* [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
@ 2014-05-16 20:38 Fabian Frederick
2014-05-19 23:09 ` Andrew Morton
0 siblings, 1 reply; 7+ messages in thread
From: Fabian Frederick @ 2014-05-16 20:38 UTC (permalink / raw)
To: linux-kernel; +Cc: akpm, Joe Perches
Replace while blocksize;shift by fls -1
Suggested-By: Joe Perches <joe@perches.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Fabian Frederick <fabf@skynet.be>
---
v2: rebased on top of mmots
compiles without including bitops
fs/hfsplus/wrapper.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c
index 284c90f..8e8cf50d 100644
--- a/fs/hfsplus/wrapper.c
+++ b/fs/hfsplus/wrapper.c
@@ -231,9 +231,7 @@ reread:
if (blocksize < HFSPLUS_SECTOR_SIZE || ((blocksize - 1) & blocksize))
goto out_free_backup_vhdr;
sbi->alloc_blksz = blocksize;
- sbi->alloc_blksz_shift = 0;
- while ((blocksize >>= 1) != 0)
- sbi->alloc_blksz_shift++;
+ sbi->alloc_blksz_shift = fls(blocksize) - 1;
blocksize = min_t(u32, sbi->alloc_blksz, PAGE_SIZE);
/*
--
1.8.4.5
^ permalink raw reply related [flat|nested] 7+ messages in thread* Re: [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
2014-05-16 20:38 [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls Fabian Frederick
@ 2014-05-19 23:09 ` Andrew Morton
2014-05-19 23:22 ` Joe Perches
0 siblings, 1 reply; 7+ messages in thread
From: Andrew Morton @ 2014-05-19 23:09 UTC (permalink / raw)
To: Fabian Frederick; +Cc: linux-kernel, Joe Perches
On Fri, 16 May 2014 22:38:15 +0200 Fabian Frederick <fabf@skynet.be> wrote:
> Replace while blocksize;shift by fls -1
>
> Suggested-By: Joe Perches <joe@perches.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Signed-off-by: Fabian Frederick <fabf@skynet.be>
> ---
> v2: rebased on top of mmots
> compiles without including bitops
>
> fs/hfsplus/wrapper.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c
> index 284c90f..8e8cf50d 100644
> --- a/fs/hfsplus/wrapper.c
> +++ b/fs/hfsplus/wrapper.c
> @@ -231,9 +231,7 @@ reread:
> if (blocksize < HFSPLUS_SECTOR_SIZE || ((blocksize - 1) & blocksize))
> goto out_free_backup_vhdr;
> sbi->alloc_blksz = blocksize;
> - sbi->alloc_blksz_shift = 0;
> - while ((blocksize >>= 1) != 0)
> - sbi->alloc_blksz_shift++;
> + sbi->alloc_blksz_shift = fls(blocksize) - 1;
> blocksize = min_t(u32, sbi->alloc_blksz, PAGE_SIZE);
fls() always makes my brain hurt. ilog() is nicer?
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
2014-05-19 23:09 ` Andrew Morton
@ 2014-05-19 23:22 ` Joe Perches
2014-05-19 23:40 ` Andrew Morton
0 siblings, 1 reply; 7+ messages in thread
From: Joe Perches @ 2014-05-19 23:22 UTC (permalink / raw)
To: Andrew Morton; +Cc: Fabian Frederick, linux-kernel
On Mon, 2014-05-19 at 16:09 -0700, Andrew Morton wrote:
> On Fri, 16 May 2014 22:38:15 +0200 Fabian Frederick <fabf@skynet.be> wrote:
> > Replace while blocksize;shift by fls -1
[]
> fls() always makes my brain hurt. ilog() is nicer?
Maybe take an aspirin?
ilog2(n) is effectively fls(n) - 1 but n can be
either u64 or u64 and ilog2 is compile-time
optimized when n is constant.
blocksize is a non-const u32 here.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
2014-05-19 23:22 ` Joe Perches
@ 2014-05-19 23:40 ` Andrew Morton
2014-05-19 23:50 ` Joe Perches
0 siblings, 1 reply; 7+ messages in thread
From: Andrew Morton @ 2014-05-19 23:40 UTC (permalink / raw)
To: Joe Perches; +Cc: Fabian Frederick, linux-kernel
On Mon, 19 May 2014 16:22:28 -0700 Joe Perches <joe@perches.com> wrote:
> On Mon, 2014-05-19 at 16:09 -0700, Andrew Morton wrote:
> > On Fri, 16 May 2014 22:38:15 +0200 Fabian Frederick <fabf@skynet.be> wrote:
> > > Replace while blocksize;shift by fls -1
> []
> > fls() always makes my brain hurt. ilog() is nicer?
>
> Maybe take an aspirin?
>
> ilog2(n) is effectively fls(n) - 1
Precisely. ilog2() means "integer log 2". Whereas fls() is some weird
low-level bit-scanning thing.
What we want at this callsite is "integer log 2". So use it.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
2014-05-19 23:40 ` Andrew Morton
@ 2014-05-19 23:50 ` Joe Perches
2014-05-20 5:02 ` Fabian Frederick
0 siblings, 1 reply; 7+ messages in thread
From: Joe Perches @ 2014-05-19 23:50 UTC (permalink / raw)
To: Andrew Morton; +Cc: Fabian Frederick, linux-kernel
On Mon, 2014-05-19 at 16:40 -0700, Andrew Morton wrote:
> On Mon, 19 May 2014 16:22:28 -0700 Joe Perches <joe@perches.com> wrote:
>
> > On Mon, 2014-05-19 at 16:09 -0700, Andrew Morton wrote:
> > > On Fri, 16 May 2014 22:38:15 +0200 Fabian Frederick <fabf@skynet.be> wrote:
> > > > Replace while blocksize;shift by fls -1
> > []
> > > fls() always makes my brain hurt. ilog() is nicer?
> > Maybe take an aspirin?
> > ilog2(n) is effectively fls(n) - 1
> Precisely. ilog2() means "integer log 2". Whereas fls() is some weird
> low-level bit-scanning thing.
<shrug> find-last[bit]-set is pretty obvious to me.
ilog2(n) makes me think just a little more.
> What we want at this callsite is "integer log 2". So use it.
Same thing. Same ratio of use too.
$ git grep -w fls | wc -l
457
$ git grep -w ilog2 | wc -l
391
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls
2014-05-19 23:50 ` Joe Perches
@ 2014-05-20 5:02 ` Fabian Frederick
2014-05-20 5:10 ` Joe Perches
0 siblings, 1 reply; 7+ messages in thread
From: Fabian Frederick @ 2014-05-20 5:02 UTC (permalink / raw)
To: Joe Perches; +Cc: Andrew Morton, linux-kernel
On Mon, 19 May 2014 16:50:05 -0700
Joe Perches <joe@perches.com> wrote:
> On Mon, 2014-05-19 at 16:40 -0700, Andrew Morton wrote:
> > On Mon, 19 May 2014 16:22:28 -0700 Joe Perches <joe@perches.com> wrote:
> >
> > > On Mon, 2014-05-19 at 16:09 -0700, Andrew Morton wrote:
> > > > On Fri, 16 May 2014 22:38:15 +0200 Fabian Frederick <fabf@skynet.be> wrote:
> > > > > Replace while blocksize;shift by fls -1
> > > []
> > > > fls() always makes my brain hurt. ilog() is nicer?
> > > Maybe take an aspirin?
> > > ilog2(n) is effectively fls(n) - 1
> > Precisely. ilog2() means "integer log 2". Whereas fls() is some weird
> > low-level bit-scanning thing.
>
> <shrug> find-last[bit]-set is pretty obvious to me.
> ilog2(n) makes me think just a little more.
>
> > What we want at this callsite is "integer log 2". So use it.
>
> Same thing. Same ratio of use too.
>
> $ git grep -w fls | wc -l
> 457
> $ git grep -w ilog2 | wc -l
> 391
and the winner is ? :)
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2014-05-20 5:10 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-05-16 20:38 [PATCH V2 mmots] fs/hfsplus/wrapper.c: replace shift loop by fls Fabian Frederick
2014-05-19 23:09 ` Andrew Morton
2014-05-19 23:22 ` Joe Perches
2014-05-19 23:40 ` Andrew Morton
2014-05-19 23:50 ` Joe Perches
2014-05-20 5:02 ` Fabian Frederick
2014-05-20 5:10 ` Joe Perches
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox