exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Hi Chander, Abhilash,

I'm using the exynos5422 odroid-xu3 and only 6 of 8 cores are coming
online when using MCPM[1].  All the A15s are booting but only 2 of the
A7s, one of which is the primary CPU.

After booting, if I try to manually bring one of the CPUs online via
/sys/devices/.../cpuX/online, I hit the BUG() in
mcpm-exynos.c:exynos_power_up().

Any ideas why 2 of the A7s aren't booting up?

Kevin

[1]
[    0.001719] CPU: Testing write buffer coherency: ok
[    0.002171] CPU0: update cpu_capacity 448
[    0.002187] CPU0: thread -1, cpu 0, socket 1, mpidr 80000100
[    0.002429] Setting up static identity map for 0x4046df70 - 0x4046dfc8
[    0.002732] ARM CCI driver probed
[    0.002815] exynos_mcpm_usage_count_init: cpu 0 cluster 1
[    0.003010] Exynos MCPM support installed
[    0.030310] mcpm_boot_secondary: logical CPU 1 is physical CPU 0 cluster 0
[    0.030326] exynos_power_up: cpu 0 cluster 0
[    0.030626] CPU1: Booted secondary processor
[    0.030686] CPU1: update cpu_capacity 1535
[    0.030693] CPU1: thread -1, cpu 0, socket 0, mpidr 80000000
[    0.040314] mcpm_boot_secondary: logical CPU 2 is physical CPU 1 cluster 0
[    0.040340] exynos_power_up: cpu 1 cluster 0
[    0.040629] CPU2: Booted secondary processor
[    0.040672] CPU2: update cpu_capacity 1535
[    0.040679] CPU2: thread -1, cpu 1, socket 0, mpidr 80000001
[    0.050321] mcpm_boot_secondary: logical CPU 3 is physical CPU 2 cluster 0
[    0.050347] exynos_power_up: cpu 2 cluster 0
[    0.050659] CPU3: Booted secondary processor
[    0.050703] CPU3: update cpu_capacity 1535
[    0.050710] CPU3: thread -1, cpu 2, socket 0, mpidr 80000002
[    0.060316] mcpm_boot_secondary: logical CPU 4 is physical CPU 3 cluster 0
[    0.060343] exynos_power_up: cpu 3 cluster 0
[    0.060631] CPU4: Booted secondary processor
[    0.060675] CPU4: update cpu_capacity 1535
[    0.060682] CPU4: thread -1, cpu 3, socket 0, mpidr 80000003
[    0.070316] mcpm_boot_secondary: logical CPU 5 is physical CPU 1 cluster 1
[    0.070343] exynos_power_up: cpu 1 cluster 1
[    1.070007] CPU5: failed to come online
[    1.080322] mcpm_boot_secondary: logical CPU 6 is physical CPU 2 cluster 1
[    1.080337] exynos_power_up: cpu 2 cluster 1
[    1.080625] CPU6: Booted secondary processor
[    1.080691] CPU6: update cpu_capacity 448
[    1.080699] CPU6: thread -1, cpu 2, socket 1, mpidr 80000102
[    1.090331] mcpm_boot_secondary: logical CPU 7 is physical CPU 3 cluster 1
[    1.090356] exynos_power_up: cpu 3 cluster 1
[    2.090023] CPU7: failed to come online
[    2.090055] Brought up 6 CPUs
[    2.090067] SMP: Total of 6 processors activated.
[    2.090079] CPU: WARNING: CPU(s) started in wrong/inconsistent
modes (primary CPU mode 0x13)
[    2.090089] CPU: This may indicate a broken bootloader or firmware.

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Abhilash Kesavan]

Hi Kevin,

I had noticed a similar issue where only 6 cores would come up during development on 5420. This was because there was a check in our u-boot secondary boot-up path which would put a core into WFI if a flag was not present and hence fail to boot. I wonder if it is the same situation in your case too.
Is the u-boot for odroid-xu3 publically available or can you check if there is any such code present in your secondary bring-up path.

Regards,
Abhilash
------- Original Message -------
Sender : Kevin Hilman<khilman@linaro.org>
Date : Oct 09, 2014 00:28 (GMT+05:30)
Title : exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

Hi Chander, Abhilash,

I'm using the exynos5422 odroid-xu3 and only 6 of 8 cores are coming
online when using MCPM[1].  All the A15s are booting but only 2 of the
A7s, one of which is the primary CPU.

After booting, if I try to manually bring one of the CPUs online via
/sys/devices/.../cpuX/online, I hit the BUG() in
mcpm-exynos.c:exynos_power_up().

Any ideas why 2 of the A7s aren't booting up?

Kevin

[1]
[    0.001719] CPU: Testing write buffer coherency: ok
[    0.002171] CPU0: update cpu_capacity 448
[    0.002187] CPU0: thread -1, cpu 0, socket 1, mpidr 80000100
[    0.002429] Setting up static identity map for 0x4046df70 - 0x4046dfc8
[    0.002732] ARM CCI driver probed
[    0.002815] exynos_mcpm_usage_count_init: cpu 0 cluster 1
[    0.003010] Exynos MCPM support installed
[    0.030310] mcpm_boot_secondary: logical CPU 1 is physical CPU 0 cluster 0
[    0.030326] exynos_power_up: cpu 0 cluster 0
[    0.030626] CPU1: Booted secondary processor
[    0.030686] CPU1: update cpu_capacity 1535
[    0.030693] CPU1: thread -1, cpu 0, socket 0, mpidr 80000000
[    0.040314] mcpm_boot_secondary: logical CPU 2 is physical CPU 1 cluster 0
[    0.040340] exynos_power_up: cpu 1 cluster 0
[    0.040629] CPU2: Booted secondary processor
[    0.040672] CPU2: update cpu_capacity 1535
[    0.040679] CPU2: thread -1, cpu 1, socket 0, mpidr 80000001
[    0.050321] mcpm_boot_secondary: logical CPU 3 is physical CPU 2 cluster 0
[    0.050347] exynos_power_up: cpu 2 cluster 0
[    0.050659] CPU3: Booted secondary processor
[    0.050703] CPU3: update cpu_capacity 1535
[    0.050710] CPU3: thread -1, cpu 2, socket 0, mpidr 80000002
[    0.060316] mcpm_boot_secondary: logical CPU 4 is physical CPU 3 cluster 0
[    0.060343] exynos_power_up: cpu 3 cluster 0
[    0.060631] CPU4: Booted secondary processor
[    0.060675] CPU4: update cpu_capacity 1535
[    0.060682] CPU4: thread -1, cpu 3, socket 0, mpidr 80000003
[    0.070316] mcpm_boot_secondary: logical CPU 5 is physical CPU 1 cluster 1
[    0.070343] exynos_power_up: cpu 1 cluster 1
[    1.070007] CPU5: failed to come online
[    1.080322] mcpm_boot_secondary: logical CPU 6 is physical CPU 2 cluster 1
[    1.080337] exynos_power_up: cpu 2 cluster 1
[    1.080625] CPU6: Booted secondary processor
[    1.080691] CPU6: update cpu_capacity 448
[    1.080699] CPU6: thread -1, cpu 2, socket 1, mpidr 80000102
[    1.090331] mcpm_boot_secondary: logical CPU 7 is physical CPU 3 cluster 1
[    1.090356] exynos_power_up: cpu 3 cluster 1
[    2.090023] CPU7: failed to come online
[    2.090055] Brought up 6 CPUs
[    2.090067] SMP: Total of 6 processors activated.
[    2.090079] CPU: WARNING: CPU(s) started in wrong/inconsistent
modes (primary CPU mode 0x13)
[    2.090089] CPU: This may indicate a broken bootloader or firmware.

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Hi Abhilash,

Abhilash Kesavan <a.kesavan@samsung.com> writes:

> I had noticed a similar issue where only 6 cores would come up during
> development on 5420. This was because there was a check in our u-boot
> secondary boot-up path which would put a core into WFI if a flag was
> not present and hence fail to boot. I wonder if it is the same
> situation in your case too.
> Is the u-boot for odroid-xu3 publically available or can you check if there is any such code present in your secondary bring-up path.

I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
currently using the pre-built binaries that are in the
sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
needed.

I'm not very familiar with u-boot, but if f you have some pointers where
to start looking in that code, that would be helpful.

Kevin

[1] https://github.com/hardkernel/u-boot.git

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

Hi,

Kevin Hilman <khilman <at> kernel.org> writes:

> 
> Hi Abhilash,
> 
> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
> 
> > I had noticed a similar issue where only 6 cores would come up during
> > development on 5420. This was because there was a check in our u-boot
> > secondary boot-up path which would put a core into WFI if a flag was
> > not present and hence fail to boot. I wonder if it is the same
> > situation in your case too.
> > Is the u-boot for odroid-xu3 publically available or can you check if
there is any such code present in your
> secondary bring-up path.
> 
> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
> currently using the pre-built binaries that are in the
> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
> needed.
> 
> I'm not very familiar with u-boot, but if f you have some pointers where
> to start looking in that code, that would be helpful.
> 
> Kevin
> 
> [1] https://github.com/hardkernel/u-boot.git
> 
> 

It seems to need magic code.

Please refer
https://github.com/abhijeet-dev/linux-samsung/commit
/553b7f40465ab4fad6c270f289cd4780942f7fae

Thanks.

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
> Hi,
>
> Kevin Hilman <khilman <at> kernel.org> writes:
>
>>
>> Hi Abhilash,
>>
>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>
>> > I had noticed a similar issue where only 6 cores would come up during
>> > development on 5420. This was because there was a check in our u-boot
>> > secondary boot-up path which would put a core into WFI if a flag was
>> > not present and hence fail to boot. I wonder if it is the same
>> > situation in your case too.
>> > Is the u-boot for odroid-xu3 publically available or can you check if
> there is any such code present in your
>> secondary bring-up path.
>>
>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>> currently using the pre-built binaries that are in the
>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>> needed.
>>
>> I'm not very familiar with u-boot, but if f you have some pointers where
>> to start looking in that code, that would be helpful.
>>
>> Kevin
>>
>> [1] https://github.com/hardkernel/u-boot.git
>>
>>
>
> It seems to need magic code.
>
> Please refer
> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae

Thanks, Abhilash had pointed me to the same code, and it works.  But...

