linux-kernel - Re: [PATCHv2 1/2] perf tools: Collect tracing event data files directly

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]

Message-ID: <1317044192.26514.1.camel@gandalf.stny.rr.com>
Date:	Mon, 26 Sep 2011 09:36:31 -0400
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Jiri Olsa <jolsa@...hat.com>
Cc:	acme@...hat.com, eric.dumazet@...il.com, fweisbec@...il.com,
	a.p.zijlstra@...llo.nl, mingo@...e.hu, paulus@...ba.org,
	linux-kernel@...r.kernel.org, nhorman@...driver.com
Subject: Re: [PATCHv2 1/2] perf tools: Collect tracing event data files
 directly

On Mon, 2011-09-26 at 11:11 +0200, Jiri Olsa wrote:
> Changing the way the event files are searched by quering specified
> event files directly, instead of walking the events directory.
> 
> Hopefully this way is more straightforward and faster.

Have you looked at my code I posted earlier that uses the libparsevents?

It uses globs such that you could do -e sched:sched* and it will enable
all sched events.

-- Steve



> +	 * and one safety byte ;)
> +	 */
> +	file = malloc_or_die(strlen(path) + strlen(sys) +
> +			     strlen(name) + strlen("format") + 5);
>  
> -		free(format);
> -	}
> -	closedir(dir);
> +	sprintf(file, "%s/%s/%s/format", path, sys, name);
> +	return file;
>  }
>  
> -static void read_ftrace_files(struct tracepoint_path *tps)
> +static void put_format_file(char *file)
>  {
> -	char *path;
> -
> -	path = get_tracing_file("events/ftrace");
> -
> -	copy_event_system(path, tps);
> -
> -	put_tracing_file(path);
> +	free(file);
>  }
>  
> -static bool system_in_tp_list(char *sys, struct tracepoint_path *tps)
> +/*
> + * Walk tracepoint event objects and store them.
> + * Only those matching the sys parameter are stored
> + * and marked as done.
> + */
> +static void read_event_files_system(struct tracepoint_path *tps,
> +				    const char *sys)
>  {
> +	off_t count_pos;
> +	u32 count = 0;
> +
> +	/* Place for number of events under single system. */
> +	count_pos = lseek(output_fd, 0, SEEK_CUR);
> +	write_or_die(&count, 4);
> +
>  	while (tps) {
> -		if (!strcmp(sys, tps->system))
> -			return true;
> +		if ((!tps->done) &&
> +		    (!strcmp(sys, tps->system))) {
> +			char *file;
> +
> +			file = get_format_file(tps->system, tps->name);
> +			if (file) {
> +				record_file(file, 8);
> +				count++;
> +			}
> +
> +			put_format_file(file);
> +			tps->done = 1;
> +		}
> +
>  		tps = tps->next;


--
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/