on a lan port ?
Mar 18 20:48:03 <blogic_> yes yes
Mar 18 20:48:10 <blogic_> ralink has a 4/1 and 1/4 mapping
Mar 18 20:48:19 <lfcorreia> let me connect the cable to the wan port
Mar 18 20:48:22 <lfcorreia> one sec
Mar 18 20:48:24 <blogic_> ok
Mar 18 20:48:29 <blogic_> wan should therefor be lan
Mar 18 20:48:40 <blogic_> the current driver does a 4/1 config
Mar 18 20:48:47 <blogic_> we need to make the swconfig stub
Mar 18 20:49:15 <lfcorreia> let me reset the board
Mar 18 20:49:28 <blogic_> ok
Mar 18 20:49:43 <blogic_> i am afraid i need to leave in a few sec
Mar 18 20:49:44 <lfcorreia> the default serial port is 115200
Mar 18 20:49:48 <blogic_> is it ?
Mar 18 20:49:50 <blogic_> ok
Mar 18 20:49:57 <blogic_> normally ralink uses 57600
Mar 18 20:49:59 <lfcorreia> ok, i'm also going to be away next week
Mar 18 20:50:04 <blogic_> oki
Mar 18 20:50:11 <lfcorreia> but when I return we'll sort this out
Mar 18 20:50:13 <lfcorreia> ah
Mar 18 20:50:14 <lfcorreia> also
Mar 18 20:50:16 <blogic_> i saw there is already a soc module in the mac80211 driver
Mar 18 20:50:19 <lfcorreia> the manufacturer
Mar 18 20:50:29 <lfcorreia> has the firmware source in the website
Mar 18 20:50:33 <blogic_> ok
Mar 18 20:50:35 <blogic_> link ?
Mar 18 20:50:36 <lfcorreia> including the crap ralink driver
Mar 18 20:50:37 <[florian]> rtz2: not bad: http://openwrt.pastebin.com/KRRMPsgH
Mar 18 20:50:44 <lfcorreia> just a sec
Mar 18 20:50:44 <blogic_> lfcorreia: haha
Mar 18 20:50:47 <blogic_> goodie
Mar 18 20:50:50 <blogic_> gimme
Mar 18 20:50:51 <blogic_> :)
Mar 18 20:51:12 <blogic_> see if they did anything usefull
Mar 18 20:51:18 <blogic_> lets see if they did anything usefull
Mar 18 20:51:22 <lfcorreia> not really
Mar 18 20:51:29 <blogic_> ok
Mar 18 20:51:30 <lfcorreia> the kernel they use is 2.6.21
Mar 18 20:51:37 <blogic_> yes, default ralink
Mar 18 20:51:50 <blogic_> how much work do you think it will be to make those chips work ?
Mar 18 20:52:05 <rtz2> [florian]: could you upload the image somewhere?
Mar 18 20:52:07 <blogic_> i am sort of willing to help
Mar 18 20:52:09 <[florian]> rtz2: sure
Mar 18 20:52:15 <lfcorreia> actually it will depend on how hard it is to do testis using owrt development system :)
Mar 18 20:52:19 <blogic_> not done much wifi yet, but u gotta start one day i guess
Mar 18 20:52:35 <lfcorreia> blogic_, firmware source http://www.smc.com/files/AC/GPL_SMCWBR14S-N4.zip
Mar 18 20:52:39 <[florian]> rtz2: http://alphacore.org/~florian/openwrt/rdc/openwrt-rdc-squashfs-ar525w.img-new
Mar 18 20:52:45 <blogic_> k
Mar 18 20:52:54 <blogic_> thx
Mar 18 20:53:06 <lfcorreia> we are working on soc
Mar 18 20:53:16 <lfcorreia> i've got a friend, Helmut Schaa
Mar 18 20:53:33 <lfcorreia> which is also interested in it, and since i'm one of the rt2x00 team...
Mar 18 20:53:33 <blogic_> ok
Mar 18 20:53:39 <blogic_> indeed
Mar 18 20:53:42 <rtz2> [florian]: and does it work after a reboot?
Mar 18 20:53:42 <blogic_> well
Mar 18 20:53:47 <blogic_> willing to help
Mar 18 20:53:58 <blogic_> even if only testing
Mar 18 20:53:59 <[florian]> rtz2: let me check that
Mar 18 20:54:00 <lfcorreia> and I thank you for that
Mar 18 20:54:04 <rtz2> [florian]: by work I mean still boot
Mar 18 20:54:09 <lfcorreia> just help me with owrt
Mar 18 20:54:09 <rtz2> or start to boot
Mar 18 20:54:19 <[florian]> rtz2: nope, it's waiting for a new firmware
Mar 18 20:54:26 <blogic_> yes, but i want to follow the wifi stuff to understand it better
Mar 18 20:54:32 <[florian]> # Firmware Checksum Fail!! expected[A75DC0B6], got[3DD7E984]
Mar 18 20:54:35 <blogic_> one of the few areas i have not poked around in yet
Mar 18 20:54:39 <rtz2> [florian]: ok, give me a minute
Mar 18 20:54:47 <lfcorreia> blogic_, on how to create and add patches mainly local for testing
Mar 18 20:55:05 <lfcorreia> blogic_, wifi for ralink devices is kinda easy
Mar 18 20:55:10 <blogic_> lfcorreia: good
Mar 18 20:55:19 <lfcorreia> the drvier is well structured, nothing like ralink's own crap
Mar 18 20:55:37 <lfcorreia> although i'm not really the man to explain it all
Mar 18 20:55:42 <blogic_> lfcorreia: let me know your mail addr again please and i will send you a small "howto make patches in owrt the proper way" :)
Mar 18 20:55:52 <blogic_> yes
Mar 18 20:55:56 <blogic_> i will just read it i guess
Mar 18 21:05:27 <rtz2> [florian]: this is a new makefile snipped: http://openwrt.pastebin.com/4yfJpy6v
Mar 18 21:05:42 <rtz2> [florian]: and could you add this to ar525w.c around line 66-67?
Mar 18 21:05:43 <rtz2> 	printk(KERN_INFO "Fixing up AR525W header, old image size: %u, new image size: %u\n",
Mar 18 21:05:43 <rtz2> 		   header->imagesz, bufferlength);
Mar 18 21:07:10 <thepeople> cshore: ping
Mar 18 21:09:13 <thepeople> cshore: would you be able to look at https://dev.openwrt.org/ticket/6719
Mar 18 21:09:22 <rtz2> blogic_: if you write such a thing, you could post it into the wiki ;)
Mar 18 21:10:17 <blogic_> yes
Mar 18 21:10:59 <CIA-39> jow * r20290 /trunk/scripts/remote-gdb:
Mar 18 21:10:59 <CIA-39> [scripts] remote-gdb:
Mar 18 21:10:59 <CIA-39> * use sysroot instead of solib search paths
Mar 18 21:10:59 <CIA-39> * don't query arch if there's only one choice
Mar 18 21:11:04 <blogic_> 22:09 <blogic_> make package/abc/{clean,prepare,compile,install} V=99 QUILT=1
Mar 18 21:11:13 <blogic_> 22:09 <blogic_> cd build_dir..../abc/
Mar 18 21:11:14 <blogic_> 22:09 <blogic_> quilt series .....
Mar 18 21:11:14 <blogic_> 22:09 <blogic_> quilt refresch
Mar 18 21:11:14 <blogic_> 22:09 <blogic_> cd -
Mar 18 21:11:15 <blogic_> 22:09 <blogic_> make package/abc/update
Mar 18 21:11:31 <blogic_> 22:10 <blogic_> ls package/abc/patches/ now has your quilt series in it
Mar 18 21:11:34 <blogic_> that is all really :)
Mar 18 21:11:45 <xMff> compile + install are not needed when prearing :)
Mar 18 21:14:49 <blogic_> xMff: jaja
Mar 18 21:18:28 <builder> build #44 of ppc44x is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/44
Mar 18 21:19:56 <xMff> blogic_: ;)
Mar 18 21:20:12 <[florian]> rtz2: http://openwrt.pastebin.com/eagM75se
Mar 18 21:22:48 <rtz2> [florian]: does it survive a reboot?
