2019-05-27 08:55:01 +02:00
/* SPDX-License-Identifier: GPL-2.0-or-later */
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
/*
* PowerPC atomic bit operations .
*
* Merged version by David Gibson < david @ gibson . dropbear . id . au > .
* Based on ppc64 versions by : Dave Engebretsen , Todd Inglett , Don
* Reed , Pat McCarthy , Peter Bergner , Anton Blanchard . They
* originally took it from the ppc32 code .
*
* Within a word , bits are numbered LSB first . Lot ' s of places make
* this assumption by directly testing bits with ( val & ( 1 < < nr ) ) .
* This can cause confusion for large ( > 1 word ) bitmaps on a
* big - endian system because , unlike little endian , the number of each
* bit depends on the word size .
*
* The bitop functions are defined to work on unsigned longs , so for a
* ppc64 system the bits end up numbered :
2014-11-11 12:50:22 +08:00
* | 63. . . . . . . . . . . . . .0 | 127. . . . . . . . . . . .64 | 191. . . . . . . . . . .128 | 255. . . . . . . . . . .192 |
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
* and on ppc32 :
2014-11-11 12:50:22 +08:00
* | 31. . . . .0 | 63. . . .32 | 95. . . .64 | 127. . .96 | 159. .128 | 191. .160 | 223. .192 | 255. .224 |
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
*
* There are a few little - endian macros used mostly for filesystem
* bitmaps , these work on similar bit arrays layouts , but
* byte - oriented :
* | 7. . .0 | 15. . .8 | 23. . .16 | 31. . .24 | 39. . .32 | 47. . .40 | 55. . .48 | 63. . .56 |
*
* The main difference is that bit 3 - 5 ( 64 b ) or 3 - 4 ( 32 b ) in the bit
* number field needs to be reversed compared to the big - endian bit
* fields . This can be achieved by XOR with 0x38 ( 64 b ) or 0x18 ( 32 b ) .
*/
# ifndef _ASM_POWERPC_BITOPS_H
# define _ASM_POWERPC_BITOPS_H
# ifdef __KERNEL__
2007-10-18 23:40:26 -07:00
# ifndef _LINUX_BITOPS_H
# error only <linux / bitops.h> can be included directly
# endif
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
# include <linux/compiler.h>
[PATCH] powerpc: Consolidate asm compatibility macros
This patch consolidates macros used to generate assembly for
compatibility across different CPUs or configs. A new header,
asm-powerpc/asm-compat.h contains the main compatibility macros. It
uses some preprocessor magic to make the macros suitable both for use
in .S files, and in inline asm in .c files. Headers (bitops.h,
uaccess.h, atomic.h, bug.h) which had their own such compatibility
macros are changed to use asm-compat.h.
ppc_asm.h is now for use in .S files *only*, and a #error enforces
that. As such, we're a lot more careless about namespace pollution
here than in asm-compat.h.
While we're at it, this patch adds a call to the PPC405_ERR77 macro in
futex.h which should have had it already, but didn't.
Built and booted on pSeries, Maple and iSeries (ARCH=powerpc). Built
for 32-bit powermac (ARCH=powerpc) and Walnut (ARCH=ppc).
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-10 12:56:55 +11:00
# include <asm/asm-compat.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
# include <asm/synch.h>
2013-10-30 20:05:11 +05:30
/* PPC bit number conversion */
# define PPC_BITLSHIFT(be) (BITS_PER_LONG - 1 - (be))
# define PPC_BIT(bit) (1UL << PPC_BITLSHIFT(bit))
# define PPC_BITMASK(bs, be) ((PPC_BIT(bs) - PPC_BIT(be)) | PPC_BIT(bs))
2017-02-28 12:00:48 +10:00
/* Put a PPC bit into a "normal" bit position */
# define PPC_BITEXTRACT(bits, ppc_bit, dst_bit) \
( ( ( ( bits ) > > PPC_BITLSHIFT ( ppc_bit ) ) & 1 ) < < ( dst_bit ) )
2017-04-05 17:54:47 +10:00
# define PPC_BITLSHIFT32(be) (32 - 1 - (be))
# define PPC_BIT32(bit) (1UL << PPC_BITLSHIFT32(bit))
# define PPC_BITMASK32(bs, be) ((PPC_BIT32(bs) - PPC_BIT32(be))|PPC_BIT32(bs))
# define PPC_BITLSHIFT8(be) (8 - 1 - (be))
# define PPC_BIT8(bit) (1UL << PPC_BITLSHIFT8(bit))
# define PPC_BITMASK8(bs, be) ((PPC_BIT8(bs) - PPC_BIT8(be))|PPC_BIT8(bs))
2014-03-13 19:00:35 +01:00
# include <asm/barrier.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
2009-07-07 15:23:56 +00:00
/* Macro for generating the ***_bits() functions */
2013-02-21 17:25:41 +00:00
# define DEFINE_BITOP(fn, op, prefix) \
2019-08-20 12:49:41 +10:00
static inline void fn ( unsigned long mask , \
2009-07-07 15:23:56 +00:00
volatile unsigned long * _p ) \
{ \
unsigned long old ; \
unsigned long * p = ( unsigned long * ) _p ; \
__asm__ __volatile__ ( \
prefix \
2021-03-02 08:48:11 +00:00
" 1: " PPC_LLARX " %0,0,%3,0 \n " \
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
# op "%I2 %0,%0,%2\n" \
2009-07-07 15:23:56 +00:00
PPC_STLCX " %0,0,%3 \n " \
" bne- 1b \n " \
: " =&r " ( old ) , " +m " ( * p ) \
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
: " rK " ( mask ) , " r " ( p ) \
2009-07-07 15:23:56 +00:00
: " cc " , " memory " ) ; \
}
2013-02-21 17:25:41 +00:00
DEFINE_BITOP ( set_bits , or , " " )
DEFINE_BITOP ( change_bits , xor , " " )
2009-07-07 15:23:56 +00:00
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
static __always_inline bool is_rlwinm_mask_valid ( unsigned long x )
{
if ( ! x )
return false ;
if ( x & 1 )
x = ~ x ; // make the mask non-wrapping
x + = x & - x ; // adding the low set bit results in at most one bit set
return ! ( x & ( x - 1 ) ) ;
}
# define DEFINE_CLROP(fn, prefix) \
static inline void fn ( unsigned long mask , volatile unsigned long * _p ) \
{ \
unsigned long old ; \
unsigned long * p = ( unsigned long * ) _p ; \
\
if ( IS_ENABLED ( CONFIG_PPC32 ) & & \
__builtin_constant_p ( mask ) & & is_rlwinm_mask_valid ( ~ mask ) ) { \
asm volatile ( \
prefix \
" 1: " " lwarx %0,0,%3 \n " \
" rlwinm %0,%0,0,%2 \n " \
" stwcx. %0,0,%3 \n " \
" bne- 1b \n " \
: " =&r " ( old ) , " +m " ( * p ) \
: " n " ( ~ mask ) , " r " ( p ) \
: " cc " , " memory " ) ; \
} else { \
asm volatile ( \
prefix \
" 1: " PPC_LLARX " %0,0,%3,0 \n " \
" andc %0,%0,%2 \n " \
PPC_STLCX " %0,0,%3 \n " \
" bne- 1b \n " \
: " =&r " ( old ) , " +m " ( * p ) \
: " r " ( mask ) , " r " ( p ) \
: " cc " , " memory " ) ; \
