**** BEGIN LOGGING AT Tue May 29 02:59:57 2007 May 29 11:15:16 florian * r7380 /trunk/target/linux/ar7-2.6/files/ (arch/mips/ar7/platform.c include/asm-mips/ar7/gpio.h): Restore the gpio.h file, platform.c uses the generic GPIO API May 29 11:32:32 nbd: no change in data throughput with new madwifi code (using minstrel) May 29 11:32:44 nbd: will try sample now/soon May 29 11:32:44 ok May 29 11:36:25 nbd: but I recognized a strange behaviour, /etc/init.d/network seems not to be executed at boottime anymore. Could this be, because I flash one device and customize it, then dd if=/dev/mtdblock1 of=/tmp/r7314.trx and write the resulting trx-file (7.3mb) on my productive system? May 29 11:38:20 florian * r7381 /trunk/target/linux/ar7-2.6/files/arch/mips/ar7/platform.c: Include the architecture specific gpio file for now May 29 11:50:04 nbd: no change with sample. I think it's even worse May 29 11:50:45 100mb, ETA > 18m May 29 11:50:54 from lan to wlan May 29 12:39:59 nbd: For me using madwifi - minstrel is more stable than it was (lastest over a day but still got the same segfault) - sample seems ok (probably switch to using it exclusively later today if I don't get a segfault) May 29 12:41:25 h3sp4wn: how is your data throughput? May 29 12:42:02 loswillios: give me 10 mins and I will tell you May 29 12:43:46 loswillios: 1.5MB/s May 29 12:43:58 (No real difference) May 29 12:45:43 h3sp4wn: hm May 29 12:45:56 h3sp4wn: wgt634u righ? May 29 12:54:22 Yep May 29 12:55:02 going to a broadcom ap (it will be switched to xscale some time soon, then hopefully I will get better) May 29 12:59:32 loswillios: are you using the card that came with your wgt or a different one? May 29 13:00:53 nbd: I haven't replaced it May 29 13:01:30 hm May 29 13:06:53 rwhitby * r7382 /trunk/package/kernel/modules/block.mk: pata-artop: no longer requires ide modules May 29 13:44:46 florian * r7383 /trunk/target/linux/adm5120-2.6/files/arch/mips/adm5120/memory.c: Fix .21 compilation on memory.c (#1778) May 29 14:09:08 rwhitby * r7384 /trunk/target/linux/ixp4xx-2.6/config/default: Remove the disabling of bluetooth for ixp4xx-2.6 compared to generic-2.6 May 29 14:23:24 nbd * r7385 /trunk/ (.gitignore scripts/config/.gitignore): Add gitignore files May 29 14:27:20 [florian]: Sorry, i have problems with my internet connection, since Friday. May 29 14:38:04 rwhitby * r7386 /trunk/target/linux/ixp4xx-2.6/config/default: Remove the disabling of sound support for ixp4xx-2.6 compared to generic-2.6 May 29 15:29:05 <[florian]> juhosg: no problem May 29 16:56:14 florian * r7387 /trunk/target/linux/adm5120-2.6/files/drivers/ (mtd/maps/adm5120_mtd.c serial/adm5120_uart.c): .21 fixes May 29 17:28:05 florian * r7388 /trunk/target/linux/adm5120-2.6/files/drivers/mtd/nand/rbmipsnand.c: Add the rbmipsnand fix from #1778 May 29 19:20:53 Force feedback support (EXPERIMENTAL) (HID_FF) [N/y/?] (NEW) <- bcm47xx with 2.6 May 29 19:21:20 * Kaloz slaps [florian] May 29 19:23:04 same error on x86 May 29 20:08:34 btw, for example usbmodules will be included into the images for wrt54gs, maybe it's a good idea to add a way to exclude some packages being included into images for some devices May 29 20:11:15 <[mbm]> that's what the profiles setting is for in the target/linux/* May 29 20:11:32 <[mbm]> meaning that the target shouldn't assume usb, it should be a profile May 29 20:14:10 <[florian]> what am I slaped for ? May 29 20:15:26 hmmm, well, but if i build the images for wl500gp with usb included and also want to put the image which was generated for the wrt54gs (via tft), the usbstuff shouldn't be packed into the image for the wrt54gs imho May 29 20:16:22 [florian]: guess because in the generic config for 2.6 the line for HID_FF doesn't exists May 29 20:17:09 <[mbm]> crazy_imp: any time you select a broadcom platform, it will compile all the various *.bin files with all the