Normally __swp_entry_to_pte() is never called with a value translating
to a valid PTE. The only known exception is pte_swap_tests(), resulting
in a WARN splat in Xen PV guests, as __pte_to_swp_entry() did
translate the PFN of the valid PTE to a guest local PFN, while
__swp_entry_to_pte() doesn't do the opposite translation.
Fix that by using __pte() in __swp_entry_to_pte() instead of open
coding the native variant of it.
For correctness do the similar conversion for __swp_entry_to_pmd().
Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
Signed-off-by: Juergen Gross <[email protected]>
---
arch/x86/include/asm/pgtable_64.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
index 7929327abe00..a629b1b9f65a 100644
--- a/arch/x86/include/asm/pgtable_64.h
+++ b/arch/x86/include/asm/pgtable_64.h
@@ -237,8 +237,8 @@ static inline void native_pgd_clear(pgd_t *pgd)
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val((pte)) })
#define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val((pmd)) })
-#define __swp_entry_to_pte(x) ((pte_t) { .pte = (x).val })
-#define __swp_entry_to_pmd(x) ((pmd_t) { .pmd = (x).val })
+#define __swp_entry_to_pte(x) (__pte((x).val))
+#define __swp_entry_to_pmd(x) (__pmd((x).val))
extern void cleanup_highmap(void);
--
2.35.3
On 3/6/23 04:32, Juergen Gross wrote:
> Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
> Signed-off-by: Juergen Gross <[email protected]>
This looks sane. Any reason this shouldn't be cc:stable@?
On 06.03.23 15:08, Dave Hansen wrote:
> On 3/6/23 04:32, Juergen Gross wrote:
>> Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
>> Signed-off-by: Juergen Gross <[email protected]>
>
> This looks sane. Any reason this shouldn't be cc:stable@?
It wasn't missed for several years now, and production kernels will
probably not have CONFIG_DEBUG_VM_PGTABLE set.
In the end I don't mind either way. It was just a bug I stumbled over
during development of my MTRR series (someone mentioned a problem and
I enabled said config option).
Juergen
On 06.03.23 13:32, Juergen Gross wrote:
> Normally __swp_entry_to_pte() is never called with a value translating
> to a valid PTE. The only known exception is pte_swap_tests(), resulting
> in a WARN splat in Xen PV guests, as __pte_to_swp_entry() did
> translate the PFN of the valid PTE to a guest local PFN, while
> __swp_entry_to_pte() doesn't do the opposite translation.
>
> Fix that by using __pte() in __swp_entry_to_pte() instead of open
> coding the native variant of it.
>
> For correctness do the similar conversion for __swp_entry_to_pmd().
>
> Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
> Signed-off-by: Juergen Gross <[email protected]>
Any further comments (apart from "It looks sane")?
Juergen
Ping?
On 06.03.23 13:32, Juergen Gross wrote:
> Normally __swp_entry_to_pte() is never called with a value translating
> to a valid PTE. The only known exception is pte_swap_tests(), resulting
> in a WARN splat in Xen PV guests, as __pte_to_swp_entry() did
> translate the PFN of the valid PTE to a guest local PFN, while
> __swp_entry_to_pte() doesn't do the opposite translation.
>
> Fix that by using __pte() in __swp_entry_to_pte() instead of open
> coding the native variant of it.
>
> For correctness do the similar conversion for __swp_entry_to_pmd().
>
> Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
> Signed-off-by: Juergen Gross <[email protected]>
> ---
> arch/x86/include/asm/pgtable_64.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
> index 7929327abe00..a629b1b9f65a 100644
> --- a/arch/x86/include/asm/pgtable_64.h
> +++ b/arch/x86/include/asm/pgtable_64.h
> @@ -237,8 +237,8 @@ static inline void native_pgd_clear(pgd_t *pgd)
>
> #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val((pte)) })
> #define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val((pmd)) })
> -#define __swp_entry_to_pte(x) ((pte_t) { .pte = (x).val })
> -#define __swp_entry_to_pmd(x) ((pmd_t) { .pmd = (x).val })
> +#define __swp_entry_to_pte(x) (__pte((x).val))
> +#define __swp_entry_to_pmd(x) (__pmd((x).val))
>
> extern void cleanup_highmap(void);
>
PING (3rd one, after 2 months of getting no feedback)!
