search for: 4109679a

Displaying 3 results from an estimated 3 matches for "4109679a".

Did you mean: 1109679
2016 May 17
0
[PATCH v2 3/6] powerpc: lib/locks.c: cpu yield/wake helper function
...h_rwlock_t *lock); #else /* SPLPAR */ #define __spin_yield(x) barrier() +#define __spin_yield_cpu(x) barrier() +#define __spin_wake_cpu(x) barrier() #define __rw_yield(x) barrier() #define SHARED_PROCESSOR 0 #endif diff --git a/arch/powerpc/lib/locks.c b/arch/powerpc/lib/locks.c index a9ebd71..4109679a 100644 --- a/arch/powerpc/lib/locks.c +++ b/arch/powerpc/lib/locks.c @@ -23,6 +23,38 @@ #include <asm/hvcall.h> #include <asm/smp.h> +void __spin_yield_cpu(int cpu) +{ + unsigned int holder_cpu = cpu, yield_count; + + if (cpu == -1) { + plpar_hcall_norets(H_CEDE); + return; + } +...
2016 May 17
6
[PATCH v3 0/6] powerpc use pv-qpsinlock instead of spinlock
change fome v1: separate into 6 pathes from one patch some minor code changes. benchmark test results are below. run 3 tests on pseries IBM,8408-E8E with 32cpus, 64GB memory perf bench futex hash perf bench futex lock-pi perf record -advRT || perf bench sched messaging -g 1000 || perf report summary: _____test________________spinlcok______________pv-qspinlcok_____ |futex hash | 556370 ops |
2016 May 17
6
[PATCH v3 0/6] powerpc use pv-qpsinlock instead of spinlock
change fome v1: separate into 6 pathes from one patch some minor code changes. benchmark test results are below. run 3 tests on pseries IBM,8408-E8E with 32cpus, 64GB memory perf bench futex hash perf bench futex lock-pi perf record -advRT || perf bench sched messaging -g 1000 || perf report summary: _____test________________spinlcok______________pv-qspinlcok_____ |futex hash | 556370 ops |