From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5847E30EF94; Mon, 12 Jan 2026 05:20:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195218; cv=none; b=NoBYWubzsVB2ujZQtRrsT5gdETW6+FehtZ+PaDhNrAkm+6VulRaqlHdi6B2fTvUWXB8aBcc+fa3m5Ejd00x54746TOUN9kZudOLBm8fGcVXeW8hrXZkDHwsoK9e5FaEhfRDLWZl+1rTjy2DJtWtZEh2wAg96Nz75nJNcDWr7PTo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195218; c=relaxed/simple; bh=mD4cUKeT80Gmm4SvuP3lmdsk/KPKwm5ADKsv7AVp93k=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=B0akzEltkM2k5Dy6CuZlhVwvr3Il2S8J4nsTJaR0Kjo/FWMYOKZ8pRVeRDS2/2U+Ljb+bo79UJ1xcsavogSCe/5CbCQnxMLZVDmVa6x39rz4vSDEdkjD3VAgqJu8jxh1YnW8KjRnmHs7AzkdNmrhb7i2mKnvqXvyhER5us9k6dk= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=hAGekp/y; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="hAGekp/y" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195216; x=1799731216; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=mD4cUKeT80Gmm4SvuP3lmdsk/KPKwm5ADKsv7AVp93k=; b=hAGekp/yubQohzcvSnNTfAMYhOitxORdO0G3eEHmrq0xnWItxTbnVDnZ t/X5XGstBdNQqtPOizX/Yutbf1jV8NXHCymj+9LLEgYm8gPJKCZ9kFuph VyPo7gvSFZ/4OHbDfkhsssky1RENuNtQYsUKZgWumeBvc2Yo/Rr3633Ao wp14O1KWJWGt69JCC/DAbvlaGHudZu6yBp+ksOiW7b/077wWeiVmpr5CZ L3a6/05u6t0zmWk4sR2B/Bl90EUESaPfOh41pvQBExGjE+1q8IU1BeoAd HiG6HgDltb8JzymnXPW6vjZAUe5nPu8m05caCf3YxfE0fzJLnAyScopeS w==; X-CSE-ConnectionGUID: VZ+UrKHSRfmp+tTLbT7gUQ== X-CSE-MsgGUID: /yfBdoaQR1KUWzpVxmnh9A== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673213" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673213" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:16 -0800 X-CSE-ConnectionGUID: MAbXLbs0RUO/C1pd8Pq+Bg== X-CSE-MsgGUID: iHbo50JDTIqOmxUa9f1KEA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233378" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:13 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 1/7] perf/x86/intel: Support the 4 new OMR MSRs introduced in DMR and NVL Date: Mon, 12 Jan 2026 13:16:43 +0800 Message-Id: <20260112051649.1113435-2-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Diamond Rapids (DMR) and Nova Lake (NVL) introduce an enhanced Off-Module Response (OMR) facility, replacing the Off-Core Response (OCR) Performance Monitoring of previous processors. Legacy microarchitectures used the OCR facility to evaluate off-core and multi-core off-module transactions. The newly named OMR facility improves OCR capabilities for scalable coverage of new memory systems in multi-core module systems. Similar to OCR, 4 additional off-module configuration MSRs (OFFMODULE_RSP_0 to OFFMODULE_RSP_3) are introduced to specify attributes of off-module transactions. When multiple identical OMR events are created, they need to occupy the same OFFMODULE_RSP_x MSR. To ensure these multiple identical OMR events can work simultaneously, the intel_alt_er() and intel_fixup_er() helpers are enhanced to rotate these OMR events across different OFFMODULE_RSP_* MSRs, similar to previous OCR events. For more details about OMR, please refer to section 16.1 "OFF-MODULE RESPONSE (OMR) FACILITY" in ISE documentation. ISE link: https://www.intel.com/content/www/us/en/content-details/869288/in= tel-architecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- v2: Optimize intel_fixup_er(). arch/x86/events/intel/core.c | 55 +++++++++++++++++++++----------- arch/x86/events/perf_event.h | 5 +++ arch/x86/include/asm/msr-index.h | 5 +++ 3 files changed, 47 insertions(+), 18 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 1840ca1918d1..6ea3260f6422 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -3532,17 +3532,28 @@ static int intel_alt_er(struct cpu_hw_events *cpuc, struct extra_reg *extra_regs =3D hybrid(cpuc->pmu, extra_regs); int alt_idx =3D idx; =20 - if (!(x86_pmu.flags & PMU_FL_HAS_RSP_1)) - return idx; - - if (idx =3D=3D EXTRA_REG_RSP_0) - alt_idx =3D EXTRA_REG_RSP_1; - - if (idx =3D=3D EXTRA_REG_RSP_1) - alt_idx =3D EXTRA_REG_RSP_0; + if (idx =3D=3D EXTRA_REG_RSP_0 || idx =3D=3D EXTRA_REG_RSP_1) { + if (!(x86_pmu.flags & PMU_FL_HAS_RSP_1)) + return idx; + if (++alt_idx > EXTRA_REG_RSP_1) + alt_idx =3D EXTRA_REG_RSP_0; + if (config & ~extra_regs[alt_idx].valid_mask) + return idx; + } =20 - if (config & ~extra_regs[alt_idx].valid_mask) - return idx; + if (idx >=3D EXTRA_REG_OMR_0 && idx <=3D EXTRA_REG_OMR_3) { + if (!(x86_pmu.flags & PMU_FL_HAS_OMR)) + return idx; + if (++alt_idx > EXTRA_REG_OMR_3) + alt_idx =3D EXTRA_REG_OMR_0; + /* + * Subtracting EXTRA_REG_OMR_0 ensures to get correct + * OMR extra_reg entries which start from 0. + */ + if (config & + ~extra_regs[alt_idx - EXTRA_REG_OMR_0].valid_mask) + return idx; + } =20 return alt_idx; } @@ -3550,16 +3561,24 @@ static int intel_alt_er(struct cpu_hw_events *cpuc, static void intel_fixup_er(struct perf_event *event, int idx) { struct extra_reg *extra_regs =3D hybrid(event->pmu, extra_regs); - event->hw.extra_reg.idx =3D idx; + int er_idx; =20 - if (idx =3D=3D EXTRA_REG_RSP_0) { - event->hw.config &=3D ~INTEL_ARCH_EVENT_MASK; - event->hw.config |=3D extra_regs[EXTRA_REG_RSP_0].event; - event->hw.extra_reg.reg =3D MSR_OFFCORE_RSP_0; - } else if (idx =3D=3D EXTRA_REG_RSP_1) { + event->hw.extra_reg.idx =3D idx; + switch (idx) { + case EXTRA_REG_RSP_0 ... EXTRA_REG_RSP_1: + er_idx =3D idx - EXTRA_REG_RSP_0; event->hw.config &=3D ~INTEL_ARCH_EVENT_MASK; - event->hw.config |=3D extra_regs[EXTRA_REG_RSP_1].event; - event->hw.extra_reg.reg =3D MSR_OFFCORE_RSP_1; + event->hw.config |=3D extra_regs[er_idx].event; + event->hw.extra_reg.reg =3D MSR_OFFCORE_RSP_0 + er_idx; + break; + case EXTRA_REG_OMR_0 ... EXTRA_REG_OMR_3: + er_idx =3D idx - EXTRA_REG_OMR_0; + event->hw.config &=3D ~ARCH_PERFMON_EVENTSEL_UMASK; + event->hw.config |=3D 1ULL << (8 + er_idx); + event->hw.extra_reg.reg =3D MSR_OMR_0 + er_idx; + break; + default: + pr_warn("The extra reg idx %d is not supported.\n", idx); } } =20 diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index 3161ec0a3416..586e3fdfe6d8 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -45,6 +45,10 @@ enum extra_reg_type { EXTRA_REG_FE =3D 4, /* fe_* */ EXTRA_REG_SNOOP_0 =3D 5, /* snoop response 0 */ EXTRA_REG_SNOOP_1 =3D 6, /* snoop response 1 */ + EXTRA_REG_OMR_0 =3D 7, /* OMR 0 */ + EXTRA_REG_OMR_1 =3D 8, /* OMR 1 */ + EXTRA_REG_OMR_2 =3D 9, /* OMR 2 */ + EXTRA_REG_OMR_3 =3D 10, /* OMR 3 */ =20 EXTRA_REG_MAX /* number of entries needed */ }; @@ -1099,6 +1103,7 @@ do { \ #define PMU_FL_RETIRE_LATENCY 0x200 /* Support Retire Latency in PEBS */ #define PMU_FL_BR_CNTR 0x400 /* Support branch counter logging */ #define PMU_FL_DYN_CONSTRAINT 0x800 /* Needs dynamic constraint */ +#define PMU_FL_HAS_OMR 0x1000 /* has 4 equivalent OMR regs */ =20 #define EVENT_VAR(_id) event_attr_##_id #define EVENT_PTR(_id) &event_attr_##_id.attr.attr diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-in= dex.h index 3d0a0950d20a..6d1b69ea01c2 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -263,6 +263,11 @@ #define MSR_SNOOP_RSP_0 0x00001328 #define MSR_SNOOP_RSP_1 0x00001329 =20 +#define MSR_OMR_0 0x000003e0 +#define MSR_OMR_1 0x000003e1 +#define MSR_OMR_2 0x000003e2 +#define MSR_OMR_3 0x000003e3 + #define MSR_LBR_SELECT 0x000001c8 #define MSR_LBR_TOS 0x000001c9 =20 --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4D48030E84F; Mon, 12 Jan 2026 05:20:20 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195225; cv=none; b=LYdKAmMvNpI0Ok5eEL7HWbdjI2Gp443mIOotyKiG9OtbrDiJzofcXpVLqXkTtVYE53+7YnIriACmntyFE1lkJ1FRXfcnjE0DzDmnoN3yzWIGEHEA3/jMClLH+mrQeRiMF0dqABvzOFCOi+HgFH6fpx42Wmzfb9EwlV81JUZ6CI4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195225; c=relaxed/simple; bh=v8IElGX8i0ySe3uIAhT6UmH/LZEkqPLuMWeaQIuHFnM=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=te9w9jGJopkIM3oHublCAbOXA4D3RfGkYcQPTTMUnBp+FhnZmc/v4FcBGPZz/T2DXrZ6AVgz0mCWuH+6je09iNM/nytkujP1+rm+yfWeh8UvaL07AhUvsWIYIOtBCO1+rACEhZ+r+AnaKlO+qZEruIjzjJcGS5qYfwQQid7a+vQ= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=P1nMVbzw; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="P1nMVbzw" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195220; x=1799731220; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=v8IElGX8i0ySe3uIAhT6UmH/LZEkqPLuMWeaQIuHFnM=; b=P1nMVbzwHzQum8nJpP2CL/3xzNzIk0r2YVO2khMna2n60N2waUn7cNPq