(Experimental) Trace2 base plus GVFS extensions #28

jeffhostetler · 2018-09-26T20:07:21Z

Experimental merge of basic Trace2 plus GVFS extensions into our GVFS branch for testing purposes.
The basic Trace2 commits (the ones starting with "trace2:") will eventually be submitted upstream and
should be considered experimental at this time.

The GVFS extension commits (the ones starting with "gvfs:trace2:") depend on GVFS-specific code
and might not be up-streamed until much later.

The "t0051" and "mingw" commits fix a bug introduced shortly before 2.19.0 was released. These
commits have already been submitted upstream.

benpeart

s/higer/higher

benpeart

In general I like the direction this is taking but... it is a large change and was hard to review effectively. I feel like it needs a design document added in Documentation\trace2.txt so that I can get an overview of what it is trying to accomplish and how it goes about it. The closest thing I could find was the trace2 header file.

Some questions I have include:

what providers are available?
how do they differ?
when would I use one vs another?
how are they configured/used?

Some examples of sample output would also be very helpful. I see lots of tracing added to the source and I'm sure you have a specific goal you were trying to accomplish but I just have a hard time figuring out what it is.

benpeart · 2018-09-27T14:09:47Z

config.c

@@ -1504,6 +1504,11 @@ int git_default_config(const char *var, const char *value, void *cb)
 		return 0;
 	}

+	if (starts_with(var, "gvfs.")) {


Do you have any thoughts on all the other git_config_get_* functions that are in this file? I think it's fine to say they can get tracing added "as needed."

I think I want to move this somewhere else. probably in a separate iteration so that we don't alter the main init sequence and so that we can cluster all of the fields we want to echo to trace2.

benpeart · 2018-09-27T14:21:43Z

trace2.c

+
+	tr2_sid_get();
+
+	atexit(tr2main_atexit_handler);


Do you need to protect against multiple initialize calls here?

yeah, i should.

benpeart · 2018-09-27T14:22:52Z

trace2.c

+
+	code &= 0xff;
+
+	if (!trace2_initialized)


I wonder if moving this initialization test (and the others in this file) out into the macro so that it is inlined would make any difference in the perf overhead when tracing isn't enabled?

i thought i would. i put it inside the functions for now to avoid cluttering up the .h with more macros. i'll revisit now that things have settled down.

trace2.h

benpeart · 2018-09-27T14:54:41Z

trace2.h

+ *
+ * Emits a 'worktree' event for this repo instance.
+ */
+void trace2_def_repo_fl(const char *file, int line,


I'm trying to understand when it makes sense to do a generic trace event vs create a new custom event (like the repo event). Do you have any guidelines on when each is appropriate? I ask because I was surprised when I saw this was a custom event.

I'm hoping that most new calls can be region-enter/leave or data.
def-repo is special because I wanted to add a new column to the perf and event views
saying what repo is being operated on. that is, when google gets the in-proc submodule stuff finished, we'll get "do_read_index" calls for each repo they visit. with this, we'll get a repo# column for each do_read_index call.

benpeart · 2018-09-27T16:28:05Z

trace2.h

+ * repo-id and printf message.
+ *
+ * Enter a new nesting level on the current thread and remember the
+ * current time.  This controls the indenting of subsequent thread


by "subsequent thread events" I assume you mean "all subsequent events on this thread" and not just thread specific events (ie thread_start/thread_exit)

yes, thanks.

benpeart · 2018-09-27T16:55:14Z

trace2.h

+ * This event implicitly contains information about thread, nesting region,
+ * and optional repo-id.
+ *
+ * On event-based TRACE2 targets, this generates a 'data' event suitable


Is there some documentation somewhere on what TRACE2 targets exist and how they behave? This information on how event-based targets and printf-based targets is interesting/good to know but is buried here where it is unlikely someone will discover it.

not yet. i need to write the design doc/summary.

derrickstolee · 2018-10-09T14:06:11Z

trace2.c

+
+	for_each_wanted_builtin(j, tgt_j) {
+		if (tgt_j->pfn_error_va_fl)
+			tgt_j->pfn_error_va_fl(file, line, fmt, ap);


@jeffhostetler I got a segfault here when running GIT_TR2_PERF=1 git push origin topic when a force-push was required.

Here is my GDB data:

(gdb) run push gh-stolee push-trace Starting program: /home/stolee/git/git push gh-stolee push-trace [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". 14:04:16.895179 common-main.c:36 | d0 | main | version | | | | | 2.19.0.gvfs.1.23.g86fac7d809 14:04:16.900169 common-main.c:36 | d0 | main | start | | | | | /home/stolee/git/git push gh-stolee push-trace 14:04:16.900257 repository.c:117 | d0 | main | def_repo | r1 | | | | worktree:/home/stolee/git 14:04:16.910344 git.c:486 | d0 | main | cmd_verb | | | | | push 14:04:17.133183 run-command.c:710 | d0 | main | child_start | | 0.238003 | | | [ch0] class:? argv: ssh git@github.com 'git-receive-pack '\''derrickstolee/git.git'\''' 14:04:19.032353 run-command.c:956 | d0 | main | child_exit | | 2.137168 | 1.899165 | | [ch0] code:0 To github.com:derrickstolee/git.git ! [rejected] push-trace -> push-trace (non-fast-forward) error: failed to push some refs to 'git@github.com:derrickstolee/git.git' Program received signal SIGSEGV, Segmentation fault. __strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:62 62 ../sysdeps/x86_64/multiarch/strlen-avx2.S: No such file or directory. (gdb) bt #0 __strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:62 #1 0x00007ffff73fd4d3 in _IO_vfprintf_internal (s=s@entry=0x7fffffffd6b0, format=format@entry=0x55555575dce0 "failed to push some refs to '%s'", ap=ap@entry=0x7fffffffd830) at vfprintf.c:1643 #2 0x00007ffff74d2169 in ___vsnprintf_chk (s=0x555555a4c080 "failed to push some refs to 'git", maxlen=<optimized out>, flags=flags@entry=1, slen=slen@entry=18446744073709551615, format=format@entry=0x55555575dce0 "failed to push some refs to '%s'", args=args@entry=0x7fffffffd830) at vsnprintf_chk.c:63 #3 0x00005555556f86d4 in vsnprintf (__ap=0x7fffffffd830, __fmt=0x55555575dce0 "failed to push some refs to '%s'", __n=<optimized out>, __s=<optimized out>) at /usr/include/x86_64-linux-gnu/bits/stdio2.h:77 #4 strbuf_vaddf (sb=sb@entry=0x7fffffffd870, fmt=0x55555575dce0 "failed to push some refs to '%s'", ap=0x7fffffffd900) at strbuf.c:343 #5 0x0000555555709b1c in fn_error_va_fl (file=0x55555578f9de "usage.c", line=59, fmt=<optimized out>, ap=<optimized out>) at trace2/tr2_tgt_perf.c:226 #6 0x0000555555705566 in trace2_cmd_error_va_fl (file=0x55555578f9de "usage.c", line=59, fmt=0x55555575dce0 "failed to push some refs to '%s'", ap=0x7fffffffd900) at trace2.c:213 #7 0x00005555557202ba in error (err=<optimized out>) at usage.c:213 #8 0x00005555555d25c4 in push_with_options (transport=0x555555a54460, rs=rs@entry=0x5555559f5320 <rs>, flags=flags@entry=0) at builtin/push.c:361 #9 0x00005555555d338c in do_push (push_options=0x5555559e3140 <push_options_config>, flags=0, repo=<optimized out>) at builtin/push.c:426 #10 cmd_push (argc=<optimized out>, argv=<optimized out>, prefix=<optimized out>) at builtin/push.c:636 #11 0x000055555556f2c6 in run_builtin (argv=<optimized out>, argc=<optimized out>, p=<optimized out>) at git.c:490 #12 handle_builtin (argc=<optimized out>, argv=<optimized out>) at git.c:724 #13 0x000055555557032f in run_argv (argv=0x7fffffffe2c0, argcp=0x7fffffffe2cc) at git.c:792 #14 cmd_main (argc=<optimized out>, argv=<optimized out>) at git.c:914 #15 0x000055555556edf2 in main (argc=4, argv=0x7fffffffe528) at common-main.c:46 (gdb) f 6 #6 0x0000555555705566 in trace2_cmd_error_va_fl (file=0x55555578f9de "usage.c", line=59, fmt=0x55555575dce0 "failed to push some refs to '%s'", ap=0x7fffffffd900) at trace2.c:213 213 tgt_j->pfn_error_va_fl(file, line, fmt, ap);

I ran valgrind which may help a little bit:

==33024== Invalid read of size 1 ==33024== at 0x4C32CF2: strlen (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==33024== by 0x54DD4D2: vfprintf (vfprintf.c:1643) ==33024== by 0x55B2168: __vsnprintf_chk (vsnprintf_chk.c:63) ==33024== by 0x2AC6F3: vsnprintf (stdio2.h:77) ==33024== by 0x2AC6F3: strbuf_vaddf (strbuf.c:343) ==33024== by 0x2BDB3B: fn_error_va_fl (tr2_tgt_perf.c:226) ==33024== by 0x2B9585: trace2_cmd_error_va_fl (trace2.c:213) ==33024== by 0x2D42D9: error (usage.c:213) ==33024== by 0x1865E1: push_with_options (push.c:362) ==33024== by 0x1873AB: do_push (push.c:427) ==33024== by 0x1873AB: cmd_push (push.c:637) ==33024== by 0x1232C5: run_builtin (git.c:490) ==33024== by 0x1232C5: handle_builtin (git.c:724) ==33024== by 0x12432E: run_argv (git.c:792) ==33024== by 0x12432E: cmd_main (git.c:914) ==33024== by 0x122DF1: main (common-main.c:46) ==33024== Address 0x1 is not stack'd, malloc'd or (recently) free'd ==33024== ==33024== ==33024== Process terminating with default action of signal 11 (SIGSEGV) ==33024== Access not within mapped region at address 0x1 ==33024== at 0x4C32CF2: strlen (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==33024== by 0x54DD4D2: vfprintf (vfprintf.c:1643) ==33024== by 0x55B2168: __vsnprintf_chk (vsnprintf_chk.c:63) ==33024== by 0x2AC6F3: vsnprintf (stdio2.h:77) ==33024== by 0x2AC6F3: strbuf_vaddf (strbuf.c:343) ==33024== by 0x2BDB3B: fn_error_va_fl (tr2_tgt_perf.c:226) ==33024== by 0x2B9585: trace2_cmd_error_va_fl (trace2.c:213) ==33024== by 0x2D42D9: error (usage.c:213) ==33024== by 0x1865E1: push_with_options (push.c:362) ==33024== by 0x1873AB: do_push (push.c:427) ==33024== by 0x1873AB: cmd_push (push.c:637) ==33024== by 0x1232C5: run_builtin (git.c:490) ==33024== by 0x1232C5: handle_builtin (git.c:724) ==33024== by 0x12432E: run_argv (git.c:792) ==33024== by 0x12432E: cmd_main (git.c:914) ==33024== by 0x122DF1: main (common-main.c:46)

@jeffhostetler The problem is that vsnprintf mutates the va_args as it executes. I verified this by swapping the order of the vreportf and the trace2_cmd_error_va in error_builtin. If I swap them, then your error message appears and the segfault happens during the vreportf.

I omitted the va_copy() on the va_list args when I moved code over from my SLOG version.
I just pushed an update to fix this.

derrickstolee · 2018-10-09T15:34:04Z

usage.c

+	 * TODO It would be nice to update the call sites to pass both
+	 * the static usage string and the detailed error message.
+	 */
+
 	exit(129);
 }

 static NORETURN void die_builtin(const char *err, va_list params)
 {
 	vreportf("fatal: ", err, params);


copy params before this vreportf or else they will be mutated. Same with error_builtin below.

Create a test-tool helper to create the server side of a windows named pipe, wait for a client connection, and copy data written to the pipe to stdout. Create t0051 test to route GIT_TRACE output of a command to a named pipe using the above test-tool helper. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

derrickstolee · 2018-10-09T18:14:03Z

trace2.c

+	for_each_wanted_builtin(j, tgt_j) {
+		if (tgt_j->pfn_region_enter_printf_va_fl) {
+			va_list copy_ap;
+			va_copy(copy_ap, ap);


if (ap) va_copy(copy_ap, ap);

or else nullrefs when called form trace2_region_enter_f1.

derrickstolee · 2018-10-09T18:14:16Z

trace2.c

+	for_each_wanted_builtin(j, tgt_j) {
+		if (tgt_j->pfn_region_leave_printf_va_fl) {
+			va_list copy_ap;
+			va_copy(copy_ap, ap);


if (ap) va_copy(copy_ap, ap);

or else nullrefs when called form trace2_region_leave_f1.

