Re: AM335x EVM and core-image-sato

[Denys Dmytriyenko <denys@ti.com>]

[-- Attachment #1: Type: text/plain, Size: 31994 bytes --]

Josh,

Yes, in your case it was clear from the beginning you had a problem with SK 
and it is kind of known... In Rudi's case he was saying he had an actual EVM, 
which is known to boot fine at the moment, hence I was confirming with him.

For your case, please try the DTB file attached. Thanks.

-- 
Denys


On Tue, Oct 22, 2013 at 08:36:13PM +0000, Vanderpool Josh wrote:
> It is a SK and the revision of silicon is 1.0 as it is a XAM3359ZCZ no A or 
> B after the AM335x.
> 
> ________________________________________
> From: meta-ti-bounces@yoctoproject.org [meta-ti-bounces@yoctoproject.org] on behalf of meta-ti-request@yoctoproject.org [meta-ti-request@yoctoproject.org]
> Sent: Tuesday, October 22, 2013 4:17 PM
> To: meta-ti@yoctoproject.org
> Subject: meta-ti Digest, Vol 25, Issue 36
> 
> Send meta-ti mailing list submissions to
>         meta-ti@yoctoproject.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://lists.yoctoproject.org/listinfo/meta-ti
> or, via email, send a message with subject or body 'help' to
>         meta-ti-request@yoctoproject.org
> 
> You can reach the person managing the list at
>         meta-ti-owner@yoctoproject.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of meta-ti digest..."
> 
> 
> Today's Topics:
> 
>    1. Re: AM335x EVM and core-image-sato (Denys Dmytriyenko)
>    2. Re: AM335x EVM and core-image-sato (Denys Dmytriyenko)
>    3. Re: AM335x-SK kernel panic with Poky(master), meta-ti(master)
>       core-image-base (Vanderpool Josh)
>    4. Re: AM335x EVM and core-image-sato (Denys Dmytriyenko)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Tue, 22 Oct 2013 15:36:21 -0400
> From: Denys Dmytriyenko <denys@ti.com>
> To: Rudolf Streif <rstreif@linuxfoundation.org>
> Cc: meta-ti@yoctoproject.org
> Subject: Re: [meta-ti] AM335x EVM and core-image-sato
> Message-ID: <20131022193620.GB9301@edge>
> Content-Type: text/plain; charset="us-ascii"
> 
> Huh, I'm wondering if 1.2 is still supported... There were 2.0 and 2.1
> versions of the silicon and I believe those are the latest ones. Need to
> double-check that.
> 
> Denys
> 
> 
> On Tue, Oct 22, 2013 at 08:57:51AM +0100, Rudolf Streif wrote:
> > Where do I find the PG version of the board? The barcode sticker on the
> > board says rev 1.2.
> >
> >
> > On Tue, Oct 22, 2013 at 12:19 AM, Denys Dmytriyenko <denys@ti.com> wrote:
> >
> > > Now I'm just wondering what PG version of the board you've got, if it
> > > fails in
> > > the same way with our 3.8 kernel and a known good set of 3.12 images...
> > >
> > >
> > > On Mon, Oct 21, 2013 at 11:19:35PM +0100, Rudolf Streif wrote:
> > > > Thanks, Denys.
> > > >
> > > > Interesting point about gcc. It's version 4.8.1 which is the default for
> > > > poky master. Which is what I used all along.
> > > >
> > > > Thanks for sending your binaries. They fail too with the exact same
> > > > problem. See attached log. I found [1] on gmane which exactly describes
> > > the
> > > > problem I am seeing.
> > > >
> > > > If that is the case then I was wondering why my earlier builds with
> > > > 3.12-rc3 actually worked. And I tried them again and they are now failing
> > > > too. Of course I don't know anything about the specifics of the gpio and
> > > > the timers on the chip but it looks like a weird hardware initialization
> > > > condition. And I don't know if the proposed patch in [1] was ever
> > > > incorporated in your kernel tree.
> > > >
> > > > I don't think it's the hardware (or at least not alone) because the SD
> > > that
> > > > ships with it with the 3.2 kernel works just fine.