Can someone explain this "magic", ideally in the form of a changelog
and patch submitted upstream?

Kevin

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

Hi,

On 10/17/2014 09:06 PM, Kevin Hilman wrote:
> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>> Hi,
>>
>> Kevin Hilman <khilman <at> kernel.org> writes:
>>
>>>
>>> Hi hdmi_resources_init,
>>>
>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>
>>>> I had noticed a similar issue where only 6 cores would come up during
>>>> development on 5420. This was because there was a check in our u-boot
>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>> not present and hence fail to boot. I wonder if it is the same
>>>> situation in your case too.
>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>> there is any such code present in your
>>> secondary bring-up path.
>>>
>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>> currently using the pre-built binaries that are in the
>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>> needed.
>>>
>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>> to start looking in that code, that would be helpful.
>>>
>>> Kevin
>>>
>>> [1] https://github.com/hardkernel/u-boot.git
>>>
>>>
>>
>> It seems to need magic code.
>>
>> Please refer
>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
> 
> Thanks, Abhilash had pointed me to the same code, and it works.  But...
> 
> Can someone explain this "magic", ideally in the form of a changelog
> and patch submitted upstream?
> 

I also really wonder.

One more problem, cpu0 doesn't boot after offline and system is halted.

# echo 0 > /sys/devices/system/cpu/cpu0/online
[   28.027852] IRQ152 no longer affine to CPU0
[   28.028087] IRQ256 no longer affine to CPU0
[   28.028344] CPU0: shutdown
# 
# echo 1 > /sys/devices/system/cpu/cpu0/online

So, it's failed script[1] to test MCPM functionalities

Thanks.

[1]: http://www.spinics.net/lists/arm-kernel/msg330946.html

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

Add Cc the patch author and committer from github.

On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
> Hi,
> 
> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>> Hi,
>>>
>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>
>>>>
>>>> Hi hdmi_resources_init,
>>>>
>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>
>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>> development on 5420. This was because there was a check in our u-boot
>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>> situation in your case too.
>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>> there is any such code present in your
>>>> secondary bring-up path.
>>>>
>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>> currently using the pre-built binaries that are in the
>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>> needed.
>>>>
>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>> to start looking in that code, that would be helpful.
>>>>
>>>> Kevin
>>>>
>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>
>>>>
>>>
>>> It seems to need magic code.
>>>
>>> Please refer
>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>
>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>
>> Can someone explain this "magic", ideally in the form of a changelog
>> and patch submitted upstream?
>>
> 
> I also really wonder.
> 
> One more problem, cpu0 doesn't boot after offline and system is halted.
> 
> # echo 0 > /sys/devices/system/cpu/cpu0/online
> [   28.027852] IRQ152 no longer affine to CPU0
> [   28.028087] IRQ256 no longer affine to CPU0
> [   28.028344] CPU0: shutdown
> # 
> # echo 1 > /sys/devices/system/cpu/cpu0/online
> 
> So, it's failed script[1] to test MCPM functionalities
> 
> Thanks.
> 
> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
> 

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Mauro Ribeiro]

Joonyoung,

This patch was provided by Samsung and if someone can explain is you
guys from Samsung.
I have no idea of the logics behind this or what it would take to fix it.
Keep in mind that our u-boot and kernel (downstream 3.10) is part of
Samsung provided BSP.
Would be nice if you guys could explain this.



On Wed, Oct 22, 2014 at 11:17 PM, Joonyoung Shim
<jy0922.shim@samsung.com> wrote:
> Add Cc the patch author and committer from github.
>
> On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
>> Hi,
>>
>> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>>> Hi,
>>>>
>>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>>
>>>>>
>>>>> Hi hdmi_resources_init,
>>>>>
>>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>>
>>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>>> development on 5420. This was because there was a check in our u-boot
>>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>>> situation in your case too.
>>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>>> there is any such code present in your
>>>>> secondary bring-up path.
>>>>>
>>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>>> currently using the pre-built binaries that are in the
>>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>>> needed.
>>>>>
>>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>>> to start looking in that code, that would be helpful.
>>>>>
>>>>> Kevin
>>>>>
>>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>>
>>>>>
>>>>
>>>> It seems to need magic code.
>>>>
>>>> Please refer
>>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>
>>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>>
>>> Can someone explain this "magic", ideally in the form of a changelog
>>> and patch submitted upstream?
>>>
>>
>> I also really wonder.
>>
>> One more problem, cpu0 doesn't boot after offline and system is halted.
>>
>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>> [   28.027852] IRQ152 no longer affine to CPU0
>> [   28.028087] IRQ256 no longer affine to CPU0
>> [   28.028344] CPU0: shutdown
>> #
>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>>
>> So, it's failed script[1] to test MCPM functionalities
>>
>> Thanks.
>>
>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
>>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

On 10/23/2014 10:25 AM, Mauro Ribeiro wrote:
> Joonyoung,
> 
> This patch was provided by Samsung and if someone can explain is you
> guys from Samsung.
> I have no idea of the logics behind this or what it would take to fix it.
> Keep in mind that our u-boot and kernel (downstream 3.10) is part of
> Samsung provided BSP.
> Would be nice if you guys could explain this.
> 

Mauro,

Thanks to reply, do you know author of this patch written by Samsung?


