Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[PATCH] udf: use ext2_find_next_bit

2 views
Skip to first unread message

Akinobu Mita

unread,
Feb 23, 2010, 9:20:02 AM2/23/10
to
Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
in little endian bitmap region.

Signed-off-by: Akinobu Mita <akinob...@gmail.com>
Cc: Jan Kara <ja...@suse.cz>
---
fs/udf/balloc.c | 50 ++------------------------------------------------
1 files changed, 2 insertions(+), 48 deletions(-)

diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index 82372e3..05347f4 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -31,55 +31,9 @@
#define udf_clear_bit(nr, addr) ext2_clear_bit(nr, addr)
#define udf_set_bit(nr, addr) ext2_set_bit(nr, addr)
#define udf_test_bit(nr, addr) ext2_test_bit(nr, addr)
-#define udf_find_first_one_bit(addr, size) find_first_one_bit(addr, size)
+#define udf_find_first_one_bit(addr, size) ext2_find_first_bit(addr, size)
#define udf_find_next_one_bit(addr, size, offset) \
- find_next_one_bit(addr, size, offset)
-
-#define leBPL_to_cpup(x) leNUM_to_cpup(BITS_PER_LONG, x)
-#define leNUM_to_cpup(x, y) xleNUM_to_cpup(x, y)
-#define xleNUM_to_cpup(x, y) (le ## x ## _to_cpup(y))
-#define uintBPL_t uint(BITS_PER_LONG)
-#define uint(x) xuint(x)
-#define xuint(x) __le ## x
-
-static inline int find_next_one_bit(void *addr, int size, int offset)
-{
- uintBPL_t *p = ((uintBPL_t *) addr) + (offset / BITS_PER_LONG);
- int result = offset & ~(BITS_PER_LONG - 1);
- unsigned long tmp;
-
- if (offset >= size)
- return size;
- size -= result;
- offset &= (BITS_PER_LONG - 1);
- if (offset) {
- tmp = leBPL_to_cpup(p++);
- tmp &= ~0UL << offset;
- if (size < BITS_PER_LONG)
- goto found_first;
- if (tmp)
- goto found_middle;
- size -= BITS_PER_LONG;
- result += BITS_PER_LONG;
- }
- while (size & ~(BITS_PER_LONG - 1)) {
- tmp = leBPL_to_cpup(p++);
- if (tmp)
- goto found_middle;
- result += BITS_PER_LONG;
- size -= BITS_PER_LONG;
- }
- if (!size)
- return result;
- tmp = leBPL_to_cpup(p);
-found_first:
- tmp &= ~0UL >> (BITS_PER_LONG - size);
-found_middle:
- return result + ffz(~tmp);
-}
-
-#define find_first_one_bit(addr, size)\
- find_next_one_bit((addr), (size), 0)
+ ext2_find_next_bit(addr, size, offset)

static int read_block_bitmap(struct super_block *sb,
struct udf_bitmap *bitmap, unsigned int block,
--
1.6.0.6

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Jan Kara

unread,
Feb 24, 2010, 11:30:02 AM2/24/10
to
On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
> in little endian bitmap region.
>
> Signed-off-by: Akinobu Mita <akinob...@gmail.com>
> Cc: Jan Kara <ja...@suse.cz>
I've looked at the code and I think this is wrong. UDF uses 1 for free
block in the bitmap and 0 for used one. So you need to use
generic_find_next_le_bit...
Something like the patch below?

Honza

--
Jan Kara <ja...@suse.cz>
SUSE Labs, CR
---

From bfc8c674188d4be5856e445f9b57e1b7a3dbff6d Mon Sep 17 00:00:00 2001
From: Jan Kara <ja...@suse.cz>
Date: Wed, 24 Feb 2010 17:18:57 +0100
Subject: [PATCH] udf: Use generic function for bit searching

Use generic_find_next_le_bit to search for bits in bitmap so
that we don't duplicate code unnecessarily.

Signed-off-by: Jan Kara <ja...@suse.cz>
---
fs/udf/balloc.c | 51 ++-------------------------------------------------
1 files changed, 2 insertions(+), 49 deletions(-)

diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index 82372e3..f5ff06c 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -31,55 +31,8 @@


#define udf_clear_bit(nr, addr) ext2_clear_bit(nr, addr)
#define udf_set_bit(nr, addr) ext2_set_bit(nr, addr)
#define udf_test_bit(nr, addr) ext2_test_bit(nr, addr)
-#define udf_find_first_one_bit(addr, size) find_first_one_bit(addr, size)

+ generic_find_next_le_bit((unsigned long *)addr, size, offset)



static int read_block_bitmap(struct super_block *sb,
struct udf_bitmap *bitmap, unsigned int block,

@@ -356,7 +309,7 @@ repeat:
break;
}
} else {
- bit = udf_find_next_one_bit((char *)bh->b_data,
+ bit = udf_find_next_one_bit(bh->b_data,
sb->s_blocksize << 3,
group_start << 3);
if (bit < sb->s_blocksize << 3)
--
1.6.4.2

Akinobu Mita

unread,
Feb 24, 2010, 8:40:02 PM2/24/10
to
2010/2/25 Jan Kara <ja...@suse.cz>:

> On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
>> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
>> in little endian bitmap region.
>>
>> Signed-off-by: Akinobu Mita <akinob...@gmail.com>
>> Cc: Jan Kara <ja...@suse.cz>
> �I've looked at the code and I think this is wrong. UDF uses 1 for free
> block in the bitmap and 0 for used one. So you need to use
> generic_find_next_le_bit...
> �Something like the patch below?

According to include/asm-generic/bitops/ext2-non-atomic.h,
generic_find_next_le_bit() == ext2_find_next_bit().

So, is there any difference between my patch and yours?
Or am I missing something?

BTW, I realized that udf_find_first_one_bit() macro is not used.
So you can remove it in this patch.

> � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �Honza

Geert Uytterhoeven

unread,
Feb 25, 2010, 4:00:02 AM2/25/10
to
On Thu, Feb 25, 2010 at 02:38, Akinobu Mita <akinob...@gmail.com> wrote:
> 2010/2/25 Jan Kara <ja...@suse.cz>:
>> On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
>>> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
>>> in little endian bitmap region.

Is any of this in linux-next now?

http://kisskb.ellerman.id.au/kisskb/buildresult/2208481/
| fs/udf/balloc.c:274: error: implicit declaration of function
'generic_find_next_le_bit'

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Akinobu Mita

unread,
Feb 25, 2010, 9:40:02 AM2/25/10
to
2010/2/25 Geert Uytterhoeven <ge...@linux-m68k.org>:

> On Thu, Feb 25, 2010 at 02:38, Akinobu Mita <akinob...@gmail.com> wrote:
>> 2010/2/25 Jan Kara <ja...@suse.cz>:
>>> On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
>>>> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
>>>> in little endian bitmap region.
>
> Is any of this in linux-next now?
>
> http://kisskb.ellerman.id.au/kisskb/buildresult/2208481/
> | fs/udf/balloc.c:274: error: implicit declaration of function
> 'generic_find_next_le_bit'

Yep, Jan's patch caused the build breakage.

Because generic_find_next_le_bit() is not available for all
archtectures. So we should use ext2_find_next_bit() here.

Geert Uytterhoeven

unread,
Feb 28, 2010, 5:10:03 AM2/28/10
to
On Thu, Feb 25, 2010 at 15:29, Akinobu Mita <akinob...@gmail.com> wrote:
> 2010/2/25 Geert Uytterhoeven <ge...@linux-m68k.org>:
>> On Thu, Feb 25, 2010 at 02:38, Akinobu Mita <akinob...@gmail.com> wrote:
>>> 2010/2/25 Jan Kara <ja...@suse.cz>:
>>>> On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
>>>>> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
>>>>> in little endian bitmap region.
>>
>> Is any of this in linux-next now?
>>
>> http://kisskb.ellerman.id.au/kisskb/buildresult/2208481/
>> | fs/udf/balloc.c:274: error: implicit declaration of function
>> 'generic_find_next_le_bit'
>
> Yep, Jan's patch caused the build breakage.
>
> Because generic_find_next_le_bit() is not available for all
> archtectures. So we should use ext2_find_next_bit() here.

Most architectures use the definitions in asm-generic, so they're OK.
M68k doesn't. S390 is also affected, and I think arm as well (but there's no
arm all-modconfig build in linux-next, so I'm not 100% sure).

I'm cooking a patch...

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Geert Uytterhoeven

unread,
Feb 28, 2010, 9:10:02 AM2/28/10
to
On Sun, Feb 28, 2010 at 11:04, Geert Uytterhoeven <ge...@linux-m68k.org> wrote:
> On Thu, Feb 25, 2010 at 15:29, Akinobu Mita <akinob...@gmail.com> wrote:
>> 2010/2/25 Geert Uytterhoeven <ge...@linux-m68k.org>:
>>> On Thu, Feb 25, 2010 at 02:38, Akinobu Mita <akinob...@gmail.com> wrote:
>>>> 2010/2/25 Jan Kara <ja...@suse.cz>:
>>>>> On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
>>>>>> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
>>>>>> in little endian bitmap region.
>>>
>>> Is any of this in linux-next now?
>>>
>>> http://kisskb.ellerman.id.au/kisskb/buildresult/2208481/
>>> | fs/udf/balloc.c:274: error: implicit declaration of function
>>> 'generic_find_next_le_bit'
>>
>> Yep, Jan's patch caused the build breakage.
>>
>> Because generic_find_next_le_bit() is not available for all
>> archtectures. So we should use ext2_find_next_bit() here.
>
> Most architectures use the definitions in asm-generic, so they're OK.
> M68k doesn't. S390 is also affected, and I think arm as well (but there's no
> arm all-modconfig build in linux-next, so I'm not 100% sure).
>
> I'm cooking a patch...

From c9b5c7e6ef2092be822778a0b6b3d3032c058f5b Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <ge...@linux-m68k.org>
Date: Sun, 28 Feb 2010 13:06:27 +0100
Subject: [PATCH] m68k: Implement generic_find_next_{zero_,}le_bit()

linux-next:
fs/udf/balloc.c: In function 'udf_bitmap_new_block':


fs/udf/balloc.c:274: error: implicit declaration of function
'generic_find_next_le_bit'

Convert ext2_find_next_{zero_,}bit() into generic_find_next_{zero_,}le_bit(),
and wrap the ext2_find_next_{zero_,}bit() around the latter.

Signed-off-by: Geert Uytterhoeven <ge...@linux-m68k.org>
---
arch/m68k/include/asm/bitops_mm.h | 14 ++++++++------
1 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/arch/m68k/include/asm/bitops_mm.h
b/arch/m68k/include/asm/bitops_mm.h
index 9bde784..b4ecdaa 100644
--- a/arch/m68k/include/asm/bitops_mm.h
+++ b/arch/m68k/include/asm/bitops_mm.h
@@ -365,6 +365,10 @@ static inline int minix_test_bit(int nr, const void *vaddr)
#define ext2_set_bit_atomic(lock, nr, addr) test_and_set_bit((nr) ^
24, (unsigned long *)(addr))
#define ext2_clear_bit(nr, addr) __test_and_clear_bit((nr) ^ 24,
(unsigned long *)(addr))
#define ext2_clear_bit_atomic(lock, nr, addr) test_and_clear_bit((nr)
^ 24, (unsigned long *)(addr))
+#define ext2_find_next_zero_bit(addr, size, offset) \
+ generic_find_next_zero_le_bit((unsigned long *)addr, size, offset)
+#define ext2_find_next_bit(addr, size, offset) \


+ generic_find_next_le_bit((unsigned long *)addr, size, offset)

static inline int ext2_test_bit(int nr, const void *vaddr)
{
@@ -394,10 +398,9 @@ static inline int ext2_find_first_zero_bit(const
void *vaddr, unsigned size)
return (p - addr) * 32 + res;
}

-static inline int ext2_find_next_zero_bit(const void *vaddr, unsigned size,
- unsigned offset)
+static inline unsigned long generic_find_next_zero_le_bit(const
unsigned long *addr,
+ unsigned long size, unsigned long offset)
{
- const unsigned long *addr = vaddr;
const unsigned long *p = addr + (offset >> 5);
int bit = offset & 31UL, res;

@@ -437,10 +440,9 @@ static inline int ext2_find_first_bit(const void
*vaddr, unsigned size)
return (p - addr) * 32 + res;
}

-static inline int ext2_find_next_bit(const void *vaddr, unsigned size,
- unsigned offset)
+static inline unsigned long generic_find_next_le_bit(const unsigned long *addr,
+ unsigned long size, unsigned long offset)
{
- const unsigned long *addr = vaddr;
const unsigned long *p = addr + (offset >> 5);
int bit = offset & 31UL, res;

--
1.6.0.4

Jan Kara

unread,
Mar 1, 2010, 6:00:02 AM3/1/10
to
I've now used Akinobu's patch so yours isn't really needed but it's a
good cleanup anyway in my opinion.

Honza
--
Jan Kara <ja...@suse.cz>
SUSE Labs, CR

Jan Kara

unread,
Mar 1, 2010, 6:00:02 AM3/1/10
to
On Thu 25-02-10 10:38:25, Akinobu Mita wrote:
> 2010/2/25 Jan Kara <ja...@suse.cz>:
> > On Tue 23-02-10 23:11:13, Akinobu Mita wrote:
> >> Use ext2_find_next_bit (generic_find_next_le_bit) to find the set bit
> >> in little endian bitmap region.
> >>
> >> Signed-off-by: Akinobu Mita <akinob...@gmail.com>
> >> Cc: Jan Kara <ja...@suse.cz>
> > �I've looked at the code and I think this is wrong. UDF uses 1 for free
> > block in the bitmap and 0 for used one. So you need to use
> > generic_find_next_le_bit...
> > �Something like the patch below?
>
> According to include/asm-generic/bitops/ext2-non-atomic.h,
> generic_find_next_le_bit() == ext2_find_next_bit().
Ah, I'm really sorry. I've originally misread the header and thought that
ext2_find_next_bit == generic_find_next_le_zero_bit. So my patch does
exactly the same thing as yours. I've now merged your patch so that you get
proper credit.

> BTW, I realized that udf_find_first_one_bit() macro is not used.
> So you can remove it in this patch.

Yes, I've removed it now.

Honza

--
Jan Kara <ja...@suse.cz>
SUSE Labs, CR

0 new messages