Mar 18 21:24:52 <[florian]> rtz2: nop
Mar 18 21:25:17 <rtz2> [florian]: ok, could you upload the image again?
Mar 18 21:25:49 <[florian]> rtz2: same location
Mar 18 21:41:05 <rtz2> [florian]: cound it
Mar 18 21:41:15 <rtz2> [florian]: check line 78 of ar525w.c
Mar 18 21:41:39 <rtz2> [florian]: there are 4 lines starting with header and 3 memcpy afterwards
Mar 18 21:41:52 <rtz2> [florian]: you have to swap the blocks
Mar 18 21:42:06 <CIA-39> bud * r20291 /packages/utils/ntfs-3g/Makefile:
Mar 18 21:42:06 <CIA-39> - bump version to 2010.3.6 for uClibc 0.9.30.2+
Mar 18 21:42:06 <CIA-39> - added ntfs-3g.probe to package ntfs-3g
Mar 18 21:42:06 <CIA-39> - menuconfig reworked: external fuse now bool selector
Mar 18 21:42:06 <CIA-39> - new package ntfs-3g-utils containing all binaries except ntfs-3g, ntfs-3g.probe
Mar 18 21:42:16 <rtz2> [florian]: or wait, one moemtn
Mar 18 21:42:57 <[florian]> rtz2: of course
Mar 18 21:43:39 <rtz2> [florian]: they are right the way the are (I think)
Mar 18 21:43:49 <rtz2> there has to be another problem
Mar 18 21:50:15 <[florian]> rtz2: shall I try swapping the blocks like you suggested?
Mar 18 21:50:55 <rtz2> [florian]: no
Mar 18 21:51:18 <rtz2> [florian]: I suspect it's some kind of byte endianess problem
Mar 18 21:59:03 <rtz2> [florian]: could you swab32 the checksums?
Mar 18 21:59:09 <rtz2> header->fastcksum = swab32(crc32(0xFFFFFFFF, crcbuf, sizeof(crcbuf)));
Mar 18 21:59:15 <rtz2> somehting like this
Mar 18 21:59:18 <rtz2> for both of them
Mar 18 21:59:27 <[florian]> sure
Mar 18 22:02:55 <[florian]> rtz2: this is not too good: jffs2: Flash size not aligned to erasesize, reducing to 704KiB
Mar 18 22:04:50 <[florian]> rtz2: I know get # Firmware Checksum Fail!! expected[94C1F40F], got[1C6DBF59]
Mar 18 22:07:11 <rtz2> [florian]: the first error is another problem it tries to mount the kernel image as the root filesystem
Mar 18 22:07:27 <[florian]> rtz2: you are right
Mar 18 22:11:41 <CIA-39> florian * r20292 /packages/ipv6/dibbler/Makefile: [package] fix dibbler dependencies (#6879)
Mar 18 22:12:12 <rtz2> [florian]: ok, this should be fixed by this change: http://openwrt.pastebin.com/3a4NtFt1
Mar 18 22:12:52 <rtz2> [florian]: and I think I understand the other problem, give me a moment
Mar 18 22:15:31 <cshore> [florian]: could you look at the /jffs to /overlay patch I sent to the mailing list a couple weeks ago.  Felix hasn't had to time to look at it and commit
Mar 18 22:18:34 <rtz2> [florian]: here is the complete ar525w.c file again: http://openwrt.pastebin.com/nzgAJjE8
Mar 18 22:20:55 <[florian]> cshore: ok, why not
Mar 18 22:20:58 <[florian]> rtz2: hold on
Mar 18 22:22:47 <[florian]> rtz2: line 56, this should be __initdata, you cannot annoate a variable __init
Mar 18 22:23:03 <rtz2> [florian]: ohh, right
Mar 18 22:24:07 <[florian]> rtz2: it just works :)
Mar 18 22:24:18 <[florian]> rtz2: let me check if it survives a reboot
Mar 18 22:25:24 <[florian]> rtz2: but it does not yet survive a reboot
Mar 18 22:25:52 <[florian]> rtz2: # Firmware Checksum Fail!! expected[010D5C73], got[9155A5CD]
Mar 18 22:26:46 <rtz2> [florian]: could you flash again and create a dump of the firmware partition and upload both the dump and the original image?