To Kukjin,

The patch[1] seems to come from S.LSI guys, could you contact about this?

[1]: https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae

> 
> 
> On Wed, Oct 22, 2014 at 11:17 PM, Joonyoung Shim
> <jy0922.shim@samsung.com> wrote:
>> Add Cc the patch author and committer from github.
>>
>> On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
>>> Hi,
>>>
>>> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>>>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>>>> Hi,
>>>>>
>>>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>>>
>>>>>>
>>>>>> Hi hdmi_resources_init,
>>>>>>
>>>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>>>
>>>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>>>> development on 5420. This was because there was a check in our u-boot
>>>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>>>> situation in your case too.
>>>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>>>> there is any such code present in your
>>>>>> secondary bring-up path.
>>>>>>
>>>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>>>> currently using the pre-built binaries that are in the
>>>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>>>> needed.
>>>>>>
>>>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>>>> to start looking in that code, that would be helpful.
>>>>>>
>>>>>> Kevin
>>>>>>
>>>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>>>
>>>>>>
>>>>>
>>>>> It seems to need magic code.
>>>>>
>>>>> Please refer
>>>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>>
>>>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>>>
>>>> Can someone explain this "magic", ideally in the form of a changelog
>>>> and patch submitted upstream?
>>>>
>>>
>>> I also really wonder.
>>>
>>> One more problem, cpu0 doesn't boot after offline and system is halted.
>>>
>>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>>> [   28.027852] IRQ152 no longer affine to CPU0
>>> [   28.028087] IRQ256 no longer affine to CPU0
>>> [   28.028344] CPU0: shutdown
>>> #
>>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>>>
>>> So, it's failed script[1] to test MCPM functionalities
>>>
>>> Thanks.
>>>
>>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
>>>
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Mauro Ribeiro]

Joonyoung,

Abhilash looks to be the father of it...
Also I have no idea why I'm appearing as a commiter of that.. as me or
anyone from Hardkernel doesn't have access to that github account or
contributed to it.
I do have a 3.17 branch here locally on my computer but its not public anywhere.

On Wed, Oct 22, 2014 at 11:36 PM, Joonyoung Shim
<jy0922.shim@samsung.com> wrote:
> On 10/23/2014 10:25 AM, Mauro Ribeiro wrote:
>> Joonyoung,
>>
>> This patch was provided by Samsung and if someone can explain is you
>> guys from Samsung.
>> I have no idea of the logics behind this or what it would take to fix it.
>> Keep in mind that our u-boot and kernel (downstream 3.10) is part of
>> Samsung provided BSP.
>> Would be nice if you guys could explain this.
>>
>
> Mauro,
>
> Thanks to reply, do you know author of this patch written by Samsung?
>
>
> To Kukjin,
>
> The patch[1] seems to come from S.LSI guys, could you contact about this?
>
> [1]: https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>
>>
>>
>> On Wed, Oct 22, 2014 at 11:17 PM, Joonyoung Shim
>> <jy0922.shim@samsung.com> wrote:
>>> Add Cc the patch author and committer from github.
>>>
>>> On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
>>>> Hi,
>>>>
>>>> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>>>>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>>>>> Hi,
>>>>>>
>>>>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>>>>
>>>>>>>
>>>>>>> Hi hdmi_resources_init,
>>>>>>>
>>>>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>>>>
>>>>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>>>>> development on 5420. This was because there was a check in our u-boot
>>>>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>>>>> situation in your case too.
>>>>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>>>>> there is any such code present in your
>>>>>>> secondary bring-up path.
>>>>>>>
>>>>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>>>>> currently using the pre-built binaries that are in the
>>>>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>>>>> needed.
>>>>>>>
>>>>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>>>>> to start looking in that code, that would be helpful.
>>>>>>>
>>>>>>> Kevin
>>>>>>>
>>>>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> It seems to need magic code.
>>>>>>
>>>>>> Please refer
>>>>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>>>
>>>>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>>>>
>>>>> Can someone explain this "magic", ideally in the form of a changelog
>>>>> and patch submitted upstream?
>>>>>
>>>>
>>>> I also really wonder.
>>>>
>>>> One more problem, cpu0 doesn't boot after offline and system is halted.
>>>>
>>>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>>>> [   28.027852] IRQ152 no longer affine to CPU0
>>>> [   28.028087] IRQ256 no longer affine to CPU0
>>>> [   28.028344] CPU0: shutdown
>>>> #
>>>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>>>>
>>>> So, it's failed script[1] to test MCPM functionalities
>>>>
>>>> Thanks.
>>>>
>>>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
>>>>
>>>
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

Hi,

On 10/23/2014 10:42 AM, Mauro Ribeiro wrote:
> Joonyoung,
> 
> Abhilash looks to be the father of it...
> Also I have no idea why I'm appearing as a commiter of that.. as me or
> anyone from Hardkernel doesn't have access to that github account or
> contributed to it.
> I do have a 3.17 branch here locally on my computer but its not public anywhere.

You are just author of the patch on github and it was committed to github by Abhilash.