> > > >
> > > > Cheers,
> > > > Rudi
> > > >
> > > >
> > > > [1]  http://permalink.gmane.org/gmane.linux.ports.arm.omap/96044
> > > >
> > > >
> > > > On Mon, Oct 21, 2013 at 10:15 PM, Denys Dmytriyenko <denis@denix.org>
> > > wrote:
> > > >
> > > > > Rudi,
> > > > >
> > > > > As we discussed earlier today, let my send you my binaries for the
> > > kernel
> > > > > and
> > > > > u-boot to try on your board. I just wanted to ask you about the
> > > toolchain
> > > > > that
> > > > > you used to build yours. In our builds we use Linaro gcc-4.7 hardfp
> > > > > toolchain
> > > > > and I vaguely remember hearing about some problems with the recent
> > > gcc-4.8
> > > > > toolchain from oe-core... I'm wondering if you are having the same
> > > issue.
> > > > >
> > > > > --
> > > > > Denys
> > > > >
> > > > >
> > > > > On Mon, Oct 21, 2013 at 10:42:35AM -0700, Rudolf Streif wrote:
> > > > > > Since the 3.12-rc5 kernel does not work I decided to give the 3.8
> > > kernel
> > > > > a
> > > > > > try. The issue with this one is that the linux-ti-staging_3.8.bbrecipe
> > > > > > explicitly sets KERNEL_IMAGETYPE="uImage". However, the u-boot
> > > > > environment
> > > > > > has now moved on to using zImage which consequently means u-boot
> > > will not
> > > > > > find the kernel file.
> > > > > >
> > > > > > I do not know why KERNEL_IMAGETYPE = "uImage" was hacked into the
> > > > > > linux-ti-staging_3.8.bb recipe but whoever did it did not think
> > > about
> > > > > > u-boot. Does zImage not work with 3.8? I did not try it but it should
> > > > > > normally have no effect. Since there is no dependency of the u-boot
> > > > > recipes
> > > > > > on the kernel image types changes to the kernel image type for newer
> > > > > > versions of the kernel then automatically means that it must be
> > > changed
> > > > > for
> > > > > > the previous versions too.
> > > > > >
> > > > > > I mocked around with the u-boot environment and booted the 3.8 uImage
> > > > > > kernel. To my surprise the 3.8 kernel from the TI staging repository
> > > > > fails
> > > > > > with the exact same problem and stops booting with a kernel not
> > > syncing
> > > > > > message during the init calls. Apparently a patch must have been
> > > applied
> > > > > to
> > > > > > the kernel repo that breaks it. It has nothing possibly to do with
> > > the
> > > > > > build system.
> > > > > >
> > > > > > I would appreciate any hints to get this kernel built and boot on the
> > > > > board.
> > > > > >
> > > > > > Thanks,
> > > > > > Rudi
> > > > > >
> > > > > >
> > > > > > On Mon, Oct 21, 2013 at 1:16 AM, Rudolf Streif
> > > > > > <rstreif@linuxfoundation.org>wrote:
> > > > > >
> > > > > > > Denys,
> > > > > > >
> > > > > > > I recompiled the 3.12-rc5 kernel again but it still stalls during
> > > > > > > initialization. I attached the boot log. I don't know much about
> > > the
> > > > > > > platform details but maybe that's an issue you know about.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Rudi
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >>
> > > > > > >> [1]
> > > > > > >> Build Configuration:
> > > > > > >> BB_VERSION        = "1.21.0"
> > > > > > >> BUILD_SYS         = "x86_64-linux"
> > > > > > >> NATIVELSBSTRING   = "Fedora-18"
> > > > > > >> TARGET_SYS        = "arm-poky-linux-gnueabi"
> > > > > > >> MACHINE           = "am335x-evm"
> > > > > > >> DISTRO            = "poky"
> > > > > > >> DISTRO_VERSION    = "1.5+snapshot-20131019"
> > > > > > >> TUNE_FEATURES     = "armv7a vfp neon"
> > > > > > >> TARGET_FPU        = "vfp-neon"
> > > > > > >> meta-yocto        =
> > > "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > > > >> meta-ti           =
