Re: [PATCH v2 0/3] special_mapping_fault() is broken

2015-07-11 Thread Oleg Nesterov
On 07/10, Andrew Morton wrote: > > On Fri, 10 Jul 2015 18:51:21 +0200 Oleg Nesterov wrote: > > > special_mapping_fault() is absolutely broken. It seems it was always > > wrong, but this didn't matter until vdso/vvar started to use more than > > one page. > > > > The patches are the same, just 1/3

Re: [PATCH v2 0/3] special_mapping_fault() is broken

2015-07-10 Thread Andrew Morton
On Fri, 10 Jul 2015 18:51:21 +0200 Oleg Nesterov wrote: > special_mapping_fault() is absolutely broken. It seems it was always > wrong, but this didn't matter until vdso/vvar started to use more than > one page. > > The patches are the same, just 1/3 was re-diffed on top of the recent > 6b7339f4

Re: [PATCH v2 0/3] special_mapping_fault() is broken

2015-07-10 Thread Davidlohr Bueso
On Fri, 2015-07-10 at 18:51 +0200, Oleg Nesterov wrote: > And after this change vma_is_anonymous() becomes really trivial, it > simply checks vm_ops == NULL. However, I do think the helper makes > sense. There are a lot of ->vm_ops != NULL checks, the helper makes > the caller's code more understan

Re: [PATCH v2 0/3] special_mapping_fault() is broken

2015-07-10 Thread Kirill A. Shutemov
On Fri, Jul 10, 2015 at 06:51:21PM +0200, Oleg Nesterov wrote: > Hello, > > special_mapping_fault() is absolutely broken. It seems it was always > wrong, but this didn't matter until vdso/vvar started to use more than > one page. > > The patches are the same, just 1/3 was re-diffed on top of the

[PATCH v2 0/3] special_mapping_fault() is broken

2015-07-10 Thread Oleg Nesterov
Hello, special_mapping_fault() is absolutely broken. It seems it was always wrong, but this didn't matter until vdso/vvar started to use more than one page. The patches are the same, just 1/3 was re-diffed on top of the recent 6b7339f4c31ad "mm: avoid setting up anonymous pages into file mapping"