aboutsummaryrefslogtreecommitdiff
path: root/src/nvim/msgpack_rpc
Commit message (Collapse)AuthorAge
...
* API: nvim_unsubscribe(): Handle unknown events #8745Daniel Hahler2018-08-28
| | | | close #8745
* channel.c: refactor spaghetti codemicha2018-07-14
| | | | | | | | | | | | | | | | | | | | channel.c: WIP remove redundant method check and added FUNC_ATTR_NONNULL_ALL macro channel.c channel_defs.h helpers.c: added Error field to RequestEvent, added no_op handler func channel.c: use const char* instead of string and cleanup channel.c; channel_defs.h; helpers.c: removed error from event again; send errors directly to the channel without using handlers and events channel.c: fixed memory leak and lint errors api/private/dispatch.c; api/vim.c; msgpack_rpc/channel.c msgpack_rpc/helpers.c added Error* field to msgpack_get_handler_for; further refactored channel.c channel.c:323 changed order of evaluation in if statement channel.c: removed superflous whitespace dispatch.c: review comment
* API: validation: mention invalid method name (#8489)Justin M. Keyes2018-06-07
|
* api: list information about all channels/jobs.Björn Linse2018-05-23
| | | | | Fire autocmd when channel opens or its info changes. Add a way for API clients can describe themselves.
* server: init v:servername if $NVIM_LISTEN_ADDRESS is invalidJustin M. Keyes2018-04-11
| | | | | Before this change, if $NVIM_LISTEN_ADDRESS was invalid, v:servername was left empty.
* server: introduce --listen, deprecate $NVIM_LISTEN_ADDRESSJustin M. Keyes2018-04-11
|
* serverstop(): return FALSE for invalid addressJustin M. Keyes2018-04-11
|
* channels: generalize jobclose()Björn Linse2017-11-25
|
* channels: allow bytes sockets and stdio, and buffered bytes outputBjörn Linse2017-11-24
|
* channels: refactorBjörn Linse2017-11-24
|
* Revert channel logging, rebased on new code belowBjörn Linse2017-11-24
|
* server.c: Fix bug in release mode (#7594)Phlosioneer2017-11-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When compiling with CMAKE_BUILD_TYPE=RelWithDebInfo, several -Wmaybe-uninitialized warnings are printed. These were thought to be false positives (#5061); there are no control paths that lead to an uninitialized value. However, when gcc is run in -O2 mode, it makes a mistake while generating the necessary logic. Specifically, for the code: ... int = 0; // Index of the server whose address equals addr. for (; i < watchers.ga_len; i++) { watcher = ((SocketWatcher **)watchers.ga_data)[i]; // <snip> } if (i >= watchers.ga_len) { ELOG("Not listening on %s", addr); return; } ... Gcc generates: ... <+98>: cmp %ebx, %ebp <+100>: jg 0x530f13 <server_stop+55> <+102>: cmp %ebp, ebx <+104>: jl 0x530f7e <server_stop+162> ... Normally, the if statement should catch the only control path where watcher is not assigned: watchers.ga_len <= 0. When compiled, the assembly lines 98 and 100 correspond to checking if i < watchers.ga_len, and the lines 102 and 104 correspond to checking if i >= watchers.ga_len. The assembly seems to compare ebp (which is watchers.ga_len) with ebx (which is i), and jump if greater; then do the same comparison and jump if less. This is where gcc makes a mistake: it flips the order of the cmp instruction. This means that the REAL behavior is first check if i < watchers.ga_len and then check if i < watchers.ga_len. Which means the code inside the if statement is NEVER executed; no combination of i and watchers.ga_len will ever trigger ELOG(). So not only is this a use of an uninitialized value if watchers.ga_len == 0 (or technically, if it's less than zero too), it also clobbers any error detection if the for loop reaches the last entry (which would normally cause i == watchers.ga_len too). This commit fixes this issue by adding a bool to keep track of whether a watcher was found during the loop. This makes gcc generate the correct code, avoiding both bugs.
