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: <20150502153814.GA21242@krava.redhat.com>
Date:	Sat, 2 May 2015 17:38:14 +0200
From:	Jiri Olsa <jolsa@...hat.com>
To:	Sukadev Bhattiprolu <sukadev@...ux.vnet.ibm.com>
Cc:	mingo@...hat.com, ak@...ux.intel.com,
	Michael Ellerman <mpe@...erman.id.au>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Paul Mackerras <paulus@...ba.org>,
	linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC][PATCH 4/4] perf: Create aliases for PMU events

On Fri, May 01, 2015 at 12:05:41AM -0700, Sukadev Bhattiprolu wrote:

SNIP

> +
> +static int pmu_add_cpu_aliases(char *cpustr, void *data)
> +{
> +	struct list_head *head = (struct list_head *)data;
> +	struct perf_pmu_alias *alias;
> +	int i;
> +	struct perf_pmu_event *events_table, *event;
> +	struct parse_events_term *term;
> +
> +	events_table = arch_get_events_table(cpustr);
> +	if (!events_table)
> +		return 0;
> +
> +	for (i = 0; events_table[i].name != NULL; i++) {
> +		event = &events_table[i];
> +
> +		alias = malloc(sizeof(*alias));
> +		if (!alias)
> +			return -ENOMEM;
> +
> +		term = malloc(sizeof(*term));
> +		if (!term) {
> +			/*
> +			 * TODO: cleanup aliases allocated so far?
> +			 */
> +			free(alias);
> +			return -ENOMEM;
> +		}
> +
> +		/* ->config is not const; cast to override */
> +		term->config = (char *)"event";
> +		term->val.num = event->code;
> +		term->type_val = PARSE_EVENTS__TERM_TYPE_NUM;
> +		term->type_term = PARSE_EVENTS__TERM_TYPE_USER;
> +		INIT_LIST_HEAD(&term->list);
> +		term->used = 0;

hum, so I checked and for x86 the 'struct perf_pmu_event::code' is not enough

Andi introduced following JSON notation for event:

[
  {
    "EventCode": "0x00",
    "UMask": "0x01",
    "EventName": "INST_RETIRED.ANY",
    "BriefDescription": "Instructions retired from execution.",
    "PublicDescription": "Instructions retired from execution.",
    "Counter": "Fixed counter 1",
    "CounterHTOff": "Fixed counter 1",
    "SampleAfterValue": "2000003",
    "MSRIndex": "0",
    "MSRValue": "0",
    "TakenAlone": "0",
    "CounterMask": "0",
    "Invert": "0",
    "AnyThread": "0",
    "EdgeDetect": "0",
    "PEBS": "0",
    "PRECISE_STORE": "0",
    "Errata": "null",
    "Offcore": "0"
  }

which gets processed/translated into string of terms "event=..,umask=,..."
which is used to create alias 'struct perf_pmu_alias'

please check Andi's patch:
  [PATCH v9 04/11] perf, tools: Add support for reading JSON event files

  function json_events


I wonder we should stay with JSON description during build time
translate all (for architecture) events into strings in term format
"event=..,umask=,...'  and this array of string would be loaded
as aliases in runtime

so we would have architecture specific tool that would translate
JSON events data into array of strings (events in terms form for
given architecture) which would get compiled into perf

I personaly like having set of event files in JSON notation
rather than having them directly in C structure

thoughts?

jirka
--
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

Powered by Openwall GNU/*/Linux Powered by OpenVZ