(there are two more that need to be protected.)

@jeffhostetler here is a fixup commit for you

Thanks! I just ran some tests on Linux and I saw some of these failures. They don't seem to be a problem on Windows.

I'm pushing a better fix for this.

derrickstolee · 2018-10-09T19:25:18Z

I'm running make coverage-test using all your trace2 environment variables. When that's done, I'll share the coverage-diff to see if we are missing any coverage.

derrickstolee · 2018-10-09T23:44:31Z

Uncovered lines in trace2.c:

    #####:  125:static void tr2main_signal_handler(int signo)
    #####:  132:        us_now = getnanotime() / 1000;
    #####:  133:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  135:        for_each_wanted_builtin(j, tgt_j) {
    #####:  136:                if (tgt_j->pfn_signal)
    #####:  137:                        tgt_j->pfn_signal(us_elapsed_absolute, signo);
    #####:  140:        sigchain_pop(signo);
    #####:  141:        raise(signo);
    #####:  142:}
    #####:  154:                return;
    #####:  211:        for_each_wanted_builtin(j, tgt_j) {
    #####:  212:                if (tgt_j->pfn_error_va_fl) {
    #####:  214:                        if (ap)
    #####:  215:                                va_copy(copy_ap, ap);
    #####:  216:                        tgt_j->pfn_error_va_fl(file, line, fmt, copy_ap);
    #####:  217:                        va_end(copy_ap);
    #####:  222:void trace2_cmd_path_fl(const char *file, int line, const char *pathname)
    #####:  227:        if (!trace2_enabled)
    #####:  228:                return;
    #####:  230:        for_each_wanted_builtin(j, tgt_j) {
    #####:  231:                if (tgt_j->pfn_command_path_fl)
    #####:  232:                        tgt_j->pfn_command_path_fl(file, line, pathname);
    #####:  244:        for_each_wanted_builtin(j, tgt_j) {
    #####:  245:                if (tgt_j->pfn_command_verb_fl)
    #####:  246:                        tgt_j->pfn_command_verb_fl(file, line, command_verb);
    #####:  259:        for_each_wanted_builtin(j, tgt_j) {
    #####:  260:                if (tgt_j->pfn_alias_fl)
    #####:  261:                        tgt_j->pfn_alias_fl(file, line, alias, argv);
    #####:  270:        tr2_cfg_list_config_fl(file, line);
    #####:  279:        tr2_cfg_set_fl(file, line, key, value);
    #####:  293:        us_now = getnanotime() / 1000;
    #####:  294:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  296:        cmd->trace2_child_id = tr2tls_locked_increment(&tr2_next_child_id);
    #####:  297:        cmd->trace2_child_us_start = us_now;
    #####:  299:        for_each_wanted_builtin(j, tgt_j) {
    #####:  300:                if (tgt_j->pfn_child_start_fl)
    #####:  301:                        tgt_j->pfn_child_start_fl(file, line,
    #####:  319:        us_now = getnanotime() / 1000;
    #####:  320:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  322:        if (cmd->trace2_child_us_start)
    #####:  323:                us_elapsed_child = us_now - cmd->trace2_child_us_start;
    #####:  325:                us_elapsed_child = 0;
    #####:  327:        for_each_wanted_builtin(j, tgt_j) {
    #####:  328:                if (tgt_j->pfn_child_exit_fl)
    #####:  329:                        tgt_j->pfn_child_exit_fl(
    #####:  347:        us_now = getnanotime() / 1000;
    #####:  348:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  350:        for_each_wanted_builtin(j, tgt_j) {
    #####:  351:                if (tgt_j->pfn_exec_fl)
    #####:  352:                        tgt_j->pfn_exec_fl(file, line, us_elapsed_absolute,
    #####:  367:        us_now = getnanotime() / 1000;
    #####:  368:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  370:        for_each_wanted_builtin(j, tgt_j) {
    #####:  371:                if (tgt_j->pfn_exec_result_fl)
    #####:  372:                        tgt_j->pfn_exec_result_fl(file, line,
    #####:  378:void trace2_thread_start_fl(const char *file, int line,
    #####:  386:        if (!trace2_enabled)
    #####:  387:                return;
    #####:  389:        if (tr2tls_is_main_thread())
    #####:  400:                trace2_region_enter_printf_fl(file, line, NULL, NULL, NULL,
    #####:  403:                return;
    #####:  406:        us_now = getnanotime() / 1000;
    #####:  407:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  409:        tr2tls_create_self(thread_name);
    #####:  411:        for_each_wanted_builtin(j, tgt_j) {