* rpc: Don't delay notifications when request is pending (#6544)Björn Linse2017-10-29
| | | | | | | | | | | | | | | | | | | | | | With the old behavior, if a GUI makes a blocking request that requires user interaction (like nvim_input()), it would not get any screen updates. The client, not nvim, should decide how to handle notifications during a pending request. If an rplugin wants to avoid async calls while a sync call is busy, it likely wants to avoid processing async calls while another async call also is handled as well. This may break the expectation of some existing rplugins. For compatibility, remote/define.vim reimplements the old behavior. Clients can opt-out by specifying `sync=urgent`. - Legacy hosts should be updated to use `sync=urgent`. They could add a flag indicating which async methods are always safe to call and which must wait until the main loop returns. - New hosts can expose the full asyncness, they don't need to offer both behaviors. ref #6532 ref #1398 d83868fe9071af1b4866594eac12f7aa0fa71b53
* doc: channel, eventloopJustin M. Keyes2017-09-05
|
* channel.c:call_set_error(): fix memory leakJustin M. Keyes2017-08-21
|
* rpc: close channel if stream was closedJustin M. Keyes2017-08-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | f_jobstop()/f_rpcstop() .. process_stop() .. process_close_in(proc) closes the write-stream of a RPC channel. But there might be a pending RPC notification on the queue, which may get processed just before the channel is closed. To handle that case, check the Stream.closed in channel.c:receive_msgpack(). Before this change, the above scenario could trigger this assert(!stream->closed) in wstream_write(): 0x00007f96e1cd3428 in __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:54 0x00007f96e1cd502a in __GI_abort () at abort.c:89 0x00007f96e1ccbbd7 in __assert_fail_base (fmt=<optimized out>, assertion=assertion@entry=0x768f9b "!stream->closed", file=file@entry=0x768f70 "../src/nvim/event/wstream.c", line=line@entry=77, function=function@entry=0x768fb0 <__PRETTY_FUNCTION__.13735> "wstream_write") at assert.c:92 0x00007f96e1ccbc82 in __GI___assert_fail (assertion=0x768f9b "!stream->closed", file=0x768f70 "../src/nvim/event/wstream.c", line=77, function=0x768fb0 <__PRETTY_FUNCTION__.13735> "wstream_write") at assert.c:101 0x00000000004d2c1f in wstream_write (stream=0x7f96e0a35078, buffer=0x7f96e09f9b40) at ../src/nvim/event/wstream.c:77 0x00000000005857b2 in channel_write (channel=0x7f96e0ae5800, buffer=0x7f96e09f9b40) at ../src/nvim/msgpack_rpc/channel.c:551 0x000000000058567d in on_request_event (argv=0x7ffed792efa0) at ../src/nvim/msgpack_rpc/channel.c:523 0x00000000005854c8 in handle_request (channel=0x7f96e0ae5800, request=0x7ffed792f1b8) at ../src/nvim/msgpack_rpc/channel.c:503 0x00000000005850cb in parse_msgpack (channel=0x7f96e0ae5800) at ../src/nvim/msgpack_rpc/channel.c:423 0x0000000000584f90 in receive_msgpack (stream=0x7f96e0a35218, rbuf=0x7f96e0d1d4c0, c=22, data=0x7f96e0ae5800, eof=false) at ../src/nvim/msgpack_rpc/channel.c:389 0x00000000004d0b20 in read_event (argv=0x7ffed792f4a8) at ../src/nvim/event/rstream.c:190 0x00000000004ce462 in multiqueue_process_events (this=0x7f96e18172d0) at ../src/nvim/event/multiqueue.c:150 0x000000000059b630 in nv_event (cap=0x7ffed792f620) at ../src/nvim/normal.c:7908 0x000000000058be69 in normal_execute (state=0x7ffed792f580, key=-25341) at ../src/nvim/normal.c:1137 0x0000000000652463 in state_enter (s=0x7ffed792f580) at ../src/nvim/state.c:61 0x000000000058a1fe in normal_enter (cmdwin=false, noexmode=false) at ../src/nvim/normal.c:467 0x00000000005500c2 in main (argc=2, argv=0x7ffed792f8d8) at ../src/nvim/main.c:554 Alternative approach suggested by bfredl is to use close_cb of the process. My unsuccessful attempt is below. (It seems close_cb is queued too late, which is the similar problem addressed by this commit): commit 75fc12c6ab15711bdb7b18c6d42ec9d157f5145e Author: Justin M. Keyes <justinkz@gmail.com> Date: Fri Aug 18 01:30:41 2017 +0200 rpc: use Stream's close_cb instead of explicit check in receive_msgpack() diff --git a/src/nvim/event/process.c b/src/nvim/event/process.c index 8371d3cd482e..e52da23cdc40 100644 --- a/src/nvim/event/process.c +++ b/src/nvim/event/process.c @@ -416,6 +416,10 @@ static void on_process_exit(Process *proc) static void on_process_stream_close(Stream *stream, void *data) { Process *proc = data; + ILOG("on_process_stream_close"); + if (proc->stream_close_cb != NULL) { + proc->stream_close_cb(stream, proc->stream_close_data); + } decref(proc); } diff --git a/src/nvim/event/process.h b/src/nvim/event/process.h index 5c00e8e7ecd5..34a8d54f6f8c 100644 --- a/src/nvim/event/process.h +++ b/src/nvim/event/process.h @@ -26,6 +26,11 @@ struct process { Stream *in, *out, *err; process_exit_cb cb; internal_process_cb internal_exit_cb, internal_close_cb; + + // Called when any of the process streams (in/out/err) closes. + stream_close_cb stream_close_cb; + void *stream_close_data; + bool closed, detach; MultiQueue *events; }; @@ -50,6 +55,8 @@ static inline Process process_init(Loop *loop, ProcessType type, void *data) .closed = false, .internal_close_cb = NULL, .internal_exit_cb = NULL, + .stream_close_cb = NULL, + .stream_close_data = NULL, .detach = false }; } diff --git a/src/nvim/event/stream.c b/src/nvim/event/stream.c index 7c865bfe1e8c..c8720d1e45d9 100644 --- a/src/nvim/event/stream.c +++ b/src/nvim/event/stream.c @@ -95,7 +95,11 @@ void stream_close(Stream *stream, stream_close_cb on_stream_close, void *data) void stream_close_handle(Stream *stream) FUNC_ATTR_NONNULL_ALL { + ILOG("stream=%d", stream); + // LOG_CALLSTACK(); if (stream->uvstream) { + // problem: this schedules on the queue, but channel.c:receive_msgpack may + // be processed before close_cb is called by libuv. uv_close((uv_handle_t *)stream->uvstream, close_cb); } else { uv_close((uv_handle_t *)&stream->uv.idle, close_cb); @@ -105,6 +109,7 @@ void stream_close_handle(Stream *stream) static void close_cb(uv_handle_t *handle) { Stream *stream = handle->data; + ILOG(">>>>>>>>>>>>>>>>>>>>>>> stream=%p stream->internal_close_cb=%p", stream, stream->internal_close_cb); if (stream->buffer) { rbuffer_free(stream->buffer); } diff --git a/src/nvim/msgpack_rpc/channel.c b/src/nvim/msgpack_rpc/channel.c index 782eabe04e4a..dc2b794e366a 100644 --- a/src/nvim/msgpack_rpc/channel.c +++ b/src/nvim/msgpack_rpc/channel.c @@ -128,6 +128,8 @@ uint64_t channel_from_process(Process *proc, uint64_t id, char *source) source); incref(channel); // process channels are only closed by the exit_cb channel->data.proc = proc; + channel->data.proc->stream_close_cb = close_cb2; + channel->data.proc->stream_close_data = channel; wstream_init(proc->in, 0); rstream_init(proc->out, 0); @@ -387,17 +389,6 @@ static void receive_msgpack(Stream *stream, RBuffer *rbuf, size_t c, goto end; } - if ((chan_wstream(channel) != NULL && chan_wstream(channel)->closed) - || (chan_rstream(channel) != NULL && chan_rstream(channel)->closed)) { - char buf[256]; - snprintf(buf, sizeof(buf), - "ch %" PRIu64 ": stream closed unexpectedly. " - "closing channel", - channel->id); - call_set_error(channel, buf, WARN_LOG_LEVEL); - goto end; - } - size_t count = rbuffer_size(rbuf); DLOG("ch %" PRIu64 ": parsing %u bytes from msgpack Stream: %p", channel->id, count, stream); @@ -571,23 +562,6 @@ static Stream *chan_wstream(Channel *chan) abort(); } -/// Returns the Stream that a Channel reads from. -static Stream *chan_rstream(Channel *chan) -{ - switch (chan->type) { - case kChannelTypeSocket: - return &chan->data.stream; - case kChannelTypeProc: - return chan->data.proc->out; - case kChannelTypeStdio: - return &chan->data.std.in; - case kChannelTypeInternal: - return NULL; - } - abort(); -} - - static bool channel_write(Channel *channel, WBuffer *buffer) { bool success = false; @@ -799,6 +773,12 @@ static void close_cb(Stream *stream, void *data) decref(data); } +static void close_cb2(Stream *stream, void *data) +{ + ILOG("close_cb2"); + close_channel(data); +} + /// @param source description of source function, rplugin name, TCP addr, etc static Channel *register_channel(ChannelType type, uint64_t id, MultiQueue *events, char *source)
* log: some DEBUG-level stream loggingJustin M. Keyes2017-08-21
|
* log: caller provides the source detailsJustin M. Keyes2017-07-23
|
* log: channel registrationJustin M. Keyes2017-07-22
|
* log: Always enable; remove DISABLE_LOGJustin M. Keyes2017-06-07
| | | | | | - Establish ERROR log level as "critical". Such errors are rare and will be valuable when users encounter unusual circumstances. - Set -DMIN_LOG_LEVEL=3 for release-type builds
* channel_write: fix compiler warningJustin M. Keyes2017-05-31
|
* sockets: don't deadlock when connecting to own pipe addressBjörn Linse2017-05-29
|
* channels: implement sockopen() to connect to socketBjörn Linse2017-05-29
| | | | Helped-By: oni-link <knil.ino@gmail.com>
* Server: don't fall back to Unix socketsMarco Hinz2017-05-22
|
* Server: use uv_getaddrinfo() for $NVIM_LISTEN_ADDRESSMarco Hinz2017-05-22
| | | | | | | | | | | | | | | | | | | This change implicitly adds IPv6 support. If the address contains ":", we try to use a TCP socket instead of a Unix domain socket. Everything in front of the last occurrence of ":" is the hostname and everything after it the port. If the hostname lookup fails, we fall back to using a Unix domain socket. If the port is empty ("localhost:"), a random port will be assigned. Examples: NVIM_LISTEN_ADDRESS=localhost:12345 -> TCP (IPv4 or IPv6), port: 12345 NVIM_LISTEN_ADDRESS=localhost: -> TCP (IPv4 or IPv6), port: random (> 1024) NVIM_LISTEN_ADDRESS=localhost:0 -> TCP (IPv4 or IPv6), port: random (> 1024) NVIM_LISTEN_ADDRESS=localhost -> Unix domain socket "localhost" in current dir
* Merge #4411 from ZyX-I/luaviml'/luaJustin M. Keyes2017-05-09
|\
| * Merge branch 'master' into luaviml'/luaZyX2017-05-08
| |\
| * \ Merge branch 'master' into luaviml'/luaZyX2017-04-08
| |\ \
| * | | api: Also shift numbers in api_metadata outputZyX2017-03-27
| | | | | | | | | | | | | | | | | | | | | | | | Fixes problem introduced by “api: Allow kObjectTypeNil to be zero without breaking compatibility”: apparently there are clients which use metadata and there are which aren’t. For the first that commit would not be needed, for the second that commit misses this critical piece.
| * | | gendeclarations: Make declarations generator work with macros funcsZyX2017-03-27
| | | | | | | | | | | | | | | | | | | | Now it checks functions also after every semicolon and closing figure brace, possibly preceded by whitespaces (tabs and spaces). This should make messing with declarations in macros not needed.
| * | | msgpack_rpc: Fix #HANDLE_TYPE_CONVERSION_IMPLZyX2017-03-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Function declarations generator is able to handle properly only the *first* function definition that is in macros, and only if it is the first entity in the macros. So msgpack_rpc_from_* was already really a static function, additionally its attributes were useless. This commit switches to explicit declarations and makes generated functions static.
| * | | api: Allow kObjectTypeNil to be zero without breaking compatibilityZyX2017-03-27
| | | |
| * | | functests: Test for error conditionsZyX2017-03-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | During testing found the following bugs: 1. msgpack-gen.lua script is completely unprepared for Float values either in return type or in arguments. Specifically: 1. At the time of writing relevant code FLOAT_OBJ did not exist as well as FLOATING_OBJ, but it would be used by msgpack-gen.lua should return type be Float. I added FLOATING_OBJ macros later because did not know that msgpack-gen.lua uses these _OBJ macros, otherwise it would be FLOAT_OBJ. 2. msgpack-gen.lua should use .data.floating in place of .data.float. But it did not expect that .data subattribute may have name different from lowercased type name. 2. vim_replace_termcodes returned its argument as-is if it receives an empty string (as well as _vim_id*() functions did). But if something in returned argument lives in an allocated memory such action will cause double free: once when freeing arguments, then when freeing return value. It did not cause problems yet because msgpack bindings return empty string as {NULL, 0} and nothing was actually allocated. 3. New code in msgpack-gen.lua popped arguments in reversed order, making lua bindings’ signatures be different from API ones.
* | | | refactor/single-include (#6687)Carlo Abelli2017-05-08
| |_|/ |/| |
* | | coverity/161682: FP: Dereference after null check (#6630)Justin M. Keyes2017-05-01
| | |
* | | api/nvim_get_mode: Use child-queue instead of "priority".Justin M. Keyes2017-04-28
| | |
* | | input.c: Process only safe events before blocking.Justin M. Keyes2017-04-28
| | | | | | | | | | | | | | | Introduce multiqueue_process_priority() to process only events at or above a certain priority.
* | | api: nvim_get_mode()Justin M. Keyes2017-04-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Asynchronous API functions are served immediately, which means pending input could change the state of Nvim shortly after an async API function result is returned. nvim_get_mode() is different: - If RPCs are known to be blocked, it responds immediately (without flushing the input/event queue) - else it is handled just-in-time before waiting for input, after pending input was processed. This makes the result more reliable (but not perfect). Internally this is handled as a special case, but _semantically_ nothing has changed: API users never know when input flushes, so this internal special-case doesn't violate that. As far as API users are concerned, nvim_get_mode() is just another asynchronous API function. In all cases nvim_get_mode() never blocks for more than the time it takes to flush the input/event queue (~µs). Note: This doesn't address #6166; nvim_get_mode() will provoke #6166 if e.g. `d` is operator-pending. Closes #6159
* | | api: Do not translate error messages.Justin M. Keyes2017-04-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Also re-word some error messages: - "Key does not exist: %s" - "Invalid channel: %<PRIu64>" - "Request array size must be 4 (request) or 3 (notification)" - "String cannot contain newlines" References #6150
* | | api_set_error(): renameJustin M. Keyes2017-04-23
| | |
* | | api/internal: Remove `set` field from Error type.Justin M. Keyes2017-04-23
| | |
* | | api_clear_error()Justin M. Keyes2017-04-23
| | |
* | | api: Do not truncate errors <1 MB. #6237Sander Bosma2017-04-23
| | | | | | | | | | | | Closes #5984
* | | *: Add comment to all C filesZyX2017-04-19
| |/ |/|
* | Add handling for MSGPACK_OBJECT_FLOAT{32,64}James McCoy2017-03-30
| | | | | | | | | | | | msgpack-c previously only had MSGPACK_OBJECT_FLOAT, which was a 64-bit value. Now, 32-bit and 64-bit floats are supported as distinct types, but we'll simply continue to treat everything as 64-bit types.
* | eval,*: Move get_tv_string to typval.cZyX2017-03-29
|/ | | | Function was renamed and changed to return `const char *`.
* terminal: Disable some options in terminal-mode.Justin M. Keyes2017-03-01
| | | | | | | In terminal-mode these options are nonsense because cursor is placed at end of buffer to "follow" output. Closes #2259
* channel.c: loggingJustin M. Keyes2017-02-26
|
* msgpack_rpc: Allocate empty string in msgpack_rpc_to_objectJames McCoy2016-11-29
| | | | | | | | | | STR_CASE previously used a NULL data pointer for the String object, but this pushes the NULL checks to the rest of the code. Instead, allocating an empty string solves the same issue of there not being any data but ensures that we're not passing NULL to functions that don't expect it. Closes #5627
* Avoid serializing NULL string through msgpackJames McCoy2016-11-23
| | | | | | Attempting to serialize a NULL string through msgpack results in msgpack_sbuffer_write attempting to memcpy from a NULL pointer, which is undefined behavior.
generated by cgit (git 2.34.1) at 2025-06-26 04:03:05 +0000