> > > "master:c11504664990f007c3903d349e7cc135a904f8e2"
> > > > > > >> meta-oe           =
> > > "master:f8903f53db717294d63a8d79b5ef7ca34b3c189e"
> > > > > > >> meta              =
> > > "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > > > >>
> > > > > > >>
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > --
> > > > > > *Rudolf J. Streif*
> > > > > > Director of Embedded Solutions
> > > > > > The Linux Foundation
> > > > > >
> > > > > > rudolf.streif@linux.com
> > > > > > Phone: +1.619.537.6646
> > > > > > Skype: rudolfstreif
> > > > > > PGP: RSA 2048/2048 D6E7D28B
> > > > > >
> > > > > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > > > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > > > >
> > > > > > _______________________________________________
> > > > > > meta-ti mailing list
> > > > > > meta-ti@yoctoproject.org
> > > > > > https://lists.yoctoproject.org/listinfo/meta-ti
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > --
> > > > *Rudolf J. Streif*
> > > > Director of Embedded Solutions
> > > > The Linux Foundation
> > > >
> > > > rudolf.streif@linux.com
> > > > Phone: +1.619.537.6646
> > > > Skype: rudolfstreif
> > > > PGP: RSA 2048/2048 D6E7D28B
> > > >
> > > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > >
> > >
> > >
> >
> >
> > --
> > --
> > *Rudolf J. Streif*
> > Director of Embedded Solutions
> > The Linux Foundation
> >
> > rudolf.streif@linux.com
> > Phone: +1.619.537.6646
> > Skype: rudolfstreif
> > PGP: RSA 2048/2048 D6E7D28B
> >
> > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > Linux Foundation Training Schedule: training.linuxfoundation.org
> 
> 
> ------------------------------
> 
> Message: 2
> Date: Tue, 22 Oct 2013 15:37:32 -0400
> From: Denys Dmytriyenko <denys@ti.com>
> To: Rudolf Streif <rstreif@linuxfoundation.org>
> Cc: meta-ti@yoctoproject.org
> Subject: Re: [meta-ti] AM335x EVM and core-image-sato
> Message-ID: <20131022193732.GC9301@edge>
> Content-Type: text/plain; charset="us-ascii"
> 
> And just to double-check here as well - you did try all the binaries I sent
> you, including MLO, u-boot, zImage and DTB, right?
> 
> Denys
> 
> 
> On Mon, Oct 21, 2013 at 11:19:35PM +0100, Rudolf Streif wrote:
> > Thanks, Denys.
> >
> > Interesting point about gcc. It's version 4.8.1 which is the default for
> > poky master. Which is what I used all along.
> >
> > Thanks for sending your binaries. They fail too with the exact same
> > problem. See attached log. I found [1] on gmane which exactly describes the
> > problem I am seeing.
> >
> > If that is the case then I was wondering why my earlier builds with
> > 3.12-rc3 actually worked. And I tried them again and they are now failing
> > too. Of course I don't know anything about the specifics of the gpio and
> > the timers on the chip but it looks like a weird hardware initialization
> > condition. And I don't know if the proposed patch in [1] was ever
> > incorporated in your kernel tree.
> >
> > I don't think it's the hardware (or at least not alone) because the SD that
> > ships with it with the 3.2 kernel works just fine.
> >
> > Cheers,
> > Rudi
> >
> >
> > [1]  http://permalink.gmane.org/gmane.linux.ports.arm.omap/96044
> >
> >
> > On Mon, Oct 21, 2013 at 10:15 PM, Denys Dmytriyenko <denis@denix.org> wrote:
> >
> > > Rudi,
> > >
> > > As we discussed earlier today, let my send you my binaries for the kernel
> > > and
> > > u-boot to try on your board. I just wanted to ask you about the toolchain
> > > that
> > > you used to build yours. In our builds we use Linaro gcc-4.7 hardfp
> > > toolchain
> > > and I vaguely remember hearing about some problems with the recent gcc-4.8
> > > toolchain from oe-core... I'm wondering if you are having the same issue.
> > >
> > > --
> > > Denys
> > >
> > >
> > > On Mon, Oct 21, 2013 at 10:42:35AM -0700, Rudolf Streif wrote:
> > > > Since the 3.12-rc5 kernel does not work I decided to give the 3.8 kernel
> > > a
> > > > try. The issue with this one is that the linux-ti-staging_3.8.bb recipe
> > > > explicitly sets KERNEL_IMAGETYPE="uImage". However, the u-boot
> > > environment
> > > > has now moved on to using zImage which consequently means u-boot will not
> > > > find the kernel file.
> > > >
> > > > I do not know why KERNEL_IMAGETYPE = "uImage" was hacked into the
> > > > linux-ti-staging_3.8.bb recipe but whoever did it did not think about
> > > > u-boot. Does zImage not work with 3.8? I did not try it but it should
> > > > normally have no effect. Since there is no dependency of the u-boot
> > > recipes
> > > > on the kernel image types changes to the kernel image type for newer
> > > > versions of the kernel then automatically means that it must be changed
> > > for
> > > > the previous versions too.
> > > >
> > > > I mocked around with the u-boot environment and booted the 3.8 uImage
> > > > kernel. To my surprise the 3.8 kernel from the TI staging repository
> > > fails
> > > > with the exact same problem and stops booting with a kernel not syncing
> > > > message during the init calls. Apparently a patch must have been applied
> > > to
> > > > the kernel repo that breaks it. It has nothing possibly to do with the
> > > > build system.
> > > >
> > > > I would appreciate any hints to get this kernel built and boot on the
> > > board.
> > > >
> > > > Thanks,
> > > > Rudi
> > > >
> > > >
> > > > On Mon, Oct 21, 2013 at 1:16 AM, Rudolf Streif
> > > > <rstreif@linuxfoundation.org>wrote:
> > > >
> > > > > Denys,
> > > > >
> > > > > I recompiled the 3.12-rc5 kernel again but it still stalls during
> > > > > initialization. I attached the boot log. I don't know much about the
> > > > > platform details but maybe that's an issue you know about.
> > > > >
> > > > > Thanks,
> > > > > Rudi
> > > > >
> > > > >
> > > > >
> > > > >>
> > > > >> [1]
> > > > >> Build Configuration:
> > > > >> BB_VERSION        = "1.21.0"
> > > > >> BUILD_SYS         = "x86_64-linux"
> > > > >> NATIVELSBSTRING   = "Fedora-18"
> > > > >> TARGET_SYS        = "arm-poky-linux-gnueabi"
> > > > >> MACHINE           = "am335x-evm"
> > > > >> DISTRO            = "poky"
> > > > >> DISTRO_VERSION    = "1.5+snapshot-20131019"
> > > > >> TUNE_FEATURES     = "armv7a vfp neon"
> > > > >> TARGET_FPU        = "vfp-neon"
> > > > >> meta-yocto        = "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > >> meta-ti           = "master:c11504664990f007c3903d349e7cc135a904f8e2"
> > > > >> meta-oe           = "master:f8903f53db717294d63a8d79b5ef7ca34b3c189e"
> > > > >> meta              = "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > >>
> > > > >>
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > --
> > > > *Rudolf J. Streif*
> > > > Director of Embedded Solutions
> > > > The Linux Foundation
> > > >
> > > > rudolf.streif@linux.com
> > > > Phone: +1.619.537.6646
> > > > Skype: rudolfstreif
> > > > PGP: RSA 2048/2048 D6E7D28B
> > > >
> > > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > >
> > > > _______________________________________________
> > > > meta-ti mailing list
> > > > meta-ti@yoctoproject.org
> > > > https://lists.yoctoproject.org/listinfo/meta-ti
> > >
> > >
> >
> >
> > --
> > --
> > *Rudolf J. Streif*
> > Director of Embedded Solutions
> > The Linux Foundation
> >
> > rudolf.streif@linux.com
> > Phone: +1.619.537.6646
> > Skype: rudolfstreif
> > PGP: RSA 2048/2048 D6E7D28B
> >
> > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > Linux Foundation Training Schedule: training.linuxfoundation.org
> 
> 
> 
> 
> ------------------------------
> 
> Message: 3
> Date: Tue, 22 Oct 2013 19:52:13 +0000
> From: Vanderpool Josh <Joshua.Vanderpool@raymondcorp.com>
> To: "meta-ti@yoctoproject.org" <meta-ti@yoctoproject.org>
> Subject: Re: [meta-ti] AM335x-SK kernel panic with Poky(master),
>         meta-ti(master) core-image-base
> Message-ID:
>         <7D1F292BE8E4CC4F93079C1A7A58D5622B9FF6D7@USNYGRMBX4.raymondcorp.local>
> 
> Content-Type: text/plain; charset="us-ascii"
> 
> Thanks for the links Rudi, I looked for the file referenced in [2] "arch/arm/mach-omap2/cclock33xx_data.c" to see if the patch works but...
