blob: f1785bc999c3987872d543371d50e714139a8749 [file] [log] [blame]
// Copyright 2016 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "src/libsampler/sampler.h"
#if V8_OS_STARBOARD
#include "starboard/thread.h"
#endif // V8_OS_STARBOARD
#ifdef USE_SIGNALS
#include <errno.h>
#include <pthread.h>
#include <signal.h>
#include <sys/time.h>
#include <atomic>
#if !V8_OS_QNX && !V8_OS_AIX
#include <sys/syscall.h> // NOLINT
#endif
#if V8_OS_MACOSX
#include <mach/mach.h>
// OpenBSD doesn't have <ucontext.h>. ucontext_t lives in <signal.h>
// and is a typedef for struct sigcontext. There is no uc_mcontext.
#elif !V8_OS_OPENBSD
#include <ucontext.h>
#endif
#include <unistd.h>
#elif V8_OS_WIN || V8_OS_CYGWIN
#include "src/base/win32-headers.h"
#elif V8_OS_FUCHSIA
#include <zircon/process.h>
#include <zircon/syscalls.h>
#include <zircon/syscalls/debug.h>
#include <zircon/types.h>
// TODO(wez): Remove this once the Fuchsia SDK has rolled.
#if defined(ZX_THREAD_STATE_REGSET0)
#define ZX_THREAD_STATE_GENERAL_REGS ZX_THREAD_STATE_REGSET0
zx_status_t zx_thread_read_state(zx_handle_t h, uint32_t k, void* b, size_t l) {
uint32_t dummy_out_len = 0;
return zx_thread_read_state(h, k, b, static_cast<uint32_t>(l),
&dummy_out_len);
}
#if defined(__x86_64__)
using zx_thread_state_general_regs_t = zx_x86_64_general_regs_t;
#else
using zx_thread_state_general_regs_t = zx_arm64_general_regs_t;
#endif
#endif // !defined(ZX_THREAD_STATE_GENERAL_REGS)
#endif
#include <algorithm>
#include <vector>
#include "src/base/atomic-utils.h"
#include "src/base/platform/platform.h"
#if V8_OS_ANDROID && !defined(__BIONIC_HAVE_UCONTEXT_T)
// Not all versions of Android's C library provide ucontext_t.
// Detect this and provide custom but compatible definitions. Note that these
// follow the GLibc naming convention to access register values from
// mcontext_t.
//
// See http://code.google.com/p/android/issues/detail?id=34784
#if defined(__arm__)
using mcontext_t = struct sigcontext;
struct ucontext_t {
uint32_t uc_flags;
struct ucontext* uc_link;
stack_t uc_stack;
mcontext_t uc_mcontext;
// Other fields are not used by V8, don't define them here.
};
#elif defined(__aarch64__)
using mcontext_t = struct sigcontext;
struct ucontext_t {
uint64_t uc_flags;
struct ucontext *uc_link;
stack_t uc_stack;
mcontext_t uc_mcontext;
// Other fields are not used by V8, don't define them here.
};
#elif defined(__mips__)
// MIPS version of sigcontext, for Android bionic.
struct mcontext_t {
uint32_t regmask;
uint32_t status;
uint64_t pc;
uint64_t gregs[32];
uint64_t fpregs[32];
uint32_t acx;
uint32_t fpc_csr;
uint32_t fpc_eir;
uint32_t used_math;
uint32_t dsp;
uint64_t mdhi;
uint64_t mdlo;
uint32_t hi1;
uint32_t lo1;
uint32_t hi2;
uint32_t lo2;
uint32_t hi3;
uint32_t lo3;
};
struct ucontext_t {
uint32_t uc_flags;
struct ucontext* uc_link;
stack_t uc_stack;
mcontext_t uc_mcontext;
// Other fields are not used by V8, don't define them here.
};
#elif defined(__i386__)
// x86 version for Android.
struct mcontext_t {
uint32_t gregs[19];
void* fpregs;
uint32_t oldmask;
uint32_t cr2;
};
using kernel_sigset_t = uint32_t[2]; // x86 kernel uses 64-bit signal masks
struct ucontext_t {
uint32_t uc_flags;
struct ucontext* uc_link;
stack_t uc_stack;
mcontext_t uc_mcontext;
// Other fields are not used by V8, don't define them here.
};
enum { REG_EBP = 6, REG_ESP = 7, REG_EIP = 14 };
#elif defined(__x86_64__)
// x64 version for Android.
struct mcontext_t {
uint64_t gregs[23];
void* fpregs;
uint64_t __reserved1[8];
};
struct ucontext_t {
uint64_t uc_flags;
struct ucontext *uc_link;
stack_t uc_stack;
mcontext_t uc_mcontext;
// Other fields are not used by V8, don't define them here.
};
enum { REG_RBP = 10, REG_RSP = 15, REG_RIP = 16 };
#endif
#endif // V8_OS_ANDROID && !defined(__BIONIC_HAVE_UCONTEXT_T)
namespace v8 {
namespace sampler {
#if V8_OS_STARBOARD
class Sampler::PlatformData {
public:
PlatformData()
: thread_(SbThreadGetCurrent()),
thread_sampler_(kSbThreadSamplerInvalid) {}
~PlatformData() { ReleaseThreadSampler(); }
SbThreadSampler EnsureThreadSampler() {
if (thread_sampler_ == kSbThreadSamplerInvalid) {
thread_sampler_ = SbThreadSamplerCreate(thread_);
}
return thread_sampler_;
}
void ReleaseThreadSampler() {
if (thread_sampler_ != kSbThreadSamplerInvalid) {
SbThreadSamplerDestroy(thread_sampler_);
thread_sampler_ = kSbThreadSamplerInvalid;
}
}
private:
SbThread thread_;
SbThreadSampler thread_sampler_;
};
#endif // V8_OS_STARBOARD
#if defined(USE_SIGNALS)
AtomicGuard::AtomicGuard(AtomicMutex* atomic, bool is_blocking)
: atomic_(atomic), is_success_(false) {
do {
bool expected = false;
// We have to use the strong version here for the case where is_blocking
// is false, and we will only attempt the exchange once.
is_success_ = atomic->compare_exchange_strong(expected, true);
} while (is_blocking && !is_success_);
}
AtomicGuard::~AtomicGuard() {
if (!is_success_) return;
atomic_->store(false);
}
bool AtomicGuard::is_success() const { return is_success_; }
class Sampler::PlatformData {
public:
PlatformData() : vm_tid_(pthread_self()) {}
pthread_t vm_tid() const { return vm_tid_; }
private:
pthread_t vm_tid_;
};
void SamplerManager::AddSampler(Sampler* sampler) {
AtomicGuard atomic_guard(&samplers_access_counter_);
DCHECK(sampler->IsActive());
pthread_t thread_id = sampler->platform_data()->vm_tid();
auto it = sampler_map_.find(thread_id);
if (it == sampler_map_.end()) {
SamplerList samplers;
samplers.push_back(sampler);
sampler_map_.emplace(thread_id, std::move(samplers));
} else {
SamplerList& samplers = it->second;
auto it = std::find(samplers.begin(), samplers.end(), sampler);
if (it == samplers.end()) samplers.push_back(sampler);
}
}
void SamplerManager::RemoveSampler(Sampler* sampler) {
AtomicGuard atomic_guard(&samplers_access_counter_);
DCHECK(sampler->IsActive());
pthread_t thread_id = sampler->platform_data()->vm_tid();
auto it = sampler_map_.find(thread_id);
DCHECK_NE(it, sampler_map_.end());
SamplerList& samplers = it->second;
samplers.erase(std::remove(samplers.begin(), samplers.end(), sampler),
samplers.end());
if (samplers.empty()) {
sampler_map_.erase(it);
}
}
void SamplerManager::DoSample(const v8::RegisterState& state) {
AtomicGuard atomic_guard(&samplers_access_counter_, false);
// TODO(petermarshall): Add stat counters for the bailouts here.
if (!atomic_guard.is_success()) return;
pthread_t thread_id = pthread_self();
auto it = sampler_map_.find(thread_id);
if (it == sampler_map_.end()) return;
SamplerList& samplers = it->second;
for (Sampler* sampler : samplers) {
if (!sampler->ShouldRecordSample()) continue;
Isolate* isolate = sampler->isolate();
// We require a fully initialized and entered isolate.
if (isolate == nullptr || !isolate->IsInUse()) continue;
sampler->SampleStack(state);
}
}
SamplerManager* SamplerManager::instance() {
static base::LeakyObject<SamplerManager> instance;
return instance.get();
}
#elif V8_OS_WIN || V8_OS_CYGWIN
// ----------------------------------------------------------------------------
// Win32 profiler support. On Cygwin we use the same sampler implementation as
// on Win32.
class Sampler::PlatformData {
public:
// Get a handle to the calling thread. This is the thread that we are
// going to profile. We need to make a copy of the handle because we are
// going to use it in the sampler thread. Using GetThreadHandle() will
// not work in this case. We're using OpenThread because DuplicateHandle
// for some reason doesn't work in Chrome's sandbox.
PlatformData()
: profiled_thread_(OpenThread(THREAD_GET_CONTEXT |
THREAD_SUSPEND_RESUME |
THREAD_QUERY_INFORMATION,
false,
GetCurrentThreadId())) {}
~PlatformData() {
if (profiled_thread_ != nullptr) {
CloseHandle(profiled_thread_);
profiled_thread_ = nullptr;
}
}
HANDLE profiled_thread() { return profiled_thread_; }
private:
HANDLE profiled_thread_;
};
#elif V8_OS_FUCHSIA
class Sampler::PlatformData {
public:
PlatformData() {
zx_handle_duplicate(zx_thread_self(), ZX_RIGHT_SAME_RIGHTS,
&profiled_thread_);
}
~PlatformData() {
if (profiled_thread_ != ZX_HANDLE_INVALID) {
zx_handle_close(profiled_thread_);
profiled_thread_ = ZX_HANDLE_INVALID;
}
}
zx_handle_t profiled_thread() { return profiled_thread_; }
private:
zx_handle_t profiled_thread_ = ZX_HANDLE_INVALID;
};
#endif // USE_SIGNALS
#if defined(USE_SIGNALS)
class SignalHandler {
public:
static void IncreaseSamplerCount() {
base::MutexGuard lock_guard(mutex_.Pointer());
if (++client_count_ == 1) Install();
}
static void DecreaseSamplerCount() {
base::MutexGuard lock_guard(mutex_.Pointer());
if (--client_count_ == 0) Restore();
}
static bool Installed() {
base::MutexGuard lock_guard(mutex_.Pointer());
return signal_handler_installed_;
}
private:
static void Install() {
struct sigaction sa;
sa.sa_sigaction = &HandleProfilerSignal;
sigemptyset(&sa.sa_mask);
#if V8_OS_QNX
sa.sa_flags = SA_SIGINFO | SA_ONSTACK;
#else
sa.sa_flags = SA_RESTART | SA_SIGINFO | SA_ONSTACK;
#endif
signal_handler_installed_ =
(sigaction(SIGPROF, &sa, &old_signal_handler_) == 0);
}
static void Restore() {
if (signal_handler_installed_) {
sigaction(SIGPROF, &old_signal_handler_, nullptr);
signal_handler_installed_ = false;
}
}
static void FillRegisterState(void* context, RegisterState* regs);
static void HandleProfilerSignal(int signal, siginfo_t* info, void* context);
// Protects the process wide state below.
static base::LazyMutex mutex_;
static int client_count_;
static bool signal_handler_installed_;
static struct sigaction old_signal_handler_;
};
base::LazyMutex SignalHandler::mutex_ = LAZY_MUTEX_INITIALIZER;
int SignalHandler::client_count_ = 0;
struct sigaction SignalHandler::old_signal_handler_;
bool SignalHandler::signal_handler_installed_ = false;
void SignalHandler::HandleProfilerSignal(int signal, siginfo_t* info,
void* context) {
USE(info);
if (signal != SIGPROF) return;
v8::RegisterState state;
FillRegisterState(context, &state);
SamplerManager::instance()->DoSample(state);
}
void SignalHandler::FillRegisterState(void* context, RegisterState* state) {
// Extracting the sample from the context is extremely machine dependent.
