Re: [PATCH v7] perf stat: Fix wrong skipping for per-die aggregation

[Jiri Olsa <jolsa@redhat.com>]

On Mon, Jan 18, 2021 at 12:05:21PM +0800, Jin Yao wrote:

SNIP

> diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
> index 8ce1479c98f0..5aba8fa92386 100644
> --- a/tools/perf/util/stat.c
> +++ b/tools/perf/util/stat.c
> @@ -13,6 +13,7 @@
>  #include "evlist.h"
>  #include "evsel.h"
>  #include "thread_map.h"
> +#include "hashmap.h"
>  #include <linux/zalloc.h>
>  
>  void update_stats(struct stats *stats, u64 val)
> @@ -276,15 +277,27 @@ void evlist__save_aggr_prev_raw_counts(struct evlist *evlist)
>  static void zero_per_pkg(struct evsel *counter)
>  {
>  	if (counter->per_pkg_mask)
> -		memset(counter->per_pkg_mask, 0, cpu__max_cpu());
> +		hashmap__clear(counter->per_pkg_mask);
> +}
> +
> +static size_t pkg_id_hash(const void *key, void *ctx __maybe_unused)
> +{
> +	return (size_t)key & 0xffff;
> +}
> +
> +static bool pkg_id_equal(const void *key1, const void *key2,
> +			 void *ctx __maybe_unused)
> +{
> +	return (size_t)key1 == (size_t)key2;
>  }
>  
>  static int check_per_pkg(struct evsel *counter,
>  			 struct perf_counts_values *vals, int cpu, bool *skip)
>  {
> -	unsigned long *mask = counter->per_pkg_mask;
> +	struct hashmap *mask = counter->per_pkg_mask;
>  	struct perf_cpu_map *cpus = evsel__cpus(counter);
> -	int s;
> +	int s, d, ret = 0;
> +	size_t key;

please use uint32_t to make it obvious

>  
>  	*skip = false;
>  
> @@ -295,7 +308,7 @@ static int check_per_pkg(struct evsel *counter,
>  		return 0;
>  
>  	if (!mask) {
> -		mask = zalloc(cpu__max_cpu());
> +		mask = hashmap__new(pkg_id_hash, pkg_id_equal, NULL);
>  		if (!mask)
>  			return -ENOMEM;
>  
> @@ -317,8 +330,21 @@ static int check_per_pkg(struct evsel *counter,
>  	if (s < 0)
>  		return -1;
>  
> -	*skip = test_and_set_bit(s, mask) == 1;
> -	return 0;
> +	/*
> +	 * On multi-die system, die_id > 0. On no-die system, die_id = 0.
> +	 * We use hashmap(socket, die) to check the used socket+die pair.
> +	 */
> +	d = cpu_map__get_die(cpus, cpu, NULL).die;
> +	if (d < 0)
> +		return -1;
> +
> +	key = (size_t)d << 16 | s;

I'm not sure about the socket number bounds, but I guess we should at
least check that it's not over the limit

how hard would it be to allocate key values and keep the uint64_t?

thanks,
jirka