I'm inclined to carry it just via the Xen tree if x86 maintainers continue
to ignore it ...
Juergen
On 06.03.23 13:32, Juergen Gross wrote:
> Normally __swp_entry_to_pte() is never called with a value translating
> to a valid PTE. The only known exception is pte_swap_tests(), resulting
> in a WARN splat in Xen PV guests, as __pte_to_swp_entry() did
> translate the PFN of the valid PTE to a guest local PFN, while
> __swp_entry_to_pte() doesn't do the opposite translation.
>
> Fix that by using __pte() in __swp_entry_to_pte() instead of open
> coding the native variant of it.
>
> For correctness do the similar conversion for __swp_entry_to_pmd().
>
> Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
> Signed-off-by: Juergen Gross <[email protected]>
> ---
> arch/x86/include/asm/pgtable_64.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
> index 7929327abe00..a629b1b9f65a 100644
> --- a/arch/x86/include/asm/pgtable_64.h
> +++ b/arch/x86/include/asm/pgtable_64.h
> @@ -237,8 +237,8 @@ static inline void native_pgd_clear(pgd_t *pgd)
>
> #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val((pte)) })
> #define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val((pmd)) })
> -#define __swp_entry_to_pte(x) ((pte_t) { .pte = (x).val })
> -#define __swp_entry_to_pmd(x) ((pmd_t) { .pmd = (x).val })
> +#define __swp_entry_to_pte(x) (__pte((x).val))
> +#define __swp_entry_to_pmd(x) (__pmd((x).val))
>
> extern void cleanup_highmap(void);
>
The following commit has been merged into the x86/mm branch of tip:
Commit-ID: 0f88130e8a6fd185b0aeb5d8e286083735f2585a
Gitweb: https://git.kernel.org/tip/0f88130e8a6fd185b0aeb5d8e286083735f2585a
Author: Juergen Gross <[email protected]>
AuthorDate: Mon, 06 Mar 2023 13:32:59 +01:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 08 May 2023 15:25:24 +02:00
x86/mm: Fix __swp_entry_to_pte() for Xen PV guests
Normally __swp_entry_to_pte() is never called with a value translating
to a valid PTE. The only known exception is pte_swap_tests(), resulting
in a WARN splat in Xen PV guests, as __pte_to_swp_entry() did
translate the PFN of the valid PTE to a guest local PFN, while
__swp_entry_to_pte() doesn't do the opposite translation.
Fix that by using __pte() in __swp_entry_to_pte() instead of open
coding the native variant of it.
For correctness do the similar conversion for __swp_entry_to_pmd().
Fixes: 05289402d717 ("mm/debug_vm_pgtable: add tests validating arch helpers for core MM features")
Signed-off-by: Juergen Gross <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/include/asm/pgtable_64.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
index 7929327..a629b1b 100644
--- a/arch/x86/include/asm/pgtable_64.h
+++ b/arch/x86/include/asm/pgtable_64.h
@@ -237,8 +237,8 @@ static inline void native_pgd_clear(pgd_t *pgd)
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val((pte)) })
#define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val((pmd)) })
-#define __swp_entry_to_pte(x) ((pte_t) { .pte = (x).val })
-#define __swp_entry_to_pmd(x) ((pmd_t) { .pmd = (x).val })
+#define __swp_entry_to_pte(x) (__pte((x).val))
+#define __swp_entry_to_pmd(x) (__pmd((x).val))
extern void cleanup_highmap(void);