LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH] i386/PAE: fix pud_page()
@ 2008-10-30 10:37 Jan Beulich
  2008-10-30 10:49 ` Ingo Molnar
  0 siblings, 1 reply; 2+ messages in thread
From: Jan Beulich @ 2008-10-30 10:37 UTC (permalink / raw)
  To: mingo, tglx, hpa; +Cc: linux-kernel

To the unsuspecting user it is quite annoying that this broken and
inconsistent with x86-64 definition still exists.

Signed-off-by: Jan Beulich <jbeulich@novell.com>

---
 arch/x86/include/asm/pgtable-3level.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- linux-2.6.28-rc2/arch/x86/include/asm/pgtable-3level.h	2008-10-30 11:03:18.000000000 +0100
+++ 2.6.28-rc2-i386-pae-pud_page/arch/x86/include/asm/pgtable-3level.h	2008-10-29 11:35:24.000000000 +0100
@@ -120,13 +120,13 @@ static inline void pud_clear(pud_t *pudp
 		write_cr3(pgd);
 }
 
-#define pud_page(pud) ((struct page *) __va(pud_val(pud) & PTE_PFN_MASK))
+#define pud_page(pud) pfn_to_page(pud_val(pud) >> PAGE_SHIFT)
 
 #define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PTE_PFN_MASK))
 
 
 /* Find an entry in the second-level page table.. */
-#define pmd_offset(pud, address) ((pmd_t *)pud_page(*(pud)) +	\
+#define pmd_offset(pud, address) ((pmd_t *)pud_page_vaddr(*(pud)) +	\
 				  pmd_index(address))
 
 #ifdef CONFIG_SMP




^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] i386/PAE: fix pud_page()
  2008-10-30 10:37 [PATCH] i386/PAE: fix pud_page() Jan Beulich
@ 2008-10-30 10:49 ` Ingo Molnar
  0 siblings, 0 replies; 2+ messages in thread
From: Ingo Molnar @ 2008-10-30 10:49 UTC (permalink / raw)
  To: Jan Beulich; +Cc: tglx, hpa, linux-kernel, Jeremy Fitzhardinge


* Jan Beulich <jbeulich@novell.com> wrote:

> To the unsuspecting user it is quite annoying that this broken and
> inconsistent with x86-64 definition still exists.
> 
> Signed-off-by: Jan Beulich <jbeulich@novell.com>
> 
> ---
>  arch/x86/include/asm/pgtable-3level.h |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

applied, thanks Jan!

> -#define pud_page(pud) ((struct page *) __va(pud_val(pud) & PTE_PFN_MASK))
> +#define pud_page(pud) pfn_to_page(pud_val(pud) >> PAGE_SHIFT)
>  
>  #define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PTE_PFN_MASK))
>  
>  
>  /* Find an entry in the second-level page table.. */
> -#define pmd_offset(pud, address) ((pmd_t *)pud_page(*(pud)) +	\
> +#define pmd_offset(pud, address) ((pmd_t *)pud_page_vaddr(*(pud)) +	\
>  				  pmd_index(address))

we were quite lucky there. Despite being a cleanup i've put this into 
x86/urgent - i agree that it's just too ugly and confusing.

	Ingo

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2008-10-30 10:49 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-10-30 10:37 [PATCH] i386/PAE: fix pud_page() Jan Beulich
2008-10-30 10:49 ` Ingo Molnar

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).