diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-03 17:38:09 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-03 17:38:09 -0800 |
commit | b02ac6b18cd4e2c76bf0a102c20c429b973f5f76 (patch) | |
tree | 87b3648f448627d61cb9ba32511584d6318b7bb6 /arch/x86 | |
parent | 105ff3cbf225036b75a6a46c96d1ddce8e7bdc66 (diff) | |
parent | bebd23a2ed31d47e7dd746d3b125068aa2c42d85 (diff) | |
download | linux-b02ac6b18cd4e2c76bf0a102c20c429b973f5f76.tar.bz2 |
Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf updates from Ingo Molnar:
"Kernel side changes:
- Improve accuracy of perf/sched clock on x86. (Adrian Hunter)
- Intel DS and BTS updates. (Alexander Shishkin)
- Intel cstate PMU support. (Kan Liang)
- Add group read support to perf_event_read(). (Peter Zijlstra)
- Branch call hardware sampling support, implemented on x86 and
PowerPC. (Stephane Eranian)
- Event groups transactional interface enhancements. (Sukadev
Bhattiprolu)
- Enable proper x86/intel/uncore PMU support on multi-segment PCI
systems. (Taku Izumi)
- ... misc fixes and cleanups.
The perf tooling team was very busy again with 200+ commits, the full
diff doesn't fit into lkml size limits. Here's an (incomplete) list
of the tooling highlights:
New features:
- Change the default event used in all tools (record/top): use the
most precise "cycles" hw counter available, i.e. when the user
doesn't specify any event, it will try using cycles:ppp, cycles:pp,
etc and fall back transparently until it finds a working counter.
(Arnaldo Carvalho de Melo)
- Integration of perf with eBPF that, given an eBPF .c source file
(or .o file built for the 'bpf' target with clang), will get it
automatically built, validated and loaded into the kernel via the
sys_bpf syscall, which can then be used and seen using 'perf trace'
and other tools.
(Wang Nan)
Various user interface improvements:
- Automatic pager invocation on long help output. (Namhyung Kim)
- Search for more options when passing args to -h, e.g.: (Arnaldo
Carvalho de Melo)
$ perf report -h interface
Usage: perf report [<options>]
--gtk Use the GTK2 interface
--stdio Use the stdio interface
--tui Use the TUI interface
- Show ordered command line options when -h is used or when an
unknown option is specified. (Arnaldo Carvalho de Melo)
- If options are passed after -h, show just its descriptions, not all
options. (Arnaldo Carvalho de Melo)
- Implement column based horizontal scrolling in the hists browser
(top, report), making it possible to use the TUI for things like
'perf mem report' where there are many more columns than can fit in
a terminal. (Arnaldo Carvalho de Melo)
- Enhance the error reporting of tracepoint event parsing, e.g.:
$ oldperf record -e sched:sched_switc usleep 1
event syntax error: 'sched:sched_switc'
\___ unknown tracepoint
Run 'perf list' for a list of valid events
Now we get the much nicer:
$ perf record -e sched:sched_switc ls
event syntax error: 'sched:sched_switc'
\___ can't access trace events
Error: No permissions to read /sys/kernel/debug/tracing/events/sched/sched_switc
Hint: Try 'sudo mount -o remount,mode=755 /sys/kernel/debug'
And after we have those mount point permissions fixed:
$ perf record -e sched:sched_switc ls
event syntax error: 'sched:sched_switc'
\___ unknown tracepoint
Error: File /sys/kernel/debug/tracing/events/sched/sched_switc not found.
Hint: Perhaps this kernel misses some CONFIG_ setting to enable this feature?.
I.e. basically now the event parsing routing uses the strerror_open()
routines introduced by and used in 'perf trace' work. (Jiri Olsa)
- Fail properly when pattern matching fails to find a tracepoint,
i.e. '-e non:existent' was being correctly handled, with a proper
error message about that not being a valid event, but '-e
non:existent*' wasn't, fix it. (Jiri Olsa)
- Do event name substring search as last resort in 'perf list'.
(Arnaldo Carvalho de Melo)
E.g.:
# perf list clock
List of pre-defined events (to be used in -e):
cpu-clock [Software event]
task-clock [Software event]
uncore_cbox_0/clockticks/ [Kernel PMU event]
uncore_cbox_1/clockticks/ [Kernel PMU event]
kvm:kvm_pvclock_update [Tracepoint event]
kvm:kvm_update_master_clock [Tracepoint event]
power:clock_disable [Tracepoint event]
power:clock_enable [Tracepoint event]
power:clock_set_rate [Tracepoint event]
syscalls:sys_enter_clock_adjtime [Tracepoint event]
syscalls:sys_enter_clock_getres [Tracepoint event]
syscalls:sys_enter_clock_gettime [Tracepoint event]
syscalls:sys_enter_clock_nanosleep [Tracepoint event]
syscalls:sys_enter_clock_settime [Tracepoint event]
syscalls:sys_exit_clock_adjtime [Tracepoint event]
syscalls:sys_exit_clock_getres [Tracepoint event]
syscalls:sys_exit_clock_gettime [Tracepoint event]
syscalls:sys_exit_clock_nanosleep [Tracepoint event]
syscalls:sys_exit_clock_settime [Tracepoint event]
Intel PT hardware tracing enhancements:
- Accept a zero --itrace period, meaning "as often as possible". In
the case of Intel PT that is the same as a period of 1 and a unit
of 'instructions' (i.e. --itrace=i1i). (Adrian Hunter)
- Harmonize itrace's synthesized callchains with the existing
--max-stack tool option. (Adrian Hunter)
- Allow time to be displayed in nanoseconds in 'perf script'.
(Adrian Hunter)
- Fix potential infinite loop when handling Intel PT timestamps.
(Adrian Hunter)
- Slighly improve Intel PT debug logging. (Adrian Hunter)
- Warn when AUX data has been lost, just like when processing
PERF_RECORD_LOST. (Adrian Hunter)
- Further document export-to-postgresql.py script. (Adrian Hunter)
- Add option to synthesize branch stack from auxtrace data. (Adrian
Hunter)
Misc notable changes:
- Switch the default callchain output mode to 'graph,0.5,caller', to
make it look like the default for other tools, reducing the
learning curve for people used to 'caller' based viewing. (Arnaldo
Carvalho de Melo)
- various call chain usability enhancements. (Namhyung Kim)
- Introduce the 'P' event modifier, meaning 'max precision level,
please', i.e.:
$ perf record -e cycles:P usleep 1
Is now similar to:
$ perf record usleep 1
Useful, for instance, when specifying multiple events. (Jiri Olsa)
- Add 'socket' sort entry, to sort by the processor socket in 'perf
top' and 'perf report'. (Kan Liang)
- Introduce --socket-filter to 'perf report', for filtering by
processor socket. (Kan Liang)
- Add new "Zoom into Processor Socket" operation in the perf hists
browser, used in 'perf top' and 'perf report'. (Kan Liang)
- Allow probing on kmodules without DWARF. (Masami Hiramatsu)
- Fix 'perf probe -l' for probes added to kernel module functions.
(Masami Hiramatsu)
- Preparatory work for the 'perf stat record' feature that will allow
generating perf.data files with counting data in addition to the
sampling mode we have now (Jiri Olsa)
- Update libtraceevent KVM plugin. (Paolo Bonzini)
- ... plus lots of other enhancements that I failed to list properly,
by: Adrian Hunter, Alexander Shishkin, Andi Kleen, Andrzej Hajda,
Arnaldo Carvalho de Melo, Dima Kogan, Don Zickus, Geliang Tang, He
Kuang, Huaitong Han, Ingo Molnar, Jan Stancek, Jiri Olsa, Kan
Liang, Kirill Tkhai, Masami Hiramatsu, Matt Fleming, Namhyung Kim,
Paolo Bonzini, Peter Zijlstra, Rabin Vincent, Scott Wood, Stephane
Eranian, Sukadev Bhattiprolu, Taku Izumi, Vaishali Thakkar, Wang
Nan, Yang Shi and Yunlong Song"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (260 commits)
perf unwind: Pass symbol source to libunwind
tools build: Fix libiberty feature detection
perf tools: Compile scriptlets to BPF objects when passing '.c' to --event
perf record: Add clang options for compiling BPF scripts
perf bpf: Attach eBPF filter to perf event
perf tools: Make sure fixdep is built before libbpf
perf script: Enable printing of branch stack
perf trace: Add cmd string table to decode sys_bpf first arg
perf bpf: Collect perf_evsel in BPF object files
perf tools: Load eBPF object into kernel
perf tools: Create probe points for BPF programs
perf tools: Enable passing bpf object file to --event
perf ebpf: Add the libbpf glue
perf tools: Make perf depend on libbpf
perf symbols: Fix endless loop in dso__split_kallsyms_for_kcore
perf tools: Enable pre-event inherit setting by config terms
perf symbols: we can now read separate debug-info files based on a build ID
perf symbols: Fix type error when reading a build-id
perf tools: Search for more options when passing args to -h
perf stat: Cache aggregated map entries in extra cpumap
...
Diffstat (limited to 'arch/x86')
-rw-r--r-- | arch/x86/kernel/cpu/Makefile | 1 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/intel_cacheinfo.c | 8 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event.c | 39 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event.h | 2 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_bts.c | 13 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_cstate.c | 694 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_ds.c | 40 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_lbr.c | 8 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_pt.c | 7 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_uncore.c | 61 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_uncore.h | 12 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c | 16 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c | 34 | ||||
-rw-r--r-- | arch/x86/kernel/tsc.c | 35 | ||||
-rw-r--r-- | arch/x86/lib/x86-opcode-map.txt | 24 |
15 files changed, 938 insertions, 56 deletions
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 4eb065c6bed2..58031303e304 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -41,6 +41,7 @@ obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_knc.o perf_event_p4.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_rapl.o perf_event_intel_cqm.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_pt.o perf_event_intel_bts.o +obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_cstate.o obj-$(CONFIG_PERF_EVENTS_INTEL_UNCORE) += perf_event_intel_uncore.o \ perf_event_intel_uncore_snb.o \ diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index be4febc58b94..e38d338a6447 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -157,7 +157,7 @@ struct _cpuid4_info_regs { struct amd_northbridge *nb; }; -unsigned short num_cache_leaves; +static unsigned short num_cache_leaves; /* AMD doesn't have CPUID4. Emulate it here to report the same information to the user. This makes some assumptions about the machine: @@ -326,7 +326,7 @@ static void amd_calc_l3_indices(struct amd_northbridge *nb) * * @returns: the disabled index if used or negative value if slot free. */ -int amd_get_l3_disable_slot(struct amd_northbridge *nb, unsigned slot) +static int amd_get_l3_disable_slot(struct amd_northbridge *nb, unsigned slot) { unsigned int reg = 0; @@ -403,8 +403,8 @@ static void amd_l3_disable_index(struct amd_northbridge *nb, int cpu, * * @return: 0 on success, error status on failure */ -int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, unsigned slot, - unsigned long index) +static int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, + unsigned slot, unsigned long index) { int ret = 0; diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 66dd3fe99b82..4562cf070c27 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1175,7 +1175,7 @@ static int x86_pmu_add(struct perf_event *event, int flags) * skip the schedulability test here, it will be performed * at commit time (->commit_txn) as a whole. */ - if (cpuc->group_flag & PERF_EVENT_TXN) + if (cpuc->txn_flags & PERF_PMU_TXN_ADD) goto done_collect; ret = x86_pmu.schedule_events(cpuc, n, assign); @@ -1326,7 +1326,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) * XXX assumes any ->del() called during a TXN will only be on * an event added during that same TXN. */ - if (cpuc->group_flag & PERF_EVENT_TXN) + if (cpuc->txn_flags & PERF_PMU_TXN_ADD) return; /* @@ -1748,11 +1748,22 @@ static inline void x86_pmu_read(struct perf_event *event) * Start group events scheduling transaction * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time + * + * We only support PERF_PMU_TXN_ADD transactions. Save the + * transaction flags but otherwise ignore non-PERF_PMU_TXN_ADD + * transactions. */ -static void x86_pmu_start_txn(struct pmu *pmu) +static void x86_pmu_start_txn(struct pmu *pmu, unsigned int txn_flags) { + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + WARN_ON_ONCE(cpuc->txn_flags); /* txn already in flight */ + + cpuc->txn_flags = txn_flags; + if (txn_flags & ~PERF_PMU_TXN_ADD) + return; + perf_pmu_disable(pmu); - __this_cpu_or(cpu_hw_events.group_flag, PERF_EVENT_TXN); __this_cpu_write(cpu_hw_events.n_txn, 0); } @@ -1763,7 +1774,16 @@ static void x86_pmu_start_txn(struct pmu *pmu) */ static void x86_pmu_cancel_txn(struct pmu *pmu) { - __this_cpu_and(cpu_hw_events.group_flag, ~PERF_EVENT_TXN); + unsigned int txn_flags; + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + WARN_ON_ONCE(!cpuc->txn_flags); /* no txn in flight */ + + txn_flags = cpuc->txn_flags; + cpuc->txn_flags = 0; + if (txn_flags & ~PERF_PMU_TXN_ADD) + return; + /* * Truncate collected array by the number of events added in this * transaction. See x86_pmu_add() and x86_pmu_*_txn(). @@ -1786,6 +1806,13 @@ static int x86_pmu_commit_txn(struct pmu *pmu) int assign[X86_PMC_IDX_MAX]; int n, ret; + WARN_ON_ONCE(!cpuc->txn_flags); /* no txn in flight */ + + if (cpuc->txn_flags & ~PERF_PMU_TXN_ADD) { + cpuc->txn_flags = 0; + return 0; + } + n = cpuc->n_events; if (!x86_pmu_initialized()) @@ -1801,7 +1828,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu) */ memcpy(cpuc->assign, assign, n*sizeof(int)); - cpuc->group_flag &= ~PERF_EVENT_TXN; + cpuc->txn_flags = 0; perf_pmu_enable(pmu); return 0; } diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h index 165be83a7fa4..499f533dd3cc 100644 --- a/arch/x86/kernel/cpu/perf_event.h +++ b/arch/x86/kernel/cpu/perf_event.h @@ -196,7 +196,7 @@ struct cpu_hw_events { int n_excl; /* the number of exclusive events */ - unsigned int group_flag; + unsigned int txn_flags; int is_fake; /* diff --git a/arch/x86/kernel/cpu/perf_event_intel_bts.c b/arch/x86/kernel/cpu/perf_event_intel_bts.c index d1c0f254afbe..2cad71d1b14c 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_bts.c +++ b/arch/x86/kernel/cpu/perf_event_intel_bts.c @@ -495,6 +495,19 @@ static int bts_event_init(struct perf_event *event) if (x86_add_exclusive(x86_lbr_exclusive_bts)) return -EBUSY; + /* + * BTS leaks kernel addresses even when CPL0 tracing is + * disabled, so disallow intel_bts driver for unprivileged + * users on paranoid systems since it provides trace data + * to the user in a zero-copy fashion. + * + * Note that the default paranoia setting permits unprivileged + * users to profile the kernel. + */ + if (event->attr.exclude_kernel && perf_paranoid_kernel() && + !capable(CAP_SYS_ADMIN)) + return -EACCES; + ret = x86_reserve_hardware(); if (ret) { x86_del_exclusive(x86_lbr_exclusive_bts); diff --git a/arch/x86/kernel/cpu/perf_event_intel_cstate.c b/arch/x86/kernel/cpu/perf_event_intel_cstate.c new file mode 100644 index 000000000000..75a38b5a2e26 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_intel_cstate.c @@ -0,0 +1,694 @@ +/* + * perf_event_intel_cstate.c: support cstate residency counters + * + * Copyright (C) 2015, Intel Corp. + * Author: Kan Liang (kan.liang@intel.com) + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Library General Public License for more details. + * + */ + +/* + * This file export cstate related free running (read-only) counters + * for perf. These counters may be use simultaneously by other tools, + * such as turbostat. However, it still make sense to implement them + * in perf. Because we can conveniently collect them together with + * other events, and allow to use them from tools without special MSR + * access code. + * + * The events only support system-wide mode counting. There is no + * sampling support because it is not supported by the hardware. + * + * According to counters' scope and category, two PMUs are registered + * with the perf_event core subsystem. + * - 'cstate_core': The counter is available for each physical core. + * The counters include CORE_C*_RESIDENCY. + * - 'cstate_pkg': The counter is available for each physical package. + * The counters include PKG_C*_RESIDENCY. + * + * All of these counters are specified in the IntelĀ® 64 and IA-32 + * Architectures Software Developer.s Manual Vol3b. + * + * Model specific counters: + * MSR_CORE_C1_RES: CORE C1 Residency Counter + * perf code: 0x00 + * Available model: SLM,AMT + * Scope: Core (each processor core has a MSR) + * MSR_CORE_C3_RESIDENCY: CORE C3 Residency Counter + * perf code: 0x01 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_CORE_C6_RESIDENCY: CORE C6 Residency Counter + * perf code: 0x02 + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_CORE_C7_RESIDENCY: CORE C7 Residency Counter + * perf code: 0x03 + * Available model: SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_PKG_C2_RESIDENCY: Package C2 Residency Counter. + * perf code: 0x00 + * Available model: SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C3_RESIDENCY: Package C3 Residency Counter. + * perf code: 0x01 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C6_RESIDENCY: Package C6 Residency Counter. + * perf code: 0x02 + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C7_RESIDENCY: Package C7 Residency Counter. + * perf code: 0x03 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C8_RESIDENCY: Package C8 Residency Counter. + * perf code: 0x04 + * Available model: HSW ULT only + * Scope: Package (physical package) + * MSR_PKG_C9_RESIDENCY: Package C9 Residency Counter. + * perf code: 0x05 + * Available model: HSW ULT only + * Scope: Package (physical package) + * MSR_PKG_C10_RESIDENCY: Package C10 Residency Counter. + * perf code: 0x06 + * Available model: HSW ULT only + * Scope: Package (physical package) + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/perf_event.h> +#include <asm/cpu_device_id.h> +#include "perf_event.h" + +#define DEFINE_CSTATE_FORMAT_ATTR(_var, _name, _format) \ +static ssize_t __cstate_##_var##_show(struct kobject *kobj, \ + struct kobj_attribute *attr, \ + char *page) \ +{ \ + BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE); \ + return sprintf(page, _format "\n"); \ +} \ +static struct kobj_attribute format_attr_##_var = \ + __ATTR(_name, 0444, __cstate_##_var##_show, NULL) + +static ssize_t cstate_get_attr_cpumask(struct device *dev, + struct device_attribute *attr, + char *buf); + +struct perf_cstate_msr { + u64 msr; + struct perf_pmu_events_attr *attr; + bool (*test)(int idx); +}; + + +/* cstate_core PMU */ + +static struct pmu cstate_core_pmu; +static bool has_cstate_core; + +enum perf_cstate_core_id { + /* + * cstate_core events + */ + PERF_CSTATE_CORE_C1_RES = 0, + PERF_CSTATE_CORE_C3_RES, + PERF_CSTATE_CORE_C6_RES, + PERF_CSTATE_CORE_C7_RES, + + PERF_CSTATE_CORE_EVENT_MAX, +}; + +bool test_core(int idx) +{ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || + boot_cpu_data.x86 != 6) + return false; + + switch (boot_cpu_data.x86_model) { + case 30: /* 45nm Nehalem */ + case 26: /* 45nm Nehalem-EP */ + case 46: /* 45nm Nehalem-EX */ + + case 37: /* 32nm Westmere */ + case 44: /* 32nm Westmere-EP */ + case 47: /* 32nm Westmere-EX */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + case 42: /* 32nm SandyBridge */ + case 45: /* 32nm SandyBridge-E/EN/EP */ + + case 58: /* 22nm IvyBridge */ + case 62: /* 22nm IvyBridge-EP/EX */ + + case 60: /* 22nm Haswell Core */ + case 63: /* 22nm Haswell Server */ + case 69: /* 22nm Haswell ULT */ + case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ + + case 61: /* 14nm Broadwell Core-M */ + case 86: /* 14nm Broadwell Xeon D */ + case 71: /* 14nm Broadwell + GT3e (Intel Iris Pro graphics) */ + case 79: /* 14nm Broadwell Server */ + + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES || + idx == PERF_CSTATE_CORE_C7_RES) + return true; + break; + case 55: /* 22nm Atom "Silvermont" */ + case 77: /* 22nm Atom "Silvermont Avoton/Rangely" */ + case 76: /* 14nm Atom "Airmont" */ + if (idx == PERF_CSTATE_CORE_C1_RES || + idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + } + + return false; +} + +PMU_EVENT_ATTR_STRING(c1-residency, evattr_cstate_core_c1, "event=0x00"); +PMU_EVENT_ATTR_STRING(c3-residency, evattr_cstate_core_c3, "event=0x01"); +PMU_EVENT_ATTR_STRING(c6-residency, evattr_cstate_core_c6, "event=0x02"); +PMU_EVENT_ATTR_STRING(c7-residency, evattr_cstate_core_c7, "event=0x03"); + +static struct perf_cstate_msr core_msr[] = { + [PERF_CSTATE_CORE_C1_RES] = { MSR_CORE_C1_RES, &evattr_cstate_core_c1, test_core, }, + [PERF_CSTATE_CORE_C3_RES] = { MSR_CORE_C3_RESIDENCY, &evattr_cstate_core_c3, test_core, }, + [PERF_CSTATE_CORE_C6_RES] = { MSR_CORE_C6_RESIDENCY, &evattr_cstate_core_c6, test_core, }, + [PERF_CSTATE_CORE_C7_RES] = { MSR_CORE_C7_RESIDENCY, &evattr_cstate_core_c7, test_core, }, +}; + +static struct attribute *core_events_attrs[PERF_CSTATE_CORE_EVENT_MAX + 1] = { + NULL, +}; + +static struct attribute_group core_events_attr_group = { + .name = "events", + .attrs = core_events_attrs, +}; + +DEFINE_CSTATE_FORMAT_ATTR(core_event, event, "config:0-63"); +static struct attribute *core_format_attrs[] = { + &format_attr_core_event.attr, + NULL, +}; + +static struct attribute_group core_format_attr_group = { + .name = "format", + .attrs = core_format_attrs, +}; + +static cpumask_t cstate_core_cpu_mask; +static DEVICE_ATTR(cpumask, S_IRUGO, cstate_get_attr_cpumask, NULL); + +static struct attribute *cstate_cpumask_attrs[] = { + &dev_attr_cpumask.attr, + NULL, +}; + +static struct attribute_group cpumask_attr_group = { + .attrs = cstate_cpumask_attrs, +}; + +static const struct attribute_group *core_attr_groups[] = { + &core_events_attr_group, + &core_format_attr_group, + &cpumask_attr_group, + NULL, +}; + +/* cstate_core PMU end */ + + +/* cstate_pkg PMU */ + +static struct pmu cstate_pkg_pmu; +static bool has_cstate_pkg; + +enum perf_cstate_pkg_id { + /* + * cstate_pkg events + */ + PERF_CSTATE_PKG_C2_RES = 0, + PERF_CSTATE_PKG_C3_RES, + PERF_CSTATE_PKG_C6_RES, + PERF_CSTATE_PKG_C7_RES, + PERF_CSTATE_PKG_C8_RES, + PERF_CSTATE_PKG_C9_RES, + PERF_CSTATE_PKG_C10_RES, + + PERF_CSTATE_PKG_EVENT_MAX, +}; + +bool test_pkg(int idx) +{ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || + boot_cpu_data.x86 != 6) + return false; + + switch (boot_cpu_data.x86_model) { + case 30: /* 45nm Nehalem */ + case 26: /* 45nm Nehalem-EP */ + case 46: /* 45nm Nehalem-EX */ + + case 37: /* 32nm Westmere */ + case 44: /* 32nm Westmere-EP */ + case 47: /* 32nm Westmere-EX */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES || + idx == PERF_CSTATE_CORE_C7_RES) + return true; + break; + case 42: /* 32nm SandyBridge */ + case 45: /* 32nm SandyBridge-E/EN/EP */ + + case 58: /* 22nm IvyBridge */ + case 62: /* 22nm IvyBridge-EP/EX */ + + case 60: /* 22nm Haswell Core */ + case 63: /* 22nm Haswell Server */ + case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ + + case 61: /* 14nm Broadwell Core-M */ + case 86: /* 14nm Broadwell Xeon D */ + case 71: /* 14nm Broadwell + GT3e (Intel Iris Pro graphics) */ + case 79: /* 14nm Broadwell Server */ + + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + if (idx == PERF_CSTATE_PKG_C2_RES || + idx == PERF_CSTATE_PKG_C3_RES || + idx == PERF_CSTATE_PKG_C6_RES || + idx == PERF_CSTATE_PKG_C7_RES) + return true; + break; + case 55: /* 22nm Atom "Silvermont" */ + case 77: /* 22nm Atom "Silvermont Avoton/Rangely" */ + case 76: /* 14nm Atom "Airmont" */ + if (idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + case 69: /* 22nm Haswell ULT */ + if (idx == PERF_CSTATE_PKG_C2_RES || + idx == PERF_CSTATE_PKG_C3_RES || + idx == PERF_CSTATE_PKG_C6_RES || + idx == PERF_CSTATE_PKG_C7_RES || + idx == PERF_CSTATE_PKG_C8_RES || + idx == PERF_CSTATE_PKG_C9_RES || + idx == PERF_CSTATE_PKG_C10_RES) + return true; + break; + } + + return false; +} + +PMU_EVENT_ATTR_STRING(c2-residency, evattr_cstate_pkg_c2, "event=0x00"); +PMU_EVENT_ATTR_STRING(c3-residency, evattr_cstate_pkg_c3, "event=0x01"); +PMU_EVENT_ATTR_STRING(c6-residency, evattr_cstate_pkg_c6, "event=0x02"); +PMU_EVENT_ATTR_STRING(c7-residency, evattr_cstate_pkg_c7, "event=0x03"); +PMU_EVENT_ATTR_STRING(c8-residency, evattr_cstate_pkg_c8, "event=0x04"); +PMU_EVENT_ATTR_STRING(c9-residency, evattr_cstate_pkg_c9, "event=0x05"); +PMU_EVENT_ATTR_STRING(c10-residency, evattr_cstate_pkg_c10, "event=0x06"); + +static struct perf_cstate_msr pkg_msr[] = { + [PERF_CSTATE_PKG_C2_RES] = { MSR_PKG_C2_RESIDENCY, &evattr_cstate_pkg_c2, test_pkg, }, + [PERF_CSTATE_PKG_C3_RES] = { MSR_PKG_C3_RESIDENCY, &evattr_cstate_pkg_c3, test_pkg, }, + [PERF_CSTATE_PKG_C6_RES] = { MSR_PKG_C6_RESIDENCY, &evattr_cstate_pkg_c6, test_pkg, }, + [PERF_CSTATE_PKG_C7_RES] = { MSR_PKG_C7_RESIDENCY, &evattr_cstate_pkg_c7, test_pkg, }, + [PERF_CSTATE_PKG_C8_RES] = { MSR_PKG_C8_RESIDENCY, &evattr_cstate_pkg_c8, test_pkg, }, + [PERF_CSTATE_PKG_C9_RES] = { MSR_PKG_C9_RESIDENCY, &evattr_cstate_pkg_c9, test_pkg, }, + [PERF_CSTATE_PKG_C10_RES] = { MSR_PKG_C10_RESIDENCY, &evattr_cstate_pkg_c10, test_pkg, }, +}; + +static struct attribute *pkg_events_attrs[PERF_CSTATE_PKG_EVENT_MAX + 1] = { + NULL, +}; + +static struct attribute_group pkg_events_attr_group = { + .name = "events", + .attrs = pkg_events_attrs, +}; + +DEFINE_CSTATE_FORMAT_ATTR(pkg_event, event, "config:0-63"); +static struct attribute *pkg_format_attrs[] = { + &format_attr_pkg_event.attr, + NULL, +}; +static struct attribute_group pkg_format_attr_group = { + .name = "format", + .attrs = pkg_format_attrs, +}; + +static cpumask_t cstate_pkg_cpu_mask; + +static const struct attribute_group *pkg_attr_groups[] = { + &pkg_events_attr_group, + &pkg_format_attr_group, + &cpumask_attr_group, + NULL, +}; + +/* cstate_pkg PMU end*/ + +static ssize_t cstate_get_attr_cpumask(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct pmu *pmu = dev_get_drvdata(dev); + + if (pmu == &cstate_core_pmu) + return cpumap_print_to_pagebuf(true, buf, &cstate_core_cpu_mask); + else if (pmu == &cstate_pkg_pmu) + return cpumap_print_to_pagebuf(true, buf, &cstate_pkg_cpu_mask); + else + return 0; +} + +static int cstate_pmu_event_init(struct perf_event *event) +{ + u64 cfg = event->attr.config; + int ret = 0; + + if (event->attr.type != event->pmu->type) + return -ENOENT; + + /* unsupported modes and filters */ + if (event->attr.exclude_user || + event->attr.exclude_kernel || + event->attr.exclude_hv || + event->attr.exclude_idle || + event->attr.exclude_host || + event->attr.exclude_guest || + event->attr.sample_period) /* no sampling */ + return -EINVAL; + + if (event->pmu == &cstate_core_pmu) { + if (cfg >= PERF_CSTATE_CORE_EVENT_MAX) + return -EINVAL; + if (!core_msr[cfg].attr) + return -EINVAL; + event->hw.event_base = core_msr[cfg].msr; + } else if (event->pmu == &cstate_pkg_pmu) { + if (cfg >= PERF_CSTATE_PKG_EVENT_MAX) + return -EINVAL; + if (!pkg_msr[cfg].attr) + return -EINVAL; + event->hw.event_base = pkg_msr[cfg].msr; + } else + return -ENOENT; + + /* must be done before validate_group */ + event->hw.config = cfg; + event->hw.idx = -1; + + return ret; +} + +static inline u64 cstate_pmu_read_counter(struct perf_event *event) +{ + u64 val; + + rdmsrl(event->hw.event_base, val); + return val; +} + +static void cstate_pmu_event_update(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + u64 prev_raw_count, new_raw_count; + +again: + prev_raw_count = local64_read(&hwc->prev_count); + new_raw_count = cstate_pmu_read_counter(event); + + if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, + new_raw_count) != prev_raw_count) + goto again; + + local64_add(new_raw_count - prev_raw_count, &event->count); +} + +static void cstate_pmu_event_start(struct perf_event *event, int mode) +{ + local64_set(&event->hw.prev_count, cstate_pmu_read_counter(event)); +} + +static void cstate_pmu_event_stop(struct perf_event *event, int mode) +{ + cstate_pmu_event_update(event); +} + +static void cstate_pmu_event_del(struct perf_event *event, int mode) +{ + cstate_pmu_event_stop(event, PERF_EF_UPDATE); +} + +static int cstate_pmu_event_add(struct perf_event *event, int mode) +{ + if (mode & PERF_EF_START) + cstate_pmu_event_start(event, mode); + + return 0; +} + +static void cstate_cpu_exit(int cpu) +{ + int i, id, target; + + /* cpu exit for cstate core */ + if (has_cstate_core) { + id = topology_core_id(cpu); + target = -1; + + for_each_online_cpu(i) { + if (i == cpu) + continue; + if (id == topology_core_id(i)) { + target = i; + break; + } + } + if (cpumask_test_and_clear_cpu(cpu, &cstate_core_cpu_mask) && target >= 0) + cpumask_set_cpu(target, &cstate_core_cpu_mask); + WARN_ON(cpumask_empty(&cstate_core_cpu_mask)); + if (target >= 0) + perf_pmu_migrate_context(&cstate_core_pmu, cpu, target); + } + + /* cpu exit for cstate pkg */ + if (has_cstate_pkg) { + id = topology_physical_package_id(cpu); + target = -1; + + for_each_online_cpu(i) { + if (i == cpu) + continue; + if (id == topology_physical_package_id(i)) { + target = i; + break; + } + } + if (cpumask_test_and_clear_cpu(cpu, &cstate_pkg_cpu_mask) && target >= 0) + cpumask_set_cpu(target, &cstate_pkg_cpu_mask); + WARN_ON(cpumask_empty(&cstate_pkg_cpu_mask)); + if (target >= 0) + perf_pmu_migrate_context(&cstate_pkg_pmu, cpu, target); + } +} + +static void cstate_cpu_init(int cpu) +{ + int i, id; + + /* cpu init for cstate core */ + if (has_cstate_core) { + id = topology_core_id(cpu); + for_each_cpu(i, &cstate_core_cpu_mask) { + if (id == topology_core_id(i)) + break; + } + if (i >= nr_cpu_ids) + cpumask_set_cpu(cpu, &cstate_core_cpu_mask); + } + + /* cpu init for cstate pkg */ + if (has_cstate_pkg) { + id = topology_physical_package_id(cpu); + for_each_cpu(i, &cstate_pkg_cpu_mask) { + if (id == topology_physical_package_id(i)) + break; + } + if (i >= nr_cpu_ids) + cpumask_set_cpu(cpu, &cstate_pkg_cpu_mask); + } +} + +static int cstate_cpu_notifier(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + unsigned int cpu = (long)hcpu; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_UP_PREPARE: + break; + case CPU_STARTING: + cstate_cpu_init(cpu); + break; + case CPU_UP_CANCELED: + case CPU_DYING: + break; + case CPU_ONLINE: + case CPU_DEAD: + break; + case CPU_DOWN_PREPARE: + cstate_cpu_exit(cpu); + break; + default: + break; + } + + return NOTIFY_OK; +} + +/* + * Probe the cstate events and insert the available one into sysfs attrs + * Return false if there is no available events. + */ +static bool cstate_probe_msr(struct perf_cstate_msr *msr, + struct attribute **events_attrs, + int max_event_nr) +{ + int i, j = 0; + u64 val; + + /* Probe the cstate events. */ + for (i = 0; i < max_event_nr; i++) { + if (!msr[i].test(i) || rdmsrl_safe(msr[i].msr, &val)) + msr[i].attr = NULL; + } + + /* List remaining events in the sysfs attrs. */ + for (i = 0; i < max_event_nr; i++) { + if (msr[i].attr) + events_attrs[j++] = &msr[i].attr->attr.attr; + } + events_attrs[j] = NULL; + + return (j > 0) ? true : false; +} + +static int __init cstate_init(void) +{ + /* SLM has different MSR for PKG C6 */ + switch (boot_cpu_data.x86_model) { + case 55: + case 76: + case 77: + pkg_msr[PERF_CSTATE_PKG_C6_RES].msr = MSR_PKG_C7_RESIDENCY; + } + + if (cstate_probe_msr(core_msr, core_events_attrs, PERF_CSTATE_CORE_EVENT_MAX)) + has_cstate_core = true; + + if (cstate_probe_msr(pkg_msr, pkg_events_attrs, PERF_CSTATE_PKG_EVENT_MAX)) + has_cstate_pkg = true; + + return (has_cstate_core || has_cstate_pkg) ? 0 : -ENODEV; +} + +static void __init cstate_cpumask_init(void) +{ + int cpu; + + cpu_notifier_register_begin(); + + for_each_online_cpu(cpu) + cstate_cpu_init(cpu); + + __perf_cpu_notifier(cstate_cpu_notifier); + + cpu_notifier_register_done(); +} + +static struct pmu cstate_core_pmu = { + .attr_groups = core_attr_groups, + .name = "cstate_core", + .task_ctx_nr = perf_invalid_context, + .event_init = cstate_pmu_event_init, + .add = cstate_pmu_event_add, /* must have */ + .del = cstate_pmu_event_del, /* must have */ + .start = cstate_pmu_event_start, + .stop = cstate_pmu_event_stop, + .read = cstate_pmu_event_update, + .capabilities = PERF_PMU_CAP_NO_INTERRUPT, +}; + +static struct pmu cstate_pkg_pmu = { + .attr_groups = pkg_attr_groups, + .name = "cstate_pkg", + .task_ctx_nr = perf_invalid_context, + .event_init = cstate_pmu_event_init, + .add = cstate_pmu_event_add, /* must have */ + .del = cstate_pmu_event_del, /* must have */ + .start = cstate_pmu_event_start, + .stop = cstate_pmu_event_stop, + .read = cstate_pmu_event_update, + .capabilities = PERF_PMU_CAP_NO_INTERRUPT, +}; + +static void __init cstate_pmus_register(void) +{ + int err; + + if (has_cstate_core) { + err = perf_pmu_register(&cstate_core_pmu, cstate_core_pmu.name, -1); + if (WARN_ON(err)) + pr_info("Failed to register PMU %s error %d\n", + cstate_core_pmu.name, err); + } + + if (has_cstate_pkg) { + err = perf_pmu_register(&cstate_pkg_pmu, cstate_pkg_pmu.name, -1); + if (WARN_ON(err)) + pr_info("Failed to register PMU %s error %d\n", + cstate_pkg_pmu.name, err); + } +} + +static int __init cstate_pmu_init(void) +{ + int err; + + if (cpu_has_hypervisor) + return -ENODEV; + + err = cstate_init(); + if (err) + return err; + + cstate_cpumask_init(); + + cstate_pmus_register(); + + return 0; +} + +device_initcall(cstate_pmu_init); diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 84f236ab96b0..5db1c7755548 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -510,10 +510,11 @@ int intel_pmu_drain_bts_buffer(void) u64 flags; }; struct perf_event *event = cpuc->events[INTEL_PMC_IDX_FIXED_BTS]; - struct bts_record *at, *top; + struct bts_record *at, *base, *top; struct perf_output_handle handle; struct perf_event_header header; struct perf_sample_data data; + unsigned long skip = 0; struct pt_regs regs; if (!event) @@ -522,10 +523,10 @@ int intel_pmu_drain_bts_buffer(void) if (!x86_pmu.bts_active) return 0; - at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; - top = (struct bts_record *)(unsigned long)ds->bts_index; + base = (struct bts_record *)(unsigned long)ds->bts_buffer_base; + top = (struct bts_record *)(unsigned long)ds->bts_index; - if (top <= at) + if (top <= base) return 0; memset(®s, 0, sizeof(regs)); @@ -535,16 +536,43 @@ int intel_pmu_drain_bts_buffer(void) perf_sample_data_init(&data, 0, event->hw.last_period); /* + * BTS leaks kernel addresses in branches across the cpl boundary, + * such as traps or system calls, so unless the user is asking for + * kernel tracing (and right now it's not possible), we'd need to + * filter them out. But first we need to count how many of those we + * have in the current batch. This is an extra O(n) pass, however, + * it's much faster than the other one especially considering that + * n <= 2560 (BTS_BUFFER_SIZE / BTS_RECORD_SIZE * 15/16; see the + * alloc_bts_buffer()). + */ + for (at = base; at < top; at++) { + /* + * Note that right now *this* BTS code only works if + * attr::exclude_kernel is set, but let's keep this extra + * check here in case that changes. + */ + if (event->attr.exclude_kernel && + (kernel_ip(at->from) || kernel_ip(at->to))) + skip++; + } + + /* * Prepare a generic sample, i.e. fill in the invariant fields. * We will overwrite the from and to address before we output * the sample. */ perf_prepare_sample(&header, &data, event, ®s); - if (perf_output_begin(&handle, event, header.size * (top - at))) + if (perf_output_begin(&handle, event, header.size * + (top - base - skip))) return 1; - for (; at < top; at++) { + for (at = base; at < top; at++) { + /* Filter out any records that contain kernel addresses. */ + if (event->attr.exclude_kernel && + (kernel_ip(at->from) || kernel_ip(at->to))) + continue; + data.ip = at->from; data.addr = at->to; diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c index b2c9475b7ff2..bfd0b717e944 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c +++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c @@ -151,10 +151,10 @@ static void __intel_pmu_lbr_enable(bool pmi) * No need to reprogram LBR_SELECT in a PMI, as it * did not change. */ - if (cpuc->lbr_sel && !pmi) { + if (cpuc->lbr_sel) lbr_select = cpuc->lbr_sel->config; + if (!pmi) wrmsrl(MSR_LBR_SELECT, lbr_select); - } rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); orig_debugctl = debugctl; @@ -555,6 +555,8 @@ static int intel_pmu_setup_sw_lbr_filter(struct perf_event *event) if (br_type & PERF_SAMPLE_BRANCH_IND_JUMP) mask |= X86_BR_IND_JMP; + if (br_type & PERF_SAMPLE_BRANCH_CALL) + mask |= X86_BR_CALL | X86_BR_ZERO_CALL; /* * stash actual user request into reg, it may * be used by fixup code for some CPU @@ -890,6 +892,7 @@ static const int snb_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { [PERF_SAMPLE_BRANCH_IND_CALL_SHIFT] = LBR_IND_CALL, [PERF_SAMPLE_BRANCH_COND_SHIFT] = LBR_JCC, [PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP, + [PERF_SAMPLE_BRANCH_CALL_SHIFT] = LBR_REL_CALL, }; static const int hsw_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { @@ -905,6 +908,7 @@ static const int hsw_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { [PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT] = LBR_REL_CALL | LBR_IND_CALL | LBR_RETURN | LBR_CALL_STACK, [PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP, + [PERF_SAMPLE_BRANCH_CALL_SHIFT] = LBR_REL_CALL, }; /* core */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_pt.c b/arch/x86/kernel/cpu/perf_event_intel_pt.c index 42169283448b..868e1194337f 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_pt.c +++ b/arch/x86/kernel/cpu/perf_event_intel_pt.c @@ -139,9 +139,6 @@ static int __init pt_pmu_hw_init(void) long i; attrs = NULL; - ret = -ENODEV; - if (!test_cpu_cap(&boot_cpu_data, X86_FEATURE_INTEL_PT)) - goto fail; for (i = 0; i < PT_CPUID_LEAVES; i++) { cpuid_count(20, i, @@ -1130,6 +1127,10 @@ static __init int pt_init(void) int ret, cpu, prior_warn = 0; BUILD_BUG_ON(sizeof(struct topa) > PAGE_SIZE); + + if (!test_cpu_cap(&boot_cpu_data, X86_FEATURE_INTEL_PT)) + return -ENODEV; + get_online_cpus(); for_each_online_cpu(cpu) { u64 ctl; diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c index 560e5255b15e..61215a69b03d 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c @@ -7,7 +7,8 @@ struct intel_uncore_type **uncore_pci_uncores = empty_uncore; static bool pcidrv_registered; struct pci_driver *uncore_pci_driver; /* pci bus to socket mapping */ -int uncore_pcibus_to_physid[256] = { [0 ... 255] = -1, }; +DEFINE_RAW_SPINLOCK(pci2phy_map_lock); +struct list_head pci2phy_map_head = LIST_HEAD_INIT(pci2phy_map_head); struct pci_dev *uncore_extra_pci_dev[UNCORE_SOCKET_MAX][UNCORE_EXTRA_PCI_DEV_MAX]; static DEFINE_RAW_SPINLOCK(uncore_box_lock); @@ -20,6 +21,59 @@ static struct event_constraint uncore_constraint_fixed = struct event_constraint uncore_constraint_empty = EVENT_CONSTRAINT(0, 0, 0); +int uncore_pcibus_to_physid(struct pci_bus *bus) +{ + struct pci2phy_map *map; + int phys_id = -1; + + raw_spin_lock(&pci2phy_map_lock); + list_for_each_entry(map, &pci2phy_map_head, list) { + if (map->segment == pci_domain_nr(bus)) { + phys_id = map->pbus_to_physid[bus->number]; + break; + } + } + raw_spin_unlock(&pci2phy_map_lock); + + return phys_id; +} + +struct pci2phy_map *__find_pci2phy_map(int segment) +{ + struct pci2phy_map *map, *alloc = NULL; + int i; + + lockdep_assert_held(&pci2phy_map_lock); + +lookup: + list_for_each_entry(map, &pci2phy_map_head, list) { + if (map->segment == segment) + goto end; + } + + if (!alloc) { + raw_spin_unlock(&pci2phy_map_lock); + alloc = kmalloc(sizeof(struct pci2phy_map), GFP_KERNEL); + raw_spin_lock(&pci2phy_map_lock); + + if (!alloc) + return NULL; + + goto lookup; + } + + map = alloc; + alloc = NULL; + map->segment = segment; + for (i = 0; i < 256; i++) + map->pbus_to_physid[i] = -1; + list_add_tail(&map->list, &pci2phy_map_head); + +end: + kfree(alloc); + return map; +} + ssize_t uncore_event_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -809,7 +863,7 @@ static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id int phys_id; bool first_box = false; - phys_id = uncore_pcibus_to_physid[pdev->bus->number]; + phys_id = uncore_pcibus_to_physid(pdev->bus); if (phys_id < 0) return -ENODEV; @@ -856,9 +910,10 @@ static void uncore_pci_remove(struct pci_dev *pdev) { struct intel_uncore_box *box = pci_get_drvdata(pdev); struct intel_uncore_pmu *pmu; - int i, cpu, phys_id = uncore_pcibus_to_physid[pdev->bus->number]; + int i, cpu, phys_id; bool last_box = false; + phys_id = uncore_pcibus_to_physid(pdev->bus); box = pci_get_drvdata(pdev); if (!box) { for (i = 0; i < UNCORE_EXTRA_PCI_DEV_MAX; i++) { diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.h b/arch/x86/kernel/cpu/perf_event_intel_uncore.h index 72c54c2e5b1a..2f0a4a98e16b 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore.h +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.h @@ -117,6 +117,15 @@ struct uncore_event_desc { const char *config; }; +struct pci2phy_map { + struct list_head list; + int segment; + int pbus_to_physid[256]; +}; + +int uncore_pcibus_to_physid(struct pci_bus *bus); +struct pci2phy_map *__find_pci2phy_map(int segment); + ssize_t uncore_event_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf); @@ -317,7 +326,8 @@ u64 uncore_shared_reg_config(struct intel_uncore_box *box, int idx); extern struct intel_uncore_type **uncore_msr_uncores; extern struct intel_uncore_type **uncore_pci_uncores; extern struct pci_driver *uncore_pci_driver; -extern int uncore_pcibus_to_physid[256]; +extern raw_spinlock_t pci2phy_map_lock; +extern struct list_head pci2phy_map_head; extern struct pci_dev *uncore_extra_pci_dev[UNCORE_SOCKET_MAX][UNCORE_EXTRA_PCI_DEV_MAX]; extern struct event_constraint uncore_constraint_empty; diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c index f78574b3cb55..845256158a10 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c @@ -420,15 +420,25 @@ static void snb_uncore_imc_event_del(struct perf_event *event, int flags) static int snb_pci2phy_map_init(int devid) { struct pci_dev *dev = NULL; - int bus; + struct pci2phy_map *map; + int bus, segment; dev = pci_get_device(PCI_VENDOR_ID_INTEL, devid, dev); if (!dev) return -ENOTTY; bus = dev->bus->number; - - uncore_pcibus_to_physid[bus] = 0; + segment = pci_domain_nr(dev->bus); + + raw_spin_lock(&pci2phy_map_lock); + map = __find_pci2phy_map(segment); + if (!map) { + raw_spin_unlock(&pci2phy_map_lock); + pci_dev_put(dev); + return -ENOMEM; + } + map->pbus_to_physid[bus] = 0; + raw_spin_unlock(&pci2phy_map_lock); pci_dev_put(dev); diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c index 694510a887dc..f0f4fcba252e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c @@ -1087,7 +1087,8 @@ static struct pci_driver snbep_uncore_pci_driver = { static int snbep_pci2phy_map_init(int devid) { struct pci_dev *ubox_dev = NULL; - int i, bus, nodeid; + int i, bus, nodeid, segment; + struct pci2phy_map *map; int err = 0; u32 config = 0; @@ -1106,16 +1107,27 @@ static int snbep_pci2phy_map_init(int devid) err = pci_read_config_dword(ubox_dev, 0x54, &config); if (err) break; + + segment = pci_domain_nr(ubox_dev->bus); + raw_spin_lock(&pci2phy_map_lock); + map = __find_pci2phy_map(segment); + if (!map) { + raw_spin_unlock(&pci2phy_map_lock); + err = -ENOMEM; + break; + } + /* * every three bits in the Node ID mapping register maps * to a particular node. */ for (i = 0; i < 8; i++) { if (nodeid == ((config >> (3 * i)) & 0x7)) { - uncore_pcibus_to_physid[bus] = i; + map->pbus_to_physid[bus] = i; break; } } + raw_spin_unlock(&pci2phy_map_lock); } if (!err) { @@ -1123,13 +1135,17 @@ static int snbep_pci2phy_map_init(int devid) * For PCI bus with no UBOX device, find the next bus * that has UBOX device and use its mapping. */ - i = -1; - for (bus = 255; bus >= 0; bus--) { - if (uncore_pcibus_to_physid[bus] >= 0) - i = uncore_pcibus_to_physid[bus]; - else - uncore_pcibus_to_physid[bus] = i; + raw_spin_lock(&pci2phy_map_lock); + list_for_each_entry(map, &pci2phy_map_head, list) { + i = -1; + for (bus = 255; bus >= 0; bus--) { + if (map->pbus_to_physid[bus] >= 0) + i = map->pbus_to_physid[bus]; + else + map->pbus_to_physid[bus] = i; + } } + raw_spin_unlock(&pci2phy_map_lock); } pci_dev_put(ubox_dev); @@ -2444,7 +2460,7 @@ static struct intel_uncore_type *bdx_pci_uncores[] = { NULL, }; -static DEFINE_PCI_DEVICE_TABLE(bdx_uncore_pci_ids) = { +static const struct pci_device_id bdx_uncore_pci_ids[] = { { /* Home Agent 0 */ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f30), .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_HA, 0), diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index c3f7602cd038..c7c4d9c51e99 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -168,21 +168,20 @@ static void cyc2ns_write_end(int cpu, struct cyc2ns_data *data) * ns = cycles * cyc2ns_scale / SC * * And since SC is a constant power of two, we can convert the div - * into a shift. + * into a shift. The larger SC is, the more accurate the conversion, but + * cyc2ns_scale needs to be a 32-bit value so that 32-bit multiplication + * (64-bit result) can be used. * - * We can use khz divisor instead of mhz to keep a better precision, since - * cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits. + * We can use khz divisor instead of mhz to keep a better precision. * (mathieu.desnoyers@polymtl.ca) * * -johnstul@us.ibm.com "math is hard, lets go shopping!" */ -#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ - static void cyc2ns_data_init(struct cyc2ns_data *data) { data->cyc2ns_mul = 0; - data->cyc2ns_shift = CYC2NS_SCALE_FACTOR; + data->cyc2ns_shift = 0; data->cyc2ns_offset = 0; data->__count = 0; } @@ -216,14 +215,14 @@ static inline unsigned long long cycles_2_ns(unsigned long long cyc) if (likely(data == tail)) { ns = data->cyc2ns_offset; - ns += mul_u64_u32_shr(cyc, data->cyc2ns_mul, CYC2NS_SCALE_FACTOR); + ns += mul_u64_u32_shr(cyc, data->cyc2ns_mul, data->cyc2ns_shift); } else { data->__count++; barrier(); ns = data->cyc2ns_offset; - ns += mul_u64_u32_shr(cyc, data->cyc2ns_mul, CYC2NS_SCALE_FACTOR); + ns += mul_u64_u32_shr(cyc, data->cyc2ns_mul, data->cyc2ns_shift); barrier(); @@ -257,12 +256,22 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) * time function is continuous; see the comment near struct * cyc2ns_data. */ - data->cyc2ns_mul = - DIV_ROUND_CLOSEST(NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR, - cpu_khz); - data->cyc2ns_shift = CYC2NS_SCALE_FACTOR; + clocks_calc_mult_shift(&data->cyc2ns_mul, &data->cyc2ns_shift, cpu_khz, + NSEC_PER_MSEC, 0); + + /* + * cyc2ns_shift is exported via arch_perf_update_userpage() where it is + * not expected to be greater than 31 due to the original published + * conversion algorithm shifting a 32-bit value (now specifies a 64-bit + * value) - refer perf_event_mmap_page documentation in perf_event.h. + */ + if (data->cyc2ns_shift == 32) { + data->cyc2ns_shift = 31; + data->cyc2ns_mul >>= 1; + } + data->cyc2ns_offset = ns_now - - mul_u64_u32_shr(tsc_now, data->cyc2ns_mul, CYC2NS_SCALE_FACTOR); + mul_u64_u32_shr(tsc_now, data->cyc2ns_mul, data->cyc2ns_shift); cyc2ns_write_end(cpu, data); diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt index 816488c0b97e..d388de72eaca 100644 --- a/arch/x86/lib/x86-opcode-map.txt +++ b/arch/x86/lib/x86-opcode-map.txt @@ -353,8 +353,12 @@ AVXcode: 1 17: vmovhps Mq,Vq (v1) | vmovhpd Mq,Vq (66),(v1) 18: Grp16 (1A) 19: -1a: BNDCL Ev,Gv | BNDCU Ev,Gv | BNDMOV Gv,Ev | BNDLDX Gv,Ev,Gv -1b: BNDCN Ev,Gv | BNDMOV Ev,Gv | BNDMK Gv,Ev | BNDSTX Ev,GV,Gv +# Intel SDM opcode map does not list MPX instructions. For now using Gv for +# bnd registers and Ev for everything else is OK because the instruction +# decoder does not use the information except as an indication that there is +# a ModR/M byte. +1a: BNDCL Gv,Ev (F3) | BNDCU Gv,Ev (F2) | BNDMOV Gv,Ev (66) | BNDLDX Gv,Ev +1b: BNDCN Gv,Ev (F2) | BNDMOV Ev,Gv (66) | BNDMK Gv,Ev (F3) | BNDSTX Ev,Gv 1c: 1d: 1e: @@ -732,6 +736,12 @@ bd: vfnmadd231ss/d Vx,Hx,Wx (66),(v),(v1) be: vfnmsub231ps/d Vx,Hx,Wx (66),(v) bf: vfnmsub231ss/d Vx,Hx,Wx (66),(v),(v1) # 0x0f 0x38 0xc0-0xff +c8: sha1nexte Vdq,Wdq +c9: sha1msg1 Vdq,Wdq +ca: sha1msg2 Vdq,Wdq +cb: sha256rnds2 Vdq,Wdq +cc: sha256msg1 Vdq,Wdq +cd: sha256msg2 Vdq,Wdq db: VAESIMC Vdq,Wdq (66),(v1) dc: VAESENC Vdq,Hdq,Wdq (66),(v1) dd: VAESENCLAST Vdq,Hdq,Wdq (66),(v1) @@ -790,6 +800,7 @@ AVXcode: 3 61: vpcmpestri Vdq,Wdq,Ib (66),(v1) 62: vpcmpistrm Vdq,Wdq,Ib (66),(v1) 63: vpcmpistri Vdq,Wdq,Ib (66),(v1) +cc: sha1rnds4 Vdq,Wdq,Ib df: VAESKEYGEN Vdq,Wdq,Ib (66),(v1) f0: RORX Gy,Ey,Ib (F2),(v) EndTable @@ -874,7 +885,7 @@ GrpTable: Grp7 2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B) | XEND (101)(11B) | XTEST (110)(11B) 3: LIDT Ms 4: SMSW Mw/Rv -5: +5: rdpkru (110),(11B) | wrpkru (111),(11B) 6: LMSW Ew 7: INVLPG Mb | SWAPGS (o64),(000),(11B) | RDTSCP (001),(11B) EndTable @@ -888,6 +899,9 @@ EndTable GrpTable: Grp9 1: CMPXCHG8B/16B Mq/Mdq +3: xrstors +4: xsavec +5: xsaves 6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3) | RDRAND Rv (11B) 7: VMPTRST Mq | VMPTRST Mq (F3) | RDSEED Rv (11B) EndTable @@ -932,8 +946,8 @@ GrpTable: Grp15 3: vstmxcsr Md (v1) | WRGSBASE Ry (F3),(11B) 4: XSAVE 5: XRSTOR | lfence (11B) -6: XSAVEOPT | mfence (11B) -7: clflush | sfence (11B) +6: XSAVEOPT | clwb (66) | mfence (11B) +7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) EndTable GrpTable: Grp16 |