Re: [PATCH v4 3/4] perf util: Flexible to set block info output formats

From: Jin, Yao
Date: Tue Jan 28 2020 - 05:27:10 EST




On 1/28/2020 5:21 PM, Jiri Olsa wrote:
On Mon, Jan 27, 2020 at 08:31:23PM +0800, Jin, Yao wrote:


On 1/20/2020 11:00 PM, Jin, Yao wrote:


On 1/20/2020 5:47 PM, Jiri Olsa wrote:
On Thu, Jan 16, 2020 at 03:29:03AM +0800, Jin Yao wrote:

SNIP

+ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ block_hpps, nr_hpps);
-ÂÂÂ perf_hpp_list__register_sort_field(&bh->block_list,
-ÂÂÂÂÂÂÂ &block_fmts[PERF_HPP_REPORT__BLOCK_TOTAL_CYCLES_PCT].fmt);
+ÂÂÂ /* Sort by the first fmt */
+ÂÂÂ perf_hpp_list__register_sort_field(&bh->block_list,
&block_fmts[0].fmt);
 }
-static void process_block_report(struct hists *hists,
-ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ struct block_report *block_report,
-ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ u64 total_cycles)
+static int process_block_report(struct hists *hists,
+ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ struct block_report *block_report,
+ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ u64 total_cycles, int *block_hpps,
+ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ int nr_hpps)
 {
ÂÂÂÂÂ struct rb_node *next = rb_first_cached(&hists->entries);
ÂÂÂÂÂ struct block_hist *bh = &block_report->hist;
ÂÂÂÂÂ struct hist_entry *he;
-ÂÂÂ init_block_hist(bh, block_report->fmts);
+ÂÂÂ if (nr_hpps > PERF_HPP_REPORT__BLOCK_MAX_INDEX)

hum, should be '>=' above.. ?

jirka


'=' should be OK.

enum {
ÂÂÂÂPERF_HPP_REPORT__BLOCK_TOTAL_CYCLES_PCT,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_LBR_CYCLES,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_CYCLES_PCT,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_AVG_CYCLES,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_RANGE,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_DSO,
ÂÂÂÂPERF_HPP_REPORT__BLOCK_MAX_INDEX
};

PERF_HPP_REPORT__BLOCK_MAX_INDEX is 6.

If nr_hpps is 6, for example, block_hpps[] is,

ÂÂÂÂÂÂÂ int block_hpps[6] = {
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_TOTAL_CYCLES_PCT,
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_LBR_CYCLES,
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_CYCLES_PCT,
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_AVG_CYCLES,
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_RANGE,
ÂÂÂÂÂÂÂÂÂÂÂ PERF_HPP_REPORT__BLOCK_DSO,
ÂÂÂÂÂÂÂ };

ÂÂÂÂÂÂÂ block_info__create_report(session->evlist,
ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ rep->total_cycles,
ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ block_hpps, 6,
ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ &rep->nr_block_reports);

That should be legal.

Thanks
Jin Yao

Hi Jiri,

Does this explanation make sense?

ok, make sense

thanks,
jirka


Thanks Jiri!

I will post v5 which only contains the change of dropping the unnecessary release code.

Thanks
Jin Yao