Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-14 Thread Alexander Duyck
On Thu, Feb 14, 2019 at 7:13 AM Jann Horn wrote: > > On Wed, Feb 13, 2019 at 11:42 PM Alexander Duyck > wrote: > > On Wed, Feb 13, 2019 at 12:42 PM Jann Horn wrote: > > > The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum > > > number of references that we might need to crea

Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-14 Thread Jann Horn
On Wed, Feb 13, 2019 at 11:42 PM Alexander Duyck wrote: > On Wed, Feb 13, 2019 at 12:42 PM Jann Horn wrote: > > The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum > > number of references that we might need to create in the fastpath later, > > the bump-allocation fastpath onl

Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-13 Thread Alexander Duyck
On Wed, Feb 13, 2019 at 12:42 PM Jann Horn wrote: > > The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum > number of references that we might need to create in the fastpath later, > the bump-allocation fastpath only has to modify the non-atomic bias value > that tracks the num

Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-13 Thread Andrew Morton
On Wed, 13 Feb 2019 22:11:58 +0100 Jann Horn wrote: > > This is probably more a davem patch than a -mm one. > > Ah, sorry. I assumed that I just should go by which directory the > patched code is in. > > You did just add it to the -mm tree though, right? So I shouldn't > resend it to davem? Ye

Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-13 Thread Jann Horn
On Wed, Feb 13, 2019 at 9:59 PM Andrew Morton wrote: > > On Wed, 13 Feb 2019 21:41:57 +0100 Jann Horn wrote: > > > The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum > > number of references that we might need to create in the fastpath later, > > the bump-allocation fastpath

Re: [PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-13 Thread Andrew Morton
On Wed, 13 Feb 2019 21:41:57 +0100 Jann Horn wrote: > The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum > number of references that we might need to create in the fastpath later, > the bump-allocation fastpath only has to modify the non-atomic bias value > that tracks the nu

[PATCH] mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

2019-02-13 Thread Jann Horn
The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum number of references that we might need to create in the fastpath later, the bump-allocation fastpath only has to modify the non-atomic bias value that tracks the number of extra references we hold instead of the atomic refcoun