Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753477AbdFGXW7 (ORCPT ); Wed, 7 Jun 2017 19:22:59 -0400 Received: from mga02.intel.com ([134.134.136.20]:37707 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752695AbdFGXW5 (ORCPT ); Wed, 7 Jun 2017 19:22:57 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.39,312,1493708400"; d="scan'208";a="978097120" From: Andi Kleen To: peterz@infradead.org, acme@kernel.org Cc: linux-kernel@vger.kernel.org, jolsa@kernel.org, eranian@google.com, Andi Kleen Subject: [PATCH v2 2/4] perf/x86: Fix data source decoding for Skylake Date: Wed, 7 Jun 2017 16:22:24 -0700 Message-Id: <20170607232226.26365-3-andi@firstfloor.org> X-Mailer: git-send-email 2.9.4 In-Reply-To: <20170607232226.26365-1-andi@firstfloor.org> References: <20170607232226.26365-1-andi@firstfloor.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4651 Lines: 132 From: Andi Kleen Skylake changed the encoding of the PEBS data source field. Some combinations are not available anymore, but some new cases e.g. for L4 cache hit are added. Fix up the conversion table for Skylake, similar as had been done for Nehalem. On Skylake server the encoding for L4 actually means persistent memory. Handle this case too. To properly describe it in the abstracted perf format I had to add some new bits. Unfortunately the existing fields were full, so this required adding eXtension fields for mem_lvl and snoop into existing reserved space. v2: Merge with persistent memory patch. Add explicit bit for each case instead of using generic modifier. Signed-off-by: Andi Kleen --- arch/x86/events/intel/core.c | 2 ++ arch/x86/events/intel/ds.c | 13 +++++++++++++ arch/x86/events/perf_event.h | 2 ++ include/uapi/linux/perf_event.h | 22 ++++++++++++++++++++-- 4 files changed, 37 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index dec9b4bf0752..08e53f36d697 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -4052,6 +4052,8 @@ __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( + 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 7732999f5e2a..cd28c4babd36 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -79,6 +79,19 @@ 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(bool pmem) +{ + u64 pmem_or_l4; + + pmem_or_l4 = pmem ? P(LVLX, PMEM) : P(LVLX, L4); + pebs_data_source[0x08] = OP_LH | pmem_or_l4 | P(SNOOP, HIT); + pmem_or_l4 = pmem ? P(LVLX, REM_PMEM) : P(LVLX, REM_L4); + pebs_data_source[0x09] = OP_LH | pmem_or_l4 | P(SNOOP, HIT); + pebs_data_source[0x0b] = OP_LH | P(LVLX, REM_RAM) | P(SNOOP, NONE); + pebs_data_source[0x0c] = OP_LH | P(LVLX, REM_NA) | P(SNOOPX, FWD); + pebs_data_source[0x0d] = OP_LH | P(LVLX, REM_NA) | P(SNOOP, HITM); +} + static u64 precise_store_data(u64 status) { union intel_x86_pebs_dse dse; diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index a6d9d6570957..d7571f248652 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -946,6 +946,8 @@ void intel_pmu_lbr_init_knl(void); void intel_pmu_pebs_data_source_nhm(void); +void intel_pmu_pebs_data_source_skl(bool pmem); + int intel_pmu_setup_lbr_filter(struct perf_event *event); void intel_pt_interrupt(void); diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index b1c0b187acfe..95daade294d7 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -931,14 +931,18 @@ union perf_mem_data_src { mem_snoop:5, /* snoop mode */ mem_lock:2, /* lock instr */ mem_dtlb:7, /* tlb access */ - mem_rsvd:31; + mem_lvlx:8, /* memory hierarchy level, ext */ + mem_snoopx:2, /* snoop mode, ext */ + mem_rsvd:21; }; }; #elif defined(__BIG_ENDIAN_BITFIELD) union perf_mem_data_src { __u64 val; struct { - __u64 mem_rsvd:31, + __u64 mem_rsvd:21, + mem_snoopx:2, /* snoop mode, ext */ + mem_lvlx:8, /* memory hierarchy level, ext */ mem_dtlb:7, /* tlb access */ mem_lock:2, /* lock instr */ mem_snoop:5, /* snoop mode */ @@ -975,6 +979,16 @@ union perf_mem_data_src { #define PERF_MEM_LVL_UNC 0x2000 /* Uncached memory */ #define PERF_MEM_LVL_SHIFT 5 +#define PERF_MEM_LVLX_L4 0x01 /* L4 */ +#define PERF_MEM_LVLX_REM_L4 0x02 /* Remote L4 */ +#define PERF_MEM_LVLX_REM_RAM 0x04 /* Remote Ram, unknown hops */ +#define PERF_MEM_LVLX_PMEM 0x08 /* Persistent Memory */ +#define PERF_MEM_LVLX_REM_PMEM 0x10 /* Remote Persistent Memory */ +#define PERF_MEM_LVLX_REM_NA 0x20 /* Remote N/A level */ +/* 2 free */ + +#define PERF_MEM_LVLX_SHIFT 33 + /* snoop mode */ #define PERF_MEM_SNOOP_NA 0x01 /* not available */ #define PERF_MEM_SNOOP_NONE 0x02 /* no snoop */ @@ -983,6 +997,10 @@ union perf_mem_data_src { #define PERF_MEM_SNOOP_HITM 0x10 /* snoop hit modified */ #define PERF_MEM_SNOOP_SHIFT 19 +#define PERF_MEM_SNOOPX_FWD 0x01 /* forward */ +/* 1 free */ +#define PERF_MEM_SNOOPX_SHIFT 41 + /* locked instruction */ #define PERF_MEM_LOCK_NA 0x01 /* not available */ #define PERF_MEM_LOCK_LOCKED 0x02 /* locked transaction */ -- 2.9.4