kRMpz/8TES+DVKCs6zhK7rw4dr/+3fMEpbn3TVNfEFlHVnRsv5vhk3w6e 3bL/dDC/gUFD1q/NSXumi1h7eTJTLhVd8rfsmv1j0wPmx1wqQlt02AkKB LRFTw5h2u3VjXEw1EJmxRPmFF1lrouT2ysJdyDU4LgoArd/sen3b+QiJ+ XhDOHoj3fmcgYIHgfNPyxAM0TnFRm1cGlfyILUZec55/tSbCe9wq0OLwc GRU0fLkEK6BepxS0erzJZSplwlDOGumwG/ZDGERXk/4NO/xMBb7THd4dX Q==; X-CSE-ConnectionGUID: 5m2Yy3q+S4WjxuRNFFwGwA== X-CSE-MsgGUID: 8qay7tw2TkOZ8twm3JMJVQ== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673223" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673223" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:20 -0800 X-CSE-ConnectionGUID: 8rJ3zbe/RTuAsNIzTTDqjQ== X-CSE-MsgGUID: 8M1ThxrDQW2nAJjPWD0Xow== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233387" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:16 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 2/7] perf/x86/intel: Add support for PEBS memory auxiliary info field in DMR Date: Mon, 12 Jan 2026 13:16:44 +0800 Message-Id: <20260112051649.1113435-3-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" With the introduction of the OMR feature, the PEBS memory auxiliary info field for load and store latency events has been restructured for DMR. The memory auxiliary info field's bit[8] indicates whether a L2 cache miss occurred for a memory load or store instruction. If bit[8] is 0, it signifies no L2 cache miss, and bits[7:0] specify the exact cache data source (up to the L2 cache level). If bit[8] is 1, bits[7:0] represent the OMR encoding, indicating the specific L3 cache or memory region involved in the memory access. A significant enhancement is OMR encoding provides up to 8 fine-grained memory regions besides the cache region. A significant enhancement for OMR encoding is the ability to provide up to 8 fine-grained memory regions in addition to the cache region, offering more detailed insights into memory access regions. For detailed information on the memory auxiliary info encoding, please refer to section 16.2 "PEBS LOAD LATENCY AND STORE LATENCY FACILITY" in the ISE documentation. This patch ensures that the PEBS memory auxiliary info field is correctly interpreted and utilized in DMR. ISE: https://www.intel.com/content/www/us/en/content-details/869288/intel-a= rchitecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- arch/x86/events/intel/ds.c | 140 ++++++++++++++++++++++++++ arch/x86/events/perf_event.h | 2 + include/uapi/linux/perf_event.h | 27 ++++- tools/include/uapi/linux/perf_event.h | 27 ++++- 4 files changed, 190 insertions(+), 6 deletions(-) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index feb1c3cf63e4..272e652f25fc 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -34,6 +34,17 @@ struct pebs_record_32 { =20 */ =20 +union omr_encoding { + struct { + u8 omr_source : 4; + u8 omr_remote : 1; + u8 omr_hitm : 1; + u8 omr_snoop : 1; + u8 omr_promoted : 1; + }; + u8 omr_full; +}; + union intel_x86_pebs_dse { u64 val; struct { @@ -73,6 +84,18 @@ union intel_x86_pebs_dse { unsigned int lnc_addr_blk:1; unsigned int ld_reserved6:18; }; + struct { + unsigned int pnc_dse: 8; + unsigned int pnc_l2_miss:1; + unsigned int pnc_stlb_clean_hit:1; + unsigned int pnc_stlb_any_hit:1; + unsigned int pnc_stlb_miss:1; + unsigned int pnc_locked:1; + unsigned int pnc_data_blk:1; + unsigned int pnc_addr_blk:1; + unsigned int pnc_fb_full:1; + unsigned int ld_reserved8:16; + }; }; =20 =20 @@ -228,6 +251,85 @@ void __init intel_pmu_pebs_data_source_lnl(void) __intel_pmu_pebs_data_source_cmt(data_source); } =20 +/* Version for Panthercove and later */ + +/* L2 hit */ +#define PNC_PEBS_DATA_SOURCE_MAX 16 +static u64 pnc_pebs_l2_hit_data_source[PNC_PEBS_DATA_SOURCE_MAX] =3D { + P(OP, LOAD) | P(LVL, NA) | LEVEL(NA) | P(SNOOP, NA), /* 0x00: non-cache a= ccess */ + OP_LH | LEVEL(L0) | P(SNOOP, NONE), /* 0x01: L0 hit */ + OP_LH | P(LVL, L1) | LEVEL(L1) | P(SNOOP, NONE), /* 0x02: L1 hit */ + OP_LH | P(LVL, LFB) | LEVEL(LFB) | P(SNOOP, NONE), /* 0x03: L1 Miss Handl= ing Buffer hit */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, NONE), /* 0x04: L2 Hit Clean */ + 0, /* 0x05: Reserved */ + 0, /* 0x06: Reserved */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, HIT), /* 0x07: L2 Hit Snoop HI= T */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, HITM), /* 0x08: L2 Hit Snoop H= it Modified */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, MISS), /* 0x09: Prefetch Promo= tion */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, MISS), /* 0x0a: Cross Core Pre= fetch Promotion */ + 0, /* 0x0b: Reserved */ + 0, /* 0x0c: Reserved */ + 0, /* 0x0d: Reserved */ + 0, /* 0x0e: Reserved */ + OP_LH | P(LVL, UNC) | LEVEL(NA) | P(SNOOP, NONE), /* 0x0f: uncached */ +}; + +/* L2 miss */ +#define OMR_DATA_SOURCE_MAX 16 +static u64 omr_data_source[OMR_DATA_SOURCE_MAX] =3D { + P(OP, LOAD) | P(LVL, NA) | LEVEL(NA) | P(SNOOP, NA), /* 0x00: invalid */ + 0, /* 0x01: Reserved */ + OP_LH | P(LVL, L3) | LEVEL(L3) | P(REGION, L_SHARE), /* 0x02: local CA sh= ared cache */ + OP_LH | P(LVL, L3) | LEVEL(L3) | P(REGION, L_NON_SHARE),/* 0x03: local CA= non-shared cache */ + OP_LH | P(LVL, L3) | LEVEL(L3) | P(REGION, O_IO), /* 0x04: other CA IO ag= ent */ + OP_LH | P(LVL, L3) | LEVEL(L3) | P(REGION, O_SHARE), /* 0x05: other CA sh= ared cache */ + OP_LH | P(LVL, L3) | LEVEL(L3) | P(REGION, O_NON_SHARE),/* 0x06: other CA= non-shared cache */ + OP_LH | LEVEL(RAM) | P(REGION, MMIO), /* 0x07: MMIO */ + OP_LH | LEVEL(RAM) | P(REGION, MEM0), /* 0x08: Memory region 0 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM1), /* 0x09: Memory region 1 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM2), /* 0x0a: Memory region 2 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM3), /* 0x0b: Memory region 3 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM4), /* 0x0c: Memory region 4 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM5), /* 0x0d: Memory region 5 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM6), /* 0x0e: Memory region 6 */ + OP_LH | LEVEL(RAM) | P(REGION, MEM7), /* 0x0f: Memory region 7 */ +}; + +static u64 parse_omr_data_source(u8 dse) +{ + union omr_encoding omr; + u64 val =3D 0; + + omr.omr_full =3D dse; + val =3D omr_data_source[omr.omr_source]; + if (omr.omr_source > 0x1 && omr.omr_source < 0x7) + val |=3D omr.omr_remote ? P(LVL, REM_CCE1) : 0; + else if (omr.omr_source > 0x7) + val |=3D omr.omr_remote ? P(LVL, REM_RAM1) : P(LVL, LOC_RAM); + + if (omr.omr_remote) + val |=3D REM; + + val |=3D omr.omr_hitm ? P(SNOOP, HITM) : P(SNOOP, HIT); + + if (omr.omr_source =3D=3D 0x2) { + u8 snoop =3D omr.omr_snoop | omr.omr_promoted; + + if (snoop =3D=3D 0x0) + val |=3D P(SNOOP, NA); + else if (snoop =3D=3D 0x1) + val |=3D P(SNOOP, MISS); + else if (snoop =3D=3D 0x2) + val |=3D P(SNOOP, HIT); + else if (snoop =3D=3D 0x3) + val |=3D P(SNOOP, NONE); + } else if (omr.omr_source > 0x2 && omr.omr_source < 0x7) { + val |=3D omr.omr_snoop ? P(SNOOPX, FWD) : 0; + } + + return val; +} + static u64 precise_store_data(u64 status) { union intel_x86_pebs_dse dse; @@ -411,6 +513,44 @@ u64 arl_h_latency_data(struct perf_event *event, u64 s= tatus) return lnl_latency_data(event, status); } =20 +u64 pnc_latency_data(struct perf_event *event, u64 status) +{ + union intel_x86_pebs_dse dse; + union perf_mem_data_src src; + u64 val; + + dse.val =3D status; + + if (!dse.pnc_l2_miss) + val =3D pnc_pebs_l2_hit_data_source[dse.pnc_dse & 0xf]; + else + val =3D parse_omr_data_source(dse.pnc_dse); + + if (!val) + val =3D P(OP, LOAD) | LEVEL(NA) | P(SNOOP, NA); + + if (dse.pnc_stlb_miss) + val |=3D P(TLB, MISS) | P(TLB, L2); + else + val |=3D P(TLB, HIT) | P(TLB, L1) | P(TLB, L2); + + if (dse.pnc_locked) + val |=3D P(LOCK, LOCKED); + + if (dse.pnc_data_blk) + val |=3D P(BLK, DATA); + if (dse.pnc_addr_blk) + val |=3D P(BLK, ADDR); + if (!dse.pnc_data_blk && !dse.pnc_addr_blk) + val |=3D P(BLK, NA); + + src.val =3D val; + if (event->hw.flags & PERF_X86_EVENT_PEBS_ST_HSW) + src.mem_op =3D P(OP, STORE); + + return src.val; +} + static u64 load_latency_data(struct perf_event *event, 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 586e3fdfe6d8..bd501c2a0f73 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1664,6 +1664,8 @@ u64 lnl_latency_data(struct perf_event *event, u64 st= atus); =20 u64 arl_h_latency_data(struct