} \
}
DEFINE_CLROP ( clear_bits , " " )
DEFINE_CLROP ( clear_bits_unlock , PPC_RELEASE_BARRIER )
2019-08-20 12:49:41 +10:00
static inline void arch_set_bit ( int nr , volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
set_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2019-08-20 12:49:41 +10:00
static inline void arch_clear_bit ( int nr , volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
clear_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2019-08-20 12:49:41 +10:00
static inline void arch_clear_bit_unlock ( int nr , volatile unsigned long * addr )
2007-10-18 03:06:53 -07:00
{
2012-11-04 02:03:44 +00:00
clear_bits_unlock ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ;
2007-10-18 03:06:53 -07:00
}
2019-08-20 12:49:41 +10:00
static inline void arch_change_bit ( int nr , volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
change_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ;
2009-07-07 15:23:56 +00:00
}
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
2009-07-07 15:23:56 +00:00
/* Like DEFINE_BITOP(), with changes to the arguments to 'op' and the output
* operands . */
2010-02-10 01:02:36 +00:00
# define DEFINE_TESTOP(fn, op, prefix, postfix, eh) \
2019-08-20 12:49:41 +10:00
static inline unsigned long fn ( \
2010-02-10 01:02:36 +00:00
unsigned long mask , \
volatile unsigned long * _p ) \
{ \
unsigned long old , t ; \
unsigned long * p = ( unsigned long * ) _p ; \
__asm__ __volatile__ ( \
prefix \
2021-03-02 08:48:11 +00:00
" 1: " PPC_LLARX " %0,0,%3,%4 \n " \
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
# op "%I2 %1,%0,%2\n" \
2010-02-10 01:02:36 +00:00
PPC_STLCX " %1,0,%3 \n " \
" bne- 1b \n " \
postfix \
: " =&r " ( old ) , " =&r " ( t ) \
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
: " rK " ( mask ) , " r " ( p ) , " i " ( IS_ENABLED ( CONFIG_PPC64 ) ? eh : 0 ) \
2010-02-10 01:02:36 +00:00
: " cc " , " memory " ) ; \
return ( old & mask ) ; \
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2011-11-15 17:11:27 +00:00
DEFINE_TESTOP ( test_and_set_bits , or , PPC_ATOMIC_ENTRY_BARRIER ,
PPC_ATOMIC_EXIT_BARRIER , 0 )
2010-02-10 01:04:06 +00:00
DEFINE_TESTOP ( test_and_set_bits_lock , or , " " ,
PPC_ACQUIRE_BARRIER , 1 )
2011-11-15 17:11:27 +00:00
DEFINE_TESTOP ( test_and_change_bits , xor , PPC_ATOMIC_ENTRY_BARRIER ,
PPC_ATOMIC_EXIT_BARRIER , 0 )
2009-07-07 15:23:56 +00:00
powerpc/bitops: Use immediate operand when possible
Today we get the following code generation for bitops like
set or clear bit:
c0009fe0: 39 40 08 00 li r10,2048
c0009fe4: 7c e0 40 28 lwarx r7,0,r8
c0009fe8: 7c e7 53 78 or r7,r7,r10
c0009fec: 7c e0 41 2d stwcx. r7,0,r8
c000d568: 39 00 18 00 li r8,6144
c000d56c: 7c c0 38 28 lwarx r6,0,r7
c000d570: 7c c6 40 78 andc r6,r6,r8
c000d574: 7c c0 39 2d stwcx. r6,0,r7
Most set bits are constant on lower 16 bits, so it can easily
be replaced by the "immediate" version of the operation. Allow
GCC to choose between the normal or immediate form.
For clear bits, on 32 bits 'rlwinm' can be used instead of 'andc' for
when all bits to be cleared are consecutive.
On 64 bits we don't have any equivalent single operation for clearing,
single bits or a few bits, we'd need two 'rldicl' so it is not
worth it, the li/andc sequence is doing the same.
With this patch we get:
c0009fe0: 7d 00 50 28 lwarx r8,0,r10
c0009fe4: 61 08 08 00 ori r8,r8,2048
c0009fe8: 7d 00 51 2d stwcx. r8,0,r10
c000d558: 7c e0 40 28 lwarx r7,0,r8
c000d55c: 54 e7 05 64 rlwinm r7,r7,0,21,18
c000d560: 7c e0 41 2d stwcx. r7,0,r8
On pmac32_defconfig, it reduces the text by approx 10 kbytes.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/e6f815d9181bab09df3b350af51149437863e9f9.1632236981.git.christophe.leroy@csgroup.eu
2021-09-21 17:09:47 +02:00
static inline unsigned long test_and_clear_bits ( unsigned long mask , volatile unsigned long * _p )
{
unsigned long old , t ;
unsigned long * p = ( unsigned long * ) _p ;
if ( IS_ENABLED ( CONFIG_PPC32 ) & &
__builtin_constant_p ( mask ) & & is_rlwinm_mask_valid ( ~ mask ) ) {
asm volatile (
PPC_ATOMIC_ENTRY_BARRIER
" 1: " " lwarx %0,0,%3 \n "
" rlwinm %1,%0,0,%2 \n "
" stwcx. %1,0,%3 \n "
" bne- 1b \n "
PPC_ATOMIC_EXIT_BARRIER
: " =&r " ( old ) , " =&r " ( t )
: " n " ( ~ mask ) , " r " ( p )
: " cc " , " memory " ) ;
} else {
asm volatile (
PPC_ATOMIC_ENTRY_BARRIER
" 1: " PPC_LLARX " %0,0,%3,0 \n "
" andc %1,%0,%2 \n "
PPC_STLCX " %1,0,%3 \n "
" bne- 1b \n "
PPC_ATOMIC_EXIT_BARRIER
: " =&r " ( old ) , " =&r " ( t )
: " r " ( mask ) , " r " ( p )
: " cc " , " memory " ) ;
}
return ( old & mask ) ;
}
2019-08-20 12:49:41 +10:00
static inline int arch_test_and_set_bit ( unsigned long nr ,
volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
return test_and_set_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ! = 0 ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2019-08-20 12:49:41 +10:00
static inline int arch_test_and_set_bit_lock ( unsigned long nr ,
volatile unsigned long * addr )
2007-10-18 03:06:53 -07:00
{
2012-11-04 02:03:44 +00:00
return test_and_set_bits_lock ( BIT_MASK ( nr ) ,
addr + BIT_WORD ( nr ) ) ! = 0 ;
2007-10-18 03:06:53 -07:00
}
2019-08-20 12:49:41 +10:00
static inline int arch_test_and_clear_bit ( unsigned long nr ,
volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
return test_and_clear_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ! = 0 ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2019-08-20 12:49:41 +10:00
static inline int arch_test_and_change_bit ( unsigned long nr ,
volatile unsigned long * addr )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
2012-11-04 02:03:44 +00:00
return test_and_change_bits ( BIT_MASK ( nr ) , addr + BIT_WORD ( nr ) ) ! = 0 ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
powerpc/64: Implement clear_bit_unlock_is_negative_byte()
Commit b91e1302ad9b8 ("mm: optimize PageWaiters bit use for
unlock_page()") added a special bitop function to speed up
unlock_page(). Implement this for 64-bit powerpc.
This improves the unlock_page() core code from this:
li 9,1
lwsync
1: ldarx 10,0,3,0
andc 10,10,9
stdcx. 10,0,3
bne- 1b
ori 2,2,0
ld 9,0(3)
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
To this:
li 10,1
lwsync
1: ldarx 9,0,3,0
andc 9,9,10
stdcx. 9,0,3
bne- 1b
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
In a test of elapsed time for dd writing into 16GB of already-dirty
pagecache on a POWER8 with 4K pages, which has one unlock_page per 4kB
this patch reduced overhead by 1.1%:
N Min Max Median Avg Stddev
x 19 2.578 2.619 2.594 2.595 0.011
+ 19 2.552 2.592 2.564 2.565 0.008
Difference at 95.0% confidence
-0.030 +/- 0.006
-1.142% +/- 0.243%
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Made 64-bit only until I can test it properly on 32-bit]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-01-04 04:58:28 +10:00
# ifdef CONFIG_PPC64
2019-08-20 12:49:41 +10:00
static inline unsigned long
clear_bit_unlock_return_word ( int nr , volatile unsigned long * addr )
powerpc/64: Implement clear_bit_unlock_is_negative_byte()
Commit b91e1302ad9b8 ("mm: optimize PageWaiters bit use for
unlock_page()") added a special bitop function to speed up
unlock_page(). Implement this for 64-bit powerpc.
This improves the unlock_page() core code from this:
li 9,1
lwsync
1: ldarx 10,0,3,0
andc 10,10,9
stdcx. 10,0,3
bne- 1b
ori 2,2,0
ld 9,0(3)
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
To this:
li 10,1
lwsync
1: ldarx 9,0,3,0
andc 9,9,10
stdcx. 9,0,3
bne- 1b
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
In a test of elapsed time for dd writing into 16GB of already-dirty
pagecache on a POWER8 with 4K pages, which has one unlock_page per 4kB
this patch reduced overhead by 1.1%:
N Min Max Median Avg Stddev
x 19 2.578 2.619 2.594 2.595 0.011
+ 19 2.552 2.592 2.564 2.565 0.008
Difference at 95.0% confidence
-0.030 +/- 0.006
-1.142% +/- 0.243%
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Made 64-bit only until I can test it properly on 32-bit]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-01-04 04:58:28 +10:00
{
unsigned long old , t ;
unsigned long * p = ( unsigned long * ) addr + BIT_WORD ( nr ) ;
unsigned long mask = BIT_MASK ( nr ) ;
__asm__ __volatile__ (
PPC_RELEASE_BARRIER
2021-03-02 08:48:11 +00:00
" 1: " PPC_LLARX " %0,0,%3,0 \n "
powerpc/64: Implement clear_bit_unlock_is_negative_byte()
Commit b91e1302ad9b8 ("mm: optimize PageWaiters bit use for
unlock_page()") added a special bitop function to speed up
unlock_page(). Implement this for 64-bit powerpc.
This improves the unlock_page() core code from this:
li 9,1
lwsync
1: ldarx 10,0,3,0
andc 10,10,9
stdcx. 10,0,3
bne- 1b
ori 2,2,0
ld 9,0(3)
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
To this:
li 10,1
lwsync
1: ldarx 9,0,3,0
andc 9,9,10
stdcx. 9,0,3
bne- 1b
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
In a test of elapsed time for dd writing into 16GB of already-dirty
pagecache on a POWER8 with 4K pages, which has one unlock_page per 4kB
this patch reduced overhead by 1.1%:
N Min Max Median Avg Stddev
x 19 2.578 2.619 2.594 2.595 0.011
+ 19 2.552 2.592 2.564 2.565 0.008
Difference at 95.0% confidence
-0.030 +/- 0.006
-1.142% +/- 0.243%
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Made 64-bit only until I can test it properly on 32-bit]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-01-04 04:58:28 +10:00
" andc %1,%0,%2 \n "
PPC_STLCX " %1,0,%3 \n "
" bne- 1b \n "
: " =&r " ( old ) , " =&r " ( t )
: " r " ( mask ) , " r " ( p )
: " cc " , " memory " ) ;
return old ;
}
2019-08-20 12:49:41 +10:00
/*
* This is a special function for mm / filemap . c
* Bit 7 corresponds to PG_waiters .
*/
# define arch_clear_bit_unlock_is_negative_byte(nr, addr) \
( clear_bit_unlock_return_word ( nr , addr ) & BIT_MASK ( 7 ) )
powerpc/64: Implement clear_bit_unlock_is_negative_byte()
Commit b91e1302ad9b8 ("mm: optimize PageWaiters bit use for
unlock_page()") added a special bitop function to speed up
unlock_page(). Implement this for 64-bit powerpc.
This improves the unlock_page() core code from this:
li 9,1
lwsync
1: ldarx 10,0,3,0
andc 10,10,9
stdcx. 10,0,3
bne- 1b
ori 2,2,0
ld 9,0(3)
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
To this:
li 10,1
lwsync
1: ldarx 9,0,3,0
andc 9,9,10
stdcx. 9,0,3
bne- 1b
andi. 10,9,0x80
beqlr
li 4,0
b wake_up_page_bit
In a test of elapsed time for dd writing into 16GB of already-dirty
pagecache on a POWER8 with 4K pages, which has one unlock_page per 4kB
this patch reduced overhead by 1.1%:
N Min Max Median Avg Stddev
x 19 2.578 2.619 2.594 2.595 0.011
+ 19 2.552 2.592 2.564 2.565 0.008
Difference at 95.0% confidence
-0.030 +/- 0.006
-1.142% +/- 0.243%
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Made 64-bit only until I can test it properly on 32-bit]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-01-04 04:58:28 +10:00
# endif /* CONFIG_PPC64 */
2006-03-26 01:39:33 -08:00
# include <asm-generic/bitops/non-atomic.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
2019-08-20 12:49:41 +10:00
static inline void arch___clear_bit_unlock ( int nr , volatile unsigned long * addr )
2007-10-18 03:06:53 -07:00
{
2010-02-10 01:04:06 +00:00
__asm__ __volatile__ ( PPC_RELEASE_BARRIER " " : : : " memory " ) ;
2007-10-18 03:06:53 -07:00
__clear_bit ( nr , addr ) ;
}
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
/*
* Return the zero - based bit position ( LE , not IBM bit numbering ) of
* the most significant 1 - bit in a double word .
*/
powerpc: Remove __ilog2()s and use generic ones
With the __ilog2() function as defined in
arch/powerpc/include/asm/bitops.h, GCC will not optimise the code
in case of constant parameter.
The generic ilog2() function in include/linux/log2.h is written
to handle the case of the constant parameter.
This patch discards the three __ilog2() functions and
defines __ilog2() as ilog2()
For non constant calls, the generated code is doing the same:
int test__ilog2(unsigned long x)
{
return __ilog2(x);
}
int test__ilog2_u32(u32 n)
{
return __ilog2_u32(n);
}
int test__ilog2_u64(u64 n)
{
return __ilog2_u64(n);
}
On PPC32 before the patch:
00000000 <test__ilog2>:
0: 7c 63 00 34 cntlzw r3,r3
4: 20 63 00 1f subfic r3,r3,31
8: 4e 80 00 20 blr
0000000c <test__ilog2_u32>:
c: 7c 63 00 34 cntlzw r3,r3
10: 20 63 00 1f subfic r3,r3,31
14: 4e 80 00 20 blr
On PPC32 after the patch:
00000000 <test__ilog2>:
0: 7c 63 00 34 cntlzw r3,r3
4: 20 63 00 1f subfic r3,r3,31
8: 4e 80 00 20 blr
0000000c <test__ilog2_u32>:
c: 7c 63 00 34 cntlzw r3,r3
10: 20 63 00 1f subfic r3,r3,31
14: 4e 80 00 20 blr
On PPC64 before the patch:
0000000000000000 <.test__ilog2>:
0: 7c 63 00 74 cntlzd r3,r3
4: 20 63 00 3f subfic r3,r3,63
8: 7c 63 07 b4 extsw r3,r3
c: 4e 80 00 20 blr
0000000000000010 <.test__ilog2_u32>:
10: 7c 63 00 34 cntlzw r3,r3
14: 20 63 00 1f subfic r3,r3,31
18: 7c 63 07 b4 extsw r3,r3
1c: 4e 80 00 20 blr
0000000000000020 <.test__ilog2_u64>:
20: 7c 63 00 74 cntlzd r3,r3
24: 20 63 00 3f subfic r3,r3,63
28: 7c 63 07 b4 extsw r3,r3
2c: 4e 80 00 20 blr
On PPC64 after the patch:
0000000000000000 <.test__ilog2>:
0: 7c 63 00 74 cntlzd r3,r3
4: 20 63 00 3f subfic r3,r3,63
8: 7c 63 07 b4 extsw r3,r3
c: 4e 80 00 20 blr
0000000000000010 <.test__ilog2_u32>:
10: 7c 63 00 34 cntlzw r3,r3
14: 20 63 00 1f subfic r3,r3,31
18: 7c 63 07 b4 extsw r3,r3
1c: 4e 80 00 20 blr
0000000000000020 <.test__ilog2_u64>:
20: 7c 63 00 74 cntlzd r3,r3
24: 20 63 00 3f subfic r3,r3,63
28: 7c 63 07 b4 extsw r3,r3
2c: 4e 80 00 20 blr
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-21 13:18:52 +02:00
# define __ilog2(x) ilog2(x)
2006-12-08 02:37:53 -08:00
powerpc: Replace ffz() by equivalent generic function
With the ffz() function as defined in arch/powerpc/include/asm/bitops.h
GCC will not optimise the code in case of constant parameter.
This patch replaces ffz() by the generic function.
The generic ffz(x) expects to never be called with ~x == 0
as written in the comment in include/asm-generic/bitops/ffz.h
The only user of ffz() within arch/powerpc/ is
platforms/512x/mpc5121_ads_cpld.c, which checks if x is not 0xff
For non constant calls, the generated code is doing the same:
unsigned long testffz(unsigned long x)
{
return ffz(x);
}
On PPC32, before the patch:
00000018 <testffz>:
18: 7c 63 18 f9 not. r3,r3
1c: 40 82 00 0c bne 28 <testffz+0x10>
20: 38 60 00 20 li r3,32
24: 4e 80 00 20 blr
28: 7d 23 00 d0 neg r9,r3
2c: 7d 23 18 38 and r3,r9,r3
30: 7c 63 00 34 cntlzw r3,r3
34: 20 63 00 1f subfic r3,r3,31
38: 4e 80 00 20 blr
On PPC32, after the patch:
00000018 <testffz>:
18: 39 23 00 01 addi r9,r3,1
1c: 7d 23 18 78 andc r3,r9,r3
20: 7c 63 00 34 cntlzw r3,r3
24: 20 63 00 1f subfic r3,r3,31
28: 4e 80 00 20 blr
On PPC64, before the patch:
0000000000000030 <.testffz>:
30: 7c 60 18 f9 not. r0,r3
34: 38 60 00 40 li r3,64
38: 4d 82 00 20 beqlr
3c: 7c 60 00 d0 neg r3,r0
40: 7c 63 00 38 and r3,r3,r0
44: 7c 63 00 74 cntlzd r3,r3
48: 20 63 00 3f subfic r3,r3,63
4c: 7c 63 07 b4 extsw r3,r3
50: 4e 80 00 20 blr
On PPC64, after the patch:
0000000000000030 <.testffz>:
30: 38 03 00 01 addi r0,r3,1
34: 7c 03 18 78 andc r3,r0,r3
38: 7c 63 00 74 cntlzd r3,r3
3c: 20 63 00 3f subfic r3,r3,63
40: 4e 80 00 20 blr
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-21 13:18:50 +02:00
# include <asm-generic/bitops/ffz.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
powerpc: Discard ffs()/__ffs() function and use builtin functions instead
With the ffs() function as defined in arch/powerpc/include/asm/bitops.h
GCC will not optimise the code in case of constant parameter, as shown
by the small exemple below.
int ffs_test(void)
{
return 4 << ffs(31);
}
c0012334 <ffs_test>:
c0012334: 39 20 00 01 li r9,1
c0012338: 38 60 00 04 li r3,4
c001233c: 7d 29 00 34 cntlzw r9,r9
c0012340: 21 29 00 20 subfic r9,r9,32
c0012344: 7c 63 48 30 slw r3,r3,r9
c0012348: 4e 80 00 20 blr
With this patch, the same function will compile as follows:
c0012334 <ffs_test>:
c0012334: 38 60 00 08 li r3,8
c0012338: 4e 80 00 20 blr
The same happens with __ffs()
For non constant calls, the generated code is doing the same,
allthought it is slightly different on 64 bits for ffs():
unsigned long test__ffs(unsigned long x)
{
return __ffs(x);
}
int testffs(int x)
{
return ffs(x);
}
On PPC32, before the patch:
0000003c <test__ffs>:
3c: 7d 23 00 d0 neg r9,r3
40: 7d 23 18 38 and r3,r9,r3
44: 7c 63 00 34 cntlzw r3,r3
48: 20 63 00 1f subfic r3,r3,31
4c: 4e 80 00 20 blr
00000050 <testffs>:
50: 7d 23 00 d0 neg r9,r3
54: 7d 23 18 38 and r3,r9,r3
58: 7c 63 00 34 cntlzw r3,r3
5c: 20 63 00 20 subfic r3,r3,32
60: 4e 80 00 20 blr
On PPC32, after the patch:
0000002c <test__ffs>:
2c: 7d 23 00 d0 neg r9,r3
30: 7d 23 18 38 and r3,r9,r3
34: 7c 63 00 34 cntlzw r3,r3
38: 20 63 00 1f subfic r3,r3,31
3c: 4e 80 00 20 blr
00000040 <testffs>:
40: 7d 23 00 d0 neg r9,r3
44: 7d 23 18 38 and r3,r9,r3
48: 7c 63 00 34 cntlzw r3,r3
4c: 20 63 00 20 subfic r3,r3,32
50: 4e 80 00 20 blr
On PPC64, before the patch:
0000000000000060 <.test__ffs>:
60: 7c 03 00 d0 neg r0,r3
64: 7c 03 18 38 and r3,r0,r3
68: 7c 63 00 74 cntlzd r3,r3
6c: 20 63 00 3f subfic r3,r3,63
70: 7c 63 07 b4 extsw r3,r3
74: 4e 80 00 20 blr
0000000000000080 <.testffs>:
80: 7c 03 00 d0 neg r0,r3
84: 7c 03 18 38 and r3,r0,r3
88: 7c 63 00 74 cntlzd r3,r3
8c: 20 63 00 40 subfic r3,r3,64
90: 7c 63 07 b4 extsw r3,r3
94: 4e 80 00 20 blr
On PPC64, after the patch:
0000000000000050 <.test__ffs>:
50: 7c 03 00 d0 neg r0,r3
54: 7c 03 18 38 and r3,r0,r3
58: 7c 63 00 74 cntlzd r3,r3
5c: 20 63 00 3f subfic r3,r3,63
60: 4e 80 00 20 blr
0000000000000070 <.testffs>:
70: 7c 03 00 d0 neg r0,r3
74: 7c 03 18 38 and r3,r0,r3
78: 7c 63 00 34 cntlzw r3,r3
7c: 20 63 00 20 subfic r3,r3,32
80: 7c 63 07 b4 extsw r3,r3
84: 4e 80 00 20 blr
(ffs() operates on an int so cntlzw is equivalent to cntlzd)
In addition, when reading the generated vmlinux, we can observe
that with the builtin functions, GCC sometimes efficiently spreads
the instructions within the generated functions while the inline
assembly force them to remain grouped together.
