| 01:22.59 | traeak | oh i *love* win32 calls |
| 01:23.02 | traeak | NOT |
| 01:23.03 | traeak | anyways |
| 02:02.22 | *** join/#arm-netbook A1Sheds (~ly@adsl-75-49-211-164.dsl.emhril.sbcglobal.net) |
| 02:02.22 | *** join/#arm-netbook A1Sheds (~ly@unaffiliated/l84supper) |
| 02:54.03 | *** join/#arm-netbook QingPei (~qingpei@inet-hqmc03-o.oracle.com) |
| 03:24.41 | CIA-91 | rhombus-tech: Donald master * r838fd8c3b305 /allwinner_a10/orders/.mdwn: |
| 03:53.34 | *** join/#arm-netbook hp__ (~hp_@p5B38A22C.dip.t-dialin.net) |
| 04:36.07 | steev | hno: where does the script.bin go? |
| 04:40.23 | steev | would be nice if the wiki would move the orders changes out of RecentChanges and into it's own area |
| 07:02.53 | *** join/#arm-netbook QingPei (~qingpei@inet-hqmc01-o.oracle.com) |
| 07:15.43 | specing | steev: Did you manage to compile the kernel per instructions? |
| 07:15.58 | specing | arch/arm/mach-sun4i/pm/standby.S:3: Error: file not found: arch/arm/mach-sun4i/pm/standby/standby.code |
| 07:45.26 | hno | steev, scipt.bin needs to be loaded at 0x43000000 |
| 07:46.56 | hno | mnemoc, the problematic area in 3.3 have been identified. Something goes wrong with the device mappings causing kernel crash when trying to address I/O. |
| 07:59.03 | hno | mach-sun4i/core.c sw_map_io() or from where it's called.in arch/arm/mm/mmu.c devicemaps_init() |
| 08:23.23 | hno | seems to be same problem but in inverse why earlyprintk do not work in 2.36.6. The early pagetable mapping of I/O seems to fail there. |
| 08:23.48 | hno | unfortunately I have no JTAG running yet so am a bit blinded. |
| 08:42.00 | steev | hno: okay, thanks for the info. perhaps i'm getting things mixed up. this is all in the wiki right? |
| 08:42.23 | steev | specing: I grabbed the one liner from the scripts/build.sh (or whatever it's name is) to build the standby.code |
| 08:43.29 | specing | steev: Its fine now, I found the build_sun4i_crane script and ran it |
| 08:43.59 | steev | yeah sorry, i moved recently so i have to borrow my neighbors wifi, and it's not the most stable (with his permission) |
| 08:54.42 | hno | steev, script.bin info is not in the wiki I think |
| 08:56.01 | steev | hno: okay, i'm guessing it's from the "Hack A10 devices" page? i'm asking because there are a few different pages out there with *very* similar (yet just different enough...) instructions on things |
| 08:56.53 | hno | which u-boot are you using for the SD-card? |
| 08:57.26 | hno | i.e. are you creating a phoenix boot card using phoenixcard, or the MMC u-boot by Tom? |
| 08:57.46 | hno | The latter is http://rhombus-tech.net/a10_mmc_boot/ |
| 08:57.51 | steev | i am using the instructions from http://rhombus-tech.net/a10_mmc_boot/ yes |
| 08:58.55 | steev | i create the sdcard following those instructions, but powering it on does nothing, so i'm assuming i'm missing some little bit somewhere |
| 09:01.55 | steev | i was looking at the paste you did earlier and noticed your uboot attempting to load script.bin, so that's why I asked about it |
| 09:05.51 | hno | you need to place script.bin in the fat partition, and load it to 0x43000000 before loading the kernel. |
| 09:08.28 | hno | Interrupt U-boot, and fix envirnoment a bit: |
| 09:09.15 | hno | setenv boot_mmc 'fatload mmc 0 0x43000000 script.bin; fatload mmc 0 0x48000000 ${kernel}; bootm 0x48000000' |
| 09:09.20 | hno | setenv kernel uImage |
| 09:09.22 | hno | saveenv |
| 09:10.06 | steev | okay, that's easy enough. i'm guessing the script.bin from http://elinux.org/Hack_A10_devices is needed? |
| 09:10.37 | hno | I am using the one from the mele firmware. |
| 09:11.06 | hno | mnemoc posted a dropbox link yesterday here. |
| 09:11.19 | steev | okay i'll scroll up and look |
| 09:11.34 | hno | <mnemoc> http://dl.dropbox.com/u/24940371/A10/uImage http://dl.dropbox.com/u/24940371/A10/script.bin |
