X-Git-Url: http://git.liburcu.org/?a=blobdiff_plain;f=urcu%2Farch%2Fppc.h;h=95393ea0be0ee89b888c64ad53322e41e30631c8;hb=48a8832be16833b1a9f2625799ac3d0ea4dafc91;hp=a03d688837344f0bb2d438c7161d43dcb208a5ef;hpb=bce63dfd0a2306452c9e39f5df01789e77f3f44a;p=urcu.git diff --git a/urcu/arch/ppc.h b/urcu/arch/ppc.h index a03d688..95393ea 100644 --- a/urcu/arch/ppc.h +++ b/urcu/arch/ppc.h @@ -32,26 +32,49 @@ extern "C" { /* Include size of POWER5+ L3 cache lines: 256 bytes */ #define CAA_CACHE_LINE_SIZE 256 -#define cmm_mb() asm volatile("sync":::"memory") +#ifdef __NO_LWSYNC__ +#define LWSYNC_OPCODE "sync\n" +#else +#define LWSYNC_OPCODE "lwsync\n" +#endif + +/* + * Use sync for all cmm_mb/rmb/wmb barriers because lwsync does not + * preserve ordering of cacheable vs. non-cacheable accesses, so it + * should not be used to order with respect to MMIO operations. An + * eieio+lwsync pair is also not enough for cmm_rmb, because it will + * order cacheable and non-cacheable memory operations separately---i.e. + * not the latter against the former. + */ +#define cmm_mb() __asm__ __volatile__ ("sync":::"memory") + +/* + * lwsync orders loads in cacheable memory with respect to other loads, + * and stores in cacheable memory with respect to other stores. + * Therefore, use it for barriers ordering accesses to cacheable memory + * only. + */ +#define cmm_smp_rmb() __asm__ __volatile__ (LWSYNC_OPCODE:::"memory") +#define cmm_smp_wmb() __asm__ __volatile__ (LWSYNC_OPCODE:::"memory") #define mftbl() \ ({ \ unsigned long rval; \ - asm volatile("mftbl %0" : "=r" (rval)); \ + __asm__ __volatile__ ("mftbl %0" : "=r" (rval)); \ rval; \ }) #define mftbu() \ ({ \ unsigned long rval; \ - asm volatile("mftbu %0" : "=r" (rval)); \ + __asm__ __volatile__ ("mftbu %0" : "=r" (rval)); \ rval; \ }) #define mftb() \ ({ \ unsigned long long rval; \ - asm volatile("mftb %0" : "=r" (rval)); \ + __asm__ __volatile__ ("mftb %0" : "=r" (rval)); \ rval; \ })