All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: peterz@infradead.org, acme@kernel.org
Cc: linux-kernel@vger.kernel.org, jolsa@kernel.org,
	Andi Kleen <ak@linux.intel.com>
Subject: [PATCH 4/6] perf/x86: Add support for PEBS sampling persistent RAM on Skylake
Date: Mon,  5 Jun 2017 15:48:36 -0700	[thread overview]
Message-ID: <20170605224838.11759-4-andi@firstfloor.org> (raw)
In-Reply-To: <20170605224838.11759-1-andi@firstfloor.org>

From: Andi Kleen <ak@linux.intel.com>

On Skylake server the L4 encoding in the PEBS data source actually
means persistent memory. Add a new perf encoding for this case,
and report it on Skylake Server.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 arch/x86/events/intel/core.c    | 3 ++-
 arch/x86/events/intel/ds.c      | 8 +++++---
 arch/x86/events/perf_event.h    | 2 +-
 include/uapi/linux/perf_event.h | 3 ++-
 4 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index b54d9022d016..f341126766cb 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -3989,7 +3989,8 @@ __init int intel_pmu_init(void)
 						  skl_format_attr);
 		WARN_ON(!x86_pmu.format_attrs);
 		x86_pmu.cpu_events = hsw_events_attrs;
-		intel_pmu_pebs_data_source_skl();
+		intel_pmu_pebs_data_source_skl(
+			boot_cpu_data.x86_model == INTEL_FAM6_SKYLAKE_X);
 		pr_cont("Skylake events, ");
 		break;
 
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index c7f2c71d74bd..881784ed0f0b 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -79,10 +79,12 @@ void __init intel_pmu_pebs_data_source_nhm(void)
 	pebs_data_source[0x07] = OP_LH | P(LVL, L3)  | P(SNOOP, HITM);
 }
 
-void __init intel_pmu_pebs_data_source_skl(void)
+void __init intel_pmu_pebs_data_source_skl(bool pmem)
 {
-	pebs_data_source[0x08] = OP_LH | P(LVLX, L4) | P(SNOOP, HIT);
-	pebs_data_source[0x09] = OP_LH | P(LVLX, L4) | P(LVLX, REMOTE) | P(SNOOP, HIT);
+	u64 pmem_or_l4 = pmem ? P(LVLX, PMEM) : P(LVLX, L4);
+
+	pebs_data_source[0x08] = OP_LH | pmem_or_l4 | P(SNOOP, HIT);
+	pebs_data_source[0x09] = OP_LH | pmem_or_l4 | P(LVLX, REMOTE) | P(SNOOP, HIT);
 	pebs_data_source[0x0b] = OP_LH | P(LVLX, RAM) | P(LVLX, REMOTE) | P(SNOOP, NONE);
 	pebs_data_source[0x0c] = OP_LH | P(LVL, NA) | P(LVLX, REMOTE) | P(SNOOPX, FWD);
 	pebs_data_source[0x0d] = OP_LH | P(LVL, NA) | P(LVLX, REMOTE) | P(SNOOP, HITM);
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index 5298debaeb32..9b05a5a04e0b 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -943,7 +943,7 @@ void intel_pmu_lbr_init_knl(void);
 
 void intel_pmu_pebs_data_source_nhm(void);
 
-void intel_pmu_pebs_data_source_skl(void);
+void intel_pmu_pebs_data_source_skl(bool pmem);
 
 int intel_pmu_setup_lbr_filter(struct perf_event *event);
 
diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index 4b5deeada34b..8d2c7c4ec881 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -982,7 +982,8 @@ union perf_mem_data_src {
 #define PERF_MEM_LVLX_REMOTE    0x01 /* Remote */
 #define PERF_MEM_LVLX_L4	0x02 /* L4 */
 #define PERF_MEM_LVLX_RAM	0x04 /* Ram */
-/* 5 free */
+#define PERF_MEM_LVLX_PMEM	0x08 /* Persistent Memory */
+/* 4 free */
 
 #define PERF_MEM_LVLX_SHIFT	33
 
-- 
2.9.4

  parent reply	other threads:[~2017-06-05 22:49 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-05 22:48 [PATCH 1/6] perf/x86: Move Nehalem PEBS code to flag Andi Kleen
2017-06-05 22:48 ` [PATCH 2/6] perf/x86: Fix data source decoding for Skylake Andi Kleen
2017-06-06 10:08   ` Peter Zijlstra
2017-06-06 13:51     ` Andi Kleen
2017-06-06 16:21       ` Peter Zijlstra
2017-06-06 17:12         ` Andi Kleen
2017-06-05 22:48 ` [PATCH 3/6] perf, tools: Add support for printing new mem_info encodings Andi Kleen
2017-06-05 22:48 ` Andi Kleen [this message]
2017-06-05 22:48 ` [PATCH 5/6] perf, tools: Support persistent memory encoding Andi Kleen
2017-06-05 22:48 ` [PATCH 6/6] perf, tools: Add test cases for new data source encoding Andi Kleen
  -- strict thread matches above, loose matches on Subject: below --
2017-06-02 20:12 [PATCH 1/6] perf/x86: Move Nehalem PEBS code to flag Andi Kleen
2017-06-02 20:12 ` [PATCH 4/6] perf/x86: Add support for PEBS sampling persistent RAM on Skylake Andi Kleen

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=20170605224838.11759-4-andi@firstfloor.org \
    --to=andi@firstfloor.org \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.