> 
> There is no cclock33xx_data.c rather there is in tmp/work/am335x_evm-poky-linux-gnueabi/linux-ti-staging/3.11+3.12-rc5-r20a+gitrAUTOINC+eb765a394c/git/arch/arm/mach-omap2/cclock3xxx_data.c.
> 
> however a search I found tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/linux-libc-headers/3.10-r0/linux-3.10/arch/arm/mach-omap2/cclock33xx_data.c
> 
> So it appears there is mis-match as the patch is in the 3.10 files that re in the armv7a-vfp-neon-poky-linux-gnueabi directory?
> 
> So is this something that is just lost from the 3.12-rc5 kernel and shouldn't the linux-linc-headers be the version being built?
> 
> Unfortunately being inexperienced with the project I am at a loss at how to fix this but hope this help point someone to the solution.
> 
> Regards,
> Josh
> 
> ________________________________________
> From: meta-ti-bounces@yoctoproject.org [meta-ti-bounces@yoctoproject.org] on behalf of meta-ti-request@yoctoproject.org [meta-ti-request@yoctoproject.org]
> Sent: Tuesday, October 22, 2013 3:00 PM
> To: meta-ti@yoctoproject.org
> Subject: meta-ti Digest, Vol 25, Issue 35
> 
> Send meta-ti mailing list submissions to
>         meta-ti@yoctoproject.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://lists.yoctoproject.org/listinfo/meta-ti
> or, via email, send a message with subject or body 'help' to
>         meta-ti-request@yoctoproject.org
> 
> You can reach the person managing the list at
>         meta-ti-owner@yoctoproject.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of meta-ti digest..."
> 
> 
> Today's Topics:
> 
>    1. Re: AM335x-SK kernel panic with Poky(master), meta-ti(master)
>       core-image-base (Rudolf Streif)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Tue, 22 Oct 2013 18:18:29 +0100
> From: Rudolf Streif <rstreif@linuxfoundation.org>
> To: Vanderpool Josh <Joshua.Vanderpool@raymondcorp.com>
> Cc: "meta-ti@yoctoproject.org" <meta-ti@yoctoproject.org>
> Subject: Re: [meta-ti] AM335x-SK kernel panic with Poky(master),
>         meta-ti(master) core-image-base
> Message-ID:
>         <CA+gsK-pNkDYCcecOftuopRu9XnxK3Xr9ZbSU23yekqQEczCYeg@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Hi Josh,
> 
> I am seeing the exact same problem with my am335x-evmsk and the latest
> meta-ti with 3.12-rc5 kernel. I posted it on this mailing list [1].
> 
> This problem looks like the one posted here [2].
> 
> The bad news is I don't have a solution for it. The good news is that Denys
> and I are attending the same conference and I keep harassing him about it.
> :)
> 
> Cheers,
> Rudi
> 
> 
> [1]
> https://lists.yoctoproject.org/pipermail/meta-ti/2013-October/003519.html
> [2] http://permalink.gmane.org/gmane.linux.ports.arm.omap/96044
> 
> 
> On Tue, Oct 22, 2013 at 5:28 PM, Vanderpool Josh <
> Joshua.Vanderpool@raymondcorp.com> wrote:
> 
> > I am having a similar issue as Rudi.
> > Attached is the bootLog.