__builtin_ffs() is already used in arch/powerpc/include/asm/page_32.h
Those builtins have been in GCC since at least 3.4.6 (see
https://gcc.gnu.org/onlinedocs/gcc-3.4.6/gcc/Other-Builtins.html )
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-21 13:18:46 +02:00
# include <asm-generic/bitops/builtin-__ffs.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
powerpc: Discard ffs()/__ffs() function and use builtin functions instead
With the ffs() function as defined in arch/powerpc/include/asm/bitops.h
GCC will not optimise the code in case of constant parameter, as shown
by the small exemple below.
int ffs_test(void)
{
return 4 << ffs(31);
}
c0012334 <ffs_test>:
c0012334: 39 20 00 01 li r9,1
c0012338: 38 60 00 04 li r3,4
c001233c: 7d 29 00 34 cntlzw r9,r9
c0012340: 21 29 00 20 subfic r9,r9,32
c0012344: 7c 63 48 30 slw r3,r3,r9
c0012348: 4e 80 00 20 blr
With this patch, the same function will compile as follows:
c0012334 <ffs_test>:
c0012334: 38 60 00 08 li r3,8
c0012338: 4e 80 00 20 blr
The same happens with __ffs()
For non constant calls, the generated code is doing the same,
allthought it is slightly different on 64 bits for ffs():
unsigned long test__ffs(unsigned long x)
{
return __ffs(x);
}
int testffs(int x)
{
return ffs(x);
}
On PPC32, before the patch:
0000003c <test__ffs>:
3c: 7d 23 00 d0 neg r9,r3
40: 7d 23 18 38 and r3,r9,r3
44: 7c 63 00 34 cntlzw r3,r3
48: 20 63 00 1f subfic r3,r3,31
4c: 4e 80 00 20 blr
00000050 <testffs>:
50: 7d 23 00 d0 neg r9,r3
54: 7d 23 18 38 and r3,r9,r3
58: 7c 63 00 34 cntlzw r3,r3
5c: 20 63 00 20 subfic r3,r3,32
60: 4e 80 00 20 blr
On PPC32, after the patch:
0000002c <test__ffs>:
2c: 7d 23 00 d0 neg r9,r3
30: 7d 23 18 38 and r3,r9,r3
34: 7c 63 00 34 cntlzw r3,r3
38: 20 63 00 1f subfic r3,r3,31
3c: 4e 80 00 20 blr
00000040 <testffs>:
40: 7d 23 00 d0 neg r9,r3
44: 7d 23 18 38 and r3,r9,r3
48: 7c 63 00 34 cntlzw r3,r3
4c: 20 63 00 20 subfic r3,r3,32
50: 4e 80 00 20 blr
On PPC64, before the patch:
0000000000000060 <.test__ffs>:
60: 7c 03 00 d0 neg r0,r3
64: 7c 03 18 38 and r3,r0,r3
68: 7c 63 00 74 cntlzd r3,r3
6c: 20 63 00 3f subfic r3,r3,63
70: 7c 63 07 b4 extsw r3,r3
74: 4e 80 00 20 blr
0000000000000080 <.testffs>:
80: 7c 03 00 d0 neg r0,r3
84: 7c 03 18 38 and r3,r0,r3
88: 7c 63 00 74 cntlzd r3,r3
8c: 20 63 00 40 subfic r3,r3,64
90: 7c 63 07 b4 extsw r3,r3
94: 4e 80 00 20 blr
On PPC64, after the patch:
0000000000000050 <.test__ffs>:
50: 7c 03 00 d0 neg r0,r3
54: 7c 03 18 38 and r3,r0,r3
58: 7c 63 00 74 cntlzd r3,r3
5c: 20 63 00 3f subfic r3,r3,63
60: 4e 80 00 20 blr
0000000000000070 <.testffs>:
70: 7c 03 00 d0 neg r0,r3
74: 7c 03 18 38 and r3,r0,r3
78: 7c 63 00 34 cntlzw r3,r3
7c: 20 63 00 20 subfic r3,r3,32
80: 7c 63 07 b4 extsw r3,r3
84: 4e 80 00 20 blr
(ffs() operates on an int so cntlzw is equivalent to cntlzd)
In addition, when reading the generated vmlinux, we can observe
that with the builtin functions, GCC sometimes efficiently spreads
the instructions within the generated functions while the inline
assembly force them to remain grouped together.
__builtin_ffs() is already used in arch/powerpc/include/asm/page_32.h
Those builtins have been in GCC since at least 3.4.6 (see
https://gcc.gnu.org/onlinedocs/gcc-3.4.6/gcc/Other-Builtins.html )
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-21 13:18:46 +02:00
# include <asm-generic/bitops/builtin-ffs.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
/*
* fls : find last ( most - significant ) bit set .
* Note fls ( 0 ) = 0 , fls ( 1 ) = 1 , fls ( 0x80000000 ) = 32.
*/
powerpc/bitops: Force inlining of fls()
Building a kernel with CONFIG_CC_OPTIMISE_FOR_SIZE leads to
the following functions being copied several times in vmlinux:
31 times __ilog2_u32()
34 times fls()
Disassembly follows:
c00f476c <fls>:
c00f476c: 7c 63 00 34 cntlzw r3,r3
c00f4770: 20 63 00 20 subfic r3,r3,32
c00f4774: 4e 80 00 20 blr
c00f4778 <__ilog2_u32>:
c00f4778: 94 21 ff f0 stwu r1,-16(r1)
c00f477c: 7c 08 02 a6 mflr r0
c00f4780: 90 01 00 14 stw r0,20(r1)
c00f4784: 4b ff ff e9 bl c00f476c <fls>
c00f4788: 80 01 00 14 lwz r0,20(r1)
c00f478c: 38 63 ff ff addi r3,r3,-1
c00f4790: 7c 08 03 a6 mtlr r0
c00f4794: 38 21 00 10 addi r1,r1,16
c00f4798: 4e 80 00 20 blr
When forcing inlining of fls(), we get
c0008b80 <__ilog2_u32>:
c0008b80: 7c 63 00 34 cntlzw r3,r3
c0008b84: 20 63 00 1f subfic r3,r3,31
c0008b88: 4e 80 00 20 blr
vmlinux size gets reduced by 1 kbyte with that change.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/adc9c9d6378f6b5008246ca717993d7870188efb.1644569473.git.christophe.leroy@csgroup.eu
2022-02-11 09:51:32 +01:00
static __always_inline int fls ( unsigned int x )
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
{
powerpc/bitops: Fix possible undefined behaviour with fls() and fls64()
fls() and fls64() are using __builtin_ctz() and _builtin_ctzll().
