diff options
| author |   Vineet Gupta <vgupta@synopsys.com> | 2015-07-14 19:50:18 +0530 |
|---|---|---|
| committer | Vineet Gupta <vgupta@synopsys.com> | 2015-08-04 09:26:34 +0530 |
| commit | e78fdfef84be13a5c2b8276e12203cdf24778596 (patch) | |
| tree | e16fad7f1d73a651d2f02deefd8f20f5bf9a018e /arch/arc/include/asm/atomic.h | |
| parent | ARC: LLOCK/SCOND based rwlock (diff) | |
| download | linux-dev-e78fdfef84be13a5c2b8276e12203cdf24778596.tar.xz linux-dev-e78fdfef84be13a5c2b8276e12203cdf24778596.zip | |
ARCv2: spinlock/rwlock/atomics: Delayed retry of failed SCOND with exponential backoff
This is to workaround the llock/scond livelock
HS38x4 could get into a LLOCK/SCOND livelock in case of multiple overlapping
coherency transactions in the SCU. The exclusive line state keeps rotating
among contenting cores leading to a never ending cycle. So break the cycle
by deferring the retry of failed exclusive access (SCOND). The actual delay
needed is function of number of contending cores as well as the unrelated
coherency traffic from other cores. To keep the code simple, start off with
small delay of 1 which would suffice most cases and in case of contention
double the delay. Eventually the delay is sufficient such that the coherency
pipeline is drained, thus a subsequent exclusive access would succeed.
Link: http://lkml.kernel.org/r/1438612568-28265-1-git-send-email-vgupta@synopsys.com
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Diffstat (limited to 'arch/arc/include/asm/atomic.h')
| -rw-r--r-- | arch/arc/include/asm/atomic.h | 49 |
1 files changed, 45 insertions, 4 deletions
diff --git a/arch/arc/include/asm/atomic.h b/arch/arc/include/asm/atomic.h index 3dd36c1efee1..629dfd0a0c6b 100644 --- a/arch/arc/include/asm/atomic.h +++ b/arch/arc/include/asm/atomic.h @@ -23,17 +23,51 @@ #define atomic_set(v, i) (((v)->counter) = (i)) +#ifdef CONFIG_ARC_STAR_9000923308 + +#define SCOND_FAIL_RETRY_VAR_DEF \ + unsigned int delay = 1, tmp; \ + +#define SCOND_FAIL_RETRY_ASM \ + " bz 4f \n" \ + " ; --- scond fail delay --- \n" \ + " mov %[tmp], %[delay] \n" /* tmp = delay */ \ + "2: brne.d %[tmp], 0, 2b \n" /* while (tmp != 0) */ \ + " sub %[tmp], %[tmp], 1 \n" /* tmp-- */ \ + " asl.f %[delay], %[delay], 1 \n" /* delay *= 2 */ \ + " mov.z %[delay], 1 \n" /* handle overflow */ \ + " b 1b \n" /* start over */ \ + "4: ; --- success --- \n" \ + +#define SCOND_FAIL_RETRY_VARS \ + ,[delay] "+&r" (delay),[tmp] "=&r" (tmp) \ + +#else /* !CONFIG_ARC_STAR_9000923308 */ + +#define SCOND_FAIL_RETRY_VAR_DEF + +#define SCOND_FAIL_RETRY_ASM \ + " bnz 1b \n" \ + +#define SCOND_FAIL_RETRY_VARS + +#endif + #define ATOMIC_OP(op, c_op, asm_op) \ static inline void atomic_##op(int i, atomic_t *v) \ { \ - unsigned int val; \ + unsigned int val; \ + SCOND_FAIL_RETRY_VAR_DEF \ \ __asm__ __volatile__( \ "1: llock %[val], [%[ctr]] \n" \ " " #asm_op " %[val], %[val], %[i] \n" \ " scond %[val], [%[ctr]] \n" \ - " bnz 1b \n" \ + " \n" \ + SCOND_FAIL_RETRY_ASM \ + \ : [val] "=&r" (val) /* Early clobber to prevent reg reuse */ \ + SCOND_FAIL_RETRY_VARS \ : [ctr] "r" (&v->counter), /* Not "m": llock only supports reg direct addr mode */ \ [i] "ir" (i) \ : "cc"); \ @@ -42,7 +76,8 @@ static inline void atomic_##op(int i, atomic_t *v) \ #define ATOMIC_OP_RETURN(op, c_op, asm_op) \ static inline int atomic_##op##_return(int i, atomic_t *v) \ { \ - unsigned int val; \ + unsigned int val; \ + SCOND_FAIL_RETRY_VAR_DEF \ \ /* \ * Explicit full memory barrier needed before/after as \ @@ -54,8 +89,11 @@ static inline int atomic_##op##_return(int i, atomic_t *v) \ "1: llock %[val], [%[ctr]] \n" \ " " #asm_op " %[val], %[val], %[i] \n" \ " scond %[val], [%[ctr]] \n" \ - " bnz 1b \n" \ + " \n" \ + SCOND_FAIL_RETRY_ASM \ + \ : [val] "=&r" (val) \ + SCOND_FAIL_RETRY_VARS \ : [ctr] "r" (&v->counter), \ [i] "ir" (i) \ : "cc"); \ @@ -142,6 +180,9 @@ ATOMIC_OP(and, &=, and) #undef ATOMIC_OPS #undef ATOMIC_OP_RETURN #undef ATOMIC_OP +#undef SCOND_FAIL_RETRY_VAR_DEF +#undef SCOND_FAIL_RETRY_ASM +#undef SCOND_FAIL_RETRY_VARS /** * __atomic_add_unless - add unless the number is a given value |