Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-03 Thread Jan Lübbe
Hi,

On Sun, 2012-12-02 at 04:24 -0500, Robert P. J. Day wrote:
 i've isolated the commit that introduces the apparent booting
 problem on my xM:
 
 $ git show 154496f
 commit 154496f736db54123c3f6daaba18809fab54ed9b
 Merge: 05d251f bd6d12b
 Author: Sascha Hauer s.ha...@pengutronix.de
 Date:   Fri Nov 16 14:02:49 2012 +0100
 
 Merge branch 'for-next/omap'

*snip*

   so something in that omap merge is causing this.

The merge itself does not actually introduce new code.

Could you try finding the commit with git bisect?

~/git/barebox $ git bisect start
~/git/barebox $ git bisect bad 154496f736db54123c3f6daaba18809fab54ed9b
~/git/barebox $ git bisect good 05d251f
Bisecting: 3 revisions left to test after this (roughly 2 steps)
[663c466794382f803a69414c935cf0a13d4f0761] ARM OMAP4 panda: switch to new 
environment
~/git/barebox $

Then test this checkout and tell git using git bisect good or git
bisect bad.

Thanks,
Jan
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |


___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox


Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-03 Thread Robert P. J. Day
On Mon, 3 Dec 2012, Jan Lübbe wrote:

 Hi,

 On Sun, 2012-12-02 at 04:24 -0500, Robert P. J. Day wrote:
  i've isolated the commit that introduces the apparent booting
  problem on my xM:
 
  $ git show 154496f
  commit 154496f736db54123c3f6daaba18809fab54ed9b
  Merge: 05d251f bd6d12b
  Author: Sascha Hauer s.ha...@pengutronix.de
  Date:   Fri Nov 16 14:02:49 2012 +0100
 
  Merge branch 'for-next/omap'

 *snip*

so something in that omap merge is causing this.

 The merge itself does not actually introduce new code.

 Could you try finding the commit with git bisect?

 ~/git/barebox $ git bisect start
 ~/git/barebox $ git bisect bad 154496f736db54123c3f6daaba18809fab54ed9b
 ~/git/barebox $ git bisect good 05d251f
 Bisecting: 3 revisions left to test after this (roughly 2 steps)
 [663c466794382f803a69414c935cf0a13d4f0761] ARM OMAP4 panda: switch to new 
 environment
 ~/git/barebox $

 Then test this checkout and tell git using git bisect good or git
 bisect bad.

  *sigh* ... i *think* i figured out what's happening.  in order to
play with barebox on my xM, until recently, i'd done nothing more than
create a single FAT partition on a micro SD card, put MLO and
barebox.bin on it, put it in my xM, and let it boot.  until that OMAP
merge, it worked fine.

  as i read it, the new default environment for the xM *requires* a
second (ext3) partition or the boot process simply hangs, am i reading
that right?  if that's true, how can i adjust that so i can once again
just boot to barebox without having a kernel and root fs created yet?

rday

-- 


Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca

Twitter:   http://twitter.com/rpjday
LinkedIn:   http://ca.linkedin.com/in/rpjday
___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox


Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-03 Thread Jan Lübbe
On Mon, 2012-12-03 at 03:31 -0500, Robert P. J. Day wrote:
 *sigh* ... i *think* i figured out what's happening.  in order to
 play with barebox on my xM, until recently, i'd done nothing more than
 create a single FAT partition on a micro SD card, put MLO and
 barebox.bin on it, put it in my xM, and let it boot.  until that OMAP
 merge, it worked fine.
 
   as i read it, the new default environment for the xM *requires* a
 second (ext3) partition or the boot process simply hangs, am i reading
 that right?  if that's true, how can i adjust that so i can once again
 just boot to barebox without having a kernel and root fs created yet?

Only the kernel uses the second partition for its rootfs, barebox should
get to the prompt where you can interrupt autoboot anyway.

So figuring out the exact commit would be useful. The bisection should
take only a few steps. Sascha build-checks every revision, so all the
bisection steps should build fine.

Regards,
Jan
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |


___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox


Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-03 Thread Robert P. J. Day
On Mon, 3 Dec 2012, Jan Lübbe wrote:

 On Mon, 2012-12-03 at 03:31 -0500, Robert P. J. Day wrote:
  *sigh* ... i *think* i figured out what's happening.  in order to
  play with barebox on my xM, until recently, i'd done nothing more than
  create a single FAT partition on a micro SD card, put MLO and
  barebox.bin on it, put it in my xM, and let it boot.  until that OMAP
  merge, it worked fine.
 
as i read it, the new default environment for the xM *requires* a
  second (ext3) partition or the boot process simply hangs, am i reading
  that right?  if that's true, how can i adjust that so i can once again
  just boot to barebox without having a kernel and root fs created yet?

 Only the kernel uses the second partition for its rootfs, barebox
 should get to the prompt where you can interrupt autoboot anyway.

