* [PATCH V2] xfs: simplify kmem_{zone_}zalloc
@ 2013-11-04 10:21 Gu Zheng
2013-11-04 20:59 ` Dave Chinner
2013-11-06 22:59 ` Ben Myers
0 siblings, 2 replies; 3+ messages in thread
From: Gu Zheng @ 2013-11-04 10:21 UTC (permalink / raw)
To: bpm, elder; +Cc: xfs, linux-kernel, guz.fnst, Dave Chinner
Introduce flag KM_ZERO which is used to alloc zeroed entry, and convert
kmem_{zone_}zalloc to call kmem_{zone_}alloc() with KM_ZERO directly,
in order to avoid the setting to zero step.
And following Dave's suggestion, make kmem_{zone_}zalloc static inline
into kmem.h as they're now just a simple wrapper.
V2:
Make kmem_{zone_}zalloc static inline into kmem.h as Dave suggested.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
---
fs/xfs/kmem.c | 22 ----------------------
fs/xfs/kmem.h | 21 ++++++++++++++++++---
2 files changed, 18 insertions(+), 25 deletions(-)
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index a02cfb9..66a36be 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -63,17 +63,6 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
}
void *
-kmem_zalloc(size_t size, xfs_km_flags_t flags)
-{
- void *ptr;
-
- ptr = kmem_alloc(size, flags);
- if (ptr)
- memset((char *)ptr, 0, (int)size);
- return ptr;
-}
-
-void *
kmem_zalloc_large(size_t size, xfs_km_flags_t flags)
{
void *ptr;
@@ -128,14 +117,3 @@ kmem_zone_alloc(kmem_zone_t *zone, xfs_km_flags_t flags)
congestion_wait(BLK_RW_ASYNC, HZ/50);
} while (1);
}
-
-void *
-kmem_zone_zalloc(kmem_zone_t *zone, xfs_km_flags_t flags)
-{
- void *ptr;
-
- ptr = kmem_zone_alloc(zone, flags);
- if (ptr)
- memset((char *)ptr, 0, kmem_cache_size(zone));
- return ptr;
-}
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index 3a7371c..64db0e5 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -32,6 +32,7 @@ typedef unsigned __bitwise xfs_km_flags_t;
#define KM_NOSLEEP ((__force xfs_km_flags_t)0x0002u)
#define KM_NOFS ((__force xfs_km_flags_t)0x0004u)
#define KM_MAYFAIL ((__force xfs_km_flags_t)0x0008u)
+#define KM_ZERO ((__force xfs_km_flags_t)0x0010u)
/*
* We use a special process flag to avoid recursive callbacks into
@@ -43,7 +44,7 @@ kmem_flags_convert(xfs_km_flags_t flags)
{
gfp_t lflags;
- BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL));
+ BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL|KM_ZERO));
if (flags & KM_NOSLEEP) {
lflags = GFP_ATOMIC | __GFP_NOWARN;
@@ -52,11 +53,14 @@ kmem_flags_convert(xfs_km_flags_t flags)
if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS))
lflags &= ~__GFP_FS;
}
+
+ if (flags & KM_ZERO)
+ lflags |= __GFP_ZERO;
+
return lflags;
}
extern void *kmem_alloc(size_t, xfs_km_flags_t);
-extern void *kmem_zalloc(size_t, xfs_km_flags_t);
extern void *kmem_zalloc_large(size_t size, xfs_km_flags_t);
extern void *kmem_realloc(const void *, size_t, size_t, xfs_km_flags_t);
extern void kmem_free(const void *);
@@ -64,6 +68,12 @@ extern void kmem_free(const void *);
extern void *kmem_zalloc_greedy(size_t *, size_t, size_t);
+static inline void *
+kmem_zalloc(size_t size, xfs_km_flags_t flags)
+{
+ return kmem_alloc(size, flags | KM_ZERO);
+}
+
/*
* Zone interfaces
*/
@@ -102,6 +112,11 @@ kmem_zone_destroy(kmem_zone_t *zone)
}
extern void *kmem_zone_alloc(kmem_zone_t *, xfs_km_flags_t);
-extern void *kmem_zone_zalloc(kmem_zone_t *, xfs_km_flags_t);
+
+static inline void *
+kmem_zone_zalloc(kmem_zone_t *zone, xfs_km_flags_t flags)
+{
+ return kmem_zone_alloc(zone, flags | KM_ZERO);
+}
#endif /* __XFS_SUPPORT_KMEM_H__ */
--
1.7.7
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH V2] xfs: simplify kmem_{zone_}zalloc
2013-11-04 10:21 [PATCH V2] xfs: simplify kmem_{zone_}zalloc Gu Zheng
@ 2013-11-04 20:59 ` Dave Chinner
2013-11-06 22:59 ` Ben Myers
1 sibling, 0 replies; 3+ messages in thread
From: Dave Chinner @ 2013-11-04 20:59 UTC (permalink / raw)
To: Gu Zheng; +Cc: bpm, elder, xfs, linux-kernel
On Mon, Nov 04, 2013 at 06:21:05PM +0800, Gu Zheng wrote:
> Introduce flag KM_ZERO which is used to alloc zeroed entry, and convert
> kmem_{zone_}zalloc to call kmem_{zone_}alloc() with KM_ZERO directly,
> in order to avoid the setting to zero step.
> And following Dave's suggestion, make kmem_{zone_}zalloc static inline
> into kmem.h as they're now just a simple wrapper.
>
> V2:
> Make kmem_{zone_}zalloc static inline into kmem.h as Dave suggested.
>
> Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Looks good. It also results in a slight reduction in code size:
text data bss dec hex filename
792234 99018 632 891884 d9bec fs/xfs/xfs.o.orig
792090 99018 632 891740 d9b5c fs/xfs/xfs.o
Which means making it inline hasn't cost us anything at individual
call sites.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH V2] xfs: simplify kmem_{zone_}zalloc
2013-11-04 10:21 [PATCH V2] xfs: simplify kmem_{zone_}zalloc Gu Zheng
2013-11-04 20:59 ` Dave Chinner
@ 2013-11-06 22:59 ` Ben Myers
1 sibling, 0 replies; 3+ messages in thread
From: Ben Myers @ 2013-11-06 22:59 UTC (permalink / raw)
To: Gu Zheng; +Cc: elder, linux-kernel, xfs
On Mon, Nov 04, 2013 at 06:21:05PM +0800, Gu Zheng wrote:
> Introduce flag KM_ZERO which is used to alloc zeroed entry, and convert
> kmem_{zone_}zalloc to call kmem_{zone_}alloc() with KM_ZERO directly,
> in order to avoid the setting to zero step.
> And following Dave's suggestion, make kmem_{zone_}zalloc static inline
> into kmem.h as they're now just a simple wrapper.
>
> V2:
> Make kmem_{zone_}zalloc static inline into kmem.h as Dave suggested.
>
> Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Applied. Thanks!
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2013-11-06 22:59 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-11-04 10:21 [PATCH V2] xfs: simplify kmem_{zone_}zalloc Gu Zheng
2013-11-04 20:59 ` Dave Chinner
2013-11-06 22:59 ` Ben Myers
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).