[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20131216131628.GA2368@infradead.org>
Date: Mon, 16 Dec 2013 11:16:28 -0200
From: Arnaldo Carvalho de Melo <acme@...stprotocols.net>
To: Ramkumar Ramachandra <artagnon@...il.com>
Cc: LKML <linux-kernel@...r.kernel.org>,
David Ahern <dsahern@...il.com>
Subject: Re: [PATCH] perf kvm: introduce --list-cmds for use by scripts
Em Thu, Dec 12, 2013 at 10:56:51PM +0530, Ramkumar Ramachandra escreveu:
> Introduce
>
> $ perf kvm --list-cmds
>
> to dump a raw list of commands for use by the completion script. While
> at it, refactor kvm_usage so that there's only one copy of the command
> listing.
>
> Cc: David Ahern <dsahern@...il.com>
> Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
> Signed-off-by: Ramkumar Ramachandra <artagnon@...il.com>
> ---
> David Ahern wrote:
> > That would work -- perhaps a #define or string near
> >
> > const char * const kvm_usage[] = {
> > "perf kvm [<options>] {top|record|report|diff|buildid-list|stat}",
> > NULL
> > };
> >
> > Building kvm_usage from the string would better - only 1 place listing the
> > commands.
>
> Something like this, perhaps? It's not too pretty though: do you have
> suggestions to prettify it?
Yes:
Don't do all those things open coded, introduce functions to print,
concat, etc.
The best thing tho, since we have all those sub sub commands in things
like 'perf kvm', 'perf bench', etc, we could have some
parse_options_subcmd, and make the parse options machinery aware of
this, so that it could receive an array of subcmds and when asked for
--list-cmds, would print that sublist, etc, i.e. make sub cmds a first
class citizen.
So I'd suggest that you first introduce functions for doing the concat
to pass to the current infrastructure, so that we have what your patch
provides, but prettified, then, as follow on patches, you could work on
making the options parsing machinery aware of sub cmds.
Ah, and try not using fixed sized arrays, or at least verify that space
is available, i.e. never use strcat, use strncat, better, take a look at
tools/perf/util/strbuf.h, I guess you can use it to build the string for
you in a safe way and expanding the buffer as needed, etc.
- Arnaldo
> tools/perf/builtin-kvm.c | 25 +++++++++++++++++++++----
> tools/perf/perf-completion.sh | 2 +-
> 2 files changed, 22 insertions(+), 5 deletions(-)
>
> diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
> index c6fa3cb..ce44a9b 100644
> --- a/tools/perf/builtin-kvm.c
> +++ b/tools/perf/builtin-kvm.c
> @@ -1672,6 +1672,7 @@ __cmd_buildid_list(const char *file_name, int argc, const char **argv)
> int cmd_kvm(int argc, const char **argv, const char *prefix __maybe_unused)
> {
> const char *file_name = NULL;
> + bool list_cmds = false;
> const struct option kvm_options[] = {
> OPT_STRING('i', "input", &file_name, "file",
> "Input file name"),
> @@ -1692,20 +1693,36 @@ int cmd_kvm(int argc, const char **argv, const char *prefix __maybe_unused)
> "file", "file saving guest os /proc/modules"),
> OPT_INCR('v', "verbose", &verbose,
> "be more verbose (show counter open errors, etc)"),
> + OPT_BOOLEAN(0, "list-cmds", &list_cmds,
> + "list commands raw for use by scripts"),
> OPT_END()
> };
>
> + const char *const commands[] = { "top", "record", "report", "diff",
> + "buildid-list", "stat", NULL };
> + char kvm_usage_str[80];
> + const char *kvm_usage[] = { NULL, NULL };
>
> - const char * const kvm_usage[] = {
> - "perf kvm [<options>] {top|record|report|diff|buildid-list|stat}",
> - NULL
> - };
> + sprintf(kvm_usage_str, "%s", "perf kvm [<options>] {");
> + for (int i = 0; commands[i]; i++) {
> + if (i)
> + strcat(kvm_usage_str, "|");
> + strcat(kvm_usage_str, commands[i]);
> + }
> + strcat(kvm_usage_str, "}");
> +
> + kvm_usage[0] = kvm_usage_str;
>
> perf_host = 0;
> perf_guest = 1;
>
> argc = parse_options(argc, argv, kvm_options, kvm_usage,
> PARSE_OPT_STOP_AT_NON_OPTION);
> + if (list_cmds) {
> + for (int i = 0; commands[i]; i++)
> + printf("%s ", commands[i]);
> + return 0;
> + }
> if (!argc)
> usage_with_options(kvm_usage, kvm_options);
>
> diff --git a/tools/perf/perf-completion.sh b/tools/perf/perf-completion.sh
> index 496e2ab..d8bfa43 100644
> --- a/tools/perf/perf-completion.sh
> +++ b/tools/perf/perf-completion.sh
> @@ -123,7 +123,7 @@ __perf_main ()
> __perfcomp_colon "$evts" "$cur"
> # List subcommands for 'perf kvm'
> elif [[ $prev == "kvm" ]]; then
> - subcmds="top record report diff buildid-list stat"
> + subcmds=$($cmd kvm --list-cmds)
> __perfcomp_colon "$subcmds" "$cur"
> # List long option names
> elif [[ $cur == --* ]]; then
> --
> 1.8.5.1.113.g8cb5bef.dirty
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists