diff options
Diffstat (limited to 'arch/powerpc/perf/core-book3s.c')
| -rw-r--r-- | arch/powerpc/perf/core-book3s.c | 550 |
1 files changed, 365 insertions, 185 deletions
diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c index c627843c5b2..a3985aee77f 100644 --- a/arch/powerpc/perf/core-book3s.c +++ b/arch/powerpc/perf/core-book3s.c @@ -13,11 +13,13 @@ #include <linux/perf_event.h> #include <linux/percpu.h> #include <linux/hardirq.h> +#include <linux/uaccess.h> #include <asm/reg.h> #include <asm/pmc.h> #include <asm/machdep.h> #include <asm/firmware.h> #include <asm/ptrace.h> +#include <asm/code-patching.h> #define BHRB_MAX_ENTRIES 32 #define BHRB_TARGET 0x0000000000000002 @@ -73,6 +75,11 @@ static unsigned int freeze_events_kernel = MMCR0_FCS; #define MMCR0_FCHV 0 #define MMCR0_PMCjCE MMCR0_PMCnCE +#define MMCR0_FC56 0 +#define MMCR0_PMAO 0 +#define MMCR0_EBE 0 +#define MMCR0_PMCC 0 +#define MMCR0_PMCC_U6 0 #define SPRN_MMCRA SPRN_MMCR2 #define MMCRA_SAMPLE_ENABLE 0 @@ -100,11 +107,24 @@ static inline int siar_valid(struct pt_regs *regs) return 1; } +static bool is_ebb_event(struct perf_event *event) { return false; } +static int ebb_event_check(struct perf_event *event) { return 0; } +static void ebb_event_add(struct perf_event *event) { } +static void ebb_switch_out(unsigned long mmcr0) { } +static unsigned long ebb_switch_in(bool ebb, unsigned long mmcr0) +{ + return mmcr0; +} + +static inline void power_pmu_bhrb_enable(struct perf_event *event) {} +static inline void power_pmu_bhrb_disable(struct perf_event *event) {} +void power_pmu_flush_branch_stack(void) {} +static inline void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) {} #endif /* CONFIG_PPC32 */ static bool regs_use_siar(struct pt_regs *regs) { - return !!(regs->result & 1); + return !!regs->result; } /* @@ -130,22 +150,30 @@ static inline unsigned long perf_ip_adjust(struct pt_regs *regs) * If we're not doing instruction sampling, give them the SDAR * (sampled data address). If we are doing instruction sampling, then * only give them the SDAR if it corresponds to the instruction - * pointed to by SIAR; this is indicated by the [POWER6_]MMCRA_SDSYNC or - * the [POWER7P_]MMCRA_SDAR_VALID bit in MMCRA. + * pointed to by SIAR; this is indicated by the [POWER6_]MMCRA_SDSYNC, the + * [POWER7P_]MMCRA_SDAR_VALID bit in MMCRA, or the SDAR_VALID bit in SIER. */ static inline void perf_get_data_addr(struct pt_regs *regs, u64 *addrp) { unsigned long mmcra = regs->dsisr; - unsigned long sdsync; + bool sdar_valid; - if (ppmu->flags & PPMU_SIAR_VALID) - sdsync = POWER7P_MMCRA_SDAR_VALID; - else if (ppmu->flags & PPMU_ALT_SIPR) - sdsync = POWER6_MMCRA_SDSYNC; - else - sdsync = MMCRA_SDSYNC; + if (ppmu->flags & PPMU_HAS_SIER) + sdar_valid = regs->dar & SIER_SDAR_VALID; + else { + unsigned long sdsync; + + if (ppmu->flags & PPMU_SIAR_VALID) + sdsync = POWER7P_MMCRA_SDAR_VALID; + else if (ppmu->flags & PPMU_ALT_SIPR) + sdsync = POWER6_MMCRA_SDSYNC; + else + sdsync = MMCRA_SDSYNC; + + sdar_valid = mmcra & sdsync; + } - if (!(mmcra & MMCRA_SAMPLE_ENABLE) || (mmcra & sdsync)) + if (!(mmcra & MMCRA_SAMPLE_ENABLE) || sdar_valid) *addrp = mfspr(SPRN_SDAR); } @@ -175,11 +203,6 @@ static bool regs_sipr(struct pt_regs *regs) return !!(regs->dsisr & sipr); } -static bool regs_no_sipr(struct pt_regs *regs) -{ - return !!(regs->result & 2); -} - static inline u32 perf_flags_from_msr(struct pt_regs *regs) { if (regs->msr & MSR_PR) @@ -202,7 +225,7 @@ static inline u32 perf_get_misc_flags(struct pt_regs *regs) * SIAR which should give slightly more reliable * results */ - if (regs_no_sipr(regs)) { + if (ppmu->flags & PPMU_NO_SIPR) { unsigned long siar = mfspr(SPRN_SIAR); if (siar >= PAGE_OFFSET) return PERF_RECORD_MISC_KERNEL; @@ -233,22 +256,9 @@ static inline void perf_read_regs(struct pt_regs *regs) int use_siar; regs->dsisr = mmcra; - regs->result = 0; - - if (ppmu->flags & PPMU_NO_SIPR) - regs->result |= 2; - - /* - * On power8 if we're in random sampling mode, the SIER is updated. - * If we're in continuous sampling mode, we don't have SIPR. - */ - if (ppmu->flags & PPMU_HAS_SIER) { - if (marked) - regs->dar = mfspr(SPRN_SIER); - else - regs->result |= 2; - } + if (ppmu->flags & PPMU_HAS_SIER) + regs->dar = mfspr(SPRN_SIER); /* * If this isn't a PMU exception (eg a software event) the SIAR is @@ -273,12 +283,12 @@ static inline void perf_read_regs(struct pt_regs *regs) use_siar = 1; else if ((ppmu->flags & PPMU_NO_CONT_SAMPLING)) use_siar = 0; - else if (!regs_no_sipr(regs) && regs_sipr(regs)) + else if (!(ppmu->flags & PPMU_NO_SIPR) && regs_sipr(regs)) use_siar = 0; else use_siar = 1; - regs->result |= use_siar; + regs->result = use_siar; } /* @@ -302,12 +312,253 @@ static inline int siar_valid(struct pt_regs *regs) unsigned long mmcra = regs->dsisr; int marked = mmcra & MMCRA_SAMPLE_ENABLE; - if ((ppmu->flags & PPMU_SIAR_VALID) && marked) - return mmcra & POWER7P_MMCRA_SIAR_VALID; + if (marked) { + if (ppmu->flags & PPMU_HAS_SIER) + return regs->dar & SIER_SIAR_VALID; + + if (ppmu->flags & PPMU_SIAR_VALID) + return mmcra & POWER7P_MMCRA_SIAR_VALID; + } return 1; } + +/* Reset all possible BHRB entries */ +static void power_pmu_bhrb_reset(void) +{ + asm volatile(PPC_CLRBHRB); +} + +static void power_pmu_bhrb_enable(struct perf_event *event) +{ + struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); + + if (!ppmu->bhrb_nr) + return; + + /* Clear BHRB if we changed task context to avoid data leaks */ + if (event->ctx->task && cpuhw->bhrb_context != event->ctx) { + power_pmu_bhrb_reset(); + cpuhw->bhrb_context = event->ctx; + } + cpuhw->bhrb_users++; +} + +static void power_pmu_bhrb_disable(struct perf_event *event) +{ + struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); + + if (!ppmu->bhrb_nr) + return; + + cpuhw->bhrb_users--; + WARN_ON_ONCE(cpuhw->bhrb_users < 0); + + if (!cpuhw->disabled && !cpuhw->bhrb_users) { + /* BHRB cannot be turned off when other + * events are active on the PMU. + */ + + /* avoid stale pointer */ + cpuhw->bhrb_context = NULL; + } +} + +/* Called from ctxsw to prevent one process's branch entries to + * mingle with the other process's entries during context switch. + */ +void power_pmu_flush_branch_stack(void) +{ + if (ppmu->bhrb_nr) + power_pmu_bhrb_reset(); +} +/* Calculate the to address for a branch */ +static __u64 power_pmu_bhrb_to(u64 addr) +{ + unsigned int instr; + int ret; + __u64 target; + + if (is_kernel_addr(addr)) + return branch_target((unsigned int *)addr); + + /* Userspace: need copy instruction here then translate it */ + pagefault_disable(); + ret = __get_user_inatomic(instr, (unsigned int __user *)addr); + if (ret) { + pagefault_enable(); + return 0; + } + pagefault_enable(); + + target = branch_target(&instr); + if ((!target) || (instr & BRANCH_ABSOLUTE)) + return target; + + /* Translate relative branch target from kernel to user address */ + return target - (unsigned long)&instr + addr; +} + +/* Processing BHRB entries */ +void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) +{ + u64 val; + u64 addr; + int r_index, u_index, pred; + + r_index = 0; + u_index = 0; + while (r_index < ppmu->bhrb_nr) { + /* Assembly read function */ + val = read_bhrb(r_index++); + if (!val) + /* Terminal marker: End of valid BHRB entries */ + break; + else { + addr = val & BHRB_EA; + pred = val & BHRB_PREDICTION; + + if (!addr) + /* invalid entry */ + continue; + + /* Branches are read most recent first (ie. mfbhrb 0 is + * the most recent branch). + * There are two types of valid entries: + * 1) a target entry which is the to address of a + * computed goto like a blr,bctr,btar. The next + * entry read from the bhrb will be branch + * corresponding to this target (ie. the actual + * blr/bctr/btar instruction). + * 2) a from address which is an actual branch. If a + * target entry proceeds this, then this is the + * matching branch for that target. If this is not + * following a target entry, then this is a branch + * where the target is given as an immediate field + * in the instruction (ie. an i or b form branch). + * In this case we need to read the instruction from + * memory to determine the target/to address. + */ + + if (val & BHRB_TARGET) { + /* Target branches use two entries + * (ie. computed gotos/XL form) + */ + cpuhw->bhrb_entries[u_index].to = addr; + cpuhw->bhrb_entries[u_index].mispred = pred; + cpuhw->bhrb_entries[u_index].predicted = ~pred; + + /* Get from address in next entry */ + val = read_bhrb(r_index++); + addr = val & BHRB_EA; + if (val & BHRB_TARGET) { + /* Shouldn't have two targets in a + row.. Reset index and try again */ + r_index--; + addr = 0; + } + cpuhw->bhrb_entries[u_index].from = addr; + } else { + /* Branches to immediate field + (ie I or B form) */ + cpuhw->bhrb_entries[u_index].from = addr; + cpuhw->bhrb_entries[u_index].to = + power_pmu_bhrb_to(addr); + cpuhw->bhrb_entries[u_index].mispred = pred; + cpuhw->bhrb_entries[u_index].predicted = ~pred; + } + u_index++; + + } + } + cpuhw->bhrb_stack.nr = u_index; + return; +} + +static bool is_ebb_event(struct perf_event *event) +{ + /* + * This could be a per-PMU callback, but we'd rather avoid the cost. We + * check that the PMU supports EBB, meaning those that don't can still + * use bit 63 of the event code for something else if they wish. + */ + return (ppmu->flags & PPMU_EBB) && + ((event->attr.config >> EVENT_CONFIG_EBB_SHIFT) & 1); +} + +static int ebb_event_check(struct perf_event *event) +{ + struct perf_event *leader = event->group_leader; + + /* Event and group leader must agree on EBB */ + if (is_ebb_event(leader) != is_ebb_event(event)) + return -EINVAL; + + if (is_ebb_event(event)) { + if (!(event->attach_state & PERF_ATTACH_TASK)) + return -EINVAL; + + if (!leader->attr.pinned || !leader->attr.exclusive) + return -EINVAL; + + if (event->attr.inherit || event->attr.sample_period || + event->attr.enable_on_exec || event->attr.freq) + return -EINVAL; + } + + return 0; +} + +static void ebb_event_add(struct perf_event *event) +{ + if (!is_ebb_event(event) || current->thread.used_ebb) + return; + + /* + * IFF this is the first time we've added an EBB event, set + * PMXE in the user MMCR0 so we can detect when it's cleared by + * userspace. We need this so that we can context switch while + * userspace is in the EBB handler (where PMXE is 0). + */ + current->thread.used_ebb = 1; + current->thread.mmcr0 |= MMCR0_PMXE; +} + +static void ebb_switch_out(unsigned long mmcr0) +{ + if (!(mmcr0 & MMCR0_EBE)) + return; + + current->thread.siar = mfspr(SPRN_SIAR); + current->thread.sier = mfspr(SPRN_SIER); + current->thread.sdar = mfspr(SPRN_SDAR); + current->thread.mmcr0 = mmcr0 & MMCR0_USER_MASK; + current->thread.mmcr2 = mfspr(SPRN_MMCR2) & MMCR2_USER_MASK; +} + +static unsigned long ebb_switch_in(bool ebb, unsigned long mmcr0) +{ + if (!ebb) + goto out; + + /* Enable EBB and read/write to all 6 PMCs for userspace */ + mmcr0 |= MMCR0_EBE | MMCR0_PMCC_U6; + + /* Add any bits from the user reg, FC or PMAO */ + mmcr0 |= current->thread.mmcr0; + + /* Be careful not to set PMXE if userspace had it cleared */ + if (!(current->thread.mmcr0 & MMCR0_PMXE)) + mmcr0 &= ~MMCR0_PMXE; + + mtspr(SPRN_SIAR, current->thread.siar); + mtspr(SPRN_SIER, current->thread.sier); + mtspr(SPRN_SDAR, current->thread.sdar); + mtspr(SPRN_MMCR2, current->thread.mmcr2); +out: + return mmcr0; +} #endif /* CONFIG_PPC64 */ static void perf_event_interrupt(struct pt_regs *regs); @@ -578,6 +829,13 @@ static void power_pmu_read(struct perf_event *event) if (!event->hw.idx) return; + + if (is_ebb_event(event)) { + val = read_pmc(event->hw.idx); + local64_set(&event->hw.prev_count, val); + return; + } + /* * Performance monitor interrupts come even when interrupts * are soft-disabled, as long as interrupts are hard-enabled. @@ -698,7 +956,7 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0) static void power_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; - unsigned long flags; + unsigned long flags, mmcr0, val; if (!ppmu) return; @@ -706,9 +964,6 @@ static void power_pmu_disable(struct pmu *pmu) cpuhw = &__get_cpu_var(cpu_hw_events); if (!cpuhw->disabled) { - cpuhw->disabled = 1; - cpuhw->n_added = 0; - /* * Check if we ever enabled the PMU on this cpu. */ @@ -718,6 +973,21 @@ static void power_pmu_disable(struct pmu *pmu) } /* + * Set the 'freeze counters' bit, clear EBE/PMCC/PMAO/FC56. + */ + val = mmcr0 = mfspr(SPRN_MMCR0); + val |= MMCR0_FC; + val &= ~(MMCR0_EBE | MMCR0_PMCC | MMCR0_PMAO | MMCR0_FC56); + + /* + * The barrier is to make sure the mtspr has been + * executed and the PMU has frozen the events etc. + * before we return. + */ + write_mmcr0(cpuhw, val); + mb(); + + /* * Disable instruction sampling if it was enabled */ if (cpuhw->mmcr[2] & MMCRA_SAMPLE_ENABLE) { @@ -726,15 +996,12 @@ static void power_pmu_disable(struct pmu *pmu) mb(); } - /* - * Set the 'freeze counters' bit. - * The barrier is to make sure the mtspr has been - * executed and the PMU has frozen the events - * before we return. - */ - write_mmcr0(cpuhw, mfspr(SPRN_MMCR0) | MMCR0_FC); - mb(); + cpuhw->disabled = 1; + cpuhw->n_added = 0; + + ebb_switch_out(mmcr0); } + local_irq_restore(flags); } @@ -749,23 +1016,36 @@ static void power_pmu_enable(struct pmu *pmu) struct cpu_hw_events *cpuhw; unsigned long flags; long i; - unsigned long val; + unsigned long val, mmcr0; s64 left; unsigned int hwc_index[MAX_HWEVENTS]; int n_lim; int idx; + bool ebb; if (!ppmu) return; local_irq_save(flags); + cpuhw = &__get_cpu_var(cpu_hw_events); - if (!cpuhw->disabled) { - local_irq_restore(flags); - return; + if (!cpuhw->disabled) + goto out; + + if (cpuhw->n_events == 0) { + ppc_set_pmu_inuse(0); + goto out; } + cpuhw->disabled = 0; /* + * EBB requires an exclusive group and all events must have the EBB + * flag set, or not set, so we can just check a single event. Also we + * know we have at least one event. + */ + ebb = is_ebb_event(cpuhw->event[0]); + + /* * If we didn't change anything, or only removed events, * no need to recalculate MMCR* settings and reset the PMCs. * Just reenable the PMU with the current MMCR* settings @@ -774,8 +1054,6 @@ static void power_pmu_enable(struct pmu *pmu) if (!cpuhw->n_added) { mtspr(SPRN_MMCRA, cpuhw->mmcr[2] & ~MMCRA_SAMPLE_ENABLE); mtspr(SPRN_MMCR1, cpuhw->mmcr[1]); - if (cpuhw->n_events == 0) - ppc_set_pmu_inuse(0); goto out_enable; } @@ -842,25 +1120,34 @@ static void power_pmu_enable(struct pmu *pmu) ++n_lim; continue; } - val = 0; - if (event->hw.sample_period) { - left = local64_read(&event->hw.period_left); - if (left < 0x80000000L) - val = 0x80000000L - left; + + if (ebb) + val = local64_read(&event->hw.prev_count); + else { + val = 0; + if (event->hw.sample_period) { + left = local64_read(&event->hw.period_left); + if (left < 0x80000000L) + val = 0x80000000L - left; + } + local64_set(&event->hw.prev_count, val); } - local64_set(&event->hw.prev_count, val); + event->hw.idx = idx; if (event->hw.state & PERF_HES_STOPPED) val = 0; write_pmc(idx, val); + perf_event_update_userpage(event); } cpuhw->n_limited = n_lim; cpuhw->mmcr[0] |= MMCR0_PMXE | MMCR0_FCECE; out_enable: + mmcr0 = ebb_switch_in(ebb, cpuhw->mmcr[0]); + mb(); - write_mmcr0(cpuhw, cpuhw->mmcr[0]); + write_mmcr0(cpuhw, mmcr0); /* * Enable instruction sampling if necessary @@ -904,47 +1191,6 @@ static int collect_events(struct perf_event *group, int max_count, return n; } -/* Reset all possible BHRB entries */ -static void power_pmu_bhrb_reset(void) -{ - asm volatile(PPC_CLRBHRB); -} - -void power_pmu_bhrb_enable(struct perf_event *event) -{ - struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); - - if (!ppmu->bhrb_nr) - return; - - /* Clear BHRB if we changed task context to avoid data leaks */ - if (event->ctx->task && cpuhw->bhrb_context != event->ctx) { - power_pmu_bhrb_reset(); - cpuhw->bhrb_context = event->ctx; - } - cpuhw->bhrb_users++; -} - -void power_pmu_bhrb_disable(struct perf_event *event) -{ - struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); - - if (!ppmu->bhrb_nr) - return; - - cpuhw->bhrb_users--; - WARN_ON_ONCE(cpuhw->bhrb_users < 0); - - if (!cpuhw->disabled && !cpuhw->bhrb_users) { - /* BHRB cannot be turned off when other - * events are active on the PMU. - */ - - /* avoid stale pointer */ - cpuhw->bhrb_context = NULL; - } -} - /* * Add a event to the PMU. * If all events are not already frozen, then we disable and @@ -999,6 +1245,8 @@ static int power_pmu_add(struct perf_event *event, int ef_flags) event->hw.config = cpuhw->events[n0]; nocheck: + ebb_event_add(event); + ++cpuhw->n_events; ++cpuhw->n_added; @@ -1180,15 +1428,6 @@ int power_pmu_commit_txn(struct pmu *pmu) return 0; } -/* Called from ctxsw to prevent one process's branch entries to - * mingle with the other process's entries during context switch. - */ -void power_pmu_flush_branch_stack(void) -{ - if (ppmu->bhrb_nr) - power_pmu_bhrb_reset(); -} - /* * Return 1 if we might be able to put event on a limited PMC, * or 0 if not. @@ -1368,6 +1607,11 @@ static int power_pmu_event_init(struct perf_event *event) } } + /* Extra checks for EBB */ + err = ebb_event_check(event); + if (err) + return err; + /* * If this is in a group, check if it can go on with all the * other hardware events in the group. We assume the event @@ -1407,6 +1651,13 @@ static int power_pmu_event_init(struct perf_event *event) local64_set(&event->hw.period_left, event->hw.last_period); /* + * For EBB events we just context switch the PMC value, we don't do any + * of the sample_period logic. We use hw.prev_count for this. + */ + if (is_ebb_event(event)) + local64_set(&event->hw.prev_count, 0); + + /* * See if we need to reserve the PMU. * If no events are currently in use, then we have to take a * mutex to ensure that we don't race with another task doing @@ -1458,77 +1709,6 @@ struct pmu power_pmu = { .flush_branch_stack = power_pmu_flush_branch_stack, }; -/* Processing BHRB entries */ -void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) -{ - u64 val; - u64 addr; - int r_index, u_index, target, pred; - - r_index = 0; - u_index = 0; - while (r_index < ppmu->bhrb_nr) { - /* Assembly read function */ - val = read_bhrb(r_index); - - /* Terminal marker: End of valid BHRB entries */ - if (val == 0) { - break; - } else { - /* BHRB field break up */ - addr = val & BHRB_EA; - pred = val & BHRB_PREDICTION; - target = val & BHRB_TARGET; - - /* Probable Missed entry: Not applicable for POWER8 */ - if ((addr == 0) && (target == 0) && (pred == 1)) { - r_index++; - continue; - } - - /* Real Missed entry: Power8 based missed entry */ - if ((addr == 0) && (target == 1) && (pred == 1)) { - r_index++; - continue; - } - - /* Reserved condition: Not a valid entry */ - if ((addr == 0) && (target == 1) && (pred == 0)) { - r_index++; - continue; - } - - /* Is a target address */ - if (val & BHRB_TARGET) { - /* First address cannot be a target address */ - if (r_index == 0) { - r_index++; - continue; - } - - /* Update target address for the previous entry */ - cpuhw->bhrb_entries[u_index - 1].to = addr; - cpuhw->bhrb_entries[u_index - 1].mispred = pred; - cpuhw->bhrb_entries[u_index - 1].predicted = ~pred; - - /* Dont increment u_index */ - r_index++; - } else { - /* Update address, flags for current entry */ - cpuhw->bhrb_entries[u_index].from = addr; - cpuhw->bhrb_entries[u_index].mispred = pred; - cpuhw->bhrb_entries[u_index].predicted = ~pred; - - /* Successfully popullated one entry */ - u_index++; - r_index++; - } - } - } - cpuhw->bhrb_stack.nr = u_index; - return; -} - /* * A counter has overflowed; update its count and record * things if requested. Note that interrupts are hard-disabled @@ -1725,7 +1905,7 @@ static void perf_event_interrupt(struct pt_regs *regs) } } } - if ((!found) && printk_ratelimit()) + if (!found && !nmi && printk_ratelimit()) printk(KERN_WARNING "Can't find PMC that caused IRQ\n"); /* @@ -1753,7 +1933,7 @@ static void power_pmu_setup(int cpu) cpuhw->mmcr[0] = MMCR0_FC; } -static int __cpuinit +static int power_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; @@ -1770,7 +1950,7 @@ power_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu return NOTIFY_OK; } -int __cpuinit register_power_pmu(struct power_pmu *pmu) +int register_power_pmu(struct power_pmu *pmu) { if (ppmu) return -EBUSY; /* something's already registered */ |