On powerpc, those builtins trivially use ctlzw and ctlzd power
instructions.
Allthough those instructions provide the expected result with
input argument 0, __builtin_ctz() and __builtin_ctzll() are
documented as undefined for value 0.
The easiest fix would be to use fls() and fls64() functions
defined in include/asm-generic/bitops/builtin-fls.h and
include/asm-generic/bitops/fls64.h, but GCC output is not optimal:
00000388 <testfls>:
388: 2c 03 00 00 cmpwi r3,0
38c: 41 82 00 10 beq 39c <testfls+0x14>
390: 7c 63 00 34 cntlzw r3,r3
394: 20 63 00 20 subfic r3,r3,32
398: 4e 80 00 20 blr
39c: 38 60 00 00 li r3,0
3a0: 4e 80 00 20 blr
000003b0 <testfls64>:
3b0: 2c 03 00 00 cmpwi r3,0
3b4: 40 82 00 1c bne 3d0 <testfls64+0x20>
3b8: 2f 84 00 00 cmpwi cr7,r4,0
3bc: 38 60 00 00 li r3,0
3c0: 4d 9e 00 20 beqlr cr7
3c4: 7c 83 00 34 cntlzw r3,r4
3c8: 20 63 00 20 subfic r3,r3,32
3cc: 4e 80 00 20 blr
3d0: 7c 63 00 34 cntlzw r3,r3
3d4: 20 63 00 40 subfic r3,r3,64
3d8: 4e 80 00 20 blr
When the input of fls(x) is a constant, just check x for nullity and
return either 0 or __builtin_clz(x). Otherwise, use cntlzw instruction
directly.
For fls64() on PPC64, do the same but with __builtin_clzll() and
cntlzd instruction. On PPC32, lets take the generic fls64() which
will use our fls(). The result is as expected:
00000388 <testfls>:
388: 7c 63 00 34 cntlzw r3,r3
38c: 20 63 00 20 subfic r3,r3,32
390: 4e 80 00 20 blr
000003a0 <testfls64>:
3a0: 2c 03 00 00 cmpwi r3,0
3a4: 40 82 00 10 bne 3b4 <testfls64+0x14>
3a8: 7c 83 00 34 cntlzw r3,r4
3ac: 20 63 00 20 subfic r3,r3,32
3b0: 4e 80 00 20 blr
3b4: 7c 63 00 34 cntlzw r3,r3
3b8: 20 63 00 40 subfic r3,r3,64
3bc: 4e 80 00 20 blr
Fixes: 2fcff790dcb4 ("powerpc: Use builtin functions for fls()/__fls()/fls64()")
Cc: stable@vger.kernel.org
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/348c2d3f19ffcff8abe50d52513f989c4581d000.1603375524.git.christophe.leroy@csgroup.eu
2020-10-22 14:05:46 +00:00
int lz ;
if ( __builtin_constant_p ( x ) )
return x ? 32 - __builtin_clz ( x ) : 0 ;
asm ( " cntlzw %0,%1 " : " =r " ( lz ) : " r " ( x ) ) ;
return 32 - lz ;
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
}
2008-04-18 14:26:08 +10:00
powerpc: Use builtin functions for fls()/__fls()/fls64()
With the fls() functions as defined in arch/powerpc/include/asm/bitops.h
GCC will not optimise the code in case of constant parameter.
This patch replaces __fls() by the builtin function, and modifies
fls() and fls64() to use builtins instead of inline assembly
For non constant calls, the generated code is doing the same:
int testfls(unsigned int x)
{
return fls(x);
}
unsigned long test__fls(unsigned long x)
{
return __fls(x);
}
int testfls64(__u64 x)
{
return fls64(x);
}
On PPC32, before the patch:
00000064 <testfls>:
64: 7c 63 00 34 cntlzw r3,r3
68: 20 63 00 20 subfic r3,r3,32
6c: 4e 80 00 20 blr
00000070 <test__fls>:
70: 7c 63 00 34 cntlzw r3,r3
74: 20 63 00 1f subfic r3,r3,31
78: 4e 80 00 20 blr
0000007c <testfls64>:
7c: 2c 03 00 00 cmpwi r3,0
80: 40 82 00 10 bne 90 <testfls64+0x14>
84: 7c 83 00 34 cntlzw r3,r4
88: 20 63 00 20 subfic r3,r3,32
8c: 4e 80 00 20 blr
90: 7c 63 00 34 cntlzw r3,r3
94: 20 63 00 40 subfic r3,r3,64
98: 4e 80 00 20 blr
On PPC32, after the patch:
00000054 <testfls>:
54: 7c 63 00 34 cntlzw r3,r3
58: 20 63 00 20 subfic r3,r3,32
5c: 4e 80 00 20 blr
00000060 <test__fls>:
60: 7c 63 00 34 cntlzw r3,r3
64: 20 63 00 1f subfic r3,r3,31
68: 4e 80 00 20 blr
0000006c <testfls64>:
6c: 2c 03 00 00 cmpwi r3,0
70: 41 82 00 10 beq 80 <testfls64+0x14>
74: 7c 63 00 34 cntlzw r3,r3
78: 20 63 00 40 subfic r3,r3,64
7c: 4e 80 00 20 blr
80: 7c 83 00 34 cntlzw r3,r4
84: 20 63 00 40 subfic r3,r3,32
88: 4e 80 00 20 blr
On PPC64, before the patch:
00000000000000a0 <.testfls>:
a0: 7c 63 00 34 cntlzw r3,r3
a4: 20 63 00 20 subfic r3,r3,32
a8: 7c 63 07 b4 extsw r3,r3
ac: 4e 80 00 20 blr
00000000000000b0 <.test__fls>:
b0: 7c 63 00 74 cntlzd r3,r3
b4: 20 63 00 3f subfic r3,r3,63
b8: 7c 63 07 b4 extsw r3,r3
bc: 4e 80 00 20 blr
00000000000000c0 <.testfls64>:
c0: 7c 63 00 74 cntlzd r3,r3
c4: 20 63 00 40 subfic r3,r3,64
c8: 7c 63 07 b4 extsw r3,r3
cc: 4e 80 00 20 blr
On PPC64, after the patch:
0000000000000090 <.testfls>:
90: 7c 63 00 34 cntlzw r3,r3
94: 20 63 00 20 subfic r3,r3,32
98: 7c 63 07 b4 extsw r3,r3
9c: 4e 80 00 20 blr
00000000000000a0 <.test__fls>:
a0: 7c 63 00 74 cntlzd r3,r3
a4: 20 63 00 3f subfic r3,r3,63
a8: 4e 80 00 20 blr
ac: 60 00 00 00 nop
00000000000000b0 <.testfls64>:
b0: 7c 63 00 74 cntlzd r3,r3
b4: 20 63 00 40 subfic r3,r3,64
b8: 7c 63 07 b4 extsw r3,r3
bc: 4e 80 00 20 blr
Those builtins have been in GCC since at least 3.4.6 (see
https://gcc.gnu.org/onlinedocs/gcc-3.4.6/gcc/Other-Builtins.html )
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-21 13:18:48 +02:00
# include <asm-generic/bitops/builtin-__fls.h>
2008-03-15 18:31:49 +01:00
powerpc/bitops: Fix possible undefined behaviour with fls() and fls64()
fls() and fls64() are using __builtin_ctz() and _builtin_ctzll().
