[PATCH] arm64: fault: Print info about page table structure when dumping pte

[PATCH] arm64: fault: Print info about page table structure when dumping pte

[Will Deacon]

Whilst debugging a remote crash, I noticed that show_pte is unhelpful
when it comes to describing the structure of the page table being walked.
This is easily fixed by printing out the page table (swapper vs user),
page size and virtual address size when displaying the PGD address.

Signed-off-by: Will Deacon <will.deacon@arm.com>
---
 arch/arm64/mm/fault.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
index 37b95dff0b07..ee6085a922b6 100644
--- a/arch/arm64/mm/fault.c
+++ b/arch/arm64/mm/fault.c
@@ -89,7 +89,9 @@ void show_pte(struct mm_struct *mm, unsigned long addr)
 	if (!mm)
 		mm = &init_mm;
 
-	pr_alert("pgd = %p\n", mm->pgd);
+	pr_alert("%s pgtable: %luk pages, %u-bit VAs, pgd = %p\n",
+		 mm == &init_mm ? "swapper" : "user", PAGE_SIZE / SZ_1K,
+		 VA_BITS, mm->pgd);
 	pgd = pgd_offset(mm, addr);
 	pr_alert("[%08lx] *pgd=%016llx", addr, pgd_val(*pgd));
 
-- 
2.1.4

[PATCH] arm64: fault: Print info about page table structure when dumping pte

[Mark Rutland]

On Tue, May 30, 2017 at 11:10:47AM +0100, Will Deacon wrote:
> Whilst debugging a remote crash, I noticed that show_pte is unhelpful
> when it comes to describing the structure of the page table being walked.
> This is easily fixed by printing out the page table (swapper vs user),
> page size and virtual address size when displaying the PGD address.
> 
> Signed-off-by: Will Deacon <will.deacon@arm.com>
> ---
>  arch/arm64/mm/fault.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
> index 37b95dff0b07..ee6085a922b6 100644
> --- a/arch/arm64/mm/fault.c
> +++ b/arch/arm64/mm/fault.c
> @@ -89,7 +89,9 @@ void show_pte(struct mm_struct *mm, unsigned long addr)
>  	if (!mm)
>  		mm = &init_mm;
>  
> -	pr_alert("pgd = %p\n", mm->pgd);
> +	pr_alert("%s pgtable: %luk pages, %u-bit VAs, pgd = %p\n",
> +		 mm == &init_mm ? "swapper" : "user", PAGE_SIZE / SZ_1K,
> +		 VA_BITS, mm->pgd);

I take it this will go with Kristina's fix [1] to dump the correct page
table?

Otherwise, this would be more misleading than presently.

Assuming that is the case, this looks very useful to me. FWIW:

Acked-by: Mark Rutland <mark.rutland@arm.com>

Mark.

>  	pgd = pgd_offset(mm, addr);
>  	pr_alert("[%08lx] *pgd=%016llx", addr, pgd_val(*pgd));
>  
> -- 
> 2.1.4
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH] arm64: fault: Print info about page table structure when dumping pte

[Will Deacon]

On Tue, May 30, 2017 at 12:16:39PM +0100, Mark Rutland wrote:
> On Tue, May 30, 2017 at 11:10:47AM +0100, Will Deacon wrote:
> > Whilst debugging a remote crash, I noticed that show_pte is unhelpful
> > when it comes to describing the structure of the page table being walked.
> > This is easily fixed by printing out the page table (swapper vs user),
> > page size and virtual address size when displaying the PGD address.
> > 
> > Signed-off-by: Will Deacon <will.deacon@arm.com>
> > ---
> >  arch/arm64/mm/fault.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
> > index 37b95dff0b07..ee6085a922b6 100644
> > --- a/arch/arm64/mm/fault.c
> > +++ b/arch/arm64/mm/fault.c
> > @@ -89,7 +89,9 @@ void show_pte(struct mm_struct *mm, unsigned long addr)
> >  	if (!mm)
> >  		mm = &init_mm;
> >  
> > -	pr_alert("pgd = %p\n", mm->pgd);
> > +	pr_alert("%s pgtable: %luk pages, %u-bit VAs, pgd = %p\n",
> > +		 mm == &init_mm ? "swapper" : "user", PAGE_SIZE / SZ_1K,
> > +		 VA_BITS, mm->pgd);
> 
> I take it this will go with Kristina's fix [1] to dump the correct page
> table?

Yup, just waiting for her to respin those patches and I'll apply this one on
top.

> Otherwise, this would be more misleading than presently.
> 
> Assuming that is the case, this looks very useful to me. FWIW:
> 
> Acked-by: Mark Rutland <mark.rutland@arm.com>

Thanks,

Will