迷你傾印是當機時的程序記憶體傾印，可用於偵錯當機原因。
本教學課程會逐步介紹偵錯工作流程，該工作流程可將 Minidump 檔案載入 然後依照偵錯步驟逐步操作，藉此瞭解 檔案內容
擷取 minidump 檔案
你可以使用 savedump 從 zxdb 中的偵錯工作階段擷取 Minidump
指令
舉例來說，如果您在 cobalt.cm 元件上使用 zxdb：
ffx component debug cobalt.cm您應該會看到類似下方的輸出內容：
Waiting for process matching "job 20999".
Type "filter" to see the current filters.
👉 To get started, try "status" or "help".
Attached Process 1 state=Running koid=21246 name=cobalt.cm component=cobalt.cm
Loading 15 modules for cobalt.cm ...Done.
如果您想擷取迷你轉儲，請先 pause 偵錯工具：
pause
您應該會看到類似以下的輸出內容：
508 const zx_port_packet_t* packet))
509
▶ 510 BLOCKING_SYSCALL(port_wait, zx_status_t, /* no attributes */, 3, (handle, deadline, packet),
511 (_ZX_SYSCALL_ANNO(use_handle("Fuchsia")) zx_handle_t handle, zx_time_t deadline,
512 zx_port_packet_t* packet))
🛑 thread 1 $elf(SYSCALL_zx_port_wait) + 0x7 • syscalls.inc:510
使用 savedump 儲存迷你轉儲：
savedump cobalt.dump
您應該會看到類似下方的輸出內容：
Saving minidump...
Minidump written to cobalt.dump
儲存迷你轉儲記錄後，您可以退出偵錯工具：
quit
您的 minidump 現已儲存在工作目錄中。您現在可以載入 minidump 檔案並進行偵錯。
載入 Minidump 檔案並進行偵錯
無論您嘗試對元件進行偵錯的語言為何，在 zxdb 中載入及偵錯迷你傾印檔案的程序都相同。不過 每種語言要留意的事項差異
荒漠油廠
如要將 minidump 檔案載入 zxdb，您可以使用
ffx debug core。例如，如要載入名為archivist_unittest.dump的迷你轉儲檔案：ffx debug core archivist_unittest.dump您應該會看到類似以下的輸出內容：
Opening dump file... Dump loaded successfully. 👉 To get started, try "status" or "help". Attached Process 1 state=Running koid=14446435 name=<_> Loading 6 modules for <_> Downloading symbols... Done. 🛑 (no location information) Attached Process 2 state=Running koid=14446435 name=<_> Attaching to previously connected processes: 14446435: <_> Loading 6 modules for <_> Done. Symbol downloading complete. 5 succeeded, 0 failed.列出堆疊框架：
frame您應該會看到類似以下的輸出內容：
▶ 0 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::test_entry_point::λ(…) • archivist.rs:547 1 core::future::future::«impl»::poll<…>(…) • future/future.rs:123 2 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:27 3 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:122 4 fuchsia_async::atomic_future::«impl»::poll<…>(…) • atomic_future.rs:72 5 fuchsia_async::atomic_future::AtomicFuture::try_poll(…) • atomic_future.rs:230 6 fuchsia_async::runtime::fuchsia::executor::common::Executor::try_poll(…) • executor/common.rs:554 7 fuchsia_async::runtime::fuchsia::executor::common::Executor::poll_ready_tasks(…) • executor/common.rs:133 8 fuchsia_async::runtime::fuchsia::executor::common::Executor::worker_lifecycle<…>(…) • executor/common.rs:429 9 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run<…>(…) • executor/local.rs:104 10 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run_singlethreaded<…>(…) • executor/local.rs:70 11 fuchsia_async::test_support::«impl»::run_singlethreaded::λ() • test_support.rs:120 12 fuchsia_async::test_support::Config::in_parallel(…) • test_support.rs:215 13 fuchsia_async::test_support::«impl»::run_singlethreaded(…) • test_support.rs:117 14 fuchsia_async::test_support::run_singlethreaded_test<…>(…) • test_support.rs:227 15 fuchsia::test_singlethreaded<…>(…) • fuchsia/src/lib.rs:195 16 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log() • archivist.rs:524 17 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::λ(…) • archivist.rs:525 18 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 19 core::ops::function::FnOnce::call_once<…>(…) • library/core/src/ops/function.rs:250 (inline) 20 test::__rust_begin_short_backtrace<…>(…) • library/test/src/lib.rs:625 21 test::run_test_in_spawned_subprocess(…) • library/test/src/lib.rs:753 22 test::test_main_static_abort(…) • library/test/src/lib.rs:199 23 archivist_lib_lib_test::main() • archivist/src/lib.rs:1 24 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 25 std::sys::backtrace::__rust_begin_short_backtrace<…>(…) • fuchsia-third_party-rust/library/std/src/sys/backtrace.rs:155 26 std::rt::lang_start::λ() • fuchsia-third_party-rust/library/std/src/rt.rs:159 27 