From 39157ec04095ab012d11db23c462844634bfbb8f Mon Sep 17 00:00:00 2001 From: Luca Boccassi Date: Mon, 23 Apr 2018 14:16:57 +0100 Subject: New upstream version 16.11.5 Change-Id: I47171042629a57c6958d50251351e668ca5f3d8b Signed-off-by: Luca Boccassi --- .../common/include/arch/ppc_64/rte_atomic.h | 8 ++-- .../common/include/arch/x86/rte_atomic.h | 44 +++++++++++++++++++++- 2 files changed, 46 insertions(+), 6 deletions(-) (limited to 'lib/librte_eal/common/include/arch') diff --git a/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h b/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h index fb4fccb4..37f5eff2 100644 --- a/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h +++ b/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h @@ -64,9 +64,9 @@ extern "C" { * occur before the STORE operations generated after. */ #ifdef RTE_ARCH_64 -#define rte_wmb() {asm volatile("lwsync" : : : "memory"); } +#define rte_wmb() asm volatile("lwsync" : : : "memory") #else -#define rte_wmb() {asm volatile("sync" : : : "memory"); } +#define rte_wmb() asm volatile("sync" : : : "memory") #endif /** @@ -76,9 +76,9 @@ extern "C" { * occur before the LOAD operations generated after. */ #ifdef RTE_ARCH_64 -#define rte_rmb() {asm volatile("lwsync" : : : "memory"); } +#define rte_rmb() asm volatile("lwsync" : : : "memory") #else -#define rte_rmb() {asm volatile("sync" : : : "memory"); } +#define rte_rmb() asm volatile("sync" : : : "memory") #endif #define rte_smp_mb() rte_mb() diff --git a/lib/librte_eal/common/include/arch/x86/rte_atomic.h b/lib/librte_eal/common/include/arch/x86/rte_atomic.h index 00b1cdf5..d12b679a 100644 --- a/lib/librte_eal/common/include/arch/x86/rte_atomic.h +++ b/lib/librte_eal/common/include/arch/x86/rte_atomic.h @@ -55,12 +55,52 @@ extern "C" { #define rte_rmb() _mm_lfence() -#define rte_smp_mb() rte_mb() - #define rte_smp_wmb() rte_compiler_barrier() #define rte_smp_rmb() rte_compiler_barrier() +/* + * From Intel Software Development Manual; Vol 3; + * 8.2.2 Memory Ordering in P6 and More Recent Processor Families: + * ... + * . Reads are not reordered with other reads. + * . Writes are not reordered with older reads. + * . Writes to memory are not reordered with other writes, + * with the following exceptions: + * . streaming stores (writes) executed with the non-temporal move + * instructions (MOVNTI, MOVNTQ, MOVNTDQ, MOVNTPS, and MOVNTPD); and + * . string operations (see Section 8.2.4.1). + * ... + * . Reads may be reordered with older writes to different locations but not + * with older writes to the same location. + * . Reads or writes cannot be reordered with I/O instructions, + * locked instructions, or serializing instructions. + * . Reads cannot pass earlier LFENCE and MFENCE instructions. + * . Writes ... cannot pass earlier LFENCE, SFENCE, and MFENCE instructions. + * . LFENCE instructions cannot pass earlier reads. + * . SFENCE instructions cannot pass earlier writes ... + * . MFENCE instructions cannot pass earlier reads, writes ... + * + * As pointed by Java guys, that makes possible to use lock-prefixed + * instructions to get the same effect as mfence and on most modern HW + * that gives a better perfomance then using mfence: + * https://shipilev.net/blog/2014/on-the-fence-with-dependencies/ + * Basic idea is to use lock prefixed add with some dummy memory location + * as the destination. From their experiments 128B(2 cache lines) below + * current stack pointer looks like a good candidate. + * So below we use that techinque for rte_smp_mb() implementation. + */ + +static inline void __attribute__((always_inline)) +rte_smp_mb(void) +{ +#ifdef RTE_ARCH_I686 + asm volatile("lock addl $0, -128(%%esp); " ::: "memory"); +#else + asm volatile("lock addl $0, -128(%%rsp); " ::: "memory"); +#endif +} + /*------------------------- 16 bit atomic operations -------------------------*/ #ifndef RTE_FORCE_INTRINSICS -- cgit 1.2.3-korg