diff mbox series

[net-next,v5,12/21] bitmap: introduce generic optimized bitmap_size()

Message ID 20240201122216.2634007-13-aleksander.lobakin@intel.com (mailing list archive)
State Not Applicable
Delegated to: Netdev Maintainers
Headers show
Series ice: add PFCP filter support | expand

Checks

Context Check Description
netdev/series_format fail Series longer than 15 patches (and no cover letter)
netdev/tree_selection success Clearly marked for net-next, async
netdev/ynl success Generated files up to date; no warnings/errors; no diff in generated;
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 15605 this patch: 15605
netdev/build_tools success Errors and warnings before: 1 this patch: 0
netdev/cc_maintainers success CCed 0 of 0 maintainers
netdev/build_clang success Errors and warnings before: 3260 this patch: 3260
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 16751 this patch: 16751
netdev/checkpatch success total: 0 errors, 0 warnings, 0 checks, 87 lines checked
netdev/build_clang_rust success No Rust files in patch. Skipping build
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0

Commit Message

Alexander Lobakin Feb. 1, 2024, 12:22 p.m. UTC 
The number of times yet another open coded
`BITS_TO_LONGS(nbits) * sizeof(long)` can be spotted is huge.
Some generic helper is long overdue.

Add one, bitmap_size(), but with one detail.
BITS_TO_LONGS() uses DIV_ROUND_UP(). The latter works well when both
divident and divisor are compile-time constants or when the divisor
is not a pow-of-2. When it is however, the compilers sometimes tend
to generate suboptimal code (GCC 13):

48 83 c0 3f          	add    $0x3f,%rax
48 c1 e8 06          	shr    $0x6,%rax
48 8d 14 c5 00 00 00 00	lea    0x0(,%rax,8),%rdx

%BITS_PER_LONG is always a pow-2 (either 32 or 64), but GCC still does
full division of `nbits + 63` by it and then multiplication by 8.
Instead of BITS_TO_LONGS(), use ALIGN() and then divide by 8. GCC:

8d 50 3f             	lea    0x3f(%rax),%edx
c1 ea 03             	shr    $0x3,%edx
81 e2 f8 ff ff 1f    	and    $0x1ffffff8,%edx

Now it shifts `nbits + 63` by 3 positions (IOW performs fast division
by 8) and then masks bits[2:0]. bloat-o-meter:

add/remove: 0/0 grow/shrink: 20/133 up/down: 156/-773 (-617)

Clang does it better and generates the same code before/after starting
from -O1, except that with the ALIGN() approach it uses %edx and thus
still saves some bytes:

add/remove: 0/0 grow/shrink: 9/133 up/down: 18/-538 (-520)

Note that we can't expand DIV_ROUND_UP() by adding a check and using
this approach there, as it's used in array declarations where
expressions are not allowed.
Add this helper to tools/ as well.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Acked-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 include/linux/bitmap.h         | 8 +++++---
 include/linux/cpumask.h        | 2 +-
 tools/include/linux/bitmap.h   | 7 ++++---
 drivers/md/dm-clone-metadata.c | 5 -----
 drivers/s390/cio/idset.c       | 2 +-
 lib/math/prime_numbers.c       | 2 --
 6 files changed, 11 insertions(+), 15 deletions(-)

Comments

Yury Norov Feb. 28, 2024, 4:31 p.m. UTC | #1 
On Thu, Feb 01, 2024 at 01:22:07PM +0100, Alexander Lobakin wrote:
> The number of times yet another open coded
> `BITS_TO_LONGS(nbits) * sizeof(long)` can be spotted is huge.
> Some generic helper is long overdue.
> 
> Add one, bitmap_size(), but with one detail.
> BITS_TO_LONGS() uses DIV_ROUND_UP(). The latter works well when both
> divident and divisor are compile-time constants or when the divisor
> is not a pow-of-2. When it is however, the compilers sometimes tend
> to generate suboptimal code (GCC 13):
> 
> 48 83 c0 3f          	add    $0x3f,%rax
> 48 c1 e8 06          	shr    $0x6,%rax
> 48 8d 14 c5 00 00 00 00	lea    0x0(,%rax,8),%rdx
> 
> %BITS_PER_LONG is always a pow-2 (either 32 or 64), but GCC still does
> full division of `nbits + 63` by it and then multiplication by 8.
> Instead of BITS_TO_LONGS(), use ALIGN() and then divide by 8. GCC:
> 
> 8d 50 3f             	lea    0x3f(%rax),%edx
> c1 ea 03             	shr    $0x3,%edx
> 81 e2 f8 ff ff 1f    	and    $0x1ffffff8,%edx
> 
> Now it shifts `nbits + 63` by 3 positions (IOW performs fast division
> by 8) and then masks bits[2:0]. bloat-o-meter:
> 
> add/remove: 0/0 grow/shrink: 20/133 up/down: 156/-773 (-617)
> 
> Clang does it better and generates the same code before/after starting
> from -O1, except that with the ALIGN() approach it uses %edx and thus
> still saves some bytes:
> 
> add/remove: 0/0 grow/shrink: 9/133 up/down: 18/-538 (-520)
> 
> Note that we can't expand DIV_ROUND_UP() by adding a check and using
> this approach there, as it's used in array declarations where
> expressions are not allowed.
> Add this helper to tools/ as well.
> 
> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
> Acked-by: Yury Norov <yury.norov@gmail.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>

Signed-off-by: Yury Norov <yury.norov@gmail.com>
diff mbox series

Patch

diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index 7ca0379be8c1..9a6a27a7f675 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -218,9 +218,11 @@  void bitmap_fold(unsigned long *dst, const unsigned long *orig,
 #define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) & (BITS_PER_LONG - 1)))
 #define BITMAP_LAST_WORD_MASK(nbits) (~0UL >> (-(nbits) & (BITS_PER_LONG - 1)))
 
+#define bitmap_size(nbits)	(ALIGN(nbits, BITS_PER_LONG) / BITS_PER_BYTE)
+
 static inline void bitmap_zero(unsigned long *dst, unsigned int nbits)
 {
-	unsigned int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+	unsigned int len = bitmap_size(nbits);
 
 	if (small_const_nbits(nbits))
 		*dst = 0;
@@ -230,7 +232,7 @@  static inline void bitmap_zero(unsigned long *dst, unsigned int nbits)
 
 static inline void bitmap_fill(unsigned long *dst, unsigned int nbits)
 {
-	unsigned int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+	unsigned int len = bitmap_size(nbits);
 
 	if (small_const_nbits(nbits))
 		*dst = ~0UL;
@@ -241,7 +243,7 @@  static inline void bitmap_fill(unsigned long *dst, unsigned int nbits)
 static inline void bitmap_copy(unsigned long *dst, const unsigned long *src,
 			unsigned int nbits)
 {
-	unsigned int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+	unsigned int len = bitmap_size(nbits);
 
 	if (small_const_nbits(nbits))
 		*dst = *src;
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index cfb545841a2c..a2064c2a9441 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -839,7 +839,7 @@  static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
  */
 static inline unsigned int cpumask_size(void)
 {
-	return BITS_TO_LONGS(large_cpumask_bits) * sizeof(long);
+	return bitmap_size(large_cpumask_bits);
 }
 
 /*
diff --git a/tools/include/linux/bitmap.h b/tools/include/linux/bitmap.h
index 8c6852dba04f..210c13b1b857 100644
--- a/tools/include/linux/bitmap.h
+++ b/tools/include/linux/bitmap.h
@@ -26,13 +26,14 @@  bool __bitmap_intersects(const unsigned long *bitmap1,
 #define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) & (BITS_PER_LONG - 1)))
 #define BITMAP_LAST_WORD_MASK(nbits) (~0UL >> (-(nbits) & (BITS_PER_LONG - 1)))
 
+#define bitmap_size(nbits)	(ALIGN(nbits, BITS_PER_LONG) / BITS_PER_BYTE)
+
 static inline void bitmap_zero(unsigned long *dst, unsigned int nbits)
 {
 	if (small_const_nbits(nbits))
 		*dst = 0UL;
 	else {
-		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
-		memset(dst, 0, len);
+		memset(dst, 0, bitmap_size(nbits));
 	}
 }
 
@@ -84,7 +85,7 @@  static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
  */
 static inline unsigned long *bitmap_zalloc(int nbits)
 {
-	return calloc(1, BITS_TO_LONGS(nbits) * sizeof(unsigned long));
+	return calloc(1, bitmap_size(nbits));
 }
 
 /*
diff --git a/drivers/md/dm-clone-metadata.c b/drivers/md/dm-clone-metadata.c
index c43d55672bce..47c1fa7aad8b 100644
--- a/drivers/md/dm-clone-metadata.c
+++ b/drivers/md/dm-clone-metadata.c
@@ -465,11 +465,6 @@  static void __destroy_persistent_data_structures(struct dm_clone_metadata *cmd)
 
 /*---------------------------------------------------------------------------*/
 
-static size_t bitmap_size(unsigned long nr_bits)
-{
-	return BITS_TO_LONGS(nr_bits) * sizeof(long);
-}
-
 static int __dirty_map_init(struct dirty_map *dmap, unsigned long nr_words,
 			    unsigned long nr_regions)
 {
diff --git a/drivers/s390/cio/idset.c b/drivers/s390/cio/idset.c
index 0a1105a483bf..e5f28370a903 100644
--- a/drivers/s390/cio/idset.c
+++ b/drivers/s390/cio/idset.c
@@ -18,7 +18,7 @@  struct idset {
 
 static inline unsigned long idset_bitmap_size(int num_ssid, int num_id)
 {
-	return BITS_TO_LONGS(num_ssid * num_id) * sizeof(unsigned long);
+	return bitmap_size(size_mul(num_ssid, num_id));
 }
 
 static struct idset *idset_new(int num_ssid, int num_id)
diff --git a/lib/math/prime_numbers.c b/lib/math/prime_numbers.c
index d42cebf7407f..d3b64b10da1c 100644
--- a/lib/math/prime_numbers.c
+++ b/lib/math/prime_numbers.c
@@ -6,8 +6,6 @@ 
 #include <linux/prime_numbers.h>
 #include <linux/slab.h>
 
-#define bitmap_size(nbits) (BITS_TO_LONGS(nbits) * sizeof(unsigned long))
-
 struct primes {
 	struct rcu_head rcu;
 	unsigned long last, sz;