Re: [PATCHv5 0/6] x86: 32-bit compatible C/R on x86_64

2016-09-14 Thread Dmitry Safonov
On 09/05/2016 04:33 PM, Dmitry Safonov wrote: Changes from v4: - check both vm_ops and vm_private_data to avoid (unlikely) confusion with some other vma in map_vdso_once (as Andy noticed) - which would lead to unable to use this API in that unlikely-case (vm_private_data may be

Re: [PATCHv5 0/6] x86: 32-bit compatible C/R on x86_64

2016-09-14 Thread Dmitry Safonov
On 09/05/2016 04:33 PM, Dmitry Safonov wrote: Changes from v4: - check both vm_ops and vm_private_data to avoid (unlikely) confusion with some other vma in map_vdso_once (as Andy noticed) - which would lead to unable to use this API in that unlikely-case (vm_private_data may be

[PATCHv5 0/6] x86: 32-bit compatible C/R on x86_64

2016-09-05 Thread Dmitry Safonov
Changes from v4: - check both vm_ops and vm_private_data to avoid (unlikely) confusion with some other vma in map_vdso_once (as Andy noticed) - which would lead to unable to use this API in that unlikely-case (vm_private_data may be uninitialized and be the same as vvar_mapping or

[PATCHv5 0/6] x86: 32-bit compatible C/R on x86_64

2016-09-05 Thread Dmitry Safonov
Changes from v4: - check both vm_ops and vm_private_data to avoid (unlikely) confusion with some other vma in map_vdso_once (as Andy noticed) - which would lead to unable to use this API in that unlikely-case (vm_private_data may be uninitialized and be the same as vvar_mapping or