[PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE

Max Filippov posted 1 patch 4 years ago
Documentation/features/debug/debug-vm-pgtable/arch-support.txt | 2 +-
arch/xtensa/Kconfig                                            | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
[PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE
Posted by Max Filippov 4 years ago
xtensa kernels successfully build and run with
CONFIG_DEBUG_VM_PGTABLE=y, enable arch support for it.

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
---
 Documentation/features/debug/debug-vm-pgtable/arch-support.txt | 2 +-
 arch/xtensa/Kconfig                                            | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
index 83eafe1a7f68..ff21a83abe62 100644
--- a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
+++ b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
@@ -27,5 +27,5 @@
     |       sparc: | TODO |
     |          um: | TODO |
     |         x86: |  ok  |
-    |      xtensa: | TODO |
+    |      xtensa: |  ok  |
     -----------------------
diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index 2ed897299f19..903b910ebc10 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -4,6 +4,7 @@ config XTENSA
 	select ARCH_32BIT_OFF_T
 	select ARCH_HAS_BINFMT_FLAT if !MMU
 	select ARCH_HAS_CURRENT_STACK_POINTER
+	select ARCH_HAS_DEBUG_VM_PGTABLE
 	select ARCH_HAS_DMA_PREP_COHERENT if MMU
 	select ARCH_HAS_SYNC_DMA_FOR_CPU if MMU
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if MMU
-- 
2.30.2
Re: [PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE
Posted by Anshuman Khandual 4 years ago

On 4/25/22 23:35, Max Filippov wrote:
> xtensa kernels successfully build and run with
> CONFIG_DEBUG_VM_PGTABLE=y, enable arch support for it.
> 
> Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>

Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>

> ---
>  Documentation/features/debug/debug-vm-pgtable/arch-support.txt | 2 +-
>  arch/xtensa/Kconfig                                            | 1 +
>  2 files changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> index 83eafe1a7f68..ff21a83abe62 100644
> --- a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> +++ b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> @@ -27,5 +27,5 @@
>      |       sparc: | TODO |
>      |          um: | TODO |
>      |         x86: |  ok  |
> -    |      xtensa: | TODO |
> +    |      xtensa: |  ok  |
>      -----------------------
> diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
> index 2ed897299f19..903b910ebc10 100644
> --- a/arch/xtensa/Kconfig
> +++ b/arch/xtensa/Kconfig
> @@ -4,6 +4,7 @@ config XTENSA
>  	select ARCH_32BIT_OFF_T
>  	select ARCH_HAS_BINFMT_FLAT if !MMU
>  	select ARCH_HAS_CURRENT_STACK_POINTER
> +	select ARCH_HAS_DEBUG_VM_PGTABLE
>  	select ARCH_HAS_DMA_PREP_COHERENT if MMU
>  	select ARCH_HAS_SYNC_DMA_FOR_CPU if MMU
>  	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if MMU
Re: [PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE
Posted by Anshuman Khandual 4 years ago
Hi Max,

Thanks for the patch. Just wondering if you ran with different config
options (which might impact test case selection) on/off during testing ?

- Anshuman

On 4/25/22 23:35, Max Filippov wrote:
> xtensa kernels successfully build and run with
> CONFIG_DEBUG_VM_PGTABLE=y, enable arch support for it.
> 
> Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
> ---
>  Documentation/features/debug/debug-vm-pgtable/arch-support.txt | 2 +-
>  arch/xtensa/Kconfig                                            | 1 +
>  2 files changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> index 83eafe1a7f68..ff21a83abe62 100644
> --- a/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> +++ b/Documentation/features/debug/debug-vm-pgtable/arch-support.txt
> @@ -27,5 +27,5 @@
>      |       sparc: | TODO |
>      |          um: | TODO |
>      |         x86: |  ok  |
> -    |      xtensa: | TODO |
> +    |      xtensa: |  ok  |
>      -----------------------
> diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
> index 2ed897299f19..903b910ebc10 100644
> --- a/arch/xtensa/Kconfig
> +++ b/arch/xtensa/Kconfig
> @@ -4,6 +4,7 @@ config XTENSA
>  	select ARCH_32BIT_OFF_T
>  	select ARCH_HAS_BINFMT_FLAT if !MMU
>  	select ARCH_HAS_CURRENT_STACK_POINTER
> +	select ARCH_HAS_DEBUG_VM_PGTABLE
>  	select ARCH_HAS_DMA_PREP_COHERENT if MMU
>  	select ARCH_HAS_SYNC_DMA_FOR_CPU if MMU
>  	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if MMU
Re: [PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE
Posted by Max Filippov 4 years ago
Anshuman,

On Mon, Apr 25, 2022 at 11:07 PM Anshuman Khandual
<anshuman.khandual@arm.com> wrote:
> Thanks for the patch. Just wondering if you ran with different config
> options (which might impact test case selection) on/off during testing ?

I went through the code in the mm/debug_vm_pgtable.c and saw that most
variations are related to the huge page support which xtensa doesn't have.
I've tested it with and without CMA.
I haven't found anything else interesting that I could tweak -- have I missed
anything?

-- 
Thanks.
-- Max
Re: [PATCH] xtensa: enable ARCH_HAS_DEBUG_VM_PGTABLE
Posted by Anshuman Khandual 4 years ago

On 4/26/22 18:12, Max Filippov wrote:
> Anshuman,
> 
> On Mon, Apr 25, 2022 at 11:07 PM Anshuman Khandual
> <anshuman.khandual@arm.com> wrote:
>> Thanks for the patch. Just wondering if you ran with different config
>> options (which might impact test case selection) on/off during testing ?
> 
> I went through the code in the mm/debug_vm_pgtable.c and saw that most
> variations are related to the huge page support which xtensa doesn't have.
> I've tested it with and without CMA.
> I haven't found anything else interesting that I could tweak -- have I missed
> anything?

I dont think so, but thanks for confirming.