ucontext_t* ucontext = reinterpret_cast<ucontext_t*>(context);
#if !(V8_OS_OPENBSD || \
(V8_OS_LINUX && \
(V8_HOST_ARCH_PPC || V8_HOST_ARCH_S390 || V8_HOST_ARCH_PPC64)))
mcontext_t& mcontext = ucontext->uc_mcontext;
#endif
#if V8_OS_LINUX
#if V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(mcontext.gregs[REG_EIP]);
state->sp = reinterpret_cast<void*>(mcontext.gregs[REG_ESP]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[REG_EBP]);
#elif V8_HOST_ARCH_X64
state->pc = reinterpret_cast<void*>(mcontext.gregs[REG_RIP]);
state->sp = reinterpret_cast<void*>(mcontext.gregs[REG_RSP]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[REG_RBP]);
#elif V8_HOST_ARCH_ARM
#if V8_LIBC_GLIBC && !V8_GLIBC_PREREQ(2, 4)
// Old GLibc ARM versions used a gregs[] array to access the register
// values from mcontext_t.
state->pc = reinterpret_cast<void*>(mcontext.gregs[R15]);
state->sp = reinterpret_cast<void*>(mcontext.gregs[R13]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[R11]);
state->lr = reinterpret_cast<void*>(mcontext.gregs[R14]);
#else
state->pc = reinterpret_cast<void*>(mcontext.arm_pc);
state->sp = reinterpret_cast<void*>(mcontext.arm_sp);
state->fp = reinterpret_cast<void*>(mcontext.arm_fp);
state->lr = reinterpret_cast<void*>(mcontext.arm_lr);
#endif // V8_LIBC_GLIBC && !V8_GLIBC_PREREQ(2, 4)
#elif V8_HOST_ARCH_ARM64
state->pc = reinterpret_cast<void*>(mcontext.pc);
state->sp = reinterpret_cast<void*>(mcontext.sp);
// FP is an alias for x29.
state->fp = reinterpret_cast<void*>(mcontext.regs[29]);
// LR is an alias for x30.
state->lr = reinterpret_cast<void*>(mcontext.regs[30]);
#elif V8_HOST_ARCH_MIPS
state->pc = reinterpret_cast<void*>(mcontext.pc);
state->sp = reinterpret_cast<void*>(mcontext.gregs[29]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[30]);
#elif V8_HOST_ARCH_MIPS64
state->pc = reinterpret_cast<void*>(mcontext.pc);
state->sp = reinterpret_cast<void*>(mcontext.gregs[29]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[30]);
#elif V8_HOST_ARCH_PPC || V8_HOST_ARCH_PPC64
#if V8_LIBC_GLIBC
state->pc = reinterpret_cast<void*>(ucontext->uc_mcontext.regs->nip);
state->sp =
reinterpret_cast<void*>(ucontext->uc_mcontext.regs->gpr[PT_R1]);
state->fp =
reinterpret_cast<void*>(ucontext->uc_mcontext.regs->gpr[PT_R31]);
state->lr = reinterpret_cast<void*>(ucontext->uc_mcontext.regs->link);
#else
// Some C libraries, notably Musl, define the regs member as a void pointer
state->pc = reinterpret_cast<void*>(ucontext->uc_mcontext.gp_regs[32]);
state->sp = reinterpret_cast<void*>(ucontext->uc_mcontext.gp_regs[1]);
state->fp = reinterpret_cast<void*>(ucontext->uc_mcontext.gp_regs[31]);
state->lr = reinterpret_cast<void*>(ucontext->uc_mcontext.gp_regs[36]);
#endif
#elif V8_HOST_ARCH_S390
#if V8_TARGET_ARCH_32_BIT
// 31-bit target will have bit 0 (MSB) of the PSW set to denote addressing
// mode. This bit needs to be masked out to resolve actual address.
state->pc =
reinterpret_cast<void*>(ucontext->uc_mcontext.psw.addr & 0x7FFFFFFF);
#else
state->pc = reinterpret_cast<void*>(ucontext->uc_mcontext.psw.addr);
#endif // V8_TARGET_ARCH_32_BIT
state->sp = reinterpret_cast<void*>(ucontext->uc_mcontext.gregs[15]);
state->fp = reinterpret_cast<void*>(ucontext->uc_mcontext.gregs[11]);
state->lr = reinterpret_cast<void*>(ucontext->uc_mcontext.gregs[14]);
#endif // V8_HOST_ARCH_*
#elif V8_OS_IOS || (V8_OS_STARBOARD && defined(__APPLE__))
#if V8_TARGET_ARCH_ARM64
// Building for the iOS device.
state->pc = reinterpret_cast<void*>(mcontext->__ss.__pc);
state->sp = reinterpret_cast<void*>(mcontext->__ss.__sp);
state->fp = reinterpret_cast<void*>(mcontext->__ss.__fp);
#elif V8_TARGET_ARCH_X64
// Building for the iOS simulator.
state->pc = reinterpret_cast<void*>(mcontext->__ss.__rip);
state->sp = reinterpret_cast<void*>(mcontext->__ss.__rsp);
state->fp = reinterpret_cast<void*>(mcontext->__ss.__rbp);
#else
#error Unexpected iOS target architecture.
#endif // V8_TARGET_ARCH_ARM64
#elif V8_OS_MACOSX
#if V8_HOST_ARCH_X64
state->pc = reinterpret_cast<void*>(mcontext->__ss.__rip);
state->sp = reinterpret_cast<void*>(mcontext->__ss.__rsp);
state->fp = reinterpret_cast<void*>(mcontext->__ss.__rbp);
#elif V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(mcontext->__ss.__eip);
state->sp = reinterpret_cast<void*>(mcontext->__ss.__esp);
state->fp = reinterpret_cast<void*>(mcontext->__ss.__ebp);
#elif V8_HOST_ARCH_ARM64
state->pc =
reinterpret_cast<void*>(arm_thread_state64_get_pc(mcontext->__ss));
state->sp =
reinterpret_cast<void*>(arm_thread_state64_get_sp(mcontext->__ss));
state->fp =
reinterpret_cast<void*>(arm_thread_state64_get_fp(mcontext->__ss));
#endif // V8_HOST_ARCH_*
#elif V8_OS_FREEBSD
#if V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(mcontext.mc_eip);
state->sp = reinterpret_cast<void*>(mcontext.mc_esp);
state->fp = reinterpret_cast<void*>(mcontext.mc_ebp);
#elif V8_HOST_ARCH_X64
state->pc = reinterpret_cast<void*>(mcontext.mc_rip);
state->sp = reinterpret_cast<void*>(mcontext.mc_rsp);
state->fp = reinterpret_cast<void*>(mcontext.mc_rbp);
#elif V8_HOST_ARCH_ARM
state->pc = reinterpret_cast<void*>(mcontext.__gregs[_REG_PC]);
state->sp = reinterpret_cast<void*>(mcontext.__gregs[_REG_SP]);
state->fp = reinterpret_cast<void*>(mcontext.__gregs[_REG_FP]);
#endif // V8_HOST_ARCH_*
#elif V8_OS_NETBSD
#if V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(mcontext.__gregs[_REG_EIP]);
state->sp = reinterpret_cast<void*>(mcontext.__gregs[_REG_ESP]);
state->fp = reinterpret_cast<void*>(mcontext.__gregs[_REG_EBP]);
#elif V8_HOST_ARCH_X64
state->pc = reinterpret_cast<void*>(mcontext.__gregs[_REG_RIP]);
state->sp = reinterpret_cast<void*>(mcontext.__gregs[_REG_RSP]);
state->fp = reinterpret_cast<void*>(mcontext.__gregs[_REG_RBP]);
#endif // V8_HOST_ARCH_*
#elif V8_OS_OPENBSD
#if V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(ucontext->sc_eip);
state->sp = reinterpret_cast<void*>(ucontext->sc_esp);
state->fp = reinterpret_cast<void*>(ucontext->sc_ebp);
#elif V8_HOST_ARCH_X64
state->pc = reinterpret_cast<void*>(ucontext->sc_rip);
state->sp = reinterpret_cast<void*>(ucontext->sc_rsp);
state->fp = reinterpret_cast<void*>(ucontext->sc_rbp);
#endif // V8_HOST_ARCH_*
#elif V8_OS_SOLARIS
state->pc = reinterpret_cast<void*>(mcontext.gregs[REG_PC]);
state->sp = reinterpret_cast<void*>(mcontext.gregs[REG_SP]);
state->fp = reinterpret_cast<void*>(mcontext.gregs[REG_FP]);
#elif V8_OS_QNX
#if V8_HOST_ARCH_IA32
state->pc = reinterpret_cast<void*>(mcontext.cpu.eip);
state->sp = reinterpret_cast<void*>(mcontext.cpu.esp);
state->fp = reinterpret_cast<void*>(mcontext.cpu.ebp);
#elif V8_HOST_ARCH_ARM
state->pc = reinterpret_cast<void*>(mcontext.cpu.gpr[ARM_REG_PC]);
state->sp = reinterpret_cast<void*>(mcontext.cpu.gpr[ARM_REG_SP]);
state->fp = reinterpret_cast<void*>(mcontext.cpu.gpr[ARM_REG_FP]);
#endif // V8_HOST_ARCH_*
#elif V8_OS_AIX
state->pc = reinterpret_cast<void*>(mcontext.jmp_context.iar);
state->sp = reinterpret_cast<void*>(mcontext.jmp_context.gpr[1]);
state->fp = reinterpret_cast<void*>(mcontext.jmp_context.gpr[31]);
state->lr = reinterpret_cast<void*>(mcontext.jmp_context.lr);
#endif // V8_OS_AIX
}
#endif // USE_SIGNALS
Sampler::Sampler(Isolate* isolate)
: isolate_(isolate), data_(std::make_unique<PlatformData>()) {}
Sampler::~Sampler() {
DCHECK(!IsActive());
}
void Sampler::Start() {
DCHECK(!IsActive());
SetActive(true);
#if V8_OS_STARBOARD
platform_data()->EnsureThreadSampler();
#endif // V8_OS_STARBOARD
#if defined(USE_SIGNALS)
SignalHandler::IncreaseSamplerCount();
SamplerManager::instance()->AddSampler(this);
#endif
}
void Sampler::Stop() {
#if V8_OS_STARBOARD
platform_data()->ReleaseThreadSampler();
#endif // V8_OS_STARBOARD
#if defined(USE_SIGNALS)
SamplerManager::instance()->RemoveSampler(this);
SignalHandler::DecreaseSamplerCount();
#endif
DCHECK(IsActive());
SetActive(false);
}
#if V8_OS_STARBOARD
void Sampler::DoSample() {
SbThreadSampler thread_sampler = platform_data()->EnsureThreadSampler();
SbThreadContext context = SbThreadSamplerFreeze(thread_sampler);
if (!SbThreadContextIsValid(context)) {
return;
}
v8::RegisterState state;
if (SbThreadContextGetPointer(context, kSbThreadContextInstructionPointer,
&state.pc) &&
SbThreadContextGetPointer(context, kSbThreadContextStackPointer,
&state.sp) &&
SbThreadContextGetPointer(context, kSbThreadContextFramePointer,
&state.fp)
#if SB_API_VERSION >= SB_EXPERIMENTAL_API_VERSION
&& SbThreadContextGetPointer(context, kSbThreadContextLinkRegister,
&state.lr)
#endif // SB_API_VERSION >= SB_EXPERIMENTAL_API_VERSION
) {
SampleStack(state);
}
SbThreadSamplerThaw(thread_sampler);
}
#endif // V8_OS_STARBOARD
#if defined(USE_SIGNALS)
void Sampler::DoSample() {
if (!SignalHandler::Installed()) return;
DCHECK(IsActive());
SetShouldRecordSample();
pthread_kill(platform_data()->vm_tid(), SIGPROF);
}
#elif V8_OS_WIN || V8_OS_CYGWIN
void Sampler::DoSample() {
HANDLE profiled_thread = platform_data()->profiled_thread();
if (profiled_thread == nullptr) return;
const DWORD kSuspendFailed = static_cast<DWORD>(-1);
if (SuspendThread(profiled_thread) == kSuspendFailed) return;
// Context used for sampling the register state of the profiled thread.
CONTEXT context;
memset(&context, 0, sizeof(context));
context.ContextFlags = CONTEXT_FULL;
if (GetThreadContext(profiled_thread, &context) != 0) {
v8::RegisterState state;
#if V8_HOST_ARCH_X64
state.pc = reinterpret_cast<void*>(context.Rip);
state.sp = reinterpret_cast<void*>(context.Rsp);
state.fp = reinterpret_cast<void*>(context.Rbp);
#elif V8_HOST_ARCH_ARM64
state.pc = reinterpret_cast<void*>(context.Pc);
state.sp = reinterpret_cast<void*>(context.Sp);
state.fp = reinterpret_cast<void*>(context.Fp);
#else
state.pc = reinterpret_cast<void*>(context.Eip);
state.sp = reinterpret_cast<void*>(context.Esp);
state.fp = reinterpret_cast<void*>(context.Ebp);
#endif
SampleStack(state);
}
ResumeThread(profiled_thread);
}
#elif V8_OS_FUCHSIA
void Sampler::DoSample() {
zx_handle_t profiled_thread = platform_data()->profiled_thread();
if (profiled_thread == ZX_HANDLE_INVALID) return;
zx_handle_t suspend_token = ZX_HANDLE_INVALID;
if (zx_task_suspend_token(profiled_thread, &suspend_token) != ZX_OK) return;
// Wait for the target thread to become suspended, or to exit.
// TODO(wez): There is currently no suspension count for threads, so there
// is a risk that some other caller resumes the thread in-between our suspend
// and wait calls, causing us to miss the SUSPENDED signal. We apply a 100ms
// deadline to protect against hanging the sampler thread in this case.
zx_signals_t signals = 0;
zx_status_t suspended = zx_object_wait_one(
profiled_thread, ZX_THREAD_SUSPENDED | ZX_THREAD_TERMINATED,
zx_deadline_after(ZX_MSEC(100)), &signals);
if (suspended != ZX_OK || (signals & ZX_THREAD_SUSPENDED) == 0) {
zx_handle_close(suspend_token);
return;
}
// Fetch a copy of its "general register" states.
zx_thread_state_general_regs_t thread_state = {};
if (zx_thread_read_state(profiled_thread, ZX_THREAD_STATE_GENERAL_REGS,
&thread_state, sizeof(thread_state)) == ZX_OK) {
v8::RegisterState state;
#if V8_HOST_ARCH_X64
state.pc = reinterpret_cast<void*>(thread_state.rip);
state.sp = reinterpret_cast<void*>(thread_state.rsp);
state.fp = reinterpret_cast<void*>(thread_state.rbp);
#elif V8_HOST_ARCH_ARM64
state.pc = reinterpret_cast<void*>(thread_state.pc);
state.sp = reinterpret_cast<void*>(thread_state.sp);
state.fp = reinterpret_cast<void*>(thread_state.r[29]);
#endif
SampleStack(state);
}
zx_handle_close(suspend_token);
}
// TODO(wez): Remove this once the Fuchsia SDK has rolled.
#if defined(ZX_THREAD_STATE_REGSET0)
#undef ZX_THREAD_STATE_GENERAL_REGS
#endif
#endif // USE_SIGNALS
} // namespace sampler
} // namespace v8