Re: [PATCH v4] mm: pmd dirty emulation in page fault handler

[Michal Hocko]

On Sat 24-12-16 00:14:58, Minchan Kim wrote:
> Andreas reported [1] made a test in jemalloc hang in THP mode in arm64.
> http://lkml.kernel.org/r/mvmmvfy37g1.fsf@xxxxxxxxxxxxxxx
> 
> The problem is currently page fault handler doesn't supports dirty bit
> emulation of pmd for non-HW dirty-bit architecture so that application
> stucks until VM marked the pmd dirty.
> 
> How the emulation work depends on the architecture. In case of arm64,
> when it set up pte firstly, it sets pte PTE_RDONLY to get a chance to
> mark the pte dirty via triggering page fault when store access happens.
> Once the page fault occurs, VM marks the pmd dirty and arch code for
> setting pmd will clear PTE_RDONLY for application to proceed.
> 
> IOW, if VM doesn't mark the pmd dirty, application hangs forever by
> repeated fault(i.e., store op but the pmd is PTE_RDONLY).
> 
> This patch enables pmd dirty-bit emulation for those architectures.

Thanks for extending the patch description again!

> [1] b8d3c4c3009d, mm/huge_memory.c: don't split THP page when MADV_FREE syscall is called
> 
> Cc: Jason Evans <je@xxxxxx>
> Cc: Michal Hocko <mhocko@xxxxxxxx> 
> Cc: Will Deacon <will.deacon@xxxxxxx>
> Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
> Cc: linux-arch@xxxxxxxxxxxxxxx
> Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
> Cc: <stable@xxxxxxxxxxxxxxx> [4.5+]
> Fixes: b8d3c4c3009d ("mm/huge_memory.c: don't split THP page when MADV_FREE syscall is called")
> Reported-and-Tested-by: Andreas Schwab <schwab@xxxxxxx>
> Acked-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
> Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>

Acked-by: Michal Hocko <mhocko@xxxxxxxx>

> ---
>   Merry Xmas!
> 
> * from v3
>   * Elaborate description
> * from v2
>   * Add acked-by/tested-by
> * from v1
>   * Remove __handle_mm_fault part - Kirill
>  mm/huge_memory.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 10eedbf..29ec8a4 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -883,15 +883,17 @@ void huge_pmd_set_accessed(struct vm_fault *vmf, pmd_t orig_pmd)
>  {
>  	pmd_t entry;
>  	unsigned long haddr;
> +	bool write = vmf->flags & FAULT_FLAG_WRITE;
>  
>  	vmf->ptl = pmd_lock(vmf->vma->vm_mm, vmf->pmd);
>  	if (unlikely(!pmd_same(*vmf->pmd, orig_pmd)))
>  		goto unlock;
>  
>  	entry = pmd_mkyoung(orig_pmd);
> +	if (write)
> +		entry = pmd_mkdirty(entry);
>  	haddr = vmf->address & HPAGE_PMD_MASK;
> -	if (pmdp_set_access_flags(vmf->vma, haddr, vmf->pmd, entry,
> -				vmf->flags & FAULT_FLAG_WRITE))
> +	if (pmdp_set_access_flags(vmf->vma, haddr, vmf->pmd, entry, write))
>  		update_mmu_cache_pmd(vmf->vma, vmf->address, vmf->pmd);
>  
>  unlock:
> -- 
> 2.7.4
> 

-- 
Michal Hocko
SUSE Labs