Mar 18 22:27:28 <[florian]> rtz2: sir yes sir
Mar 18 22:27:39 <rtz2> :) thanks
Mar 18 22:31:53 <[florian]> rtz2: http://alphacore.org/~florian/openwrt/rdc/firmware
Mar 18 22:35:23 <rtz2> [florian]: and this prints out expected[010D5C73], got[9155A5CD]?
Mar 18 22:35:46 <[florian]> # Firmware Checksum Fail!! expected[9B447AAA], got[9155A5CD]
Mar 18 22:37:20 <rtz2> [florian]: and this was the same image, that also created the mesage expected[010D5C73], got[9155A5CD]?
**** BEGIN LOGGING AT Thu Mar 18 22:39:53 2010
Mar 18 22:45:49 <[florian]> rtz2: /win
Mar 18 22:46:00 <[florian]> rtz2: sorry, I was without connection for a couple of minutes
Mar 18 22:46:15 <[florian]> rtz2: it is the same image yes
Mar 18 22:51:58 <rtz2> [florian]: stupid
Mar 18 22:52:08 <rtz2> [florian]: take a look at the second memcpy line
Mar 18 22:52:20 <rtz2> [florian]: it has to be memcpy(crcbuf + 0x100, buffe
Mar 18 22:52:39 <[florian]> yeah
Mar 18 22:52:50 <[florian]> otherwise you are overwriting it
Mar 18 22:57:38 <[florian]> rtz2: it survives a reboot, really good job!
Mar 18 22:57:57 <[florian]> rtz2: let me commit your changes
Mar 18 22:58:47 <rtz2> [florian]: could you try, if the in kernel crc function works?
Mar 18 22:59:16 <rtz2> [florian]: this would avoid having to duplicate this thing
Mar 18 22:59:18 <[florian]> rtz2: oh, the one in lib/ ?
Mar 18 22:59:23 <rtz2> [florian]: yes
Mar 18 22:59:25 <[florian]> rtz2: sure, let me try that
Mar 18 22:59:33 <[florian]> rtz2: it could save a couple of bytes
Mar 18 23:00:14 <rtz2> [florian]: readd the crc32.h include and add 0xFFFFFFFF as first argument to the crc32 function calls
Mar 18 23:00:48 <rtz2> but it may end up with the wrong endianess, not sure
Mar 18 23:01:01 <[florian]> rtz2: we will see
Mar 18 23:01:31 <[florian]> rtz2: you can choose between crc32_le and be
Mar 18 23:02:51 <rtz2> [florian]: I know, but be uses a complete different table and the le table is the same as the one in ar525w.c but with different endianess
Mar 18 23:03:01 <rtz2> [florian]: not sure, if this will make a difference
Mar 18 23:03:12 <[florian]> rtz2: ok, let's ee
Mar 18 23:03:51 <[florian]> rtz2: I do not know if you saw it, but I actually sent the southbridge, gpio and watchdog changes upstream
Mar 18 23:04:06 <rtz2> [florian]: I didn't
Mar 18 23:04:16 <rtz2> [florian]: did they end up in .34?
