LSK merge linux-stable review

LSK merge linux-stable review

[Catalin Marinas]

On Fri, Jan 10, 2014 at 02:58:20AM +0000, Alex Shi wrote:
> commit 450cdfa9c5713551ab1438ab210679868b52a994
> Merge: bf78886 8b4ed85
> Author: Alex Shi <alex.shi@linaro.org>
> Date:   Fri Jan 10 10:50:57 2014 +0800
> 
>     Merge remote-tracking branch 'stable/linux-3.10.y' into linux-linaro-lsk
>     
>     Conflicts:
>         arch/arm64/kernel/smp.c
>     
>     Signed-off-by: Alex Shi <alex.shi@linaro.org>
> 
> diff --cc arch/arm64/kernel/smp.c
> index 5e9d274,9c93e12..87340fd
> --- a/arch/arm64/kernel/smp.c
> +++ b/arch/arm64/kernel/smp.c
> @@@ -150,17 -187,17 +150,10 @@@ asmlinkage void __cpuinit secondary_sta
>         preempt_disable();
>         trace_hardirqs_off();
>   
>  -      /*
>  -       * Let the primary processor know we're out of the
>  -       * pen, then head off into the C entry point
>  -       */
>  -      write_pen_release(INVALID_HWID);
>  +      if (cpu_ops[cpu]->cpu_postboot)
>  +              cpu_ops[cpu]->cpu_postboot();
>   
> --      /*
> -        * Enable local interrupts.
> -        */
> -       notify_cpu_starting(cpu);

This kind of diff always confuses me but notify_cpu_starting() should be
present in the resulting file. Is it further down?

-- 
Catalin

LSK merge linux-stable review

[Alex Shi]

On 01/10/2014 06:56 PM, Catalin Marinas wrote:
> On Fri, Jan 10, 2014 at 02:58:20AM +0000, Alex Shi wrote:
>> commit 450cdfa9c5713551ab1438ab210679868b52a994
>> Merge: bf78886 8b4ed85
>> Author: Alex Shi <alex.shi@linaro.org>
>> Date:   Fri Jan 10 10:50:57 2014 +0800
>>
>>     Merge remote-tracking branch 'stable/linux-3.10.y' into linux-linaro-lsk
>>     
>>     Conflicts:
>>         arch/arm64/kernel/smp.c
>>     
>>     Signed-off-by: Alex Shi <alex.shi@linaro.org>
>>
>> diff --cc arch/arm64/kernel/smp.c
>> index 5e9d274,9c93e12..87340fd
>> --- a/arch/arm64/kernel/smp.c
>> +++ b/arch/arm64/kernel/smp.c
>> @@@ -150,17 -187,17 +150,10 @@@ asmlinkage void __cpuinit secondary_sta
>>         preempt_disable();
>>         trace_hardirqs_off();
>>   
>>  -      /*
>>  -       * Let the primary processor know we're out of the
>>  -       * pen, then head off into the C entry point
>>  -       */
>>  -      write_pen_release(INVALID_HWID);
>>  +      if (cpu_ops[cpu]->cpu_postboot)
>>  +              cpu_ops[cpu]->cpu_postboot();
>>   
>> --      /*
>> -        * Enable local interrupts.
>> -        */
>> -       notify_cpu_starting(cpu);
> 
> This kind of diff always confuses me but notify_cpu_starting() should be
> present in the resulting file. Is it further down?
> 

the first column '-/+' means contents was removed/added from file
5e9d274. the 2nd column '-/+' means contents was removed/added from file
9c93e12.

Yes, The notify_cpu_starting kept in final file later.

-- 
Thanks
    Alex

LSK merge linux-stable review

[Catalin Marinas]

On Fri, Jan 10, 2014 at 03:31:00PM +0000, Alex Shi wrote:
> On 01/10/2014 06:56 PM, Catalin Marinas wrote:
> > On Fri, Jan 10, 2014 at 02:58:20AM +0000, Alex Shi wrote:
> >> commit 450cdfa9c5713551ab1438ab210679868b52a994
> >> Merge: bf78886 8b4ed85
> >> Author: Alex Shi <alex.shi@linaro.org>
> >> Date:   Fri Jan 10 10:50:57 2014 +0800
> >>
> >>     Merge remote-tracking branch 'stable/linux-3.10.y' into linux-linaro-lsk
> >>     
> >>     Conflicts:
> >>         arch/arm64/kernel/smp.c
> >>     
> >>     Signed-off-by: Alex Shi <alex.shi@linaro.org>
> >>
> >> diff --cc arch/arm64/kernel/smp.c
> >> index 5e9d274,9c93e12..87340fd
> >> --- a/arch/arm64/kernel/smp.c
> >> +++ b/arch/arm64/kernel/smp.c
> >> @@@ -150,17 -187,17 +150,10 @@@ asmlinkage void __cpuinit secondary_sta
> >>         preempt_disable();
> >>         trace_hardirqs_off();
> >>   
> >>  -      /*
> >>  -       * Let the primary processor know we're out of the
> >>  -       * pen, then head off into the C entry point
> >>  -       */
> >>  -      write_pen_release(INVALID_HWID);
> >>  +      if (cpu_ops[cpu]->cpu_postboot)
> >>  +              cpu_ops[cpu]->cpu_postboot();
> >>   
> >> --      /*
> >> -        * Enable local interrupts.
> >> -        */
> >> -       notify_cpu_starting(cpu);
> > 
> > This kind of diff always confuses me but notify_cpu_starting() should be
> > present in the resulting file. Is it further down?
> > 
> 
> the first column '-/+' means contents was removed/added from file
> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
> 9c93e12.
> 
> Yes, The notify_cpu_starting kept in final file later.

Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
enabling of GIC before CPUs are set online)

-- 
Catalin

LSK merge linux-stable review

[Alex Shi]

>>
>> the first column '-/+' means contents was removed/added from file
>> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
>> 9c93e12.
>>
>> Yes, The notify_cpu_starting kept in final file later.
> 
> Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
> enabling of GIC before CPUs are set online)
> 

Give you the part of modified smp.c file here :)
====

        printk("CPU%u: Booted secondary processor\n", cpu);

        /*
         * All kernel threads share the same mm context; grab a
         * reference and switch to it.
         */
        atomic_inc(&mm->mm_count);
        current->active_mm = mm;
        cpumask_set_cpu(cpu, mm_cpumask(mm));

        /*
         * TTBR0 is only used for the identity mapping at this stage. Make it
         * point to zero page to avoid speculatively fetching new entries.
         */
        cpu_set_reserved_ttbr0();
        flush_tlb_all();

        preempt_disable();
        trace_hardirqs_off();

        if (cpu_ops[cpu]->cpu_postboot)
                cpu_ops[cpu]->cpu_postboot();

        smp_store_cpu_info(cpu);

        /*
         * OK, now it's safe to let the boot CPU continue.  Wait for
         * the CPU migration code to notice that the CPU is online
         * before we continue.
         */
        set_cpu_online(cpu, true);
        complete(&cpu_running);

        /*
         * Enable GIC and timers.
         */
        notify_cpu_starting(cpu);

        local_irq_enable();
        local_fiq_enable();

        /*
         * OK, it's off to the idle thread for us
         */
        cpu_startup_entry(CPUHP_ONLINE);
}

-- 
Thanks
    Alex

LSK merge linux-stable review

[Catalin Marinas]

On Sat, Jan 11, 2014 at 01:35:52AM +0000, Alex Shi wrote:
> 
> >>
> >> the first column '-/+' means contents was removed/added from file
> >> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
> >> 9c93e12.
> >>
> >> Yes, The notify_cpu_starting kept in final file later.
> > 
> > Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
> > enabling of GIC before CPUs are set online)
> 
> Give you the part of modified smp.c file here :)

As I said, you may want to cherry pick 7ade67b5984d0a. It comes in handy
with GICv3 support (if you later plan to back-port it to this version of
LSK).

-- 
Catalin

LSK merge linux-stable review

[Alex Shi]

On 01/13/2014 11:55 PM, Catalin Marinas wrote:
> On Sat, Jan 11, 2014 at 01:35:52AM +0000, Alex Shi wrote:
>>
>>>>
>>>> the first column '-/+' means contents was removed/added from file
>>>> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
>>>> 9c93e12.
>>>>
>>>> Yes, The notify_cpu_starting kept in final file later.
>>>
>>> Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
>>> enabling of GIC before CPUs are set online)
>>
>> Give you the part of modified smp.c file here :)
> 
> As I said, you may want to cherry pick 7ade67b5984d0a. It comes in handy
> with GICv3 support (if you later plan to back-port it to this version of
> LSK).
> 

Sorry for not make it clear.
We had picked up 7ade67b5984d0a as I showed the contents of function
secondary_start_kernel().

