Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1571290

Re: [PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option

From Taeung Song <treeze.taeung@gmail.com>
Newsgroups linux.kernel
Subject Re: [PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option
Date 2017-02-01 08:30 +0100
Message-ID <t5XwB-2To-3@gated-at.bofh.it> (permalink)
References <t5EX0-8E-3@gated-at.bofh.it> <t5EX0-8E-1@gated-at.bofh.it> <t5GvN-19j-41@gated-at.bofh.it> <t5GvO-19j-57@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw



On 01/31/2017 10:15 PM, Arnaldo Carvalho de Melo wrote:
> Em Tue, Jan 31, 2017 at 09:46:28AM -0300, Arnaldo Carvalho de Melo escreveu:
>> Also you are silently ignoring any unknown variable in this section, so
>> if someone has this:
>>
>> cat ~/.perfconfig
>>
>> [ftrace]
>>
>> 	trace = function
>>
>>
>> I.e. forgets the 'r', it will keep using the current default,
>> "function_graph" till the user checks the config and adds that missing
>> 'r'...
>>
>> I haven't changed this because I think this is so common that we should
>> instead have a different error return for those config callbacks that
>> will tell perf_config() to say something like:
>>
>>   "unknown variable %s.%s\n", section_name, variable_name
>
> So I read a bit more the config code and ended up with this:
>
> static int perf_ftrace_config(const char *var, const char *value, void *cb)
> {
>         struct perf_ftrace *ftrace = cb;
>
>         if (prefixcmp(var, "ftrace."))
>                 return 0;
>
>         if (strcmp(var, "ftrace.tracer"))
>                 return -1;
>
>         if (!strcmp(value, "function_graph") ||
>             !strcmp(value, "function")) {
>                 ftrace->tracer = value;
>                 return 0;
>         }
>
> 	pr_err("Please select \"function_graph\" (default) or \"function\"\n");
>         return -1;
> }
>
> I.e. it will ignore variables for other sections and will stop at an unknown
> variable or value for ftrace.tracer:
>
> [root@jouet ~]# cat ~/.perfconfig
> [ftrace]
> 	trace = function
> [root@jouet ~]# perf ftrace usleep 1
> Error: wrong config key-value pair ftrace.trace=function
> [root@jouet ~]# cat ~/.perfconfig
> [ftrace]
> 	tracer = functin
> [root@jouet ~]# perf ftrace usleep 1
> Please select "function_graph" (default) or "function"
> Error: wrong config key-value pair ftrace.tracer=functin
> [root@jouet ~]# cat ~/.perfconfig
> [ftrace]
> 	tracer = function
> [root@jouet ~]# perf ftrace usleep 1 | head -5
>           <idle>-0     [000] d...  3855.820847: switch_mm_irqs_off <-__schedule
>            <...>-18550 [000] d...  3855.820849: finish_task_switch <-__schedule
>            <...>-18550 [000] d...  3855.820851: smp_irq_work_interrupt <-irq_work_interrupt
>            <...>-18550 [000] d...  3855.820851: irq_enter <-smp_irq_work_interrupt
>            <...>-18550 [000] d...  3855.820851: rcu_irq_enter <-irq_enter
> [root@jouet ~]#
>

This is great!! I agonized about error message for
each other cases as above! But I think this is more precise than my patch!

Thanks,
Taeung

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option Taeung Song <treeze.taeung@gmail.com> - 2017-01-31 12:40 +0100
  Re: [PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option Arnaldo Carvalho de Melo <acme@kernel.org> - 2017-01-31 14:20 +0100
    Re: [PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option Arnaldo Carvalho de Melo <acme@kernel.org> - 2017-01-31 14:20 +0100
      Re: [PATCH v2 2/4] perf ftrace: Add ftrace.tracer config option Taeung Song <treeze.taeung@gmail.com> - 2017-02-01 08:30 +0100
  [tip:perf/core] perf ftrace: Add ftrace.tracer config option tip-bot for Taeung Song <tipbot@zytor.com> - 2017-02-01 15:50 +0100

csiph-web