Mar 18 23:04:36 <[florian]> http://patchwork.kernel.org/patch/84820/
Mar 18 23:04:48 <[florian]> rtz2: unfortunately no, the maintainer still has not applied them
Mar 18 23:06:02 <rtz2> [florian]: well, maybe you can convince him, by telling him, that the current wdt driver writes to the registers of random pci devices
Mar 18 23:06:12 <rtz2> so it's clearly a bugfix :)
Mar 18 23:06:44 <[florian]> rtz2: I was willing to first get him merge the southbridge, then send your watchdog fixes
Mar 18 23:06:57 <rtz2> [florian]: ok
Mar 18 23:06:58 <[florian]> rtz2: such that we do not end-up bugfixing with the "old" pci accessors
Mar 18 23:08:07 <rtz2> [florian]: you know better how to do this
Mar 18 23:08:21 <[florian]> rtz2: so you are correc, the kernel crc32 function did not work out
Mar 18 23:08:28 <[florian]> rtz2: let's stick with yours
Mar 18 23:08:53 <rtz2> [florian]: you checked, if the endianess of the expected and actual crc was wrong?
Mar 18 23:09:05 <[florian]> # Firmware Checksum Fail!! expected[A373C04D], got[5C8C3FB2]
Mar 18 23:09:25 <rtz2> [florian]: completly different
Mar 18 23:09:31 <[florian]> definitively
Mar 18 23:09:38 <rtz2> [florian]: ok, it was worth the test
Mar 18 23:09:43 <[florian]> yup
Mar 18 23:09:56 <[florian]> let's commit the right things now ;)
Mar 18 23:15:33 <rtz2> [florian]: by the way, do you have any detailed knowldege about the ssb bus used on broadcom cpus?
Mar 18 23:19:21 <[florian]> rtz2: are you aware of some bcm5365 datasheet?
Mar 18 23:19:44 <[florian]> rtz2: there are quite some details about how ssb works inside
Mar 18 23:24:37 <[florian]> rtz2: did you find it?
Mar 18 23:25:08 <rtz2> [florian]: no
Mar 18 23:25:19 <rtz2> [florian]: only some 2page abstracts
Mar 18 23:27:43 <[florian]> seems like it's harder to find
Mar 18 23:28:34 <rtz2> [florian]: because on the asus rt n16, the default method to scan the bus doesn't seem to work
Mar 18 23:28:44 <[florian]> rtz2: http://voodoowarez.com/bcm5365p.pdf
Mar 18 23:30:03 <rtz2> [florian]: thanks
Mar 18 23:35:18 <CIA-39> florian * r20293 /trunk/tools/firmware-utils/src/airlink.c: [tools] fix file creation mode, thanks rtz
Mar 18 23:35:23 <CIA-39> florian * r20294 /trunk/ (16 files in 10 dirs):
Mar 18 23:35:23 <CIA-39> [rdc] rework board detection
Mar 18 23:35:23 <CIA-39> Rework board detection, separate board specific code into its own file. As a
Mar 18 23:35:23 <CIA-39> result we also change the way rdc images are generated.
Mar 18 23:35:23 <CIA-39> Support for board which required binary tools, like AMIT are dropped. Patch by
Mar 18 23:35:23 <CIA-39> Bernhard Loos.
Mar 18 23:36:24 <[florian]> rtz2: so now there is the ethernet driver to take care of :)
Mar 18 23:41:21 <[florian]> rtz2: I am off to bed now, can you check with a new build that I merged everything correctly?
Mar 18 23:41:52 <rtz2> [florian]: no, I think you forgot the header file
Mar 18 23:42:02 <rtz2> [florian]: one moment
Mar 18 23:43:43 <rtz2> [florian]: http://openwrt.pastebin.com/Gv9UWkrx
Mar 18 23:44:00 <rtz2> [florian]: this also contains the fix for the wdt driver
Mar 18 23:44:20 <xMff> l50-51 look duplicated
Mar 18 23:45:01 <rtz2> xMff: right, they are
Mar 18 23:45:54 <rtz2> [florian]: deleteing the amit target may not be a good idea yet, not sure how the buildsystem will react to those files missing
Mar 18 23:46:15 <rtz2> [florian]: I can prepare a generall cleanup patch till tomorrow
Mar 18 23:46:29 <[florian]> rtz2: cool
Mar 19 00:21:33 <builder> build #28 of kirkwood is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/kirkwood/builds/28
**** ENDING LOGGING AT Fri Mar 19 02:59:56 2010