linux/arch/hexagon/include/asm/atomic.h
Peter Zijlstra 9d664c0aec locking/atomic: Fix atomic_set_release() for 'funny' architectures
Those architectures that have a special atomic_set implementation also
need a special atomic_set_release(), because for the very same reason
WRITE_ONCE() is broken for them, smp_store_release() is too.

The vast majority is architectures that have spinlock hash based atomic
implementation except hexagon which seems to have a hardware 'feature'.

The spinlock based atomics should be SC, that is, none of them appear to
place extra barriers in atomic_cmpxchg() or any of the other SC atomic
primitives and therefore seem to rely on their spinlock implementation
being SC (I did not fully validate all that).

Therefore, the normal atomic_set() is SC and can be used at
atomic_set_release().

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Chris Metcalf <cmetcalf@mellanox.com> [for tile]
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: davem@davemloft.net
Cc: james.hogan@imgtec.com
Cc: jejb@parisc-linux.org
Cc: rkuo@codeaurora.org
Cc: vgupta@synopsys.com
Link: http://lkml.kernel.org/r/20170609110506.yod47flaav3wgoj5@hirez.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 12:28:54 +02:00

214 lines
5.3 KiB
C

/*
* Atomic operations for the Hexagon architecture
*
* Copyright (c) 2010-2013, The Linux Foundation. All rights reserved.
*
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 and
* only version 2 as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
* 02110-1301, USA.
*/
#ifndef _ASM_ATOMIC_H
#define _ASM_ATOMIC_H
#include <linux/types.h>
#include <asm/cmpxchg.h>
#include <asm/barrier.h>
#define ATOMIC_INIT(i) { (i) }
/* Normal writes in our arch don't clear lock reservations */
static inline void atomic_set(atomic_t *v, int new)
{
asm volatile(
"1: r6 = memw_locked(%0);\n"
" memw_locked(%0,p0) = %1;\n"
" if (!P0) jump 1b;\n"
:
: "r" (&v->counter), "r" (new)
: "memory", "p0", "r6"
);
}
#define atomic_set_release(v, i) atomic_set((v), (i))
/**
* atomic_read - reads a word, atomically
* @v: pointer to atomic value
*
* Assumes all word reads on our architecture are atomic.
*/
#define atomic_read(v) READ_ONCE((v)->counter)
/**
* atomic_xchg - atomic
* @v: pointer to memory to change
* @new: new value (technically passed in a register -- see xchg)
*/
#define atomic_xchg(v, new) (xchg(&((v)->counter), (new)))
/**
* atomic_cmpxchg - atomic compare-and-exchange values
* @v: pointer to value to change
* @old: desired old value to match
* @new: new value to put in
*
* Parameters are then pointer, value-in-register, value-in-register,
* and the output is the old value.
*
* Apparently this is complicated for archs that don't support
* the memw_locked like we do (or it's broken or whatever).
*
* Kind of the lynchpin of the rest of the generically defined routines.
* Remember V2 had that bug with dotnew predicate set by memw_locked.
*
* "old" is "expected" old val, __oldval is actual old value
*/
static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
{
int __oldval;
asm volatile(
"1: %0 = memw_locked(%1);\n"
" { P0 = cmp.eq(%0,%2);\n"
" if (!P0.new) jump:nt 2f; }\n"
" memw_locked(%1,P0) = %3;\n"
" if (!P0) jump 1b;\n"
"2:\n"
: "=&r" (__oldval)
: "r" (&v->counter), "r" (old), "r" (new)
: "memory", "p0"
);
return __oldval;
}
#define ATOMIC_OP(op) \
static inline void atomic_##op(int i, atomic_t *v) \
{ \
int output; \
\
__asm__ __volatile__ ( \
"1: %0 = memw_locked(%1);\n" \
" %0 = "#op "(%0,%2);\n" \
" memw_locked(%1,P3)=%0;\n" \
" if !P3 jump 1b;\n" \
: "=&r" (output) \
: "r" (&v->counter), "r" (i) \
: "memory", "p3" \
); \
} \
#define ATOMIC_OP_RETURN(op) \
static inline int atomic_##op##_return(int i, atomic_t *v) \
{ \
int output; \
\
__asm__ __volatile__ ( \
"1: %0 = memw_locked(%1);\n" \
" %0 = "#op "(%0,%2);\n" \
" memw_locked(%1,P3)=%0;\n" \
" if !P3 jump 1b;\n" \
: "=&r" (output) \
: "r" (&v->counter), "r" (i) \
: "memory", "p3" \
); \
return output; \
}
#define ATOMIC_FETCH_OP(op) \
static inline int atomic_fetch_##op(int i, atomic_t *v) \
{ \
int output, val; \
\
__asm__ __volatile__ ( \
"1: %0 = memw_locked(%2);\n" \
" %1 = "#op "(%0,%3);\n" \
" memw_locked(%2,P3)=%1;\n" \
" if !P3 jump 1b;\n" \
: "=&r" (output), "=&r" (val) \
: "r" (&v->counter), "r" (i) \
: "memory", "p3" \
); \
return output; \
}
#define ATOMIC_OPS(op) ATOMIC_OP(op) ATOMIC_OP_RETURN(op) ATOMIC_FETCH_OP(op)
ATOMIC_OPS(add)
ATOMIC_OPS(sub)
#undef ATOMIC_OPS
#define ATOMIC_OPS(op) ATOMIC_OP(op) ATOMIC_FETCH_OP(op)
ATOMIC_OPS(and)
ATOMIC_OPS(or)
ATOMIC_OPS(xor)
#undef ATOMIC_OPS
#undef ATOMIC_FETCH_OP
#undef ATOMIC_OP_RETURN
#undef ATOMIC_OP
/**
* __atomic_add_unless - add unless the number is a given value
* @v: pointer to value
* @a: amount to add
* @u: unless value is equal to u
*
* Returns old value.
*
*/
static inline int __atomic_add_unless(atomic_t *v, int a, int u)
{
int __oldval;
register int tmp;
asm volatile(
"1: %0 = memw_locked(%2);"
" {"
" p3 = cmp.eq(%0, %4);"
" if (p3.new) jump:nt 2f;"
" %1 = add(%0, %3);"
" }"
" memw_locked(%2, p3) = %1;"
" {"
" if !p3 jump 1b;"
" }"
"2:"
: "=&r" (__oldval), "=&r" (tmp)
: "r" (v), "r" (a), "r" (u)
: "memory", "p3"
);
return __oldval;
}
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
#define atomic_inc(v) atomic_add(1, (v))
#define atomic_dec(v) atomic_sub(1, (v))
#define atomic_inc_and_test(v) (atomic_add_return(1, (v)) == 0)
#define atomic_dec_and_test(v) (atomic_sub_return(1, (v)) == 0)
#define atomic_sub_and_test(i, v) (atomic_sub_return(i, (v)) == 0)
#define atomic_add_negative(i, v) (atomic_add_return(i, (v)) < 0)
#define atomic_inc_return(v) (atomic_add_return(1, v))
#define atomic_dec_return(v) (atomic_sub_return(1, v))
#endif