model names May 29 20:17:11 <[florian]> hum did I add that ? May 29 20:17:34 [florian]: dunno :D May 29 20:17:35 <[mbm]> crazy_imp: that's because they're all based on the same trx May 29 20:17:56 [mbm]: ack May 29 20:18:01 <[mbm]> meaning that you can't possibly compile with usb modules for a wl500gp and without for a wrt54gs in the same compile May 29 20:18:24 <[mbm]> and that when you compile with the usb modules, it will still produce a openwrt-wrt54gs.bin with the usb modules May 29 20:19:36 [mbm]: you could build different .trx files to generate the .bin files for the devices and after all .bins are there, build a final .trx with all selected packages May 29 20:20:04 crazy_imp: the different target profiles are used for support of various products, and besides USB is hypothetically possible on the WRT54GS if the right pins on the SoC are mapped out May 29 20:20:54 Bartman007: so i just need to solder cables to it and i have usb on it? May 29 20:21:17 crazy_imp: may need a few extra components (caps, etc) May 29 20:21:41 Bartman007: and 5v i think May 29 20:21:54 <[mbm]> crazy_imp: you should have the choice of if you want usb or not, but it's not going to check if the board actually has usb before it spits out the bin May 29 20:22:48 but lspci indicates that it does have a USB 1.1 host controller, and the SE505 has a similar mod. May 29 20:31:05 * Kaloz pokes Bartman007 :) May 29 20:34:04 Kaloz: I'm poking the dev board seller occaisionally too. May 29 20:34:51 he is having issues with paypal choking on his old email address (even though he has switched it to his new one) May 29 20:35:11 I just wanted to ask you to grab the money I've sent for the memory modules :) May 29 20:35:25 ah, probably a good idea. May 29 20:39:09 yay, was finally able to pay for the dev board. May 29 20:40:01 w00t May 29 20:40:02 :) May 29 20:40:18 wil lyou pick it up or will he send it to you? May 29 20:40:31 he wanted to send it to me. May 29 20:40:37 (btw, I'm wondering if it will come with all 3 flash chips to play with) May 29 20:42:15 If you want, I can send the money for that or send it once including the shipping here when we figure out how much will it be May 29 20:43:29 all three flash chips? May 29 20:43:48 nah, you can hold off on the payment until I get everything shipped. May 29 20:44:40 okie :) please check fedex as well just to see how much do they want for it May 29 20:45:00 well, theoritically the kit has 3 socketed flash chips May 29 20:45:10 one for linux, one for vxworks and one for something else May 29 20:46:16 oh! cool. the pdf I was looking at only showed a 16MB chip, didn't realize it was socketed. May 29 20:46:37 256MB ram is pretty roomy. May 29 20:47:20 well, at least i will have a board which is knows to play nicely with all jtag toys May 29 20:47:43 so I can go play with bootloader development more May 29 20:50:01 will this help with your Greenphone development also? (kinda assuming it won't contribute much) May 29 20:51:11 <[mbm]> I'm still of the opinion that we should try to work with the vendor bootloaders as much as possible, meaning that I'm having a hard time with the concept of changing them May 29 20:52:22 Bartman007: nope, that's pxa May 29 20:53:05 Kaloz: heh, I forgot. May 29 20:53:08 [mbm]: on some boards you have no choices for real.. and to be honest where it's optional, I don't want to enforce it on people May 29 20:53:18 [mbm]: but i do want to give the the choice to do so May 29 20:53:25 [mbm]: I agree on most boards, some that isn't an option. May 29 20:54:20 <[mbm]> Kaloz: yeah, I suppose what i'm really trying to avoid is the "you must change this, and btw this could damage your device" May 29 20:55:23 [mbm]: well, we only enforce such stuff where there is no other option, like for converting from vxworks May 29 20:55:48 <[mbm]> right **** ENDING LOGGING AT Wed May 30 02:59:57 2007