Re: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-10 Thread Mel Gorman
On Wed, Feb 08, 2012 at 04:14:46PM +0100, Marek Szyprowski wrote: SNIP +static int __reclaim_pages(struct zone *zone, gfp_t gfp_mask, int count) +{ + enum zone_type high_zoneidx = gfp_zone(gfp_mask); + struct zonelist *zonelist = node_zonelist(0, gfp_mask); + int

RE: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-10 Thread Marek Szyprowski
Hello, On Friday, February 10, 2012 12:19 PM Mel Gorman wrote: On Wed, Feb 08, 2012 at 04:14:46PM +0100, Marek Szyprowski wrote: SNIP +static int __reclaim_pages(struct zone *zone, gfp_t gfp_mask, int count) +{ + enum zone_type high_zoneidx = gfp_zone(gfp_mask); +

Re: [Linaro-mm-sig] [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-08 Thread Michal Nazarewicz
On Wed, 08 Feb 2012 03:04:18 +0100, sandeep patil psandee...@gmail.com wrote: There's another problem I am facing with zone watermarks and CMA. Test details: Memory : 480 MB of total memory, 128 MB CMA region Test case : around 600 MB of file transfer over USB RNDIS onto target System Load :

RE: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-08 Thread Marek Szyprowski
Hello, On Friday, February 03, 2012 3:04 PM Mel Gorman wrote: On Fri, Feb 03, 2012 at 01:18:54PM +0100, Marek Szyprowski wrote: alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to

Re: [Linaro-mm-sig] [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-08 Thread sandeep patil
2012/2/8 Michal Nazarewicz min...@mina86.com: On Wed, 08 Feb 2012 03:04:18 +0100, sandeep patil psandee...@gmail.com wrote: There's another problem I am facing with zone watermarks and CMA. Test details: Memory  : 480 MB of total memory, 128 MB CMA region Test case : around 600 MB of file

[PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-03 Thread Marek Szyprowski
alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to *_slowpath style reclaim to grab enough pages to make sure that the final collection of contiguous pages from freelists will not starve the

Re: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-02-03 Thread Mel Gorman
On Fri, Feb 03, 2012 at 01:18:54PM +0100, Marek Szyprowski wrote: alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to *_slowpath style reclaim to grab enough pages to make sure that the final

RE: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-01-31 Thread Marek Szyprowski
Hello, On Monday, January 30, 2012 2:06 PM Mel Gorman wrote: On Thu, Jan 26, 2012 at 10:00:53AM +0100, Marek Szyprowski wrote: alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to

Re: [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-01-30 Thread Mel Gorman
On Thu, Jan 26, 2012 at 10:00:53AM +0100, Marek Szyprowski wrote: alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to *_slowpath style reclaim to grab enough pages to make sure that the final

[PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks

2012-01-26 Thread Marek Szyprowski
alloc_contig_range() performs memory allocation so it also should keep track on keeping the correct level of memory watermarks. This commit adds a call to *_slowpath style reclaim to grab enough pages to make sure that the final collection of contiguous pages from freelists will not starve the