perf_event *event, u64 status); =20 +u64 pnc_latency_data(struct perf_event *event, u64 status); + extern struct event_constraint intel_core2_pebs_event_constraints[]; =20 extern struct event_constraint intel_atom_pebs_event_constraints[]; diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_even= t.h index c44a8fb3e418..533393ec94d0 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -1330,14 +1330,16 @@ union perf_mem_data_src { mem_snoopx : 2, /* Snoop mode, ext */ mem_blk : 3, /* Access blocked */ mem_hops : 3, /* Hop level */ - mem_rsvd : 18; + mem_region : 5, /* cache/memory regions */ + mem_rsvd : 13; }; }; #elif defined(__BIG_ENDIAN_BITFIELD) union perf_mem_data_src { __u64 val; struct { - __u64 mem_rsvd : 18, + __u64 mem_rsvd : 13, + mem_region : 5, /* cache/memory regions */ mem_hops : 3, /* Hop level */ mem_blk : 3, /* Access blocked */ mem_snoopx : 2, /* Snoop mode, ext */ @@ -1394,7 +1396,7 @@ union perf_mem_data_src { #define PERF_MEM_LVLNUM_L4 0x0004 /* L4 */ #define PERF_MEM_LVLNUM_L2_MHB 0x0005 /* L2 Miss Handling Buffer */ #define PERF_MEM_LVLNUM_MSC 0x0006 /* Memory-side Cache */ -/* 0x007 available */ +#define PERF_MEM_LVLNUM_L0 0x0007 /* L0 */ #define PERF_MEM_LVLNUM_UNC 0x0008 /* Uncached */ #define PERF_MEM_LVLNUM_CXL 0x0009 /* CXL */ #define PERF_MEM_LVLNUM_IO 0x000a /* I/O */ @@ -1447,6 +1449,25 @@ union perf_mem_data_src { /* 5-7 available */ #define PERF_MEM_HOPS_SHIFT 43 =20 +/* Cache/Memory region */ +#define PERF_MEM_REGION_NA 0x0 /* Invalid */ +#define PERF_MEM_REGION_RSVD 0x01 /* Reserved */ +#define PERF_MEM_REGION_L_SHARE 0x02 /* Local CA shared cache */ +#define PERF_MEM_REGION_L_NON_SHARE 0x03 /* Local CA non-shared cache */ +#define PERF_MEM_REGION_O_IO 0x04 /* Other CA IO agent */ +#define PERF_MEM_REGION_O_SHARE 0x05 /* Other CA shared cache */ +#define PERF_MEM_REGION_O_NON_SHARE 0x06 /* Other CA non-shared cache */ +#define PERF_MEM_REGION_MMIO 0x07 /* MMIO */ +#define PERF_MEM_REGION_MEM0 0x08 /* Memory region 0 */ +#define PERF_MEM_REGION_MEM1 0x09 /* Memory region 1 */ +#define PERF_MEM_REGION_MEM2 0x0a /* Memory region 2 */ +#define PERF_MEM_REGION_MEM3 0x0b /* Memory region 3 */ +#define PERF_MEM_REGION_MEM4 0x0c /* Memory region 4 */ +#define PERF_MEM_REGION_MEM5 0x0d /* Memory region 5 */ +#define PERF_MEM_REGION_MEM6 0x0e /* Memory region 6 */ +#define PERF_MEM_REGION_MEM7 0x0f /* Memory region 7 */ +#define PERF_MEM_REGION_SHIFT 46 + #define PERF_MEM_S(a, s) \ (((__u64)PERF_MEM_##a##_##s) << PERF_MEM_##a##_SHIFT) =20 diff --git a/tools/include/uapi/linux/perf_event.h b/tools/include/uapi/lin= ux/perf_event.h index c44a8fb3e418..d4b99610a3b0 100644 --- a/tools/include/uapi/linux/perf_event.h +++ b/tools/include/uapi/linux/perf_event.h @@ -1330,14 +1330,16 @@ union perf_mem_data_src { mem_snoopx : 2, /* Snoop mode, ext */ mem_blk : 3, /* Access blocked */ mem_hops : 3, /* Hop level */ - mem_rsvd : 18; + mem_region : 5, /* cache/memory regions */ + mem_rsvd : 13; }; }; #elif defined(__BIG_ENDIAN_BITFIELD) union perf_mem_data_src { __u64 val; struct { - __u64 mem_rsvd : 18, + __u64 mem_rsvd : 13, + mem_region : 5, /* cache/memory regions */ mem_hops : 3, /* Hop level */ mem_blk : 3, /* Access blocked */ mem_snoopx : 2, /* Snoop mode, ext */ @@ -1394,7 +1396,7 @@ union perf_mem_data_src { #define PERF_MEM_LVLNUM_L4 0x0004 /* L4 */ #define PERF_MEM_LVLNUM_L2_MHB 0x0005 /* L2 Miss Handling Buffer */ #define PERF_MEM_LVLNUM_MSC 0x0006 /* Memory-side Cache */ -/* 0x007 available */ +#define PERF_MEM_LVLNUM_L0 0x0007 /* L0 */ #define PERF_MEM_LVLNUM_UNC 0x0008 /* Uncached */ #define PERF_MEM_LVLNUM_CXL 0x0009 /* CXL */ #define PERF_MEM_LVLNUM_IO 0x000a /* I/O */ @@ -1447,6 +1449,25 @@ union perf_mem_data_src { /* 5-7 available */ #define PERF_MEM_HOPS_SHIFT 43 =20 +/* Cache/Memory region */ +#define PERF_MEM_REGION_NA 0x0 /* Invalid */ +#define PERF_MEM_REGION_RSVD 0x01 /* Reserved */ +#define PERF_MEM_REGION_L_SHARE 0x02 /* Local CA shared cache */ +#define PERF_MEM_REGION_L_NON_SHARE 0x03 /* Local CA non-shared cache */ +#define PERF_MEM_REGION_O_IO 0x04 /* Other CA IO agent */ +#define PERF_MEM_REGION_O_SHARE 0x05 /* Other CA shared cache */ +#define PERF_MEM_REGION_O_NON_SHARE 0x06 /* Other CA non-shared cache */ +#define PERF_MEM_REGION_MMIO 0x07 /* MMIO */ +#define PERF_MEM_REGION_MEM0 0x08 /* Memory region 0 */ +#define PERF_MEM_REGION_MEM1 0x09 /* Memory region 1 */ +#define PERF_MEM_REGION_MEM2 0x0a /* Memory region 2 */ +#define PERF_MEM_REGION_MEM3 0x0b /* Memory region 3 */ +#define PERF_MEM_REGION_MEM4 0x0c /* Memory region 4 */ +#define PERF_MEM_REGION_MEM5 0x0d /* Memory region 5 */ +#define PERF_MEM_REGION_MEM6 0x0e /* Memory region 6 */ +#define PERF_MEM_REGION_MEM7 0x0f /* Memory region 7 */ +#define PERF_MEM_REGION_SHIFT 46 + #define PERF_MEM_S(a, s) \ (((__u64)PERF_MEM_##a##_##s) << PERF_MEM_##a##_SHIFT) =20 --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7BB4824679F; Mon, 12 Jan 2026 05:20:24 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195226; cv=none; b=V2Evb/zJ38JWypcLAEw2EnZKUT/BCqeTnl1GYjEQ9X+rVCOJMUSUHN6rSiuDtKGZ3p+7/QWUutAbXNCGbPNASkI6oJrEOPY5ipmHbYSEviog4PGqE2rTyn38Wha0xI2D9W/dWDNpQiVPpFi8jFIcAAzmqkAsIJ1yQkfOqBBCz6U= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195226; c=relaxed/simple; bh=WiuHANaCzPcuk+5JCe/Nf/yD8EHO5u9Wjn3tHS/XVe0=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=R49gS1nWQHEr+ApoRQKXUT6G2nRZ1CdgqNW8dNg8CK/q7Ogb73uDi3ZnUqWR5prfoeM9aM+dGW7kWzdri8FuGUh3wq5vNZ1pfAGorBKg+gJ3AAuJDW+9A+nDVLICnn51aCTkbQkpT9OxMAic05ebifl3oIiEminc5uGZX6jexCQ= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=ezRtEINa; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="ezRtEINa" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195225; x=1799731225; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=WiuHANaCzPcuk+5JCe/Nf/yD8EHO5u9Wjn3tHS/XVe0=; b=ezRtEINaRqxik5ho9UN28BtIkNt/AzLpsPHWlT7jaiJZj+MRCaB0ncah GSN8Jc2+JYlqUORLZ9DnOENo5QEexKarkx7KlJRnx5fI5H9rgO/+A9kAs sA9S2fZn7/WcPmsPuEBalBotTlK5TILGggAFVSqA1KT0WsXIT1MAlybfb bNWUe4SZ4ESGsVolJ42Quwj61avdIW3tpCbzsIIai9wBcarihmConZHRo cT9G2oomurQohMyC9za7/liw5kkYr++SXo/hZI90KiHVePshsGdOCrQwG HEmFGIHFInKE4piStOsn7nJOgmKf9xFIaeHecZvaykE4mSkILlqWPaGgr g==; X-CSE-ConnectionGUID: 1Z5L5m5CSWKy3NtRsDP8kA== X-CSE-MsgGUID: Fwp5V52JSqCDHyDEW+ssKg== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673228" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673228" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:24 -0800 X-CSE-ConnectionGUID: TlX05a0MQJmvBgSq0iHQKg== X-CSE-MsgGUID: CwYWVYDEQOOKvzDdbtCmvA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233415" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:20 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 3/7] perf/x86/intel: Add core PMU support for DMR Date: Mon, 12 Jan 2026 13:16:45 +0800 Message-Id: <20260112051649.1113435-4-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" This patch enables core PMU features for Diamond Rapids (Panther Cove microarchitecture), including Panther Cove specific counter and PEBS constraints, a new cache events ID table, and the model-specific OMR events extra registers table. For detailed information about counter constraints, please refer to section 16.3 "COUNTER RESTRICTIONS" in the ISE documentation. ISE: https://www.intel.com/content/www/us/en/content-details/869288/intel-a= rchitecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- arch/x86/events/intel/core.c | 182 +++++++++++++++++++++++++++++++++++ arch/x86/events/intel/ds.c | 27 ++++++ arch/x86/events/perf_event.h | 2 + 3 files changed, 211 insertions(+) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 6ea3260f6422..82fac6e8731b 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -435,6 +435,62 @@ static struct extra_reg intel_lnc_extra_regs[] __read_= mostly =3D { EVENT_EXTRA_END }; =20 +static struct event_constraint intel_pnc_event_constraints[] =3D { + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x0100, 0), /* INST_RETIRED.PREC_DIST */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ + FIXED_EVENT_CONSTRAINT(0x013c, 2), /* CPU_CLK_UNHALTED.REF_TSC_P */ + FIXED_EVENT_CONSTRAINT(0x0400, 3), /* SLOTS */ + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_RETIRING, 0), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_BAD_SPEC, 1), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_FE_BOUND, 2), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_BE_BOUND, 3), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_HEAVY_OPS, 4), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_BR_MISPREDICT, 5), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_FETCH_LAT, 6), + METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_MEM_BOUND, 7), + + INTEL_EVENT_CONSTRAINT(0x20, 0xf), + INTEL_EVENT_CONSTRAINT(0x79, 0xf), + + INTEL_UEVENT_CONSTRAINT(0x0275, 0xf), + INTEL_UEVENT_CONSTRAINT(0x0176, 0xf), + INTEL_UEVENT_CONSTRAINT(0x04a4, 0x1), + INTEL_UEVENT_CONSTRAINT(0x08a4, 0x1), + INTEL_UEVENT_CONSTRAINT(0x01cd, 0xfc), + INTEL_UEVENT_CONSTRAINT(0x02cd, 0x3), + + INTEL_EVENT_CONSTRAINT(0xd0, 0xf), + INTEL_EVENT_CONSTRAINT(0xd1, 0xf), + INTEL_EVENT_CONSTRAINT(0xd4, 0xf), + INTEL_EVENT_CONSTRAINT(0xd6, 0xf), + INTEL_EVENT_CONSTRAINT(0xdf, 0xf), + INTEL_EVENT_CONSTRAINT(0xce, 0x1), + + INTEL_UEVENT_CONSTRAINT(0x01b1, 0x8), + INTEL_UEVENT_CONSTRAINT(0x0847, 0xf), + INTEL_UEVENT_CONSTRAINT(0x0446, 0xf), + INTEL_UEVENT_CONSTRAINT(0x0846, 0xf), + INTEL_UEVENT_CONSTRAINT(0x0148, 0xf), + + EVENT_CONSTRAINT_END +}; + +static struct extra_reg intel_pnc_extra_regs[] __read_mostly =3D { + /* must define OMR_X first, see intel_alt_er() */ + INTEL_UEVENT_EXTRA_REG(0x012a, MSR_OMR_0, 0x40ffffff0000ffffull, OMR_0), + INTEL_UEVENT_EXTRA_REG(0x022a, MSR_OMR_1, 0x40ffffff0000ffffull, OMR_1), + INTEL_UEVENT_EXTRA_REG(0x042a, MSR_OMR_2, 0x40ffffff0000ffffull, OMR_2), + INTEL_UEVENT_EXTRA_REG(0x082a, MSR_OMR_3, 0x40ffffff0000ffffull, OMR_3), + INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), + INTEL_UEVENT_EXTRA_REG(0x02c6, MSR_PEBS_FRONTEND, 0x9, FE), + INTEL_UEVENT_EXTRA_REG(0x03c6, MSR_PEBS_FRONTEND, 0x7fff1f, FE), + INTEL_UEVENT_EXTRA_REG(0x40ad, MSR_PEBS_FRONTEND, 0xf, FE), + INTEL_UEVENT_EXTRA_REG(0x04c2, MSR_PEBS_FRONTEND, 0x8, FE), + EVENT_EXTRA_END +}; + EVENT_ATTR_STR(mem-loads, mem_ld_nhm, "event=3D0x0b,umask=3D0x10,ldlat=3D3= "); EVENT_ATTR_STR(mem-loads, mem_ld_snb, "event=3D0xcd,umask=3D0x1,ldlat=3D3"= ); EVENT_ATTR_STR(mem-stores, mem_st_snb, "event=3D0xcd,umask=3D0x2"); @@ -650,6 +706,102 @@ static __initconst const u64 glc_hw_cache_extra_regs }, }; =20 +static __initconst const u64 pnc_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] =3D +{ + [ C(L1D ) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x81d0, + [ C(RESULT_MISS) ] =3D 0xe124, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x82d0, + }, + }, + [ C(L1I ) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_MISS) ] =3D 0xe424, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + }, + [ C(LL ) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x12a, + [ C(RESULT_MISS) ] =3D 0x12a, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x12a, + [ C(RESULT_MISS) ] =3D 0x12a, + }, + }, + [ C(DTLB) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x81d0, + [ C(RESULT_MISS) ] =3D 0xe12, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x82d0, + [ C(RESULT_MISS) ] =3D 0xe13, + }, + }, + [ C(ITLB) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D 0xe11, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + [ C(OP_PREFETCH) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + }, + [ C(BPU ) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x4c4, + [ C(RESULT_MISS) ] =3D 0x4c5, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + [ C(OP_PREFETCH) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + }, + [ C(NODE) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D -1, + [ C(RESULT_MISS) ] =3D -1, + }, + }, +}; + +static __initconst const u64 pnc_hw_cache_extra_regs + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] =3D +{ + [ C(LL ) ] =3D { + [ C(OP_READ) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x4000000000000001, + [ C(RESULT_MISS) ] =3D 0xFFFFF000000001, + }, + [ C(OP_WRITE) ] =3D { + [ C(RESULT_ACCESS) ] =3D 0x4000000000000002, + [ C(RESULT_MISS) ] =3D 0xFFFFF000000002, + }, + }, +}; + /* * Notes on the events: * - data reads do not include code reads (comparable to earlier tables) @@ -7230,6 +7382,20 @@ static __always_inline void intel_pmu_init_lnc(struc= t pmu *pmu) hybrid(pmu, extra_regs) =3D intel_lnc_extra_regs; } =20 +static __always_inline void intel_pmu_init_pnc(struct pmu *pmu) +{ + intel_pmu_init_glc(pmu); + x86_pmu.flags &=3D ~PMU_FL_HAS_RSP_1; + x86_pmu.flags |=3D PMU_FL_HAS_OMR; + memcpy(hybrid_var(pmu, hw_cache_event_ids), + pnc_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + memcpy(hybrid_var(pmu, hw_cache_extra_regs), + pnc_hw_cache_extra_regs, sizeof(hw_cache_extra_regs)); + hybrid(pmu, event_constraints) =3D intel_pnc_event_constraints; + hybrid(pmu, pebs_constraints) =3D intel_pnc_pebs_event_constraints; + hybrid(pmu, extra_regs) =3D intel_pnc_extra_regs; +} + static __always_inline void intel_pmu_init_skt(struct pmu *pmu) { intel_pmu_init_grt(pmu); @@ -7906,6 +8072,22 @@ __init int intel_pmu_init(void) intel_pmu_pebs_data_source_skl(true); break; =20 + case INTEL_DIAMONDRAPIDS_X: + intel_pmu_init_pnc(NULL); + x86_pmu.pebs_ept =3D 1; + x86_pmu.hw_config =3D hsw_hw_config; + x86_pmu.pebs_latency_data =3D pnc_latency_data; + x86_pmu.get_event_constraints =3D glc_get_event_constraints; + extra_attr =3D boot_cpu_has(X86_FEATURE_RTM) ? + hsw_format_attr : nhm_format_attr; + extra_skl_attr =3D skl_format_attr; + mem_attr =3D glc_events_attrs; + td_attr =3D glc_td_events_attrs; + tsx_attr =3D glc_tsx_events_attrs; + pr_cont("Panthercove events, "); + name =3D "panthercove"; + break; + case INTEL_ALDERLAKE: case INTEL_ALDERLAKE_L: case INTEL_RAPTORLAKE: diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 272e652f25fc..06e42ac33749 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -1425,6 +1425,33 @@ struct event_constraint intel_lnc_pebs_event_constra= ints[] =3D { EVENT_CONSTRAINT_END }; =20 +struct event_constraint intel_pnc_pebs_event_constraints[] =3D { + INTEL_FLAGS_UEVENT_CONSTRAINT(0x100, 0x100000000ULL), /* INST_RETIRED.PRE= C_DIST */ + INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL), + + INTEL_HYBRID_LDLAT_CONSTRAINT(0x1cd, 0xfc), + INTEL_HYBRID_STLAT_CONSTRAINT(0x2cd, 0x3), + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf), /* MEM_INST_RETIRED= .STLB_MISS_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf), /* MEM_INST_RETIRED= .STLB_MISS_STORES */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x21d0, 0xf), /* MEM_INST_RETIRED= .LOCK_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x41d0, 0xf), /* MEM_INST_RETIRED= .SPLIT_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x42d0, 0xf), /* MEM_INST_RETIRED= .SPLIT_STORES */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x81d0, 0xf), /* MEM_INST_RETIRED= .ALL_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x82d0, 0xf), /* MEM_INST_RETIRED= .ALL_STORES */ + + INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD_RANGE(0xd1, 0xd4, 0xf), + + INTEL_FLAGS_EVENT_CONSTRAINT(0xd0, 0xf), + INTEL_FLAGS_EVENT_CONSTRAINT(0xd6, 0xf), + + /* + * Everything else is handled by PMU_FL_PEBS_ALL, because we + * need the full constraints from the main table. + */ + + EVENT_CONSTRAINT_END +}; + struct event_constraint *intel_pebs_constraints(struct perf_event *event) { struct event_constraint *pebs_constraints =3D hybrid(event->pmu, pebs_con= straints); diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index bd501c2a0f73..cbca1888e8f7 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1698,6 +1698,8 @@ extern struct event_constraint intel_glc_pebs_event_c= onstraints[]; =20 extern struct event_constraint intel_lnc_pebs_event_constraints[]; =20 +extern struct event_constraint intel_pnc_pebs_event_constraints[]; + struct event_constraint *intel_pebs_constraints(struct perf_event *event); =20 void intel_pmu_pebs_add(struct perf_event *event); --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 29C5C30F529; Mon, 12 Jan 2026 05:20:27 