Re: [PATCH] thp: fix data loss when splitting a file pmd

2018-07-12 Thread Kirill A. Shutemov
On Thu, Jul 12, 2018 at 12:48:54AM +, Hugh Dickins wrote: > __split_huge_pmd_locked() must check if the cleared huge pmd was dirty, > and propagate that to PageDirty: otherwise, data may be lost when a huge > tmpfs page is modified then split then reclaimed. > > How has this taken so long to

Re: [PATCH] thp: fix data loss when splitting a file pmd

2018-07-12 Thread Kirill A. Shutemov
On Thu, Jul 12, 2018 at 12:48:54AM +, Hugh Dickins wrote: > __split_huge_pmd_locked() must check if the cleared huge pmd was dirty, > and propagate that to PageDirty: otherwise, data may be lost when a huge > tmpfs page is modified then split then reclaimed. > > How has this taken so long to

Re: [PATCH] thp: fix data loss when splitting a file pmd

2018-07-11 Thread Yang Shi
On 7/11/18 5:48 PM, Hugh Dickins wrote: __split_huge_pmd_locked() must check if the cleared huge pmd was dirty, and propagate that to PageDirty: otherwise, data may be lost when a huge tmpfs page is modified then split then reclaimed. How has this taken so long to be noticed? Because there

Re: [PATCH] thp: fix data loss when splitting a file pmd

2018-07-11 Thread Yang Shi
On 7/11/18 5:48 PM, Hugh Dickins wrote: __split_huge_pmd_locked() must check if the cleared huge pmd was dirty, and propagate that to PageDirty: otherwise, data may be lost when a huge tmpfs page is modified then split then reclaimed. How has this taken so long to be noticed? Because there

[PATCH] thp: fix data loss when splitting a file pmd

2018-07-11 Thread Hugh Dickins
__split_huge_pmd_locked() must check if the cleared huge pmd was dirty, and propagate that to PageDirty: otherwise, data may be lost when a huge tmpfs page is modified then split then reclaimed. How has this taken so long to be noticed? Because there was no problem when the huge page is written

[PATCH] thp: fix data loss when splitting a file pmd

2018-07-11 Thread Hugh Dickins
__split_huge_pmd_locked() must check if the cleared huge pmd was dirty, and propagate that to PageDirty: otherwise, data may be lost when a huge tmpfs page is modified then split then reclaimed. How has this taken so long to be noticed? Because there was no problem when the huge page is written