Re: [PATCH v2 6/6] mm, hugetlb: improve page-fault scalability

2014-02-02 Thread Joonsoo Kim
On Fri, Jan 31, 2014 at 09:36:46AM -0800, Davidlohr Bueso wrote: > From: Davidlohr Bueso > > The kernel can currently only handle a single hugetlb page fault at a time. > This is due to a single mutex that serializes the entire path. This lock > protects from spurious OOM errors under conditions

Re: [PATCH v2 6/6] mm, hugetlb: improve page-fault scalability

2014-01-31 Thread Davidlohr Bueso
On Fri, 2014-01-31 at 13:01 -0800, Andrew Morton wrote: > On Fri, 31 Jan 2014 09:36:46 -0800 Davidlohr Bueso wrote: > > > From: Davidlohr Bueso > > > > The kernel can currently only handle a single hugetlb page fault at a time. > > This is due to a single mutex that serializes the entire path.

Re: [PATCH v2 6/6] mm, hugetlb: improve page-fault scalability

2014-01-31 Thread Andrew Morton
On Fri, 31 Jan 2014 09:36:46 -0800 Davidlohr Bueso wrote: > From: Davidlohr Bueso > > The kernel can currently only handle a single hugetlb page fault at a time. > This is due to a single mutex that serializes the entire path. This lock > protects from spurious OOM errors under conditions of lo

[PATCH v2 6/6] mm, hugetlb: improve page-fault scalability

2014-01-31 Thread Davidlohr Bueso
From: Davidlohr Bueso The kernel can currently only handle a single hugetlb page fault at a time. This is due to a single mutex that serializes the entire path. This lock protects from spurious OOM errors under conditions of low of low availability of free hugepages. This problem is specific to h