1
0
Fork 0
alistair23-linux/arch/mips/include/asm/atomic.h

441 lines
12 KiB
C
Raw Normal View History

/*
* Atomic operations that C can't guarantee us. Useful for
* resource counting etc..
*
* But use these as seldom as possible since they are much more slower
* than regular operations.
*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 1996, 97, 99, 2000, 03, 04, 06 by Ralf Baechle
*/
#ifndef _ASM_ATOMIC_H
#define _ASM_ATOMIC_H
#include <linux/irqflags.h>
#include <linux/types.h>
#include <asm/barrier.h>
MIPS: Fix microMIPS LL/SC immediate offsets In the microMIPS encoding some memory access instructions have their immediate offset reduced to 12 bits only. That does not match the GCC `R' constraint we use in some places to satisfy the requirement, resulting in build failures like this: {standard input}: Assembler messages: {standard input}:720: Error: macro used $at after ".set noat" {standard input}:720: Warning: macro instruction expanded into multiple instructions Fix the problem by defining a macro, `GCC_OFF12_ASM', that expands to the right constraint depending on whether microMIPS or standard MIPS code is produced. Also apply the fix to where `m' is used as in the worst case this change does nothing, e.g. where the pointer was already in a register such as a function argument and no further offset was requested, and in the best case it avoids an extraneous sequence of up to two instructions to load the high 20 bits of the address in the LL/SC loop. This reduces the risk of lock contention that is the higher the more instructions there are in the critical section between LL and SC. Strictly speaking we could just bulk-replace `R' with `ZC' as the latter constraint adjusts automatically depending on the ISA selected. However it was only introduced with GCC 4.9 and we keep supporing older compilers for the standard MIPS configuration, hence the slightly more complicated approach I chose. The choice of a zero-argument function-like rather than an object-like macro was made so that it does not look like a function call taking the C expression used for the constraint as an argument. This is so as not to confuse the reader or formatting checkers like `checkpatch.pl' and follows previous practice. Signed-off-by: Maciej W. Rozycki <macro@codesourcery.com> Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/8482/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2014-11-15 15:08:48 -07:00
#include <asm/compiler.h>
#include <asm/cpu-features.h>
#include <asm/cmpxchg.h>
#include <asm/war.h>
/*
* Using a branch-likely instruction to check the result of an sc instruction
* works around a bug present in R10000 CPUs prior to revision 3.0 that could
* cause ll-sc sequences to execute non-atomically.
*/
#if R10000_LLSC_WAR
# define __scbeqz "beqzl"
#else
# define __scbeqz "beqz"
#endif
#define ATOMIC_INIT(i) { (i) }
/*
* atomic_read - read atomic variable
* @v: pointer of type atomic_t
*
* Atomically reads the value of @v.
*/
#define atomic_read(v) READ_ONCE((v)->counter)
/*
* atomic_set - set atomic variable
* @v: pointer of type atomic_t
* @i: required value
*
* Atomically sets the value of @v to @i.
