diff options
Diffstat (limited to 'arch/arc/include/asm')
| -rw-r--r-- | arch/arc/include/asm/Kbuild | 1 | ||||
| -rw-r--r-- | arch/arc/include/asm/arcregs.h | 1 | ||||
| -rw-r--r-- | arch/arc/include/asm/atomic.h | 8 | ||||
| -rw-r--r-- | arch/arc/include/asm/cache.h | 8 | ||||
| -rw-r--r-- | arch/arc/include/asm/cmpxchg.h | 22 | ||||
| -rw-r--r-- | arch/arc/include/asm/futex.h | 72 | ||||
| -rw-r--r-- | arch/arc/include/asm/perf_event.h | 23 |
7 files changed, 87 insertions, 48 deletions
diff --git a/arch/arc/include/asm/Kbuild b/arch/arc/include/asm/Kbuild index 7611b10a2..0b10ef2a4 100644 --- a/arch/arc/include/asm/Kbuild +++ b/arch/arc/include/asm/Kbuild @@ -48,4 +48,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h index c8f57b844..d8023bc8d 100644 --- a/arch/arc/include/asm/arcregs.h +++ b/arch/arc/include/asm/arcregs.h @@ -35,6 +35,7 @@ #define ARC_REG_RTT_BCR 0xF2 #define ARC_REG_IRQ_BCR 0xF3 #define ARC_REG_SMART_BCR 0xFF +#define ARC_REG_CLUSTER_BCR 0xcf /* status32 Bits Positions */ #define STATUS_AE_BIT 5 /* Exception active */ diff --git a/arch/arc/include/asm/atomic.h b/arch/arc/include/asm/atomic.h index 87d18ae53..c3ecda023 100644 --- a/arch/arc/include/asm/atomic.h +++ b/arch/arc/include/asm/atomic.h @@ -172,9 +172,13 @@ static inline int atomic_##op##_return(int i, atomic_t *v) \ ATOMIC_OPS(add, +=, add) ATOMIC_OPS(sub, -=, sub) -ATOMIC_OP(and, &=, and) -#define atomic_clear_mask(mask, v) atomic_and(~(mask), (v)) +#define atomic_andnot atomic_andnot + +ATOMIC_OP(and, &=, and) +ATOMIC_OP(andnot, &= ~, bic) +ATOMIC_OP(or, |=, or) +ATOMIC_OP(xor, ^=, xor) #undef ATOMIC_OPS #undef ATOMIC_OP_RETURN diff --git a/arch/arc/include/asm/cache.h b/arch/arc/include/asm/cache.h index d67345d3e..e23ea6e76 100644 --- a/arch/arc/include/asm/cache.h +++ b/arch/arc/include/asm/cache.h @@ -53,6 +53,8 @@ extern void arc_cache_init(void); extern char *arc_cache_mumbojumbo(int cpu_id, char *buf, int len); extern void read_decode_cache_bcr(void); +extern int ioc_exists; + #endif /* !__ASSEMBLY__ */ /* Instruction cache related Auxiliary registers */ @@ -94,4 +96,10 @@ extern void read_decode_cache_bcr(void); #define SLC_CTRL_BUSY 0x100 #define SLC_CTRL_RGN_OP_INV 0x200 +/* IO coherency related Auxiliary registers */ +#define ARC_REG_IO_COH_ENABLE 0x500 +#define ARC_REG_IO_COH_PARTIAL 0x501 +#define ARC_REG_IO_COH_AP0_BASE 0x508 +#define ARC_REG_IO_COH_AP0_SIZE 0x509 + #endif /* _ASM_CACHE_H */ diff --git a/arch/arc/include/asm/cmpxchg.h b/arch/arc/include/asm/cmpxchg.h index 44fd531f4..af7a2db13 100644 --- a/arch/arc/include/asm/cmpxchg.h +++ b/arch/arc/include/asm/cmpxchg.h @@ -110,18 +110,18 @@ static inline unsigned long __xchg(unsigned long val, volatile void *ptr, sizeof(*(ptr)))) /* - * On ARC700, EX insn is inherently atomic, so by default "vanilla" xchg() need - * not require any locking. However there's a quirk. - * ARC lacks native CMPXCHG, thus emulated (see above), using external locking - - * incidently it "reuses" the same atomic_ops_lock used by atomic APIs. - * Now, llist code uses cmpxchg() and xchg() on same data, so xchg() needs to - * abide by same serializing rules, thus ends up using atomic_ops_lock as well. + * xchg() maps directly to ARC EX instruction which guarantees atomicity. + * However in !LLSC config, it also needs to be use @atomic_ops_lock spinlock + * due to a subtle reason: + * - For !LLSC, cmpxchg() needs to use that lock (see above) and there is lot + * of kernel code which calls xchg()/cmpxchg() on same data (see llist.h) + * Hence xchg() needs to follow same locking rules. * - * This however is only relevant if SMP and/or ARC lacks LLSC - * if (UP or LLSC) - * xchg doesn't need serialization - * else <==> !