diff options
Diffstat (limited to 'arch/sparc/kernel')
32 files changed, 1166 insertions, 440 deletions
diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h index 0c218e4c088..cc3c5cb47cd 100644 --- a/arch/sparc/kernel/entry.h +++ b/arch/sparc/kernel/entry.h @@ -59,6 +59,13 @@ struct popc_6insn_patch_entry { extern struct popc_6insn_patch_entry __popc_6insn_patch, __popc_6insn_patch_end; +struct pause_patch_entry { + unsigned int addr; + unsigned int insns[3]; +}; +extern struct pause_patch_entry __pause_3insn_patch, + __pause_3insn_patch_end; + extern void __init per_cpu_patch(void); extern void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *, struct sun4v_1insn_patch_entry *); diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S index b42ddbf9651..2feb15c35d9 100644 --- a/arch/sparc/kernel/head_64.S +++ b/arch/sparc/kernel/head_64.S @@ -559,10 +559,10 @@ niagara_tlb_fixup: be,pt %xcc, niagara2_patch nop cmp %g1, SUN4V_CHIP_NIAGARA4 - be,pt %xcc, niagara2_patch + be,pt %xcc, niagara4_patch nop cmp %g1, SUN4V_CHIP_NIAGARA5 - be,pt %xcc, niagara2_patch + be,pt %xcc, niagara4_patch nop call generic_patch_copyops @@ -573,6 +573,16 @@ niagara_tlb_fixup: nop ba,a,pt %xcc, 80f +niagara4_patch: + call niagara4_patch_copyops + nop + call niagara4_patch_bzero + nop + call niagara4_patch_pageops + nop + + ba,a,pt %xcc, 80f + niagara2_patch: call niagara2_patch_copyops nop diff --git a/arch/sparc/kernel/hvapi.c b/arch/sparc/kernel/hvapi.c index 8593672838f..c0a2de0fd62 100644 --- a/arch/sparc/kernel/hvapi.c +++ b/arch/sparc/kernel/hvapi.c @@ -45,6 +45,7 @@ static struct api_info api_table[] = { { .group = HV_GRP_NIU, }, { .group = HV_GRP_VF_CPU, }, { .group = HV_GRP_KT_CPU, }, + { .group = HV_GRP_VT_CPU, }, { .group = HV_GRP_DIAG, .flags = FLAG_PRE_API }, }; @@ -193,7 +194,7 @@ void __init sun4v_hvapi_init(void) bad: prom_printf("HVAPI: Cannot register API group " - "%lx with major(%u) minor(%u)\n", + "%lx with major(%lu) minor(%lu)\n", group, major, minor); prom_halt(); } diff --git a/arch/sparc/kernel/hvcalls.S b/arch/sparc/kernel/hvcalls.S index 58d60de4d65..f3ab509b76a 100644 --- a/arch/sparc/kernel/hvcalls.S +++ b/arch/sparc/kernel/hvcalls.S @@ -805,3 +805,19 @@ ENTRY(sun4v_reboot_data_set) retl nop ENDPROC(sun4v_reboot_data_set) + +ENTRY(sun4v_vt_get_perfreg) + mov %o1, %o4 + mov HV_FAST_VT_GET_PERFREG, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop +ENDPROC(sun4v_vt_get_perfreg) + +ENTRY(sun4v_vt_set_perfreg) + mov HV_FAST_VT_SET_PERFREG, %o5 + ta HV_FAST_TRAP + retl + nop +ENDPROC(sun4v_vt_set_perfreg) diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S index 79f31036484..0746e5e32b3 100644 --- a/arch/sparc/kernel/ktlb.S +++ b/arch/sparc/kernel/ktlb.S @@ -188,31 +188,26 @@ valid_addr_bitmap_patch: be,pn %xcc, kvmap_dtlb_longpath 2: sethi %hi(kpte_linear_bitmap), %g2 - or %g2, %lo(kpte_linear_bitmap), %g2 /* Get the 256MB physical address index. */ sllx %g4, 21, %g5 - mov 1, %g7 + or %g2, %lo(kpte_linear_bitmap), %g2 srlx %g5, 21 + 28, %g5 + and %g5, (32 - 1), %g7 - /* Don't try this at home kids... this depends upon srlx - * only taking the low 6 bits of the shift count in %g5. - */ - sllx %g7, %g5, %g7 - - /* Divide by 64 to get the offset into the bitmask. */ - srlx %g5, 6, %g5 + /* Divide by 32 to get the offset into the bitmask. */ + srlx %g5, 5, %g5 + add %g7, %g7, %g7 sllx %g5, 3, %g5 - /* kern_linear_pte_xor[((mask & bit) ? 1 : 0)] */ + /* kern_linear_pte_xor[(mask >> shift) & 3)] */ ldx [%g2 + %g5], %g2 - andcc %g2, %g7, %g0 + srlx %g2, %g7, %g7 sethi %hi(kern_linear_pte_xor), %g5 + and %g7, 3, %g7 or %g5, %lo(kern_linear_pte_xor), %g5 - bne,a,pt %xcc, 1f - add %g5, 8, %g5 - -1: ldx [%g5], %g2 + sllx %g7, 3, %g7 + ldx [%g5 + %g7], %g2 .globl kvmap_linear_patch kvmap_linear_patch: diff --git a/arch/sparc/kernel/leon_kernel.c b/arch/sparc/kernel/leon_kernel.c index f8b6eee40bd..87f60ee6543 100644 --- a/arch/sparc/kernel/leon_kernel.c +++ b/arch/sparc/kernel/leon_kernel.c @@ -56,11 +56,13 @@ static inline unsigned int leon_eirq_get(int cpu) static void leon_handle_ext_irq(unsigned int irq, struct irq_desc *desc) { unsigned int eirq; + struct irq_bucket *p; int cpu = sparc_leon3_cpuid(); eirq = leon_eirq_get(cpu); - if ((eirq & 0x10) && irq_map[eirq]->irq) /* bit4 tells if IRQ happened */ - generic_handle_irq(irq_map[eirq]->irq); + p = irq_map[eirq]; + if ((eirq & 0x10) && p && p->irq) /* bit4 tells if IRQ happened */ + generic_handle_irq(p->irq); } /* The extended IRQ controller has been found, this function registers it */ diff --git a/arch/sparc/kernel/mdesc.c b/arch/sparc/kernel/mdesc.c index 6dc79628058..831c001604e 100644 --- a/arch/sparc/kernel/mdesc.c +++ b/arch/sparc/kernel/mdesc.c @@ -817,6 +817,30 @@ void __cpuinit mdesc_populate_present_mask(cpumask_t *mask) mdesc_iterate_over_cpus(record_one_cpu, NULL, mask); } +static void * __init check_one_pgsz(struct mdesc_handle *hp, u64 mp, int cpuid, void *arg) +{ + const u64 *pgsz_prop = mdesc_get_property(hp, mp, "mmu-page-size-list", NULL); + unsigned long *pgsz_mask = arg; + u64 val; + + val = (HV_PGSZ_MASK_8K | HV_PGSZ_MASK_64K | + HV_PGSZ_MASK_512K | HV_PGSZ_MASK_4MB); + if (pgsz_prop) + val = *pgsz_prop; + + if (!*pgsz_mask) + *pgsz_mask = val; + else + *pgsz_mask &= val; + return NULL; +} + +void __init mdesc_get_page_sizes(cpumask_t *mask, unsigned long *pgsz_mask) +{ + *pgsz_mask = 0; + mdesc_iterate_over_cpus(check_one_pgsz, pgsz_mask, mask); +} + static void * __cpuinit fill_in_one_cpu(struct mdesc_handle *hp, u64 mp, int cpuid, void *arg) { const u64 *cfreq = mdesc_get_property(hp, mp, "clock-frequency", NULL); diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index eb1c1f010a4..6479256fd5a 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -22,7 +22,6 @@ #include <asm/perf_event.h> #include <asm/ptrace.h> #include <asm/pcr.h> -#include <asm/perfctr.h> #include "kstack.h" @@ -109,7 +108,7 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) touched = 1; else - pcr_ops->write(PCR_PIC_PRIV); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); sum = local_cpu_data().irq0_irqs; if (__get_cpu_var(nmi_touch)) { @@ -126,8 +125,8 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) __this_cpu_write(alert_counter, 0); } if (__get_cpu_var(wd_enabled)) { - write_pic(picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } restore_hardirq_stack(orig_sp); @@ -166,7 +165,7 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count) void stop_nmi_watchdog(void *unused) { - pcr_ops->write(PCR_PIC_PRIV); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); __get_cpu_var(wd_enabled) = 0; atomic_dec(&nmi_active); } @@ -223,10 +222,10 @@ void start_nmi_watchdog(void *unused) __get_cpu_var(wd_enabled) = 1; atomic_inc(&nmi_active); - pcr_ops->write(PCR_PIC_PRIV); - write_pic(picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } static void nmi_adjust_hz_one(void *unused) @@ -234,10 +233,10 @@ static void nmi_adjust_hz_one(void *unused) if (!__get_cpu_var(wd_enabled)) return; - pcr_ops->write(PCR_PIC_PRIV); - write_pic(picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } void nmi_adjust_hz(unsigned int new_hz) diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c index acc8c838ff7..75b31bcdead 100644 --- a/arch/sparc/kernel/pci.c +++ b/arch/sparc/kernel/pci.c @@ -779,7 +779,7 @@ static int __pci_mmap_make_offset(struct pci_dev *pdev, static void __pci_mmap_set_flags(struct pci_dev *dev, struct vm_area_struct *vma, enum pci_mmap_state mmap_state) { - vma->vm_flags |= (VM_IO | VM_RESERVED); + vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; } /* Set vm_page_prot of VMA, as appropriate for this architecture, for a pci diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c index 7661e84a05a..051b69caeff 100644 --- a/arch/sparc/kernel/pci_sun4v.c +++ b/arch/sparc/kernel/pci_sun4v.c @@ -594,7 +594,7 @@ static int __devinit pci_sun4v_iommu_init(struct pci_pbm_info *pbm) printk(KERN_ERR PFX "Strange virtual-dma[%08x:%08x].\n", vdma[0], vdma[1]); return -EINVAL; - }; + } dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL); num_tsb_entries = vdma[1] / IO_PAGE_SIZE; diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index 0ce0dd2332a..269af58497a 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -13,23 +13,14 @@ #include <asm/pil.h> #include <asm/pcr.h> #include <asm/nmi.h> +#include <asm/asi.h> #include <asm/spitfire.h> -#include <asm/perfctr.h> /* This code is shared between various users of the performance * counters. Users will be oprofile, pseudo-NMI watchdog, and the * perf_event support layer. */ -#define PCR_SUN4U_ENABLE (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE) -#define PCR_N2_ENABLE (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE | \ - PCR_N2_TOE_OV1 | \ - (2 << PCR_N2_SL1_SHIFT) | \ - (0xff << PCR_N2_MASK1_SHIFT)) - -u64 pcr_enable; -unsigned int picl_shift; - /* Performance counter interrupts run unmasked at PIL level 15. * Therefore we can't do things like wakeups and other work * that expects IRQ disabling to be adhered to in locking etc. @@ -60,39 +51,144 @@ void arch_irq_work_raise(void) const struct pcr_ops *pcr_ops; EXPORT_SYMBOL_GPL(pcr_ops); -static u64 direct_pcr_read(void) +static u64 direct_pcr_read(unsigned long reg_num) { u64 val; - read_pcr(val); + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("rd %%pcr, %0" : "=r" (val)); return val; } -static void direct_pcr_write(u64 val) +static void direct_pcr_write(unsigned long reg_num, u64 val) +{ + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("wr %0, 0x0, %%pcr" : : "r" (val)); +} + +static u64 direct_pic_read(unsigned long reg_num) { - write_pcr(val); + u64 val; + + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("rd %%pic, %0" : "=r" (val)); + return val; +} + +static void direct_pic_write(unsigned long reg_num, u64 val) +{ + WARN_ON_ONCE(reg_num != 0); + + /* Blackbird errata workaround. See commentary in + * arch/sparc64/kernel/smp.c:smp_percpu_timer_interrupt() + * for more information. + */ + __asm__ __volatile__("ba,pt %%xcc, 99f\n\t" + " nop\n\t" + ".align 64\n" + "99:wr %0, 0x0, %%pic\n\t" + "rd %%pic, %%g0" : : "r" (val)); +} + +static u64 direct_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / nmi_hz; + + return ((u64)((0 - delta) & 0xffffffff)) << 32; } static const struct pcr_ops direct_pcr_ops = { - .read = direct_pcr_read, - .write = direct_pcr_write, + .read_pcr = direct_pcr_read, + .write_pcr = direct_pcr_write, + .read_pic = direct_pic_read, + .write_pic = direct_pic_write, + .nmi_picl_value = direct_picl_value, + .pcr_nmi_enable = (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE), + .pcr_nmi_disable = PCR_PIC_PRIV, }; -static void n2_pcr_write(u64 val) +static void n2_pcr_write(unsigned long reg_num, u64 val) { unsigned long ret; + WARN_ON_ONCE(reg_num != 0); if (val & PCR_N2_HTRACE) { ret = sun4v_niagara2_setperf(HV_N2_PERF_SPARC_CTL, val); if (ret != HV_EOK) - write_pcr(val); + direct_pcr_write(reg_num, val); } else - write_pcr(val); + direct_pcr_write(reg_num, val); +} + +static u64 n2_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / (nmi_hz << 2); + + return ((u64)((0 - delta) & 0xffffffff)) << 32; } static const struct pcr_ops n2_pcr_ops = { - .read = direct_pcr_read, - .write = n2_pcr_write, + .read_pcr = direct_pcr_read, + .write_pcr = n2_pcr_write, + .read_pic = direct_pic_read, + .write_pic = direct_pic_write, + .nmi_picl_value = n2_picl_value, + .pcr_nmi_enable = (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE | + PCR_N2_TOE_OV1 | + (2 << PCR_N2_SL1_SHIFT) | + (0xff << PCR_N2_MASK1_SHIFT)), + .pcr_nmi_disable = PCR_PIC_PRIV, +}; + +static u64 n4_pcr_read(unsigned long reg_num) +{ + unsigned long val; + + (void) sun4v_vt_get_perfreg(reg_num, &val); + + return val; +} + +static void n4_pcr_write(unsigned long reg_num, u64 val) +{ + (void) sun4v_vt_set_perfreg(reg_num, val); +} + +static u64 n4_pic_read(unsigned long reg_num) +{ + unsigned long val; + + __asm__ __volatile__("ldxa [%1] %2, %0" + : "=r" (val) + : "r" (reg_num * 0x8UL), "i" (ASI_PIC)); + + return val; +} + +static void n4_pic_write(unsigned long reg_num, u64 val) +{ + __asm__ __volatile__("stxa %0, [%1] %2" + : /* no outputs */ + : "r" (val), "r" (reg_num * 0x8UL), "i" (ASI_PIC)); +} + +static u64 n4_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / (nmi_hz << 2); + + return ((u64)((0 - delta) & 0xffffffff)); +} + +static const struct pcr_ops n4_pcr_ops = { + .read_pcr = n4_pcr_read, + .write_pcr = n4_pcr_write, + .read_pic = n4_pic_read, + .write_pic = n4_pic_write, + .nmi_picl_value = n4_picl_value, + .pcr_nmi_enable = (PCR_N4_PICNPT | PCR_N4_STRACE | + PCR_N4_UTRACE | PCR_N4_TOE | + (26 << PCR_N4_SL_SHIFT)), + .pcr_nmi_disable = PCR_N4_PICNPT, }; static unsigned long perf_hsvc_group; @@ -115,6 +211,10 @@ static int __init register_perf_hsvc(void) perf_hsvc_group = HV_GRP_KT_CPU; break; + case SUN4V_CHIP_NIAGARA4: + perf_hsvc_group = HV_GRP_VT_CPU; + break; + default: return -ENODEV; } @@ -139,6 +239,29 @@ static void __init unregister_perf_hsvc(void) sun4v_hvapi_unregister(perf_hsvc_group); } +static int __init setup_sun4v_pcr_ops(void) +{ + int ret = 0; + + switch (sun4v_chip_type) { + case SUN4V_CHIP_NIAGARA1: + case SUN4V_CHIP_NIAGARA2: + case SUN4V_CHIP_NIAGARA3: + pcr_ops = &n2_pcr_ops; + break; + + case SUN4V_CHIP_NIAGARA4: + pcr_ops = &n4_pcr_ops; + break; + + default: + ret = -ENODEV; + break; + } + + return ret; +} + int __init pcr_arch_init(void) { int err = register_perf_hsvc(); @@ -148,15 +271,14 @@ int __init pcr_arch_init(void) switch (tlb_type) { case hypervisor: - pcr_ops = &n2_pcr_ops; - pcr_enable = PCR_N2_ENABLE; - picl_shift = 2; + err = setup_sun4v_pcr_ops(); + if (err) + goto out_unregister; break; case cheetah: case cheetah_plus: pcr_ops = &direct_pcr_ops; - pcr_enable = PCR_SUN4U_ENABLE; break; case spitfire: diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 5713957dcb8..b5c38faa4ea 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -25,36 +25,48 @@ #include <linux/atomic.h> #include <asm/nmi.h> #include <asm/pcr.h> -#include <asm/perfctr.h> #include <asm/cacheflush.h> #include "kernel.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. +/* Two classes of sparc64 chips currently exist. All of which have + * 32-bit counters which can generate overflow interrupts on the + * transition from 0xffffffff to 0. * - * Both counters are controlled using a single control register. The - * only way to stop all sampling is to clear all of the context (user, - * supervisor, hypervisor) sampling enable bits. But these bits apply - * to both counters, thus the two counters can't be enabled/disabled - * individually. + * All chips upto and including SPARC-T3 have two performance + * counters. The two 32-bit counters are accessed in one go using a + * single 64-bit register. * - * The control register has two event fields, one for each of the two - * counters. It's thus nearly impossible to have one counter going - * while keeping the other one stopped. Therefore it is possible to - * get overflow interrupts for counters not currently "in use" and - * that condition must be checked in the overflow interrupt handler. + * On these older chips both counters are controlled using a single + * control register. The only way to stop all sampling is to clear + * all of the context (user, supervisor, hypervisor) sampling enable + * bits. But these bits apply to both counters, thus the two counters + * can't be enabled/disabled individually. + * + * Furthermore, the control register on these older chips have two + * event fields, one for each of the two counters. It's thus nearly + * impossible to have one counter going while keeping the other one + * stopped. Therefore it is possible to get overflow interrupts for + * counters not currently "in use" and that condition must be checked + * in the overflow interrupt handler. * * So we use a hack, in that we program inactive counters with the * "sw_count0" and "sw_count1" events. These count how many times * the instruction "sethi %hi(0xfc000), %g0" is executed. It's an * unusual way to encode a NOP and therefore will not trigger in * normal code. + * + * Starting with SPARC-T4 we have one control register per counter. + * And the counters are stored in individual registers. The registers + * for the counters are 64-bit but only a 32-bit counter is + * implemented. The event selections on SPARC-T4 lack any + * restrictions, therefore we can elide all of the complicated + * conflict resolution code we have for SPARC-T3 and earlier chips. */ -#define MAX_HWEVENTS 2 +#define MAX_HWEVENTS 4 +#define MAX_PCRS 4 #define MAX_PERIOD ((1UL << 32) - 1) #define PIC_UPPER_INDEX 0 @@ -90,8 +102,8 @@ struct cpu_hw_events { */ int current_idx[MAX_HWEVENTS]; - /* Software copy of %pcr register on this cpu. */ - u64 pcr; + /* Software copy of %pcr register(s) on this cpu. */ + u64 pcr[MAX_HWEVENTS]; /* Enabled/disable state. */ int enabled; @@ -103,6 +115,8 @@ 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. + * + * The mask is unused on SPARC-T4 and later. */ struct perf_event_map { u16 encoding; @@ -142,15 +156,53 @@ struct sparc_pmu { const struct perf_event_map *(*event_map)(int); const cache_map_t *cache_map; int max_events; + u32 (*read_pmc)(int); + void (*write_pmc)(int, u64); int upper_shift; int lower_shift; int event_mask; + int user_bit; + int priv_bit; int hv_bit; int irq_bit; int upper_nop; int lower_nop; + unsigned int flags; +#define SPARC_PMU_ALL_EXCLUDES_SAME 0x00000001 +#define SPARC_PMU_HAS_CONFLICTS 0x00000002 + int max_hw_events; + int num_pcrs; + int num_pic_regs; }; +static u32 sparc_default_read_pmc(int idx) +{ + u64 val; + + val = pcr_ops->read_pic(0); + if (idx == PIC_UPPER_INDEX) + val >>= 32; + + return val & 0xffffffff; +} + +static void sparc_default_write_pmc(int idx, u64 val) +{ + u64 shift, mask, pic; + + shift = 0; + if (idx == PIC_UPPER_INDEX) + shift = 32; + + mask = ((u64) 0xffffffff) << shift; + val <<= shift; + + pic = pcr_ops->read_pic(0); + pic &= ~mask; + pic |= val; + pcr_ops->write_pic(0, pic); +} + static const struct perf_event_map ultra3_perfmon_event_map[] = { [PERF_COUNT_HW_CPU_CYCLES] = { 0x0000, PIC_UPPER | PIC_LOWER }, [PERF_COUNT_HW_INSTRUCTIONS] = { 0x0001, PIC_UPPER | PIC_LOWER }, @@ -268,11 +320,20 @@ static const struct sparc_pmu ultra3_pmu = { .event_map = ultra3_event_map, .cache_map = &ultra3_cache_map, .max_events = ARRAY_SIZE(ultra3_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 11, .lower_shift = 4, .event_mask = 0x3f, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, .upper_nop = 0x1c, .lower_nop = 0x14, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, }; /* Niagara1 is very limited. The upper PIC is hard-locked to count @@ -397,11 +458,20 @@ static const struct sparc_pmu niagara1_pmu = { .event_map = niagara1_event_map, .cache_map = &niagara1_cache_map, .max_events = ARRAY_SIZE(niagara1_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 0, .lower_shift = 4, .event_mask = 0x7, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, .upper_nop = 0x0, .lower_nop = 0x0, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, }; static const struct perf_event_map niagara2_perfmon_event_map[] = { @@ -523,13 +593,203 @@ static const struct sparc_pmu niagara2_pmu = { .event_map = niagara2_event_map, .cache_map = &niagara2_cache_map, .max_events = ARRAY_SIZE(niagara2_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 19, .lower_shift = 6, .event_mask = 0xfff, - .hv_bit = 0x8, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, + .hv_bit = PCR_N2_HTRACE, .irq_bit = 0x30, .upper_nop = 0x220, .lower_nop = 0x220, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, +}; + +static const struct perf_event_map niagara4_perfmon_event_map[] = { + [PERF_COUNT_HW_CPU_CYCLES] = { (26 << 6) }, + [PERF_COUNT_HW_INSTRUCTIONS] = { (3 << 6) | 0x3f }, + [PERF_COUNT_HW_CACHE_REFERENCES] = { (3 << 6) | 0x04 }, + [PERF_COUNT_HW_CACHE_MISSES] = { (16 << 6) | 0x07 }, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { (4 << 6) | 0x01 }, + [PERF_COUNT_HW_BRANCH_MISSES] = { (25 << 6) | 0x0f }, +}; + +static const struct perf_event_map *niagara4_event_map(int event_id) +{ + return &niagara4_perfmon_event_map[event_id]; +} + +static const cache_map_t niagara4_cache_map = { +[C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x04 }, + [C(RESULT_MISS)] = { (16 << 6) | 0x07 }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x08 }, + [C(RESULT_MISS)] = { (16 << 6) | 0x07 }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x3f }, + [C(RESULT_MISS)] = { (11 << 6) | 0x03 }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_NONSENSE }, + [ C(RESULT_MISS) ] = { CACHE_OP_NONSENSE }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x04 }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x08 }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(DTLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { (17 << 6) | 0x3f }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { (6 << 6) | 0x3f }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(NODE)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +}; + +static u32 sparc_vt_read_pmc(int idx) +{ + u64 val = pcr_ops->read_pic(idx); + + return val & 0xffffffff; +} + +static void sparc_vt_write_pmc(int idx, u64 val) +{ + u64 pcr; + + /* There seems to be an internal latch on the overflow event + * on SPARC-T4 that prevents it from triggering unless you + * update the PIC exactly as we do here. The requirement + * seems to be that you have to turn off event counting in the + * PCR around the PIC update. + * + * For example, after the following sequence: + * + * 1) set PIC to -1 + * 2) enable event counting and overflow reporting in PCR + * 3) overflow triggers, softint 15 handler invoked + * 4) clear OV bit in PCR + * 5) write PIC to -1 + * + * a subsequent overflow event will not trigger. This + * sequence works on SPARC-T3 and previous chips. + */ + pcr = pcr_ops->read_pcr(idx); + pcr_ops->write_pcr(idx, PCR_N4_PICNPT); + + pcr_ops->write_pic(idx, val & 0xffffffff); + + pcr_ops->write_pcr(idx, pcr); +} + +static const struct sparc_pmu niagara4_pmu = { + .event_map = niagara4_event_map, + .cache_map = &niagara4_cache_map, + .max_events = ARRAY_SIZE(niagara4_perfmon_event_map), + .read_pmc = sparc_vt_read_pmc, + .write_pmc = sparc_vt_write_pmc, + .upper_shift = 5, + .lower_shift = 5, + .event_mask = 0x7ff, + .user_bit = PCR_N4_UTRACE, + .priv_bit = PCR_N4_STRACE, + + /* We explicitly don't support hypervisor tracing. The T4 + * generates the overflow event for precise events via a trap + * which will not be generated (ie. it's completely lost) if + * we happen to be in the hypervisor when the event triggers. + * Essentially, the overflow event reporting is completely + * unusable when you have hypervisor mode tracing enabled. + */ + .hv_bit = 0, + + .irq_bit = PCR_N4_TOE, + .upper_nop = 0, + .lower_nop = 0, + .flags = 0, + .max_hw_events = 4, + .num_pcrs = 4, + .num_pic_regs = 4, }; static const struct sparc_pmu *sparc_pmu __read_mostly; @@ -557,56 +817,38 @@ static u64 nop_for_index(int idx) static inline void sparc_pmu_enable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { - u64 val, mask = mask_for_index(idx); + u64 enc, val, mask = mask_for_index(idx); + int pcr_index = 0; + + if (sparc_pmu->num_pcrs > 1) + pcr_index = idx; - val = cpuc->pcr; + enc = perf_event_get_enc(cpuc->events[idx]); + + val = cpuc->pcr[pcr_index]; val &= ~mask; - val |= hwc->config; - cpuc->pcr = val; + val |= event_encoding(enc, idx); + cpuc->pcr[pcr_index] = val; - pcr_ops->write(cpuc->pcr); + pcr_ops->write_pcr(pcr_index, cpuc->pcr[pcr_index]); } static inline void sparc_pmu_disable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { u64 mask = mask_for_index(idx); u64 nop = nop_for_index(idx); + int pcr_index = 0; u64 val; - val = cpuc->pcr; + if (sparc_pmu->num_pcrs > 1) + pcr_index = idx; + + val = cpuc->pcr[pcr_index]; val &= ~mask; val |= nop; - cpuc->pcr = val; - - pcr_ops->write(cpuc->pcr); -} - -static u32 read_pmc(int idx) -{ - u64 val; - - read_pic(val); - if (idx == PIC_UPPER_INDEX) - val >>= 32; + cpuc->pcr[pcr_index] = val; - return val & 0xffffffff; -} - -static void write_pmc(int idx, u64 val) -{ - u64 shift, mask, pic; - - shift = 0; - if (idx == PIC_UPPER_INDEX) - shift = 32; - - mask = ((u64) 0xffffffff) << shift; - val <<= shift; - - read_pic(pic); - pic &= ~mask; - pic |= val; - write_pic(pic); + pcr_ops->write_pcr(pcr_index, cpuc->pcr[pcr_index]); } static u64 sparc_perf_event_update(struct perf_event *event, @@ -618,7 +860,7 @@ static u64 sparc_perf_event_update(struct perf_event *event, again: prev_raw_count = local64_read(&hwc->prev_count); - new_raw_count = read_pmc(idx); + new_raw_count = sparc_pmu->read_pmc(idx); if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, new_raw_count) != prev_raw_count) @@ -658,25 +900,17 @@ static int sparc_perf_event_set_period(struct perf_event *event, local64_set(&hwc->prev_count, (u64)-left); - write_pmc(idx, (u64)(-left) & 0xffffffff); + sparc_pmu->write_pmc(idx, (u64)(-left) & 0xffffffff); perf_event_update_userpage(event); return ret; } -/* 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) +static void read_in_all_counters(struct cpu_hw_events *cpuc) { int i; - if (!cpuc->n_added) - goto out; - - /* Read in the counters which are moving. */ for (i = 0; i < cpuc->n_events; i++) { struct perf_event *cp = cpuc->event[i]; @@ -687,6 +921,20 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) cpuc->current_idx[i] = PIC_NO_INDEX; } } +} + +/* On this PMU all PICs are programmed using a single PCR. Calculate + * the combined control register value. + * + * For such chips we require that all of the events have the same + * configuration, so just fetch the settings from the first entry. + */ +static void calculate_single_pcr(struct cpu_hw_events *cpuc) +{ + int i; + + if (!cpuc->n_added) + goto out; /* Assign to counters all unassigned events. */ for (i = 0; i < cpuc->n_events; i++) { @@ -702,20 +950,71 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) cpuc->current_idx[i] = idx; enc = perf_event_get_enc(cpuc->events[i]); - pcr &= ~mask_for_index(idx); + cpuc->pcr[0] &= ~mask_for_index(idx); if (hwc->state & PERF_HES_STOPPED) - pcr |= nop_for_index(idx); + cpuc->pcr[0] |= nop_for_index(idx); else - pcr |= event_encoding(enc, idx); + cpuc->pcr[0] |= event_encoding(enc, idx); } out: - return pcr; + cpuc->pcr[0] |= cpuc->event[0]->hw.config_base; +} + +/* On this PMU each PIC has it's own PCR control register. */ +static void calculate_multiple_pcrs(struct cpu_hw_events *cpuc) +{ + int i; + + if (!cpuc->n_added) + goto out; + + 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]); + cpuc->pcr[idx] &= ~mask_for_index(idx); + if (hwc->state & PERF_HES_STOPPED) + cpuc->pcr[idx] |= nop_for_index(idx); + else + cpuc->pcr[idx] |= event_encoding(enc, idx); + } +out: + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *cp = cpuc->event[i]; + int idx = cp->hw.idx; + + cpuc->pcr[idx] |= cp->hw.config_base; + } +} + +/* If performance event entries have been added, move existing events + * around (if necessary) and then assign new entries to counters. + */ +static void update_pcrs_for_enable(struct cpu_hw_events *cpuc) +{ + if (cpuc->n_added) + read_in_all_counters(cpuc); + + if (sparc_pmu->num_pcrs == 1) { + calculate_single_pcr(cpuc); + } else { + calculate_multiple_pcrs(cpuc); + } } static void sparc_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 pcr; + int i; if (cpuc->enabled) return; @@ -723,26 +1022,17 @@ static void sparc_pmu_enable(struct pmu *pmu) cpuc->enabled = 1; barrier(); - pcr = cpuc->pcr; - if (!cpuc->n_events) { - pcr = 0; - } else { - pcr = maybe_change_configuration(cpuc, pcr); - - /* 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; - } + if (cpuc->n_events) + update_pcrs_for_enable(cpuc); - pcr_ops->write(cpuc->pcr); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + pcr_ops->write_pcr(i, cpuc->pcr[i]); } static void sparc_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; + int i; if (!cpuc->enabled) return; @@ -750,12 +1040,14 @@ static void sparc_pmu_disable(struct pmu *pmu) 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; + for (i = 0; i < sparc_pmu->num_pcrs; i++) { + u64 val = cpuc->pcr[i]; - pcr_ops->write(cpuc->pcr); + val &= ~(sparc_pmu->user_bit | sparc_pmu->priv_bit | + sparc_pmu->hv_bit | sparc_pmu->irq_bit); + cpuc->pcr[i] = val; + pcr_ops->write_pcr(i, cpuc->pcr[i]); + } } static int active_event_index(struct cpu_hw_events *cpuc, @@ -854,9 +1146,11 @@ static DEFINE_MUTEX(pmc_grab_mutex); static void perf_stop_nmi_watchdog(void *unused) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int i; stop_nmi_watchdog(NULL); - cpuc->pcr = pcr_ops->read(); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + cpuc->pcr[i] = pcr_ops->read_pcr(i); } void perf_event_grab_pmc(void) @@ -942,9 +1236,17 @@ static int sparc_check_constraints(struct perf_event **evts, if (!n_ev) return 0; - if (n_ev > MAX_HWEVENTS) + if (n_ev > sparc_pmu->max_hw_events) return -1; + if (!(sparc_pmu->flags & SPARC_PMU_HAS_CONFLICTS)) { + int i; + + for (i = 0; i < n_ev; i++) + evts[i]->hw.idx = i; + return 0; + } + msk0 = perf_event_get_msk(events[0]); if (n_ev == 1) { if (msk0 & PIC_LOWER) @@ -1000,6 +1302,9 @@ static int check_excludes(struct perf_event **evts, int n_prev, int n_new) struct perf_event *event; int i, n, first; + if (!(sparc_pmu->flags & SPARC_PMU_ALL_EXCLUDES_SAME)) + return 0; + n = n_prev + n_new; if (n <= 1) return 0; @@ -1059,7 +1364,7 @@ static int sparc_pmu_add(struct perf_event *event, int ef_flags) perf_pmu_disable(event->pmu); n0 = cpuc->n_events; - if (n0 >= MAX_HWEVENTS) + if (n0 >= sparc_pmu->max_hw_events) goto out; cpuc->event[n0] = event; @@ -1146,16 +1451,16 @@ static int sparc_pmu_event_init(struct perf_event *event) /* 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; + hwc->config_base |= sparc_pmu->user_bit; if (!attr->exclude_kernel) - hwc->config_base |= PCR_STRACE; + hwc->config_base |= sparc_pmu->priv_bit; if (!attr->exclude_hv) hwc->config_base |= sparc_pmu->hv_bit; n = 0; if (event->group_leader != event) { n = collect_events(event->group_leader, - MAX_HWEVENTS - 1, + sparc_pmu->max_hw_events - 1, evts, events, current_idx_dmy); if (n < 0) return -EINVAL; @@ -1254,8 +1559,7 @@ static struct pmu pmu = { void perf_event_print_debug(void) { unsigned long flags; - u64 pcr, pic; - int cpu; + int cpu, i; if (!sparc_pmu) return; @@ -1264,12 +1568,13 @@ void perf_event_print_debug(void) cpu = smp_processor_id(); - pcr = pcr_ops->read(); - read_pic(pic); - pr_info("\n"); - pr_info("CPU#%d: PCR[%016llx] PIC[%016llx]\n", - cpu, pcr, pic); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + pr_info("CPU#%d: PCR%d[%016llx]\n", + cpu, i, pcr_ops->read_pcr(i)); + for (i = 0; i < sparc_pmu->num_pic_regs; i++) + pr_info("CPU#%d: PIC%d[%016llx]\n", + cpu, i, pcr_ops->read_pic(i)); local_irq_restore(flags); } @@ -1305,8 +1610,9 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, * 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); + if (sparc_pmu->irq_bit && + sparc_pmu->num_pcrs == 1) + pcr_ops->write_pcr(0, cpuc->pcr[0]); for (i = 0; i < cpuc->n_events; i++) { struct perf_event *event = cpuc->event[i]; @@ -1314,6 +1620,10 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, struct hw_perf_event *hwc; u64 val; + if (sparc_pmu->irq_bit && + sparc_pmu->num_pcrs > 1) + pcr_ops->write_pcr(idx, cpuc->pcr[idx]); + hwc = &event->hw; val = sparc_perf_event_update(event, hwc, idx); if (val & (1ULL << 31)) @@ -1352,6 +1662,10 @@ static bool __init supported_pmu(void) sparc_pmu = &niagara2_pmu; return true; } + if (!strcmp(sparc_pmu_type, "niagara4")) { + sparc_pmu = &niagara4_pmu; + return true; + } return false; } @@ -1426,8 +1740,6 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_callchain_store(entry, regs->tpc); - ufp = regs->u_regs[UREG_I6] + STACK_BIAS; do { struct sparc_stackf *usf, sf; @@ -1448,19 +1760,27 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_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; + if (thread32_stack_is_64bit(ufp)) { + struct sparc_stackf *usf, sf; - pc = sf.callers_pc; - ufp = (unsigned long)sf.fp; + ufp += STACK_BIAS; + usf = (struct sparc_stackf *) ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + pc = sf.callers_pc & 0xffffffff; + ufp = ((unsigned long) sf.fp) & 0xffffffff; + } else { + struct sparc_stackf32 *usf, sf; + usf = (struct sparc_stackf32 *) ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + pc = sf.callers_pc; + ufp = (unsigned long)sf.fp; + } perf_callchain_store(entry, pc); } while (entry->nr < PERF_MAX_STACK_DEPTH); } @@ -1468,6 +1788,11 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, void perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { + perf_callchain_store(entry, regs->tpc); + + if (!current->mm) + return; + flushw_user(); if (test_thread_flag(TIF_32BIT)) perf_callchain_user_32(entry, regs); diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c index 14006d8aca2..487bffb36f5 100644 --- a/arch/sparc/kernel/process_32.c +++ b/arch/sparc/kernel/process_32.c @@ -482,7 +482,7 @@ int dump_fpu (struct pt_regs * regs, elf_fpregset_t * fpregs) asmlinkage int sparc_execve(struct pt_regs *regs) { int error, base = 0; - char *filename; + struct filename *filename; /* Check for indirect call. */ if(regs->u_regs[UREG_G1] == 0) @@ -492,7 +492,7 @@ asmlinkage int sparc_execve(struct pt_regs *regs) error = PTR_ERR(filename); if(IS_ERR(filename)) goto out; - error = do_execve(filename, + error = do_execve(filename->name, (const char __user *const __user *) regs->u_regs[base + UREG_I1], (const char __user *const __user *) diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index aff0c72fac0..c6e0c291004 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -27,6 +27,7 @@ #include <linux/tick.h> #include <linux/init.h> #include <linux/cpu.h> +#include <linux/perf_event.h> #include <linux/elfcore.h> #include <linux/sysrq.h> #include <linux/nmi.h> @@ -47,6 +48,7 @@ #include <asm/syscalls.h> #include <asm/irq_regs.h> #include <asm/smp.h> +#include <asm/pcr.h> #include "kstack.h" @@ -204,18 +206,22 @@ void show_regs(struct pt_regs *regs) show_stack(current, (unsigned long *) regs->u_regs[UREG_FP]); } -struct global_reg_snapshot global_reg_snapshot[NR_CPUS]; -static DEFINE_SPINLOCK(global_reg_snapshot_lock); +union global_cpu_snapshot global_cpu_snapshot[NR_CPUS]; +static