*/
#define atomic_set(v, i) WRITE_ONCE((v)->counter, (i))
#define ATOMIC_OP(op, c_op, asm_op) \
static __inline__ void atomic_##op(int i, atomic_t * v) \
{ \
if (kernel_uses_llsc) { \
int temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: ll %0, %1 # atomic_" #op " \n" \
" " #asm_op " %0, %2 \n" \
" sc %0, %1 \n" \
"\t" __scbeqz " %0, 1b \n" \
" .set pop \n" \
: "=&r" (temp), "+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
v->counter c_op i; \
raw_local_irq_restore(flags); \
} \
}
#define ATOMIC_OP_RETURN(op, c_op, asm_op) \
static __inline__ int atomic_##op##_return_relaxed(int i, atomic_t * v) \
{ \
int result; \
\
if (kernel_uses_llsc) { \
int temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: ll %1, %2 # atomic_" #op "_return \n" \
" " #asm_op " %0, %1, %3 \n" \
" sc %0, %2 \n" \
"\t" __scbeqz " %0, 1b \n" \
" " #asm_op " %0, %1, %3 \n" \
" .set pop \n" \
: "=&r" (result), "=&r" (temp), \
"+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
result = v->counter; \
result c_op i; \
v->counter = result; \
raw_local_irq_restore(flags); \
} \
\
return result; \
}
#define ATOMIC_FETCH_OP(op, c_op, asm_op) \
static __inline__ int atomic_fetch_##op##_relaxed(int i, atomic_t * v) \
{ \
int result; \
\
if (kernel_uses_llsc) { \
int temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: ll %1, %2 # atomic_fetch_" #op " \n" \
" " #asm_op " %0, %1, %3 \n" \
" sc %0, %2 \n" \
"\t" __scbeqz " %0, 1b \n" \
" .set pop \n" \
MIPS: Avoid move psuedo-instruction whilst using MIPS_ISA_LEVEL MIPS_ISA_LEVEL is always defined as the 64 bit ISA that is a compatible superset of the ISA that the kernel build is targeting, and is used to allow us to emit instructions that we may detect support for at runtime. When we use a .set MIPS_ISA_LEVEL directive & are building a 32-bit kernel, we therefore are temporarily allowing the assembler to generate MIPS64 instructions. Using the move pseudo-instruction whilst this is the case is problematic because the assembler is likely to emit a daddu instruction which will generate a reserved instruction exception when executed on a MIPS32 machine. Unfortunately the combination of commit a0a5ac3ce8fe ("MIPS: Fix delay slot bug in `atomic*_sub_if_positive' for R10000_LLSC_WAR") and commit 4936084c2ee2 ("MIPS: Cleanup R10000_LLSC_WAR logic in atomic.h") causes us to do exactly this in atomic_sub_if_positive(), and the result is MIPS64 daddu instructions in 32-bit kernels. Fix this by using .set mips0 to restore the default ISA after the ll instruction, and use .set MIPS_ISA_LEVEL again prior to the sc. This ensures everything but the ll & sc are assembled using the default ISA for the kernel build & the move pseudo-instruction is emitted as a MIPS32 addu instruction. We appear to have another pre-existing instance of the same issue in our atomic_fetch_*_relaxed() functions, and fix that up too by moving our .set move0 such that it occurs prior to use of the move pseudo-instruction. Signed-off-by: Paul Burton <paul.burton@mips.com> Fixes: a0a5ac3ce8fe ("MIPS: Fix delay slot bug in `atomic*_sub_if_positive' for R10000_LLSC_WAR") Fixes: 4936084c2ee2 ("MIPS: Cleanup R10000_LLSC_WAR logic in atomic.h") Patchwork: https://patchwork.linux-mips.org/patch/20253/ Cc: James Hogan <jhogan@kernel.org> Cc: Joshua Kinard <kumba@gentoo.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org
2018-08-17 16:36:24 -06:00
" move %0, %1 \n" \
: "=&r" (result), "=&r" (temp), \
"+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
result = v->counter; \
v->counter c_op i; \
raw_local_irq_restore(flags); \
} \
\
return result; \
}
#define ATOMIC_OPS(op, c_op, asm_op) \
ATOMIC_OP(op, c_op, asm_op) \
ATOMIC_OP_RETURN(op, c_op, asm_op) \
ATOMIC_FETCH_OP(op, c_op, asm_op)
ATOMIC_OPS(add, +=, addu)
ATOMIC_OPS(sub, -=, subu)
#define atomic_add_return_relaxed atomic_add_return_relaxed
#define atomic_sub_return_relaxed atomic_sub_return_relaxed
#define atomic_fetch_add_relaxed atomic_fetch_add_relaxed
#define atomic_fetch_sub_relaxed atomic_fetch_sub_relaxed
#undef ATOMIC_OPS
#define ATOMIC_OPS(op, c_op, asm_op) \
ATOMIC_OP(op, c_op, asm_op) \
ATOMIC_FETCH_OP(op, c_op, asm_op)
ATOMIC_OPS(and, &=, and)
ATOMIC_OPS(or, |=, or)
ATOMIC_OPS(xor, ^=, xor)
#define atomic_fetch_and_relaxed atomic_fetch_and_relaxed
#define atomic_fetch_or_relaxed atomic_fetch_or_relaxed
#define atomic_fetch_xor_relaxed atomic_fetch_xor_relaxed
#undef ATOMIC_OPS
#undef ATOMIC_FETCH_OP
#undef ATOMIC_OP_RETURN
#undef ATOMIC_OP
/*
* atomic_sub_if_positive - conditionally subtract integer from atomic variable
* @i: integer value to subtract
* @v: pointer of type atomic_t
*
* Atomically test @v and subtract @i if @v is greater or equal than @i.
* The function returns the old value of @v minus @i.
*/
static __inline__ int atomic_sub_if_positive(int i, atomic_t * v)
{
int result;
smp_mb__before_llsc();
if (kernel_uses_llsc) {
int temp;
__asm__ __volatile__(
" .set push \n"
" .set "MIPS_ISA_LEVEL" \n"
"1: ll %1, %2 # atomic_sub_if_positive\n"
" .set pop \n"
" subu %0, %1, %3 \n"
" move %1, %0 \n"
" bltz %0, 1f \n"
" .set push \n"
MIPS: Avoid move psuedo-instruction whilst using MIPS_ISA_LEVEL MIPS_ISA_LEVEL is always defined as the 64 bit ISA that is a compatible superset of the ISA that the kernel build is targeting, and is used to allow us to emit instructions that we may detect support for at runtime. When we use a .set MIPS_ISA_LEVEL directive & are building a 32-bit kernel, we therefore are temporarily allowing the assembler to generate MIPS64 instructions. Using the move pseudo-instruction whilst this is the case is problematic because the assembler is likely to emit a daddu instruction which will generate a reserved instruction exception when executed on a MIPS32 machine. Unfortunately the combination of commit a0a5ac3ce8fe ("MIPS: Fix delay slot bug in `atomic*_sub_if_positive' for R10000_LLSC_WAR") and commit 4936084c2ee2 ("MIPS: Cleanup R10000_LLSC_WAR logic in atomic.h") causes us to do exactly this in atomic_sub_if_positive(), and the result is MIPS64 daddu instructions in 32-bit kernels. Fix this by using .set mips0 to restore the default ISA after the ll instruction, and use .set MIPS_ISA_LEVEL again prior to the sc. This ensures everything but the ll & sc are assembled using the default ISA for the kernel build & the move pseudo-instruction is emitted as a MIPS32 addu instruction. We appear to have another pre-existing instance of the same issue in our atomic_fetch_*_relaxed() functions, and fix that up too by moving our .set move0 such that it occurs prior to use of the move pseudo-instruction. Signed-off-by: Paul Burton <paul.burton@mips.com> Fixes: a0a5ac3ce8fe ("MIPS: Fix delay slot bug in `atomic*_sub_if_positive' for R10000_LLSC_WAR") Fixes: 4936084c2ee2 ("MIPS: Cleanup R10000_LLSC_WAR logic in atomic.h") Patchwork: https://patchwork.linux-mips.org/patch/20253/ Cc: James Hogan <jhogan@kernel.org> Cc: Joshua Kinard <kumba@gentoo.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org
2018-08-17 16:36:24 -06:00
" .set "MIPS_ISA_LEVEL" \n"
" sc %1, %2 \n"
"\t" __scbeqz " %1, 1b \n"
"1: \n"
" .set pop \n"
MIPS: Fix microMIPS LL/SC immediate offsets In the microMIPS encoding some memory access instructions have their immediate offset reduced to 12 bits only. That does not match the GCC `R' constraint we use in some places to satisfy the requirement, resulting in build failures like this: {standard input}: Assembler messages: {standard input}:720: Error: macro used $at after ".set noat" {standard input}:720: Warning: macro instruction expanded into multiple instructions Fix the problem by defining a macro, `GCC_OFF12_ASM', that expands to the right constraint depending on whether microMIPS or standard MIPS code is produced. Also apply the fix to where `m' is used as in the worst case this change does nothing, e.g. where the pointer was already in a register such as a function argument and no further offset was requested, and in the best case it avoids an extraneous sequence of up to two instructions to load the high 20 bits of the address in the LL/SC loop. This reduces the risk of lock contention that is the higher the more instructions there are in the critical section between LL and SC. Strictly speaking we could just bulk-replace `R' with `ZC' as the latter constraint adjusts automatically depending on the ISA selected. However it was only introduced with GCC 4.9 and we keep supporing older compilers for the standard MIPS configuration, hence the slightly more complicated approach I chose. The choice of a zero-argument function-like rather than an object-like macro was made so that it does not look like a function call taking the C expression used for the constraint as an argument. This is so as not to confuse the reader or formatting checkers like `checkpatch.pl' and follows previous practice. Signed-off-by: Maciej W. Rozycki <macro@codesourcery.com> Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/8482/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2014-11-15 15:08:48 -07:00
: "=&r" (result), "=&r" (temp),
"+" GCC_OFF_SMALL_ASM() (v->counter)
: "Ir" (i));
} else {
unsigned long flags;
raw_local_irq_save(flags);
result = v->counter;
result -= i;
if (result >= 0)
v->counter = result;
raw_local_irq_restore(flags);
}
smp_llsc_mb();
return result;
}
#define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n)))
#define atomic_xchg(v, new) (xchg(&((v)->counter), (new)))
/*
* atomic_dec_if_positive - decrement by 1 if old value positive
* @v: pointer of type atomic_t
*/
#define atomic_dec_if_positive(v) atomic_sub_if_positive(1, v)
#ifdef CONFIG_64BIT
#define ATOMIC64_INIT(i) { (i) }
/*
* atomic64_read - read atomic variable
* @v: pointer of type atomic64_t
*
*/
#define atomic64_read(v) READ_ONCE((v)->counter)
/*
* atomic64_set - set atomic variable
* @v: pointer of type atomic64_t
* @i: required value
*/
#define atomic64_set(v, i) WRITE_ONCE((v)->counter, (i))
#define ATOMIC64_OP(op, c_op, asm_op) \
static __inline__ void atomic64_##op(long i, atomic64_t * v) \
{ \
if (kernel_uses_llsc) { \
long temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: lld %0, %1 # atomic64_" #op " \n" \
" " #asm_op " %0, %2 \n" \
" scd %0, %1 \n" \
"\t" __scbeqz " %0, 1b \n" \
" .set pop \n" \
: "=&r" (temp), "+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
v->counter c_op i; \
raw_local_irq_restore(flags); \
} \
}
#define ATOMIC64_OP_RETURN(op, c_op, asm_op) \
static __inline__ long atomic64_##op##_return_relaxed(long i, atomic64_t * v) \
{ \
long result; \
\
if (kernel_uses_llsc) { \
long temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: lld %1, %2 # atomic64_" #op "_return\n" \
" " #asm_op " %0, %1, %3 \n" \
" scd %0, %2 \n" \
"\t" __scbeqz " %0, 1b \n" \
" " #asm_op " %0, %1, %3 \n" \
" .set pop \n" \
: "=&r" (result), "=&r" (temp), \
"+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
result = v->counter; \
result c_op i; \
v->counter = result; \
raw_local_irq_restore(flags); \
} \
\
return result; \
}
#define ATOMIC64_FETCH_OP(op, c_op, asm_op) \
static __inline__ long atomic64_fetch_##op##_relaxed(long i, atomic64_t * v) \
{ \
long result; \
\
if (kernel_uses_llsc) { \
long temp; \
\
MIPS: Loongson: Introduce and use loongson_llsc_mb() On the Loongson-2G/2H/3A/3B there is a hardware flaw that ll/sc and lld/scd is very weak ordering. We should add sync instructions "before each ll/lld" and "at the branch-target between ll/sc" to workaround. Otherwise, this flaw will cause deadlock occasionally (e.g. when doing heavy load test with LTP). Below is the explaination of CPU designer: "For Loongson 3 family, when a memory access instruction (load, store, or prefetch)'s executing occurs between the execution of LL and SC, the success or failure of SC is not predictable. Although programmer would not insert memory access instructions between LL and SC, the memory instructions before LL in program-order, may dynamically executed between the execution of LL/SC, so a memory fence (SYNC) is needed before LL/LLD to avoid this situation. Since Loongson-3A R2 (3A2000), we have improved our hardware design to handle this case. But we later deduce a rarely circumstance that some speculatively executed memory instructions due to branch misprediction between LL/SC still fall into the above case, so a memory fence (SYNC) at branch-target (if its target is not between LL/SC) is needed for Loongson 3A1000, 3B1500, 3A2000 and 3A3000. Our processor is continually evolving and we aim to to remove all these workaround-SYNCs around LL/SC for new-come processor." Here is an example: Both cpu1 and cpu2 simutaneously run atomic_add by 1 on same atomic var, this bug cause both 'sc' run by two cpus (in atomic_add) succeed at same time('sc' return 1), and the variable is only *added by 1*, sometimes, which is wrong and unacceptable(it should be added by 2). Why disable fix-loongson3-llsc in compiler? Because compiler fix will cause problems in kernel's __ex_table section. This patch fix all the cases in kernel, but: +. the fix at the end of futex_atomic_cmpxchg_inatomic is for branch-target of 'bne', there other cases which smp_mb__before_llsc() and smp_llsc_mb() fix the ll and branch-target coincidently such as atomic_sub_if_positive/ cmpxchg/xchg, just like this one. +. Loongson 3 does support CONFIG_EDAC_ATOMIC_SCRUB, so no need to touch edac.h +. local_ops and cmpxchg_local should not be affected by this bug since only the owner can write. +. mips_atomic_set for syscall.c is deprecated and rarely used, just let it go Signed-off-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Huang Pei <huangpei@loongson.cn> [paul.burton@mips.com: - Simplify the addition of -mno-fix-loongson3-llsc to cflags, and add a comment describing why it's there. - Make loongson_llsc_mb() a no-op when CONFIG_CPU_LOONGSON3_WORKAROUNDS=n, rather than a compiler memory barrier. - Add a comment describing the bug & how loongson_llsc_mb() helps in asm/barrier.h.] Signed-off-by: Paul Burton <paul.burton@mips.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: ambrosehua@gmail.com Cc: Steven J . Hill <Steven.Hill@cavium.com> Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Xu Chenghua <xuchenghua@loongson.cn>
2019-01-15 01:04:54 -07:00
loongson_llsc_mb(); \
__asm__ __volatile__( \
" .set push \n" \
" .set "MIPS_ISA_LEVEL" \n" \
"1: lld %1, %2 # atomic64_fetch_" #op "\n" \
" " #asm_op " %0, %1, %3 \n" \
" scd %0, %2 \n" \
"\t" __scbeqz " %0, 1b \n" \
" move %0, %1 \n" \
" .set pop \n" \
: "=&r" (result), "=&r" (temp), \
"+" GCC_OFF_SMALL_ASM() (v->counter) \
: "Ir" (i)); \
} else { \
unsigned long flags; \
\
raw_local_irq_save(flags); \
result = v->counter; \
v->counter c_op i; \
raw_local_irq_restore(flags); \
} \
\
return result; \
}
#define ATOMIC64_OPS(op, c_op, asm_op) \
ATOMIC64_OP(op, c_op, asm_op) \
ATOMIC64_OP_RETURN(op, c_op, asm_op) \
ATOMIC64_FETCH_OP(op, c_op, asm_op)
ATOMIC64_OPS(add, +=, daddu)
ATOMIC64_OPS(sub, -=, dsubu)
#define atomic64_add_return_relaxed atomic64_add_return_relaxed
#define atomic64_sub_return_relaxed atomic64_sub_return_relaxed
#define atomic64_fetch_add_relaxed atomic64_fetch_add_relaxed
#define atomic64_fetch_sub_relaxed atomic64_fetch_sub_relaxed
#undef ATOMIC64_OPS
#define ATOMIC64_OPS(op, c_op, asm_op) \
ATOMIC64_OP(op, c_op, asm_op) \
ATOMIC64_FETCH_OP(op, c_op, asm_op)
ATOMIC64_OPS(and, &=, and)
ATOMIC64_OPS(or, |=, or)
ATOMIC64_OPS(xor, ^=, xor)
#define atomic64_fetch_and_relaxed atomic64_fetch_and_relaxed
#define atomic64_fetch_or_relaxed atomic64_fetch_or_relaxed
#define atomic64_fetch_xor_relaxed atomic64_fetch_xor_relaxed
#undef ATOMIC64_OPS
#undef ATOMIC64_FETCH_OP
#undef ATOMIC64_OP_RETURN
#undef ATOMIC64_OP
/*
* atomic64_sub_if_positive - conditionally subtract integer from atomic
* variable
* @i: integer value to subtract
* @v: pointer of type atomic64_t
*
* Atomically test @v and subtract @i if @v is greater or equal than @i.
* The function returns the old value of @v minus @i.
*/
static __inline__ long atomic64_sub_if_positive(long i, atomic64_t * v)
{
long result;
smp_mb__before_llsc();
if (kernel_uses_llsc) {
long temp;
__asm__ __volatile__(
" .set push \n"
" .set "MIPS_ISA_LEVEL" \n"
"1: lld %1, %2 # atomic64_sub_if_positive\n"
" dsubu %0, %1, %3 \n"
" move %1, %0 \n"
" bltz %0, 1f \n"
" scd %1, %2 \n"
"\t" __scbeqz " %1, 1b \n"
"1: \n"
" .set pop \n"
MIPS: Fix microMIPS LL/SC immediate offsets In the microMIPS encoding some memory access instructions have their immediate offset reduced to 12 bits only. That does not match the GCC `R' constraint we use in some places to satisfy the requirement, resulting in build failures like this: {standard input}: Assembler messages: {standard input}:720: Error: macro used $at after ".set noat" {standard input}:720: Warning: macro instruction expanded into multiple instructions Fix the problem by defining a macro, `GCC_OFF12_ASM', that expands to the right constraint depending on whether microMIPS or standard MIPS code is produced. Also apply the fix to where `m' is used as in the worst case this change does nothing, e.g. where the pointer was already in a register such as a function argument and no further offset was requested, and in the best case it avoids an extraneous sequence of up to two instructions to load the high 20 bits of the address in the LL/SC loop. This reduces the risk of lock contention that is the higher the more instructions there are in the critical section between LL and SC. Strictly speaking we could just bulk-replace `R' with `ZC' as the latter constraint adjusts automatically depending on the ISA selected. However it was only introduced with GCC 4.9 and we keep supporing older compilers for the standard MIPS configuration, hence the slightly more complicated approach I chose. The choice of a zero-argument function-like rather than an object-like macro was made so that it does not look like a function call taking the C expression used for the constraint as an argument. This is so as not to confuse the reader or formatting checkers like `checkpatch.pl' and follows previous practice. Signed-off-by: Maciej W. Rozycki <macro@codesourcery.com> Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/8482/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2014-11-15 15:08:48 -07:00
: "=&r" (result), "=&r" (temp),
"+" GCC_OFF_SMALL_ASM() (v->counter)
: "Ir" (i));
} else {
unsigned long flags;
raw_local_irq_save(flags);
result = v->counter;
result -= i;
if (result >= 0)
v->counter = result;
raw_local_irq_restore(flags);
}
smp_llsc_mb();
return result;
}
#define atomic64_cmpxchg(v, o, n) \
((__typeof__((v)->counter))cmpxchg(&((v)->counter), (o), (n)))
#define atomic64_xchg(v, new) (xchg(&((v)->counter), (new)))
/*
* atomic64_dec_if_positive - decrement by 1 if old value positive
* @v: pointer of type atomic64_t
*/
#define atomic64_dec_if_positive(v) atomic64_sub_if_positive(1, v)
#endif /* CONFIG_64BIT */
#endif /* _ASM_ATOMIC_H */