| 09:11.53 | steev | oh even easier :D |
| 09:12.12 | steev | (thanks, i'm on my old macbook pro and i have no idea how to page up on this crazy keyboard anymore) |
| 09:14.53 | *** part/#arm-netbook QingPei (~qingpei@inet-hqmc01-o.oracle.com) |
| 09:17.28 | mnemoc | hno: might be related to the evil gps/g2d table allocation in head.S ? |
| 09:20.37 | mnemoc | specing: that issue with standby.code integration is fixed in the branches in github |
| 09:44.01 | hno | mnemoc, I don't think so. |
| 09:45.17 | hno | Really need to get that JTAG header fixed, or at a minimum locate a RESET signal. The nRESET pin in the JTAG header seems unconnected unfortunately. |
| 09:46.29 | mnemoc | hno: what about the jtag over SD? |
| 09:46.45 | hno | I need the SD for boot media. |
| 09:46.53 | hno | and there also is no RESET signal there. |
| 09:47.09 | mnemoc | narf |
| 09:49.19 | mnemoc | can the lichee-dev (not mmc) branch be blessed with usb support? |
| 09:49.25 | mnemoc | as booting media |
| 09:50.04 | hno | There probably is a reset signal beetween the power controller and CPU. |
| 09:51.03 | hno | mnemoc, I think so. |
| 09:53.22 | hno | but I also do not have an SD breakout board, and easier to fix the JTAG header. |
| 09:55.06 | mnemoc | :) |
| 10:12.19 | *** join/#arm-netbook MaDMaLKaV (~IceChat9@188.86.253.69) |
| 10:24.31 | hno | Nice, I do have an USB cable with compatible header. |
| 10:25.21 | hno | and found the FEL switch. |
| 10:25.39 | hno | 1K1 is FEL |
| 10:26.51 | hno | Mar 31 12:25:01 home kernel: [180729.325171] usb 2-3: New USB device found, idVendor=1f3a, idProduct=efe8 |
| 10:37.30 | hno | and fastboot usb gadget support works in android u-boot, so should be easy to add USB gadget serial console support. |
| 10:37.57 | hno | enabling fast download of kernel over USB. |
| 11:20.39 | CIA-91 | rhombus-tech: hno master * rf5747b5cffaf /allwinner_a10/hacking_the_mele_a1000.mdwn: Document FEL jumper location |
| 11:34.17 | mnemoc | neat |
| 11:34.30 | mnemoc | if only I could solder... :-/ |
| 11:36.25 | mnemoc | hno: to search for a cable, what's the name of that usb connector? |
| 11:40.24 | mnemoc | hno: can the FEL be triggered by using a m/m jumper cable without desoldering what's covering the holes? |
| 11:41.04 | mnemoc | but I would still need the usb cable anyway :< |
| 11:41.55 | mnemoc | with the jumper cable I meant about how long does it have to be closed |
| 12:08.42 | CIA-91 | rhombus-tech: lkcl master * r66d2ed35b454 /allwinner_a10/orders/snita.mdwn: order processing cleanup |
| 12:10.20 | hno | mnemoc, I used a unsoldered cable to trigger it. |
| 12:13.18 | hno | regarding USB it's not a standard connector for USB. The connector type is quite often seen on internal connectors in electronics, but usually not in computers. The same 2.5->2mm jumper cables should work fine however. You only neeed GND, D+ and D-. |
| 12:14.25 | hno | to avoid soldering pick an internal USB cable from a computer. populate the connector with suitable pins (i.e. pieces of wire) and connect 2.5-2mm jumper wires to the USB port. |
| 12:16.54 | hno | I just happened to have an old USB LCD module which had the right density (but thinner model) on the USB connector, only had to use a needle to remove the 5V cable and it fits just fine (but not very steady, to thin) in the mele usb port. |
| 12:17.02 | mnemoc | thanks, I will... when my pack of 2.54-2.00 cables arrives :) |
| 12:17.21 | hno | that is, I use the USB cable from that module. |
| 12:17.28 | mnemoc | I have plenty use cables to mutilate |
| 12:17.35 | mnemoc | usb* |
| 12:17.49 | hno | Me too, but not many of those small connectors. |
| 12:18.44 | hno | Actually... the right USB cable can quite likely be found in USB mice and keyboards when I think about it. |
| 12:19.00 | hno | but don't have many of those to mutilate. |
