linux-kernel - Re: [PATCH] buitin-kvm: Update cmd_kvm() to check perf

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Wed, 22 May 2024 21:23:30 -0700
From: Namhyung Kim <namhyung@...nel.org>
To: yskelg@...il.com
Cc: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 
	Arnaldo Carvalho de Melo <acme@...nel.org>, Mark Rutland <mark.rutland@....com>, 
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>, 
	Ian Rogers <irogers@...gle.com>, Adrian Hunter <adrian.hunter@...el.com>, 
	Austin Kim <austindh.kim@...il.com>, shjy180909@...il.com, 
	linux-perf-users@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] buitin-kvm: Update cmd_kvm() to check perf_host value

Hello,

On Wed, May 15, 2024 at 12:54 PM <yskelg@...il.com> wrote:
>
> From: Yunseong Kim <yskelg@...il.com>
>
> According to perf-kvm.txt, the default operation is --guest, meaning
> perf_host should be set to false and perf_guest to true. Currently,
> the incorrect usage of --no-host and --no-guest still results in
> the default --guest operation.

Hmm, right.. This is not good.

>
> This patch updates the usage messages to include the --no-guest option
> for those who only want host data, as well as the --no-host and
> --no-guest case.

I'm not sure if it's worth mentioning the host-only use case which is
not expected for this command.  I think it's enough to warn the
no-host and no-guest use cases.

>
> Signed-off-by: Yunseong Kim <yskelg@...il.com>
> ---
>  tools/perf/builtin-kvm.c | 19 +++++++++++++------
>  1 file changed, 13 insertions(+), 6 deletions(-)
>
> diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
> index 71165036e4ca..654a07658cdd 100644
> --- a/tools/perf/builtin-kvm.c
> +++ b/tools/perf/builtin-kvm.c
> @@ -2129,7 +2129,7 @@ int cmd_kvm(int argc, const char **argv)
>                 OPT_BOOLEAN(0, "guest", &perf_guest,
>                             "Collect guest os data"),
>                 OPT_BOOLEAN(0, "host", &perf_host,
> -                           "Collect host os data"),
> +                          "Collect host os data. Host only with --host --no-guest"),
>                 OPT_STRING(0, "guestmount", &symbol_conf.guestmount, "directory",
>                            "guest mount directory under which every guest os"
>                            " instance has a subdir"),
> @@ -2150,16 +2150,23 @@ int cmd_kvm(int argc, const char **argv)
>                                                 "buildid-list", "stat", NULL };
>         const char *kvm_usage[] = { NULL, NULL };
>
> -       perf_host  = 0;
> -       perf_guest = 1;
> +       /*
> +        * tools/perf/Documentation/perf-kvm.txt
> +        *
> +        *    Default('')         ->  perf.data.guest
> +        *    --host              ->  perf.data.kvm
> +        *    --guest             ->  perf.data.guest
> +        *    --host --guest      ->  perf.data.kvm
> +        *    --host --no-guest   ->  perf.data.host
> +        */
> +       perf_host = false;
> +       perf_guest = true;

Probably not needed.

>
>         argc = parse_options_subcommand(argc, argv, kvm_options, kvm_subcommands, kvm_usage,
>                                         PARSE_OPT_STOP_AT_NON_OPTION);
> -       if (!argc)
> +       if (!argc || (!perf_host && !perf_guest))
>                 usage_with_options(kvm_usage, kvm_options);
>
> -       if (!perf_host)
> -               perf_guest = 1;

Instead of showing the whole usage, I think it's better to
print a specific message for this case.

Oh, I've realized that we have 'h' modifier for hypervisors
(on ARM64?)  Then maybe no-host + no-guest is a valid
combination.  Probably we need to add --hypervisor option
then.

Thanks,
Namhyung


>
>         if (!file_name) {
>                 file_name = get_filename_for_perf_kvm();
> --
> 2.34.1
>