Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-23 Thread Andy Whitcroft
Andrew Morton wrote: > On Sat, 21 Apr 2007 01:28:43 -0700 Andrew Morton <[EMAIL PROTECTED]> wrote: > >> It would have been better to have patched page_alloc.c independently, then >> to have used HIGH_ORDER in "lumpy: increase pressure at the end of the >> inactive >> list". > > Actually that

Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-23 Thread Andy Whitcroft
Andrew Morton wrote: On Sat, 21 Apr 2007 01:28:43 -0700 Andrew Morton [EMAIL PROTECTED] wrote: It would have been better to have patched page_alloc.c independently, then to have used HIGH_ORDER in lumpy: increase pressure at the end of the inactive list. Actually that doesn't matter,

Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-21 Thread Andrew Morton
On Sat, 21 Apr 2007 01:28:43 -0700 Andrew Morton <[EMAIL PROTECTED]> wrote: > It would have been better to have patched page_alloc.c independently, then > to have used HIGH_ORDER in "lumpy: increase pressure at the end of the > inactive > list". Actually that doesn't matter, because I plan on

Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-21 Thread Andrew Morton
On Fri, 20 Apr 2007 16:04:36 +0100 Andy Whitcroft <[EMAIL PROTECTED]> wrote: > The memory allocator treats lower order (order <= 3) and higher order > (order >= 4) allocations in slightly different ways. As lower orders > are much more likely to be available and also more likely to be > simply

Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-21 Thread Andrew Morton
On Fri, 20 Apr 2007 16:04:36 +0100 Andy Whitcroft [EMAIL PROTECTED] wrote: The memory allocator treats lower order (order = 3) and higher order (order = 4) allocations in slightly different ways. As lower orders are much more likely to be available and also more likely to be simply reclaimed

Re: [PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-21 Thread Andrew Morton
On Sat, 21 Apr 2007 01:28:43 -0700 Andrew Morton [EMAIL PROTECTED] wrote: It would have been better to have patched page_alloc.c independently, then to have used HIGH_ORDER in lumpy: increase pressure at the end of the inactive list. Actually that doesn't matter, because I plan on lumping

[PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-20 Thread Andy Whitcroft
The memory allocator treats lower order (order <= 3) and higher order (order >= 4) allocations in slightly different ways. As lower orders are much more likely to be available and also more likely to be simply reclaimed it is deemed reasonable to wait longer for those. Lumpy reclaim also changes

[PATCH 3/3] introduce HIGH_ORDER delineating easily reclaimable orders

2007-04-20 Thread Andy Whitcroft
The memory allocator treats lower order (order = 3) and higher order (order = 4) allocations in slightly different ways. As lower orders are much more likely to be available and also more likely to be simply reclaimed it is deemed reasonable to wait longer for those. Lumpy reclaim also changes