**** BEGIN LOGGING AT Wed May 01 02:59:58 2013 May 01 05:14:41 build #228 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/228 May 01 05:17:27 build #233 of xburst is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/233 May 01 05:33:06 build #238 of avr32 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/238 May 01 07:05:18 build #207 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/207 May 01 13:40:08 build #202 of gemini is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/202 May 01 15:45:31 build #241 of ar7 is complete: Failure [failed shell shell_7] Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/241 May 01 16:14:42 i'm helping Christian Lamparter to improve the r92su driver (https://github.com/chunkeey/rtl8192su) using a BCM6328 based router, but appeared an issue with the USB 2.0 port May 01 16:14:46 http://pastebin.com/Aia5yy39 May 01 16:15:46 Are there problems with unaligned dma transfers with the bcm6328? May 01 16:16:54 or hardware limitations? May 01 16:21:38 the issue is: [ 1149.704000] ehci-platform ehci-platform: IAA with IAAD still set? May 01 16:58:07 build #183 of octeon is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/183 May 01 18:27:23 hi folks. here is a quick question. i upgraded the ram from 8-32 MiB and flash from 2-4 MiB in dsl-502t indian edition, with ar7. now adam2 fails completely. i tried to jtag the bootloader, but failed. can anyone help me with the bootloader? May 01 18:31:29 which bootloader is ideal. clearly adam2 failed. if i replace the 32 MiB with 16 MiB it works. with 32 MiB ram boot loader doesnt work. any help? May 01 18:39:23 is the hard limit 16 MiB? did i hit the ceiling? because there are a few ar7 modems which can have 32 MiB, IIRC. or i could be wrong. May 01 18:42:17 quackquack: if you change the RAM chips make sure that the new ones have the same organization than the old ones May 01 18:43:31 wait a minute. i still got one pristine dsl-502t. let me see where i goofed up. May 01 18:43:51 no problem May 01 18:46:06 Can you put the bootlog in pastebin? May 01 18:49:05 hynnix hy57v281620hct -> old ram and uT48lc32m8a2 -> new ram May 01 18:49:32 yep confident. no i can not. since both the device are with out boot loader. May 01 18:49:51 so i cannot read either device via serial. May 01 18:50:04 check the datasheets May 01 18:50:09 no problem May 01 18:50:18 give me a moment... May 01 18:50:23 sure thing. May 01 18:50:32 quackquack: two RAM chips or only one? May 01 18:51:36 device has legroom for only one ram chip. 2 devices. dsl-502t. May 01 18:52:02 this is wierd. both says 502t. but with different ram and flash in both the device. May 01 18:52:48 i mean flash chips are different. ram are same. hy57v281620hc, in both the devices. but i replaced the ram and flash in one and just ram in another. May 01 18:53:02 and now none of them boots. May 01 18:55:50 seems that the problem is the ram: the original is 16bits wide an the new is 8 bits wide May 01 18:55:52 the bootloader is from a router with 32 MiB ram 1 chip (16bit) ? May 01 18:55:52 quackquack: do you have more info about the new ram? the name does not give any google search results May 01 18:55:57 ah May 01 18:56:10 yes, 8 vs 16 bit is likely the problem May 01 18:56:36 yes i am checking the spec sheet. May 01 18:56:44 http://download.micron.com/pdf/datasheets/dram/sdram/512MbSDRAM.pdf May 01 18:56:47 this is the new ram datasheet: http://www.alldatasheet.com/datasheet-pdf/pdf/75880/MICRON/MT48LC32M8A2.html May 01 18:58:11 i will be back in 5 mins. hold on. May 01 18:58:34 quackquack: and check the latency too May 01 19:00:03 KanjiMonster: which means "[ 5102.140000] ehci-platform ehci-platform: IAA with IAAD still set?" ? bcm6328 SoC May 01 19:00:21 Ginkgo: no idea, I'm no usb expert May 01 19:00:37 no problem, thanks May 01 19:05:58 sorry was a little busy. May 01 19:06:09 no problem May 01 19:06:52 so there is an issue wrt RAM? May 01 19:07:41 sure May 01 19:08:15 hmmm. May 01 19:08:40 which is the closest replacement for hy57v281620hct? May 01 19:08:46 32 MiB May 01 19:09:02 google xD May 01 19:09:11 sorry, bad joke May 01 19:09:42 