Re: am3517: failed to boot 3.10-rc1

[Kevin Hilman <khilman@linaro.org>]

Yegor Yefremov <yegor_sub1@visionsystems.de> writes:

> On 15.05.2013 10:37, Yegor Yefremov wrote:
>> On 14.05.2013 15:01, Yegor Yefremov wrote:
>>> On 14.05.2013 14:52, Felipe Balbi wrote:
>>>> On Tue, May 14, 2013 at 11:24:44AM +0200, Yegor Yefremov wrote:
>>>>> Trying to boot 3.10-rc1 on an am3515 based board. With the same
>>>>> .config as 3.7 the system comes to RTC stops there. I've also tried
>>>>> make omap2plus_defconfig with no visible difference. I'm booting from
>>>>> MMC card and it will be detected by the system. Kernel is from
>>>>> http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git, latest
>>>>> commit f722406faae2d073cc1d01063d1123c35425939e. Are there any pending
>>>>> patches needed to boot on am3517?
>>>> does v3.9 work ? Can you bisect ?
>>>
>>> 'git checkout v3.9' version is working, will try to bisect.
>> 
>> This is bisect's result:
>> 
>> f7b861b7a6d9d1838cbbb5f4053e61578b86d134 is the first bad commit
>> commit f7b861b7a6d9d1838cbbb5f4053e61578b86d134
>> Author: Thomas Gleixner <tglx@linutronix.de>
>> Date:   Thu Mar 21 22:49:38 2013 +0100
>> 
>>     arm: Use generic idle loop
>> 
>>     Use the generic idle loop and replace enable/disable_hlt with the
>>     respective core functions.
>> 
>>     Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
>>     Cc: Linus Torvalds <torvalds@linux-foundation.org>
>>     Cc: Rusty Russell <rusty@rustcorp.com.au>
>>     Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
>>     Cc: Peter Zijlstra <peterz@infradead.org>
>>     Reviewed-by: Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
>>     Cc: Magnus Damm <magnus.damm@gmail.com>
>>     Cc: Russell King <linux@arm.linux.org.uk>
>>     Tested-by: Kevin Hilman <khilman@linaro.org> # OMAP
>>     Link: http://lkml.kernel.org/r/20130321215233.826238797@linutronix.de
>> 
>> :040000 040000 887517403888ff3947bce216aa4c1ff3ce3af434 8dc5236688a80fc64fa4112ce6136ecf3a19bac9 M      arch
>
> We had such a problem long ago
> (https://patchwork.kernel.org/patch/1220061/ and
> http://www.spinics.net/lists/arm-kernel/msg168865.html) and the
> workaround was to use hohlt in kernel parameters. But with this patch
> the "nohlt" has gone. So now we need a solution, that fixes the root
> of the problem.

The broken 'nohlt' feature was indeed a regression caused by the above
patch, but there's already a fix queued for v3.10-rc2[1], so -rc1 will
need the patch applied manually.

Kevin

[1] https://lkml.org/lkml/2013/5/5/6