Messages in this thread | | | Date | Tue, 26 Apr 2011 21:15:00 +1000 | From | NeilBrown <> | Subject | Re: [PATCH 02/13] mm: sl[au]b: Add knowledge of PFMEMALLOC reserve pages |
| |
On Tue, 26 Apr 2011 08:36:43 +0100 Mel Gorman <mgorman@suse.de> wrote:
> +bool gfp_pfmemalloc_allowed(gfp_t gfp_mask) > +{ > + return gfp_to_alloc_flags(gfp_mask) & ALLOC_PFMEMALLOC; > +} > + > static inline struct page * > __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, > struct zonelist *zonelist, enum zone_type high_zoneidx, > @@ -2202,8 +2211,16 @@ nopage: > got_pg: > if (kmemcheck_enabled) > kmemcheck_pagealloc_alloc(page, order, gfp_mask); > - return page; > > + /* > + * page->pfmemalloc is set when the caller had PFMEMALLOC set or is > + * been OOM killed. The expectation is that the caller is taking > + * steps that will free more memory. The caller should avoid the > + * page being used for !PFMEMALLOC purposes. > + */ > + page->pfmemalloc = (alloc_flags & ALLOC_PFMEMALLOC); > + > + return page;
Linus doesn't seem to be a fan of this construct: https://lkml.org/lkml/2011/4/1/255
pfmemalloc is a bool, and the value on the right is either 0 or 0x1000.
If bool happens to be typedefed to 'char' or even 'short', pfmemalloc would always be set to 0. Ditto for the gfp_pfmemalloc_allowed function.
Prefixing with '!!' would make it safe.
NeilBrown
| |