Re: [discuss] Re: BUG in x86_64 hugepage support

[Nishanth Aravamudan <nacc@us.ibm.com>]

On 15.03.2006 [02:03:00 -0800], Chen, Kenneth W wrote:
> Nishanth Aravamudan wrote on Tuesday, March 14, 2006 11:31 PM
> > Description: We currently fail mprotect testing in libhugetlbfs
> > because the PSE bit in the hugepage PTEs gets unset. In the case
> > where we know that a filled hugetlb PTE is going to have its
> > protection changed, make sure it stays a hugetlb PTE by setting the
> > PSE bit in the new protection flags.
> 
> Jan Beulich wrote on Wednesday, March 15, 2006 12:50 AM
> > This is architecture independent code - you shouldn't be using
> > _PAGE_PSE here. Probably x86-64 (and then likely also i386) should
> > define their own set_huge_pte_at(), and use that# to or in the
> > needed flag?
> 
> 
> Yeah, that will do.  i386, x86_64 should also clean up pte_mkhuge()
> macro.  The unconditional setting of _PAGE_PRESENT bit was a leftover
> stuff from the good'old day of pre-faulting hugetlb page.  
> 
> 
> 
> [patch] fix i386/x86-64 _PAGE_PSE bit when changing page protection
> 
> On i386 and x86-64, pte flag _PAGE_PSE collides with _PAGE_PROTNONE.
> The identify of hugetlb pte is lost when changing page protection via
> mprotect. A page fault occurs later will trigger a bug check in
> huge_pte_alloc().
> 
> The fix is to always make new pte a hugetlb pte and also to clean up
> legacy code where _PAGE_PRESENT is forced on in the pre-faulting day.
> 
> 
> Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>

I can confirm this fixes the BUGs I was seeing on x86_64 testing of
libhugetlbfs' mprotect support.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>

Thanks,
Nish

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>