Thanks.

> 
> On Wed, Oct 22, 2014 at 11:36 PM, Joonyoung Shim
> <jy0922.shim@samsung.com> wrote:
>> On 10/23/2014 10:25 AM, Mauro Ribeiro wrote:
>>> Joonyoung,
>>>
>>> This patch was provided by Samsung and if someone can explain is you
>>> guys from Samsung.
>>> I have no idea of the logics behind this or what it would take to fix it.
>>> Keep in mind that our u-boot and kernel (downstream 3.10) is part of
>>> Samsung provided BSP.
>>> Would be nice if you guys could explain this.
>>>
>>
>> Mauro,
>>
>> Thanks to reply, do you know author of this patch written by Samsung?
>>
>>
>> To Kukjin,
>>
>> The patch[1] seems to come from S.LSI guys, could you contact about this?
>>
>> [1]: https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>
>>>
>>>
>>> On Wed, Oct 22, 2014 at 11:17 PM, Joonyoung Shim
>>> <jy0922.shim@samsung.com> wrote:
>>>> Add Cc the patch author and committer from github.
>>>>
>>>> On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
>>>>> Hi,
>>>>>
>>>>> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>>>>>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>>>>>> Hi,
>>>>>>>
>>>>>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>>>>>
>>>>>>>>
>>>>>>>> Hi hdmi_resources_init,
>>>>>>>>
>>>>>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>>>>>
>>>>>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>>>>>> development on 5420. This was because there was a check in our u-boot
>>>>>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>>>>>> situation in your case too.
>>>>>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>>>>>> there is any such code present in your
>>>>>>>> secondary bring-up path.
>>>>>>>>
>>>>>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>>>>>> currently using the pre-built binaries that are in the
>>>>>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>>>>>> needed.
>>>>>>>>
>>>>>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>>>>>> to start looking in that code, that would be helpful.
>>>>>>>>
>>>>>>>> Kevin
>>>>>>>>
>>>>>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> It seems to need magic code.
>>>>>>>
>>>>>>> Please refer
>>>>>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>>>>
>>>>>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>>>>>
>>>>>> Can someone explain this "magic", ideally in the form of a changelog
>>>>>> and patch submitted upstream?
>>>>>>
>>>>>
>>>>> I also really wonder.
>>>>>
>>>>> One more problem, cpu0 doesn't boot after offline and system is halted.
>>>>>
>>>>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>>>>> [   28.027852] IRQ152 no longer affine to CPU0
>>>>> [   28.028087] IRQ256 no longer affine to CPU0
>>>>> [   28.028344] CPU0: shutdown
>>>>> #
>>>>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>>>>>
>>>>> So, it's failed script[1] to test MCPM functionalities
>>>>>
>>>>> Thanks.
>>>>>
>>>>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
>>>>>
>>>>
>>>> --
>>>> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
>>>> the body of a message to majordomo@vger.kernel.org
>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Joonyoung Shim]

On 10/23/2014 10:49 AM, Joonyoung Shim wrote:
> Hi,
> 
> On 10/23/2014 10:42 AM, Mauro Ribeiro wrote:
>> Joonyoung,
>>
>> Abhilash looks to be the father of it...
>> Also I have no idea why I'm appearing as a commiter of that.. as me or
>> anyone from Hardkernel doesn't have access to that github account or
>> contributed to it.
>> I do have a 3.17 branch here locally on my computer but its not public anywhere.
> 
> You are just author of the patch on github and it was committed to github by Abhilash.

Sorry, not Abhilash, by Abhijeet Dev <abhijeet@abhijeet-dev.net>, i don't know who he is.

> 
> Thanks.
> 
>>
>> On Wed, Oct 22, 2014 at 11:36 PM, Joonyoung Shim
>> <jy0922.shim@samsung.com> wrote:
>>> On 10/23/2014 10:25 AM, Mauro Ribeiro wrote:
>>>> Joonyoung,
>>>>
>>>> This patch was provided by Samsung and if someone can explain is you
>>>> guys from Samsung.
>>>> I have no idea of the logics behind this or what it would take to fix it.
>>>> Keep in mind that our u-boot and kernel (downstream 3.10) is part of
>>>> Samsung provided BSP.
>>>> Would be nice if you guys could explain this.
>>>>
>>>
>>> Mauro,
>>>
>>> Thanks to reply, do you know author of this patch written by Samsung?
>>>
>>>
>>> To Kukjin,
>>>
>>> The patch[1] seems to come from S.LSI guys, could you contact about this?
>>>
>>> [1]: https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>
>>>>
>>>>
>>>> On Wed, Oct 22, 2014 at 11:17 PM, Joonyoung Shim
>>>> <jy0922.shim@samsung.com> wrote:
>>>>> Add Cc the patch author and committer from github.
>>>>>
>>>>> On 10/23/2014 10:04 AM, Joonyoung Shim wrote:
>>>>>> Hi,
>>>>>>
>>>>>> On 10/17/2014 09:06 PM, Kevin Hilman wrote:
>>>>>>> On Wed, Oct 15, 2014 at 4:35 AM, Joonyoung Shim <jy0922.shim@samsung.com> wrote:
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> Kevin Hilman <khilman <at> kernel.org> writes:
>>>>>>>>
>>>>>>>>>
>>>>>>>>> Hi hdmi_resources_init,
>>>>>>>>>
>>>>>>>>> Abhilash Kesavan <a.kesavan <at> samsung.com> writes:
>>>>>>>>>
>>>>>>>>>> I had noticed a similar issue where only 6 cores would come up during
>>>>>>>>>> development on 5420. This was because there was a check in our u-boot
>>>>>>>>>> secondary boot-up path which would put a core into WFI if a flag was
>>>>>>>>>> not present and hence fail to boot. I wonder if it is the same
>>>>>>>>>> situation in your case too.
>>>>>>>>>> Is the u-boot for odroid-xu3 publically available or can you check if
>>>>>>>> there is any such code present in your
>>>>>>>>> secondary bring-up path.
>>>>>>>>>
>>>>>>>>> I'm using the hardkernel u-boot[1], branch: odroidxu3-v2012.07.  I'm
>>>>>>>>> currently using the pre-built binaries that are in the
>>>>>>>>> sd_fuse/hardkernel directory of that branch, but I can rebuild u-boot if
>>>>>>>>> needed.
>>>>>>>>>
>>>>>>>>> I'm not very familiar with u-boot, but if f you have some pointers where
>>>>>>>>> to start looking in that code, that would be helpful.
>>>>>>>>>
>>>>>>>>> Kevin
>>>>>>>>>
>>>>>>>>> [1] https://github.com/hardkernel/u-boot.git
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> It seems to need magic code.
>>>>>>>>
>>>>>>>> Please refer
>>>>>>>> https://github.com/abhijeet-dev/linux-samsung/commit/553b7f40465ab4fad6c270f289cd4780942f7fae
>>>>>>>
>>>>>>> Thanks, Abhilash had pointed me to the same code, and it works.  But...
>>>>>>>
>>>>>>> Can someone explain this "magic", ideally in the form of a changelog
>>>>>>> and patch submitted upstream?
>>>>>>>
>>>>>>
>>>>>> I also really wonder.
>>>>>>
>>>>>> One more problem, cpu0 doesn't boot after offline and system is halted.
>>>>>>
>>>>>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>>>>>> [   28.027852] IRQ152 no longer affine to CPU0
>>>>>> [   28.028087] IRQ256 no longer affine to CPU0
>>>>>> [   28.028344] CPU0: shutdown
>>>>>> #
>>>>>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>>>>>>
>>>>>> So, it's failed script[1] to test MCPM functionalities
>>>>>>
>>>>>> Thanks.
>>>>>>
>>>>>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html
>>>>>>
>>>>>
>>>>> --
>>>>> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
>>>>> the body of a message to majordomo@vger.kernel.org
>>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>>
>>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Abhijeet Dev]

Joonyoung Shim <jy0922.shim <at> samsung.com> writes:
> 
> On 10/23/2014 10:49 AM, Joonyoung Shim wrote:
> > Hi,
> > 
> > On 10/23/2014 10:42 AM, Mauro Ribeiro wrote:
> >> Joonyoung,
> >>
> >> Abhilash looks to be the father of it...
> >> Also I have no idea why I'm appearing as a commiter of that.. as me or
> >> anyone from Hardkernel doesn't have access to that github account or
> >> contributed to it.
> >> I do have a 3.17 branch here locally on my computer but its not public 
anywhere.
> > 
> > You are just author of the patch on github and it was committed to github 
by Abhilash.
> 
> Sorry, not Abhilash, by Abhijeet Dev <abhijeet <at> abhijeet-dev.net>, i 
don't know who he is.
> 
I had received it from Mauro. As Mauro mentioned, Abhilash seems to be the 
father of it - and the patch has to have Samsung origin - one can't dream of 
such magic :)

At the moment we can trace it as far as Abhilash - he looks like the person 
who actually knows this magic:

Abhilash(Samsung) -> Kevin -> Mauro -> Abhijeet (Me)

As I could only trace as far as Mauro earlier, I marked him as the author of 
this patch. I will change it in the next one. As for the tree on my 
github account, I maintain it to support various boards on Kukjin's for-next 
branch on kernel.org.

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Abhilash Kesavan <a.kesavan@samsung.com> writes:

> The change was based on my reading of the platsmp file of the
> Odroid-xu3 3.10 kernel and issues experienced previously. From what I
> understand, the platsmp code in 3.10 kernel is trying to reset the A7s
> and my change just bypasses the check (no magic here ). The change was
> only suggested as a workaround to get 8 cores up using MCPM and not an
> actual fix.  I had explained this in my mail to Kevin. So, if this
> were to be mainlined we would probably need to use something like
> exynos_core_restart in mcpm along with the workaround. But, this is
> all guesswork on my part and someone with knowledge of the 3.10
> kernel/bootloader needs to comment.

Unfortunately, the harkernel tree git repo doesn't help understand where
the change came from because an entire v3.10.y based Samsung BSP is
checked into the tree on top of a v3.8.y kernel in a single commit !!!

Abhilash, in the absence of the original author, could write up a patch
with your change?