> >
> >
> > Build Configuration:
> > BB_VERSION        = "1.21.0"
> > BUILD_SYS         = "x86_64-linux"
> > NATIVELSBSTRING   = "Ubuntu-12.04"
> > TARGET_SYS        = "arm-poky-linux-gnueabi"
> > MACHINE           = "am335x-evm"
> > DISTRO            = "poky"
> > DISTRO_VERSION    = "1.5+snapshot-20131022"
> > TUNE_FEATURES     = "armv7a vfp neon"
> > TARGET_FPU        = "vfp-neon"
> > meta
> > meta-yocto        = "master:505114c13d29799a25379f7ce720d632616d8776"
> > meta-ti           = "master:c11504664990f007c3903d349e7cc135a904f8e2"
> > meta-qt5          = "master:84cd4be2491895b0c3a0af64e4c83d58910d1f6c"
> >
> > Josh Vanderpool
> > Electrical Engineer
> > Control Systems
> > The Raymond Corporation
> > 20 South Canal Street
> > Greene, New York 13778-0130
> > Phone: 607-656-2617
> > Fax: 607-656-4845
> > joshua.vanderpool@raymondcorp.com
> > http://www.raymondcorp.com
> >
> >
> > Confidentiality Notice:
> >
> > The preceding e-mail message (including any attachments) contains
> > information that may be confidential, protected by applicable legal
> > privileges, or constitute non-public information. It is intended to be
> > conveyed only to the designated recipient(s). If you are not an intended
> > recipient of this message, please notify the sender by replying to this
> > message and then delete it from your system. Use, dissemination,
> > distribution or reproduction of this message by unintended recipients is
> > not authorized and may be unlawful.
> >
> >
> >
> > _______________________________________________
> > meta-ti mailing list
> > meta-ti@yoctoproject.org
> > https://lists.yoctoproject.org/listinfo/meta-ti
> >
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://lists.yoctoproject.org/pipermail/meta-ti/attachments/20131022/34dc4b0c/attachment-0001.html>
> 
> ------------------------------
> 
> _______________________________________________
> meta-ti mailing list
> meta-ti@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/meta-ti
> 
> 
> End of meta-ti Digest, Vol 25, Issue 35
> ***************************************
> Confidentiality Notice:
> 
> The preceding e-mail message (including any attachments) contains information that may be confidential, protected by applicable legal privileges, or constitute non-public information. It is intended to be conveyed only to the designated recipient(s). If you are not an intended recipient of this message, please notify the sender by replying to this message and then delete it from your system. Use, dissemination, distribution or reproduction of this message by unintended recipients is
> not authorized and may be unlawful.
> 
> 
> 
> 
> 
> ------------------------------
> 
> Message: 4
> Date: Tue, 22 Oct 2013 16:16:52 -0400
> From: Denys Dmytriyenko <denys@ti.com>
> To: Rudolf Streif <rstreif@linuxfoundation.org>,
>         meta-ti@yoctoproject.org
> Subject: Re: [meta-ti] AM335x EVM and core-image-sato
> Message-ID: <20131022201652.GB404@denix.org>
> Content-Type: text/plain; charset=us-ascii
> 
> Hey, Rudi,
> 
> Just one more quick check about your board - are you sure yours is an EVM? Or
> is it an SK-EVM i.e. a "Starter Kit"? Basically, what color is the PCB of your
> board - is it green (EVM) or purple (SK)?
> 
> For example, here's the EVM:
> https://plus.google.com/u/0/110579591067038943673/posts/SLB35o1JgBu
> 
> --
> Denys
> 
> 
> 
> On Tue, Oct 22, 2013 at 03:36:21PM -0400, Denys Dmytriyenko wrote:
> > Huh, I'm wondering if 1.2 is still supported... There were 2.0 and 2.1
> > versions of the silicon and I believe those are the latest ones. Need to
> > double-check that.
> >
> > Denys
> >
> >
> > On Tue, Oct 22, 2013 at 08:57:51AM +0100, Rudolf Streif wrote:
> > > Where do I find the PG version of the board? The barcode sticker on the
> > > board says rev 1.2.
> > >
> > >
> > > On Tue, Oct 22, 2013 at 12:19 AM, Denys Dmytriyenko <denys@ti.com> wrote:
> > >
> > > > Now I'm just wondering what PG version of the board you've got, if it
> > > > fails in
> > > > the same way with our 3.8 kernel and a known good set of 3.12 images...