 So figuring out the exact commit would be useful. The bisection
 should take only a few steps. Sascha build-checks every revision, so
 all the bisection steps should build fine.

  actually, i *did* use bisection to isolate it -- here's the commit
that introduces the issue:

commit 5f2074bdd13ea69af5a10c3bcccb648720d6937a
Author: Sascha Hauer s.ha...@pengutronix.de
Date:   Wed Oct 31 10:16:48 2012 +0100

ARM OMAP3 beagle: switch to new environment

Signed-off-by: Sascha Hauer s.ha...@pengutronix.de

  did i forget to post that?  oops ...

rday

-- 


Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca

Twitter:   http://twitter.com/rpjday
LinkedIn:   http://ca.linkedin.com/in/rpjday
___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox


Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-03 Thread Robert P. J. Day
On Mon, 3 Dec 2012, Jan Lübbe wrote:

 On Mon, 2012-12-03 at 03:31 -0500, Robert P. J. Day wrote:
  *sigh* ... i *think* i figured out what's happening.  in order to
  play with barebox on my xM, until recently, i'd done nothing more than
  create a single FAT partition on a micro SD card, put MLO and
  barebox.bin on it, put it in my xM, and let it boot.  until that OMAP
  merge, it worked fine.
 
as i read it, the new default environment for the xM *requires* a
  second (ext3) partition or the boot process simply hangs, am i reading
  that right?  if that's true, how can i adjust that so i can once again
  just boot to barebox without having a kernel and root fs created yet?

 Only the kernel uses the second partition for its rootfs, barebox should
 get to the prompt where you can interrupt autoboot anyway.

 So figuring out the exact commit would be useful. The bisection should
 take only a few steps. Sascha build-checks every revision, so all the
 bisection steps should build fine.

  here's a copy-and-paste from one of my other sessions, where i build
MLO/barebox.bin using two consecutive commits:

= start =

barebox 2012.10.0-00286-g36969d3 #6 Sun Dec 2 05:31:40 EST 2012


Board: Texas Instrument's Beagle
I2C probe
i2c-omap i2c-omap0: bus 0 rev4.0 at 100 kHz
ehci ehci0: USB EHCI 1.00
NAND type unknown: ff,ff
No NAND device found (-19)!
omap-hsmmc omap-hsmmc0: registered as omap-hsmmc0
mci mci0: registered disk0
malloc space: 0x80c0 - 0x80ff (size  4 MB)
stack space:  0x80bf8000 - 0x80c0 (size 32 kB)
Open /dev/env0 No such file or directory
no valid environment found on /dev/env0. Using default environment
running /env/bin/init...

Hit any key to stop autoboot:  1
dhcp failed: Network is down
dhcp: Network is down
do_tftpb: /.tftp_tmp_path/zImage- - /image
could not open /.tftp_tmp_path/zImage-: Network is down

barebox@Texas Instrument's Beagle:/

barebox 2012.10.0-00287-g5f2074b #7 Sun Dec 2 05:33:23 EST 2012


Board: Texas Instrument's Beagle
NAND type unknown: ff,ff
No NAND device found (-19)!
omap-hsmmc omap-hsmmc0: registered as omap-hsmmc0
mci mci0: registered disk0
malloc space: 0x87bfff10 - 0x870f (size  4 MB)
stack space:  0x4020f000 - 0x4020fc00 (size  3 kB)
booting from MMC1

= end =

  as you can see, commit 36969d3 has the old behaviour (allowing
booting), while the next commit 5f2074b which introduces the new
beagle environment causes the boot to hang at booting from MMC1.

rday

-- 


Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca

Twitter:   http://twitter.com/rpjday
LinkedIn:   http://ca.linkedin.com/in/rpjday
___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox


Re: suddenly, i can't boot barebox on my beagle xM anymore [ISOLATED]

2012-12-02 Thread Robert P. J. Day

  i've isolated the commit that introduces the apparent booting
problem on my xM:

$ git show 154496f
commit 154496f736db54123c3f6daaba18809fab54ed9b
Merge: 05d251f bd6d12b
Author: Sascha Hauer s.ha...@pengutronix.de
Date:   Fri Nov 16 14:02:49 2012 +0100

Merge branch 'for-next/omap'

$

  if i build MLO and barebox.bin with that commit, booting produces:

barebox 2012.11.0-00237-g154496f #1 Sat Dec 1 18:43:27 EST 2012


Board: Texas Instrument's Beagle
NAND type unknown: ff,ff
No NAND device found (-19)!
gpmc_nand gpmc_nand0: probe failed: error 6
omap-hsmmc omap-hsmmc0: registered as omap-hsmmc0
mci mci0: registered disk0
malloc space: 0x87bfff10 - 0x870f (size  4 MB)
stack space:  0x4020f000 - 0x4020fc00 (size  3 kB)
booting from MMC1
870f (size  4 MB)
stack space:  0x4020f000 - 0x4020fc00 (size  3 kB)
booting from MMC1

... and hangs ...

  if, however, i build from the parent commit 05d251f, then i'm back
to a full boot to barebox:

barebox 2012.11.0-00230-g05d251f #2 Sun Dec 2 04:04:37 EST 2012


Board: Texas Instrument's Beagle
I2C probe
i2c-omap i2c-omap0: bus 0 rev4.0 at 100 kHz
ehci ehci0: USB EHCI 1.00
NAND type unknown: ff,ff
No NAND device found (-19)!
gpmc_nand gpmc_nand0: probe failed: No such device or address
omap-hsmmc omap-hsmmc0: registered as omap-hsmmc0
mci mci0: registered disk0
malloc space: 0x80c0 - 0x80ff (size  4 MB)
stack space:  0x80bf8000 - 0x80c0 (size 32 kB)
Open /dev/env0 No such file or directory
no valid environment found on /dev/env0. Using default environment
running /env/bin/init...

Hit any key to stop autoboot:  0
dhcp failed: Network is down
dhcp: Network is down
do_tftpb: /.tftp_tmp_path/zImage- - /image
could not open /.tftp_tmp_path/zImage-: Network is down

barebox@Texas Instrument's Beagle:/

  so something in that omap merge is causing this.

rday

-- 


Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca

Twitter:   http://twitter.com/rpjday
LinkedIn:   http://ca.linkedin.com/in/rpjday


___
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox