diff options
Diffstat (limited to 'arch/sparc/kernel')
61 files changed, 962 insertions, 881 deletions
diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile index c6316142db4..0c2dc1f24a9 100644 --- a/arch/sparc/kernel/Makefile +++ b/arch/sparc/kernel/Makefile @@ -13,6 +13,14 @@ extra-y += init_task.o CPPFLAGS_vmlinux.lds := -Usparc -m$(BITS) extra-y += vmlinux.lds +ifdef CONFIG_FUNCTION_TRACER +# Do not profile debug and lowlevel utilities +CFLAGS_REMOVE_ftrace.o := -pg +CFLAGS_REMOVE_time_$(BITS).o := -pg +CFLAGS_REMOVE_perf_event.o := -pg +CFLAGS_REMOVE_pcr.o := -pg +endif + obj-$(CONFIG_SPARC32) += entry.o wof.o wuf.o obj-$(CONFIG_SPARC32) += etrap_32.o obj-$(CONFIG_SPARC32) += rtrap_32.o @@ -85,7 +93,7 @@ obj-$(CONFIG_KGDB) += kgdb_$(BITS).o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o -CFLAGS_REMOVE_ftrace.o := -pg +obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_EARLYFB) += btext.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/arch/sparc/kernel/central.c b/arch/sparc/kernel/central.c index f3b5466c389..415c86d5a8d 100644 --- a/arch/sparc/kernel/central.c +++ b/arch/sparc/kernel/central.c @@ -5,6 +5,7 @@ #include <linux/kernel.h> #include <linux/types.h> +#include <linux/slab.h> #include <linux/string.h> #include <linux/init.h> #include <linux/of_device.h> @@ -99,7 +100,7 @@ static int __devinit clock_board_probe(struct of_device *op, p->leds_resource.start = (unsigned long) (p->clock_regs + CLOCK_CTRL); - p->leds_resource.end = p->leds_resource.end; + p->leds_resource.end = p->leds_resource.start; p->leds_resource.name = "leds"; p->leds_pdev.name = "sunfire-clockboard-leds"; @@ -194,7 +195,7 @@ static int __devinit fhc_probe(struct of_device *op, if (!p->central) { p->leds_resource.start = (unsigned long) (p->pregs + FHC_PREGS_CTRL); - p->leds_resource.end = p->leds_resource.end; + p->leds_resource.end = p->leds_resource.start; p->leds_resource.name = "leds"; p->leds_pdev.name = "sunfire-fhc-leds"; diff --git a/arch/sparc/kernel/cpumap.c b/arch/sparc/kernel/cpumap.c index 7430ed080b2..8de64c8126b 100644 --- a/arch/sparc/kernel/cpumap.c +++ b/arch/sparc/kernel/cpumap.c @@ -4,6 +4,7 @@ */ #include <linux/module.h> +#include <linux/slab.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/cpumask.h> diff --git a/arch/sparc/kernel/devices.c b/arch/sparc/kernel/devices.c index b171ae8de90..62dc7a02141 100644 --- a/arch/sparc/kernel/devices.c +++ b/arch/sparc/kernel/devices.c @@ -59,7 +59,7 @@ static int __cpu_find_by(int (*compare)(int, int, void *), void *compare_arg, cur_inst = 0; for_each_node_by_type(dp, "cpu") { - int err = check_cpu_node(dp->node, &cur_inst, + int err = check_cpu_node(dp->phandle, &cur_inst, compare, compare_arg, prom_node, mid); if (!err) { @@ -143,6 +143,4 @@ void __init device_scan(void) if (ARCH_SUN4C) sun4c_probe_memerr_reg(); - - return; } diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h index 4f53a2395ac..c011b932bb1 100644 --- a/arch/sparc/kernel/entry.h +++ b/arch/sparc/kernel/entry.h @@ -48,7 +48,6 @@ extern void __init boot_cpu_id_too_large(int cpu); extern unsigned int dcache_parity_tl1_occurred; extern unsigned int icache_parity_tl1_occurred; -extern asmlinkage void update_perfctrs(void); extern asmlinkage void sparc_breakpoint(struct pt_regs *regs); extern void timer_interrupt(int irq, struct pt_regs *regs); diff --git a/arch/sparc/kernel/ftrace.c b/arch/sparc/kernel/ftrace.c index 29973daa993..03ab022e51c 100644 --- a/arch/sparc/kernel/ftrace.c +++ b/arch/sparc/kernel/ftrace.c @@ -13,7 +13,7 @@ static const u32 ftrace_nop = 0x01000000; static u32 ftrace_call_replace(unsigned long ip, unsigned long addr) { - static u32 call; + u32 call; s32 off; off = ((s32)addr - (s32)ip); @@ -92,13 +92,60 @@ int __init ftrace_dyn_arch_init(void *data) } #endif -#ifdef CONFIG_FTRACE_SYSCALLS +#ifdef CONFIG_FUNCTION_GRAPH_TRACER -extern unsigned int sys_call_table[]; +#ifdef CONFIG_DYNAMIC_FTRACE +extern void ftrace_graph_call(void); -unsigned long __init arch_syscall_addr(int nr) +int ftrace_enable_ftrace_graph_caller(void) { - return (unsigned long)sys_call_table[nr]; + unsigned long ip = (unsigned long)(&ftrace_graph_call); + u32 old, new; + + old = *(u32 *) &ftrace_graph_call; + new = ftrace_call_replace(ip, (unsigned long) &ftrace_graph_caller); + return ftrace_modify_code(ip, old, new); } -#endif +int ftrace_disable_ftrace_graph_caller(void) +{ + unsigned long ip = (unsigned long)(&ftrace_graph_call); + u32 old, new; + + old = *(u32 *) &ftrace_graph_call; + new = ftrace_call_replace(ip, (unsigned long) &ftrace_stub); + + return ftrace_modify_code(ip, old, new); +} + +#endif /* !CONFIG_DYNAMIC_FTRACE */ + +/* + * Hook the return address and push it in the stack of return addrs + * in current thread info. + */ +unsigned long prepare_ftrace_return(unsigned long parent, + unsigned long self_addr, + unsigned long frame_pointer) +{ + unsigned long return_hooker = (unsigned long) &return_to_handler; + struct ftrace_graph_ent trace; + + if (unlikely(atomic_read(¤t->tracing_graph_pause))) + return parent + 8UL; + + if (ftrace_push_return_trace(parent, self_addr, &trace.depth, + frame_pointer) == -EBUSY) + return parent + 8UL; + + trace.func = self_addr; + + /* Only trace if the calling function expects to */ + if (!ftrace_graph_entry(&trace)) { + current->curr_ret_stack--; + return parent + 8UL; + } + + return return_hooker; +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/arch/sparc/kernel/helpers.S b/arch/sparc/kernel/helpers.S index 314dd0c9fc5..92090cc9e82 100644 --- a/arch/sparc/kernel/helpers.S +++ b/arch/sparc/kernel/helpers.S @@ -46,6 +46,81 @@ stack_trace_flush: nop .size stack_trace_flush,.-stack_trace_flush +#ifdef CONFIG_PERF_EVENTS + .globl perf_arch_fetch_caller_regs + .type perf_arch_fetch_caller_regs,#function +perf_arch_fetch_caller_regs: + /* We always read the %pstate into %o5 since we will use + * that to construct a fake %tstate to store into the regs. + */ + rdpr %pstate, %o5 + brz,pn %o2, 50f + mov %o2, %g7 + + /* Turn off interrupts while we walk around the register + * window by hand. + */ + wrpr %o5, PSTATE_IE, %pstate + + /* The %canrestore tells us how many register windows are + * still live in the chip above us, past that we have to + * walk the frame as saved on the stack. We stash away + * the %cwp in %g1 so we can return back to the original + * register window. + */ + rdpr %cwp, %g1 + rdpr %canrestore, %g2 + sub %g1, 1, %g3 + + /* We have the skip count in %g7, if it hits zero then + * %fp/%i7 are the registers we need. Otherwise if our + * %canrestore count maintained in %g2 hits zero we have + * to start traversing the stack. + */ +10: brz,pn %g2, 4f + sub %g2, 1, %g2 + wrpr %g3, %cwp + subcc %g7, 1, %g7 + bne,pt %xcc, 10b + sub %g3, 1, %g3 + + /* We found the values we need in the cpu's register + * windows. + */ + mov %fp, %g3 + ba,pt %xcc, 3f + mov %i7, %g2 + +50: mov %fp, %g3 + ba,pt %xcc, 2f + mov %i7, %g2 + + /* We hit the end of the valid register windows in the + * cpu, start traversing the stack frame. + */ +4: mov %fp, %g3 + +20: ldx [%g3 + STACK_BIAS + RW_V9_I7], %g2 + subcc %g7, 1, %g7 + bne,pn %xcc, 20b + ldx [%g3 + STACK_BIAS + RW_V9_I6], %g3 + + /* Restore the current register window position and + * re-enable interrupts. + */ +3: wrpr %g1, %cwp + wrpr %o5, %pstate + +2: stx %g3, [%o0 + PT_V9_FP] + sllx %o5, 8, %o5 + stx %o5, [%o0 + PT_V9_TSTATE] + stx %g2, [%o0 + PT_V9_TPC] + add %g2, 4, %g2 + retl + stx %g2, [%o0 + PT_V9_TNPC] + .size perf_arch_fetch_caller_regs,.-perf_arch_fetch_caller_regs +#endif /* CONFIG_PERF_EVENTS */ + #ifdef CONFIG_SMP .globl hard_smp_processor_id .type hard_smp_processor_id,#function diff --git a/arch/sparc/kernel/hvapi.c b/arch/sparc/kernel/hvapi.c index 1d272c3b574..7c60afb835b 100644 --- a/arch/sparc/kernel/hvapi.c +++ b/arch/sparc/kernel/hvapi.c @@ -5,7 +5,6 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <asm/hypervisor.h> #include <asm/oplib.h> diff --git a/arch/sparc/kernel/iommu.c b/arch/sparc/kernel/iommu.c index 5fad94950e7..47977a77f6c 100644 --- a/arch/sparc/kernel/iommu.c +++ b/arch/sparc/kernel/iommu.c @@ -6,6 +6,7 @@ #include <linux/kernel.h> #include <linux/module.h> +#include <linux/slab.h> #include <linux/delay.h> #include <linux/device.h> #include <linux/dma-mapping.h> @@ -862,13 +863,3 @@ int dma_supported(struct device *dev, u64 device_mask) return 0; } EXPORT_SYMBOL(dma_supported); - -int dma_set_mask(struct device *dev, u64 dma_mask) -{ -#ifdef CONFIG_PCI - if (dev->bus == &pci_bus_type) - return pci_set_dma_mask(to_pci_dev(dev), dma_mask); -#endif - return -EINVAL; -} -EXPORT_SYMBOL(dma_set_mask); diff --git a/arch/sparc/kernel/ioport.c b/arch/sparc/kernel/ioport.c index 3c8c44f6a41..84e5386714c 100644 --- a/arch/sparc/kernel/ioport.c +++ b/arch/sparc/kernel/ioport.c @@ -676,17 +676,6 @@ int dma_supported(struct device *dev, u64 mask) } EXPORT_SYMBOL(dma_supported); -int dma_set_mask(struct device *dev, u64 dma_mask) -{ -#ifdef CONFIG_PCI - if (dev->bus == &pci_bus_type) - return pci_set_dma_mask(to_pci_dev(dev), dma_mask); -#endif - return -EOPNOTSUPP; -} -EXPORT_SYMBOL(dma_set_mask); - - #ifdef CONFIG_PROC_FS static int sparc_io_proc_show(struct seq_file *m, void *v) diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index 8d6882bb480..830d70a3e20 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -20,7 +20,9 @@ #include <linux/delay.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/ftrace.h> #include <linux/irq.h> +#include <linux/kmemleak.h> #include <asm/ptrace.h> #include <asm/processor.h> @@ -45,6 +47,7 @@ #include "entry.h" #include "cpumap.h" +#include "kstack.h" #define NUM_IVECS (IMAP_INR + 1) @@ -250,12 +253,12 @@ struct irq_handler_data { }; #ifdef CONFIG_SMP -static int irq_choose_cpu(unsigned int virt_irq) +static int irq_choose_cpu(unsigned int virt_irq, const struct cpumask *affinity) { cpumask_t mask; int cpuid; - cpumask_copy(&mask, irq_desc[virt_irq].affinity); + cpumask_copy(&mask, affinity); if (cpus_equal(mask, cpu_online_map)) { cpuid = map_to_cpu(virt_irq); } else { @@ -268,10 +271,8 @@ static int irq_choose_cpu(unsigned int virt_irq) return cpuid; } #else -static int irq_choose_cpu(unsigned int virt_irq) -{ - return real_hard_smp_processor_id(); -} +#define irq_choose_cpu(virt_irq, affinity) \ + real_hard_smp_processor_id() #endif static void sun4u_irq_enable(unsigned int virt_irq) @@ -282,7 +283,8 @@ static void sun4u_irq_enable(unsigned int virt_irq) unsigned long cpuid, imap, val; unsigned int tid; - cpuid = irq_choose_cpu(virt_irq); + cpuid = irq_choose_cpu(virt_irq, + irq_desc[virt_irq].affinity); imap = data->imap; tid = sun4u_compute_tid(imap, cpuid); @@ -299,7 +301,24 @@ static void sun4u_irq_enable(unsigned int virt_irq) static int sun4u_set_affinity(unsigned int virt_irq, const struct cpumask *mask) { - sun4u_irq_enable(virt_irq); + struct irq_handler_data *data = get_irq_chip_data(virt_irq); + + if (likely(data)) { + unsigned long cpuid, imap, val; + unsigned int tid; + + cpuid = irq_choose_cpu(virt_irq, mask); + imap = data->imap; + + tid = sun4u_compute_tid(imap, cpuid); + + val = upa_readq(imap); + val &= ~(IMAP_TID_UPA | IMAP_TID_JBUS | + IMAP_AID_SAFARI | IMAP_NID_SAFARI); + val |= tid | IMAP_VALID; + upa_writeq(val, imap); + upa_writeq(ICLR_IDLE, data->iclr); + } return 0; } @@ -340,7 +359,8 @@ static void sun4u_irq_eoi(unsigned int virt_irq) static void sun4v_irq_enable(unsigned int virt_irq) { unsigned int ino = virt_irq_table[virt_irq].dev_ino; - unsigned long cpuid = irq_choose_cpu(virt_irq); + unsigned long cpuid = irq_choose_cpu(virt_irq, + irq_desc[virt_irq].affinity); int err; err = sun4v_intr_settarget(ino, cpuid); @@ -361,7 +381,7 @@ static int sun4v_set_affinity(unsigned int virt_irq, const struct cpumask *mask) { unsigned int ino = virt_irq_table[virt_irq].dev_ino; - unsigned long cpuid = irq_choose_cpu(virt_irq); + unsigned long cpuid = irq_choose_cpu(virt_irq, mask); int err; err = sun4v_intr_settarget(ino, cpuid); @@ -403,7 +423,7 @@ static void sun4v_virq_enable(unsigned int virt_irq) unsigned long cpuid, dev_handle, dev_ino; int err; - cpuid = irq_choose_cpu(virt_irq); + cpuid = irq_choose_cpu(virt_irq, irq_desc[virt_irq].affinity); dev_handle = virt_irq_table[virt_irq].dev_handle; dev_ino = virt_irq_table[virt_irq].dev_ino; @@ -433,7 +453,7 @@ static int sun4v_virt_set_affinity(unsigned int virt_irq, unsigned long cpuid, dev_handle, dev_ino; int err; - cpuid = irq_choose_cpu(virt_irq); + cpuid = irq_choose_cpu(virt_irq, mask); dev_handle = virt_irq_table[virt_irq].dev_handle; dev_ino = virt_irq_table[virt_irq].dev_ino; @@ -630,6 +650,14 @@ unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino) bucket = kzalloc(sizeof(struct ino_bucket), GFP_ATOMIC); if (unlikely(!bucket)) return 0; + + /* The only reference we store to the IRQ bucket is + * by physical address which kmemleak can't see, tell + * it that this object explicitly is not a leak and + * should be scanned. + */ + kmemleak_not_leak(bucket); + __flush_dcache_range((unsigned long) bucket, ((unsigned long) bucket + sizeof(struct ino_bucket))); @@ -686,25 +714,7 @@ void ack_bad_irq(unsigned int virt_irq) void *hardirq_stack[NR_CPUS]; void *softirq_stack[NR_CPUS]; -static __attribute__((always_inline)) void *set_hardirq_stack(void) -{ - void *orig_sp, *sp = hardirq_stack[smp_processor_id()]; - - __asm__ __volatile__("mov %%sp, %0" : "=r" (orig_sp)); - if (orig_sp < sp || - orig_sp > (sp + THREAD_SIZE)) { - sp += THREAD_SIZE - 192 - STACK_BIAS; - __asm__ __volatile__("mov %0, %%sp" : : "r" (sp)); - } - - return orig_sp; -} -static __attribute__((always_inline)) void restore_hardirq_stack(void *orig_sp) -{ - __asm__ __volatile__("mov %0, %%sp" : : "r" (orig_sp)); -} - -void handler_irq(int irq, struct pt_regs *regs) +void __irq_entry handler_irq(int irq, struct pt_regs *regs) { unsigned long pstate, bucket_pa; struct pt_regs *old_regs; diff --git a/arch/sparc/kernel/kgdb_64.c b/arch/sparc/kernel/kgdb_64.c index f5a0fd490b5..0a2bd0f99fc 100644 --- a/arch/sparc/kernel/kgdb_64.c +++ b/arch/sparc/kernel/kgdb_64.c @@ -5,6 +5,7 @@ #include <linux/kgdb.h> #include <linux/kdebug.h> +#include <linux/ftrace.h> #include <asm/kdebug.h> #include <asm/ptrace.h> @@ -108,7 +109,7 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) } #ifdef CONFIG_SMP -void smp_kgdb_capture_client(int irq, struct pt_regs *regs) +void __irq_entry smp_kgdb_capture_client(int irq, struct pt_regs *regs) { unsigned long flags; diff --git a/arch/sparc/kernel/kprobes.c b/arch/sparc/kernel/kprobes.c index 6716584e48a..a39d1ba5a11 100644 --- a/arch/sparc/kernel/kprobes.c +++ b/arch/sparc/kernel/kprobes.c @@ -7,6 +7,7 @@ #include <linux/kprobes.h> #include <linux/module.h> #include <linux/kdebug.h> +#include <linux/slab.h> #include <asm/signal.h> #include <asm/cacheflush.h> #include <asm/uaccess.h> diff --git a/arch/sparc/kernel/kstack.h b/arch/sparc/kernel/kstack.h index 4248d969272..53dfb92e09f 100644 --- a/arch/sparc/kernel/kstack.h +++ b/arch/sparc/kernel/kstack.h @@ -11,6 +11,10 @@ static inline bool kstack_valid(struct thread_info *tp, unsigned long sp) { unsigned long base = (unsigned long) tp; + /* Stack pointer must be 16-byte aligned. */ + if (sp & (16UL - 1)) + return false; + if (sp >= (base + sizeof(struct thread_info)) && sp <= (base + THREAD_SIZE - sizeof(struct sparc_stackf))) return true; @@ -57,4 +61,23 @@ check_magic: } +static inline __attribute__((always_inline)) void *set_hardirq_stack(void) +{ + void *orig_sp, *sp = hardirq_stack[smp_processor_id()]; + + __asm__ __volatile__("mov %%sp, %0" : "=r" (orig_sp)); + if (orig_sp < sp || + orig_sp > (sp + THREAD_SIZE)) { + sp += THREAD_SIZE - 192 - STACK_BIAS; + __asm__ __volatile__("mov %0, %%sp" : : "r" (sp)); + } + + return orig_sp; +} + +static inline __attribute__((always_inline)) void restore_hardirq_stack(void *orig_sp) +{ + __asm__ __volatile__("mov %0, %%sp" : : "r" (orig_sp)); +} + #endif /* _KSTACK_H */ diff --git a/arch/sparc/kernel/led.c b/arch/sparc/kernel/led.c index 00d034ea216..3ae36f36e75 100644 --- a/arch/sparc/kernel/led.c +++ b/arch/sparc/kernel/led.c @@ -3,6 +3,7 @@ #include <linux/init.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/slab.h> #include <linux/string.h> #include <linux/jiffies.h> #include <linux/timer.h> diff --git a/arch/sparc/kernel/leon_kernel.c b/arch/sparc/kernel/leon_kernel.c index 87f1760c0aa..6a7b4dbc8e0 100644 --- a/arch/sparc/kernel/leon_kernel.c +++ b/arch/sparc/kernel/leon_kernel.c @@ -7,7 +7,6 @@ #include <linux/module.h> #include <linux/errno.h> #include <linux/mutex.h> -#include <linux/slab.h> #include <linux/of.h> #include <linux/of_platform.h> #include <linux/interrupt.h> @@ -124,7 +123,7 @@ void __init leon_init_timers(irq_handler_t counter_fn) if (!(LEON3_BYPASS_LOAD_PA(&leon3_gptimer_regs->config) & (1<<LEON3_GPTIMER_SEPIRQ))) { - prom_printf("irq timer not configured with seperate irqs \n"); + prom_printf("irq timer not configured with separate irqs\n"); BUG(); } diff --git a/arch/sparc/kernel/leon_smp.c b/arch/sparc/kernel/leon_smp.c index 05c0dadd637..e1656fc41cc 100644 --- a/arch/sparc/kernel/leon_smp.c +++ b/arch/sparc/kernel/leon_smp.c @@ -22,6 +22,7 @@ #include <linux/profile.h> #include <linux/pm.h> #include <linux/delay.h> +#include <linux/gfp.h> #include <asm/cacheflush.h> #include <asm/tlbflush.h> @@ -177,7 +178,7 @@ void __init leon_boot_cpus(void) int nrcpu = leon_smp_nrcpus(); int me = smp_processor_id(); - printk(KERN_INFO "%d:(%d:%d) cpus mpirq at 0x%x \n", (unsigned int)me, + printk(KERN_INFO "%d:(%d:%d) cpus mpirq at 0x%x\n", (unsigned int)me, (unsigned int)nrcpu, (unsigned int)NR_CPUS, (unsigned int)&(leon3_irqctrl_regs->mpstatus)); @@ -226,7 +227,7 @@ int __cpuinit leon_boot_one_cpu(int i) break; udelay(200); } - printk(KERN_INFO "Started CPU %d \n", (unsigned int)i); + printk(KERN_INFO "Started CPU %d\n", (unsigned int)i); if (!(cpu_callin_map[i])) { printk(KERN_ERR "Processor %d is stuck.\n", i); diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c index 0ee642f6323..f848aadf54d 100644 --- a/arch/sparc/kernel/module.c +++ b/arch/sparc/kernel/module.c @@ -9,9 +9,9 @@ #include <linux/elf.h> #include <linux/vmalloc.h> #include <linux/fs.h> +#include <linux/gfp.h> #include <linux/string.h> #include <linux/ctype.h> -#include <linux/slab.h> #include <linux/mm.h> #include <asm/processor.h> diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index f30f4a1ead2..a4bd7ba74c8 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -21,9 +21,10 @@ #include <asm/perf_event.h> #include <asm/ptrace.h> -#include <asm/local.h> #include <asm/pcr.h> +#include "kstack.h" + /* We don't have a real NMI on sparc64, but we can fake one * up using profiling counter overflow interrupts and interrupt * levels. @@ -93,38 +94,43 @@ static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) { unsigned int sum, touched = 0; - int cpu = smp_processor_id(); + void *orig_sp; clear_softint(1 << irq); - pcr_ops->write(PCR_PIC_PRIV); local_cpu_data().__nmi_count++; nmi_enter(); + orig_sp = set_hardirq_stack(); + if (notify_die(DIE_NMI, "nmi", regs, 0, pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) touched = 1; + else + pcr_ops->write(PCR_PIC_PRIV); - sum = kstat_irqs_cpu(0, cpu); + sum = local_cpu_data().irq0_irqs; if (__get_cpu_var(nmi_touch)) { __get_cpu_var(nmi_touch) = 0; touched = 1; } if (!touched && __get_cpu_var(last_irq_sum) == sum) { - __this_cpu_inc(per_cpu_var(alert_counter)); - if (__this_cpu_read(per_cpu_var(alert_counter)) == 30 * nmi_hz) + __this_cpu_inc(alert_counter); + if (__this_cpu_read(alert_counter) == 30 * nmi_hz) die_nmi("BUG: NMI Watchdog detected LOCKUP", regs, panic_on_timeout); } else { __get_cpu_var(last_irq_sum) = sum; - __this_cpu_write(per_cpu_var(alert_counter), 0); + __this_cpu_write(alert_counter, 0); } if (__get_cpu_var(wd_enabled)) { write_pic(picl_value(nmi_hz)); pcr_ops->write(pcr_enable); } + restore_hardirq_stack(orig_sp); + nmi_exit(); } diff --git a/arch/sparc/kernel/of_device_32.c b/arch/sparc/kernel/of_device_32.c index 4c26eb59e74..da527b33ebc 100644 --- a/arch/sparc/kernel/of_device_32.c +++ b/arch/sparc/kernel/of_device_32.c @@ -105,7 +105,7 @@ static unsigned long of_bus_sbus_get_flags(const u32 *addr, unsigned long flags) static int of_bus_ambapp_match(struct device_node *np) { - return !strcmp(np->name, "ambapp"); + return !strcmp(np->type, "ambapp"); } static void of_bus_ambapp_count_cells(struct device_node *child, @@ -433,7 +433,7 @@ build_resources: if (!parent) dev_set_name(&op->dev, "root"); else - dev_set_name(&op->dev, "%08x", dp->node); + dev_set_name(&op->dev, "%08x", dp->phandle); if (of_device_register(op)) { printk("%s: Could not register of device.\n", diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c index 0a6f2d1798d..b3d4cb5d21b 100644 --- a/arch/sparc/kernel/of_device_64.c +++ b/arch/sparc/kernel/of_device_64.c @@ -676,7 +676,7 @@ static struct of_device * __init scan_one_device(struct device_node *dp, if (!parent) dev_set_name(&op->dev, "root"); else - dev_set_name(&op->dev, "%08x", dp->node); + dev_set_name(&op->dev, "%08x", dp->phandle); if (of_device_register(op)) { printk("%s: Could not register of device.\n", diff --git a/arch/sparc/kernel/of_device_common.c b/arch/sparc/kernel/of_device_common.c index cb8eb799bb6..0247e68210b 100644 --- a/arch/sparc/kernel/of_device_common.c +++ b/arch/sparc/kernel/of_device_common.c @@ -4,7 +4,6 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/mod_devicetable.h> -#include <linux/slab.h> #include <linux/errno.h> #include <linux/irq.h> #include <linux/of_device.h> diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c index 539e83f8e08..5ac539a5930 100644 --- a/arch/sparc/kernel/pci.c +++ b/arch/sparc/kernel/pci.c @@ -247,6 +247,7 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, struct pci_bus *bus, int devfn) { struct dev_archdata *sd; + struct pci_slot *slot; struct of_device *op; struct pci_dev *dev; const char *type; @@ -286,6 +287,11 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, dev->dev.bus = &pci_bus_type; dev->devfn = devfn; dev->multifunction = 0; /* maybe a lie? */ + set_pcie_port_type(dev); + + list_for_each_entry(slot, &dev->bus->slots, list) + if (PCI_SLOT(dev->devfn) == slot->number) + dev->slot = slot; dev->vendor = of_getintprop_default(node, "vendor-id", 0xffff); dev->device = of_getintprop_default(node, "device-id", 0xffff); @@ -322,6 +328,7 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, dev->current_state = 4; /* unknown power state */ dev->error_state = pci_channel_io_normal; + dev->dma_mask = 0xffffffff; if (!strcmp(node->name, "pci")) { /* a PCI-PCI bridge */ @@ -715,9 +722,10 @@ void pcibios_update_irq(struct pci_dev *pdev, int irq) { } -void pcibios_align_resource(void *data, struct resource *res, - resource_size_t size, resource_size_t align) +resource_size_t pcibios_align_resource(void *data, const struct resource *res, + resource_size_t size, resource_size_t align) { + return res->start; } int pcibios_enable_device(struct pci_dev *dev, int mask) @@ -1087,3 +1095,78 @@ static int __init pcibios_init(void) return 0; } subsys_initcall(pcibios_init); + +#ifdef CONFIG_SYSFS +static void __devinit pci_bus_slot_names(struct device_node *node, + struct pci_bus *bus) +{ + const struct pci_slot_names { + u32 slot_mask; + char names[0]; + } *prop; + const char *sp; + int len, i; + u32 mask; + + prop = of_get_property(node, "slot-names", &len); + if (!prop) + return; + + mask = prop->slot_mask; + sp = prop->names; + + if (ofpci_verbose) + printk("PCI: Making slots for [%s] mask[0x%02x]\n", + node->full_name, mask); + + i = 0; + while (mask) { + struct pci_slot *pci_slot; + u32 this_bit = 1 << i; + + if (!(mask & this_bit)) { + i++; + continue; + } + + if (ofpci_verbose) + printk("PCI: Making slot [%s]\n", sp); + + pci_slot = pci_create_slot(bus, i, sp, NULL); + if (IS_ERR(pci_slot)) + printk(KERN_ERR "PCI: pci_create_slot returned %ld\n", + PTR_ERR(pci_slot)); + + sp += strlen(sp) + 1; + mask &= ~this_bit; + i++; + } +} + +static int __init of_pci_slot_init(void) +{ + struct pci_bus *pbus = NULL; + + while ((pbus = pci_find_next_bus(pbus)) != NULL) { + struct device_node *node; + + if (pbus->self) { + struct dev_archdata *sd = pbus->self->sysdata; + + /* PCI->PCI bridge */ + node = sd->prom_node; + } else { + struct pci_pbm_info *pbm = pbus->sysdata; + + /* Host PCI controller */ + node = pbm->op->node; + } + + pci_bus_slot_names(node, pbus); + } + + return 0; +} + +module_init(of_pci_slot_init); +#endif diff --git a/arch/sparc/kernel/pci_common.c b/arch/sparc/kernel/pci_common.c index b775658a927..8a000583b5c 100644 --- a/arch/sparc/kernel/pci_common.c +++ b/arch/sparc/kernel/pci_common.c @@ -371,14 +371,19 @@ static void pci_register_iommu_region(struct pci_pbm_info *pbm) struct resource *rp = kzalloc(sizeof(*rp), GFP_KERNEL); if (!rp) { - prom_printf("Cannot allocate IOMMU resource.\n"); - prom_halt(); + pr_info("%s: Cannot allocate IOMMU resource.\n", + pbm->name); + return; } rp->name = "IOMMU"; rp->start = pbm->mem_space.start + (unsigned long) vdma[0]; rp->end = rp->start + (unsigned long) vdma[1] - 1UL; rp->flags = IORESOURCE_BUSY; - request_resource(&pbm->mem_space, rp); + if (request_resource(&pbm->mem_space, rp)) { + pr_info("%s: Unable to request IOMMU resource.\n", + pbm->name); + kfree(rp); + } } } diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c index e1b0541feb1..e0ef847219c 100644 --- a/arch/sparc/kernel/pci_msi.c +++ b/arch/sparc/kernel/pci_msi.c @@ -4,6 +4,7 @@ */ #include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/slab.h> #include <linux/irq.h> #include "pci_impl.h" diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c index 85e7037429b..d36a8d391ca 100644 --- a/arch/sparc/kernel/pcic.c +++ b/arch/sparc/kernel/pcic.c @@ -30,6 +30,7 @@ #include <asm/oplib.h> #include <asm/prom.h> #include <asm/pcic.h> +#include <asm/timex.h> #include <asm/timer.h> #include <asm/uaccess.h> #include <asm/irq_regs.h> @@ -163,8 +164,6 @@ void __iomem *pcic_regs; volatile int pcic_speculative; volatile int pcic_trapped; -static void pci_do_gettimeofday(struct timeval *tv); -static int pci_do_settimeofday(struct timespec *tv); #define CONFIG_CMD(bus, device_fn, where) (0x80000000 | (((unsigned int)bus) << 16) | (((unsigned int)device_fn) << 8) | (where & ~3)) @@ -586,8 +585,6 @@ pcic_fill_irq(struct linux_pcic *pcic, struct pci_dev *dev, int node) writew(ivec, pcic->pcic_regs+PCI_INT_SELECT_LO); } } - - return; } /* @@ -716,19 +713,27 @@ static irqreturn_t pcic_timer_handler (int irq, void *h) #define USECS_PER_JIFFY 10000 /* We have 100HZ "standard" timer for sparc */ #define TICK_TIMER_LIMIT ((100*1000000/4)/100) +u32 pci_gettimeoffset(void) +{ + /* + * We divide all by 100 + * to have microsecond resolution and to avoid overflow + */ + unsigned long count = + readl(pcic0.pcic_regs+PCI_SYS_COUNTER) & ~PCI_SYS_COUNTER_OVERFLOW; + count = ((count/100)*USECS_PER_JIFFY) / (TICK_TIMER_LIMIT/100); + return count * 1000; +} + + void __init pci_time_init(void) { struct linux_pcic *pcic = &pcic0; unsigned long v; int timer_irq, irq; - /* A hack until do_gettimeofday prototype is moved to arch specific headers - and btfixupped. Patch do_gettimeofday with ba pci_do_gettimeofday; nop */ - ((unsigned int *)do_gettimeofday)[0] = - 0x10800000 | ((((unsigned long)pci_do_gettimeofday - - (unsigned long)do_gettimeofday) >> 2) & 0x003fffff); - ((unsigned int *)do_gettimeofday)[1] = 0x01000000; - BTFIXUPSET_CALL(bus_do_settimeofday, pci_do_settimeofday, BTFIXUPCALL_NORM); + do_arch_gettimeoffset = pci_gettimeoffset; + btfixup(); writel (TICK_TIMER_LIMIT, pcic->pcic_regs+PCI_SYS_LIMIT); @@ -746,84 +751,6 @@ void __init pci_time_init(void) local_irq_enable(); } -static inline unsigned long do_gettimeoffset(void) -{ - /* - * We divide all by 100 - * to have microsecond resolution and to avoid overflow - */ - unsigned long count = - readl(pcic0.pcic_regs+PCI_SYS_COUNTER) & ~PCI_SYS_COUNTER_OVERFLOW; - count = ((count/100)*USECS_PER_JIFFY) / (TICK_TIMER_LIMIT/100); - return count; -} - -static void pci_do_gettimeofday(struct timeval *tv) -{ - unsigned long flags; - unsigned long seq; - unsigned long usec, sec; - unsigned long max_ntp_tick = tick_usec - tickadj; - - do { - seq = read_seqbegin_irqsave(&xtime_lock, flags); - usec = do_gettimeoffset(); - - /* - * If time_adjust is negative then NTP is slowing the clock - * so make sure not to go into next possible interval. - * Better to lose some accuracy than have time go backwards.. - */ - if (unlikely(time_adjust < 0)) - usec = min(usec, max_ntp_tick); - - sec = xtime.tv_sec; - usec += (xtime.tv_nsec / 1000); - } while (read_seqretry_irqrestore(&xtime_lock, seq, flags)); - - while (usec >= 1000000) { - usec -= 1000000; - sec++; - } - - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -static int pci_do_settimeofday(struct timespec *tv) -{ - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - /* - * This is revolting. We need to set "xtime" correctly. However, the - * value in this location is the value at the most recent update of - * wall time. Discover what correction gettimeofday() would have - * made, and then undo it! - */ - tv->tv_nsec -= 1000 * do_gettimeoffset(); - while (tv->tv_nsec < 0) { - tv->tv_nsec += NSEC_PER_SEC; - tv->tv_sec--; - } - - wall_to_monotonic.tv_sec += xtime.tv_sec - tv->tv_sec; - wall_to_monotonic.tv_nsec += xtime.tv_nsec - tv->tv_nsec; - - if (wall_to_monotonic.tv_nsec > NSEC_PER_SEC) { - wall_to_monotonic.tv_nsec -= NSEC_PER_SEC; - wall_to_monotonic.tv_sec++; - } - if (wall_to_monotonic.tv_nsec < 0) { - wall_to_monotonic.tv_nsec += NSEC_PER_SEC; - wall_to_monotonic.tv_sec--; - } - - xtime.tv_sec = tv->tv_sec; - xtime.tv_nsec = tv->tv_nsec; - ntp_clear(); - return 0; -} #if 0 static void watchdog_reset() { @@ -839,9 +766,10 @@ char * __devinit pcibios_setup(char *str) return str; } -void pcibios_align_resource(void *data, struct resource *res, - resource_size_t size, resource_size_t align) +resource_size_t pcibios_align_resource(void *data, const struct resource *res, + resource_size_t size, resource_size_t align) { + return res->start; } int pcibios_enable_device(struct pci_dev *pdev, int mask) diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index 2d94e7a03af..c4a6a50b484 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -8,6 +8,7 @@ #include <linux/irq.h> #include <linux/perf_event.h> +#include <linux/ftrace.h> #include <asm/pil.h> #include <asm/pcr.h> @@ -34,7 +35,7 @@ unsigned int picl_shift; * Therefore in such situations we defer the work by signalling * a lower level cpu IRQ. */ -void deferred_pcr_work_irq(int irq, struct pt_regs *regs) +void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs) { struct pt_regs *old_regs; diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index fa5936e1c3b..e2771939341 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1,6 +1,6 @@ /* Performance event support for sparc64. * - * Copyright (C) 2009 David S. Miller <davem@davemloft.net> + * Copyright (C) 2009, 2010 David S. Miller <davem@davemloft.net> * * This code is based almost entirely upon the x86 perf event * code, which is: @@ -18,11 +18,15 @@ #include <linux/kdebug.h> #include <linux/mutex.h> +#include <asm/stacktrace.h> #include <asm/cpudata.h> +#include <asm/uaccess.h> #include <asm/atomic.h> #include <asm/nmi.h> #include <asm/pcr.h> +#include "kstack.h" + /* Sparc64 chips have two performance counters, 32-bits each, with * overflow interrupts generated on transition from 0xffffffff to 0. * The counters are accessed in one go using a 64-bit register. @@ -51,16 +55,49 @@ #define PIC_UPPER_INDEX 0 #define PIC_LOWER_INDEX 1 +#define PIC_NO_INDEX -1 struct cpu_hw_events { - struct perf_event *events[MAX_HWEVENTS]; - unsigned long used_mask[BITS_TO_LONGS(MAX_HWEVENTS)]; - unsigned long active_mask[BITS_TO_LONGS(MAX_HWEVENTS)]; + /* Number of events currently scheduled onto this cpu. + * This tells how many entries in the arrays below + * are valid. + */ + int n_events; + + /* Number of new events added since the last hw_perf_disable(). + * This works because the perf event layer always adds new + * events inside of a perf_{disable,enable}() sequence. + */ + int n_added; + + /* Array of events current scheduled on this cpu. */ + struct perf_event *event[MAX_HWEVENTS]; + + /* Array of encoded longs, specifying the %pcr register + * encoding and the mask of PIC counters this even can + * be scheduled on. See perf_event_encode() et al. + */ + unsigned long events[MAX_HWEVENTS]; + + /* The current counter index assigned to an event. When the + * event hasn't been programmed into the cpu yet, this will + * hold PIC_NO_INDEX. The event->hw.idx value tells us where + * we ought to schedule the event. + */ + int current_idx[MAX_HWEVENTS]; + + /* Software copy of %pcr register on this cpu. */ u64 pcr; + + /* Enabled/disable state. */ int enabled; }; DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; +/* An event map describes the characteristics of a performance + * counter event. In particular it gives the encoding as well as + * a mask telling which counters the event can be measured on. + */ struct perf_event_map { u16 encoding; u8 pic_mask; @@ -69,15 +106,20 @@ struct perf_event_map { #define PIC_LOWER 0x02 }; +/* Encode a perf_event_map entry into a long. */ static unsigned long perf_event_encode(const struct perf_event_map *pmap) { return ((unsigned long) pmap->encoding << 16) | pmap->pic_mask; } -static void perf_event_decode(unsigned long val, u16 *enc, u8 *msk) +static u8 perf_event_get_msk(unsigned long val) { - *msk = val & 0xff; - *enc = val >> 16; + return val & 0xff; +} + +static u64 perf_event_get_enc(unsigned long val) +{ + return val >> 16; } #define C(x) PERF_COUNT_HW_CACHE_##x @@ -491,53 +533,6 @@ static inline void sparc_pmu_disable_event(struct cpu_hw_events *cpuc, struct hw pcr_ops->write(cpuc->pcr); } -void hw_perf_enable(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; - int i; - - if (cpuc->enabled) - return; - - cpuc->enabled = 1; - barrier(); - - val = cpuc->pcr; - - for (i = 0; i < MAX_HWEVENTS; i++) { - struct perf_event *cp = cpuc->events[i]; - struct hw_perf_event *hwc; - - if (!cp) - continue; - hwc = &cp->hw; - val |= hwc->config_base; - } - - cpuc->pcr = val; - - pcr_ops->write(cpuc->pcr); -} - -void hw_perf_disable(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; - - if (!cpuc->enabled) - return; - - cpuc->enabled = 0; - - val = cpuc->pcr; - val &= ~(PCR_UTRACE | PCR_STRACE | - sparc_pmu->hv_bit | sparc_pmu->irq_bit); - cpuc->pcr = val; - - pcr_ops->write(cpuc->pcr); -} - static u32 read_pmc(int idx) { u64 val; @@ -566,6 +561,30 @@ static void write_pmc(int idx, u64 val) write_pic(pic); } +static u64 sparc_perf_event_update(struct perf_event *event, + struct hw_perf_event *hwc, int idx) +{ + int shift = 64 - 32; + u64 prev_raw_count, new_raw_count; + s64 delta; + +again: + prev_raw_count = atomic64_read(&hwc->prev_count); + new_raw_count = read_pmc(idx); + + if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, + new_raw_count) != prev_raw_count) + goto again; + + delta = (new_raw_count << shift) - (prev_raw_count << shift); + delta >>= shift; + + atomic64_add(delta, &event->count); + atomic64_sub(delta, &hwc->period_left); + + return new_raw_count; +} + static int sparc_perf_event_set_period(struct perf_event *event, struct hw_perf_event *hwc, int idx) { @@ -598,81 +617,166 @@ static int sparc_perf_event_set_period(struct perf_event *event, return ret; } -static int sparc_pmu_enable(struct perf_event *event) +/* If performance event entries have been added, move existing + * events around (if necessary) and then assign new entries to + * counters. + */ +static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; + int i; - if (test_and_set_bit(idx, cpuc->used_mask)) - return -EAGAIN; + if (!cpuc->n_added) + goto out; - sparc_pmu_disable_event(cpuc, hwc, idx); + /* Read in the counters which are moving. */ + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *cp = cpuc->event[i]; - cpuc->events[idx] = event; - set_bit(idx, cpuc->active_mask); + if (cpuc->current_idx[i] != PIC_NO_INDEX && + cpuc->current_idx[i] != cp->hw.idx) { + sparc_perf_event_update(cp, &cp->hw, + cpuc->current_idx[i]); + cpuc->current_idx[i] = PIC_NO_INDEX; + } + } - sparc_perf_event_set_period(event, hwc, idx); - sparc_pmu_enable_event(cpuc, hwc, idx); - perf_event_update_userpage(event); - return 0; + /* Assign to counters all unassigned events. */ + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *cp = cpuc->event[i]; + struct hw_perf_event *hwc = &cp->hw; + int idx = hwc->idx; + u64 enc; + + if (cpuc->current_idx[i] != PIC_NO_INDEX) + continue; + + sparc_perf_event_set_period(cp, hwc, idx); + cpuc->current_idx[i] = idx; + + enc = perf_event_get_enc(cpuc->events[i]); + pcr |= event_encoding(enc, idx); + } +out: + return pcr; } -static u64 sparc_perf_event_update(struct perf_event *event, - struct hw_perf_event *hwc, int idx) +void hw_perf_enable(void) { - int shift = 64 - 32; - u64 prev_raw_count, new_raw_count; - s64 delta; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + u64 pcr; -again: - prev_raw_count = atomic64_read(&hwc->prev_count); - new_raw_count = read_pmc(idx); + if (cpuc->enabled) + return; - if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, - new_raw_count) != prev_raw_count) - goto again; + cpuc->enabled = 1; + barrier(); - delta = (new_raw_count << shift) - (prev_raw_count << shift); - delta >>= shift; + pcr = cpuc->pcr; + if (!cpuc->n_events) { + pcr = 0; + } else { + pcr = maybe_change_configuration(cpuc, pcr); - atomic64_add(delta, &event->count); - atomic64_sub(delta, &hwc->period_left); + /* We require that all of the events have the same + * configuration, so just fetch the settings from the + * first entry. + */ + cpuc->pcr = pcr | cpuc->event[0]->hw.config_base; + } - return new_raw_count; + pcr_ops->write(cpuc->pcr); +} + +void hw_perf_disable(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + u64 val; + + if (!cpuc->enabled) + return; + + cpuc->enabled = 0; + cpuc->n_added = 0; + + val = cpuc->pcr; + val &= ~(PCR_UTRACE | PCR_STRACE | + sparc_pmu->hv_bit | sparc_pmu->irq_bit); + cpuc->pcr = val; + + pcr_ops->write(cpuc->pcr); } static void sparc_pmu_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; + unsigned long flags; + int i; - clear_bit(idx, cpuc->active_mask); - sparc_pmu_disable_event(cpuc, hwc, idx); + local_irq_save(flags); + perf_disable(); + + for (i = 0; i < cpuc->n_events; i++) { + if (event == cpuc->event[i]) { + int idx = cpuc->current_idx[i]; + + /* Shift remaining entries down into + * the existing slot. + */ + while (++i < cpuc->n_events) { + cpuc->event[i - 1] = cpuc->event[i]; + cpuc->events[i - 1] = cpuc->events[i]; + cpuc->current_idx[i - 1] = + cpuc->current_idx[i]; + } + + /* Absorb the final count and turn off the + * event. + */ + sparc_pmu_disable_event(cpuc, hwc, idx); + barrier(); + sparc_perf_event_update(event, hwc, idx); - barrier(); + perf_event_update_userpage(event); - sparc_perf_event_update(event, hwc, idx); - cpuc->events[idx] = NULL; - clear_bit(idx, cpuc->used_mask); + cpuc->n_events--; + break; + } + } - perf_event_update_userpage(event); + perf_enable(); + local_irq_restore(flags); +} + +static int active_event_index(struct cpu_hw_events *cpuc, + struct perf_event *event) +{ + int i; + + for (i = 0; i < cpuc->n_events; i++) { + if (cpuc->event[i] == event) + break; + } + BUG_ON(i == cpuc->n_events); + return cpuc->current_idx[i]; } static void sparc_pmu_read(struct perf_event *event) { + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx = active_event_index(cpuc, event); struct hw_perf_event *hwc = &event->hw; - sparc_perf_event_update(event, hwc, hwc->idx); + sparc_perf_event_update(event, hwc, idx); } static void sparc_pmu_unthrottle(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx = active_event_index(cpuc, event); struct hw_perf_event *hwc = &event->hw; - sparc_pmu_enable_event(cpuc, hwc, hwc->idx); + sparc_pmu_enable_event(cpuc, hwc, idx); } static atomic_t active_events = ATOMIC_INIT(0); @@ -750,43 +854,75 @@ static void hw_perf_event_destroy(struct perf_event *event) /* Make sure all events can be scheduled into the hardware at * the same time. This is simplified by the fact that we only * need to support 2 simultaneous HW events. + * + * As a side effect, the evts[]->hw.idx values will be assigned + * on success. These are pending indexes. When the events are + * actually programmed into the chip, these values will propagate + * to the per-cpu cpuc->current_idx[] slots, see the code in + * maybe_change_configuration() for details. */ -static int sparc_check_constraints(unsigned long *events, int n_ev) +static int sparc_check_constraints(struct perf_event **evts, + unsigned long *events, int n_ev) { - if (n_ev <= perf_max_events) { - u8 msk1, msk2; - u16 dummy; - - if (n_ev == 1) - return 0; - BUG_ON(n_ev != 2); - perf_event_decode(events[0], &dummy, &msk1); - perf_event_decode(events[1], &dummy, &msk2); - - /* If both events can go on any counter, OK. */ - if (msk1 == (PIC_UPPER | PIC_LOWER) && - msk2 == (PIC_UPPER | PIC_LOWER)) - return 0; - - /* If one event is limited to a specific counter, - * and the other can go on both, OK. - */ - if ((msk1 == PIC_UPPER || msk1 == PIC_LOWER) && - msk2 == (PIC_UPPER | PIC_LOWER)) - return 0; - if ((msk2 == PIC_UPPER || msk2 == PIC_LOWER) && - msk1 == (PIC_UPPER | PIC_LOWER)) - return 0; - - /* If the events are fixed to different counters, OK. */ - if ((msk1 == PIC_UPPER && msk2 == PIC_LOWER) || - (msk1 == PIC_LOWER && msk2 == PIC_UPPER)) - return 0; - - /* Otherwise, there is a conflict. */ + u8 msk0 = 0, msk1 = 0; + int idx0 = 0; + + /* This case is possible when we are invoked from + * hw_perf_group_sched_in(). + */ + if (!n_ev) + return 0; + + if (n_ev > perf_max_events) + return -1; + + msk0 = perf_event_get_msk(events[0]); + if (n_ev == 1) { + if (msk0 & PIC_LOWER) + idx0 = 1; + goto success; + } + BUG_ON(n_ev != 2); + msk1 = perf_event_get_msk(events[1]); + + /* If both events can go on any counter, OK. */ + if (msk0 == (PIC_UPPER | PIC_LOWER) && + msk1 == (PIC_UPPER | PIC_LOWER)) + goto success; + + /* If one event is limited to a specific counter, + * and the other can go on both, OK. + */ + if ((msk0 == PIC_UPPER || msk0 == PIC_LOWER) && + msk1 == (PIC_UPPER | PIC_LOWER)) { + if (msk0 & PIC_LOWER) + idx0 = 1; + goto success; } + if ((msk1 == PIC_UPPER || msk1 == PIC_LOWER) && + msk0 == (PIC_UPPER | PIC_LOWER)) { + if (msk1 & PIC_UPPER) + idx0 = 1; + goto success; + } + + /* If the events are fixed to different counters, OK. */ + if ((msk0 == PIC_UPPER && msk1 == PIC_LOWER) || + (msk0 == PIC_LOWER && msk1 == PIC_UPPER)) { + if (msk0 & PIC_LOWER) + idx0 = 1; + goto success; + } + + /* Otherwise, there is a conflict. */ return -1; + +success: + evts[0]->hw.idx = idx0; + if (n_ev == 2) + evts[1]->hw.idx = idx0 ^ 1; + return 0; } static int check_excludes(struct perf_event **evts, int n_prev, int n_new) @@ -818,7 +954,8 @@ static int check_excludes(struct perf_event **evts, int n_prev, int n_new) } static int collect_events(struct perf_event *group, int max_count, - struct perf_event *evts[], unsigned long *events) + struct perf_event *evts[], unsigned long *events, + int *current_idx) { struct perf_event *event; int n = 0; @@ -827,7 +964,8 @@ static int collect_events(struct perf_event *group, int max_count, if (n >= max_count) return -1; evts[n] = group; - events[n++] = group->hw.event_base; + events[n] = group->hw.event_base; + current_idx[n++] = PIC_NO_INDEX; } list_for_each_entry(event, &group->sibling_list, group_entry) { if (!is_software_event(event) && @@ -835,20 +973,100 @@ static int collect_events(struct perf_event *group, int max_count, if (n >= max_count) return -1; evts[n] = event; - events[n++] = event->hw.event_base; + events[n] = event->hw.event_base; + current_idx[n++] = PIC_NO_INDEX; } } return n; } +static void event_sched_in(struct perf_event *event) +{ + event->state = PERF_EVENT_STATE_ACTIVE; + event->oncpu = smp_processor_id(); + event->tstamp_running += event->ctx->time - event->tstamp_stopped; + if (is_software_event(event)) + event->pmu->enable(event); +} + +int hw_perf_group_sched_in(struct perf_event *group_leader, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct perf_event *sub; + int n0, n; + + if (!sparc_pmu) + return 0; + + n0 = cpuc->n_events; + n = collect_events(group_leader, perf_max_events - n0, + &cpuc->event[n0], &cpuc->events[n0], + &cpuc->current_idx[n0]); + if (n < 0) + return -EAGAIN; + if (check_excludes(cpuc->event, n0, n)) + return -EINVAL; + if (sparc_check_constraints(cpuc->event, cpuc->events, n + n0)) + return -EAGAIN; + cpuc->n_events = n0 + n; + cpuc->n_added += n; + + cpuctx->active_oncpu += n; + n = 1; + event_sched_in(group_leader); + list_for_each_entry(sub, &group_leader->sibling_list, group_entry) { + if (sub->state != PERF_EVENT_STATE_OFF) { + event_sched_in(sub); + n++; + } + } + ctx->nr_active += n; + + return 1; +} + +static int sparc_pmu_enable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int n0, ret = -EAGAIN; + unsigned long flags; + + local_irq_save(flags); + perf_disable(); + + n0 = cpuc->n_events; + if (n0 >= perf_max_events) + goto out; + + cpuc->event[n0] = event; + cpuc->events[n0] = event->hw.event_base; + cpuc->current_idx[n0] = PIC_NO_INDEX; + + if (check_excludes(cpuc->event, n0, 1)) + goto out; + if (sparc_check_constraints(cpuc->event, cpuc->events, n0 + 1)) + goto out; + + cpuc->n_events++; + cpuc->n_added++; + + ret = 0; +out: + perf_enable(); + local_irq_restore(flags); + return ret; +} + static int __hw_perf_event_init(struct perf_event *event) { struct perf_event_attr *attr = &event->attr; struct perf_event *evts[MAX_HWEVENTS]; struct hw_perf_event *hwc = &event->hw; unsigned long events[MAX_HWEVENTS]; + int current_idx_dmy[MAX_HWEVENTS]; const struct perf_event_map *pmap; - u64 enc; int n; if (atomic_read(&nmi_active) < 0) @@ -865,10 +1083,7 @@ static int __hw_perf_event_init(struct perf_event *event) } else return -EOPNOTSUPP; - /* We save the enable bits in the config_base. So to - * turn off sampling just write 'config', and to enable - * things write 'config | config_base'. - */ + /* We save the enable bits in the config_base. */ hwc->config_base = sparc_pmu->irq_bit; if (!attr->exclude_user) hwc->config_base |= PCR_UTRACE; @@ -879,13 +1094,11 @@ static int __hw_perf_event_init(struct perf_event *event) hwc->event_base = perf_event_encode(pmap); - enc = pmap->encoding; - n = 0; if (event->group_leader != event) { n = collect_events(event->group_leader, perf_max_events - 1, - evts, events); + evts, events, current_idx_dmy); if (n < 0) return -EINVAL; } @@ -895,9 +1108,11 @@ static int __hw_perf_event_init(struct perf_event *event) if (check_excludes(evts, n, 1)) return -EINVAL; - if (sparc_check_constraints(events, n + 1)) + if (sparc_check_constraints(evts, events, n + 1)) return -EINVAL; + hwc->idx = PIC_NO_INDEX; + /* Try to do all error checking before this point, as unwinding * state after grabbing the PMC is difficult. */ @@ -910,15 +1125,6 @@ static int __hw_perf_event_init(struct perf_event *event) atomic64_set(&hwc->period_left, hwc->sample_period); } - if (pmap->pic_mask & PIC_UPPER) { - hwc->idx = PIC_UPPER_INDEX; - enc <<= sparc_pmu->upper_shift; - } else { - hwc->idx = PIC_LOWER_INDEX; - enc <<= sparc_pmu->lower_shift; - } - - hwc->config |= enc; return 0; } @@ -968,7 +1174,7 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, struct perf_sample_data data; struct cpu_hw_events *cpuc; struct pt_regs *regs; - int idx; + int i; if (!atomic_read(&active_events)) return NOTIFY_DONE; @@ -983,16 +1189,26 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, regs = args->regs; - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); - for (idx = 0; idx < MAX_HWEVENTS; idx++) { - struct perf_event *event = cpuc->events[idx]; + + /* If the PMU has the TOE IRQ enable bits, we need to do a + * dummy write to the %pcr to clear the overflow bits and thus + * the interrupt. + * + * Do this before we peek at the counters to determine + * overflow so we don't lose any events. + */ + if (sparc_pmu->irq_bit) + pcr_ops->write(cpuc->pcr); + + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *event = cpuc->event[i]; + int idx = cpuc->current_idx[i]; struct hw_perf_event *hwc; u64 val; - if (!test_bit(idx, cpuc->active_mask)) - continue; hwc = &event->hw; val = sparc_perf_event_update(event, hwc, idx); if (val & (1ULL << 31)) @@ -1044,10 +1260,122 @@ void __init init_hw_perf_events(void) pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type); - /* All sparc64 PMUs currently have 2 events. But this simple - * driver only supports one active event at a time. - */ - perf_max_events = 1; + /* All sparc64 PMUs currently have 2 events. */ + perf_max_events = 2; register_die_notifier(&perf_event_nmi_notifier); } + +static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip) +{ + if (entry->nr < PERF_MAX_STACK_DEPTH) + entry->ip[entry->nr++] = ip; +} + +static void perf_callchain_kernel(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + unsigned long ksp, fp; + + callchain_store(entry, PERF_CONTEXT_KERNEL); + callchain_store(entry, regs->tpc); + + ksp = regs->u_regs[UREG_I6]; + fp = ksp + STACK_BIAS; + do { + struct sparc_stackf *sf; + struct pt_regs *regs; + unsigned long pc; + + if (!kstack_valid(current_thread_info(), fp)) + break; + + sf = (struct sparc_stackf *) fp; + regs = (struct pt_regs *) (sf + 1); + + if (kstack_is_trap_frame(current_thread_info(), regs)) { + if (user_mode(regs)) + break; + pc = regs->tpc; + fp = regs->u_regs[UREG_I6] + STACK_BIAS; + } else { + pc = sf->callers_pc; + fp = (unsigned long)sf->fp + STACK_BIAS; + } + callchain_store(entry, pc); + } while (entry->nr < PERF_MAX_STACK_DEPTH); +} + +static void perf_callchain_user_64(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + unsigned long ufp; + + callchain_store(entry, PERF_CONTEXT_USER); + callchain_store(entry, regs->tpc); + + ufp = regs->u_regs[UREG_I6] + STACK_BIAS; + do { + struct sparc_stackf *usf, sf; + unsigned long pc; + + usf = (struct sparc_stackf *) ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + + pc = sf.callers_pc; + ufp = (unsigned long)sf.fp + STACK_BIAS; + callchain_store(entry, pc); + } while (entry->nr < PERF_MAX_STACK_DEPTH); +} + +static void perf_callchain_user_32(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + unsigned long ufp; + + callchain_store(entry, PERF_CONTEXT_USER); + callchain_store(entry, regs->tpc); + + ufp = regs->u_regs[UREG_I6] & 0xffffffffUL; + do { + struct sparc_stackf32 *usf, sf; + unsigned long pc; + + usf = (struct sparc_stackf32 *) ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + + pc = sf.callers_pc; + ufp = (unsigned long)sf.fp; + callchain_store(entry, pc); + } while (entry->nr < PERF_MAX_STACK_DEPTH); +} + +/* Like powerpc we can't get PMU interrupts within the PMU handler, + * so no need for separate NMI and IRQ chains as on x86. + */ +static DEFINE_PER_CPU(struct perf_callchain_entry, callchain); + +struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + struct perf_callchain_entry *entry = &__get_cpu_var(callchain); + + entry->nr = 0; + if (!user_mode(regs)) { + stack_trace_flush(); + perf_callchain_kernel(regs, entry); + if (current->mm) + regs = task_pt_regs(current); + else + regs = NULL; + } + if (regs) { + flushw_user(); + if (test_thread_flag(TIF_32BIT)) + perf_callchain_user_32(regs, entry); + else + perf_callchain_user_64(regs, entry); + } + return entry; +} diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c index 2830b415e21..40e29fc8a4d 100644 --- a/arch/sparc/kernel/process_32.c +++ b/arch/sparc/kernel/process_32.c @@ -17,13 +17,13 @@ #include <linux/mm.h> #include <linux/stddef.h> #include <linux/ptrace.h> -#include <linux/slab.h> #include <linux/user.h> #include <linux/smp.h> #include <linux/reboot.h> #include <linux/delay.h> #include <linux/pm.h> #include <linux/init.h> +#include <linux/slab.h> #include <asm/auxio.h> #include <asm/oplib.h> @@ -526,7 +526,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, * Set some valid stack frames to give to the child. */ childstack = (struct sparc_stackf __user *) - (sp & ~0x7UL); + (sp & ~0xfUL); parentstack = (struct sparc_stackf __user *) regs->u_regs[UREG_FP]; diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index 18d67854a1b..a5cf3864b31 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -352,12 +352,6 @@ void exit_thread(void) else t->utraps[0]--; } - - if (test_and_clear_thread_flag(TIF_PERFCTR)) { - t->user_cntd0 = t->user_cntd1 = NULL; - t->pcr_reg = 0; - write_pcr(0); - } } void flush_thread(void) @@ -365,27 +359,12 @@ void flush_thread(void) struct thread_info *t = current_thread_info(); struct mm_struct *mm; - if (test_ti_thread_flag(t, TIF_ABI_PENDING)) { - clear_ti_thread_flag(t, TIF_ABI_PENDING); - if (test_ti_thread_flag(t, TIF_32BIT)) - clear_ti_thread_flag(t, TIF_32BIT); - else - set_ti_thread_flag(t, TIF_32BIT); - } - mm = t->task->mm; if (mm) tsb_context_switch(mm); set_thread_wsaved(0); - /* Turn off performance counters if on. */ - if (test_and_clear_thread_flag(TIF_PERFCTR)) { - t->user_cntd0 = t->user_cntd1 = NULL; - t->pcr_reg = 0; - write_pcr(0); - } - /* Clear FPU register state. */ t->fpsaved[0] = 0; @@ -406,11 +385,11 @@ static unsigned long clone_stackframe(unsigned long csp, unsigned long psp) } else __get_user(fp, &(((struct reg_window32 __user *)psp)->ins[6])); - /* Now 8-byte align the stack as this is mandatory in the - * Sparc ABI due to how register windows work. This hides - * the restriction from thread libraries etc. -DaveM + /* Now align the stack as this is mandatory in the Sparc ABI + * due to how register windows work. This hides the + * restriction from thread libraries etc. */ - csp &= ~7UL; + csp &= ~15UL; distance = fp - psp; rval = (csp - distance); @@ -599,16 +578,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, t->kregs->u_regs[UREG_FP] = ((unsigned long) child_sf) - STACK_BIAS; - /* Special case, if we are spawning a kernel thread from - * a userspace task (usermode helper, NFS or similar), we - * must disable performance counters in the child because - * the address space and protection realm are changing. - */ - if (t->flags & _TIF_PERFCTR) { - t->user_cntd0 = t->user_cntd1 = NULL; - t->pcr_reg = 0; - t->flags &= ~_TIF_PERFCTR; - } t->flags |= ((long)ASI_P << TI_FLAG_CURRENT_DS_SHIFT); t->kregs->u_regs[UREG_G6] = (unsigned long) t; t->kregs->u_regs[UREG_G4] = (unsigned long) t->task; diff --git a/arch/sparc/kernel/prom.h b/arch/sparc/kernel/prom.h index 453397fe5e1..a8591ef2636 100644 --- a/arch/sparc/kernel/prom.h +++ b/arch/sparc/kernel/prom.h @@ -4,9 +4,6 @@ #include <linux/spinlock.h> #include <asm/prom.h> -extern struct device_node *allnodes; /* temporary while merging */ -extern rwlock_t devtree_lock; /* temporary while merging */ - extern void * prom_early_alloc(unsigned long size); extern void irq_trans_init(struct device_node *dp); diff --git a/arch/sparc/kernel/prom_common.c b/arch/sparc/kernel/prom_common.c index d80a65d9e89..57ac9e28be0 100644 --- a/arch/sparc/kernel/prom_common.c +++ b/arch/sparc/kernel/prom_common.c @@ -37,18 +37,6 @@ EXPORT_SYMBOL(of_console_path); char *of_console_options; EXPORT_SYMBOL(of_console_options); -struct device_node *of_find_node_by_phandle(phandle handle) -{ - struct device_node *np; - - for (np = allnodes; np; np = np->allnext) - if (np->node == handle) - break; - - return np; -} -EXPORT_SYMBOL(of_find_node_by_phandle); - int of_getintprop_default(struct device_node *np, const char *name, int def) { struct property *prop; @@ -89,7 +77,7 @@ int of_set_property(struct device_node *dp, const char *name, void *val, int len void *old_val = prop->value; int ret; - ret = prom_setprop(dp->node, name, val, len); + ret = prom_setprop(dp->phandle, name, val, len); err = -EINVAL; if (ret >= 0) { @@ -236,7 +224,7 @@ static struct device_node * __init prom_create_node(phandle node, dp->name = get_one_property(node, "name"); dp->type = get_one_property(node, "device_type"); - dp->node = node; + dp->phandle = node; dp->properties = build_prop_list(node); @@ -313,7 +301,7 @@ void __init prom_build_devicetree(void) nextp = &allnodes->allnext; allnodes->child = prom_build_tree(allnodes, - prom_getchild(allnodes->node), + prom_getchild(allnodes->phandle), &nextp); of_console_init(); diff --git a/arch/sparc/kernel/ptrace_32.c b/arch/sparc/kernel/ptrace_32.c index 7e3dfd9bb97..e608f397e11 100644 --- a/arch/sparc/kernel/ptrace_32.c +++ b/arch/sparc/kernel/ptrace_32.c @@ -65,6 +65,7 @@ static int genregs32_get(struct task_struct *target, *k++ = regs->u_regs[pos++]; reg_window = (unsigned long __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; for (; count > 0 && pos < 32; count--) { if (get_user(*k++, ®_window[pos++])) return -EFAULT; @@ -76,6 +77,7 @@ static int genregs32_get(struct task_struct *target, } reg_window = (unsigned long __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; for (; count > 0 && pos < 32; count--) { if (get_user(reg, ®_window[pos++]) || put_user(reg, u++)) @@ -141,6 +143,7 @@ static int genregs32_set(struct task_struct *target, regs->u_regs[pos++] = *k++; reg_window = (unsigned long __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; for (; count > 0 && pos < 32; count--) { if (put_user(*k++, ®_window[pos++])) return -EFAULT; @@ -153,6 +156,7 @@ static int genregs32_set(struct task_struct *target, } reg_window = (unsigned long __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; for (; count > 0 && pos < 32; count--) { if (get_user(reg, u++) || put_user(reg, ®_window[pos++])) diff --git a/arch/sparc/kernel/ptrace_64.c b/arch/sparc/kernel/ptrace_64.c index 2f6524d1a81..aa90da08bf6 100644 --- a/arch/sparc/kernel/ptrace_64.c +++ b/arch/sparc/kernel/ptrace_64.c @@ -492,6 +492,7 @@ static int genregs32_get(struct task_struct *target, *k++ = regs->u_regs[pos++]; reg_window = (compat_ulong_t __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; if (target == current) { for (; count > 0 && pos < 32; count--) { if (get_user(*k++, ®_window[pos++])) @@ -516,6 +517,7 @@ static int genregs32_get(struct task_struct *target, } reg_window = (compat_ulong_t __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; if (target == current) { for (; count > 0 && pos < 32; count--) { if (get_user(reg, ®_window[pos++]) || @@ -599,6 +601,7 @@ static int genregs32_set(struct task_struct *target, regs->u_regs[pos++] = *k++; reg_window = (compat_ulong_t __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; if (target == current) { for (; count > 0 && pos < 32; count--) { if (put_user(*k++, ®_window[pos++])) @@ -625,6 +628,7 @@ static int genregs32_set(struct task_struct *target, } reg_window = (compat_ulong_t __user *) regs->u_regs[UREG_I6]; + reg_window -= 16; if (target == current) { for (; count > 0 && pos < 32; count--) { if (get_user(reg, u++) || diff --git a/arch/sparc/kernel/rtrap_64.S b/arch/sparc/kernel/rtrap_64.S index fd3cee4d117..090b9e9ad5e 100644 --- a/arch/sparc/kernel/rtrap_64.S +++ b/arch/sparc/kernel/rtrap_64.S @@ -65,48 +65,6 @@ __handle_user_windows: ba,pt %xcc, __handle_user_windows_continue andn %l1, %l4, %l1 -__handle_perfctrs: - call update_perfctrs - wrpr %g0, RTRAP_PSTATE, %pstate - wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate - ldub [%g6 + TI_WSAVED], %o2 - brz,pt %o2, 1f - nop - /* Redo userwin+sched+sig checks */ - call fault_in_user_windows - - wrpr %g0, RTRAP_PSTATE, %pstate - wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate - ldx [%g6 + TI_FLAGS], %l0 - andcc %l0, _TIF_NEED_RESCHED, %g0 - be,pt %xcc, 1f - - nop - call schedule - wrpr %g0, RTRAP_PSTATE, %pstate - wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate - ldx [%g6 + TI_FLAGS], %l0 -1: andcc %l0, _TIF_DO_NOTIFY_RESUME_MASK, %g0 - - be,pt %xcc, __handle_perfctrs_continue - sethi %hi(TSTATE_PEF), %o0 - mov %l5, %o1 - add %sp, PTREGS_OFF, %o0 - mov %l0, %o2 - call do_notify_resume - - wrpr %g0, RTRAP_PSTATE, %pstate - wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate - /* Signal delivery can modify pt_regs tstate, so we must - * reload it. - */ - ldx [%sp + PTREGS_OFF + PT_V9_TSTATE], %l1 - sethi %hi(0xf << 20), %l4 - and %l1, %l4, %l4 - andn %l1, %l4, %l1 - ba,pt %xcc, __handle_perfctrs_continue - - sethi %hi(TSTATE_PEF), %o0 __handle_userfpu: rd %fprs, %l5 andcc %l5, FPRS_FEF, %g0 @@ -149,11 +107,11 @@ rtrap_nmi: ldx [%sp + PTREGS_OFF + PT_V9_TSTATE], %l1 rtrap_irq: rtrap: #ifndef CONFIG_SMP - sethi %hi(per_cpu____cpu_data), %l0 - lduw [%l0 + %lo(per_cpu____cpu_data)], %l1 + sethi %hi(__cpu_data), %l0 + lduw [%l0 + %lo(__cpu_data)], %l1 #else - sethi %hi(per_cpu____cpu_data), %l0 - or %l0, %lo(per_cpu____cpu_data), %l0 + sethi %hi(__cpu_data), %l0 + or %l0, %lo(__cpu_data), %l0 lduw [%l0 + %g5], %l1 #endif cmp %l1, 0 @@ -172,7 +130,17 @@ rtrap_xcall: nop call trace_hardirqs_on nop - wrpr %l4, %pil + /* Do not actually set the %pil here. We will do that + * below after we clear PSTATE_IE in the %pstate register. + * If we re-enable interrupts here, we can recurse down + * the hardirq stack potentially endlessly, causing a + * stack overflow. + * + * It is tempting to put this test and trace_hardirqs_on + * call at the 'rt_continue' label, but that will not work + * as that path hits unconditionally and we do not want to + * execute this in NMI return paths, for example. + */ #endif rtrap_no_irq_enable: andcc %l1, TSTATE_PRIV, %l3 @@ -191,9 +159,9 @@ rtrap_no_irq_enable: * take until the next local IRQ before the signal/resched * event would be handled. * - * This also means that if we have to deal with performance - * counters or user windows, we have to redo all of these - * sched+signal checks with IRQs disabled. + * This also means that if we have to deal with user + * windows, we have to redo all of these sched+signal checks + * with IRQs disabled. */ to_user: wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate wrpr 0, %pil @@ -214,12 +182,8 @@ __handle_signal_continue: brnz,pn %o2, __handle_user_windows nop __handle_user_windows_continue: - ldx [%g6 + TI_FLAGS], %l5 - andcc %l5, _TIF_PERFCTR, %g0 sethi %hi(TSTATE_PEF), %o0 - bne,pn %xcc, __handle_perfctrs -__handle_perfctrs_continue: - andcc %l1, %o0, %g0 + andcc %l1, %o0, %g0 /* This fpdepth clear is necessary for non-syscall rtraps only */ user_nowork: diff --git a/arch/sparc/kernel/setup_32.c b/arch/sparc/kernel/setup_32.c index 9be2af55c5c..b22ce610040 100644 --- a/arch/sparc/kernel/setup_32.c +++ b/arch/sparc/kernel/setup_32.c @@ -95,8 +95,6 @@ static void prom_sync_me(void) "nop\n\t" "nop\n\t" : : "r" (prom_tbr)); local_irq_restore(flags); - - return; } static unsigned int boot_flags __initdata = 0; diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index a2a79e76344..5f72de67588 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -12,7 +12,6 @@ #include <linux/stddef.h> #include <linux/unistd.h> #include <linux/ptrace.h> -#include <linux/slab.h> #include <asm/smp.h> #include <linux/user.h> #include <linux/screen_info.h> diff --git a/arch/sparc/kernel/signal32.c b/arch/sparc/kernel/signal32.c index ba5b09ad666..ea22cd373c6 100644 --- a/arch/sparc/kernel/signal32.c +++ b/arch/sparc/kernel/signal32.c @@ -120,8 +120,8 @@ struct rt_signal_frame32 { }; /* Align macros */ -#define SF_ALIGNEDSZ (((sizeof(struct signal_frame32) + 7) & (~7))) -#define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame32) + 7) & (~7))) +#define SF_ALIGNEDSZ (((sizeof(struct signal_frame32) + 15) & (~15))) +#define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame32) + 15) & (~15))) int copy_siginfo_to_user32(compat_siginfo_t __user *to, siginfo_t *from) { @@ -420,15 +420,17 @@ static void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, uns sp = current->sas_ss_sp + current->sas_ss_size; } + sp -= framesize; + /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack * alignment. Second, if we took this signal because the stack * is not aligned properly, we'd like to take the signal cleanly * and report that. */ - sp &= ~7UL; + sp &= ~15UL; - return (void __user *)(sp - framesize); + return (void __user *) sp; } static int save_fpu_state32(struct pt_regs *regs, __siginfo_fpu_t __user *fpu) diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c index 7ce1a1005b1..9882df92ba0 100644 --- a/arch/sparc/kernel/signal_32.c +++ b/arch/sparc/kernel/signal_32.c @@ -267,15 +267,17 @@ static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *re sp = current->sas_ss_sp + current->sas_ss_size; } + sp -= framesize; + /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack * alignment. Second, if we took this signal because the stack * is not aligned properly, we'd like to take the signal cleanly * and report that. */ - sp &= ~7UL; + sp &= ~15UL; - return (void __user *)(sp - framesize); + return (void __user *) sp; } static inline int diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c index 647afbda7ae..9fa48c30037 100644 --- a/arch/sparc/kernel/signal_64.c +++ b/arch/sparc/kernel/signal_64.c @@ -353,7 +353,7 @@ segv: /* Checks if the fp is valid */ static int invalid_frame_pointer(void __user *fp, int fplen) { - if (((unsigned long) fp) & 7) + if (((unsigned long) fp) & 15) return 1; return 0; } @@ -396,15 +396,17 @@ static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs * sp = current->sas_ss_sp + current->sas_ss_size; } + sp -= framesize; + /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack * alignment. Second, if we took this signal because the stack * is not aligned properly, we'd like to take the signal cleanly * and report that. */ - sp &= ~7UL; + sp &= ~15UL; - return (void __user *)(sp - framesize); + return (void __user *) sp; } static inline void diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index aa36223497b..b6a2b8f4704 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -22,7 +22,9 @@ #include <linux/profile.h> #include <linux/bootmem.h> #include <linux/vmalloc.h> +#include <linux/ftrace.h> #include <linux/cpu.h> +#include <linux/slab.h> #include <asm/head.h> #include <asm/ptrace.h> @@ -370,7 +372,7 @@ static int __cpuinit smp_boot_one_cpu(unsigned int cpu) } else { struct device_node *dp = of_find_node_by_cpuid(cpu); - prom_startcpu(dp->node, entry, cookie); + prom_startcpu(dp->phandle, entry, cookie); } for (timeout = 0; timeout < 50000; timeout++) { @@ -822,13 +824,13 @@ void arch_send_call_function_single_ipi(int cpu) &cpumask_of_cpu(cpu)); } -void smp_call_function_client(int irq, struct pt_regs *regs) +void __irq_entry smp_call_function_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); generic_smp_call_function_interrupt(); } -void smp_call_function_single_client(int irq, struct pt_regs *regs) +void __irq_entry smp_call_function_single_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); generic_smp_call_function_single_interrupt(); @@ -964,7 +966,7 @@ void flush_dcache_page_all(struct mm_struct *mm, struct page *page) put_cpu(); } -void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) +void __irq_entry smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) { struct mm_struct *mm; unsigned long flags; @@ -1148,7 +1150,7 @@ void smp_release(void) */ extern void prom_world(int); -void smp_penguin_jailcell(int irq, struct pt_regs *regs) +void __irq_entry smp_penguin_jailcell(int irq, struct pt_regs *regs) { clear_softint(1 << irq); @@ -1364,7 +1366,7 @@ void smp_send_reschedule(int cpu) &cpumask_of_cpu(cpu)); } -void smp_receive_signal_client(int irq, struct pt_regs *regs) +void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); } diff --git a/arch/sparc/kernel/sun4c_irq.c b/arch/sparc/kernel/sun4c_irq.c index bc3adbf79c6..892fb884910 100644 --- a/arch/sparc/kernel/sun4c_irq.c +++ b/arch/sparc/kernel/sun4c_irq.c @@ -16,7 +16,6 @@ #include <linux/sched.h> #include <linux/ptrace.h> #include <linux/interrupt.h> -#include <linux/slab.h> #include <linux/init.h> #include <linux/of.h> #include <linux/of_device.h> diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index 68791cad7b7..482f2ab9269 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -194,7 +194,7 @@ int __cpuinit smp4d_boot_one_cpu(int i) smp_penguin_ctable.reg_size = 0; /* whirrr, whirrr, whirrrrrrrrr... */ - SMP_PRINTK(("Starting CPU %d at %p \n", i, entry)); + SMP_PRINTK(("Starting CPU %d at %p\n", i, entry)); local_flush_cache_all(); prom_startcpu(cpu_node, &smp_penguin_ctable, 0, (char *)entry); diff --git a/arch/sparc/kernel/sun4m_irq.c b/arch/sparc/kernel/sun4m_irq.c index 301892e2d71..7f3b97ff62c 100644 --- a/arch/sparc/kernel/sun4m_irq.c +++ b/arch/sparc/kernel/sun4m_irq.c @@ -17,7 +17,6 @@ #include <linux/ptrace.h> #include <linux/smp.h> #include <linux/interrupt.h> -#include <linux/slab.h> #include <linux/init.h> #include <linux/ioport.h> #include <linux/of.h> diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index e7061138c98..46a76ba3fb4 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -51,7 +51,6 @@ SIGN1(sys32_exit_group, sys_exit_group, %o0) SIGN1(sys32_wait4, compat_sys_wait4, %o2) SIGN1(sys32_creat, sys_creat, %o1) SIGN1(sys32_mknod, sys_mknod, %o1) -SIGN1(sys32_perfctr, sys_perfctr, %o0) SIGN1(sys32_umount, sys_umount, %o1) SIGN1(sys32_signal, sys_signal, %o0) SIGN1(sys32_access, sys_access, %o1) diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c index dc0ac197e7e..c0ca87553e1 100644 --- a/arch/sparc/kernel/sys_sparc32.c +++ b/arch/sparc/kernel/sys_sparc32.c @@ -21,7 +21,6 @@ #include <linux/sem.h> #include <linux/msg.h> #include <linux/shm.h> -#include <linux/slab.h> #include <linux/uio.h> #include <linux/nfs_fs.h> #include <linux/quota.h> @@ -43,8 +42,8 @@ #include <linux/security.h> #include <linux/compat.h> #include <linux/vfs.h> -#include <linux/netfilter_ipv4/ip_tables.h> #include <linux/ptrace.h> +#include <linux/slab.h> #include <asm/types.h> #include <asm/uaccess.h> diff --git a/arch/sparc/kernel/sys_sparc_32.c b/arch/sparc/kernel/sys_sparc_32.c index 3a82e65d8db..ee995b7dae7 100644 --- a/arch/sparc/kernel/sys_sparc_32.c +++ b/arch/sparc/kernel/sys_sparc_32.c @@ -98,119 +98,6 @@ out: return error; } -/* - * sys_ipc() is the de-multiplexer for the SysV IPC calls.. - * - * This is really horribly ugly. - */ - -asmlinkage int sys_ipc (uint call, int first, int second, int third, void __user *ptr, long fifth) -{ - int version, err; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - if (call <= SEMCTL) - switch (call) { - case SEMOP: - err = sys_semtimedop (first, (struct sembuf __user *)ptr, second, NULL); - goto out; - case SEMTIMEDOP: - err = sys_semtimedop (first, (struct sembuf __user *)ptr, second, (const struct timespec __user *) fifth); - goto out; - case SEMGET: - err = sys_semget (first, second, third); - goto out; - case SEMCTL: { - union semun fourth; - err = -EINVAL; - if (!ptr) - goto out; - err = -EFAULT; - if (get_user(fourth.