**** BEGIN LOGGING AT Fri Nov 01 02:59:59 2013
Nov 01 04:27:57 <builder_> build #4 of imx6 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/imx6/builds/4
Nov 01 05:10:34 <builder_> build #3 of mvebu is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/mvebu/builds/3
Nov 01 10:03:51 <builder_> build #127 of mpc85xx is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/127
Nov 01 15:35:56 <builder_> build #295 of mpc52xx is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/295
Nov 01 15:37:56 <builder_> build #396 of rb532 is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/396
Nov 01 15:39:52 <builder_> build #396 of ppc44x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/396
Nov 01 18:34:55 <Guest19371> juhosg r38634 trunk/target/linux/ar71xx/modules.mk * ar71xx: always enable hw ecc support for the ar934x-nfc module
Nov 01 18:34:57 <Guest19371> juhosg r38635 trunk/target/linux/ar71xx/image/Makefile * ar71xx: image: add IMAGE_PROFILE variable
Nov 01 18:34:59 <Guest19371> juhosg r38636 trunk/target/linux/ar71xx/image/Makefile * ar71xx: image: only define profiles for the generic subtarget
Nov 01 19:13:11 <JyZyXEL> where can i select that i want a ramdisk image to be built?
Nov 01 20:06:26 <[florian]> JyZyXEL: in make menuconfig -> Targets images -> ramdisk
Nov 01 20:07:01 <JyZyXEL> [florian]: i guess my target doesn't have support for that
Nov 01 20:07:29 <JyZyXEL> oh wait, it was the first one so i never saw it, haha :P
Nov 01 20:09:37 <JyZyXEL> figured i would try the ramdisk before i go overwriting any of the flash memory
Nov 01 20:26:22 <JyZyXEL> hmm im quite not sure which one of the resulting files is the ramdisk one
Nov 01 20:30:44 <JyZyXEL> i wonder if i should have make clean
Nov 01 20:37:03 <JyZyXEL> figured it would be openwrt-lantiq-xrx200-TDW8970-vmlinux-initramfs but when i 'bootm' with that it just resets the uboot
Nov 01 21:25:54 <JyZyXEL> openwrt-lantiq-xrx200-vmlinux-initramfs.elf: ELF 32-bit MSB executable, MIPS, MIPS32 rel2 version 1 (SYSV), statically linked, with unknown capability 0x41000000 = 0xf676e75, with unknown capability 0x10000 = 0x70403, stripped
Nov 01 21:26:01 <JyZyXEL> looks like a good candidate
Nov 01 21:26:17 <JyZyXEL> however when i try to boot that, the u-boot simply resets the box
Nov 01 21:31:29 <JyZyXEL> or am i supposed to boot the uImage and give some of those initramfs files as argument?
Nov 01 21:31:39 <JyZyXEL> im having trouble finding any documentation for this
Nov 01 21:32:07 <[florian]> you should boot the uImage
Nov 01 21:32:25 <[florian]> I doubt there is any sort of consistent documentation about all of this
Nov 01 21:32:51 <JyZyXEL> the uImage isn't the one that is bootable from ram though?
Nov 01 21:34:05 <JyZyXEL> the uImage does boot up but it kernel panics pretty quick saying: [    1.948000] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(31,2)
Nov 01 21:34:23 <JyZyXEL> and [    1.940000] No filesystem could mount root, tried:  squashfs
Nov 01 21:40:12 <JyZyXEL> maybe i just have to risk it and try to put this on the flash and try boot that way
Nov 01 21:47:31 <[florian]> that's because you used an uImage which does not contain an initramfs
Nov 01 21:47:41 <[florian]> uImage is just a format for the bootloader
Nov 01 21:47:52 <[florian]> whether this is an initramfs kernel or something else is up to you
Nov 01 21:48:50 <JyZyXEL> make only made me: openwrt-lantiq-xrx200-TDW8970-sysupgrade.image, openwrt-lantiq-xrx200-TDW8970-uImage, openwrt-lantiq-xrx200-TDW8970-vmlinux-initramfs, openwrt-lantiq-xrx200-TDW8970-vmlinux-initramfs-loader and openwrt-lantiq-xrx200-vmlinux-initramfs.elf
Nov 01 21:49:24 <[florian]> ok so you'd need an uImage-initramfs one
Nov 01 21:49:37 <[florian]> or maybe you can convince uboot to load an ELF file instead
Nov 01 21:50:14 <JyZyXEL> figured 'go' would do that but it didn't
Nov 01 21:51:24 <[florian]> that would work witht he binary initramfs probably though
Nov 01 21:51:26 <[florian]> *the binary
Nov 01 21:52:32 <JyZyXEL> nope
Nov 01 21:52:38 <JyZyXEL> it just resets the box
Nov 01 21:52:46 <builder_> build #370 of sibyte is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/370
Nov 01 21:54:47 <JyZyXEL> ill try adding the  cpio.gz and tar.gz into my make menyconfig
Nov 01 21:54:57 <JyZyXEL> perhaps that generates some additional images that might help
Nov 01 21:55:09 <[florian]> you'd rather hack target/linux/xway/image/Makefile to generate an initramsf uImage instead
Nov 01 21:55:32 <[florian]> uboot does not accept much more than raw(binary), elf and uImage
Nov 01 21:56:43 <JyZyXEL> well this uboot only seems to accept uImage
Nov 01 21:56:57 <JyZyXEL> binary and elf just cause a reboot
Nov 01 21:58:09 <[florian]> you did use load before and speficied the address for go to jump to right?
