linux-kernel - Re: [PATCH V2 3/5] perf stat: Support per-die aggregation

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: <cc4eea9a-e18e-af7e-2290-98727fce2f6a@linux.intel.com>
Date:   Mon, 3 Jun 2019 13:42:58 -0400
From:   "Liang, Kan" <kan.liang@...ux.intel.com>
To:     Jiri Olsa <jolsa@...hat.com>
Cc:     acme@...nel.org, jolsa@...nel.org, mingo@...hat.com,
        linux-kernel@...r.kernel.org, peterz@...radead.org,
        ak@...ux.intel.com
Subject: Re: [PATCH V2 3/5] perf stat: Support per-die aggregation



On 6/3/2019 12:36 PM, Jiri Olsa wrote:
> On Thu, May 30, 2019 at 07:53:47AM -0700, kan.liang@...ux.intel.com wrote:
> 
> SNIP
> 
>> +
>>   static int perf_env__get_core(struct cpu_map *map, int idx, void *data)
>>   {
>>   	struct perf_env *env = data;
>>   	int core = -1, cpu = perf_env__get_cpu(env, map, idx);
>>   
>>   	if (cpu != -1) {
>> -		int socket_id = env->cpu[cpu].socket_id;
>> -
>>   		/*
>> -		 * Encode socket in upper 16 bits
>> -		 * core_id is relative to socket, and
>> +		 * Encode socket in upper 24 bits
> 
> please note we use upper 8 bits for socket number,
> the comments suggests it's 24 bits

I will fix it in v3.

> 
>> +		 * encode die id in upper 16 bits
>> +		 * core_id is relative to socket and die,
>>   		 * we need a global id. So we combine
>> -		 * socket + core id.
>> +		 * socket + die id + core id
>>   		 */
>> -		core = (socket_id << 16) | (env->cpu[cpu].core_id & 0xffff);
>> +		if (WARN_ONCE(env->cpu[cpu].socket_id >> 8,
>> +		    "The socket_id number is too big. Please upgrade the perf tool.\n"))
> 
> hum, how's perf tool upgrade going to help in here?


I think there is nothing we can do for current encoding, if the socket 
number or die number is bigger than 8 bits. We have to design a new 
encoding method, which needs to update the perf tool.
So I use a similar expression from process_cpu_topology().
		if (do_core_id_test && nr != (u32)-1 && nr > (u32)cpu_nr) {
			pr_debug("socket_id number is too big."
				 "You may need to upgrade the perf tool.\n");
			goto free_cpu;
		}

Any suggestions for the warning message?



> 
>> +			return -1;
>> +
>> +		if (WARN_ONCE(env->cpu[cpu].die_id >> 8,
>> +		    "The die_id number is too big. Please upgrade the perf tool.\n"))
>> +			return -1;
>> +
>> +		core = (env->cpu[cpu].socket_id << 24) |
>> +		       (env->cpu[cpu].die_id << 16) |
>> +		       (env->cpu[cpu].core_id & 0xffff);
>>   	}
> 
> other than comments above, the patchset looks good to me
>

Thanks for the review. I will send out V3 after the comments are addressed.

Thanks,
Kan