haha great May 01 19:10:50 if you have a jtag cable you can try with the bootloader of other ar7 routers May 01 19:11:02 Ginkgo, i did try many times. May 01 19:11:23 bootloader went smooth via jtag. then there is no console output. May 01 19:11:38 quackquack: another one with 16 bit bus with? May 01 19:11:55 that's a pity May 01 19:13:00 yes. i thought the ram replacement went smooth. afaik if ram is nor initated then jtag = blank. May 01 19:13:04 am i right? May 01 19:13:17 yes May 01 19:13:35 then it means the ram is recognised. May 01 19:13:56 then is it because of bit width bootloader failed? May 01 19:14:43 what? no. just because you could flash through jtag does not mean ram is recognised; when flashing through jtag you communicate directly with the flash chip without accessing ram on the target May 01 19:15:34 oh okay i didnt know that. i thought the reverse. May 01 19:16:06 you can still use jtag if ram not reconigzed May 01 19:16:35 oh damn it. this i didnt know. May 01 19:16:46 could be a bad soldered pin in flash? May 01 19:17:07 If you could flash through jtag, then no May 01 19:17:19 solder work is okay. my friend checked with a multimeter. May 01 19:18:35 yes i could flash through jtag. May 01 19:18:47 but i didnt get any serial output. May 01 19:19:40 than likely means the bootloader failed to initialize the ram or something earlier broke. May 01 19:19:41 then the bootloader failed early May 01 19:20:29 hmmm May 01 19:20:31 if you have JTAG you should be able to step through the bootloder and find the place where it hangs (requires mips assembly knowledge unless you have sources) May 01 19:20:39 HY57V561620BT is a 32MB RAM 16bit, something like this may work May 01 19:20:44 dinner time 15-20 mins please. May 01 20:01:52 aah back. May 01 20:02:10 HY57V561620BT? closest match? May 01 20:07:45 sure May 01 20:10:08 thanks danitool. i am going to market day after tomorrow. time to visit old junk yard and dig the old shack. May 01 20:10:22 and the shopkeeper as usual will give me a wierd look. May 01 20:10:40 btw, similar models should be fine too May 01 20:11:23 i downloaded the spec sheet. and i am reading it. and making a note of the closest match. May 01 20:13:08 I don't know if speed is important, if you can choose it use the same as the other May 01 20:15:02 speed is important. since it will be my adsl gateway. May 01 20:15:26 i mean important to be correctly recognized :P May 01 20:15:40 aah aah i get it. May 01 20:15:59 HY57V561620CT is another one close to it May 01 20:16:50 http://tinyurl.com/c38xco4 May 01 20:16:52 HY57V561620(L)T too. May 01 20:17:15 so this means HY57V561620 is the family May 01 20:17:39 thanks danitool ;-) May 01 20:18:57 I think even SDR from other manufacturers should work May 01 20:22:19 yeah i got it. i will hunt for it day after. May 01 21:10:53 hey blogic you got the u-boot dumped from dgn1000 May 01 21:13:32 florian r36504 trunk/target/linux/ (119 files in 2 dirs) * brcm63xx: remove 3.7 kernel support May 01 21:13:39 florian r36505 trunk/target/linux/brcm63xx/patches-3.8/419-MIPS-BCM63XX-enable-enet-for-BCM6345.patch * brcm63xx: add missing binary operands to BCM6345 ethernet patch May 01 21:37:15 build #265 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/265 May 01 21:46:56 good night everyone. May 01 21:53:25 florian r36506 trunk/target/linux/brcm63xx/patches-3.8/419-MIPS-BCM63XX-enable-enet-for-BCM6345.patch * brcm63xx: fix typo and missing BCM6328 check in BCM6345 ethernet patch May 01 22:16:17 juhosg r36507 trunk/ (32 files in 30 dirs) * kernel: update linux 3.8 to 3.8.11 May 01 22:56:58 any hints on undeleting files in overlayfs/squashfs? May 01 22:59:37 ..please? May 01 23:00:06 dzubey: the original squashfs contents are in /rom May 01 23:01:21 okay..i see that. So any changes to the filesystem are committed immediately by overlay then. May 01 23:02:19 so, i am sol. my fault. May 01 23:02:27 thx. May 02 00:50:04 luka r36508 packages/net/ tinc/files/tinc.config tinc/files/tinc.init * [packages] tinc: fix init.d/tinc parameter generation and logfile config name **** ENDING LOGGING AT Thu May 02 02:59:58 2013