-- 
Thanks
    Alex

LSK merge linux-stable review

[Catalin Marinas]

On Tue, Jan 14, 2014 at 12:32:06AM +0000, Alex Shi wrote:
> On 01/13/2014 11:55 PM, Catalin Marinas wrote:
> > On Sat, Jan 11, 2014 at 01:35:52AM +0000, Alex Shi wrote:
> >>>> the first column '-/+' means contents was removed/added from file
> >>>> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
> >>>> 9c93e12.
> >>>>
> >>>> Yes, The notify_cpu_starting kept in final file later.
> >>>
> >>> Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
> >>> enabling of GIC before CPUs are set online)
> >>
> >> Give you the part of modified smp.c file here :)
> > 
> > As I said, you may want to cherry pick 7ade67b5984d0a. It comes in handy
> > with GICv3 support (if you later plan to back-port it to this version of
> > LSK).
> 
> Sorry for not make it clear.
> We had picked up 7ade67b5984d0a as I showed the contents of function
> secondary_start_kernel().

I don't think you picked it since this commit moves
notify_cpu_starting() before set_cpu_online() which is not the case for
the code you pasted.

-- 
Catalin

LSK merge linux-stable review

[Alex Shi]

On 01/14/2014 05:45 PM, Catalin Marinas wrote:
> On Tue, Jan 14, 2014 at 12:32:06AM +0000, Alex Shi wrote:
>> On 01/13/2014 11:55 PM, Catalin Marinas wrote:
>>> On Sat, Jan 11, 2014 at 01:35:52AM +0000, Alex Shi wrote:
>>>>>> the first column '-/+' means contents was removed/added from file
>>>>>> 5e9d274. the 2nd column '-/+' means contents was removed/added from file
>>>>>> 9c93e12.
>>>>>>
>>>>>> Yes, The notify_cpu_starting kept in final file later.
>>>>>
>>>>> Just make sure it is in line with commit 7ade67b5984d0a (arm64: move
>>>>> enabling of GIC before CPUs are set online)
>>>>
>>>> Give you the part of modified smp.c file here :)
>>>
>>> As I said, you may want to cherry pick 7ade67b5984d0a. It comes in handy
>>> with GICv3 support (if you later plan to back-port it to this version of
>>> LSK).
>>
>> Sorry for not make it clear.
>> We had picked up 7ade67b5984d0a as I showed the contents of function
>> secondary_start_kernel().
> 
> I don't think you picked it since this commit moves
> notify_cpu_starting() before set_cpu_online() which is not the case for
> the code you pasted.
> 

Ops, my stupid mistaken. Just cherry-picked this patch into lsk.

Thanks again!

-- 
Thanks
    Alex

LSK merge linux-stable review

[Alex Shi]

On 01/10/2014 06:41 PM, Mark Brown wrote:
> On 10 January 2014 02:58, Alex Shi <alex.shi@linaro.org
> <mailto:alex.shi@linaro.org>> wrote:
> 
>     Mark & Catalin,
> 
>     Could you like to review the merge conflict patch, which at:
>     git://git.linaro.org/kernel/linux-linaro-stable.git
>     <http://git.linaro.org/kernel/linux-linaro-stable.git>
>     linux-linaro-lsk-test
> 
> 
> This looks good (if nothing else it ends up with the same thing we have
> in mainline) providing it tests out OK.

Thanks for review. The build for -test branch is fine. guess that means
compile and reboot is fine.
But test result link still can not show test cases results. :(

https://ci.linaro.org/jenkins/job/linux-linaro-stable-lsk-pre-merge/

-- 
Thanks
    Alex

LSK merge linux-stable review

[Alex Shi]

On 01/10/2014 11:51 PM, Alex Shi wrote:
> On 01/10/2014 06:41 PM, Mark Brown wrote:
>> On 10 January 2014 02:58, Alex Shi <alex.shi@linaro.org
>> <mailto:alex.shi@linaro.org>> wrote:
>>
>>     Mark & Catalin,
>>
>>     Could you like to review the merge conflict patch, which at:
>>     git://git.linaro.org/kernel/linux-linaro-stable.git
>>     <http://git.linaro.org/kernel/linux-linaro-stable.git>
>>     linux-linaro-lsk-test
>>
>>
>> This looks good (if nothing else it ends up with the same thing we have
>> in mainline) providing it tests out OK.
> 
> Thanks for review. The build for -test branch is fine. guess that means
> compile and reboot is fine.

Anyway, I updated this merge into lsk.
> But test result link still can not show test cases results. :(
> 
> https://ci.linaro.org/jenkins/job/linux-linaro-stable-lsk-pre-merge/
> 


-- 
Thanks
    Alex