(UP or LLSC) <==> (!UP and !LLSC) <==> (SMP and !LLSC) - * xchg needs serialization + * Technically the lock is also needed for UP (boils down to irq save/restore) + * but we can cheat a bit since cmpxchg() atomic_ops_lock() would cause irqs to + * be disabled thus can't possibly be interrpted/preempted/clobbered by xchg() + * Other way around, xchg is one instruction anyways, so can't be interrupted + * as such */ #if !defined(CONFIG_ARC_HAS_LLSC) && defined(CONFIG_SMP) diff --git a/arch/arc/include/asm/futex.h b/arch/arc/include/asm/futex.h index 70cfe16b7..11e1b1f3a 100644 --- a/arch/arc/include/asm/futex.h +++ b/arch/arc/include/asm/futex.h @@ -20,6 +20,7 @@ #define __futex_atomic_op(insn, ret, oldval, uaddr, oparg)\ \ + smp_mb(); \ __asm__ __volatile__( \ "1: llock %1, [%2] \n" \ insn "\n" \ @@ -30,7 +31,7 @@ " .section .fixup,\"ax\" \n" \ " .align 4 \n" \ "4: mov %0, %4 \n" \ - " b 3b \n" \ + " j 3b \n" \ " .previous \n" \ " .section __ex_table,\"a\" \n" \ " .align 4 \n" \ @@ -40,12 +41,14 @@ \ : "=&r" (ret), "=&r" (oldval) \ : "r" (uaddr), "r" (oparg), "ir" (-EFAULT) \ - : "cc", "memory") + : "cc", "memory"); \ + smp_mb() \ #else /* !CONFIG_ARC_HAS_LLSC */ #define __futex_atomic_op(insn, ret, oldval, uaddr, oparg)\ \ + smp_mb(); \ __asm__ __volatile__( \ "1: ld %1, [%2] \n" \ insn "\n" \ @@ -55,7 +58,7 @@ " .section .fixup,\"ax\" \n" \ " .align 4 \n" \ "4: mov %0, %4 \n" \ - " b 3b \n" \ + " j 3b \n" \ " .previous \n" \ " .section __ex_table,\"a\" \n" \ " .align 4 \n" \ @@ -65,7 +68,8 @@ \ : "=&r" (ret), "=&r" (oldval) \ : "r" (uaddr), "r" (oparg), "ir" (-EFAULT) \ - : "cc", "memory") + : "cc", "memory"); \ + smp_mb() \ #endif @@ -83,6 +87,9 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; +#ifndef CONFIG_ARC_HAS_LLSC + preempt_disable(); /* to guarantee atomic r-m-w of futex op */ +#endif pagefault_disable(); switch (op) { @@ -90,6 +97,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) __futex_atomic_op("mov %0, %3", ret, oldval, uaddr, oparg); break; case FUTEX_OP_ADD: + /* oldval = *uaddr; *uaddr += oparg ; ret = *uaddr */ __futex_atomic_op("add %0, %1, %3", ret, oldval, uaddr, oparg); break; case FUTEX_OP_OR: @@ -106,6 +114,9 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) } pagefault_enable(); +#ifndef CONFIG_ARC_HAS_LLSC + preempt_enable(); +#endif if (!ret) { switch (cmp) { @@ -134,54 +145,57 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) return ret; } -/* Compare-xchg with pagefaults disabled. - * Notes: - * -Best-Effort: Exchg happens only if compare succeeds. - * If compare fails, returns; leaving retry/looping to upper layers - * -successful cmp-xchg: return orig value in @addr (same as cmp val) - * -Compare fails: return orig value in @addr - * -user access r/w fails: return -EFAULT +/* + * cmpxchg of futex (pagefaults disabled by caller) + * Return 0 for success, -EFAULT otherwise */ static inline int -futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, u32 oldval, - u32 newval) +futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, u32 expval, + u32 newval) { - u32 val; + int