    #####:  412:                if (tgt_j->pfn_thread_start_fl)
    #####:  413:                        tgt_j->pfn_thread_start_fl(file, line,
    #####:  418:void trace2_thread_exit_fl(const char *file, int line)
    #####:  426:        if (!trace2_enabled)
    #####:  427:                return;
    #####:  429:        if (tr2tls_is_main_thread())
    #####:  440:                trace2_region_leave_printf_fl(file, line, NULL, NULL, NULL,
    #####:  442:                return;
    #####:  445:        us_now = getnanotime() / 1000;
    #####:  446:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  453:        tr2tls_pop_unwind_self();
    #####:  454:        us_elapsed_thread = tr2tls_region_elasped_self(us_now);
    #####:  456:        for_each_wanted_builtin(j, tgt_j) {
    #####:  457:                if (tgt_j->pfn_thread_exit_fl)
    #####:  458:                        tgt_j->pfn_thread_exit_fl(file, line,
    #####:  463:        tr2tls_unset_self();
    #####:  466:void trace2_def_param_fl(const char *file, int line,
    #####:  472:        if (!trace2_enabled)
    #####:  473:                return;
    #####:  475:        for_each_wanted_builtin(j, tgt_j) {
    #####:  476:                if (tgt_j->pfn_param_fl)
    #####:  477:                        tgt_j->pfn_param_fl(file, line, param, value);
    #####:  490:        if (repo->trace2_repo_id)
    #####:  491:                return;
    #####:  493:        repo->trace2_repo_id = tr2tls_locked_increment(&tr2_next_repo_id);
    #####:  495:        for_each_wanted_builtin(j, tgt_j) {
    #####:  496:                if (tgt_j->pfn_repo_fl)
    #####:  497:                        tgt_j->pfn_repo_fl(file, line, repo);
    #####:  515:        us_now = getnanotime() / 1000;
    #####:  516:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  522:        for_each_wanted_builtin(j, tgt_j) {
    #####:  523:                if (tgt_j->pfn_region_enter_printf_va_fl) {
    #####:  525:                        if (ap)
    #####:  526:                                va_copy(copy_ap, ap);
    #####:  527:                        tgt_j->pfn_region_enter_printf_va_fl(
    #####:  530:                        va_end(copy_ap);
    #####:  534:        tr2tls_push_self(us_now);
    #####:  592:        us_now = getnanotime() / 1000;
    #####:  593:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  601:        us_elapsed_region = tr2tls_region_elasped_self(us_now);
    #####:  603:        tr2tls_pop_self();
    #####:  605:        for_each_wanted_builtin(j, tgt_j) {
    #####:  606:                if (tgt_j->pfn_region_leave_printf_va_fl) {
    #####:  608:                        if (ap)
    #####:  609:                                va_copy(copy_ap, ap);
    #####:  610:                        tgt_j->pfn_region_leave_printf_va_fl(
    #####:  615:                        va_end(copy_ap);
    #####:  675:        us_now = getnanotime() / 1000;
    #####:  676:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  677:        us_elapsed_region = tr2tls_region_elasped_self(us_now);
    #####:  679:        for_each_wanted_builtin(j, tgt_j) {
    #####:  680:                if (tgt_j->pfn_data_fl)
    #####:  681:                        tgt_j->pfn_data_fl(file, line, us_elapsed_absolute,
    #####:  698:        strbuf_addf(&buf_string, "%"PRIdMAX, value);
    #####:  699:        trace2_data_string_fl(file, line, category, repo, key, buf_string.buf);
    #####:  700:        strbuf_release(&buf_string);
    #####:  703:void trace2_printf_va_fl(const char *file, int line,
    #####:  711:        if (!trace2_enabled)
    #####:  712:                return;
    #####:  714:        us_now = getnanotime() / 1000;
    #####:  715:        us_elapsed_absolute = tr2tls_absolute_elapsed(us_now);
    #####:  717:        for_each_wanted_builtin(j, tgt_j) {
    #####:  718:                if (tgt_j->pfn_printf_va_fl) {
    #####:  720:                        if (ap)
    #####:  721:                                va_copy(copy_ap, ap);
    #####:  722:                        tgt_j->pfn_printf_va_fl(file, line, us_elapsed_absolute,
    #####:  724:                        va_end(copy_ap);
    #####:  729:void trace2_printf_fl(const char *file, int line,
    #####:  734:        va_start(ap, fmt);
    #####:  735:        trace2_printf_va_fl(file, line, fmt, ap);
    #####:  736:        va_end(ap);
    #####:  737:}

Looks like there are a lot of methods that we haven't used in enough places to feel they are exercised.

Create a new unified tracing facility for git. The eventual intent is to replace the current trace_printf* and trace_performance* routines with a unified set of git_trace2* routines. In addition to the usual printf-style API, trace2 provides higer-level event verbs with fixed-fields allowing structured data to be written. This makes post-processing and analysis easier for external tools. Trace2 defines 3 output targets. These are set using the environment variables "GIT_TR2", "GIT_TR2_PERF", and "GIT_TR2_EVENT". These may be set to "1" or to an absolute pathname (just like the current GIT_TRACE). * GIT_TR2 is intended to be a replacement for GIT_TRACE and logs command summary data. * GIT_TR2_PERF is intended as a replacement for GIT_TRACE_PERFORMANCE. It extends the output with columns for the command process, thread, repo, absolute and relative elapsed times. It reports events for child process start/stop, thread start/stop, and per-thread function nesting. * GIT_TR2_EVENT is a new structured format. It writes event data as a series of JSON records. Calls to trace2 functions log to any of the 3 output targets enabled without the need to call different trace_printf* or trace_performance* routines. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Add trace2_region_enter() and trace2_region_leave() calls around the various phases of a status scan. This gives elapsed time for each phase in the GIT_TR2_PERF and GIT_TR2_EVENT trace target. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Classify editor, pager, and sub-process child processes. The former two can be used to identify interactive commands, for example. A later effort could classify other types of commands, such as "gc", "hook", and etc. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Add trace2 events when reading and writing the index. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Add trace2 region and data events describing attempts to deserialize status data using a status cache. A category:status, label:deserialize region is pushed around the deserialize code. Deserialization results when reading from a file are: category:status, path = <path> category:status, polled = <number_of_attempts> category:status, result = "ok" | "reject" When reading from STDIN are: category:status, path = "STDIN" category:status, result = "ok" | "reject" Status will fallback and run a normal status scan when a "reject" is reported (unless "--deserialize-wait=fail"). If "ok" is reported, status was able to use the status cache and avoid scanning the workdir. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

jeffhostetler · 2018-10-10T20:18:16Z

@derrickstolee I'm not surprised that we don't have coverage on some of those functions. I omitted thread-start/-exit calls for now. I'll address this more when I add unit tests for trace2.

kewillford

Haven't looked this over thoroughly.

kewillford · 2018-10-10T20:39:14Z

compat/mingw.c

+ */
+static int mingw_is_local_named_pipe_path(const char *filename)
+{
+	return (IS_SBS(filename[0]) &&


Check filename length before indexing?

…racked files Update the package for GitForWindows to include tracing information. Includes the following Git PRs: * [36 Avoid `sane_execvp` in `git rebase` and `git stash`](microsoft/git#36) * [34 Add Trace2 regions to 'pack-objects'](microsoft/git#34) * [28 Trace2 base plus GVFS extensions](microsoft/git#28) * [33 virtualfilesystem: check if directory is included](microsoft/git#33) * [31 Revert "gvfs: add a perf test for reading the index"](microsoft/git#31) * [32 compat/poll: prepare for targeting Windows Vista](microsoft/git#32) * [22 Enable the filesystem cache (fscache) in refresh_index()](microsoft/git#22) * [27 virtualfilesystem: fix bug with symlinks being ignored](microsoft/git#27) * [23 Unpack trees with cache tree gvfs](microsoft/git#23) * [15 virtualfilesystem: don't run the virtual file system hook if the index has been redirected](microsoft/git#15) * Includes Git 2.19.0. This also includes VFS for Git updates from Kevin Willford's work: **Fix missing untracked files when created in subfolder** Add to the test to create multiple level of folders and files in those folder to make sure they show up as untracked files. Update the git for windows version that has the fix. Fixes #358

(Experimental) Trace2 base plus GVFS extensions

jeffhostetler requested review from benpeart, derrickstolee, jamill, jrbriggs and kewillford September 26, 2018 20:07

benpeart reviewed Sep 27, 2018

View reviewed changes

jeffhostetler force-pushed the gvfs-trace2-v1 branch 6 times, most recently from 4d83a13 to d983757 Compare October 4, 2018 15:44

jeffhostetler force-pushed the gvfs-trace2-v1 branch 5 times, most recently from 001c214 to 7bb7dbe Compare October 8, 2018 20:50

derrickstolee reviewed Oct 9, 2018

View reviewed changes

jeffhostetler added 2 commits October 9, 2018 12:36

mingw: fix mingw_open_append to work with named pipes

09379ff

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

jeffhostetler force-pushed the gvfs-trace2-v1 branch from 7bb7dbe to 64b3251 Compare October 9, 2018 17:09

derrickstolee reviewed Oct 9, 2018

View reviewed changes

derrickstolee mentioned this pull request Oct 9, 2018

Add Trace2 regions to 'pack-objects' #34

Merged

jeffhostetler added 2 commits October 10, 2018 16:02

jeffhostetler added 5 commits October 10, 2018 16:02

trace2: add child classification for transport processes

52036b9

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

trace2: instrument reading and writing the index

8b6e8e9

Add trace2 events when reading and writing the index. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

gvfs:trace2: add trace2 tracing around read_object_process

8645713

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>

jeffhostetler force-pushed the gvfs-trace2-v1 branch from 92ee37e to 8645713 Compare October 10, 2018 20:08

kewillford approved these changes Oct 10, 2018

View reviewed changes

jeffhostetler merged commit 6ee82ac into microsoft:gvfs-2.19.0 Oct 10, 2018

kewillford reviewed Oct 10, 2018

View reviewed changes

derrickstolee mentioned this pull request Oct 11, 2018

Update Git to include tracing and fix for untracked files microsoft/VFSForGit#373

Merged

dscho pushed a commit that referenced this pull request Oct 15, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

d80b11e

(Experimental) Trace2 base plus GVFS extensions

dscho pushed a commit that referenced this pull request Oct 18, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

77e9be7

(Experimental) Trace2 base plus GVFS extensions

dscho pushed a commit that referenced this pull request Nov 21, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

f8dfa70

(Experimental) Trace2 base plus GVFS extensions

dscho pushed a commit that referenced this pull request Nov 26, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

d208dcd

(Experimental) Trace2 base plus GVFS extensions

dscho pushed a commit that referenced this pull request Dec 2, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

d79acfe

(Experimental) Trace2 base plus GVFS extensions

dscho pushed a commit that referenced this pull request Dec 10, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

68fb1f2

(Experimental) Trace2 base plus GVFS extensions

derrickstolee pushed a commit that referenced this pull request Dec 17, 2018

Merge pull request #28 from jeffhostetler/gvfs-trace2-v1

36e6b20

(Experimental) Trace2 base plus GVFS extensions

jeffhostetler mentioned this pull request Dec 21, 2018

Trace2 V3 (squash/fixup of V1+V2) #100

Merged

jeffhostetler deleted the gvfs-trace2-v1 branch August 16, 2019 21:46

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

(Experimental) Trace2 base plus GVFS extensions #28

(Experimental) Trace2 base plus GVFS extensions #28

jeffhostetler commented Sep 26, 2018 •

edited

Loading

benpeart left a comment

benpeart left a comment

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

benpeart Sep 27, 2018

jeffhostetler Oct 1, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

jeffhostetler Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

derrickstolee Oct 9, 2018

jeffhostetler Oct 10, 2018

derrickstolee commented Oct 9, 2018

derrickstolee commented Oct 9, 2018

jeffhostetler commented Oct 10, 2018

kewillford left a comment

kewillford Oct 10, 2018

(Experimental) Trace2 base plus GVFS extensions #28

(Experimental) Trace2 base plus GVFS extensions #28

Conversation

jeffhostetler commented Sep 26, 2018 • edited Loading

benpeart left a comment

Choose a reason for hiding this comment

benpeart left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

derrickstolee commented Oct 9, 2018

derrickstolee commented Oct 9, 2018

jeffhostetler commented Oct 10, 2018

kewillford left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jeffhostetler commented Sep 26, 2018 •

edited

Loading