linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>,
	Ian Rogers <irogers@google.com>,
	Kan Liang <kan.liang@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	linux-perf-users@vger.kernel.org,
	Stephane Eranian <eranian@google.com>,
	Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Subject: [PATCH 6/6] perf mem: Add -T/--data-type option to report subcommand
Date: Wed, 31 Jul 2024 16:55:05 -0700	[thread overview]
Message-ID: <20240731235505.710436-7-namhyung@kernel.org> (raw)
In-Reply-To: <20240731235505.710436-1-namhyung@kernel.org>

This is just a shortcut to have 'type' in the sort key and use more
compact output format like below.

  $ perf mem report -T
  ...
  #
  # Overhead       Samples  Memory access                            Snoop         TLB access              Data Type
  # ........  ............  .......................................  ............  ......................  .........
  #
      14.84%            22  L1 hit                                   None          L1 or L2 hit            (unknown)
       7.68%             8  LFB/MAB hit                              None          L1 or L2 hit            (unknown)
       7.17%             3  RAM hit                                  Hit           L2 miss                 (unknown)
       6.29%            12  L1 hit                                   None          L1 or L2 hit            (stack operation)
       4.85%             5  RAM hit                                  Hit           L1 or L2 hit            (unknown)
       3.97%             5  LFB/MAB hit                              None          L1 or L2 hit            struct psi_group_cpu
       3.18%             3  LFB/MAB hit                              None          L1 or L2 hit            (stack operation)
       2.58%             3  L1 hit                                   None          L1 or L2 hit            unsigned int
       2.36%             2  L1 hit                                   None          L1 or L2 hit            struct
       2.31%             2  L1 hit                                   None          L1 or L2 hit            struct psi_group_cpu
  ...

Users also can use their own sort keys and -T option makes sure it has
the 'type' sort key at the end.

  $ perf mem report -T -s mem

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/builtin-mem.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/tools/perf/builtin-mem.c b/tools/perf/builtin-mem.c
index 24a4f0084f49..efa700d14c98 100644
--- a/tools/perf/builtin-mem.c
+++ b/tools/perf/builtin-mem.c
@@ -38,6 +38,7 @@ struct perf_mem {
 	bool			data_page_size;
 	bool			all_kernel;
 	bool			all_user;
+	bool			data_type;
 	int			operation;
 	const char		*cpu_list;
 	DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
@@ -317,6 +318,8 @@ static char *get_sort_order(struct perf_mem *mem)
 
 	if (mem->sort_key)
 		scnprintf(sort, sizeof(sort), "--sort=%s", mem->sort_key);
+	else if (mem->data_type)
+		strcpy(sort, "--sort=mem,snoop,tlb,type");
 	/*
 	 * there is no weight (cost) associated with stores, so don't print
 	 * the column
@@ -336,6 +339,10 @@ static char *get_sort_order(struct perf_mem *mem)
 	if (mem->data_page_size)
 		strcat(sort, ",data_page_size");
 
+	/* make sure it has 'type' sort key even -s option is used */
+	if (mem->data_type && !strstr(sort, "type"))
+		strcat(sort, ",type");
+
 	return strdup(sort);
 }
 
@@ -508,6 +515,8 @@ int cmd_mem(int argc, const char **argv)
 		   " between columns '.' is reserved."),
 	OPT_STRING('s', "sort", &mem.sort_key, "key[,key2...]",
 		   sort_order_help),
+	OPT_BOOLEAN('T', "type-profile", &mem.data_type,
+		    "Show data-type profile result"),
 	OPT_PARENT(mem_options)
 	};
 	const char *const mem_subcommands[] = { "record", "report", NULL };
-- 
2.46.0.rc1.232.g9752f9e123-goog


  parent reply	other threads:[~2024-07-31 23:55 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-31 23:54 [PATCHSET 0/6] perf mem: Basic support for data type profiling (v1) Namhyung Kim
2024-07-31 23:55 ` [PATCH 1/6] perf hist: Correct hist_entry->mem_info refcounts Namhyung Kim
2024-07-31 23:55 ` [PATCH 2/6] perf mem: Free the allocated sort string Namhyung Kim
2024-08-01 15:20   ` Arnaldo Carvalho de Melo
2024-07-31 23:55 ` [PATCH 3/6] perf mem: Rework command option handling Namhyung Kim
2024-07-31 23:55 ` [PATCH 4/6] perf tools: Add mode argument to sort_help() Namhyung Kim
2024-07-31 23:55 ` [PATCH 5/6] perf mem: Add -s/--sort option Namhyung Kim
2024-07-31 23:55 ` Namhyung Kim [this message]
2024-08-01 15:23 ` [PATCHSET 0/6] perf mem: Basic support for data type profiling (v1) Ian Rogers
2024-08-02  0:24   ` Namhyung Kim

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240731235505.710436-7-namhyung@kernel.org \
    --to=namhyung@kernel.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=atrajeev@linux.vnet.ibm.com \
    --cc=eranian@google.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).