DEFINE_SPINLOCK(global_cpu_snapshot_lock); static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs, int this_cpu) { + struct global_reg_snapshot *rp; + flushw_all(); - global_reg_snapshot[this_cpu].tstate = regs->tstate; - global_reg_snapshot[this_cpu].tpc = regs->tpc; - global_reg_snapshot[this_cpu].tnpc = regs->tnpc; - global_reg_snapshot[this_cpu].o7 = regs->u_regs[UREG_I7]; + rp = &global_cpu_snapshot[this_cpu].reg; + + rp->tstate = regs->tstate; + rp->tpc = regs->tpc; + rp->tnpc = regs->tnpc; + rp->o7 = regs->u_regs[UREG_I7]; if (regs->tstate & TSTATE_PRIV) { struct reg_window *rw; @@ -223,17 +229,17 @@ static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs, rw = (struct reg_window *) (regs->u_regs[UREG_FP] + STACK_BIAS); if (kstack_valid(tp, (unsigned long) rw)) { - global_reg_snapshot[this_cpu].i7 = rw->ins[7]; + rp->i7 = rw->ins[7]; rw = (struct reg_window *) (rw->ins[6] + STACK_BIAS); if (kstack_valid(tp, (unsigned long) rw)) - global_reg_snapshot[this_cpu].rpc = rw->ins[7]; + rp->rpc = rw->ins[7]; } } else { - global_reg_snapshot[this_cpu].i7 = 0; - global_reg_snapshot[this_cpu].rpc = 0; + rp->i7 = 0; + rp->rpc = 0; } - global_reg_snapshot[this_cpu].thread = tp; + rp->thread = tp; } /* In order to avoid hangs we do not try to synchronize with the @@ -261,9 +267,9 @@ void arch_trigger_all_cpu_backtrace(void) if (!regs) regs = tp->kregs; - spin_lock_irqsave(&global_reg_snapshot_lock, flags); + spin_lock_irqsave(&global_cpu_snapshot_lock, flags); - memset(global_reg_snapshot, 0, sizeof(global_reg_snapshot)); + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); this_cpu = raw_smp_processor_id(); @@ -272,7 +278,7 @@ void arch_trigger_all_cpu_backtrace(void) smp_fetch_global_regs(); for_each_online_cpu(cpu) { - struct global_reg_snapshot *gp = &global_reg_snapshot[cpu]; + struct global_reg_snapshot *gp = &global_cpu_snapshot[cpu].reg; __global_reg_poll(gp); @@ -295,9 +301,9 @@ void arch_trigger_all_cpu_backtrace(void) } } - memset(global_reg_snapshot, 0, sizeof(global_reg_snapshot)); + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); - spin_unlock_irqrestore(&global_reg_snapshot_lock, flags); + spin_unlock_irqrestore(&global_cpu_snapshot_lock, flags); } #ifdef CONFIG_MAGIC_SYSRQ @@ -309,16 +315,90 @@ static void sysrq_handle_globreg(int key) static struct sysrq_key_op sparc_globalreg_op = { .handler = sysrq_handle_globreg, - .help_msg = "Globalregs", + .help_msg = "global-regs(Y)", .action_msg = "Show Global CPU Regs", }; -static int __init sparc_globreg_init(void) +static void __global_pmu_self(int this_cpu) { - return register_sysrq_key('y', &sparc_globalreg_op); + struct global_pmu_snapshot *pp; + int i, num; + + pp = &global_cpu_snapshot[this_cpu].pmu; + + num = 1; + if (tlb_type == hypervisor && + sun4v_chip_type >= SUN4V_CHIP_NIAGARA4) + num = 4; + + for (i = 0; i < num; i++) { + pp->pcr[i] = pcr_ops->read_pcr(i); + pp->pic[i] = pcr_ops->read_pic(i); + } } -core_initcall(sparc_globreg_init); +static void __global_pmu_poll(struct global_pmu_snapshot *pp) +{ + int limit = 0; + + while (!pp->pcr[0] && ++limit < 100) { + barrier(); + udelay(1); + } +} + +static void pmu_snapshot_all_cpus(void) +{ + unsigned long flags; + int this_cpu, cpu; + + spin_lock_irqsave(&global_cpu_snapshot_lock, flags); + + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); + + this_cpu = raw_smp_processor_id(); + + __global_pmu_self(this_cpu); + + smp_fetch_global_pmu(); + + for_each_online_cpu(cpu) { + struct global_pmu_snapshot *pp = &global_cpu_snapshot[cpu].pmu; + + __global_pmu_poll(pp); + + printk("%c CPU[%3d]: PCR[%08lx:%08lx:%08lx:%08lx] PIC[%08lx:%08lx:%08lx:%08lx]\n", + (cpu == this_cpu ? '*' : ' '), cpu, + pp->pcr[0], pp->pcr[1], pp->pcr[2], pp->pcr[3], + pp->pic[0], pp->pic[1], pp->pic[2], pp->pic[3]); + } + + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); + + spin_unlock_irqrestore(&global_cpu_snapshot_lock, flags); +} + +static void sysrq_handle_globpmu(int key) +{ + pmu_snapshot_all_cpus(); +} + +static struct sysrq_key_op sparc_globalpmu_op = { + .handler = sysrq_handle_globpmu, + .help_msg = "global-pmu(X)", + .action_msg = "Show Global PMU Regs", +}; + +static int __init sparc_sysrq_init(void) +{ + int ret = register_sysrq_key('y', &sparc_globalreg_op); + + if (!ret) + ret = register_sysrq_key('x', &sparc_globalpmu_op); + return ret; +} + +core_initcall(sparc_sysrq_init); #endif @@ -372,13 +452,16 @@ void flush_thread(void) /* It's a bit more tricky when 64-bit tasks are involved... */ static unsigned long clone_stackframe(unsigned long csp, unsigned long psp) { + bool stack_64bit = test_thread_64bit_stack(psp); unsigned long fp, distance, rval; - if (!(test_thread_flag(TIF_32BIT))) { + if (stack_64bit) { csp += STACK_BIAS; psp += STACK_BIAS; __get_user(fp, &(((struct reg_window __user *)psp)->ins[6])); fp += STACK_BIAS; + if (test_thread_flag(TIF_32BIT)) + fp &= 0xffffffff; } else __get_user(fp, &(((struct reg_window32 __user *)psp)->ins[6])); @@ -392,7 +475,7 @@ static unsigned long clone_stackframe(unsigned long csp, unsigned long psp) rval = (csp - distance); if (copy_in_user((void __user *) rval, (void __user *) psp, distance)) rval = 0; - else if (test_thread_flag(TIF_32BIT)) { + else if (!stack_64bit) { if (put_user(((u32)csp), &(((struct reg_window32 __user *)rval)->ins[6]))) rval = 0; @@ -427,18 +510,18 @@ void synchronize_user_stack(void) flush_user_windows(); if ((window = get_thread_wsaved()) != 0) { - int winsize = sizeof(struct reg_window); - int bias = 0; - - if (test_thread_flag(TIF_32BIT)) - winsize = sizeof(struct reg_window32); - else - bias = STACK_BIAS; - window -= 1; do { - unsigned long sp = (t->rwbuf_stkptrs[window] + bias); struct reg_window *rwin = &t->reg_window[window]; + int winsize = sizeof(struct reg_window); + unsigned long sp; + + sp = t->rwbuf_stkptrs[window]; + + if (test_thread_64bit_stack(sp)) + sp += STACK_BIAS; + else + winsize = sizeof(struct reg_window32); if (!copy_to_user((char __user *)sp, rwin, winsize)) { shift_window_buffer(window, get_thread_wsaved() - 1, t); @@ -464,13 +547,6 @@ void fault_in_user_windows(void) { struct thread_info *t = current_thread_info(); unsigned long window; - int winsize = sizeof(struct reg_window); - int bias = 0; - - if (test_thread_flag(TIF_32BIT)) - winsize = sizeof(struct reg_window32); - else - bias = STACK_BIAS; flush_user_windows(); window = get_thread_wsaved(); @@ -478,8 +554,16 @@ void fault_in_user_windows(void) if (likely(window != 0)) { window -= 1; do { - unsigned long sp = (t->rwbuf_stkptrs[window] + bias); struct reg_window *rwin = &t->reg_window[window]; + int winsize = sizeof(struct reg_window); + unsigned long sp; + + sp = t->rwbuf_stkptrs[window]; + + if (test_thread_64bit_stack(sp)) + sp += STACK_BIAS; + else + winsize = sizeof(struct reg_window32); if (unlikely(sp & 0x7UL)) stack_unaligned(sp); @@ -722,7 +806,7 @@ EXPORT_SYMBOL(dump_fpu); asmlinkage int sparc_execve(struct pt_regs *regs) { int error, base = 0; - char *filename; + struct filename *filename; /* User register window flush is done by entry.S */ @@ -734,7 +818,7 @@ asmlinkage int sparc_execve(struct pt_regs *regs) error = PTR_ERR(filename); if (IS_ERR(filename)) goto out; - error = do_execve(filename, + error = do_execve(filename->name, (const char __user *const __user *) regs->u_regs[base + UREG_I1], (const char __user *const __user *) diff --git a/arch/sparc/kernel/prom_64.c b/arch/sparc/kernel/prom_64.c index 340c5b976d2..d397d7fc5c2 100644 --- a/arch/sparc/kernel/prom_64.c +++ b/arch/sparc/kernel/prom_64.c @@ -37,7 +37,7 @@ void * __init prom_early_alloc(unsigned long size) void *ret; if (!paddr) { - prom_printf("prom_early_alloc(%lu) failed\n"); + prom_printf("prom_early_alloc(%lu) failed\n", size); prom_halt(); } diff --git a/arch/sparc/kernel/ptrace_64.c b/arch/sparc/kernel/ptrace_64.c index 484dabac704..7ff45e4ba68 100644 --- a/arch/sparc/kernel/ptrace_64.c +++ b/arch/sparc/kernel/ptrace_64.c @@ -151,7 +151,7 @@ static int regwindow64_get(struct task_struct *target, { unsigned long rw_addr = regs->u_regs[UREG_I6]; - if (test_tsk_thread_flag(current, TIF_32BIT)) { + if (!test_thread_64bit_stack(rw_addr)) { struct reg_window32 win32; int i; @@ -176,7 +176,7 @@ static int regwindow64_set(struct task_struct *target, { unsigned long rw_addr = regs->u_regs[UREG_I6]; - if (test_tsk_thread_flag(current, TIF_32BIT)) { + if (!test_thread_64bit_stack(rw_addr)) { struct reg_window32 win32; int i; diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index 1414d16712b..0eaf0059aae 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -316,6 +316,25 @@ static void __init popc_patch(void) } } +static void __init pause_patch(void) +{ + struct pause_patch_entry *p; + + p = &__pause_3insn_patch; + while (p < &__pause_3insn_patch_end) { + unsigned long i, addr = p->addr; + + for (i = 0; i < 3; i++) { + *(unsigned int *) (addr + (i * 4)) = p->insns[i]; + wmb(); + __asm__ __volatile__("flush %0" + : : "r" (addr + (i * 4))); + } + + p++; + } +} + #ifdef CONFIG_SMP void __init boot_cpu_id_too_large(int cpu) { @@ -340,7 +359,12 @@ static const char *hwcaps[] = { */ "mul32", "div32", "fsmuld", "v8plus", "popc", "vis", "vis2", "ASIBlkInit", "fmaf", "vis3", "hpc", "random", "trans", "fjfmau", - "ima", "cspare", + "ima", "cspare", "pause", "cbcond", +}; + +static const char *crypto_hwcaps[] = { + "aes", "des", "kasumi", "camellia", "md5", "sha1", "sha256", + "sha512", "mpmul", "montmul", "montsqr", "crc32c", }; void cpucap_info(struct seq_file *m) @@ -357,27 +381,61 @@ void cpucap_info(struct seq_file *m) printed++; } } + if (caps & HWCAP_SPARC_CRYPTO) { + unsigned long cfr; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + unsigned long bit = 1UL << i; + if (cfr & bit) { + seq_printf(m, "%s%s", + printed ? "," : "", crypto_hwcaps[i]); + printed++; + } + } + } seq_putc(m, '\n'); } +static void __init report_one_hwcap(int *printed, const char *name) +{ + if ((*printed) == 0) + printk(KERN_INFO "CPU CAPS: ["); + printk(KERN_CONT "%s%s", + (*printed) ? "," : "", name); + if (++(*printed) == 8) { + printk(KERN_CONT "]\n"); + *printed = 0; + } +} + +static void __init report_crypto_hwcaps(int *printed) +{ + unsigned long cfr; + int i; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + unsigned long bit = 1UL << i; + if (cfr & bit) + report_one_hwcap(printed, crypto_hwcaps[i]); + } +} + static void __init report_hwcaps(unsigned long caps) { int i, printed = 0; - printk(KERN_INFO "CPU CAPS: ["); for (i = 0; i < ARRAY_SIZE(hwcaps); i++) { unsigned long bit = 1UL << i; - if (caps & bit) { - printk(KERN_CONT "%s%s", - printed ? "," : "", hwcaps[i]); - if (++printed == 8) { - printk(KERN_CONT "]\n"); - printk(KERN_INFO "CPU CAPS: ["); - printed = 0; - } - } + if (caps & bit) + report_one_hwcap(&printed, hwcaps[i]); } - printk(KERN_CONT "]\n"); + if (caps & HWCAP_SPARC_CRYPTO) + report_crypto_hwcaps(&printed); + if (printed != 0) + printk(KERN_CONT "]\n"); } static unsigned long __init mdesc_cpu_hwcap_list(void) @@ -411,6 +469,10 @@ static unsigned long __init mdesc_cpu_hwcap_list(void) break; } } + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + if (!strcmp(prop, crypto_hwcaps[i])) + caps |= HWCAP_SPARC_CRYPTO; + } plen = strlen(prop) + 1; prop += plen; @@ -485,6 +547,8 @@ static void __init init_sparc64_elf_hwcap(void) if (sparc64_elf_hwcap & AV_SPARC_POPC) popc_patch(); + if (sparc64_elf_hwcap & AV_SPARC_PAUSE) + pause_patch(); } void __init setup_arch(char **cmdline_p) diff --git a/arch/sparc/kernel/signal32.c b/arch/sparc/kernel/signal32.c index a53e0a5fd3a..53e48f721ce 100644 --- a/arch/sparc/kernel/signal32.c +++ b/arch/sparc/kernel/signal32.c @@ -54,58 +54,6 @@ struct signal_frame32 { /* __siginfo_rwin_t * */u32 rwin_save; } __attribute__((aligned(8))); -typedef struct compat_siginfo{ - int si_signo; - int si_errno; - int si_code; - - union { - int _pad[SI_PAD_SIZE32]; - - /* kill() */ - struct { - compat_pid_t _pid; /* sender's pid */ - unsigned int _uid; /* sender's uid */ - } _kill; - - /* POSIX.1b timers */ - struct { - compat_timer_t _tid; /* timer id */ - int _overrun; /* overrun count */ - compat_sigval_t _sigval; /* same as below */ - int _sys_private; /* not to be passed to user */ - } _timer; - - /* POSIX.1b signals */ - struct { - compat_pid_t _pid; /* sender's pid */ - unsigned int _uid; /* sender's uid */ - compat_sigval_t _sigval; - } _rt; - - /* SIGCHLD */ - struct { - compat_pid_t _pid; /* which child */ - unsigned int _uid; /* sender's uid */ - int _status; /* exit code */ - compat_clock_t _utime; - compat_clock_t _stime; - } _sigchld; - - /* SIGILL, SIGFPE, SIGSEGV, SIGBUS, SIGEMT */ - struct { - u32 _addr; /* faulting insn/memory ref. */ - int _trapno; - } _sigfault; - - /* SIGPOLL */ - struct { - int _band; /* POLL_IN, POLL_OUT, POLL_MSG */ - int _fd; - } _sigpoll; - } _sifields; -}compat_siginfo_t; - struct rt_signal_frame32 { struct sparc_stackf32 ss; compat_siginfo_t info; diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index 781bcb10b8b..d94b878577b 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -852,6 +852,8 @@ extern unsigned long xcall_flush_tlb_mm; extern unsigned long xcall_flush_tlb_pending; extern unsigned long xcall_flush_tlb_kernel_range; extern unsigned long xcall_fetch_glob_regs; +extern unsigned long xcall_fetch_glob_pmu; +extern unsigned long xcall_fetch_glob_pmu_n4; extern unsigned long xcall_receive_signal; extern unsigned long xcall_new_mmu_context_version; #ifdef CONFIG_KGDB @@ -1000,6 +1002,15 @@ void smp_fetch_global_regs(void) smp_cross_call(&xcall_fetch_glob_regs, 0, 0, 0); } +void smp_fetch_global_pmu(void) +{ + if (tlb_type == hypervisor && + sun4v_chip_type >= SUN4V_CHIP_NIAGARA4) + smp_cross_call(&xcall_fetch_glob_pmu_n4, 0, 0, 0); + else + smp_cross_call(&xcall_fetch_glob_pmu, 0, 0, 0); +} + /* We know that the window frames of the user have been flushed * to the stack before we get here because all callers of us * are flush_tlb_*() routines, and these run after flush_cache_*() diff --git a/arch/sparc/kernel/sun4v_tlb_miss.S b/arch/sparc/kernel/sun4v_tlb_miss.S index e1fbf8c7578..bde867fd71e 100644 --- a/arch/sparc/kernel/sun4v_tlb_miss.S +++ b/arch/sparc/kernel/sun4v_tlb_miss.S @@ -176,7 +176,7 @@ sun4v_tsb_miss_common: sub %g2, TRAP_PER_CPU_FAULT_INFO, %g2 -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) mov SCRATCHPAD_UTSBREG2, %g5 ldxa [%g5] ASI_SCRATCHPAD, %g5 cmp %g5, -1 diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index d97f3eb72e0..44025f4ba41 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -90,7 +90,7 @@ SIGN1(sys32_mkdir, sys_mkdir, %o1) SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5) SIGN1(sys32_sysfs, compat_sys_sysfs, %o0) SIGN2(sys32_sendfile, compat_sys_sendfile, %o0, %o1) -SIGN2(sys32_sendfile64, compat_sys_sendfile64, %o0, %o1) +SIGN2(sys32_sendfile64, sys_sendfile, %o0, %o1) SIGN1(sys32_prctl, sys_prctl, %o0) SIGN1(sys32_sched_rr_get_interval, compat_sys_sched_rr_get_interval, %o0) SIGN2(sys32_waitpid, sys_waitpid, %o0, %o2) diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c index f7392336961..c3239811a1b 100644 --- a/arch/sparc/kernel/sys_sparc32.c +++ b/arch/sparc/kernel/sys_sparc32.c @@ -403,7 +403,7 @@ asmlinkage long compat_sys_rt_sigaction(int sig, asmlinkage long sparc32_execve(struct pt_regs *regs) { int error, base = 0; - char *filename; + struct filename *filename; /* User register window flush is done by entry.S */ @@ -416,7 +416,7 @@ asmlinkage long sparc32_execve(struct pt_regs *regs) if (IS_ERR(filename)) goto out; - error = compat_do_execve(filename, + error = compat_do_execve(filename->name, compat_ptr(regs->u_regs[base + UREG_I1]), compat_ptr(regs->u_regs[base + UREG_I2]), regs); @@ -506,52 +506,6 @@ long compat_sys_fadvise64_64(int fd, advice); } -asmlinkage long compat_sys_sendfile(int out_fd, int in_fd, - compat_off_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - off_t of; - - if (offset && get_user(of, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile(out_fd, in_fd, - offset ? (off_t __user *) &of : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(of, offset)) - return -EFAULT; - - return ret; -} - -asmlinkage long compat_sys_sendfile64(int out_fd, int in_fd, - compat_loff_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - loff_t lof; - - if (offset && get_user(lof, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile64(out_fd, in_fd, - offset ? (loff_t __user *) &lof : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(lof, offset)) - return -EFAULT; - - return ret; -} - /* This is just a version for 32-bit applications which does * not force O_LARGEFILE on. */ diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index 11c6c9603e7..878ef3d5fec 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -751,3 +751,8 @@ int kernel_execve(const char *filename, : "cc"); return __res; } + +asmlinkage long sys_kern_features(void) +{ + return KERN_FEATURE_MIXED_MODE_STACK; +} diff --git a/arch/sparc/kernel/syscalls.S b/arch/sparc/kernel/syscalls.S index 1d7e274f3f2..7f5f65d0b3f 100644 --- a/arch/sparc/kernel/syscalls.S +++ b/arch/sparc/kernel/syscalls.S @@ -212,24 +212,20 @@ linux_sparc_syscall: 3: stx %o0, [%sp + PTREGS_OFF + PT_V9_I0] ret_sys_call: ldx [%sp + PTREGS_OFF + PT_V9_TSTATE], %g3 - ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc sra %o0, 0, %o0 mov %ulo(TSTATE_XCARRY | TSTATE_ICARRY), %g2 sllx %g2, 32, %g2 - /* Check if force_successful_syscall_return() - * was invoked. - */ - ldub [%g6 + TI_SYS_NOERROR], %l2 - brnz,a,pn %l2, 80f - stb %g0, [%g6 + TI_SYS_NOERROR] - cmp %o0, -ERESTART_RESTARTBLOCK bgeu,pn %xcc, 1f - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %l6 -80: + andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0 + ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc + +2: + stb %g0, [%g6 + TI_SYS_NOERROR] /* System call success, clear Carry condition code. */ andn %g3, %g2, %g3 +3: stx %g3, [%sp + PTREGS_OFF + PT_V9_TSTATE] bne,pn %icc, linux_syscall_trace2 add %l1, 0x4, %l2 ! npc = npc+4 @@ -238,20 +234,20 @@ ret_sys_call: stx %l2, [%sp + PTREGS_OFF + PT_V9_TNPC] 1: + /* Check if force_successful_syscall_return() + * was invoked. + */ + ldub [%g6 + TI_SYS_NOERROR], %l2 + brnz,pn %l2, 2b + ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc /* System call failure, set Carry condition code. * Also, get abs(errno) to return to the process. */ - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %l6 sub %g0, %o0, %o0 - or %g3, %g2, %g3 stx %o0, [%sp + PTREGS_OFF + PT_V9_I0] - stx %g3, [%sp + PTREGS_OFF + PT_V9_TSTATE] - bne,pn %icc, linux_syscall_trace2 - add %l1, 0x4, %l2 ! npc = npc+4 - stx %l1, [%sp + PTREGS_OFF + PT_V9_TPC] + ba,pt %xcc, 3b + or %g3, %g2, %g3 - b,pt %xcc, rtrap - stx %l2, [%sp + PTREGS_OFF + PT_V9_TNPC] linux_syscall_trace2: call syscall_trace_leave add %sp, PTREGS_OFF, %o0 diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S index 63402f9e9f5..5147f574f12 100644 --- a/arch/sparc/kernel/systbls_32.S +++ b/arch/sparc/kernel/systbls_32.S @@ -85,3 +85,4 @@ sys_call_table: /*325*/ .long sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init /*330*/ .long sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime /*335*/ .long sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev +/*340*/ .long sys_ni_syscall, sys_kcmp diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S index 3a58e0d66f5..1c9af9fa38e 100644 --- a/arch/sparc/kernel/systbls_64.S +++ b/arch/sparc/kernel/systbls_64.S @@ -86,6 +86,7 @@ sys_call_table32: .word compat_sys_pwritev, compat_sys_rt_tgsigqueueinfo, sys_perf_event_open, compat_sys_recvmmsg, sys_fanotify_init /*330*/ .word sys32_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime .word sys_syncfs, compat_sys_sendmmsg, sys_setns, compat_sys_process_vm_readv, compat_sys_process_vm_writev +/*340*/ .word sys_kern_features, sys_kcmp #endif /* CONFIG_COMPAT */ @@ -163,3 +164,4 @@ sys_call_table: .word sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init /*330*/ .word sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime .word sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev +/*340*/ .word sys_kern_features, sys_kcmp diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 3b05e669771..b66a77968f3 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -1,6 +1,6 @@ /* arch/sparc64/kernel/traps.c * - * Copyright (C) 1995,1997,2008,2009 David S. Miller (davem@davemloft.net) + * Copyright (C) 1995,1997,2008,2009,2012 David S. Miller (davem@davemloft.net) * Copyright (C) 1997,1999,2000 Jakub Jelinek (jakub@redhat.com) */ @@ -18,6 +18,7 @@ #include <linux/init.h> #include <linux/kdebug.h> #include <linux/ftrace.h> +#include <linux/reboot.h> #include <linux/gfp.h> #include <asm/smp.h> @@ -850,7 +851,7 @@ void __init cheetah_ecache_flush_init(void) ecache_flush_physbase = find_ecache_flush_span(ecache_flush_size); if (ecache_flush_physbase == ~0UL) { - prom_printf("cheetah_ecache_flush_init: Cannot find %d byte " + prom_printf("cheetah_ecache_flush_init: Cannot find %ld byte " "contiguous physical memory.\n", ecache_flush_size); prom_halt(); @@ -1760,85 +1761,223 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs) } struct sun4v_error_entry { - u64 err_handle; - u64 err_stick; + /* Unique error handle */ +/*0x00*/u64 err_handle; - u32 err_type; + /* %stick value at the time of the error */ +/*0x08*/u64 err_stick; + +/*0x10*/u8 reserved_1[3]; + + /* Error type */ +/*0x13*/u8 err_type; #define SUN4V_ERR_TYPE_UNDEFINED 0 #define SUN4V_ERR_TYPE_UNCORRECTED_RES 1 #define SUN4V_ERR_TYPE_PRECISE_NONRES 2 #define SUN4V_ERR_TYPE_DEFERRED_NONRES 3 -#define SUN4V_ERR_TYPE_WARNING_RES 4 +#define SUN4V_ERR_TYPE_SHUTDOWN_RQST 4 +#define SUN4V_ERR_TYPE_DUMP_CORE 5 +#define SUN4V_ERR_TYPE_SP_STATE_CHANGE 6 +#define SUN4V_ERR_TYPE_NUM 7 - u32 err_attrs; + /* Error attributes */ +/*0x14*/u32 err_attrs; #define SUN4V_ERR_ATTRS_PROCESSOR 0x00000001 #define SUN4V_ERR_ATTRS_MEMORY 0x00000002 #define SUN4V_ERR_ATTRS_PIO 0x00000004 #define SUN4V_ERR_ATTRS_INT_REGISTERS 0x00000008 #define SUN4V_ERR_ATTRS_FPU_REGISTERS 0x00000010 -#define SUN4V_ERR_ATTRS_USER_MODE 0x01000000 -#define SUN4V_ERR_ATTRS_PRIV_MODE 0x02000000 +#define SUN4V_ERR_ATTRS_SHUTDOWN_RQST 0x00000020 +#define SUN4V_ERR_ATTRS_ASR 0x00000040 +#define SUN4V_ERR_ATTRS_ASI 0x00000080 +#define SUN4V_ERR_ATTRS_PRIV_REG 0x00000100 +#define SUN4V_ERR_ATTRS_SPSTATE_MSK 0x00000600 +#define SUN4V_ERR_ATTRS_SPSTATE_SHFT 9 +#define SUN4V_ERR_ATTRS_MODE_MSK 0x03000000 +#define SUN4V_ERR_ATTRS_MODE_SHFT 24 #define SUN4V_ERR_ATTRS_RES_QUEUE_FULL 0x80000000 - u64 err_raddr; - u32 err_size; - u16 err_cpu; - u16 err_pad; +#define SUN4V_ERR_SPSTATE_FAULTED 0 +#define SUN4V_ERR_SPSTATE_AVAILABLE 1 +#define SUN4V_ERR_SPSTATE_NOT_PRESENT 2 + +#define SUN4V_ERR_MODE_USER 1 +#define SUN4V_ERR_MODE_PRIV 2 + + /* Real address of the memory region or PIO transaction */ +/*0x18*/u64 err_raddr; + + /* Size of the operation triggering the error, in bytes */ +/*0x20*/u32 err_size; + + /* ID of the CPU */ +/*0x24*/u16 err_cpu; + + /* Grace periof for shutdown, in seconds */ +/*0x26*/u16 err_secs; + + /* Value of the %asi register */ +/*0x28*/u8 err_asi; + +/*0x29*/u8 reserved_2; + + /* Value of the ASR register number */ +/*0x2a*/u16 err_asr; +#define SUN4V_ERR_ASR_VALID 0x8000 + +/*0x2c*/u32 reserved_3; +/*0x30*/u64 reserved_4; +/*0x38*/u64 reserved_5; }; static atomic_t sun4v_resum_oflow_cnt = ATOMIC_INIT(0); static atomic_t sun4v_nonresum_oflow_cnt = ATOMIC_INIT(0); -static const char *sun4v_err_type_to_str(u32 type) -{ - switch (type) { - case SUN4V_ERR_TYPE_UNDEFINED: - return "undefined"; - case SUN4V_ERR_TYPE_UNCORRECTED_RES: - return "uncorrected resumable"; - case SUN4V_ERR_TYPE_PRECISE_NONRES: - return "precise nonresumable"; - case SUN4V_ERR_TYPE_DEFERRED_NONRES: - return "deferred nonresumable"; - case SUN4V_ERR_TYPE_WARNING_RES: - return "warning resumable"; - default: - return "unknown"; +static const char *sun4v_err_type_to_str(u8 type) +{ + static const char *types[SUN4V_ERR_TYPE_NUM] = { + "undefined", + "uncorrected resumable", + "precise nonresumable", + "deferred nonresumable", + "shutdown request", + "dump core", + "SP state change", + }; + + if (type < SUN4V_ERR_TYPE_NUM) + return types[type]; + + return "unknown"; +} + +static void sun4v_emit_err_attr_strings(u32 attrs) +{ + static const char *attr_names[] = { + "processor", + "memory", + "PIO", + "int-registers", + "fpu-registers", + "shutdown-request", + "ASR", + "ASI", + "priv-reg", + }; + static const char *sp_states[] = { + "sp-faulted", + "sp-available", + "sp-not-present", + "sp-state-reserved", + }; + static const char *modes[] = { + "mode-reserved0", + "user", + "priv", + "mode-reserved1", + }; + u32 sp_state, mode; + int i; + + for (i = 0; i < ARRAY_SIZE(attr_names); i++) { + if (attrs & (1U << i)) { + const char *s = attr_names[i]; + + pr_cont("%s ", s); + } } + + sp_state = ((attrs & SUN4V_ERR_ATTRS_SPSTATE_MSK) >> + SUN4V_ERR_ATTRS_SPSTATE_SHFT); + pr_cont("%s ", sp_states[sp_state]); + + mode = ((attrs & SUN4V_ERR_ATTRS_MODE_MSK) >> + SUN4V_ERR_ATTRS_MODE_SHFT); + pr_cont("%s ", modes[mode]); + + if (attrs & SUN4V_ERR_ATTRS_RES_QUEUE_FULL) + pr_cont("res-queue-full "); } -static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, int cpu, const char *pfx, atomic_t *ocnt) +/* When the report contains a real-address of "-1" it means that the + * hardware did not provide the address. So we compute the effective + * address of the load or store instruction at regs->tpc and report + * that. Usually when this happens it's a PIO and in such a case we + * are using physical addresses with bypass ASIs anyways, so what we + * report here is exactly what we want. + */ +static void sun4v_report_real_raddr(const char *pfx, struct pt_regs *regs) { + unsigned int insn; + u64 addr; + + if (!