Re: [HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-11 Thread Andrew Morton
On Mon, 10 Apr 2017 21:33:51 -0400 (EDT) Jerome Glisse wrote: > > On Mon, 10 Apr 2017 10:43:26 +0200 Michal Hocko wrote: > > > > > There are more for alpha allmodconfig > > > > HMM is rather a compile catastrophe, as was the earlier version I > > merged. > > > > Jerome, I'm thinking you need

Re: [HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-10 Thread Jerome Glisse
> On Mon, 10 Apr 2017 10:43:26 +0200 Michal Hocko wrote: > > > There are more for alpha allmodconfig > > HMM is rather a compile catastrophe, as was the earlier version I > merged. > > Jerome, I'm thinking you need to install some cross-compilers! Sorry about that. I tested some but obviously

Re: [HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-10 Thread Andrew Morton
On Mon, 10 Apr 2017 10:43:26 +0200 Michal Hocko wrote: > There are more for alpha allmodconfig HMM is rather a compile catastrophe, as was the earlier version I merged. Jerome, I'm thinking you need to install some cross-compilers!

Re: [HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-10 Thread Michal Hocko
There are more for alpha allmodconfig === Config /home/mhocko/work/build-test/configs/alpha/allmodconfig mm/hmm.c: In function 'hmm_vma_walk_pmd': mm/hmm.c:370:4: error: implicit declaration of function 'pmd_pfn' [-Werror=implicit-function-declaration] unsigned long pfn = pmd_pfn(pmd) + pte_i

Re: [HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-10 Thread Michal Hocko
On Wed 05-04-17 16:40:20, Jérôme Glisse wrote: > This does not use existing page table walker because we want to share > same code for our page fault handler. I am getting the following compilation error with sparc32 allmodconfig. I didn't check more closely yet. mm/hmm.c: In function 'hmm_vma_wa

[HMM 10/16] mm/hmm/mirror: helper to snapshot CPU page table v2

2017-04-05 Thread Jérôme Glisse
This does not use existing page table walker because we want to share same code for our page fault handler. Changes since v1: - Use spinlock instead of rcu synchronized list traversal Signed-off-by: Jérôme Glisse Signed-off-by: Evgeny Baskakov Signed-off-by: John Hubbard Signed-off-by: Mark