| 12:21.26 | hno | picks up a random Dell USB mouse and ithas the right 5-pin 2mm USB cable internally. |
| 12:24.14 | hno | but probably needs the pins rearranged slightly. |
| 12:24.47 | mnemoc | please take some pictures :) |
| 12:28.27 | hno | Of a mouse? |
| 12:31.00 | mnemoc | nah, the connections (G,D+,D-) |
| 12:34.58 | hno | These? http://rhombus-tech.net/allwinner_a10/hacking_the_mele_a1000/A10_device_a1000_inside.jpg |
| 12:36.42 | hno | Standard color coding on USB cables seem to be Black (G), Green (D+), White (D-), Red (+5V, not needed) |
| 12:37.36 | hno | Do you have an USB mouse you can slaughter? |
| 12:39.39 | mnemoc | nope, only BT and USB/wireless :-/ |
| 12:41.07 | mnemoc | I hadn't noticed the usb pins are labeled on the picture |
| 12:41.58 | mnemoc | but I'll wait for the wire jumpers before taking the knife out |
| 12:44.15 | hno | no need fo a knife if you have something with the right kind of internal connevtor |
| 12:44.27 | hno | only a needle to move pins around a bit. |
| 12:54.57 | hno | <PROTECTED> |
| 13:08.57 | hno | got a pair of extra small hands helping him type today. |
| 13:23.19 | hno | mnemoc, no other usb gadget with cable? |
| 13:31.29 | mnemoc | how old are the small hands? :) |
| 13:31.55 | mnemoc | all usb things I have use m/m cables |
| 13:35.20 | mnemoc | but I can most probably find a cheap mouse on a chinese bazar on monday |
| 13:41.33 | hno | that pair of hands is 16 months. |
| 13:42.32 | hno | the cheapest ones may have cables soldered. |
| 13:44.51 | hno | a used big brand one is best bet for finding a matching cable. |
| 15:45.20 | *** join/#arm-netbook QingPei (~qingpei@114.241.187.241) |
| 16:47.04 | specing | steev: Did you specify rootfs on the kernel command line? |
| 16:47.13 | specing | steev: What device do you have? |
| 17:02.02 | specing | steev: my tablet goes completely dead when the boot SD card is inside |
| 17:04.03 | specing | steev: the default kernel command line is completely wrong |
| 17:09.23 | *** join/#arm-netbook MaDMaLKaV (~IceChat9@188.86.253.69) |
| 17:15.06 | specing | On a side note, would anyone know why /data (nandd) doesen't get mounted at boot? (android) |
| 17:19.28 | CIA-91 | rhombus-tech: Dmitriy master * rc5007beccb5a /allwinner_a10/orders/ermito.mdwn: |
| 17:29.42 | *** join/#arm-netbook penguin42 (~dg@tu006.demon.co.uk) |
| 17:36.05 | specing | script.fex is missing from nanda |
| 17:37.26 | mnemoc | .fex is a source file, machines use the compiled version, script.bin |
| 17:39.54 | specing | ah |
| 17:41.47 | specing | Do you maybe know about my /data problem? |
| 17:42.19 | specing | It started happening when it once ran out of power after having lots of /dev/block/nand* partitions mounted |
| 17:42.33 | specing | Im not sure how it could corrupt ext4 partitions... |
| 17:43.20 | specing | hmm |
| 17:43.34 | specing | there is init: do_umount /data in dmesg... |
| 17:43.37 | specing | wtf? |
| 17:45.11 | mnemoc | boot it with serial console connected and log/review the whole process? |
| 17:45.55 | specing | Im not sure where the serial port is located |
| 17:47.27 | mnemoc | on a10-based tablets the best shot is to get one out of the uSD slot |
| 17:48.14 | mnemoc | and recompiling the script.fex enabling those pins as uart |
| 17:49.34 | specing | D: but I want the SD |
| 17:49.48 | specing | and Im not sure where to get an adapter |
| 17:50.01 | specing | I can do PCBs myself but not that thin |
| 17:50.54 | mnemoc | http://www.sparkfun.com/products/9419 |
| 17:51.41 | mnemoc | you can also ask Tom to sell you one of his' ... with ttl/usb adapter and jtag connector |
| 17:53.39 | mnemoc | http://www.aliexpress.com/product-fm/527673076-A10-tablet-macro-sd-breakout-card-debug-cable-wholesalers.html |
| 17:57.35 | mnemoc | A10-based devices are very hack friendly :) |