core::ops::function::impls::«impl»::call_once<…>(…) • library/core/src/ops/function.rs:284 (inline) 28 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) 29 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 30 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 31 std::rt::lang_start_internal::λ() • library/std/src/rt.rs:141 (inline) 32 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) 33 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 34 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 35 std::rt::lang_start_internal(…) • library/std/src/rt.rs:141 36 std::rt::lang_start<…>(…) • fuchsia-third_party-rust/library/std/src/rt.rs:158 37 $elf(main) + 0x21 38…40 «libc startup» (-r expands)你現在可以使用快速鍵在所有執行緒中快速執行動作。
列出每個執行緒的框架：
thread * frame您應該會看到類似以下的輸出內容：
Thread 1 state="Core Dump" koid=14450503 name="" ▶ 0 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::test_entry_point::λ(…) • archivist.rs:547您可以將
thread與backtrace動詞結合，列出每個執行緒的框架，但會提供更詳細的資訊：thread * backtrace這個指令可協助您查看每個堆疊框架的本機變數。請注意，為了縮短擷取時間並減少擷取檔案大小，迷你轉儲檔案只會擷取堆疊記憶體。堆積變數的指標大多不會解析為任何有用的內容。
您應該會看到類似下方的輸出內容：
Thread 1 state="Core Dump" koid=14450503 name="" ▶ 0 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::test_entry_point::λ(…) • archivist.rs:547 (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 1 core::future::future::«impl»::poll<…>(…) • future/future.rs:123 self = Pin<&mut core::pin::Pin<alloc…>{__pointer: (*)0xa2987a4580} cx = (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 2 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:27 (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 3 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:122 (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 4 fuchsia_async::atomic_future::«impl»::poll<…>(…) • atomic_future.rs:72 self = (*)0xa2987a4520 cx = (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 5 fuchsia_async::atomic_future::AtomicFuture::try_poll(…) • atomic_future.rs:230 self = (*)0x9d587a4118 cx = (*)0x5c626904e0 ➔ Context{waker: (*)0x5c626904c8, local_waker: (*)0x5c626904c8, ext: AssertUnwindSafe<core::task::wake::ExtData>(…), _marker: PhantomData<fn(&())->&()>, _marker2: PhantomData<*mut()>} 6 fuchsia_async::runtime::fuchsia::executor::common::Executor::try_poll(…) • executor/common.rs:554 self = (*)0x9b187a9300 task = (*)0x5c62690588 ➔ (*)0x9d587a40f0 7 fuchsia_async::runtime::fuchsia::executor::common::Executor::poll_ready_tasks(…) • executor/common.rs:133 self = (*)0x9b187a9300 local_collector = (*)0x5c62690600 ➔ LocalCollector{collector: (*)0x9b187a9438, last_ticks: 1050997948718604, polls: 1, tasks_pending_max: 10} 8 fuchsia_async::runtime::fuchsia::executor::common::Executor::worker_lifecycle<…>(…) • executor/common.rs:429 self = (*)0xa4187a90a0 9 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run<…>(…) • executor/local.rs:104 self = (*)0x5c62690878 ➔ LocalExecutor{ehandle: EHandle{…}} main_future = AtomicFuture{state: AtomicUsize{1}, future: UnsafeCell{$(alloc::boxed::Box<dyn fuchsia_async::atomic_future::FutureOrResultAccess, alloc::alloc::Global>){…}}} 10 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run_singlethreaded<…>(…) • executor/local.rs:70 self = (*)0x5c62690878 ➔ LocalExecutor{ehandle: EHandle{…}} main_future = Unresumed{run_stream: (*)0xa3187a86a0, test: λ{…}} 11 fuchsia_async::test_support::«impl»::run_singlethreaded::λ() • test_support.rs:120 12 fuchsia_async::test_support::Config::in_parallel(…) • test_support.rs:215 self = (*)0x5c62690ae8 ➔ Config{repeat_count: 1, max_concurrency: 0, max_threads: 0, timeout: …} f = <The value of type '*const alloc::sync::ArcInner<(dyn core::ops::function::Fn<(), Output=()> + core::marker::Send + core::marker::Sync)>' is the incorrect size (expecting 8, got 16). Please file a bug.> 13 fuchsia_async::test_support::«impl»::run_singlethreaded(…) • test_support.rs:117 test = <The value of type '*const alloc::sync::ArcInner<(dyn core::ops::function::Fn<(usize), Output=core::pin::Pin<alloc::boxed::Box<dyn core::future::future::Future<Output=()>, alloc::alloc::Global>>> + core::marker::Send + core::marker::Sync)>' is the incorrect size (expecting 8, got 16). Please file a bug.> cfg = Config{repeat_count: 1, max_concurrency: 0, max_threads: 0, timeout: None<Duration>} 14 fuchsia_async::test_support::run_singlethreaded_test<…>(…) • test_support.rs:227 test = <Unavailable> 15 fuchsia::test_singlethreaded<…>(…) • fuchsia/src/lib.rs:195 f = <Register rdi not available.> 16 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log() • archivist.rs:524 17 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::λ(…) • archivist.rs:525 (*)0x5c62690bde ➔ λ 18 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 λ <Value has no data.> 19 core::ops::function::FnOnce::call_once<…>(…) • library/core/src/ops/function.rs:250 (inline) <Register rsi not available.> <Optimized out> 20 test::__rust_begin_short_backtrace<…>(…) • library/test/src/lib.rs:625 f = <Register rsi not available.> 21 test::run_test_in_spawned_subprocess(…) • library/test/src/lib.rs:753 desc = <Register rdi not available.> runnable_test = Static(&core::ops::function::FnOnce::call_once<archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::{closure_env#0}, ()>) 22 test::test_main_static_abort(…) • library/test/src/lib.rs:199 tests = <Unavailable> 23 archivist_lib_lib_test::main() • archivist/src/lib.rs:1 24 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 &archivist_lib_lib_test::main <Value has no data.> 25 std::sys::backtrace::__rust_begin_short_backtrace<…>(…) • fuchsia-third_party-rust/library/std/src/sys/backtrace.rs:155 f = &archivist_lib_lib_test::main 26 std::rt::lang_start::λ() • fuchsia-third_party-rust/library/std/src/rt.rs:159 27 core::ops::function::impls::«impl»::call_once<…>(…) • library/core/src/ops/function.rs:284 (inline) self = $(&(dyn core::ops::function::Fn<(), Output=i32> + core::marker::Sync + core::panic::unwind_safe::RefUnwindSafe)){pointer: (*)0x5c62690f50 ➔ $((dyn core::ops::function::Fn<(), Output=i32> + core::marker::Sync + core::panic::unwind_safe::RefUnwindSafe)), vtable = <Invalid data offset 8 in object of size 8.>} args = <Optimized out> 28 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) data = <Optimized out> 29 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 30 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 31 std::rt::lang_start_internal::λ() • library/std/src/rt.rs:141 (inline) 32 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) data = <Optimized out> 33 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 34 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 35 std::rt::lang_start_internal(…) • library/std/src/rt.rs:141 main = $(&(dyn core::ops::function::Fn<(), Output=i32> + core::marker::Sync + core::panic::unwind_safe::RefUnwindSafe)){pointer: (*)0x5c62690f50 ➔ $((dyn core::ops::function::Fn<(), Output=i32> + core::marker::Sync + core::panic::unwind_safe::RefUnwindSafe)), vtable: (*)0x13a66000958} argc = <Register rdx not available.> argv = <Register rcx not available.> sigpipe = <Register r8 not available.> 36 std::rt::lang_start<…>(…) • fuchsia-third_party-rust/library/std/src/rt.rs:158 main = &archivist_lib_lib_test::main argc = 2 argv = (*)0xecf9d19fb0 sigpipe = 0 37 $elf(main) + 0x21 38…40 «libc startup» (-r expands)列出目前的執行緒：
thread您應該會看到類似以下的輸出內容：
# state koid name ▶ 1 Core Dump 14450503列出特定執行緒。例如
thread 1：thread 1您應該會看到類似以下的輸出內容：
Thread 1 state="Core Dump" koid=14450503 name=""接著，您可以使用
frame查看thread 1中的這個特定堆疊框架：frame您應該會看到類似以下的輸出內容：
▶ 0 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::test_entry_point::λ(…) • archivist.rs:547 1 core::future::future::«impl»::poll<…>(…) • future/future.rs:123 2 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:27 3 fuchsia_async::test_support::«impl»::run_singlethreaded::λ::λ(…) • test_support.rs:122 4 fuchsia_async::atomic_future::«impl»::poll<…>(…) • atomic_future.rs:72 5 fuchsia_async::atomic_future::AtomicFuture::try_poll(…) • atomic_future.rs:230 6 fuchsia_async::runtime::fuchsia::executor::common::Executor::try_poll(…) • executor/common.rs:554 7 fuchsia_async::runtime::fuchsia::executor::common::Executor::poll_ready_tasks(…) • executor/common.rs:133 8 fuchsia_async::runtime::fuchsia::executor::common::Executor::worker_lifecycle<…>(…) • executor/common.rs:429 9 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run<…>(…) • executor/local.rs:104 10 fuchsia_async::runtime::fuchsia::executor::local::LocalExecutor::run_singlethreaded<…>(…) • executor/local.rs:70 11 fuchsia_async::test_support::«impl»::run_singlethreaded::λ() • test_support.rs:120 12 fuchsia_async::test_support::Config::in_parallel(…) • test_support.rs:215 13 fuchsia_async::test_support::«impl»::run_singlethreaded(…) • test_support.rs:117 14 fuchsia_async::test_support::run_singlethreaded_test<…>(…) • test_support.rs:227 15 fuchsia::test_singlethreaded<…>(…) • fuchsia/src/lib.rs:195 16 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log() • archivist.rs:524 17 archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::λ(…) • archivist.rs:525 18 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 19 core::ops::function::FnOnce::call_once<…>(…) • library/core/src/ops/function.rs:250 (inline) 20 test::__rust_begin_short_backtrace<…>(…) • library/test/src/lib.rs:625 21 test::run_test_in_spawned_subprocess(…) • library/test/src/lib.rs:753 22 test::test_main_static_abort(…) • library/test/src/lib.rs:199 23 archivist_lib_lib_test::main() • archivist/src/lib.rs:1 24 core::ops::function::FnOnce::call_once<…>(…) • fuchsia-third_party-rust/library/core/src/ops/function.rs:250 25 std::sys::backtrace::__rust_begin_short_backtrace<…>(…) • fuchsia-third_party-rust/library/std/src/sys/backtrace.rs:155 26 std::rt::lang_start::λ() • fuchsia-third_party-rust/library/std/src/rt.rs:159 27 core::ops::function::impls::«impl»::call_once<…>(…) • library/core/src/ops/function.rs:284 (inline) 28 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) 29 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 30 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 31 std::rt::lang_start_internal::λ() • library/std/src/rt.rs:141 (inline) 32 std::panicking::try::do_call<…>(…) • library/std/src/panicking.rs:553 (inline) 33 std::panicking::try<…>() • library/std/src/panicking.rs:517 (inline) 34 std::panic::catch_unwind<…>() • library/std/src/panic.rs:350 (inline) 35 std::rt::lang_start_internal(…) • library/std/src/rt.rs:141 36 std::rt::lang_start<…>(…) • fuchsia-third_party-rust/library/std/src/rt.rs:158 37 $elf(main) + 0x21 38…40 «libc startup» (-r expands)根據上一個步驟的輸出內容，您可能需要深入瞭解
frame 0：frame 0您應該會看到類似下方的輸出內容：
archivist_lib_lib_test::archivist::tests::can_log_and_retrieve_log::test_entry_point::λ(…) • archivist.rs:547如要查看目前頁框的其他幾行程式碼，請使用
list：list您應該會看到類似下方的輸出內容：
542 543 let mut expected = vec!["my msg1".to_owned(), "my msg2".to_owned()]; 544 expected.sort(); 545 546 let mut actual = vec![recv_logs.next().await.unwrap(), recv_logs.next().await.unwrap()]; ▶ 547 actual.sort(); 548 549 assert_eq!(expected, actual); 550 551 // can log after killing log sink proxy 552 log_helper.kill_log_sink(); 553 log_helper.write_log("my msg1"); 554 log_helper.write_log("my msg2"); 555 556 assert_eq!( 557 expected,這個框架實際上含有一些封存主義程式碼，因此您可以使用
locals可查看這個堆疊框架中的本機變數。如要查看目前堆疊框架中的所有本機變數，請使用
locals：locals您應該會看到類似以下的輸出內容：
_task_context = (*)0x5c626904e0 ➔ Context{ waker: (*)0x5c626904c8 local_waker: (*)0x5c626904c8 ext: AssertUnwindSafe<core::task::wake::ExtData>(None(<Value has no data.>)) _marker: PhantomData<fn(&())->&()> _marker2: PhantomData<*mut()> } actual = <Invalid pointer 0x9c187a04e8> directory = <Invalid pointer 0x9c187a0460> expected = <Invalid pointer 0x9c187a04d0> log_helper = <Invalid pointer 0x9c187a0470> log_helper2 = <Invalid pointer 0x9c187a04c0> recv_logs = <Invalid pointer 0x9c187a0468>接著，您可以使用
print進一步瞭解特定變數的相關資訊。例如：print _task_context您應該會看到類似下方的輸出內容：
(*)0x5c626904e0 ➔ Context{ waker: (*)0x5c626904c8 local_waker: (*)0x5c626904c8 ext: AssertUnwindSafe<core::task::wake::ExtData>(None(<Value has no data.>)) _marker: PhantomData<fn(&())->&()> _marker2: PhantomData<*mut()> }
您已成功載入並分析 Rust 程式碼的 Minidump。
C++
如要將 Minidump 檔案載入 zxdb，可以使用
ffx debug core。 例如，如要載入名為cobalt_minidump.dump的 Minidump 檔案：ffx debug core cobalt_minidump.dump您應該會看到類似以下的輸出內容：
Opening dump file... Dump loaded successfully. 👉 To get started, try "status" or "help". Attached Process 1 state=Running koid=15650768 name=<_> Loading 15 modules for <_> ...Done. Attached Process 2 state=Running koid=15650768 name=<_> Attaching to previously connected processes: 15650768: <_> Loading 15 modules for <_> Done.載入迷你快照檔案後，您現在可以檢查迷你快照擷取時顯示的所有執行緒的堆疊。
列出堆疊框架：
frame您應該會看到類似下方的輸出內容：
▶ 0…4 «Waiting for event in async::Loop::Run()» (-r expands) 5 main(…) • cobalt_main.cc:349 6…8 «libc startup» (-r expands)您現在可以使用快速鍵在所有執行緒中快速執行動作：
列出每個執行緒的框架：
thread * frame您應該會看到類似以下的輸出內容：
Thread 1 state="Core Dump" koid=19601 name="" ▶ 0…4 «Waiting for event in async::Loop::Run()» (-r expands) 5 main(…) • cobalt_main.cc:349 6…8 «libc startup» (-r expands) Thread 2 state="Core Dump" koid=26944 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 2 __timedwait_assign_owner(…) • __timedwait.c:23 3 __timedwait(…) • threads_impl.h:322 (inline) 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 5 std::__2::__libcpp_condvar_timedwait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:127 (inline) 6 std::__2::condition_variable::__do_timed_wait(…) • condition_variable.cpp:54 7 std::__2::condition_variable::wait_for<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:196 8 std::__2::condition_variable::__do_timed_wait<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:235 9 std::__2::condition_variable::wait_until<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:161 10 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:240 11 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:247 12 std::__2::condition_variable_any::wait_for<…>(…) • condition_variable:260 13 cobalt::local_aggregation::DelayedLocalAggregateStorage::Run(…) • delayed_local_aggregate_storage.cc:200 14 λ(…) • delayed_local_aggregate_storage.cc:45 15 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 16 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 17 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 18…19 «pthread startup» (-r expands) Thread 3 state="Core Dump" koid=26975 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 2 __timedwait_assign_owner(…) • __timedwait.c:23 3 __timedwait(…) • threads_impl.h:322 (inline) 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 5 std::__2::__libcpp_condvar_wait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:122 (inline) 6 std::__2::condition_variable::wait(…) • condition_variable.cpp:30 7 std::__2::condition_variable_any::wait<…>(…) • condition_variable:225 8 std::__2::condition_variable_any::wait<…>(…) • condition_variable:231 9 cobalt::uploader::ShippingManager::Run(…) • shipping_manager.cc:217 10 λ(…) • shipping_manager.cc:76 11 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 12 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 13 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 14…15 «pthread startup» (-r expands) Thread 4 state="Core Dump" koid=26983 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 2 __timedwait_assign_owner(…) • __timedwait.c:23 3 __timedwait(…) • threads_impl.h:322 (inline) 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 5 std::__2::__libcpp_condvar_timedwait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:127 (inline) 6 std::__2::condition_variable::__do_timed_wait(…) • condition_variable.cpp:54 7 std::__2::condition_variable::wait_for<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:196 8 std::__2::condition_variable::__do_timed_wait<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:235 9 std::__2::condition_variable::wait_until<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:161 10 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:240 11 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:247 12 cobalt::local_aggregation::ObservationGenerator::Run(…) • observation_generator.cc:92 13 λ(…) • observation_generator.cc:65 14 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 15 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 16 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 17…18 «pthread startup» (-r expands)如要列出每個執行緒的幀，並提供更詳細的資訊，您可以將
thread與backtrace動詞結合：thread * backtrace這個指令可協助您查看每個堆疊框架的本機變數。保留 請記住，只有堆疊記憶體會擷取至 Minidump 中，以便改善 擷取的時間和大小堆積變數的指標大多不會解析為任何實用的內容，這些指標的格式類似
<Invalid pointer 0x285f3fd5a18>。您應該會看到類似下方的輸出內容：
Thread 1 state="Core Dump" koid=19601 name="" ▶ 0…4 «Waiting for event in async::Loop::Run()» (-r expands) 5 main(…) • cobalt_main.cc:349 argc = 2 argv = (*)0x26866821fc0 6…8 «libc startup» (-r expands) Thread 2 state="Core Dump" koid=26944 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 value_ptr = (*)0x285f3fd5904 current_value = 2 new_futex_owner = 0 deadline = 612921722073211 2 __timedwait_assign_owner(…) • __timedwait.c:23 futex = <Register rdi not available.> val = <Register rsi not available.> clk = <Register rdx not available.> at = <Register rcx not available.> new_owner = <Register r8 not available.> 3 __timedwait(…) • threads_impl.h:322 (inline) futex = (*)0x285f3fd5904 val = 2 clk = 0 at = (*)0x285f3fd5918 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 c = (*)0x370ff24ae20 m = (*)0x3707f2b76c8 ts = (*)0x285f3fd5918 5 std::__2::__libcpp_condvar_timedwait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:127 (inline) __cv = <Register rdi not available.> __m = <Register rsi not available.> __ts = <Optimized out> 6 std::__2::condition_variable::__do_timed_wait(…) • condition_variable.cpp:54 this = <Register rdi not available.> lk = <Register rsi not available.> tp = <Register rdx not available.> 7 std::__2::condition_variable::wait_for<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:196 this = (*)0x370ff24ae20 __lk = <Invalid pointer 0x285f3fd5a18> __d = <Invalid pointer 0x285f3fd59a8> 8 std::__2::condition_variable::__do_timed_wait<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:235 this = (*)0x370ff24ae20 __lk = <Invalid pointer 0x285f3fd5a18> __tp = <Invalid pointer 0x285f3fd59b0> 9 std::__2::condition_variable::wait_until<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:161 this = (*)0x370ff24ae20 __lk = <Invalid pointer 0x285f3fd5a18> __t = <Invalid pointer 0x285f3fd5a60> 10 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:240 this = (*)0x370ff24ae20 __lock = <Invalid pointer 0x285f3fd5ab0> __t = <Invalid pointer 0x285f3fd5a60> 11 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:247 this = (*)0x370ff24ae20 __lock = <Invalid pointer 0x285f3fd5ab0> __t = <Invalid pointer 0x285f3fd5a60> __pred = <Invalid pointer 0x285f3fd5a40> 12 std::__2::condition_variable_any::wait_for<…>(…) • condition_variable:260 this = (*)0x370ff24ae20 __lock = <Invalid pointer 0x285f3fd5ab0> __d = <Invalid pointer 0x370ff24ae00> __pred = {locked_state = <Invalid pointer 0x285f3fd5ab0>} 13 cobalt::local_aggregation::DelayedLocalAggregateStorage::Run(…) • delayed_local_aggregate_storage.cc:200 this = (*)0x370ff24acf0 14 λ(…) • delayed_local_aggregate_storage.cc:45 this = (*)0x36bff27b228 15 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 __f = <Invalid pointer 0x36bff27b228> 16 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 __t = <Invalid pointer 0x36bff27b220> {} 17 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 __vp = (*)0x36bff27b220 18…19 «pthread startup» (-r expands) Thread 3 state="Core Dump" koid=26975 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 value_ptr = (*)0x23b6ad77434 current_value = 2 new_futex_owner = 0 deadline = 9223372036854775807 2 __timedwait_assign_owner(…) • __timedwait.c:23 futex = <Register rdi not available.> val = <Register rsi not available.> clk = <Register rdx not available.> at = <Register rcx not available.> new_owner = <Register r8 not available.> 3 __timedwait(…) • threads_impl.h:322 (inline) futex = (*)0x23b6ad77434 val = 2 clk = 0 at = (*)0x0 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 c = (*)0x3673f24a610 m = (*)0x3707f2b7998 ts = (*)0x0 5 std::__2::__libcpp_condvar_wait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:122 (inline) __cv = <Register rdi not available.> __m = <Register rsi not available.> 6 std::__2::condition_variable::wait(…) • condition_variable.cpp:30 this = <Register rdi not available.> lk = <Register rsi not available.> 7 std::__2::condition_variable_any::wait<…>(…) • condition_variable:225 this = (*)0x3673f24a610 __lock = <Invalid pointer 0x23b6ad77608> 8 std::__2::condition_variable_any::wait<…>(…) • condition_variable:231 this = (*)0x3673f24a610 __lock = <Invalid pointer 0x23b6ad77608> __pred = <Invalid pointer 0x23b6ad77480> 9 cobalt::uploader::ShippingManager::Run(…) • shipping_manager.cc:217 this = (*)0x3673f24a530 10 λ(…) • shipping_manager.cc:76 this = (*)0x36bff27a7a8 11 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 __f = <Invalid pointer 0x36bff27a7a8> 12 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 __t = <Invalid pointer 0x36bff27a7a0> {} 13 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 __vp = (*)0x36bff27a7a0 14…15 «pthread startup» (-r expands) Thread 4 state="Core Dump" koid=26983 name="" ▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 value_ptr = (*)0x149c84b2bc4 current_value = 2 new_futex_owner = 0 deadline = 615784347537787 2 __timedwait_assign_owner(…) • __timedwait.c:23 futex = <Register rdi not available.> val = <Register rsi not available.> clk = <Register rdx not available.> at = <Register rcx not available.> new_owner = <Register r8 not available.> 3 __timedwait(…) • threads_impl.h:322 (inline) futex = (*)0x149c84b2bc4 val = 2 clk = 0 at = (*)0x149c84b2bd8 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 c = (*)0x3673f24b128 m = (*)0x3707f2b6548 ts = (*)0x149c84b2bd8 5 std::__2::__libcpp_condvar_timedwait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:127 (inline) __cv = <Register rdi not available.> __m = <Register rsi not available.> __ts = <Optimized out> 6 std::__2::condition_variable::__do_timed_wait(…) • condition_variable.cpp:54 this = <Register rdi not available.> lk = <Register rsi not available.> tp = <Register rdx not available.> 7 std::__2::condition_variable::wait_for<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:196 this = (*)0x3673f24b128 __lk = <Invalid pointer 0x149c84b2cd8> __d = <Invalid pointer 0x149c84b2c68> 8 std::__2::condition_variable::__do_timed_wait<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:235 this = (*)0x3673f24b128 __lk = <Invalid pointer 0x149c84b2cd8> __tp = <Invalid pointer 0x149c84b2c70> 9 std::__2::condition_variable::wait_until<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:161 this = (*)0x3673f24b128 __lk = <Invalid pointer 0x149c84b2cd8> __t = <Invalid pointer 0x3673f24b0d8> 10 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:240 this = (*)0x3673f24b128 __lock = <Invalid pointer 0x149c84b2d20> __t = <Invalid pointer 0x3673f24b0d8> 11 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:247 this = (*)0x3673f24b128 __lock = <Invalid pointer 0x149c84b2d20> __t = <Invalid pointer 0x3673f24b0d8> __pred = <Invalid pointer 0x149c84b2d00> 12 cobalt::local_aggregation::ObservationGenerator::Run(…) • observation_generator.cc:92 this = (*)0x3673f24b0a0 clock = (*)0x36bff27aaa0 13 λ(…) • observation_generator.cc:65 this = (*)0x36bff24cf68 14 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 __f = <Invalid pointer 0x36bff24cf68> 15 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 __t = <Invalid pointer 0x36bff24cf60> {} 16 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 __vp = (*)0x36bff24cf60 17…18 «pthread startup» (-r expands)列出目前的執行緒：
thread您應該會看到類似以下的輸出內容：
# state koid name ▶ 1 Core Dump 19601 2 Core Dump 26944 3 Core Dump 26975 4 Core Dump 26983列出特定執行緒。例如
thread 2：thread 2您應該會看到類似下方的輸出內容：
Thread 2 state="Core Dump" koid=26944 name=""接著，您可以使用
frame查看thread 2中的這個特定堆疊框架：frame您應該會看到類似下方的輸出內容：
▶ 0 $elf(CODE_SYSCALL_zx_futex_wait) + 0xa • syscalls.inc:210 1 _zx_futex_wait(…) • syscalls.inc:210 2 __timedwait_assign_owner(…) • __timedwait.c:23 3 __timedwait(…) • threads_impl.h:322 (inline) 4 pthread_cond_timedwait(…) • pthread_cond_timedwait.c:78 5 std::__2::__libcpp_condvar_timedwait(…) • stage2-bins/include/c++/v1/__thread/support/pthread.h:127 (inline) 6 std::__2::condition_variable::__do_timed_wait(…) • condition_variable.cpp:54 7 std::__2::condition_variable::wait_for<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:196 8 std::__2::condition_variable::__do_timed_wait<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:235 9 std::__2::condition_variable::wait_until<…>(…) • linux-x64/include/c++/v1/__condition_variable/condition_variable.h:161 10 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:240 11 std::__2::condition_variable_any::wait_until<…>(…) • condition_variable:247 12 std::__2::condition_variable_any::wait_for<…>(…) • condition_variable:260 13 cobalt::local_aggregation::DelayedLocalAggregateStorage::Run(…) • delayed_local_aggregate_storage.cc:200 14 λ(…) • delayed_local_aggregate_storage.cc:45 15 std::__2::__invoke<…>(…) • linux-x64/include/c++/v1/__type_traits/invoke.h:150 16 std::__2::__thread_execute<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:192 17 std::__2::__thread_proxy<…>(…) • linux-x64/include/c++/v1/__thread/thread.h:201 18…19 «pthread startup» (-r expands)列出其他特定執行緒。例如
thread 1：thread 1您應該會看到類似下方的輸出內容：
Thread 1 state="Core Dump" koid=19601 name=""接著，您可以使用
frame查看thread 1中的這個特定堆疊框架：frame您應該會看到類似以下的輸出內容：
▶ 0…4 «Waiting for event in async::Loop::Run()» (-r expands) 5 main(…) • cobalt_main.cc:349 6…8 «libc startup» (-r expands)根據上一個步驟的輸出內容，您可能需要深入瞭解
frame 5：frame 5您應該會看到類似下方的輸出內容：
main(…) • cobalt_main.cc:349如要查看目前頁框的其他幾行程式碼，請使用
list：list您應該會看到類似以下的輸出內容：
344 345 if (!app.ok()) { 346 FX_LOGS(FATAL) << "Failed to construct the cobalt app: " << app.status(); 347 } 348 inspector.Health().Ok(); ▶ 349 loop.Run(); 350 FX_LOGS(INFO) << "Cobalt will now shut down."; 351 return 0; 352 }由於這個影格實際上包含一些程式碼，因此您可以使用
locals查看這個堆疊影格中的本機變數。如要查看目前堆疊框架中的所有本機變數，請使用
locals：locals您應該會看到類似下方的輸出內容：
argc = 2 argv = (*)0x26866821fc0 command_line = <Invalid pointer 0x26866821418> context = <Invalid pointer 0x26866821100> event_aggregator_backfill_days = 2 flag_value = <Invalid pointer 0x268668211d8> initial_interval = <Invalid pointer 0x26866821138> inspector = <Invalid pointer 0x268668212f0> loop = <Invalid pointer 0x26866821108> max_bytes_per_observation_store = 215040 min_interval = <Invalid pointer 0x26866821118> require_lifecycle_service = true schedule_interval = <Invalid pointer 0x26866821140> start_event_aggregator_worker = true status = 0 (ZX_OK) storage_quotas = { per_project_reserved_bytes = 1024 total_capacity_bytes = 731136 } test_dont_backfill_empty_reports = false upload_jitter = 0.2 upload_schedule = { target_interval = {__rep_ = 3600} min_interval = {__rep_ = 10} initial_interval = {__rep_ = 60} jitter = 0.2 } use_fake_clock = false use_memory_observation_store = false您可以使用
print查看特定變數的詳細資訊。 例如：print upload_schedule您應該會看到類似以下的輸出內容：
{ target_interval = {__rep_ = 3600} min_interval = {__rep_ = 10} initial_interval = {__rep_ = 60} jitter = 0.2 }
您已成功載入並分析 C++ 程式碼的 Minidump。