perf/x86/intel/uncore: Add Knights Landing uncore PMU support

Knights Landing uncore performance monitoring (perfmon) is derived from
Haswell-EP uncore perfmon with several differences. One notable difference
is in PCI device IDs. Knights Landing uses common PCI device ID for
multiple instances of an uncore PMU device type. In Haswell-EP, each
instance of a PMU device type has a unique device ID.

Knights Landing uncore components that have performance monitoring units
are UBOX, CHA, EDC, MC, M2PCIe, IRP and PCU. Perfmon registers in EDC, MC,
IRP, and M2PCIe reside in the PCIe configuration space. Perfmon registers
in UBOX, CHA and PCU are accessed via the MSR interface.

For more details, please refer to the public document:

  https://software.intel.com/sites/default/files/managed/15/8d/IntelXeonPhi%E2%84%A2x200ProcessorPerformanceMonitoringReferenceManual_Volume1_Registers_v0%206.pdf

Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andi Kleen <andi.kleen@intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Harish Chegondi <harish.chegondi@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Lukasz Anaczkowski <lukasz.anaczkowski@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: http://lkml.kernel.org/r/8ac513981264c3eb10343a3f523f19cc5a2d12fe.1449470704.git.harish.chegondi@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

arch/x86/kernel/cpu/perf_event_intel_uncore.c

@@ -884,6 +884,15 @@ static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id
 	 * each box has a different function id.
 	 */
 	pmu = &type->pmus[UNCORE_PCI_DEV_IDX(id->driver_data)];
+	/* Knights Landing uses a common PCI device ID for multiple instances of
+	 * an uncore PMU device type. There is only one entry per device type in
+	 * the knl_uncore_pci_ids table inspite of multiple devices present for
+	 * some device types. Hence PCI device idx would be 0 for all devices.
+	 * So increment pmu pointer to point to an unused array element.
+	 */
+	if (boot_cpu_data.x86_model == 87)
+		while (pmu->func_id >= 0)
+			pmu++;
 	if (pmu->func_id < 0)
 		pmu->func_id = pdev->devfn;
 	else
@@ -983,6 +992,9 @@ static int __init uncore_pci_init(void)
 	case 61: /* Broadwell */
 		ret = bdw_uncore_pci_init();
 		break;
+	case 87: /* Knights Landing */
+		ret = knl_uncore_pci_init();
+		break;
 	default:
 		return 0;
 	}
@@ -1292,6 +1304,9 @@ static int __init uncore_cpu_init(void)
 	case 86: /* BDX-DE */
 		bdx_uncore_cpu_init();
 		break;
+	case 87: /* Knights Landing */
+		knl_uncore_cpu_init();
+		break;
 	default:
 		return 0;
 	}

arch/x86/kernel/cpu/perf_event_intel_uncore.h

@@ -338,6 +338,7 @@ int hsw_uncore_pci_init(void);
 int bdw_uncore_pci_init(void);
 void snb_uncore_cpu_init(void);
 void nhm_uncore_cpu_init(void);
+int snb_pci2phy_map_init(int devid);
 
 /* perf_event_intel_uncore_snbep.c */
 int snbep_uncore_pci_init(void);
@@ -348,6 +349,8 @@ int hswep_uncore_pci_init(void);
 void hswep_uncore_cpu_init(void);
 int bdx_uncore_pci_init(void);
 void bdx_uncore_cpu_init(void);
+int knl_uncore_pci_init(void);
+void knl_uncore_cpu_init(void);
 
 /* perf_event_intel_uncore_nhmex.c */
 void nhmex_uncore_cpu_init(void);

arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c

@@ -417,7 +417,7 @@ static void snb_uncore_imc_event_del(struct perf_event *event, int flags)
 	}
 }
 
-static int snb_pci2phy_map_init(int devid)
+int snb_pci2phy_map_init(int devid)
 {
 	struct pci_dev *dev = NULL;
 	struct pci2phy_map *map;