[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Pine.LNX.4.64.0906141011280.8507@melkki.cs.Helsinki.FI>
Date: Sun, 14 Jun 2009 10:12:39 +0300 (EEST)
From: Pekka J Enberg <penberg@...helsinki.fi>
To: Christoph Lameter <cl@...ux-foundation.org>
cc: torvalds@...ux-foundation.org, linux-kernel@...r.kernel.org,
akpm@...ux-foundation.org, kamezawa.hiroyu@...fujitsu.com,
lizf@...fujitsu.com, mingo@...e.hu, npiggin@...e.de,
yinghai@...nel.org, benh@...nel.crashing.org
Subject: Re: [GIT PULL v2] Early SLAB fixes for 2.6.31
On Fri, 12 Jun 2009, Christoph Lameter wrote:
> On Fri, 12 Jun 2009, Pekka J Enberg wrote:
>
> > @@ -3434,6 +3451,8 @@ __cache_alloc(struct kmem_cache *cachep, gfp_t flags, void *caller)
> > unsigned long save_flags;
> > void *objp;
> >
> > + flags &= slab_gfp_mask;
> > +
> > lockdep_trace_alloc(flags);
> >
> > if (slab_should_failslab(cachep, flags))
>
> Adds code to hot code path.
>
> > @@ -1595,6 +1601,8 @@ static __always_inline void *slab_alloc(struct kmem_cache *s,
> > unsigned long flags;
> > unsigned int objsize;
> >
> > + gfpflags &= slab_gfp_mask;
> > +
> > lockdep_trace_alloc(gfpflags);
> > might_sleep_if(gfpflags & __GFP_WAIT);
> >
>
> Adds code to hot code path.
>
> The allocators mask flags passed to the page allocator through
> GFP_RECLAIM_MASK|GFP_CONSTRAINT_MASK. This is done outside of the critical
> code paths.
>
> The might_sleep issues may be fixed by adding another check to
> __might_sleep().
How about something like this? There should be no extra code in fastpaths
for production configs with this one.
Pekka
diff --git a/mm/slub.c b/mm/slub.c
index 4d6004c..5e8cea1 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1613,6 +1613,8 @@ another_slab:
deactivate_slab(s, c);
new_slab:
+ gfpflags &= slab_gfp_mask;
+
new = get_partial(s, gfpflags, node);
if (new) {
c->page = new;
@@ -1668,13 +1670,14 @@ static __always_inline void *slab_alloc(struct kmem_cache *s,
struct kmem_cache_cpu *c;
unsigned long flags;
unsigned int objsize;
+ gfp_t real_gfp;
- gfpflags &= slab_gfp_mask;
+ real_gfp = gfpflags & slab_gfp_mask;
- lockdep_trace_alloc(gfpflags);
- might_sleep_if(gfpflags & __GFP_WAIT);
+ lockdep_trace_alloc(real_gfp);
+ might_sleep_if(real_gfp & __GFP_WAIT);
- if (should_failslab(s->objsize, gfpflags))
+ if (should_failslab(s->objsize, real_gfp))
return NULL;
local_irq_save(flags);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists