neovim/runtime/doc/dev_tools.txt
Jongwook Choi 56b7a18995
docs: how to get core dump files #28826
Problem: Docs about how to obtain backtraces on Linux is not very
beginner-friendly; some users used to have difficulties in getting
stacktrace against Nvim crash.

For instance, the `core` dump file might not appear in the current
directory on Ubuntu systems with apport, and the current docs do not
fully cover such cases.

Solution: Add more hints about where core dump files can be found. For
example, on Ubuntu where apport is managing core dump files, users would
want to find them in `/var/lib/apport/coredump`.
2024-05-21 09:31:28 -07:00

197 lines
6.2 KiB
Plaintext

*dev_tools.txt* Nvim
NVIM REFERENCE MANUAL
Tools and techniques for developing Nvim *dev-tools*
The following advice is helpful when working on or debugging issues with Nvim
itself. See also |debug.txt| for advice that applies to Vim.
Type |gO| to see the table of contents.
==============================================================================
Backtraces *dev-tools-backtrace*
LINUX
Core dumps are disabled by default on Ubuntu, CentOS and others.
To enable core dumps:
>bash
ulimit -c unlimited
<
On systemd-based systems getting a backtrace is as easy as:
>bash
coredumpctl -1 gdb
<
`coredumpctl` is an optional tool, so you may need to install it:
>bash
sudo apt install systemd-coredump
<
The full backtrace is most useful; please send us the `backtrace.txt` file
when reporting a bug related to a crash:
>bash
2>&1 coredumpctl -1 gdb | tee -a backtrace.txt
(gdb) thread apply all bt full
<
On systems without `coredumpctl`, you may find a `core` dump file appearing
in the current directory or in other locations. On Linux systems where
`apport` is installed (such as Ubuntu), the directory where core dump files
are saved can be `/var/lib/apport/coredump` or elsewhere, depending on the
system configuration (see `/proc/sys/kernel/core_pattern`). See also:
https://stackoverflow.com/a/18368068
To get a backtrace from the `./core` dump file:
>bash
gdb build/bin/nvim ./core 2>&1 | tee backtrace.txt
(gdb) thread apply all bt full
<
MACOS
If `nvim` crashes, you can see the backtrace in `Console.app` (under "Crash
Reports" or "User Diagnostic Reports" for older macOS versions).
>bash
open -a Console
<
You may also want to enable core dumps on macOS. To do this, first make sure
the `/cores/` directory exists and is writable:
>bash
sudo mkdir /cores
sudo chown root:admin /cores
sudo chmod 1775 /cores
<
Then set the core size limit to `unlimited`:
>bash
ulimit -c unlimited
<
Note that this is done per shell process. If you want to make this the default
for all shells, add the above line to your shell's init file (e.g. `~/.bashrc`
or similar).
You can then open the core file in `lldb`:
>bash
lldb -c /cores/core.12345
<
Apple's documentation archive has some other useful information
https://developer.apple.com/library/archive/technotes/tn2124/_index.html#//apple_ref/doc/uid/DTS10003391-CH1-SECCOREDUMPS,
but note that some of the things on this page are out of date (such as enabling
core dumps with `/etc/launchd.conf`).
==============================================================================
Gdb *dev-tools-gdb*
USING GDB TO STEP THROUGH FUNCTIONAL TESTS
Use `TEST_TAG` to run tests matching busted tags (of the form `#foo` e.g.
`it("test #foo ...", ...)`):
>bash
GDB=1 TEST_TAG=foo make functionaltest
<
Then, in another terminal:
>bash
gdb build/bin/nvim
(gdb) target remote localhost:7777
-- See `nvim_argv` in https://github.com/neovim/neovim/blob/master/test/functional/testnvim.lua.
USING LLDB TO STEP THROUGH UNIT TESTS
>
lldb .deps/usr/bin/luajit -- .deps/usr/bin/busted --lpath="./build/?.lua" test/unit/
<
USING GDB
To attach to a running `nvim` process with a pid of 1234 (Tip: the pid of a
running Nvim instance can be obtained by calling |getpid()|), for instance:
>bash
gdb -tui -p 1234 build/bin/nvim
<
The `gdb` interactive prompt will appear. At any time you can:
- `break foo` to set a breakpoint on the `foo()` function
- `n` to step over the next statement
- `<Enter>` to repeat the last command
- `s` to step into the next statement
- `c` to continue
- `finish` to step out of the current function
- `p zub` to print the value of `zub`
- `bt` to see a backtrace (callstack) from the current location
- `CTRL-x CTRL-a` or `tui enable` to show a TUI view of the source file in the
current debugging context. This can be extremely useful as it avoids the
need for a gdb "frontend".
- `<up>` and `<down>` to scroll the source file view
GDB REVERSE DEBUGGING
- `set record full insn-number-max unlimited`
- `continue` for a bit (at least until `main()` is executed
- `record`
- provoke the bug, then use `revert-next`, `reverse-step`, etc. to rewind the
debugger
USING GDBSERVER
You may want to connect multiple `gdb` clients to the same running `nvim`
process, or you may want to connect to a remote `nvim` process with a local
`gdb`. Using `gdbserver`, you can attach to a single process and control it
from multiple `gdb` clients.
Open a terminal and start `gdbserver` attached to `nvim` like this:
>bash
gdbserver :6666 build/bin/nvim 2> gdbserver.log
<
`gdbserver` is now listening on port 6666. You then need to attach to this
debugging session in another terminal:
>bash
gdb build/bin/nvim
<
Once you've entered `gdb`, you need to attach to the remote session:
>
(gdb) target remote localhost:6666
<
In case gdbserver puts the TUI as a background process, the TUI can become
unable to read input from pty (and receives SIGTTIN signal) and/or output data
(SIGTTOU signal). To force the TUI as the foreground process, you can add
>c
signal (SIGTTOU, SIG_IGN);
if (!tcsetpgrp(data->input.in_fd, getpid())) {
perror("tcsetpgrp failed");
}
<
to `tui.c:terminfo_start`.
USING GDBSERVER IN TMUX
Consider using a custom makefile
https://github.com/neovim/neovim/blob/master/BUILD.md#custom-makefile to
quickly start debugging sessions using the `gdbserver` method mentioned above.
This example `local.mk` will create the debugging session when you type `make
debug`.
>make
.PHONY: dbg-start dbg-attach debug build
build:
@$(MAKE) nvim
dbg-start: build
@tmux new-window -n 'dbg-neovim' 'gdbserver :6666 ./build/bin/nvim -D'
dbg-attach:
@tmux new-window -n 'dbg-cgdb' 'cgdb -x gdb_start.sh ./build/bin/nvim'
debug: dbg-start dbg-attach
<
Here `gdb_start.sh` includes `gdb` commands to be called when the debugger
starts. It needs to attach to the server started by the `dbg-start` rule. For
example:
>
(gdb) target remote localhost:6666
(gdb) br main
<
vim:tw=78:ts=8:et:ft=help:norl: