git/trace.c, branch v2.22.2

trace.h: support nested performance tracing

2018-08-18T16:47:46Z

Performance measurements are listed right now as a flat list, which is fine when we measure big blocks. But when we start adding more and more measurements, some of them could be just part of a bigger measurement and a flat list gives a wrong impression that they are executed at the same level instead of nested. Add trace_performance_enter() and trace_performance_leave() to allow indent these nested measurements. For now it does not help much because the only nested thing is (lazy) name hash initialization (e.g. called in diff-index from "git status"). This will help more because I'm going to add some more tracing that's actually nested. Signed-off-by: Nguyễn Thái Ngọc Duy Signed-off-by: Junio C Hamano

trace.c: export trace_setup_key

2018-03-30T19:49:57Z

This is so that we can print traces based on this key outside trace.c. Signed-off-by: Nguyễn Thái Ngọc Duy Signed-off-by: Jeff King Signed-off-by: Junio C Hamano

trace.c: move strbuf_release() out of print_trace_line()

2018-01-16T20:16:54Z

The function is about printing a trace line, not releasing the buffer it receives too. Move strbuf_release() back outside. This makes it easier to see how strbuf is managed. Signed-off-by: Nguyễn Thái Ngọc Duy Signed-off-by: Junio C Hamano

trace: avoid unnecessary quoting

2018-01-16T20:16:54Z

Trace output which contains arbitrary strings (e.g., the arguments to commands which we are running) is always passed through sq_quote_buf(). That function always adds single-quotes, even if the output consists of vanilla characters. This can make the output a bit hard to read. Let's avoid the quoting if there are no characters which a shell would interpret. Trace output doesn't necessarily need to be shell-compatible, but: - the shell language is a good ballpark for what humans consider readable (well, humans versed in command line tools) - the run_command bits can be cut-and-pasted to a shell, and we'll keep that property - it covers any cases which would make the output visually ambiguous (e.g., embedded whitespace or quotes) Signed-off-by: Jeff King Signed-off-by: Nguyễn Thái Ngọc Duy Signed-off-by: Junio C Hamano

sq_quote_argv: drop maxlen parameter

2018-01-16T20:16:54Z

No caller passes anything but "0" for this parameter, which requests that the function ignore it completely. In fact, in all of history there was only one such caller, and it went away in 7f51f8bc2b (alias: use run_command api to execute aliases, 2011-01-07). Signed-off-by: Jeff King Signed-off-by: Nguyễn Thái Ngọc Duy Signed-off-by: Junio C Hamano

trace: improve performance while category is disabled

2017-12-06T17:43:02Z

Move just enough code from trace.c into trace.h header so all code necessary to determine that trace is disabled could be inlined to calling functions. Then perform the check if the trace key is enabled sooner in call chain. Signed-off-by: Gennady Kupava Signed-off-by: Junio C Hamano

trace: remove trace key normalization

2017-11-27T01:19:08Z

Trace key normalization is not used, not strictly necessary, complicates the code and would negatively affect compilation speed if moved to header. New trace_default_key key or existing separate marco could be used instead of passing NULL as a key. Signed-off-by: Gennady Kupava Signed-off-by: Junio C Hamano

Replace Free Software Foundation address in license notices

2017-11-09T04:21:21Z

The mailing address for the FSF has changed over the years. Rather than updating the address across all files, refer readers to gnu.org, as the GNU GPL documentation now suggests for license notices. The mailing address is retained in the full license files (COPYING and LGPL-2.1). The old address is still present in t/diff-lib/COPYING. This is intentional, as the file is used in tests and the contents are not expected to change. Signed-off-by: Todd Zullinger Signed-off-by: Junio C Hamano

trace: do not fall back to stderr

2016-08-05T16:28:17Z

If the trace code cannot open a specified file, or does not understand the contents of the GIT_TRACE variable, it falls back to printing trace output to stderr. This is an attempt to be helpful, but in practice it just ends up annoying. The user was trying to get the output to go somewhere else, so spewing it to stderr does not really accomplish that. And as it's intended for debugging, they can presumably re-run the command with their error corrected. So instead of falling back, this patch disables bogus trace keys for the rest of the program, just as we do for write errors. We can drop the "Defaulting to..." part of the error message entirely; after seeing "cannot open '/foo'", the user can assume that tracing is skipped. Signed-off-by: Jeff King Signed-off-by: Junio C Hamano

trace: disable key after write error

2016-08-05T16:28:16Z

If we get a write error writing to a trace descriptor, the error isn't likely to go away if we keep writing. Instead, you'll just get the same error over and over. E.g., try: GIT_TRACE_PACKET=42 git ls-remote >/dev/null You don't really need to see: warning: unable to write trace for GIT_TRACE_PACKET: Bad file descriptor hundreds of times. We could fallback to tracing to stderr, as we do in the error code-path for open(), but there's not much point. If the user fed us a bogus descriptor, they're probably better off fixing their invocation. And if they didn't, and we saw a transient error (e.g., ENOSPC writing to a file), it probably doesn't help anybody to have half of the trace in a file, and half on stderr. Signed-off-by: Jeff King Signed-off-by: Junio C Hamano