On powerpc, those builtins trivially use ctlzw and ctlzd power
instructions.
Allthough those instructions provide the expected result with
input argument 0, __builtin_ctz() and __builtin_ctzll() are
documented as undefined for value 0.
The easiest fix would be to use fls() and fls64() functions
defined in include/asm-generic/bitops/builtin-fls.h and
include/asm-generic/bitops/fls64.h, but GCC output is not optimal:
00000388 <testfls>:
388: 2c 03 00 00 cmpwi r3,0
38c: 41 82 00 10 beq 39c <testfls+0x14>
390: 7c 63 00 34 cntlzw r3,r3
394: 20 63 00 20 subfic r3,r3,32
398: 4e 80 00 20 blr
39c: 38 60 00 00 li r3,0
3a0: 4e 80 00 20 blr
000003b0 <testfls64>:
3b0: 2c 03 00 00 cmpwi r3,0
3b4: 40 82 00 1c bne 3d0 <testfls64+0x20>
3b8: 2f 84 00 00 cmpwi cr7,r4,0
3bc: 38 60 00 00 li r3,0
3c0: 4d 9e 00 20 beqlr cr7
3c4: 7c 83 00 34 cntlzw r3,r4
3c8: 20 63 00 20 subfic r3,r3,32
3cc: 4e 80 00 20 blr
3d0: 7c 63 00 34 cntlzw r3,r3
3d4: 20 63 00 40 subfic r3,r3,64
3d8: 4e 80 00 20 blr
When the input of fls(x) is a constant, just check x for nullity and
return either 0 or __builtin_clz(x). Otherwise, use cntlzw instruction
directly.
For fls64() on PPC64, do the same but with __builtin_clzll() and
cntlzd instruction. On PPC32, lets take the generic fls64() which
will use our fls(). The result is as expected:
00000388 <testfls>:
388: 7c 63 00 34 cntlzw r3,r3
38c: 20 63 00 20 subfic r3,r3,32
390: 4e 80 00 20 blr
000003a0 <testfls64>:
3a0: 2c 03 00 00 cmpwi r3,0
3a4: 40 82 00 10 bne 3b4 <testfls64+0x14>
3a8: 7c 83 00 34 cntlzw r3,r4
3ac: 20 63 00 20 subfic r3,r3,32
3b0: 4e 80 00 20 blr
3b4: 7c 63 00 34 cntlzw r3,r3
3b8: 20 63 00 40 subfic r3,r3,64
3bc: 4e 80 00 20 blr
Fixes: 2fcff790dcb4 ("powerpc: Use builtin functions for fls()/__fls()/fls64()")
Cc: stable@vger.kernel.org
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/348c2d3f19ffcff8abe50d52513f989c4581d000.1603375524.git.christophe.leroy@csgroup.eu
2020-10-22 14:05:46 +00:00
/*
* 64 - bit can do this using one cntlzd ( count leading zeroes doubleword )
* instruction ; for 32 - bit we use the generic version , which does two
* 32 - bit fls calls .
*/
# ifdef CONFIG_PPC64
powerpc/bitops: Force inlining of fls()
Building a kernel with CONFIG_CC_OPTIMISE_FOR_SIZE leads to
the following functions being copied several times in vmlinux:
31 times __ilog2_u32()
34 times fls()
Disassembly follows:
c00f476c <fls>:
c00f476c: 7c 63 00 34 cntlzw r3,r3
c00f4770: 20 63 00 20 subfic r3,r3,32
c00f4774: 4e 80 00 20 blr
c00f4778 <__ilog2_u32>:
c00f4778: 94 21 ff f0 stwu r1,-16(r1)
c00f477c: 7c 08 02 a6 mflr r0
c00f4780: 90 01 00 14 stw r0,20(r1)
c00f4784: 4b ff ff e9 bl c00f476c <fls>
c00f4788: 80 01 00 14 lwz r0,20(r1)
c00f478c: 38 63 ff ff addi r3,r3,-1
c00f4790: 7c 08 03 a6 mtlr r0
c00f4794: 38 21 00 10 addi r1,r1,16
c00f4798: 4e 80 00 20 blr
When forcing inlining of fls(), we get
c0008b80 <__ilog2_u32>:
c0008b80: 7c 63 00 34 cntlzw r3,r3
c0008b84: 20 63 00 1f subfic r3,r3,31
c0008b88: 4e 80 00 20 blr
vmlinux size gets reduced by 1 kbyte with that change.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/adc9c9d6378f6b5008246ca717993d7870188efb.1644569473.git.christophe.leroy@csgroup.eu
2022-02-11 09:51:32 +01:00
static __always_inline int fls64 ( __u64 x )
2008-04-18 14:26:08 +10:00
{
powerpc/bitops: Fix possible undefined behaviour with fls() and fls64()
fls() and fls64() are using __builtin_ctz() and _builtin_ctzll().
On powerpc, those builtins trivially use ctlzw and ctlzd power
instructions.
Allthough those instructions provide the expected result with
input argument 0, __builtin_ctz() and __builtin_ctzll() are
documented as undefined for value 0.
The easiest fix would be to use fls() and fls64() functions
defined in include/asm-generic/bitops/builtin-fls.h and
include/asm-generic/bitops/fls64.h, but GCC output is not optimal:
00000388 <testfls>:
388: 2c 03 00 00 cmpwi r3,0
38c: 41 82 00 10 beq 39c <testfls+0x14>
390: 7c 63 00 34 cntlzw r3,r3
394: 20 63 00 20 subfic r3,r3,32
398: 4e 80 00 20 blr
39c: 38 60 00 00 li r3,0
3a0: 4e 80 00 20 blr
000003b0 <testfls64>:
3b0: 2c 03 00 00 cmpwi r3,0
3b4: 40 82 00 1c bne 3d0 <testfls64+0x20>
3b8: 2f 84 00 00 cmpwi cr7,r4,0
3bc: 38 60 00 00 li r3,0
3c0: 4d 9e 00 20 beqlr cr7
3c4: 7c 83 00 34 cntlzw r3,r4
3c8: 20 63 00 20 subfic r3,r3,32
3cc: 4e 80 00 20 blr
3d0: 7c 63 00 34 cntlzw r3,r3
3d4: 20 63 00 40 subfic r3,r3,64
3d8: 4e 80 00 20 blr
When the input of fls(x) is a constant, just check x for nullity and
return either 0 or __builtin_clz(x). Otherwise, use cntlzw instruction
directly.