> > > >
> > > >
> > > > On Mon, Oct 21, 2013 at 11:19:35PM +0100, Rudolf Streif wrote:
> > > > > Thanks, Denys.
> > > > >
> > > > > Interesting point about gcc. It's version 4.8.1 which is the default for
> > > > > poky master. Which is what I used all along.
> > > > >
> > > > > Thanks for sending your binaries. They fail too with the exact same
> > > > > problem. See attached log. I found [1] on gmane which exactly describes
> > > > the
> > > > > problem I am seeing.
> > > > >
> > > > > If that is the case then I was wondering why my earlier builds with
> > > > > 3.12-rc3 actually worked. And I tried them again and they are now failing
> > > > > too. Of course I don't know anything about the specifics of the gpio and
> > > > > the timers on the chip but it looks like a weird hardware initialization
> > > > > condition. And I don't know if the proposed patch in [1] was ever
> > > > > incorporated in your kernel tree.
> > > > >
> > > > > I don't think it's the hardware (or at least not alone) because the SD
> > > > that
> > > > > ships with it with the 3.2 kernel works just fine.
> > > > >
> > > > > Cheers,
> > > > > Rudi
> > > > >
> > > > >
> > > > > [1]  http://permalink.gmane.org/gmane.linux.ports.arm.omap/96044
> > > > >
> > > > >
> > > > > On Mon, Oct 21, 2013 at 10:15 PM, Denys Dmytriyenko <denis@denix.org>
> > > > wrote:
> > > > >
> > > > > > Rudi,
> > > > > >
> > > > > > As we discussed earlier today, let my send you my binaries for the
> > > > kernel
> > > > > > and
> > > > > > u-boot to try on your board. I just wanted to ask you about the
> > > > toolchain
> > > > > > that
> > > > > > you used to build yours. In our builds we use Linaro gcc-4.7 hardfp
> > > > > > toolchain
> > > > > > and I vaguely remember hearing about some problems with the recent
> > > > gcc-4.8
> > > > > > toolchain from oe-core... I'm wondering if you are having the same
> > > > issue.
> > > > > >
> > > > > > --
> > > > > > Denys
> > > > > >
> > > > > >
> > > > > > On Mon, Oct 21, 2013 at 10:42:35AM -0700, Rudolf Streif wrote:
> > > > > > > Since the 3.12-rc5 kernel does not work I decided to give the 3.8
> > > > kernel
> > > > > > a
> > > > > > > try. The issue with this one is that the linux-ti-staging_3.8.bbrecipe
> > > > > > > explicitly sets KERNEL_IMAGETYPE="uImage". However, the u-boot
> > > > > > environment
> > > > > > > has now moved on to using zImage which consequently means u-boot
> > > > will not
> > > > > > > find the kernel file.
> > > > > > >
> > > > > > > I do not know why KERNEL_IMAGETYPE = "uImage" was hacked into the
> > > > > > > linux-ti-staging_3.8.bb recipe but whoever did it did not think
> > > > about
> > > > > > > u-boot. Does zImage not work with 3.8? I did not try it but it should
> > > > > > > normally have no effect. Since there is no dependency of the u-boot
> > > > > > recipes
> > > > > > > on the kernel image types changes to the kernel image type for newer
> > > > > > > versions of the kernel then automatically means that it must be
> > > > changed
> > > > > > for
> > > > > > > the previous versions too.
> > > > > > >
> > > > > > > I mocked around with the u-boot environment and booted the 3.8 uImage
> > > > > > > kernel. To my surprise the 3.8 kernel from the TI staging repository
> > > > > > fails
> > > > > > > with the exact same problem and stops booting with a kernel not
> > > > syncing
> > > > > > > message during the init calls. Apparently a patch must have been
> > > > applied
> > > > > > to
> > > > > > > the kernel repo that breaks it. It has nothing possibly to do with
> > > > the
> > > > > > > build system.