__pad, - (void __user * __user *)ptr)) - goto out; - err = sys_semctl (first, second, third, fourth); - goto out; - } - default: - err = -ENOSYS; - goto out; - } - if (call <= MSGCTL) - switch (call) { - case MSGSND: - err = sys_msgsnd (first, (struct msgbuf __user *) ptr, - second, third); - goto out; - case MSGRCV: - switch (version) { - case 0: { - struct ipc_kludge tmp; - err = -EINVAL; - if (!ptr) - goto out; - err = -EFAULT; - if (copy_from_user(&tmp, (struct ipc_kludge __user *) ptr, sizeof (tmp))) - goto out; - err = sys_msgrcv (first, tmp.msgp, second, tmp.msgtyp, third); - goto out; - } - case 1: default: - err = sys_msgrcv (first, - (struct msgbuf __user *) ptr, - second, fifth, third); - goto out; - } - case MSGGET: - err = sys_msgget ((key_t) first, second); - goto out; - case MSGCTL: - err = sys_msgctl (first, second, (struct msqid_ds __user *) ptr); - goto out; - default: - err = -ENOSYS; - goto out; - } - if (call <= SHMCTL) - switch (call) { - case SHMAT: - switch (version) { - case 0: default: { - ulong raddr; - err = do_shmat (first, (char __user *) ptr, second, &raddr); - if (err) - goto out; - err = -EFAULT; - if (put_user (raddr, (ulong __user *) third)) - goto out; - err = 0; - goto out; - } - case 1: /* iBCS2 emulator entry point */ - err = -EINVAL; - goto out; - } - case SHMDT: - err = sys_shmdt ((char __user *)ptr); - goto out; - case SHMGET: - err = sys_shmget (first, second, third); - goto out; - case SHMCTL: - err = sys_shmctl (first, second, (struct shmid_ds __user *) ptr); - goto out; - default: - err = -ENOSYS; - goto out; - } - else - err = -ENOSYS; -out: - return err; -} - int sparc_mmap_check(unsigned long addr, unsigned long len) { if (ARCH_SUN4C && diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index cfa0e19abe3..3d435c42e6d 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -27,7 +27,6 @@ #include <asm/uaccess.h> #include <asm/utrap.h> -#include <asm/perfctr.h> #include <asm/unistd.h> #include "entry.h" @@ -365,6 +364,7 @@ EXPORT_SYMBOL(get_fb_unmapped_area); void arch_pick_mmap_layout(struct mm_struct *mm) { unsigned long random_factor = 0UL; + unsigned long gap; if (current->flags & PF_RANDOMIZE) { random_factor = get_random_int(); @@ -379,9 +379,10 @@ void arch_pick_mmap_layout(struct mm_struct *mm) * Fall back to the standard layout if the personality * bit is set, or if the expected stack growth is unlimited: */ + gap = rlimit(RLIMIT_STACK); if (!test_thread_flag(TIF_32BIT) || (current->personality & ADDR_COMPAT_LAYOUT) || - current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY || + gap == RLIM_INFINITY || sysctl_legacy_va_layout) { mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; mm->get_unmapped_area = arch_get_unmapped_area; @@ -389,9 +390,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm) } else { /* We know it's 32-bit */ unsigned long task_size = STACK_TOP32; - unsigned long gap; - gap = current->signal->rlim[RLIMIT_STACK].rlim_cur; if (gap < 128 * 1024 * 1024) gap = 128 * 1024 * 1024; if (gap > (task_size / 6 * 5)) @@ -427,7 +426,7 @@ out: * This is really horribly ugly. */ -SYSCALL_DEFINE6(ipc, unsigned int, call, int, first, unsigned long, second, +SYSCALL_DEFINE6(sparc_ipc, unsigned int, call, int, first, unsigned long, second, unsigned long, third, void __user *, ptr, long, fifth) { long err; @@ -511,17 +510,6 @@ out: return err; } -SYSCALL_DEFINE1(sparc64_newuname, struct new_utsname __user *, name) -{ - int ret = sys_newuname(name); - - if (current->personality == PER_LINUX32 && !ret) { - ret = (copy_to_user(name->machine, "sparc\0\0", 8) - ? -EFAULT : 0); - } - return ret; -} - SYSCALL_DEFINE1(sparc64_personality, unsigned long, personality) { int ret; @@ -766,109 +754,6 @@ SYSCALL_DEFINE5(rt_sigaction, int, sig, const struct sigaction __user *, act, return ret; } -/* Invoked by rtrap code to update performance counters in - * user space. - */ -asmlinkage void update_perfctrs(void) -{ - unsigned long pic, tmp; - - read_pic(pic); - tmp = (current_thread_info()->kernel_cntd0 += (unsigned int)pic); - __put_user(tmp, current_thread_info()->user_cntd0); - tmp = (current_thread_info()->kernel_cntd1 += (pic >> 32)); - __put_user(tmp, current_thread_info()->user_cntd1); - reset_pic(); -} - -SYSCALL_DEFINE4(perfctr, int, opcode, unsigned long, arg0, - unsigned long, arg1, unsigned long, arg2) -{ - int err = 0; - - switch(opcode) { - case PERFCTR_ON: - current_thread_info()->pcr_reg = arg2; - current_thread_info()->user_cntd0 = (u64 __user *) arg0; - current_thread_info()->user_cntd1 = (u64 __user *) arg1; - current_thread_info()->kernel_cntd0 = - current_thread_info()->kernel_cntd1 = 0; - write_pcr(arg2); - reset_pic(); - set_thread_flag(TIF_PERFCTR); - break; - - case PERFCTR_OFF: - err = -EINVAL; - if (test_thread_flag(TIF_PERFCTR)) { - current_thread_info()->user_cntd0 = - current_thread_info()->user_cntd1 = NULL; - current_thread_info()->pcr_reg = 0; - write_pcr(0); - clear_thread_flag(TIF_PERFCTR); - err = 0; - } - break; - - case PERFCTR_READ: { - unsigned long pic, tmp; - - if (!test_thread_flag(TIF_PERFCTR)) { - err = -EINVAL; - break; - } - read_pic(pic); - tmp = (current_thread_info()->kernel_cntd0 += (unsigned int)pic); - err |= __put_user(tmp, current_thread_info()->user_cntd0); - tmp = (current_thread_info()->kernel_cntd1 += (pic >> 32)); - err |= __put_user(tmp, current_thread_info()->user_cntd1); - reset_pic(); - break; - } - - case PERFCTR_CLRPIC: - if (!test_thread_flag(TIF_PERFCTR)) { - err = -EINVAL; - break; - } - current_thread_info()->kernel_cntd0 = - current_thread_info()->kernel_cntd1 = 0; - reset_pic(); - break; - - case PERFCTR_SETPCR: { - u64 __user *user_pcr = (u64 __user *)arg0; - - if (!test_thread_flag(TIF_PERFCTR)) { - err = -EINVAL; - break; - } - err |= __get_user(current_thread_info()->pcr_reg, user_pcr); - write_pcr(current_thread_info()->pcr_reg); - current_thread_info()->kernel_cntd0 = - current_thread_info()->kernel_cntd1 = 0; - reset_pic(); - break; - } - - case PERFCTR_GETPCR: { - u64 __user *user_pcr = (u64 __user *)arg0; - - if (!test_thread_flag(TIF_PERFCTR)) { - err = -EINVAL; - break; - } - err |= __put_user(current_thread_info()->pcr_reg, user_pcr); - break; - } - - default: - err = -EINVAL; - break; - }; - return err; -} - /* * Do a system call from kernel instead of calling sys_execve so we * end up with proper pt_regs. diff --git a/arch/sparc/kernel/syscalls.S b/arch/sparc/kernel/syscalls.S index dc4a458f74d..1d7e274f3f2 100644 --- a/arch/sparc/kernel/syscalls.S +++ b/arch/sparc/kernel/syscalls.S @@ -110,31 +110,12 @@ sys_clone: .globl ret_from_syscall ret_from_syscall: - /* Clear current_thread_info()->new_child, and - * check performance counter stuff too. - */ + /* Clear current_thread_info()->new_child. */ stb %g0, [%g6 + TI_NEW_CHILD] ldx [%g6 + TI_FLAGS], %l0 call schedule_tail mov %g7, %o0 - andcc %l0, _TIF_PERFCTR, %g0 - be,pt %icc, 1f - nop - ldx [%g6 + TI_PCR], %o7 - wr %g0, %o7, %pcr - - /* Blackbird errata workaround. See commentary in - * smp.c:smp_percpu_timer_interrupt() for more - * information. - */ - ba,pt %xcc, 99f - nop - - .align 64 -99: wr %g0, %g0, %pic - rd %pic, %g0 - -1: ba,pt %xcc, ret_sys_call + ba,pt %xcc, ret_sys_call ldx [%sp + PTREGS_OFF + PT_V9_I0], %o0 .globl sparc_exit diff --git a/arch/sparc/kernel/sysfs.c b/arch/sparc/kernel/sysfs.c index ca39c606fe8..1eb8b00aed7 100644 --- a/arch/sparc/kernel/sysfs.c +++ b/arch/sparc/kernel/sysfs.c @@ -107,12 +107,12 @@ static unsigned long run_on_cpu(unsigned long cpu, unsigned long ret; /* should return -EINVAL to userspace */ - if (set_cpus_allowed(current, cpumask_of_cpu(cpu))) + if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) return 0; ret = func(arg); - set_cpus_allowed(current, old_affinity); + set_cpus_allowed_ptr(current, &old_affinity); return ret; } diff --git a/arch/sparc/kernel/systbls.h b/arch/sparc/kernel/systbls.h index d2f999ae2b8..118759cd734 100644 --- a/arch/sparc/kernel/systbls.h +++ b/arch/sparc/kernel/systbls.h @@ -6,15 +6,12 @@ #include <asm/utrap.h> #include <asm/signal.h> -struct new_utsname; - extern asmlinkage unsigned long sys_getpagesize(void); extern asmlinkage long sparc_pipe(struct pt_regs *regs); -extern asmlinkage long sys_ipc(unsigned int call, int first, +extern asmlinkage long sys_sparc_ipc(unsigned int call, int first, unsigned long second, unsigned long third, void __user *ptr, long fifth); -extern asmlinkage long sparc64_newuname(struct new_utsname __user *name); extern asmlinkage long sparc64_personality(unsigned long personality); extern asmlinkage long sys64_munmap(unsigned long addr, size_t len); extern asmlinkage unsigned long sys64_mremap(unsigned long addr, @@ -36,8 +33,6 @@ extern asmlinkage long sys_rt_sigaction(int sig, struct sigaction __user *oact, void __user *restorer, size_t sigsetsize); -extern asmlinkage long sys_perfctr(int opcode, unsigned long arg0, - unsigned long arg1, unsigned long arg2); extern asmlinkage void sparc64_set_context(struct pt_regs *regs); extern asmlinkage void sparc64_get_context(struct pt_regs *regs); diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S index e575b46bd7a..9db058dd039 100644 --- a/arch/sparc/kernel/systbls_64.S +++ b/arch/sparc/kernel/systbls_64.S @@ -21,7 +21,7 @@ sys_call_table32: /*0*/ .word sys_restart_syscall, sys32_exit, sys_fork, sys_read, sys_write /*5*/ .word sys32_open, sys_close, sys32_wait4, sys32_creat, sys_link /*10*/ .word sys_unlink, sunos_execv, sys_chdir, sys_chown16, sys32_mknod -/*15*/ .word sys_chmod, sys_lchown16, sys_brk, sys32_perfctr, sys32_lseek +/*15*/ .word sys_chmod, sys_lchown16, sys_brk, sys_nis_syscall, sys32_lseek /*20*/ .word sys_getpid, sys_capget, sys_capset, sys_setuid16, sys_getuid16 /*25*/ .word sys32_vmsplice, compat_sys_ptrace, sys_alarm, sys32_sigaltstack, sys_pause /*30*/ .word compat_sys_utime, sys_lchown, sys_fchown, sys32_access, sys32_nice @@ -55,7 +55,7 @@ sys_call_table32: /*170*/ .word sys32_lsetxattr, sys32_fsetxattr, sys_getxattr, sys_lgetxattr, compat_sys_getdents .word sys_setsid, sys_fchdir, sys32_fgetxattr, sys_listxattr, sys_llistxattr /*180*/ .word sys32_flistxattr, sys_removexattr, sys_lremovexattr, compat_sys_sigpending, sys_ni_syscall - .word sys32_setpgid, sys32_fremovexattr, sys32_tkill, sys32_exit_group, sys_sparc64_newuname + .word sys32_setpgid, sys32_fremovexattr, sys32_tkill, sys32_exit_group, sys_newuname /*190*/ .word sys32_init_module, sys_sparc64_personality, sys_remap_file_pages, sys32_epoll_create, sys32_epoll_ctl .word sys32_epoll_wait, sys32_ioprio_set, sys_getppid, sys32_sigaction, sys_sgetmask /*200*/ .word sys32_ssetmask, sys_sigsuspend, compat_sys_newlstat, sys_uselib, compat_sys_old_readdir @@ -96,7 +96,7 @@ sys_call_table: /*0*/ .word sys_restart_syscall, sparc_exit, sys_fork, sys_read, sys_write /*5*/ .word sys_open, sys_close, sys_wait4, sys_creat, sys_link /*10*/ .word sys_unlink, sys_nis_syscall, sys_chdir, sys_chown, sys_mknod -/*15*/ .word sys_chmod, sys_lchown, sys_brk, sys_perfctr, sys_lseek +/*15*/ .word sys_chmod, sys_lchown, sys_brk, sys_nis_syscall, sys_lseek /*20*/ .word sys_getpid, sys_capget, sys_capset, sys_setuid, sys_getuid /*25*/ .word sys_vmsplice, sys_ptrace, sys_alarm, sys_sigaltstack, sys_nis_syscall /*30*/ .word sys_utime, sys_nis_syscall, sys_nis_syscall, sys_access, sys_nice @@ -130,13 +130,13 @@ sys_call_table: /*170*/ .word sys_lsetxattr, sys_fsetxattr, sys_getxattr, sys_lgetxattr, sys_getdents .word sys_setsid, sys_fchdir, sys_fgetxattr, sys_listxattr, sys_llistxattr /*180*/ .word sys_flistxattr, sys_removexattr, sys_lremovexattr, sys_nis_syscall, sys_ni_syscall - .word sys_setpgid, sys_fremovexattr, sys_tkill, sys_exit_group, sys_sparc64_newuname + .word sys_setpgid, sys_fremovexattr, sys_tkill, sys_exit_group, sys_newuname /*190*/ .word sys_init_module, sys_sparc64_personality, sys_remap_file_pages, sys_epoll_create, sys_epoll_ctl .word sys_epoll_wait, sys_ioprio_set, sys_getppid, sys_nis_syscall, sys_sgetmask /*200*/ .word sys_ssetmask, sys_nis_syscall, sys_newlstat, sys_uselib, sys_nis_syscall .word sys_readahead, sys_socketcall, sys_syslog, sys_lookup_dcookie, sys_fadvise64 /*210*/ .word sys_fadvise64_64, sys_tgkill, sys_waitpid, sys_swapoff, sys_sysinfo - .word sys_ipc, sys_nis_syscall, sys_clone, sys_ioprio_get, sys_adjtimex + .word sys_sparc_ipc, sys_nis_syscall, sys_clone, sys_ioprio_get, sys_adjtimex /*220*/ .word sys_nis_syscall, sys_ni_syscall, sys_delete_module, sys_ni_syscall, sys_getpgid .word sys_bdflush, sys_sysfs, sys_nis_syscall, sys_setfsuid, sys_setfsgid /*230*/ .word sys_select, sys_nis_syscall, sys_splice, sys_stime, sys_statfs64 diff --git a/arch/sparc/kernel/time_32.c b/arch/sparc/kernel/time_32.c index 5b2f595fe65..0d4c09b15ef 100644 --- a/arch/sparc/kernel/time_32.c +++ b/arch/sparc/kernel/time_32.c @@ -35,6 +35,7 @@ #include <linux/platform_device.h> #include <asm/oplib.h> +#include <asm/timex.h> #include <asm/timer.h> #include <asm/system.h> #include <asm/irq.h> @@ -51,7 +52,6 @@ DEFINE_SPINLOCK(rtc_lock); EXPORT_SYMBOL(rtc_lock); static int set_rtc_mmss(unsigned long); -static int sbus_do_settimeofday(struct timespec *tv); unsigned long profile_pc(struct pt_regs *regs) { @@ -76,6 +76,8 @@ EXPORT_SYMBOL(profile_pc); __volatile__ unsigned int *master_l10_counter; +u32 (*do_arch_gettimeoffset)(void); + /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -196,35 +198,14 @@ static int __init clock_init(void) { return of_register_driver(&clock_driver, &of_platform_bus_type); } - /* Must be after subsys_initcall() so that busses are probed. Must * be before device_initcall() because things like the RTC driver * need to see the clock registers. */ fs_initcall(clock_init); -static void __init sbus_time_init(void) -{ - - BTFIXUPSET_CALL(bus_do_settimeofday, sbus_do_settimeofday, BTFIXUPCALL_NORM); - btfixup(); - - sparc_init_timers(timer_interrupt); -} - -void __init time_init(void) -{ -#ifdef CONFIG_PCI - extern void pci_time_init(void); - if (pcic_present()) { - pci_time_init(); - return; - } -#endif - sbus_time_init(); -} -static inline unsigned long do_gettimeoffset(void) +u32 sbus_do_gettimeoffset(void) { unsigned long val = *master_l10_counter; unsigned long usec = (val >> 10) & 0x1fffff; @@ -233,86 +214,39 @@ static inline unsigned long do_gettimeoffset(void) if (val & 0x80000000) usec += 1000000 / HZ; - return usec; + return usec * 1000; } -/* Ok, my cute asm atomicity trick doesn't work anymore. - * There are just too many variables that need to be protected - * now (both members of xtime, et al.) - */ -void do_gettimeofday(struct timeval *tv) -{ - unsigned long flags; - unsigned long seq; - unsigned long usec, sec; - unsigned long max_ntp_tick = tick_usec - tickadj; - - do { - seq = read_seqbegin_irqsave(&xtime_lock, flags); - usec = do_gettimeoffset(); - - /* - * If time_adjust is negative then NTP is slowing the clock - * so make sure not to go into next possible interval. - * Better to lose some accuracy than have time go backwards.. - */ - if (unlikely(time_adjust < 0)) - usec = min(usec, max_ntp_tick); - - sec = xtime.tv_sec; - usec += (xtime.tv_nsec / 1000); - } while (read_seqretry_irqrestore(&xtime_lock, seq, flags)); - - while (usec >= 1000000) { - usec -= 1000000; - sec++; - } - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -EXPORT_SYMBOL(do_gettimeofday); - -int do_settimeofday(struct timespec *tv) +u32 arch_gettimeoffset(void) { - int ret; - - write_seqlock_irq(&xtime_lock); - ret = bus_do_settimeofday(tv); - write_sequnlock_irq(&xtime_lock); - clock_was_set(); - return ret; + if (unlikely(!do_arch_gettimeoffset)) + return 0; + return do_arch_gettimeoffset(); } -EXPORT_SYMBOL(do_settimeofday); - -static int sbus_do_settimeofday(struct timespec *tv) +static void __init sbus_time_init(void) { - time_t wtm_sec, sec = tv->tv_sec; - long wtm_nsec, nsec = tv->tv_nsec; + do_arch_gettimeoffset = sbus_do_gettimeoffset; - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - /* - * This is revolting. We need to set "xtime" correctly. However, the - * value in this location is the value at the most recent update of - * wall time. Discover what correction gettimeofday() would have - * made, and then undo it! - */ - nsec -= 1000 * do_gettimeoffset(); - - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); + btfixup(); - set_normalized_timespec(&xtime, sec, nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); + sparc_init_timers(timer_interrupt); +} - ntp_clear(); - return 0; +void __init time_init(void) +{ +#ifdef CONFIG_PCI + extern void pci_time_init(void); + if (pcic_present()) { + pci_time_init(); + return; + } +#endif + sbus_time_init(); } + static int set_rtc_mmss(unsigned long secs) { struct rtc_device *rtc = rtc_class_open("rtc0"); diff --git a/arch/sparc/kernel/time_64.c b/arch/sparc/kernel/time_64.c index 67e16510288..c7bbe6cf7b8 100644 --- a/arch/sparc/kernel/time_64.c +++ b/arch/sparc/kernel/time_64.c @@ -35,6 +35,7 @@ #include <linux/clocksource.h> #include <linux/of_device.h> #include <linux/platform_device.h> +#include <linux/ftrace.h> #include <asm/oplib.h> #include <asm/timer.h> @@ -717,7 +718,7 @@ static struct clock_event_device sparc64_clockevent = { }; static DEFINE_PER_CPU(struct clock_event_device, sparc64_events); -void timer_interrupt(int irq, struct pt_regs *regs) +void __irq_entry timer_interrupt(int irq, struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); unsigned long tick_mask = tick_ops->softint_mask; @@ -728,6 +729,7 @@ void timer_interrupt(int irq, struct pt_regs *regs) irq_enter(); + local_cpu_data().irq0_irqs++; kstat_incr_irqs_this_cpu(0, irq_to_desc(0)); if (unlikely(!evt->event_handler)) { diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 10f7bb9fc14..9da57f03298 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -17,6 +17,7 @@ #include <linux/mm.h> #include <linux/init.h> #include <linux/kdebug.h> +#include <linux/gfp.h> #include <asm/smp.h> #include <asm/delay.h> @@ -2202,27 +2203,6 @@ void dump_stack(void) EXPORT_SYMBOL(dump_stack); -static inline int is_kernel_stack(struct task_struct *task, - struct reg_window *rw) -{ - unsigned long rw_addr = (unsigned long) rw; - unsigned long thread_base, thread_end; - - if (rw_addr < PAGE_OFFSET) { - if (task != &init_task) - return 0; - } - - thread_base = (unsigned long) task_stack_page(task); - thread_end = thread_base + sizeof(union thread_union); - if (rw_addr >= thread_base && - rw_addr < thread_end && - !(rw_addr & 0x7UL)) - return 1; - - return 0; -} - static inline struct reg_window *kernel_stack_up(struct reg_window *rw) { unsigned long fp = rw->ins[6]; @@ -2251,6 +2231,7 @@ void die_if_kernel(char *str, struct pt_regs *regs) show_regs(regs); add_taint(TAINT_DIE); if (regs->tstate & TSTATE_PRIV) { + struct thread_info *tp = current_thread_info(); struct reg_window *rw = (struct reg_window *) (regs->u_regs[UREG_FP] + STACK_BIAS); @@ -2258,8 +2239,8 @@ void die_if_kernel(char *str, struct pt_regs *regs) * find some badly aligned kernel stack. */ while (rw && - count++ < 30&& - is_kernel_stack(current, rw)) { + count++ < 30 && + kstack_valid(tp, (unsigned long) rw)) { printk("Caller[%016lx]: %pS\n", rw->ins[7], (void *) rw->ins[7]); @@ -2548,15 +2529,6 @@ void __init trap_init(void) rwbuf_stkptrs) || TI_GSR != offsetof(struct thread_info, gsr) || TI_XFSR != offsetof(struct thread_info, xfsr) || - TI_USER_CNTD0 != offsetof(struct thread_info, - user_cntd0) || - TI_USER_CNTD1 != offsetof(struct thread_info, - user_cntd1) || - TI_KERN_CNTD0 != offsetof(struct thread_info, - kernel_cntd0) || - TI_KERN_CNTD1 != offsetof(struct thread_info, - kernel_cntd1) || - TI_PCR != offsetof(struct thread_info, pcr_reg) || TI_PRE_COUNT != offsetof(struct thread_info, preempt_count) || TI_NEW_CHILD != offsetof(struct thread_info, new_child) || diff --git a/arch/sparc/kernel/tsb.S b/arch/sparc/kernel/tsb.S index 8c91d9b29a2..db15d123f05 100644 --- a/arch/sparc/kernel/tsb.S +++ b/arch/sparc/kernel/tsb.S @@ -191,10 +191,12 @@ tsb_dtlb_load: tsb_itlb_load: /* Executable bit must be set. */ -661: andcc %g5, _PAGE_EXEC_4U, %g0 - .section .sun4v_1insn_patch, "ax" +661: sethi %hi(_PAGE_EXEC_4U), %g4 + andcc %g5, %g4, %g0 + .section .sun4v_2insn_patch, "ax" .word 661b andcc %g5, _PAGE_EXEC_4V, %g0 + nop .previous be,pn %xcc, tsb_do_fault diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index 378ca82b9cc..c752c4c479b 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -21,6 +21,7 @@ #include <linux/smp.h> #include <linux/bitops.h> #include <linux/perf_event.h> +#include <linux/ratelimit.h> #include <asm/fpumacro.h> enum direction { @@ -49,7 +50,7 @@ static inline enum direction decode_direction(unsigned int insn) } /* 16 = double-word, 8 = extra-word, 4 = word, 2 = half-word */ -static inline int decode_access_size(unsigned int insn) +static inline int decode_access_size(struct pt_regs *regs, unsigned int insn) { unsigned int tmp; @@ -65,7 +66,7 @@ static inline int decode_access_size(unsigned int insn) return 2; else { printk("Impossible unaligned trap. insn=%08x\n", insn); - die_if_kernel("Byte sized unaligned access?!?!", current_thread_info()->kregs); + die_if_kernel("Byte sized unaligned access?!?!", regs); /* GCC should never warn that control reaches the end * of this function without returning a value because @@ -274,13 +275,9 @@ static void kernel_mna_trap_fault(int fixup_tstate_asi) static void log_unaligned(struct pt_regs *regs) { - static unsigned long count, last_time; + static DEFINE_RATELIMIT_STATE(ratelimit, 5 * HZ, 5); - if (time_after(jiffies, last_time + 5 * HZ)) - count = 0; - if (count < 5) { - last_time = jiffies; - count++; + if (__ratelimit(&ratelimit)) { printk("Kernel unaligned access at TPC[%lx] %pS\n", regs->tpc, (void *) regs->tpc); } @@ -289,7 +286,7 @@ static void log_unaligned(struct pt_regs *regs) asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn) { enum direction dir = decode_direction(insn); - int size = decode_access_size(insn); + int size = decode_access_size(regs, insn); int orig_asi, asi; current_thread_info()->kern_una_regs = regs; @@ -636,7 +633,6 @@ daex: return; } advance(regs); - return; } void handle_stdfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr) @@ -685,5 +681,4 @@ daex: return; } advance(regs); - return; } diff --git a/arch/sparc/kernel/us2e_cpufreq.c b/arch/sparc/kernel/us2e_cpufreq.c index 791c15138f3..8f982b76c71 100644 --- a/arch/sparc/kernel/us2e_cpufreq.c +++ b/arch/sparc/kernel/us2e_cpufreq.c @@ -238,12 +238,12 @@ static unsigned int us2e_freq_get(unsigned int cpu) return 0; cpus_allowed = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(cpu)); + set_cpus_allowed_ptr(current, cpumask_of(cpu)); clock_tick = sparc64_get_clock_tick(cpu) / 1000; estar = read_hbreg(HBIRD_ESTAR_MODE_ADDR); - set_cpus_allowed(current, cpus_allowed); + set_cpus_allowed_ptr(current, &cpus_allowed); return clock_tick / estar_to_divisor(estar); } @@ -259,7 +259,7 @@ static void us2e_set_cpu_divider_index(unsigned int cpu, unsigned int index) return; cpus_allowed = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(cpu)); + set_cpus_allowed_ptr(current, cpumask_of(cpu)); new_freq = clock_tick = sparc64_get_clock_tick(cpu) / 1000; new_bits = index_to_estar_mode(index); @@ -281,7 +281,7 @@ static void us2e_set_cpu_divider_index(unsigned int cpu, unsigned int index) cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - set_cpus_allowed(current, cpus_allowed); + set_cpus_allowed_ptr(current, &cpus_allowed); } static int us2e_freq_target(struct cpufreq_policy *policy, diff --git a/arch/sparc/kernel/us3_cpufreq.c b/arch/sparc/kernel/us3_cpufreq.c index 365b6464e2c..f35d1e79454 100644 --- a/arch/sparc/kernel/us3_cpufreq.c +++ b/arch/sparc/kernel/us3_cpufreq.c @@ -86,12 +86,12 @@ static unsigned int us3_freq_get(unsigned int cpu) return 0; cpus_allowed = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(cpu)); + set_cpus_allowed_ptr(current, cpumask_of(cpu)); reg = read_safari_cfg(); ret = get_current_freq(cpu, reg); - set_cpus_allowed(current, cpus_allowed); + set_cpus_allowed_ptr(current, &cpus_allowed); return ret; } @@ -106,7 +106,7 @@ static void us3_set_cpu_divider_index(unsigned int cpu, unsigned int index) return; cpus_allowed = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(cpu)); + set_cpus_allowed_ptr(current, cpumask_of(cpu)); new_freq = sparc64_get_clock_tick(cpu) / 1000; switch (index) { @@ -140,7 +140,7 @@ static void us3_set_cpu_divider_index(unsigned int cpu, unsigned int index) cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - set_cpus_allowed(current, cpus_allowed); + set_cpus_allowed_ptr(current, &cpus_allowed); } static int us3_freq_target(struct cpufreq_policy *policy, diff --git a/arch/sparc/kernel/vio.c b/arch/sparc/kernel/vio.c index c28c71449a6..3cb1def9806 100644 --- a/arch/sparc/kernel/vio.c +++ b/arch/sparc/kernel/vio.c @@ -10,6 +10,7 @@ */ #include <linux/kernel.h> +#include <linux/slab.h> #include <linux/irq.h> #include <linux/init.h> diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 4e599259396..0c1e6783657 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -46,11 +46,16 @@ SECTIONS SCHED_TEXT LOCK_TEXT KPROBES_TEXT + IRQENTRY_TEXT *(.gnu.warning) } = 0 _etext = .; RO_DATA(PAGE_SIZE) + + /* Start of data section */ + _sdata = .; + .data1 : { *(.data1) } |