A race condition occurs when one thread takes more than a second to get
scheduled to handle the signal we send to ask it to dump its stack.
When this happens, the main thread will continue on, close the fd, and
then ask the next thread to dump, but the slow thread will then wake up
and try to write to the new thread's fd, or trigger an assertion in
__linker_enable_fallback_allocator.
Do a few things to make this less bad:
- encode both target tid and fd in the shared atomic, so that we know
who each fd is for
- switch __linker_enable_fallback_allocator to return success instead
of aborting, and bail out if it's already in use
- write to the output fd right when we get to it, instead of doing it
whenever the dumping code decides to, to reduce the likelihood that
the timeout expires
Test: debuggerd_test
Change-Id: Ife0f6dae388b601e7f991605f14d7a0274013f6b
326 lines
11 KiB
C++
326 lines
11 KiB
C++
/*
|
|
* Copyright (C) 2017 The Android Open Source Project
|
|
* All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* * Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* * Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in
|
|
* the documentation and/or other materials provided with the
|
|
* distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
|
|
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
|
|
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
|
|
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
|
|
* OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
|
|
* AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
|
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
|
|
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*/
|
|
|
|
#include <dirent.h>
|
|
#include <fcntl.h>
|
|
#include <poll.h>
|
|
#include <pthread.h>
|
|
#include <stddef.h>
|
|
#include <sys/ucontext.h>
|
|
#include <syscall.h>
|
|
#include <unistd.h>
|
|
|
|
#include <atomic>
|
|
#include <memory>
|
|
|
|
#include <android-base/file.h>
|
|
#include <android-base/unique_fd.h>
|
|
#include <async_safe/log.h>
|
|
#include <backtrace/BacktraceMap.h>
|
|
#include <unwindstack/Memory.h>
|
|
#include <unwindstack/Regs.h>
|
|
|
|
#include "debuggerd/handler.h"
|
|
#include "tombstoned/tombstoned.h"
|
|
#include "util.h"
|
|
|
|
#include "libdebuggerd/backtrace.h"
|
|
#include "libdebuggerd/tombstone.h"
|
|
|
|
using android::base::unique_fd;
|
|
using unwindstack::Regs;
|
|
|
|
extern "C" bool __linker_enable_fallback_allocator();
|
|
extern "C" void __linker_disable_fallback_allocator();
|
|
|
|
// This is incredibly sketchy to do inside of a signal handler, especially when libbacktrace
|
|
// uses the C++ standard library throughout, but this code runs in the linker, so we'll be using
|
|
// the linker's malloc instead of the libc one. Switch it out for a replacement, just in case.
|
|
//
|
|
// This isn't the default method of dumping because it can fail in cases such as address space
|
|
// exhaustion.
|
|
static void debuggerd_fallback_trace(int output_fd, ucontext_t* ucontext) {
|
|
if (!__linker_enable_fallback_allocator()) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "fallback allocator already in use");
|
|
return;
|
|
}
|
|
|
|
{
|
|
std::unique_ptr<Regs> regs;
|
|
|
|
ThreadInfo thread;
|
|
thread.pid = getpid();
|
|
thread.tid = gettid();
|
|
thread.thread_name = get_thread_name(gettid());
|
|
thread.registers.reset(Regs::CreateFromUcontext(Regs::CurrentArch(), ucontext));
|
|
|
|
// TODO: Create this once and store it in a global?
|
|
std::unique_ptr<BacktraceMap> map(BacktraceMap::Create(getpid()));
|
|
dump_backtrace_thread(output_fd, map.get(), thread);
|
|
}
|
|
__linker_disable_fallback_allocator();
|
|
}
|
|
|
|
static void debuggerd_fallback_tombstone(int output_fd, ucontext_t* ucontext, siginfo_t* siginfo,
|
|
void* abort_message) {
|
|
if (!__linker_enable_fallback_allocator()) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "fallback allocator already in use");
|
|
return;
|
|
}
|
|
|
|
engrave_tombstone_ucontext(output_fd, reinterpret_cast<uintptr_t>(abort_message), siginfo,
|
|
ucontext);
|
|
__linker_disable_fallback_allocator();
|
|
}
|
|
|
|
static void iterate_siblings(bool (*callback)(pid_t, int), int output_fd) {
|
|
pid_t current_tid = gettid();
|
|
char buf[BUFSIZ];
|
|
snprintf(buf, sizeof(buf), "/proc/%d/task", current_tid);
|
|
DIR* dir = opendir(buf);
|
|
|
|
if (!dir) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "failed to open %s: %s", buf, strerror(errno));
|
|
return;
|
|
}
|
|
|
|
struct dirent* ent;
|
|
while ((ent = readdir(dir))) {
|
|
char* end;
|
|
long tid = strtol(ent->d_name, &end, 10);
|
|
if (end == ent->d_name || *end != '\0') {
|
|
continue;
|
|
}
|
|
|
|
if (tid != current_tid) {
|
|
callback(tid, output_fd);
|
|
}
|
|
}
|
|
closedir(dir);
|
|
}
|
|
|
|
static bool forward_output(int src_fd, int dst_fd, pid_t expected_tid) {
|
|
// Make sure the thread actually got the signal.
|
|
struct pollfd pfd = {
|
|
.fd = src_fd, .events = POLLIN,
|
|
};
|
|
|
|
// Wait for up to a second for output to start flowing.
|
|
if (poll(&pfd, 1, 1000) != 1) {
|
|
return false;
|
|
}
|
|
|
|
pid_t tid;
|
|
if (TEMP_FAILURE_RETRY(read(src_fd, &tid, sizeof(tid))) != sizeof(tid)) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "failed to read tid");
|
|
return false;
|
|
}
|
|
|
|
if (tid != expected_tid) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "received tid %d, expected %d", tid,
|
|
expected_tid);
|
|
return false;
|
|
}
|
|
|
|
while (true) {
|
|
char buf[512];
|
|
ssize_t rc = TEMP_FAILURE_RETRY(read(src_fd, buf, sizeof(buf)));
|
|
if (rc == 0) {
|
|
return true;
|
|
} else if (rc < 0) {
|
|
return false;
|
|
}
|
|
|
|
if (!android::base::WriteFully(dst_fd, buf, rc)) {
|
|
// We failed to write to tombstoned, but there's not much we can do.
|
|
// Keep reading from src_fd to keep things going.
|
|
continue;
|
|
}
|
|
}
|
|
}
|
|
|
|
struct __attribute__((__packed__)) packed_thread_output {
|
|
int32_t tid;
|
|
int32_t fd;
|
|
};
|
|
|
|
static uint64_t pack_thread_fd(pid_t tid, int fd) {
|
|
packed_thread_output packed = {.tid = tid, .fd = fd};
|
|
uint64_t result;
|
|
static_assert(sizeof(packed) == sizeof(result));
|
|
memcpy(&result, &packed, sizeof(packed));
|
|
return result;
|
|
}
|
|
|
|
static std::pair<pid_t, int> unpack_thread_fd(uint64_t value) {
|
|
packed_thread_output result;
|
|
memcpy(&result, &value, sizeof(value));
|
|
return std::make_pair(result.tid, result.fd);
|
|
}
|
|
|
|
static void trace_handler(siginfo_t* info, ucontext_t* ucontext) {
|
|
static std::atomic<uint64_t> trace_output(pack_thread_fd(-1, -1));
|
|
|
|
if (info->si_value.sival_int == ~0) {
|
|
// Asked to dump by the original signal recipient.
|
|
uint64_t val = trace_output.load();
|
|
auto [tid, fd] = unpack_thread_fd(val);
|
|
if (tid != gettid()) {
|
|
// We received some other thread's info request?
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc",
|
|
"thread %d received output fd for thread %d?", gettid(), tid);
|
|
return;
|
|
}
|
|
|
|
if (!trace_output.compare_exchange_strong(val, pack_thread_fd(-1, -1))) {
|
|
// Presumably, the timeout in forward_output expired, and the main thread moved on.
|
|
// If this happened, the main thread closed our fd for us, so just return.
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "cmpxchg for thread %d failed", gettid());
|
|
return;
|
|
}
|
|
|
|
// Write our tid to the output fd to let the main thread know that we're working.
|
|
if (TEMP_FAILURE_RETRY(write(fd, &tid, sizeof(tid))) == sizeof(tid)) {
|
|
debuggerd_fallback_trace(fd, ucontext);
|
|
} else {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "failed to write to output fd");
|
|
}
|
|
|
|
close(fd);
|
|
return;
|
|
}
|
|
|
|
// Only allow one thread to perform a trace at a time.
|
|
static pthread_mutex_t trace_mutex = PTHREAD_MUTEX_INITIALIZER;
|
|
int ret = pthread_mutex_trylock(&trace_mutex);
|
|
if (ret != 0) {
|
|
async_safe_format_log(ANDROID_LOG_INFO, "libc", "pthread_mutex_try_lock failed: %s",
|
|
strerror(ret));
|
|
return;
|
|
}
|
|
|
|
// Fetch output fd from tombstoned.
|
|
unique_fd tombstone_socket, output_fd;
|
|
if (!tombstoned_connect(getpid(), &tombstone_socket, &output_fd, kDebuggerdNativeBacktrace)) {
|
|
goto exit;
|
|
}
|
|
|
|
dump_backtrace_header(output_fd.get());
|
|
|
|
// Dump our own stack.
|
|
debuggerd_fallback_trace(output_fd.get(), ucontext);
|
|
|
|
// Send a signal to all of our siblings, asking them to dump their stack.
|
|
iterate_siblings(
|
|
[](pid_t tid, int output_fd) {
|
|
// Use a pipe, to be able to detect situations where the thread gracefully exits before
|
|
// receiving our signal.
|
|
unique_fd pipe_read, pipe_write;
|
|
if (!Pipe(&pipe_read, &pipe_write)) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "failed to create pipe: %s",
|
|
strerror(errno));
|
|
return false;
|
|
}
|
|
|
|
uint64_t expected = pack_thread_fd(-1, -1);
|
|
if (!trace_output.compare_exchange_strong(expected,
|
|
pack_thread_fd(tid, pipe_write.release()))) {
|
|
auto [tid, fd] = unpack_thread_fd(expected);
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc",
|
|
"thread %d is already outputting to fd %d?", tid, fd);
|
|
return false;
|
|
}
|
|
|
|
siginfo_t siginfo = {};
|
|
siginfo.si_code = SI_QUEUE;
|
|
siginfo.si_value.sival_int = ~0;
|
|
siginfo.si_pid = getpid();
|
|
siginfo.si_uid = getuid();
|
|
|
|
if (syscall(__NR_rt_tgsigqueueinfo, getpid(), tid, DEBUGGER_SIGNAL, &siginfo) != 0) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "failed to send trace signal to %d: %s",
|
|
tid, strerror(errno));
|
|
return false;
|
|
}
|
|
|
|
bool success = forward_output(pipe_read.get(), output_fd, tid);
|
|
if (!success) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc",
|
|
"timeout expired while waiting for thread %d to dump", tid);
|
|
}
|
|
|
|
// Regardless of whether the poll succeeds, check to see if the thread took fd ownership.
|
|
uint64_t post_wait = trace_output.exchange(pack_thread_fd(-1, -1));
|
|
if (post_wait != pack_thread_fd(-1, -1)) {
|
|
auto [tid, fd] = unpack_thread_fd(post_wait);
|
|
if (fd != -1) {
|
|
async_safe_format_log(ANDROID_LOG_ERROR, "libc", "closing fd %d for thread %d", fd, tid);
|
|
close(fd);
|
|
}
|
|
}
|
|
|
|
return true;
|
|
},
|
|
output_fd.get());
|
|
|
|
dump_backtrace_footer(output_fd.get());
|
|
tombstoned_notify_completion(tombstone_socket.get());
|
|
|
|
exit:
|
|
pthread_mutex_unlock(&trace_mutex);
|
|
}
|
|
|
|
static void crash_handler(siginfo_t* info, ucontext_t* ucontext, void* abort_message) {
|
|
// Only allow one thread to handle a crash at a time (this can happen multiple times without
|
|
// exit, since tombstones can be requested without a real crash happening.)
|
|
static pthread_mutex_t crash_mutex = PTHREAD_MUTEX_INITIALIZER;
|
|
int ret = pthread_mutex_lock(&crash_mutex);
|
|
if (ret != 0) {
|
|
async_safe_format_log(ANDROID_LOG_INFO, "libc", "pthread_mutex_lock failed: %s", strerror(ret));
|
|
return;
|
|
}
|
|
|
|
unique_fd tombstone_socket, output_fd;
|
|
bool tombstoned_connected =
|
|
tombstoned_connect(getpid(), &tombstone_socket, &output_fd, kDebuggerdTombstone);
|
|
debuggerd_fallback_tombstone(output_fd.get(), ucontext, info, abort_message);
|
|
if (tombstoned_connected) {
|
|
tombstoned_notify_completion(tombstone_socket.get());
|
|
}
|
|
|
|
pthread_mutex_unlock(&crash_mutex);
|
|
}
|
|
|
|
extern "C" void debuggerd_fallback_handler(siginfo_t* info, ucontext_t* ucontext,
|
|
void* abort_message) {
|
|
if (info->si_signo == DEBUGGER_SIGNAL && info->si_value.sival_int != 0) {
|
|
return trace_handler(info, ucontext);
|
|
} else {
|
|
return crash_handler(info, ucontext, abort_message);
|
|
}
|
|
}
|