ret = 0; + u32 existval; - if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) + if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))) return -EFAULT; - pagefault_disable(); +#ifndef CONFIG_ARC_HAS_LLSC + preempt_disable(); /* to guarantee atomic r-m-w of futex op */ +#endif + smp_mb(); __asm__ __volatile__( #ifdef CONFIG_ARC_HAS_LLSC - "1: llock %0, [%3] \n" - " brne %0, %1, 3f \n" - "2: scond %2, [%3] \n" + "1: llock %1, [%4] \n" + " brne %1, %2, 3f \n" + "2: scond %3, [%4] \n" " bnz 1b \n" #else - "1: ld %0, [%3] \n" - " brne %0, %1, 3f \n" - "2: st %2, [%3] \n" + "1: ld %1, [%4] \n" + " brne %1, %2, 3f \n" + "2: st %3, [%4] \n" #endif "3: \n" " .section .fixup,\"ax\" \n" - "4: mov %0, %4 \n" - " b 3b \n" + "4: mov %0, %5 \n" + " j 3b \n" " .previous \n" " .section __ex_table,\"a\" \n" " .align 4 \n" " .word 1b, 4b \n" " .word 2b, 4b \n" " .previous\n" - : "=&r"(val) - : "r"(oldval), "r"(newval), "r"(uaddr), "ir"(-EFAULT) + : "+&r"(ret), "=&r"(existval) + : "r"(expval), "r"(newval), "r"(uaddr), "ir"(-EFAULT) : "cc", "memory"); - pagefault_enable(); + smp_mb(); - *uval = val; - return val; +#ifndef CONFIG_ARC_HAS_LLSC + preempt_enable(); +#endif + *uval = existval; + return ret; } #endif diff --git a/arch/arc/include/asm/perf_event.h b/arch/arc/include/asm/perf_event.h index 2b8880e95..5f071762f 100644 --- a/arch/arc/include/asm/perf_event.h +++ b/arch/arc/include/asm/perf_event.h @@ -1,6 +1,7 @@ /* * Linux performance counter support for ARC * + * Copyright (C) 2014-2015 Synopsys, Inc. (www.synopsys.com) * Copyright (C) 2011-2013 Synopsys, Inc. (www.synopsys.com) * * This program is free software; you can redistribute it and/or modify @@ -12,8 +13,8 @@ #ifndef __ASM_PERF_EVENT_H #define __ASM_PERF_EVENT_H -/* real maximum varies per CPU, this is the maximum supported by the driver */ -#define ARC_PMU_MAX_HWEVENTS 64 +/* Max number of counters that PCT block may ever have */ +#define ARC_PERF_MAX_COUNTERS 32 #define ARC_REG_CC_BUILD 0xF6 #define ARC_REG_CC_INDEX 0x240 @@ -28,15 +29,22 @@ #define ARC_REG_PCT_CONFIG 0x254 #define ARC_REG_PCT_CONTROL 0x255 #define ARC_REG_PCT_INDEX 0x256 +#define ARC_REG_PCT_INT_CNTL 0x25C +#define ARC_REG_PCT_INT_CNTH 0x25D +#define ARC_REG_PCT_INT_CTRL 0x25E +#define ARC_REG_PCT_INT_ACT 0x25F + +#define ARC_REG_PCT_CONFIG_USER (1 << 18) /* count in user mode */ +#define ARC_REG_PCT_CONFIG_KERN (1 << 19) /* count in kernel mode */ #define ARC_REG_PCT_CONTROL_CC (1 << 16) /* clear counts */ #define ARC_REG_PCT_CONTROL_SN (1 << 17) /* snapshot */ struct arc_reg_pct_build { #ifdef CONFIG_CPU_BIG_ENDIAN - unsigned int m:8, c:8, r:6, s:2, v:8; + unsigned int m:8, c:8, r:5, i:1, s:2, v:8; #else - unsigned int v:8, s:2, r:6, c:8, m:8; + unsigned int v:8, s:2, i:1, r:5, c:8, m:8; #endif }; @@ -95,10 +103,13 @@ static const char * const arc_pmu_ev_hw_map[] = { /* counts condition */ [PERF_COUNT_HW_INSTRUCTIONS] = "iall", - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = "ijmp", + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = "ijmp", /* Excludes ZOL jumps */ [PERF_COUNT_ARC_BPOK] = "bpok", /* NP-NT, PT-T, PNT-NT */ +#ifdef CONFIG_ISA_ARCV2 + [PERF_COUNT_HW_BRANCH_MISSES] = "bpmp", +#else [PERF_COUNT_HW_BRANCH_MISSES] = "bpfail", /* NP-T, PT-NT, PNT-T */ - +#endif [PERF_COUNT_ARC_LDC] = "imemrdc", /* Instr: mem read cached */ [PERF_COUNT_ARC_STC] = "imemwrc", /* Instr: mem write cached */ |