Or, if you can help me with a detailed changelog, I can work on the patch.
I don't have any exynos docs, so it would be good to have a description
of what reading/writing to that register are actually doing.

Also, can you explain what you meant by using exynos_core_restart?  I
don't see that function in mainline.

Kevin

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Joonyoung Shim <jy0922.shim@samsung.com> writes:

[...]

>> One more problem, cpu0 doesn't boot after offline and system is halted.
>> 
>> # echo 0 > /sys/devices/system/cpu/cpu0/online
>> [   28.027852] IRQ152 no longer affine to CPU0
>> [   28.028087] IRQ256 no longer affine to CPU0
>> [   28.028344] CPU0: shutdown
>> # 
>> # echo 1 > /sys/devices/system/cpu/cpu0/online
>> 
>> So, it's failed script[1] to test MCPM functionalities
>> 
>> Thanks.
>> 
>> [1]: http://www.spinics.net/lists/arm-kernel/msg330946.html

FYI... this hotplug problem happens with or without the patch to bring
all 8 CPUs online, so it's not related to that fix.

Also, it only happens with the bL switcher enabled, and not with the
switcher disabled.  Since the bL switcher is built on top of hotplug,
I'm not sure if the switcher deals well with CPUs being hotplugged
without it knowing.

Kevin

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Abhilash Kesavan]

Hello Kevin,

On Thu, Oct 23, 2014 at 9:26 PM, Kevin Hilman <khilman@kernel.org> wrote:
> Abhilash Kesavan <a.kesavan@samsung.com> writes:
>
>> The change was based on my reading of the platsmp file of the
>> Odroid-xu3 3.10 kernel and issues experienced previously. From what I
>> understand, the platsmp code in 3.10 kernel is trying to reset the A7s
>> and my change just bypasses the check (no magic here ). The change was
>> only suggested as a workaround to get 8 cores up using MCPM and not an
>> actual fix.  I had explained this in my mail to Kevin. So, if this
>> were to be mainlined we would probably need to use something like
>> exynos_core_restart in mcpm along with the workaround. But, this is
>> all guesswork on my part and someone with knowledge of the 3.10
>> kernel/bootloader needs to comment.
>
> Unfortunately, the harkernel tree git repo doesn't help understand where
> the change came from because an entire v3.10.y based Samsung BSP is
> checked into the tree on top of a v3.8.y kernel in a single commit !!!
>
> Abhilash, in the absence of the original author, could write up a patch
> with your change?
>
> Or, if you can help me with a detailed changelog, I can work on the patch.
> I don't have any exynos docs, so it would be good to have a description
> of what reading/writing to that register are actually doing.

I could post a patch with an explanation of how I think the workaround
works, but there are a few issues:
- I am not completely confident that this is the best/right change.
- We are adding code to the exynos mcpm back-end to work-around a
bootloader setting for Odroid-xu3. We would have to make it specific
to odroid somehow as we do not require it for 5420/5800 chromebooks.
- This might solve the 8 core boot-up issue but other scenarios might
still fail like the one Joonyoung has already reported.

>
> Also, can you explain what you meant by using exynos_core_restart?  I
> don't see that function in mainline.

It is present in linux-next and kgene's for-next branch as part of
this commit "ARM: EXYNOS: SWRESET is needed to boot secondary CPU on
exynos3250".


Regards,
Abhilash

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Abhilash Kesavan <kesavan.abhilash@gmail.com> writes:

> On Thu, Oct 23, 2014 at 9:26 PM, Kevin Hilman <khilman@kernel.org> wrote:
>> Abhilash Kesavan <a.kesavan@samsung.com> writes:
>>
>>> The change was based on my reading of the platsmp file of the
>>> Odroid-xu3 3.10 kernel and issues experienced previously. From what I
>>> understand, the platsmp code in 3.10 kernel is trying to reset the A7s
>>> and my change just bypasses the check (no magic here ). The change was
>>> only suggested as a workaround to get 8 cores up using MCPM and not an
>>> actual fix.  I had explained this in my mail to Kevin. So, if this
>>> were to be mainlined we would probably need to use something like
>>> exynos_core_restart in mcpm along with the workaround. But, this is
>>> all guesswork on my part and someone with knowledge of the 3.10
>>> kernel/bootloader needs to comment.
>>
>> Unfortunately, the harkernel tree git repo doesn't help understand where
>> the change came from because an entire v3.10.y based Samsung BSP is
>> checked into the tree on top of a v3.8.y kernel in a single commit !!!
>>
>> Abhilash, in the absence of the original author, could write up a patch
>> with your change?
>>
>> Or, if you can help me with a detailed changelog, I can work on the patch.
>> I don't have any exynos docs, so it would be good to have a description
>> of what reading/writing to that register are actually doing.
>
> I could post a patch with an explanation of how I think the workaround
> works, but there are a few issues:
> - I am not completely confident that this is the best/right change.

Maybe post as RFC.

> - We are adding code to the exynos mcpm back-end to work-around a
> bootloader setting for Odroid-xu3. We would have to make it specific
> to odroid somehow as we do not require it for 5420/5800 chromebooks.

Hmm, if you have an idea how/where to fix it in u-boot, I'd be glad to
give that a try.  I'm using the hardkernel u-boot repo:
https://github.com/hardkernel/u-boot.git, branch odroidxu3-v2012.07, and
would be happy to test any fixes there.

> - This might solve the 8 core boot-up issue but other scenarios might
> still fail like the one Joonyoung has already reported.

I replied yesterday to that report, and the one Joonyoung reported
exists with or without your proposed fix, so I don't think they're
directly related.

>>
>> Also, can you explain what you meant by using exynos_core_restart?  I
>> don't see that function in mainline.
>
> It is present in linux-next and kgene's for-next branch as part of
> this commit "ARM: EXYNOS: SWRESET is needed to boot secondary CPU on
> exynos3250".

OK, I see it now.

Kevin

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Mauro Ribeiro]

Kevin,

As I wrote you previously outside the list this is the commit "ARM:
EXYNOS: SWRESET is needed to boot secondary CPU on exynos3250" that I
was expecting it to fix XU3 as well and apparently it is..

Next week if everything goes as I plan I'll check if I can use the
same code to enable the A7's on XU3 as well.
If its possible to use the same code I'll try to upstream this next week.

It would be nice if Samsung could tell what in our u-boot (provided by
LSI) is needed to be changed as I can just push any changes to our
u-boot.

BR,
Mauro


On Fri, Oct 24, 2014 at 2:00 PM, Kevin Hilman <khilman@kernel.org> wrote:
> Abhilash Kesavan <kesavan.abhilash@gmail.com> writes:
>
>> On Thu, Oct 23, 2014 at 9:26 PM, Kevin Hilman <khilman@kernel.org> wrote:
>>> Abhilash Kesavan <a.kesavan@samsung.com> writes:
>>>
>>>> The change was based on my reading of the platsmp file of the
>>>> Odroid-xu3 3.10 kernel and issues experienced previously. From what I
>>>> understand, the platsmp code in 3.10 kernel is trying to reset the A7s
>>>> and my change just bypasses the check (no magic here ). The change was
>>>> only suggested as a workaround to get 8 cores up using MCPM and not an
>>>> actual fix.  I had explained this in my mail to Kevin. So, if this
>>>> were to be mainlined we would probably need to use something like
>>>> exynos_core_restart in mcpm along with the workaround. But, this is
>>>> all guesswork on my part and someone with knowledge of the 3.10
>>>> kernel/bootloader needs to comment.
>>>
>>> Unfortunately, the harkernel tree git repo doesn't help understand where
>>> the change came from because an entire v3.10.y based Samsung BSP is
>>> checked into the tree on top of a v3.8.y kernel in a single commit !!!
>>>
>>> Abhilash, in the absence of the original author, could write up a patch
>>> with your change?
>>>
>>> Or, if you can help me with a detailed changelog, I can work on the patch.
>>> I don't have any exynos docs, so it would be good to have a description
>>> of what reading/writing to that register are actually doing.
>>
>> I could post a patch with an explanation of how I think the workaround
>> works, but there are a few issues:
>> - I am not completely confident that this is the best/right change.
>
> Maybe post as RFC.
>
>> - We are adding code to the exynos mcpm back-end to work-around a
>> bootloader setting for Odroid-xu3. We would have to make it specific
>> to odroid somehow as we do not require it for 5420/5800 chromebooks.
>
> Hmm, if you have an idea how/where to fix it in u-boot, I'd be glad to
> give that a try.  I'm using the hardkernel u-boot repo:
> https://github.com/hardkernel/u-boot.git, branch odroidxu3-v2012.07, and
> would be happy to test any fixes there.
>
>> - This might solve the 8 core boot-up issue but other scenarios might
>> still fail like the one Joonyoung has already reported.
>
> I replied yesterday to that report, and the one Joonyoung reported
> exists with or without your proposed fix, so I don't think they're
> directly related.
>
>>>
>>> Also, can you explain what you meant by using exynos_core_restart?  I
>>> don't see that function in mainline.
>>
>> It is present in linux-next and kgene's for-next branch as part of
>> this commit "ARM: EXYNOS: SWRESET is needed to boot secondary CPU on
>> exynos3250".
>
> OK, I see it now.
>
> Kevin

Re: exynos5422-odroid-xu3: MCPM: only 6 of 8 CPUs coming online

[Kevin Hilman]

Mauro Ribeiro <mdrjr0@gmail.com> writes:

> As I wrote you previously outside the list this is the commit "ARM:
> EXYNOS: SWRESET is needed to boot secondary CPU on exynos3250" that I
> was expecting it to fix XU3 as well and apparently it is..

I'm not sure about that, since I haven't been able to get that to work.

> Next week if everything goes as I plan I'll check if I can use the
> same code to enable the A7's on XU3 as well.
> If its possible to use the same code I'll try to upstream this next week.

Excellent.  I'd love to see a patch against mainline that using that
function that works on the XU3.  Note that mainline is using MCPM and
that patch solves the problem for the older platsmp method.

> It would be nice if Samsung could tell what in our u-boot (provided by
> LSI) is needed to be changed as I can just push any changes to our
> u-boot.

Yes, I completely agree.  This seems to me like a bootloader bug and the
best solution here would be to have a bootloader fix.

Kevin