> > > > > > >
> > > > > > > I would appreciate any hints to get this kernel built and boot on the
> > > > > > board.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Rudi
> > > > > > >
> > > > > > >
> > > > > > > On Mon, Oct 21, 2013 at 1:16 AM, Rudolf Streif
> > > > > > > <rstreif@linuxfoundation.org>wrote:
> > > > > > >
> > > > > > > > Denys,
> > > > > > > >
> > > > > > > > I recompiled the 3.12-rc5 kernel again but it still stalls during
> > > > > > > > initialization. I attached the boot log. I don't know much about
> > > > the
> > > > > > > > platform details but maybe that's an issue you know about.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Rudi
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >>
> > > > > > > >> [1]
> > > > > > > >> Build Configuration:
> > > > > > > >> BB_VERSION        = "1.21.0"
> > > > > > > >> BUILD_SYS         = "x86_64-linux"
> > > > > > > >> NATIVELSBSTRING   = "Fedora-18"
> > > > > > > >> TARGET_SYS        = "arm-poky-linux-gnueabi"
> > > > > > > >> MACHINE           = "am335x-evm"
> > > > > > > >> DISTRO            = "poky"
> > > > > > > >> DISTRO_VERSION    = "1.5+snapshot-20131019"
> > > > > > > >> TUNE_FEATURES     = "armv7a vfp neon"
> > > > > > > >> TARGET_FPU        = "vfp-neon"
> > > > > > > >> meta-yocto        =
> > > > "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > > > > >> meta-ti           =
> > > > "master:c11504664990f007c3903d349e7cc135a904f8e2"
> > > > > > > >> meta-oe           =
> > > > "master:f8903f53db717294d63a8d79b5ef7ca34b3c189e"
> > > > > > > >> meta              =
> > > > "master:9a1dce10bdc9254bb38e0e54199f23ae55e209a4"
> > > > > > > >>
> > > > > > > >>
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > --
> > > > > > > *Rudolf J. Streif*
> > > > > > > Director of Embedded Solutions
> > > > > > > The Linux Foundation
> > > > > > >
> > > > > > > rudolf.streif@linux.com
> > > > > > > Phone: +1.619.537.6646
> > > > > > > Skype: rudolfstreif
> > > > > > > PGP: RSA 2048/2048 D6E7D28B
> > > > > > >
> > > > > > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > > > > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > > > > >
> > > > > > > _______________________________________________
> > > > > > > meta-ti mailing list
> > > > > > > meta-ti@yoctoproject.org
> > > > > > > https://lists.yoctoproject.org/listinfo/meta-ti
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > --
> > > > > *Rudolf J. Streif*
> > > > > Director of Embedded Solutions
> > > > > The Linux Foundation
> > > > >
> > > > > rudolf.streif@linux.com
> > > > > Phone: +1.619.537.6646
> > > > > Skype: rudolfstreif
> > > > > PGP: RSA 2048/2048 D6E7D28B
> > > > >
> > > > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > > >
> > > >
> > > >
> > >
> > >
> > > --
> > > --
> > > *Rudolf J. Streif*
> > > Director of Embedded Solutions
> > > The Linux Foundation
> > >
> > > rudolf.streif@linux.com
> > > Phone: +1.619.537.6646
> > > Skype: rudolfstreif
> > > PGP: RSA 2048/2048 D6E7D28B
> > >
> > > Linux Foundation Events Schedule:  events.linuxfoundation.org
> > > Linux Foundation Training Schedule: training.linuxfoundation.org
> > _______________________________________________
> > meta-ti mailing list
> > meta-ti@yoctoproject.org
> > https://lists.yoctoproject.org/listinfo/meta-ti
> >
> 
> 
> ------------------------------
> 
> _______________________________________________
> meta-ti mailing list
> meta-ti@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/meta-ti
> 
> 
> End of meta-ti Digest, Vol 25, Issue 36
> ***************************************
> Confidentiality Notice:
> 
> The preceding e-mail message (including any attachments) contains information that may be confidential, protected by applicable legal privileges, or constitute non-public information. It is intended to be conveyed only to the designated recipient(s). If you are not an intended recipient of this message, please notify the sender by replying to this message and then delete it from your system. Use, dissemination, distribution or reproduction of this message by unintended recipients is 
> not authorized and may be unlawful.
> 
> 
> 
> _______________________________________________
> meta-ti mailing list
> meta-ti@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/meta-ti

[-- Attachment #2: am335x-evmsk.dtb --]
[-- Type: application/octet-stream, Size: 34722 bytes --]