Re: [FSL P50x0] Xorg always restarts again and again after the the PowerPC updates 5.13-1

[Christophe Leroy <christophe.leroy@csgroup.eu>]

Le 09/05/2021 à 14:27, Christian Zigotzky a écrit :
> On 08 May 2021 at 06:39pm, Christian Zigotzky wrote:
>> On 06 May 2021 at 03:58 pm, Christian Zigotzky wrote:
>>> I have started bisecting again.
>>>
>>> Link: https://forum.hyperion-entertainment.com/viewtopic.php?p=53106#p53106 
>>> <https://forum.hyperion-entertainment.com/viewtopic.php?p=53106#p53106>
>>>
>>>
>>>> On 6. May 2021, at 10:09, Christophe Leroy <christophe.leroy@csgroup.eu> wrote:
>>>>
>>>>
>>>> - Can you check that 887f3ceb51cd with cherry-picked 525642624783 has Xorg working ?
>> git checkout 887f3ceb51cd
>> git cherry-pick 525642624783
>>
>> Result: Xorg works.
>>>> - Can you bisect between 887f3ceb51cd[good] and 56bec2f9d4d0[bad] to identify first bad commit 
>>>> that stops after loading the dtb and uImage ?
>>>> - Once that first bad commit is identified, can you check whether the preceeding commit with 
>>>> cherry-picked 525642624783 has Xorg working or not ?
>>>>
>>>> Thanks
>>>> Christophe
>> git bisect start
>> git bisect good 887f3ceb51cd
>> git bisect bad 56bec2f9d4d0
>> git bisect good -- Xorg restarts again and again but we are looking for the first bad commit that 
>> stops the boot after loading the dtb and uImage.
>> git bisect good -- Xorg restarts again and again.
>> git bisect good -- Xorg restarts again and again.
>> git bisect good -- Xorg restarts again and again.
>>
>> Result:
>>
>> 56bec2f9d4d05675cada96772a8a93010f4d82bf is the first bad commit
>> commit 56bec2f9d4d05675cada96772a8a93010f4d82bf
>> Author: Michael Ellerman <mpe@ellerman.id.au>
>> Date:   Wed Mar 31 11:38:40 2021 +1100
>>
>>     powerpc/mm/64s: Add _PAGE_KERNEL_ROX
>>
>>     In the past we had a fallback definition for _PAGE_KERNEL_ROX, but we
>>     removed that in commit d82fd29c5a8c ("powerpc/mm: Distribute platform
>>     specific PAGE and PMD flags and definitions") and added definitions
>>     for each MMU family.
>>
>>     However we missed adding a definition for 64s, which was not really a
>>     bug because it's currently not used.
>>
>>     But we'd like to use PAGE_KERNEL_ROX in a future patch so add a
>>     definition now.
>>
>>     Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
>>     Link: https://lore.kernel.org/r/20210331003845.216246-1-mpe@ellerman.id.au
>>
>> :040000 040000 ff8171830c08e4f99852947a5c3b62e784220a26 85aff144e5219bce4eb6adb2ac32c6459cac22d0 
>> M    arch
>>
>> ---
>>
>> git cherry-pick 525642624783
>>
>> Output:
>>
>> powerpc/signal32: Fix erroneous SIGSEGV on RT signal return
>>  Author: Christophe Leroy <christophe.leroy@csgroup.eu>
>>  Date: Fri Apr 23 13:52:10 2021 +0000
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> ---
>>
>> Xorg works after compiling with the cherry-pick of 525642624783.
> Hi All,
> 
> I compiled and tested the latest git kernel with the new PowerPC updates 5.13-2 today. Unfortunately 
> the Xorg issue still exists.
> If I revert the PowerPC updates 5.13-1 and 5.13-2 then Xorg works without any problems.
> 
> Please check the BookE changes in the PowerPC updates 5.13-1 because my Book3S machines aren't 
> affected by this issue.
> 

On my side, book3e (corenet64_smp_defconfig) built with GCC 10.1 works well with QEMU 2.11.2

A kernel built with the configuration you provided doesn't boot on QEMU, no output at all, even with 
kernel v5.12.

What versions of GCC and QEMU are you using ?

Thanks
Christophe