Nov 01 21:59:10 <JyZyXEL> go takes the address as an argument
Nov 01 21:59:35 <JyZyXEL> so: go 0x80800000 in my case
Nov 01 22:00:12 <JyZyXEL> seems plain 'bootm' does exactly the same
Nov 01 22:00:55 <[florian]> that's what I am trying to say, yes
Nov 01 22:00:55 <[florian]> e.g: go 0x80010000
Nov 01 22:00:55 <[florian]> provided that where the raw binary was loaded, yes
Nov 01 22:01:26 <JyZyXEL> yeah it should work but for some reason it just resets the box
Nov 01 22:06:01 <JyZyXEL> https://dev.openwrt.org/browser/trunk/target/linux/lantiq/image/Makefile?rev=
Nov 01 22:06:17 <JyZyXEL> its supposed to generate a initramfs uimage it seems
Nov 01 22:06:19 <JyZyXEL> $(CP) $(KDIR)/uImage-$(1)-initramfs $(BIN_DIR)/$(IMG_PREFIX)-$(1)-uImage-initramfs
Nov 01 22:08:29 <JyZyXEL> i wonder if in the makefile i have to ONLY set the ramdisk
Nov 01 22:08:38 <JyZyXEL> or in the menuconfig
Nov 01 22:10:08 <JyZyXEL> oh wait, the ramdisk option has sub options too
Nov 01 22:10:12 <JyZyXEL> i can select a compression
Nov 01 22:10:37 <JyZyXEL> uImage is lzma?
Nov 01 22:11:49 <[florian]> lzma is a compression algorithm, uImage is a container
Nov 01 22:12:09 <[florian]> uImage can embedd compress or uncompressed payloads
Nov 01 22:14:05 <JyZyXEL> in the Makefile it seems all the uImages contain lzma compressed payloads
Nov 01 22:15:38 <[florian]> that's quite a possibility
Nov 01 22:15:43 <JyZyXEL> ill try to set the make menuconfig only to make me a lzma compressed ramdisk
Nov 01 22:17:50 <_trine> http://pastebin.mozilla.org/3399596
Nov 01 22:19:57 <JyZyXEL> damnit still no uimage
Nov 01 22:20:12 <_trine> http://pastebin.mozilla.org/3399596
Nov 01 22:20:30 <_trine> can someone commit this before I get committed
Nov 01 22:21:20 <[florian]> commit what?
Nov 01 22:21:26 <JyZyXEL> blogic: can you add a -uImage-initramfs into https://dev.openwrt.org/browser/trunk/target/linux/lantiq/image/Makefile?
Nov 01 22:21:45 <_trine> its a patch for GMP
Nov 01 22:21:50 <[florian]> this is a build log which fails
Nov 01 22:22:02 <[florian]> wait it already has a line which does build an uImage-initramfs for xrx200
Nov 01 22:22:50 <JyZyXEL> well it does not get triggered for some reason
Nov 01 22:22:59 <[florian]> _trine: you pasted twice the same link, where is the patch?
Nov 01 22:23:25 <_trine> https://dev.openwrt.org/ticket/14231
Nov 01 22:23:27 <_trine> opps
Nov 01 22:25:40 <JyZyXEL> perhaps i should remove line 125: ifneq ($(CONFIG_TARGET_ROOTFS_INITRAMFS),)
Nov 01 22:25:56 <JyZyXEL> im thinking then the -uImage-initramfs code would get executed
Nov 01 22:27:00 <[florian]> no, this is what should actually trigger the build
Nov 01 22:27:13 <[florian]> if this variable is not empty, meaning it was enabled, do what's enclosed
Nov 01 22:27:26 <JyZyXEL> well the thing above it gets executed
Nov 01 22:27:53 <JyZyXEL> so that must mean ifneq ($(CONFIG_TARGET_ROOTFS_INITRAMFS),) is the problem
Nov 01 22:28:08 <[florian]> you could try to add positive logic instead
Nov 01 22:28:20 <[florian]> like ifeq ($(CONFIG_TARGET_ROOTFS_INITRAMFS),y)
Nov 01 22:28:27 <[florian]> although I doubt this would change anything
Nov 01 22:28:43 <_trine> [florian], I am off to bed,, hope it gets fixed goodnight
Nov 01 22:29:23 <JyZyXEL> i wouldn't add the y just to make sure you don't reverse the logic two times
Nov 01 22:29:51 <[florian]> oh well, do whatever you want
Nov 01 22:32:27 <JyZyXEL> ill add a whole new define for it
Nov 01 22:40:41 <JyZyXEL> i still don't get the -uImage-initramfs
Nov 01 22:41:38 <JyZyXEL> some of those commands must silently just fail
Nov 01 23:32:47 <JyZyXEL> im not sure if i should try to flash this uimage because it seems suspiciously small size
Nov 01 23:32:52 <JyZyXEL> 1.2M
Nov 01 23:55:53 <builder_> build #380 of avr32 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/380
Nov 01 23:58:58 <builder_> build #319 of iop32x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/319
Nov 02 00:09:33 <builder_> build #409 of uml is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/409
**** ENDING LOGGING AT Sat Nov 02 02:59:57 2013