Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Aneesh Kumar K.V
Vlastimil Babka writes: > On 10/21/2016 01:34 PM, Aneesh Kumar K.V wrote: >> Michal Hocko writes: >>> >> >> For both MPOL_PREFERED and MPOL_INTERLEAVE we pick the zone list from >> the node other than the current running node. Why don't we do that for >>

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Aneesh Kumar K.V
Vlastimil Babka writes: > On 10/21/2016 01:34 PM, Aneesh Kumar K.V wrote: >> Michal Hocko writes: >>> >> >> For both MPOL_PREFERED and MPOL_INTERLEAVE we pick the zone list from >> the node other than the current running node. Why don't we do that for >> MPOL_BIND ?ie, if the current node is

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Vlastimil Babka
On 10/21/2016 01:34 PM, Aneesh Kumar K.V wrote: Michal Hocko writes: For both MPOL_PREFERED and MPOL_INTERLEAVE we pick the zone list from the node other than the current running node. Why don't we do that for MPOL_BIND ?ie, if the current node is not part of the policy

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Vlastimil Babka
On 10/21/2016 01:34 PM, Aneesh Kumar K.V wrote: Michal Hocko writes: For both MPOL_PREFERED and MPOL_INTERLEAVE we pick the zone list from the node other than the current running node. Why don't we do that for MPOL_BIND ?ie, if the current node is not part of the policy node mask why are we

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Michal Hocko
On Fri 21-10-16 17:04:50, Aneesh Kumar K.V wrote: > Michal Hocko writes: > > > From: Michal Hocko > > > > __GFP_THISNODE is documented to enforce the allocation to be satisified > > from the requested node with no fallbacks or placement policy > >

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Michal Hocko
On Fri 21-10-16 17:04:50, Aneesh Kumar K.V wrote: > Michal Hocko writes: > > > From: Michal Hocko > > > > __GFP_THISNODE is documented to enforce the allocation to be satisified > > from the requested node with no fallbacks or placement policy > > enforcements. policy_zonelist seemingly breaks

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Aneesh Kumar K.V
Michal Hocko writes: > From: Michal Hocko > > __GFP_THISNODE is documented to enforce the allocation to be satisified > from the requested node with no fallbacks or placement policy > enforcements. policy_zonelist seemingly breaks this semantic if the >

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-21 Thread Aneesh Kumar K.V
Michal Hocko writes: > From: Michal Hocko > > __GFP_THISNODE is documented to enforce the allocation to be satisified > from the requested node with no fallbacks or placement policy > enforcements. policy_zonelist seemingly breaks this semantic if the > current policy is MPOL_MBIND and instead

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-18 Thread Vlastimil Babka
On 10/13/2016 02:59 PM, Michal Hocko wrote: From: Michal Hocko __GFP_THISNODE is documented to enforce the allocation to be satisified from the requested node with no fallbacks or placement policy enforcements. policy_zonelist seemingly breaks this semantic if the current

Re: [PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-18 Thread Vlastimil Babka
On 10/13/2016 02:59 PM, Michal Hocko wrote: From: Michal Hocko __GFP_THISNODE is documented to enforce the allocation to be satisified from the requested node with no fallbacks or placement policy enforcements. policy_zonelist seemingly breaks this semantic if the current policy is MPOL_MBIND

[PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-13 Thread Michal Hocko
From: Michal Hocko __GFP_THISNODE is documented to enforce the allocation to be satisified from the requested node with no fallbacks or placement policy enforcements. policy_zonelist seemingly breaks this semantic if the current policy is MPOL_MBIND and instead of taking the

[PATCH] mm, mempolicy: clean up __GFP_THISNODE confusion in policy_zonelist

2016-10-13 Thread Michal Hocko
From: Michal Hocko __GFP_THISNODE is documented to enforce the allocation to be satisified from the requested node with no fallbacks or placement policy enforcements. policy_zonelist seemingly breaks this semantic if the current policy is MPOL_MBIND and instead of taking the node it will