For fls64() on PPC64, do the same but with __builtin_clzll() and
cntlzd instruction. On PPC32, lets take the generic fls64() which
will use our fls(). The result is as expected:
00000388 <testfls>:
388: 7c 63 00 34 cntlzw r3,r3
38c: 20 63 00 20 subfic r3,r3,32
390: 4e 80 00 20 blr
000003a0 <testfls64>:
3a0: 2c 03 00 00 cmpwi r3,0
3a4: 40 82 00 10 bne 3b4 <testfls64+0x14>
3a8: 7c 83 00 34 cntlzw r3,r4
3ac: 20 63 00 20 subfic r3,r3,32
3b0: 4e 80 00 20 blr
3b4: 7c 63 00 34 cntlzw r3,r3
3b8: 20 63 00 40 subfic r3,r3,64
3bc: 4e 80 00 20 blr
Fixes: 2fcff790dcb4 ("powerpc: Use builtin functions for fls()/__fls()/fls64()")
Cc: stable@vger.kernel.org
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/348c2d3f19ffcff8abe50d52513f989c4581d000.1603375524.git.christophe.leroy@csgroup.eu
2020-10-22 14:05:46 +00:00
int lz ;
if ( __builtin_constant_p ( x ) )
return x ? 64 - __builtin_clzll ( x ) : 0 ;
asm ( " cntlzd %0,%1 " : " =r " ( lz ) : " r " ( x ) ) ;
return 64 - lz ;
2008-04-18 14:26:08 +10:00
}
powerpc/bitops: Fix possible undefined behaviour with fls() and fls64()
fls() and fls64() are using __builtin_ctz() and _builtin_ctzll().
On powerpc, those builtins trivially use ctlzw and ctlzd power
instructions.
Allthough those instructions provide the expected result with
input argument 0, __builtin_ctz() and __builtin_ctzll() are
documented as undefined for value 0.
The easiest fix would be to use fls() and fls64() functions
defined in include/asm-generic/bitops/builtin-fls.h and
include/asm-generic/bitops/fls64.h, but GCC output is not optimal:
00000388 <testfls>:
388: 2c 03 00 00 cmpwi r3,0
38c: 41 82 00 10 beq 39c <testfls+0x14>
390: 7c 63 00 34 cntlzw r3,r3
394: 20 63 00 20 subfic r3,r3,32
398: 4e 80 00 20 blr
39c: 38 60 00 00 li r3,0
3a0: 4e 80 00 20 blr
000003b0 <testfls64>:
3b0: 2c 03 00 00 cmpwi r3,0
3b4: 40 82 00 1c bne 3d0 <testfls64+0x20>
3b8: 2f 84 00 00 cmpwi cr7,r4,0
3bc: 38 60 00 00 li r3,0
3c0: 4d 9e 00 20 beqlr cr7
3c4: 7c 83 00 34 cntlzw r3,r4
3c8: 20 63 00 20 subfic r3,r3,32
3cc: 4e 80 00 20 blr
3d0: 7c 63 00 34 cntlzw r3,r3
3d4: 20 63 00 40 subfic r3,r3,64
3d8: 4e 80 00 20 blr
When the input of fls(x) is a constant, just check x for nullity and
return either 0 or __builtin_clz(x). Otherwise, use cntlzw instruction
directly.
For fls64() on PPC64, do the same but with __builtin_clzll() and
cntlzd instruction. On PPC32, lets take the generic fls64() which
will use our fls(). The result is as expected:
00000388 <testfls>:
388: 7c 63 00 34 cntlzw r3,r3
38c: 20 63 00 20 subfic r3,r3,32
390: 4e 80 00 20 blr
000003a0 <testfls64>:
3a0: 2c 03 00 00 cmpwi r3,0
3a4: 40 82 00 10 bne 3b4 <testfls64+0x14>
3a8: 7c 83 00 34 cntlzw r3,r4
3ac: 20 63 00 20 subfic r3,r3,32
3b0: 4e 80 00 20 blr
3b4: 7c 63 00 34 cntlzw r3,r3
3b8: 20 63 00 40 subfic r3,r3,64
3bc: 4e 80 00 20 blr
Fixes: 2fcff790dcb4 ("powerpc: Use builtin functions for fls()/__fls()/fls64()")
Cc: stable@vger.kernel.org
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/348c2d3f19ffcff8abe50d52513f989c4581d000.1603375524.git.christophe.leroy@csgroup.eu
2020-10-22 14:05:46 +00:00
# else
# include <asm-generic/bitops/fls64.h>
# endif
2008-04-18 14:26:08 +10:00
2010-08-12 16:28:09 +00:00
# ifdef CONFIG_PPC64
unsigned int __arch_hweight8 ( unsigned int w ) ;
unsigned int __arch_hweight16 ( unsigned int w ) ;
unsigned int __arch_hweight32 ( unsigned int w ) ;
unsigned long __arch_hweight64 ( __u64 w ) ;
# include <asm-generic/bitops/const_hweight.h>
# else
2006-03-26 01:39:33 -08:00
# include <asm-generic/bitops/hweight.h>
2010-08-12 16:28:09 +00:00
# endif
2019-08-20 12:49:41 +10:00
/* wrappers that deal with KASAN instrumentation */
# include <asm-generic/bitops/instrumented-atomic.h>
# include <asm-generic/bitops/instrumented-lock.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
/* Little-endian versions */
2012-11-04 02:03:45 +00:00
# include <asm-generic/bitops/le.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
/* Bitmap functions for the ext2 filesystem */
2011-07-26 16:09:04 -07:00
# include <asm-generic/bitops/ext2-atomic-setbit.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
2006-03-26 01:39:33 -08:00
# include <asm-generic/bitops/sched.h>
[PATCH] powerpc: Merge bitops.h
Here's a revised version. This re-introduces the set_bits() function
from ppc64, which I removed because I thought it was unused (it exists
on no other arch). In fact it is used in the powermac interrupt code
(but not on pSeries).
- We use LARXL/STCXL macros to generate the right (32 or 64 bit)
instructions, similar to LDL/STL from ppc_asm.h, used in fpu.S
- ppc32 previously used a full "sync" barrier at the end of
test_and_*_bit(), whereas ppc64 used an "isync". The merged version
uses "isync", since I believe that's sufficient.
- The ppc64 versions of then minix_*() bitmap functions have changed
semantics. Previously on ppc64, these functions were big-endian
(that is bit 0 was the LSB in the first 64-bit, big-endian word).
On ppc32 (and x86, for that matter, they were little-endian. As far
as I can tell, the big-endian usage was simply wrong - I guess
no-one ever tried to use minixfs on ppc64.
- On ppc32 find_next_bit() and find_next_zero_bit() are no longer
inline (they were already out-of-line on ppc64).
- For ppc64, sched_find_first_bit() has moved from mmu_context.h to
the merged bitops. What it was doing in mmu_context.h in the first
place, I have no idea.
- The fls() function is now implemented using the cntlzw instruction
on ppc64, instead of generic_fls(), as it already was on ppc32.
- For ARCH=ppc, this patch requires adding arch/powerpc/lib to the
arch/ppc/Makefile. This in turn requires some changes to
arch/powerpc/lib/Makefile which didn't correctly handle ARCH=ppc.
Built and running on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-11-01 17:28:10 +11:00
# endif /* __KERNEL__ */
# endif /* _ASM_POWERPC_BITOPS_H */