(regs->tstate & TSTATE_PRIV)) + return; + + insn = *(unsigned int *) regs->tpc; + + addr = compute_effective_address(regs, insn, 0); + + printk("%s: insn effective address [0x%016llx]\n", + pfx, addr); +} + +static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, + int cpu, const char *pfx, atomic_t *ocnt) +{ + u64 *raw_ptr = (u64 *) ent; + u32 attrs; int cnt; printk("%s: Reporting on cpu %d\n", pfx, cpu); - printk("%s: err_handle[%llx] err_stick[%llx] err_type[%08x:%s]\n", - pfx, - ent->err_handle, ent->err_stick, - ent->err_type, - sun4v_err_type_to_str(ent->err_type)); - printk("%s: err_attrs[%08x:%s %s %s %s %s %s %s %s]\n", - pfx, - ent->err_attrs, - ((ent->err_attrs & SUN4V_ERR_ATTRS_PROCESSOR) ? - "processor" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_MEMORY) ? - "memory" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_PIO) ? - "pio" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_INT_REGISTERS) ? - "integer-regs" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_FPU_REGISTERS) ? - "fpu-regs" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_USER_MODE) ? - "user" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_PRIV_MODE) ? - "privileged" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_RES_QUEUE_FULL) ? - "queue-full" : "")); - printk("%s: err_raddr[%016llx] err_size[%u] err_cpu[%u]\n", - pfx, - ent->err_raddr, ent->err_size, ent->err_cpu); + printk("%s: TPC [0x%016lx] <%pS>\n", + pfx, regs->tpc, (void *) regs->tpc); + + printk("%s: RAW [%016llx:%016llx:%016llx:%016llx\n", + pfx, raw_ptr[0], raw_ptr[1], raw_ptr[2], raw_ptr[3]); + printk("%s: %016llx:%016llx:%016llx:%016llx]\n", + pfx, raw_ptr[4], raw_ptr[5], raw_ptr[6], raw_ptr[7]); + + printk("%s: handle [0x%016llx] stick [0x%016llx]\n", + pfx, ent->err_handle, ent->err_stick); + + printk("%s: type [%s]\n", pfx, sun4v_err_type_to_str(ent->err_type)); + + attrs = ent->err_attrs; + printk("%s: attrs [0x%08x] < ", pfx, attrs); + sun4v_emit_err_attr_strings(attrs); + pr_cont(">\n"); + + /* Various fields in the error report are only valid if + * certain attribute bits are set. + */ + if (attrs & (SUN4V_ERR_ATTRS_MEMORY | + SUN4V_ERR_ATTRS_PIO | + SUN4V_ERR_ATTRS_ASI)) { + printk("%s: raddr [0x%016llx]\n", pfx, ent->err_raddr); + + if (ent->err_raddr == ~(u64)0) + sun4v_report_real_raddr(pfx, regs); + } + + if (attrs & (SUN4V_ERR_ATTRS_MEMORY | SUN4V_ERR_ATTRS_ASI)) + printk("%s: size [0x%x]\n", pfx, ent->err_size); + + if (attrs & (SUN4V_ERR_ATTRS_PROCESSOR | + SUN4V_ERR_ATTRS_INT_REGISTERS | + SUN4V_ERR_ATTRS_FPU_REGISTERS | + SUN4V_ERR_ATTRS_PRIV_REG)) + printk("%s: cpu[%u]\n", pfx, ent->err_cpu); + + if (attrs & SUN4V_ERR_ATTRS_ASI) + printk("%s: asi [0x%02x]\n", pfx, ent->err_asi); + + if ((attrs & (SUN4V_ERR_ATTRS_INT_REGISTERS | + SUN4V_ERR_ATTRS_FPU_REGISTERS | + SUN4V_ERR_ATTRS_PRIV_REG)) && + (ent->err_asr & SUN4V_ERR_ASR_VALID) != 0) + printk("%s: reg [0x%04x]\n", + pfx, ent->err_asr & ~SUN4V_ERR_ASR_VALID); show_regs(regs); @@ -1874,13 +2013,15 @@ void sun4v_resum_error(struct pt_regs *regs, unsigned long offset) put_cpu(); - if (ent->err_type == SUN4V_ERR_TYPE_WARNING_RES) { - /* If err_type is 0x4, it's a powerdown request. Do - * not do the usual resumable error log because that - * makes it look like some abnormal error. + if (local_copy.err_type == SUN4V_ERR_TYPE_SHUTDOWN_RQST) { + /* We should really take the seconds field of + * the error report and use it for the shutdown + * invocation, but for now do the same thing we + * do for a DS shutdown request. */ - printk(KERN_INFO "Power down request...\n"); - kill_cad_pid(SIGINT, 1); + pr_info("Shutdown request, %u seconds...\n", + local_copy.err_secs); + orderly_poweroff(true); return; } diff --git a/arch/sparc/kernel/tsb.S b/arch/sparc/kernel/tsb.S index db15d123f05..d4bdc7a6237 100644 --- a/arch/sparc/kernel/tsb.S +++ b/arch/sparc/kernel/tsb.S @@ -49,7 +49,7 @@ tsb_miss_page_table_walk: /* Before committing to a full page table walk, * check the huge page TSB. */ -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) 661: ldx [%g7 + TRAP_PER_CPU_TSB_HUGE], %g5 nop @@ -110,12 +110,9 @@ tsb_miss_page_table_walk: tsb_miss_page_table_walk_sun4v_fastpath: USER_PGTABLE_WALK_TL1(%g4, %g7, %g5, %g2, tsb_do_fault) - /* Load and check PTE. */ - ldxa [%g5] ASI_PHYS_USE_EC, %g5 - brgez,pn %g5, tsb_do_fault - nop + /* Valid PTE is now in %g5. */ -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) 661: sethi %uhi(_PAGE_SZALL_4U), %g7 sllx %g7, 32, %g7 .section .sun4v_2insn_patch, "ax" diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index f81d038f734..8201c25e766 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -113,21 +113,24 @@ static inline long sign_extend_imm13(long imm) static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) { - unsigned long value; + unsigned long value, fp; if (reg < 16) return (!reg ? 0 : regs->u_regs[reg]); + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); value = win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); get_user(value, &win32->locals[reg - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); get_user(value, &win->locals[reg - 16]); } return value; @@ -135,19 +138,24 @@ static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs) { + unsigned long fp; + if (reg < 16) return ®s->u_regs[reg]; + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); return &win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 *win32; - win32 = (struct reg_window32 *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 *)((unsigned long)((u32)fp)); return (unsigned long *)&win32->locals[reg - 16]; } else { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); return &win->locals[reg - 16]; } } @@ -392,13 +400,15 @@ int handle_popc(u32 insn, struct pt_regs *regs) if (rd) regs->u_regs[rd] = ret; } else { - if (test_thread_flag(TIF_32BIT)) { + unsigned long fp = regs->u_regs[UREG_FP]; + + if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); put_user(ret, &win32->locals[rd - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); put_user(ret, &win->locals[rd - 16]); } } @@ -554,7 +564,7 @@ void handle_ld_nf(u32 insn, struct pt_regs *regs) reg[0] = 0; if ((insn & 0x780000) == 0x180000) reg[1] = 0; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(regs->u_regs[UREG_FP])) { put_user(0, (int __user *) reg); if ((insn & 0x780000) == 0x180000) put_user(0, ((int __user *) reg) + 1); diff --git a/arch/sparc/kernel/visemul.c b/arch/sparc/kernel/visemul.c index 08e074b7eb6..c096c624ac4 100644 --- a/arch/sparc/kernel/visemul.c +++ b/arch/sparc/kernel/visemul.c @@ -149,21 +149,24 @@ static inline void maybe_flush_windows(unsigned int rs1, unsigned int rs2, static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) { - unsigned long value; + unsigned long value, fp; if (reg < 16) return (!reg ? 0 : regs->u_regs[reg]); + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); value = win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); get_user(value, &win32->locals[reg - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); get_user(value, &win->locals[reg - 16]); } return value; @@ -172,16 +175,18 @@ static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) static inline unsigned long __user *__fetch_reg_addr_user(unsigned int reg, struct pt_regs *regs) { + unsigned long fp = regs->u_regs[UREG_FP]; + BUG_ON(reg < 16); BUG_ON(regs->tstate & TSTATE_PRIV); - if (test_thread_flag(TIF_32BIT)) { + if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); return (unsigned long __user *)&win32->locals[reg - 16]; } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); return &win->locals[reg - 16]; } } @@ -204,7 +209,7 @@ static void store_reg(struct pt_regs *regs, unsigned long val, unsigned long rd) } else { unsigned long __user *rd_user = __fetch_reg_addr_user(rd, regs); - if (test_thread_flag(TIF_32BIT)) + if (!test_thread_64bit_stack(regs->u_regs[UREG_FP])) __put_user((u32)val, (u32 __user *)rd_user); else __put_user(val, rd_user); diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 89c2c29f154..0bacceb1915 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -132,6 +132,11 @@ SECTIONS *(.popc_6insn_patch) __popc_6insn_patch_end = .; } + .pause_3insn_patch : { + __pause_3insn_patch = .; + *(.pause_3insn_patch) + __pause_3insn_patch_end = .; + } PERCPU_SECTION(SMP_CACHE_BYTES) . = ALIGN(PAGE_SIZE); diff --git a/arch/sparc/kernel/winfixup.S b/arch/sparc/kernel/winfixup.S index a6b0863c27d..1e67ce95836 100644 --- a/arch/sparc/kernel/winfixup.S +++ b/arch/sparc/kernel/winfixup.S @@ -43,6 +43,8 @@ spill_fixup_mna: spill_fixup_dax: TRAP_LOAD_THREAD_REG(%g6, %g1) ldx [%g6 + TI_FLAGS], %g1 + andcc %sp, 0x1, %g0 + movne %icc, 0, %g1 andcc %g1, _TIF_32BIT, %g0 ldub [%g6 + TI_WSAVED], %g1 sll %g1, 3, %g3 |