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195232; cv=none; b=CEV15zz/9njIX6MPsJBMBDd/XL0KpNJg72g7mOCZNQMUfrlHjf0DQMxWslK5gb90T2jxBsKLEaY2TzAPBiOoyDqR2TZYrMumGyLQE8CLpTHcLKdilk2HxhxAYSyzGqX0A/sbYf0fByd/XPVWaA3GyGS4kt2dDTxn+I/j2PFfbFo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195232; c=relaxed/simple; bh=Jje9Nl6kZG8JO4M8Sr6/NKQaD3pizjWUtstA+P5+e2w=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=AfDrC+Jp1uaEavjGzHAveo5QDx9C9SB3f5+I7mtjNDa9YiTLzZbzwEx5F40M5RMliNJBqHjbhvkzVNVGzv7JqUvO+2DfkqSTkFLqPOlIh28KHUjCc2UkwlnQjtRv7UBNr2jnNrW3i9268ibYO0I9qx9GXoEj9C3YW0XHwKuBngs= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=eu4XaVVv; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="eu4XaVVv" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195230; x=1799731230; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Jje9Nl6kZG8JO4M8Sr6/NKQaD3pizjWUtstA+P5+e2w=; b=eu4XaVVvl6l5k8yx5l8cosmD0m4aa7Le8SMgtfGfeRGaRwQ8fwZOgTo8 1BGm2oFxvdfhAlhCdanY2qX3j1sxSkkxF7DOogLUyXF90ZUG5KF2ldKCZ 09u/EQ+XhS86KHz4BqPAAnCQxc43GP04QWAWTcMW7fYlmKGp1jwWqpIcv DNjUht2uOc3eIMUkr3/j8/R4fULTw6BrbSp1tE/BWPVvraVSl/I0bK+UP 4i53vl8SlRmCXybp16Yvzxc18Nx18SBlG/kRX7pNCOlgnEWRVwXX5f2vu rNCsse/cxinSL1v2raqMogj017r8r3kWkCEdKsUg0SphP1Dc1majztYr4 Q==; X-CSE-ConnectionGUID: kUIurmjkQky4fnubgy4EwQ== X-CSE-MsgGUID: c9L109nOSDGzu1t8FpYDgw== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673233" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673233" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:28 -0800 X-CSE-ConnectionGUID: 5KawQ/jKQxiw0z8IiriLrQ== X-CSE-MsgGUID: JenKE254SMeoanJn0aHl5w== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233434" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:24 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 4/7] perf/x86/intel: Add support for PEBS memory auxiliary info field in NVL Date: Mon, 12 Jan 2026 13:16:46 +0800 Message-Id: <20260112051649.1113435-5-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Similar to DMR (Panther Cove uarch), both P-core (Coyote Cove uarch) and E-core (Arctic Wolf uarch) of NVL adopt the new PEBS memory auxiliary info layout. Coyote Cove microarchitecture shares the same PMU capabilities, including the memory auxiliary info layout, with Panther Cove. Arctic Wolf microarchitecture has a similar layout to Panther Cove, with the only difference being specific data source encoding for L2 hit cases (up to the L2 cache level). The OMR encoding remains the same as in Panther Cove. For detailed information on the memory auxiliary info encoding, please refer to section 16.2 "PEBS LOAD LATENCY AND STORE LATENCY FACILITY" in the latest ISE documentation. This patch defines Arctic Wolf specific data source encoding and then supports PEBS memory auxiliary info field for NVL. ISE: https://www.intel.com/content/www/us/en/content-details/869288/intel-a= rchitecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- arch/x86/events/intel/ds.c | 83 ++++++++++++++++++++++++++++++++++++ arch/x86/events/perf_event.h | 2 + 2 files changed, 85 insertions(+) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 06e42ac33749..a47f173d411b 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -96,6 +96,18 @@ union intel_x86_pebs_dse { unsigned int pnc_fb_full:1; unsigned int ld_reserved8:16; }; + struct { + unsigned int arw_dse:8; + unsigned int arw_l2_miss:1; + unsigned int arw_xq_promotion:1; + unsigned int arw_reissue:1; + unsigned int arw_stlb_miss:1; + unsigned int arw_locked:1; + unsigned int arw_data_blk:1; + unsigned int arw_addr_blk:1; + unsigned int arw_fb_full:1; + unsigned int ld_reserved9:16; + }; }; =20 =20 @@ -274,6 +286,29 @@ static u64 pnc_pebs_l2_hit_data_source[PNC_PEBS_DATA_S= OURCE_MAX] =3D { OP_LH | P(LVL, UNC) | LEVEL(NA) | P(SNOOP, NONE), /* 0x0f: uncached */ }; =20 +/* Version for Arctic Wolf and later */ + +/* L2 hit */ +#define ARW_PEBS_DATA_SOURCE_MAX 16 +static u64 arw_pebs_l2_hit_data_source[ARW_PEBS_DATA_SOURCE_MAX] =3D { + P(OP, LOAD) | P(LVL, NA) | LEVEL(NA) | P(SNOOP, NA), /* 0x00: non-cache a= ccess */ + OP_LH | P(LVL, L1) | LEVEL(L1) | P(SNOOP, NONE), /* 0x01: L1 hit */ + OP_LH | P(LVL, LFB) | LEVEL(LFB) | P(SNOOP, NONE), /* 0x02: WCB Hit */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, NONE), /* 0x03: L2 Hit Clean */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, HIT), /* 0x04: L2 Hit Snoop HI= T */ + OP_LH | P(LVL, L2) | LEVEL(L2) | P(SNOOP, HITM), /* 0x05: L2 Hit Snoop H= it Modified */ + OP_LH | P(LVL, UNC) | LEVEL(NA) | P(SNOOP, NONE), /* 0x06: uncached */ + 0, /* 0x07: Reserved */ + 0, /* 0x08: Reserved */ + 0, /* 0x09: Reserved */ + 0, /* 0x0a: Reserved */ + 0, /* 0x0b: Reserved */ + 0, /* 0x0c: Reserved */ + 0, /* 0x0d: Reserved */ + 0, /* 0x0e: Reserved */ + 0, /* 0x0f: Reserved */ +}; + /* L2 miss */ #define OMR_DATA_SOURCE_MAX 16 static u64 omr_data_source[OMR_DATA_SOURCE_MAX] =3D { @@ -458,6 +493,44 @@ u64 cmt_latency_data(struct perf_event *event, u64 sta= tus) dse.mtl_fwd_blk); } =20 +static u64 arw_latency_data(struct perf_event *event, u64 status) +{ + union intel_x86_pebs_dse dse; + union perf_mem_data_src src; + u64 val; + + dse.val =3D status; + + if (!dse.arw_l2_miss) + val =3D arw_pebs_l2_hit_data_source[dse.arw_dse & 0xf]; + else + val =3D parse_omr_data_source(dse.arw_dse); + + if (!val) + val =3D P(OP, LOAD) | LEVEL(NA) | P(SNOOP, NA); + + if (dse.arw_stlb_miss) + val |=3D P(TLB, MISS) | P(TLB, L2); + else + val |=3D P(TLB, HIT) | P(TLB, L1) | P(TLB, L2); + + if (dse.arw_locked) + val |=3D P(LOCK, LOCKED); + + if (dse.arw_data_blk) + val |=3D P(BLK, DATA); + if (dse.arw_addr_blk) + val |=3D P(BLK, ADDR); + if (!dse.arw_data_blk && !dse.arw_addr_blk) + val |=3D P(BLK, NA); + + src.val =3D val; + if (event->hw.flags & PERF_X86_EVENT_PEBS_ST_HSW) + src.mem_op =3D P(OP, STORE); + + return src.val; +} + static u64 lnc_latency_data(struct perf_event *event, u64 status) { union intel_x86_pebs_dse dse; @@ -551,6 +624,16 @@ u64 pnc_latency_data(struct perf_event *event, u64 sta= tus) return src.val; } =20 +u64 nvl_latency_data(struct perf_event *event, u64 status) +{ + struct x86_hybrid_pmu *pmu =3D hybrid_pmu(event->pmu); + + if (pmu->pmu_type =3D=3D hybrid_small) + return arw_latency_data(event, status); + + return pnc_latency_data(event, status); +} + static u64 load_latency_data(struct perf_event *event, 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 cbca1888e8f7..aedc1a7762c2 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1666,6 +1666,8 @@ u64 arl_h_latency_data(struct perf_event *event, u64 = status); =20 u64 pnc_latency_data(struct perf_event *event, u64 status); =20 +u64 nvl_latency_data(struct perf_event *event, u64 status); + extern struct event_constraint intel_core2_pebs_event_constraints[]; =20 extern struct event_constraint intel_atom_pebs_event_constraints[]; --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C23B9303C86; Mon, 12 Jan 2026 05:20:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195233; cv=none; b=GIqchMQlnHnjNbw6QdRC8CrKyRWft8WcuBb0DPQsA6hqmzdH744Leuz4ipfNZssW8wYyFgAfcS7YW+IGpXAsJgvT4QkgO+bLwr5wRPkmYblPn0aHAbN8Yl4yxzoe2F7vzZ0rjUIQJb45F0K5JIoCFpGf4BnoCHqk6bDQIBr54MI= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195233; c=relaxed/simple; bh=QuyBypBmonai9UqLRcqgxBUmxjzvXl/fe//XsNDOHZI=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=NLzNjQGCmprKtMikfZyxXXrUVcnSTXcQC0Iotq/B7T8sNKvdfrhLeNRK7tDyjUlkmI1pbJdhj4s1CLtbOuvHAt/r189+OlwqtjTvZpLp4Qjv+dxIJa2WzsEfH3HcO0zq/k4XF1MATpBIIih7NoWoNlC+GRGhqsGp1T7cd8mkK10= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=XpLWck70; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="XpLWck70" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195232; x=1799731232; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=QuyBypBmonai9UqLRcqgxBUmxjzvXl/fe//XsNDOHZI=; b=XpLWck70eP0DWf2j9rQCYwul7/IlogWXXxDZikL2ixM6BhbysMIXoL9A gO421cAubyad6JgWB8U01K3os5nHULyAhB5tWUzwGMjm4TilShAFgUo2Q CUjIbLGMSbZ21ihwQibrwSrhG3qAnq1Q50MUx1jxUlGKmbszFzKPF//JJ tU3aACODDG2SK5Zq5lMRG2oFic7CYEv1zlKQEmx6X7FeKJrLemhnfnZHK YByjRkxLF02adADCnOJkgoVB31/nQSD7QWL2v0uXeDvoADFjBFQnuVlAH r56qNa6fJVWtQdtc2mqq6AdGVxPk5kZdRWNfF9aoD0P4M4IWZQy8MUJnE Q==; X-CSE-ConnectionGUID: OqzslOTlTXGLNWxukEChlQ== X-CSE-MsgGUID: aErnCmttTeeWWCKKumWy0Q== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673238" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673238" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:31 -0800 X-CSE-ConnectionGUID: KeC/q+k0Sc2ZnPw7JxT3rA== X-CSE-MsgGUID: 