diff options
Diffstat (limited to 'arch/sparc64/kernel/smp.c')
| -rw-r--r-- | arch/sparc64/kernel/smp.c | 64 |
1 files changed, 44 insertions, 20 deletions
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c index d1adeb2c7..2fa4945d8 100644 --- a/arch/sparc64/kernel/smp.c +++ b/arch/sparc64/kernel/smp.c @@ -94,7 +94,8 @@ void __init smp_store_cpu_info(int id) cpu_data[id].udelay_val = loops_per_sec; cpu_data[id].pgcache_size = 0; - cpu_data[id].pte_cache = NULL; + cpu_data[id].pte_cache[0] = NULL; + cpu_data[id].pte_cache[1] = NULL; cpu_data[id].pgdcache_size = 0; cpu_data[id].pgd_cache = NULL; cpu_data[id].idle_volume = 1; @@ -184,7 +185,7 @@ void cpu_panic(void) extern struct prom_cpuinfo linux_cpus[64]; -extern unsigned long smp_trampoline; +extern unsigned long sparc64_cpu_startup; /* The OBP cpu startup callback truncates the 3rd arg cookie to * 32-bits (I think) so to be safe we have it read the pointer @@ -210,15 +211,13 @@ void __init smp_boot_cpus(void) continue; if(cpu_present_map & (1UL << i)) { - unsigned long entry = (unsigned long)(&smp_trampoline); + unsigned long entry = (unsigned long)(&sparc64_cpu_startup); unsigned long cookie = (unsigned long)(&cpu_new_task); struct task_struct *p; int timeout; int no; - extern unsigned long phys_base; - entry += phys_base - KERNBASE; - cookie += phys_base - KERNBASE; + prom_printf("Starting CPU %d... ", i); kernel_thread(start_secondary, NULL, CLONE_PID); cpucount++; @@ -247,9 +246,11 @@ void __init smp_boot_cpus(void) cpu_number_map[i] = cpucount; __cpu_logical_map[cpucount] = i; prom_cpu_nodes[i] = linux_cpus[no].prom_node; + prom_printf("OK\n"); } else { cpucount--; printk("Processor %d is stuck.\n", i); + prom_printf("FAILED\n"); } } if(!callin_flag) { @@ -537,14 +538,31 @@ void smp_release(void) /* Imprisoned penguins run with %pil == 15, but PSTATE_IE set, so they * can service tlb flush xcalls... */ +extern void prom_world(int); +extern void save_alternate_globals(unsigned long *); +extern void restore_alternate_globals(unsigned long *); void smp_penguin_jailcell(void) { - flushw_user(); + unsigned long global_save[24]; + + __asm__ __volatile__("flushw"); + save_alternate_globals(global_save); + prom_world(1); atomic_inc(&smp_capture_registry); membar("#StoreLoad | #StoreStore"); while(penguins_are_doing_time) membar("#LoadLoad"); + restore_alternate_globals(global_save); atomic_dec(&smp_capture_registry); + prom_world(0); +} + +extern unsigned long xcall_promstop; + +void smp_promstop_others(void) +{ + if (smp_processors_ready) + smp_cross_call(&xcall_promstop, 0, 0, 0); } static inline void sparc64_do_profile(unsigned long pc, unsigned long g3) @@ -701,14 +719,13 @@ static inline unsigned long find_flush_base(unsigned long size) /* Failure. */ if(p >= (mem_map + max_mapnr)) return 0UL; - if(PageSkip(p)) { - p = p->next_hash; - base = page_address(p); + if(PageReserved(p)) { found = size; + base = page_address(p); } else { found -= PAGE_SIZE; - p++; } + p++; } return base; } @@ -718,7 +735,7 @@ cycles_t cacheflush_time; static void __init smp_tune_scheduling (void) { unsigned long flush_base, flags, *p; - unsigned int ecache_size; + unsigned int ecache_size, order; cycles_t tick1, tick2, raw; /* Approximate heuristic for SMP scheduling. It is an @@ -733,18 +750,22 @@ static void __init smp_tune_scheduling (void) */ printk("SMP: Calibrating ecache flush... "); ecache_size = prom_getintdefault(linux_cpus[0].prom_node, - "ecache-size", (512 *1024)); - flush_base = find_flush_base(ecache_size << 1); - - if(flush_base != 0UL) { + "ecache-size", (512 * 1024)); + if (ecache_size > (4 * 1024 * 1024)) + ecache_size = (4 * 1024 * 1024); + for (order = 0UL; (PAGE_SIZE << order) < ecache_size; order++) + ; + flush_base = __get_free_pages(GFP_KERNEL, order); + + if (flush_base != 0UL) { __save_and_cli(flags); /* Scan twice the size once just to get the TLB entries * loaded and make sure the second scan measures pure misses. */ - for(p = (unsigned long *)flush_base; - ((unsigned long)p) < (flush_base + (ecache_size<<1)); - p += (64 / sizeof(unsigned long))) + for (p = (unsigned long *)flush_base; + ((unsigned long)p) < (flush_base + (ecache_size<<1)); + p += (64 / sizeof(unsigned long))) *((volatile unsigned long *)p); /* Now the real measurement. */ @@ -775,9 +796,12 @@ static void __init smp_tune_scheduling (void) * sharing the cache and fitting. */ cacheflush_time = (raw - (raw >> 2)); - } else + + free_pages(flush_base, order); + } else { cacheflush_time = ((ecache_size << 2) + (ecache_size << 1)); + } printk("Using heuristic of %d cycles.\n", (int) cacheflush_time); |