Re: Preliminary wiki starting page for i.MX53 QuickStart board
Le 03/09/2011 10:08, Zygmunt Krynicki a écrit : W dniu 03.09.2011 10:05:59, nadawca Eric Bénard,Lists Linaro-dev, Eric Miao, Christian Robottom Reis napisał: Hi Zygmunt, Le 02/09/2011 23:15, Zygmunt Krynicki a écrit : BTW: if anyone is interested in helping me out, have some work-in-progress code at lp:~zkrynicki/+junk/lava-imx53-serial-boot (serial as in "serial download mode offered by the boot-rom", not "serial cable"). here is a quick and dirty program which works in uart boot mode on i.MX25/35/51 (and should work on the i.MX53) and the corresponding configuration files for these 3 targets : it performs minimal CPU/memory initialization and upload a bootloader in RAM which allows recovery. For the i.MX53 init file, you can find a starting working base here : http://download.ronetix.info/peedi/cfg_examples/cortex-a8/mx53.cfg Eric Thanks for all of that. I'll use it to check for basic activity (that it works on my board) and then use it to finish my script. I wanted to use USB I'll be interested in feedback concerning your tests with i.MX53 as I'll soon have a 53 based board to bring up. because it allows for 0-end-user activity device provisioning. Unlike pure serial USB allows for device discovery. With a few udev rules an user with LAVA server running on their machine might simply plug an iMX53 device to the system to have it automatically detected and ready for testing. Thanks. This is what I was looking for! switching to USB is on my task list, but with a very low priority as our production testers only have serial ports for this task. Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: Preliminary wiki starting page for i.MX53 QuickStart board
Hi Zygmunt, Le 02/09/2011 23:15, Zygmunt Krynicki a écrit : BTW: if anyone is interested in helping me out, have some work-in-progress code at lp:~zkrynicki/+junk/lava-imx53-serial-boot (serial as in "serial download mode offered by the boot-rom", not "serial cable"). here is a quick and dirty program which works in uart boot mode on i.MX25/35/51 (and should work on the i.MX53) and the corresponding configuration files for these 3 targets : it performs minimal CPU/memory initialization and upload a bootloader in RAM which allows recovery. For the i.MX53 init file, you can find a starting working base here : http://download.ronetix.info/peedi/cfg_examples/cortex-a8/mx53.cfg Eric imxrecover.tar.gz Description: GNU Zip compressed data ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: [PATCH 0/3] babbage: support bootable Linaro kernel
Hi Shawn, Le 19/10/2010 18:28, Shawn Guo a écrit : If I understand correctly, you meant the tree below. git://git.pengutronix.de/git/imx/linux-2.6.git Can you please tell which branch has both your bits and Wolfram's? I can find your bits on branch imx-for-2.6.37, but Wolfram's are not there. you're right, you have to fetch them from the mmc-next branch of git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc.git Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: [PATCH 0/3] babbage: support bootable Linaro kernel
Hi, Le 19/10/2010 09:20, Amit Kucheria a écrit : Thanks for the patches. Could you please send patch 1& 2 to LAKML too for review? you will have to rebase on Sascha's tree as your first patch includes things which are already there. Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: SD/MMC for i.MX51
Hi, Le 19/10/2010 03:47, john stultz a écrit : On Mon, 2010-10-18 at 18:58 +0800, Shawn Guo wrote: Here you go. I'm using Linaro tree below. git://git.linaro.org/kernel/linux-linaro-2.6.35.git I've tried using this tree, (as well as the pengutronix.de tree and linus' upstream) with an older 9.04 disk image, but none of the booting kernels could find the sd card (no mmcblk devices located in the boot log) So sort of to echo what Nicolas said, it seems pretty difficult to follow which set of patches (as well as what .config settings) are needed to get upstream booting. on babbage you should need on top of Linaro's tree : Shawn's patch http://lists.linaro.org/pipermail/linaro-dev/2010-October/001307.html and this fix : http://lists.linaro.org/pipermail/linaro-dev/2010-October/001316.html To remove the anoying sdhci probe log, add : http://git.kernel.org/?p=linux/kernel/git/cjb/mmc.git;a=commit;h=0ca6d53ec81a13263f6643372b2a5abf7818e8fa Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: SD/MMC for i.MX51
Le 18/10/2010 12:58, Shawn Guo a écrit : Here you go. I'm using Linaro tree below. git://git.linaro.org/kernel/linux-linaro-2.6.35.git Also make sure you have the change below for /bin/auto-serial-console. line #15 of auto-serial-console: tty[A-Z]* )--> tty[a-zA-Z]* ) Thanks. Now it's booting fine on my board (using 2.6.36-rc7 + mmc & imx patches). The error was that I wasn't using the uInitrd which seems to be absolutely necessary to prepare the rootfs to boot. Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: SD/MMC for i.MX51
Hi Shawn, Le 18/10/2010 11:46, Shawn Guo a écrit : Yes, it's working. And I can get into Linaro console now. So we need the following bits to get it work. - Wolfram's esdhc driver (merged) - SDHCI_QUIRK_BROKEN_TIMEOUT_VAL patch - Babbage registration patch (based on Eric's bits) - Related kernel configurations may you please send me your .config and the link to the git tree your are using for your kernel ? I don't manage to boot the Linaro headless rootfs on my board (ureadahead crash then nothing more on the console) and now I've started to play with that I so want to get the login :-D and I think I may be missing a kernel option but which one ? Kernel : 2.6.37-rc7 + i.MX patches Command line : console=ttymxc0,115200 root=/dev/mmcblk0p3 rootfstype=ext3 ip=off rootwait rw Thanks, Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: SD/MMC for i.MX51
Hi Shawn,
Le 18/10/2010 10:10, Shawn Guo a écrit :
But when I try to login console, it gives a bunch of error messages like below.
mmcblk0: error -110 transferring data, sector 1737392, nr 2, card status 0xe00
end_request: I/O error, dev mmcblk0, sector 1737392
mmcblk0: error -84 transferring data, sector 1737394, nr 8, card status 0x900
end_request: I/O error, dev mmcblk0, sector 1737394
mmcblk0: error -84 transferring data, sector 1737402, nr 8, card status 0x900
end_request: I/O error, dev mmcblk0, sector 1737402
..
As you have tested Wolfram's bits on i.MX51 (eukrea?), probably my
babbage registration bit is causing problem. Could you please help
have a review to see if anything incorrect?
I just tested a linaro rootfs on SDCard (we have a nand flash so until now we
had our rootfs on it) and got the same errors as you.
As suggested by Richard, SDHCI_QUIRK_BROKEN_TIMEOUT_VAL seems to fix the
problem, so can you try the following patch ?
BTW, concerning Linaro's rootfs, I didn't managed to get a login prompt on the
serial port because of udev problems certainly because I'm missing options in
the kernel configuration, but at least I don't have anymore mmc error messages.
Eric
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c
b/drivers/mmc/host/sdhci-esdhc-imx.c
index 2e9cca1..e8f7048 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -134,7 +134,8 @@ static struct sdhci_ops sdhci_esdhc_ops = {
struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
.quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_NO_MULTIBLOCK
- | SDHCI_QUIRK_BROKEN_ADMA,
+ | SDHCI_QUIRK_BROKEN_ADMA
+ | SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
/* ADMA has issues. Might be fixable */
/* NO_MULTIBLOCK might be MX35 only (Errata: ENGcm07207) */
.ops = &sdhci_esdhc_ops,
___
linaro-dev mailing list
linaro-dev@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: SD/MMC for i.MX51
Hi Shawn, Le 17/10/2010 17:27, Shawn Guo a écrit : Thanks for the merging. But the bits for babbage mmc resources and driver registration are missing. I'm trying to add them for getting a bootable mmc linaro image, but running into some problem right now. Will continue digging tomorrow. you can find a working example here : http://lists.infradead.org/pipermail/linux-arm-kernel/2010-October/028846.html before, you will need this patch serie : http://lists.infradead.org/pipermail/linux-arm-kernel/2010-October/028826.html and this patch : http://lists.infradead.org/pipermail/linux-arm-kernel/2010-October/028855.html Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Re: daily snapshot - udev not working?
Hi, Le 30/09/2010 02:42, Christian Robottom Reis a écrit : > I have a similar problem, but ironically with Angstrom on my IGEPv2. It > says this when nothing is plugged in: > > r...@igep0020:~# lsusb > Bus 001 Device 001: ID 1d6b:0002 > Bus 002 Device 001: ID 1d6b:0002 > > If I plug in a keyboard, I get: > > [ 130.771484] hub 1-0:1.0: unable to enumerate USB device on 1 > if that's like on the Beagleboard, you need a USB hub to be able to use Low/Full speed USB devices : http://elinux.org/BeagleBoard#EHCI Eric ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