0QZ0sdcITqOo/fOY4kmsvA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233452" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:28 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 5/7] perf/x86/intel: Add core PMU support for Novalake Date: Mon, 12 Jan 2026 13:16:47 +0800 Message-Id: <20260112051649.1113435-6-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" This patch enables core PMU support for Novalake, covering both P-core and E-core. It includes Arctic Wolf-specific counters and PEBS constraints, and the model-specific OMR extra registers table. Since Coyote Cove shares the same PMU capabilities as Panther Cove, the existing Panther Cove PMU enabling functions are reused for Coyote Cove. For detailed information about counter constraints, please refer to section 16.3 "COUNTER RESTRICTIONS" in the ISE documentation. ISE: https://www.intel.com/content/www/us/en/content-details/869288/intel-a= rchitecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- arch/x86/events/intel/core.c | 99 ++++++++++++++++++++++++++++++++++++ arch/x86/events/intel/ds.c | 11 ++++ arch/x86/events/perf_event.h | 2 + 3 files changed, 112 insertions(+) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 82fac6e8731b..dd488a095f33 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -232,6 +232,29 @@ static struct event_constraint intel_skt_event_constra= ints[] __read_mostly =3D { EVENT_CONSTRAINT_END }; =20 +static struct event_constraint intel_arw_event_constraints[] __read_mostly= =3D { + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + FIXED_EVENT_CONSTRAINT(0x0300, 2), /* pseudo CPU_CLK_UNHALTED.REF */ + FIXED_EVENT_CONSTRAINT(0x013c, 2), /* CPU_CLK_UNHALTED.REF_TSC_P */ + FIXED_EVENT_CONSTRAINT(0x0073, 4), /* TOPDOWN_BAD_SPECULATION.ALL */ + FIXED_EVENT_CONSTRAINT(0x019c, 5), /* TOPDOWN_FE_BOUND.ALL */ + FIXED_EVENT_CONSTRAINT(0x02c2, 6), /* TOPDOWN_RETIRING.ALL */ + INTEL_UEVENT_CONSTRAINT(0x01b7, 0x1), + INTEL_UEVENT_CONSTRAINT(0x02b7, 0x2), + INTEL_UEVENT_CONSTRAINT(0x04b7, 0x4), + INTEL_UEVENT_CONSTRAINT(0x08b7, 0x8), + INTEL_UEVENT_CONSTRAINT(0x01d4, 0x1), + INTEL_UEVENT_CONSTRAINT(0x02d4, 0x2), + INTEL_UEVENT_CONSTRAINT(0x04d4, 0x4), + INTEL_UEVENT_CONSTRAINT(0x08d4, 0x8), + INTEL_UEVENT_CONSTRAINT(0x0175, 0x1), + INTEL_UEVENT_CONSTRAINT(0x0275, 0x2), + INTEL_UEVENT_CONSTRAINT(0x21d3, 0x1), + INTEL_UEVENT_CONSTRAINT(0x22d3, 0x1), + EVENT_CONSTRAINT_END +}; + static struct event_constraint intel_skl_event_constraints[] =3D { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -2319,6 +2342,26 @@ static __initconst const u64 tnt_hw_cache_extra_regs }, }; =20 +static __initconst const u64 arw_hw_cache_extra_regs + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] =3D { + [C(LL)] =3D { + [C(OP_READ)] =3D { + [C(RESULT_ACCESS)] =3D 0x4000000000000001, + [C(RESULT_MISS)] =3D 0xFFFFF000000001, + }, + [C(OP_WRITE)] =3D { + [C(RESULT_ACCESS)] =3D 0x4000000000000002, + [C(RESULT_MISS)] =3D 0xFFFFF000000002, + }, + [C(OP_PREFETCH)] =3D { + [C(RESULT_ACCESS)] =3D 0x0, + [C(RESULT_MISS)] =3D 0x0, + }, + }, +}; + EVENT_ATTR_STR(topdown-fe-bound, td_fe_bound_tnt, "event=3D0x= 71,umask=3D0x0"); EVENT_ATTR_STR(topdown-retiring, td_retiring_tnt, "event=3D0x= c2,umask=3D0x0"); EVENT_ATTR_STR(topdown-bad-spec, td_bad_spec_tnt, "event=3D0x= 73,umask=3D0x6"); @@ -2377,6 +2420,22 @@ static struct extra_reg intel_cmt_extra_regs[] __rea= d_mostly =3D { EVENT_EXTRA_END }; =20 +static struct extra_reg intel_arw_extra_regs[] __read_mostly =3D { + /* must define OMR_X first, see intel_alt_er() */ + INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OMR_0, 0xc0ffffffffffffffull, OMR_0), + INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OMR_1, 0xc0ffffffffffffffull, OMR_1), + INTEL_UEVENT_EXTRA_REG(0x04b7, MSR_OMR_2, 0xc0ffffffffffffffull, OMR_2), + INTEL_UEVENT_EXTRA_REG(0x08b7, MSR_OMR_3, 0xc0ffffffffffffffull, OMR_3), + INTEL_UEVENT_EXTRA_REG(0x01d4, MSR_OMR_0, 0xc0ffffffffffffffull, OMR_0), + INTEL_UEVENT_EXTRA_REG(0x02d4, MSR_OMR_1, 0xc0ffffffffffffffull, OMR_1), + INTEL_UEVENT_EXTRA_REG(0x04d4, MSR_OMR_2, 0xc0ffffffffffffffull, OMR_2), + INTEL_UEVENT_EXTRA_REG(0x08d4, MSR_OMR_3, 0xc0ffffffffffffffull, OMR_3), + INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x5d0), + INTEL_UEVENT_EXTRA_REG(0x0127, MSR_SNOOP_RSP_0, 0xffffffffffffffffull, SN= OOP_0), + INTEL_UEVENT_EXTRA_REG(0x0227, MSR_SNOOP_RSP_1, 0xffffffffffffffffull, SN= OOP_1), + EVENT_EXTRA_END +}; + EVENT_ATTR_STR(topdown-fe-bound, td_fe_bound_skt, "event=3D0x= 9c,umask=3D0x01"); EVENT_ATTR_STR(topdown-retiring, td_retiring_skt, "event=3D0x= c2,umask=3D0x02"); EVENT_ATTR_STR(topdown-be-bound, td_be_bound_skt, "event=3D0x= a4,umask=3D0x02"); @@ -7404,6 +7463,19 @@ static __always_inline void intel_pmu_init_skt(struc= t pmu *pmu) static_call_update(intel_pmu_enable_acr_event, intel_pmu_enable_acr); } =20 +static __always_inline void intel_pmu_init_arw(struct pmu *pmu) +{ + intel_pmu_init_grt(pmu); + x86_pmu.flags &=3D ~PMU_FL_HAS_RSP_1; + x86_pmu.flags |=3D PMU_FL_HAS_OMR; + memcpy(hybrid_var(pmu, hw_cache_extra_regs), + arw_hw_cache_extra_regs, sizeof(hw_cache_extra_regs)); + hybrid(pmu, event_constraints) =3D intel_arw_event_constraints; + hybrid(pmu, pebs_constraints) =3D intel_arw_pebs_event_constraints; + hybrid(pmu, extra_regs) =3D intel_arw_extra_regs; + static_call_update(intel_pmu_enable_acr_event, intel_pmu_enable_acr); +} + __init int intel_pmu_init(void) { struct attribute **extra_skl_attr =3D &empty_attrs; @@ -8249,6 +8321,33 @@ __init int intel_pmu_init(void) name =3D "arrowlake_h_hybrid"; break; =20 + case INTEL_NOVALAKE: + case INTEL_NOVALAKE_L: + pr_cont("Novalake Hybrid events, "); + name =3D "novalake_hybrid"; + intel_pmu_init_hybrid(hybrid_big_small); + + x86_pmu.pebs_latency_data =3D nvl_latency_data; + x86_pmu.get_event_constraints =3D mtl_get_event_constraints; + x86_pmu.hw_config =3D adl_hw_config; + + td_attr =3D lnl_hybrid_events_attrs; + mem_attr =3D mtl_hybrid_mem_attrs; + tsx_attr =3D adl_hybrid_tsx_attrs; + extra_attr =3D boot_cpu_has(X86_FEATURE_RTM) ? + mtl_hybrid_extra_attr_rtm : mtl_hybrid_extra_attr; + + /* Initialize big core specific PerfMon capabilities.*/ + pmu =3D &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_CORE_IDX]; + intel_pmu_init_pnc(&pmu->pmu); + + /* Initialize Atom core specific PerfMon capabilities.*/ + pmu =3D &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_ATOM_IDX]; + intel_pmu_init_arw(&pmu->pmu); + + intel_pmu_pebs_data_source_lnl(); + break; + default: switch (x86_pmu.version) { case 1: diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index a47f173d411b..5027afc97b65 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -1293,6 +1293,17 @@ struct event_constraint intel_grt_pebs_event_constra= ints[] =3D { EVENT_CONSTRAINT_END }; =20 +struct event_constraint intel_arw_pebs_event_constraints[] =3D { + /* Allow all events as PEBS with no flags */ + INTEL_HYBRID_LAT_CONSTRAINT(0x5d0, 0xff), + INTEL_HYBRID_LAT_CONSTRAINT(0x6d0, 0xff), + INTEL_FLAGS_UEVENT_CONSTRAINT(0x01d4, 0x1), + INTEL_FLAGS_UEVENT_CONSTRAINT(0x02d4, 0x2), + INTEL_FLAGS_UEVENT_CONSTRAINT(0x04d4, 0x4), + INTEL_FLAGS_UEVENT_CONSTRAINT(0x08d4, 0x8), + EVENT_CONSTRAINT_END +}; + struct event_constraint intel_nehalem_pebs_event_constraints[] =3D { INTEL_PLD_CONSTRAINT(0x100b, 0xf), /* MEM_INST_RETIRED.* */ INTEL_FLAGS_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */ diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index aedc1a7762c2..f7caabc5d487 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1680,6 +1680,8 @@ extern struct event_constraint intel_glp_pebs_event_c= onstraints[]; =20 extern struct event_constraint intel_grt_pebs_event_constraints[]; =20 +extern struct event_constraint intel_arw_pebs_event_constraints[]; + extern struct event_constraint intel_nehalem_pebs_event_constraints[]; =20 extern struct event_constraint intel_westmere_pebs_event_constraints[]; --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8C7F624679F; Mon, 12 Jan 2026 05:20:37 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195238; cv=none; b=QMqq8ybunBWDHibyLH/dZU1iarnl5vShJFq5ET0oc0KStEKfI1TXq0gLFOb8w6OtOYi0jySBeT9x/c8u+/mze/wNwiZVUkCLhPoQ3MN8t/48u6WW7JPP2iw27OtnFoFFBfxjOomVxRUXI/3yN1/XvUOXSmfaeKLgnI0TWjN4H7g= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195238; c=relaxed/simple; bh=8swUns+f4sPnKRPagD2A6Oy6QNSylTblYdaomHLa5sw=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=T1LPIQDzvSFPgxZPsVYaBhRGd30jCzyghBWcEGu61kk7DrWCZwFrqicnQsT+WT8dr+Ih8yL+979yQkmuEeUgcYFRq6Z7AkZC4scuws3L/d7SkxYZTYRnFm9l2cBk6xwfOS4dKxKJJJp1oC+pNjNECmcsZqqhYEvsr8Yyu0zCq6g= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=c/HH2L2Y; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="c/HH2L2Y" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195238; x=1799731238; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=8swUns+f4sPnKRPagD2A6Oy6QNSylTblYdaomHLa5sw=; b=c/HH2L2Y2H/pWse4BszJ6MPmQXwhCiW9RWFqM52EqEwFO3OSU5VDDrvH +m0GUH6H8TTPnmPx320tGtzJa84cB4+O/Tlp1EDdmvMoSz3+Gh/Gnubuz LGPGqszKkGasE0n1uCu8zs04cWQYeWHQZa4b7w/k5VLTcH/wIS45vCL3i DeJBlNnth9CVXiFjgz1cTs+PmfpQ+n4lR7GS4V6yrsNIUTmYZGeJSBVUB +V9swiR1PtLA4RiWaprLvOn+ozPusQmvZvlOBl8tYO3N38oimKnZ3CMDm ULROi0mEAZ0K7ZNHGAP9XwB1v0IVAczM5WagjZX2OLMMi64PAR09Z6YUb A==; X-CSE-ConnectionGUID: DZMkD0aQTwi0BIe6RfTD7w== X-CSE-MsgGUID: jxRWZgYXSmKHB12aODWrmw== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673246" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673246" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:35 -0800 X-CSE-ConnectionGUID: J4SI0Cr8SuKZvPNoMB6PNw== X-CSE-MsgGUID: tV7Wyz8KSNOrm8PFC5Ra4A== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233470" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:31 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 6/7] perf/x86: Use macros to replace magic numbers in attr_rdpmc Date: Mon, 12 Jan 2026 13:16:48 +0800 Message-Id: <20260112051649.1113435-7-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Replace magic numbers in attr_rdpmc with macros to improve readability and make their meanings clearer for users. Signed-off-by: Dapeng Mi --- arch/x86/events/core.c | 7 ++++--- arch/x86/events/intel/p6.c | 2 +- arch/x86/events/perf_event.h | 7 +++++++ 3 files changed, 12 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 0ecac9495d74..c2717cb5034f 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -2163,7 +2163,8 @@ static int __init init_hw_perf_events(void) =20 pr_cont("%s PMU driver.\n", x86_pmu.name); =20 - x86_pmu.attr_rdpmc =3D 1; /* enable userspace RDPMC usage by default */ + /* enable userspace RDPMC usage by default */ + x86_pmu.attr_rdpmc =3D X86_USER_RDPMC_CONDITIONAL_ENABLE; =20 for (quirk =3D x86_pmu.quirks; quirk; quirk =3D quirk->next) quirk->func(); @@ -2643,12 +2644,12 @@ static ssize_t set_attr_rdpmc(struct device *cdev, */ if (val =3D=3D 0) static_branch_inc(&rdpmc_never_available_key); - else if (x86_pmu.attr_rdpmc =3D=3D 0) + else if (x86_pmu.attr_rdpmc =3D=3D X86_USER_RDPMC_NEVER_ENABLE) static_branch_dec(&rdpmc_never_available_key); =20 if (val =3D=3D 2) static_branch_inc(&rdpmc_always_available_key); - else if (x86_pmu.attr_rdpmc =3D=3D 2) + else if (x86_pmu.attr_rdpmc =3D=3D X86_USER_RDPMC_ALWAYS_ENABLE) static_branch_dec(&rdpmc_always_available_key); =20 on_each_cpu(cr4_update_pce, NULL, 1); diff --git a/arch/x86/events/intel/p6.c b/arch/x86/events/intel/p6.c index 6e41de355bd8..fb991e0ac614 100644 --- a/arch/x86/events/intel/p6.c +++ b/arch/x86/events/intel/p6.c @@ -243,7 +243,7 @@ static __init void p6_pmu_rdpmc_quirk(void) */ pr_warn("Userspace RDPMC support disabled due to a CPU erratum\n"); x86_pmu.attr_rdpmc_broken =3D 1; - x86_pmu.attr_rdpmc =3D 0; + x86_pmu.attr_rdpmc =3D X86_USER_RDPMC_NEVER_ENABLE; } } =20 diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index f7caabc5d487..24a81d2916e9 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -187,6 +187,13 @@ struct amd_nb { (1ULL << PERF_REG_X86_R14) | \ (1ULL << PERF_REG_X86_R15)) =20 +/* user space rdpmc control values */ +enum { + X86_USER_RDPMC_NEVER_ENABLE =3D 0, + X86_USER_RDPMC_CONDITIONAL_ENABLE =3D 1, + X86_USER_RDPMC_ALWAYS_ENABLE =3D 2, +}; + /* * Per register state. */ --=20 2.34.1 From nobody Sun Feb 8 11:25:59 2026 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 0E62C30F53A; Mon, 12 Jan 2026 05:20:38 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195241; cv=none; b=mplCRQqkSLmAsJ6zGswzHaVrQOOY6g/GdzgX4dxtqDphEdGufu9AG7y7sx7SCmxElzfxq4nw15AjFz0NuGBAxCc2IJXbln96tnWOgwsO9kDqh4dHOmzNtpwYLQDcP7FVDPU4E4A7miq0k6Mn3VwtFkQ3cJQpv5BOepaur8b8jA4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1768195241; c=relaxed/simple; bh=s6K1uLWKH/5NE7aLWc20bvX3kb76eskpz3ufo3pdp0s=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=SOwBhsttPlm+CEQvnZbcD/IkfRzjrRaBmHzD7M/LtznlPeO5ytM5mBwTno8N+mHeE7gRCVqFzgmzAroVb4EinFxrzBcNtmC0Tdzw0p3K48MbZQ/E4sjW8GljOLPGHROdE+FuWtapCVcl6cmO2wCDxcBtnywEJm6ifQIYynjJvEU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com; spf=pass smtp.mailfrom=linux.intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=C11MnBjB; arc=none smtp.client-ip=192.198.163.18 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="C11MnBjB" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1768195239; x=1799731239; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=s6K1uLWKH/5NE7aLWc20bvX3kb76eskpz3ufo3pdp0s=; b=C11MnBjBfzRlFmnNeV3gTfktdSNuWm5FBk7M6BxoTCDf51eBXJtGiJVO EuS5gqPRhgyTM61wREqcUmsghbLTSX0m88q2+udcTD3iBPH/L2B/3K7b5 EMsKQPqj7l0IrfNiLhVYI7IUyLmLOsBtqE4KvS8y8IVtKBR6bOtw52vhW +S5Kkt+MRE0M3wLv2liAcaNM8bu6Dww26r5xqehbhnpnIKXfWsxRq/fTz OrCbckDUntouGtNUfXO7HkvO7ISBrxpMOssam34V8k37NGKzV4NGfklya ZeFbIeBtFHJi2mIfjHDuvmDk7poZrqQsAe0MUSbseHvOrUka2QcttnkBQ A==; X-CSE-ConnectionGUID: YXZ5/vCmSn+N9JNkups2DA== X-CSE-MsgGUID: xdsMllCHToWWTVNpWpTryQ== X-IronPort-AV: E=McAfee;i="6800,10657,11668"; a="68673251" X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="68673251" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa112.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jan 2026 21:20:39 -0800 X-CSE-ConnectionGUID: ca9yMrsaQviXUKI3vaK2og== X-CSE-MsgGUID: Tyh0JKojTsSxwC0mP9EcKg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,219,1763452800"; d="scan'208";a="235233494" Received: from spr.sh.intel.com ([10.112.229.196]) by fmviesa001.fm.intel.com with ESMTP; 11 Jan 2026 21:20:35 -0800 From: Dapeng Mi To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Ian Rogers , Adrian Hunter , Alexander Shishkin , Andi Kleen , Eranian Stephane Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Dapeng Mi , Zide Chen , Falcon Thomas , Xudong Hao , Dapeng Mi Subject: [Patch v2 7/7] perf/x86/intel: Add support for rdpmc user disable feature Date: Mon, 12 Jan 2026 13:16:49 +0800 Message-Id: <20260112051649.1113435-8-dapeng1.mi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> References: <20260112051649.1113435-1-dapeng1.mi@linux.intel.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Starting with Panther Cove, the rdpmc user disable feature is supported. This feature allows the perf system to disable user space rdpmc reads at the counter level. Currently, when a global counter is active, any user with rdpmc rights can read it, even if perf access permissions forbid it (e.g., disallow reading ring 0 counters). The rdpmc user disable feature mitigates this security concern. Details: - A new RDPMC_USR_DISABLE bit (bit 37) in each EVNTSELx MSR indicates that the GP counter cannot be read by RDPMC in ring 3. - New RDPMC_USR_DISABLE bits in IA32_FIXED_CTR_CTRL MSR (bits 33, 37, 41, 45, etc.) for fixed counters 0, 1, 2, 3, etc. - When calling rdpmc instruction for counter x, the following pseudo code demonstrates how the counter value is obtained: If (!CPL0 && RDPMC_USR_DISABLE[x] =3D=3D 1) ? 0 : counter_value; - RDPMC_USR_DISABLE is enumerated by CPUID.0x23.0.EBX[2]. This patch extends the current global user space rdpmc control logic via the sysfs interface (/sys/devices/cpu/rdpmc) as follows: - rdpmc =3D 0: Global user space rdpmc and counter-level user space rdpmc for all counters are both disabled. - rdpmc =3D 1: Global user space rdpmc is enabled during the mmap-enabled time window, and counter-level user space rdpmc is enabled only for non-system-wide events. This prevents counter data leaks as count data is cleared during context switches. - rdpmc =3D 2: Global user space rdpmc and counter-level user space rdpmc for all counters are enabled unconditionally. The new rdpmc settings only affect newly activated perf events; currently active perf events remain unaffected. This simplifies and cleans up the code. The default value of rdpmc remains unchanged at 1. For more details about rdpmc user disable, please refer to chapter 15 "RDPMC USER DISABLE" in ISE documentation. ISE: https://www.intel.com/content/www/us/en/content-details/869288/intel-a= rchitecture-instruction-set-extensions-programming-reference.html Signed-off-by: Dapeng Mi --- .../sysfs-bus-event_source-devices-rdpmc | 40 +++++++++++++++++++ arch/x86/events/core.c | 21 ++++++++++ arch/x86/events/intel/core.c | 26 ++++++++++++ arch/x86/events/perf_event.h | 6 +++ arch/x86/include/asm/perf_event.h | 8 +++- 5 files changed, 99 insertions(+), 2 deletions(-) create mode 100644 Documentation/ABI/testing/sysfs-bus-event_source-device= s-rdpmc diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-rdpmc= b/Documentation/ABI/testing/sysfs-bus-event_source-devices-rdpmc new file mode 100644 index 000000000000..d004527ab13e --- /dev/null +++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-rdpmc @@ -0,0 +1,40 @@ +What: /sys/bus/event_source/devices/cpu.../rdpmc +Date: November 2011 +KernelVersion: 3.10 +Contact: Linux kernel mailing list linux-kernel@vger.kernel.org +Description: The /sys/bus/event_source/devices/cpu.../rdpmc attribute + is used to show/manage if rdpmc instruction can be + executed in user space. This attribute supports 3 numbers. + - rdpmc =3D 0 + user space rdpmc is globally disabled for all PMU + counters. + - rdpmc =3D 1 + user space rdpmc is globally enabled only in event mmap + ioctl called time window. If the mmap region is unmapped, + user space rdpmc is disabled again. + - rdpmc =3D 2 + user space rdpmc is globally enabled for all PMU + counters. + + In the Intel platforms supporting counter level's user + space rdpmc disable feature (CPUID.23H.EBX[2] =3D 1), the + meaning of 3 numbers is extended to + - rdpmc =3D 0 + global user space rdpmc and counter level's user space + rdpmc of all counters are both disabled. + - rdpmc =3D 1 + No changes on behavior of global user space rdpmc. + counter level's rdpmc of system-wide events is disabled + but counter level's rdpmc of non-system-wide events is + enabled. + - rdpmc =3D 2 + global user space rdpmc and counter level's user space + rdpmc of all counters are both enabled unconditionally. + + The default value of rdpmc is 1. + + Please notice global user space rdpmc's behavior would + change immediately along with the rdpmc value's change, + but the behavior of counter level's user space rdpmc + won't take effect immediately until the event is + reactivated or recreated. diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index c2717cb5034f..6df73e8398cd 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -2616,6 +2616,27 @@ static ssize_t get_attr_rdpmc(struct device *cdev, return snprintf(buf, 40, "%d\n", x86_pmu.attr_rdpmc); } =20 +/* + * Behaviors of rdpmc value: + * - rdpmc =3D 0 + * global user space rdpmc and counter level's user space rdpmc of all + * counters are both disabled. + * - rdpmc =3D 1 + * global user space rdpmc is enabled in mmap enabled time window and + * counter level's user space rdpmc is enabled for only non system-wide + * events. Counter level's user space rdpmc of system-wide events is + * still disabled by default. This won't introduce counter data leak for + * non system-wide events since their count data would be cleared when + * context switches. + * - rdpmc =3D 2 + * global user space rdpmc and counter level's user space rdpmc of all + * counters are enabled unconditionally. + * + * Suppose the rdpmc value won't be changed frequently, don't dynamically + * reschedule events to make the new rpdmc value take effect on active perf + * events immediately, the new rdpmc value would only impact the new + * activated perf events. This makes code simpler and cleaner. + */ static ssize_t set_attr_rdpmc(struct device *cdev, struct device_attribute *attr, const char *buf, size_t count) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index dd488a095f33..77cf849a1381 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -3128,6 +3128,8 @@ static void intel_pmu_enable_fixed(struct perf_event = *event) bits |=3D INTEL_FIXED_0_USER; if (hwc->config & ARCH_PERFMON_EVENTSEL_OS) bits |=3D INTEL_FIXED_0_KERNEL; + if (hwc->config & ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE) + bits |=3D INTEL_FIXED_0_RDPMC_USER_DISABLE; =20 /* * ANY bit is supported in v3 and up @@ -3263,6 +3265,26 @@ static void intel_pmu_enable_event_ext(struct perf_e= vent *event) __intel_pmu_update_event_ext(hwc->idx, ext); } =20 +static void intel_pmu_update_rdpmc_user_disable(struct perf_event *event) +{ + /* + * Counter scope's user-space rdpmc is disabled by default + * except two cases. + * a. rdpmc =3D 2 (user space rdpmc enabled unconditionally) + * b. rdpmc =3D 1 and the event is not a system-wide event. + * The count of non-system-wide events would be cleared when + * context switches, so no count data is leaked. + */ + if (x86_pmu_has_rdpmc_user_disable(event->pmu)) { + if (x86_pmu.attr_rdpmc =3D=3D X86_USER_RDPMC_ALWAYS_ENABLE || + (x86_pmu.attr_rdpmc =3D=3D X86_USER_RDPMC_CONDITIONAL_ENABLE && + event->ctx->task)) + event->hw.config &=3D ~ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE; + else + event->hw.config |=3D ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE; + } +} + DEFINE_STATIC_CALL_NULL(intel_pmu_enable_event_ext, intel_pmu_enable_event= _ext); =20 static void intel_pmu_enable_event(struct perf_event *event) @@ -3271,6 +3293,8 @@ static void intel_pmu_enable_event(struct perf_event = *event) struct hw_perf_event *hwc =3D &event->hw; int idx =3D hwc->idx; =20 + intel_pmu_update_rdpmc_user_disable(event); + if (unlikely(event->attr.precise_ip)) static_call(x86_pmu_pebs_enable)(event); =20 @@ -5863,6 +5887,8 @@ static void update_pmu_cap(struct pmu *pmu) hybrid(pmu, config_mask) |=3D ARCH_PERFMON_EVENTSEL_UMASK2; if (ebx_0.split.eq) hybrid(pmu, config_mask) |=3D ARCH_PERFMON_EVENTSEL_EQ; + if (ebx_0.split.rdpmc_user_disable) + hybrid(pmu, config_mask) |=3D ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE; =20 if (eax_0.split.cntr_subleaf) { cpuid_count(ARCH_PERFMON_EXT_LEAF, ARCH_PERFMON_NUM_COUNTER_LEAF, diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index 24a81d2916e9..cd337f3ffd01 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1333,6 +1333,12 @@ static inline u64 x86_pmu_get_event_config(struct pe= rf_event *event) return event->attr.config & hybrid(event->pmu, config_mask); } =20 +static inline bool x86_pmu_has_rdpmc_user_disable(struct pmu *pmu) +{ + return !!(hybrid(pmu, config_mask) & + ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE); +} + extern struct event_constraint emptyconstraint; =20 extern struct event_constraint unconstrained; diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_= event.h index 0d9af4135e0a..ff5acb8b199b 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -33,6 +33,7 @@ #define ARCH_PERFMON_EVENTSEL_CMASK 0xFF000000ULL #define ARCH_PERFMON_EVENTSEL_BR_CNTR (1ULL << 35) #define ARCH_PERFMON_EVENTSEL_EQ (1ULL << 36) +#define ARCH_PERFMON_EVENTSEL_RDPMC_USER_DISABLE (1ULL << 37) #define ARCH_PERFMON_EVENTSEL_UMASK2 (0xFFULL << 40) =20 #define INTEL_FIXED_BITS_STRIDE 4 @@ -40,6 +41,7 @@ #define INTEL_FIXED_0_USER (1ULL << 1) #define INTEL_FIXED_0_ANYTHREAD (1ULL << 2) #define INTEL_FIXED_0_ENABLE_PMI (1ULL << 3) +#define INTEL_FIXED_0_RDPMC_USER_DISABLE (1ULL << 33) #define INTEL_FIXED_3_METRICS_CLEAR (1ULL << 2) =20 #define HSW_IN_TX (1ULL << 32) @@ -50,7 +52,7 @@ #define INTEL_FIXED_BITS_MASK \ (INTEL_FIXED_0_KERNEL | INTEL_FIXED_0_USER | \ INTEL_FIXED_0_ANYTHREAD | INTEL_FIXED_0_ENABLE_PMI | \ - ICL_FIXED_0_ADAPTIVE) + ICL_FIXED_0_ADAPTIVE | INTEL_FIXED_0_RDPMC_USER_DISABLE) =20 #define intel_fixed_bits_by_idx(_idx, _bits) \ ((_bits) << ((_idx) * INTEL_FIXED_BITS_STRIDE)) @@ -226,7 +228,9 @@ union cpuid35_ebx { unsigned int umask2:1; /* EQ-bit Supported */ unsigned int eq:1; - unsigned int reserved:30; + /* rdpmc user disable Supported */ + unsigned int rdpmc_user_disable:1; + unsigned int reserved:29; } split; unsigned int full; }; --=20 2.34.1