| 18:16.55 | specing | How come one can reconfigure pins' functions at will though? Does it include a FPGA? |
| 18:17.28 | specing | or is there peripheral multiplexing in progress? |
| 18:17.37 | mnemoc | inside the soc |
| 18:17.49 | specing | which I find hard to believe with a several hundred pin SoC |
| 18:18.18 | mnemoc | that's the purpose of the script.fex file |
| 18:28.18 | specing | mnemoc: what are my chances of finding a serial port header/connection holes on the board? |
| 18:32.39 | mnemoc | no idea... but it is quite possible the unsoldered holes are there |
| 19:38.15 | specing | Ok, I've opened it up |
| 19:39.08 | specing | There are 8 pads exposed under the label "URDL" |
| 19:44.18 | mnemoc | i'm the wrong person to help you there |
| 19:45.05 | mnemoc | i don't even know what you mean by "pads" |
| 19:46.56 | specing | it looks kinda like the thing you would solder a uSD connector on |
| 19:47.09 | specing | pads are surface equivalent of pins |
| 19:48.18 | mnemoc | thanks :) |
| 19:48.36 | mnemoc | no... pads beside those for the spare uSD? |
| 19:51.31 | specing | spare? |
| 19:53.02 | mnemoc | extra |
| 19:53.10 | specing | yes it seems like it |
| 19:53.29 | specing | I'll post the images once I figure out where |
| 19:54.33 | mnemoc | http://imgur.com/ is quite nice |
| 19:57.06 | mnemoc | mail Tom asking him to sell his sd breakout thing again, it shouldn't be a problem... he said he can buy them very easily |
| 19:57.25 | specing | I'd rather keep the SD where it is |
| 19:57.38 | mnemoc | :) |
| 19:57.50 | specing | I can solder wires manually if that pad is really a secondary SD connector |
| 19:58.22 | mnemoc | very true |
| 19:58.23 | specing | or maybe I should solder an another uSD connector... |
| 19:58.28 | specing | looks in his drawer |
| 19:58.39 | specing | looks in the second drawer |
| 19:58.44 | specing | Ah there it is |
| 20:01.33 | CIA-91 | rhombus-tech: Benjamin master * re2d6af872903 /allwinner_a10/orders/.mdwn: |
| 20:11.13 | CIA-91 | rhombus-tech: Benjamin master * r8ada9bf559a6 /allwinner_a10/orders/nagaikaminoke.mdwn: |
| 20:29.50 | specing | mnemoc: https://93.103.140.50/A10-MID/ |
| 20:32.56 | mnemoc | specing: what is it? brand/model |
| 20:33.07 | specing | A10-MID |
| 20:33.24 | specing | build.prop mentions something about "softwinners" and evb-crane |
| 20:34.09 | mnemoc | evb-crane is the devkit |
| 20:34.21 | mnemoc | softwinners is one of the alter-egos of allwinner |
| 20:34.36 | mnemoc | pretty spartan board |
| 20:34.40 | specing | I'll upload build.prop too |
| 20:34.47 | *** part/#arm-netbook QingPei (~qingpei@114.241.187.241) |
| 20:35.40 | specing | Its there |
| 20:35.59 | mnemoc | compare it with the mele :) http://i.imgur.com/1xd2j.jpg |
| 20:36.36 | specing | Mine certainly isn't that bulky ;D |
| 20:36.49 | mnemoc | :) |
| 20:38.11 | specing | What else do I need besides 1MB/s NFS over wifi? |
| 20:39.15 | mnemoc | for what? |
| 20:39.44 | specing | for what what? |
| 20:40.12 | specing | xD |
| 20:40.21 | mnemoc | ^^ |
| 20:40.33 | specing | for the lulz, but ofcourse |
| 20:40.47 | specing | NFS was in /proc/filesystems and I took the bait |
| 20:41.39 | specing | Its weird that this tablet uses ext4 on nand flash |
| 20:41.53 | specing | A journaled FS |
| 20:42.06 | specing | My mobile phone uses yaffs |
| 20:42.54 | specing | mnemoc: build.prop is up there on the webserver btw |
| 20:42.59 | mnemoc | your phone has far more soft. eng. hours than your tablet :) |
| 20:43.29 | mnemoc | s/soft. // |
| 20:43.47 | specing | Yes probably |
| 20:43.54 | specing | its alcatel, so not chinese |
| 20:44.08 | specing | actually the cheapest rootable android phone I could get |
| 23:58.42 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 23:59.08 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
