Re: [PATCH] powerpc/mm: Fix Multi hit ERAT cause by recent THP update

2016-02-07 Thread Aneesh Kumar K.V
"Kirill A. Shutemov" writes: > On Fri, Feb 05, 2016 at 11:41:40PM +0530, Aneesh Kumar K.V wrote: >> With ppc64 we use the deposted pgtable_t to store the hash pte slot >> information. We should not withdraw the deposited pgtable_t without >> marking the pmd none. This ensure that low level hash f

Re: [PATCH] powerpc/mm: Fix Multi hit ERAT cause by recent THP update

2016-02-05 Thread Kirill A. Shutemov
On Fri, Feb 05, 2016 at 11:41:40PM +0530, Aneesh Kumar K.V wrote: > With ppc64 we use the deposted pgtable_t to store the hash pte slot > information. We should not withdraw the deposited pgtable_t without > marking the pmd none. This ensure that low level hash fault handling > will skip this huge

[PATCH] powerpc/mm: Fix Multi hit ERAT cause by recent THP update

2016-02-05 Thread Aneesh Kumar K.V
With ppc64 we use the deposted pgtable_t to store the hash pte slot information. We should not withdraw the deposited pgtable_t without marking the pmd none. This ensure that low level hash fault handling will skip this huge pte and we will handle them at upper levels. We do take page table lock th