Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Kirill A. Shutemov
On Tue, Aug 18, 2015 at 06:41:13PM +0200, Michal Hocko wrote: > On Tue 18-08-15 13:20:22, Michal Hocko wrote: > > On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: > > > Hugh has pointed that compound_head() call can be unsafe in some > > > context. There's one example: > > > > > > CPU0

Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Michal Hocko
On Tue 18-08-15 13:20:22, Michal Hocko wrote: > On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: > > Hugh has pointed that compound_head() call can be unsafe in some > > context. There's one example: > > > > CPU0CPU1 > > > > isolate_migratepages_block()

Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Michal Hocko
On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: > Hugh has pointed that compound_head() call can be unsafe in some > context. There's one example: > > CPU0CPU1 > > isolate_migratepages_block() > page_count() > compound_head() >

Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Michal Hocko
On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: Hugh has pointed that compound_head() call can be unsafe in some context. There's one example: CPU0CPU1 isolate_migratepages_block() page_count() compound_head() !!PageTail() ==

Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Kirill A. Shutemov
On Tue, Aug 18, 2015 at 06:41:13PM +0200, Michal Hocko wrote: On Tue 18-08-15 13:20:22, Michal Hocko wrote: On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: Hugh has pointed that compound_head() call can be unsafe in some context. There's one example: CPU0

Re: [PATCHv2 4/4] mm: make compound_head() robust

2015-08-18 Thread Michal Hocko
On Tue 18-08-15 13:20:22, Michal Hocko wrote: On Mon 17-08-15 18:09:05, Kirill A. Shutemov wrote: Hugh has pointed that compound_head() call can be unsafe in some context. There's one example: CPU0CPU1 isolate_migratepages_block()

[PATCHv2 4/4] mm: make compound_head() robust

2015-08-17 Thread Kirill A. Shutemov
Hugh has pointed that compound_head() call can be unsafe in some context. There's one example: CPU0CPU1 isolate_migratepages_block() page_count() compound_head() !!PageTail() == true put_page()

[PATCHv2 4/4] mm: make compound_head() robust

2015-08-17 Thread Kirill A. Shutemov
Hugh has pointed that compound_head() call can be unsafe in some context. There's one example: CPU0CPU1 isolate_migratepages_block() page_count() compound_head() !!PageTail() == true put_page()