Merge tag 'perf-tools-for-v5.19-2022-05-28' of git://git.kernel.org/p…

…ub/scm/linux/kernel/git/acme/linux

Pull more perf tools updates from Arnaldo Carvalho de Melo:

 - Add BPF based off-CPU profiling

 - Improvements for system wide recording, specially for Intel PT

 - Improve DWARF unwinding on arm64

 - Support Arm CoreSight trace data disassembly in 'perf script' python

 - Fix build with new libbpf version, related to supporting older
   versions of distro released libbpf packages

 - Fix event syntax error caused by ExtSel in the JSON events infra

 - Use stdio interface if slang is not supported in 'perf c2c'

 - Add 'perf test' checking for perf stat CSV output

 - Sync the msr-index.h copy with the kernel sources

* tag 'perf-tools-for-v5.19-2022-05-28' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (38 commits)
  tools arch x86: Sync the msr-index.h copy with the kernel sources
  perf scripts python: Support Arm CoreSight trace data disassembly
  perf scripting python: Expose dso and map information
  perf jevents: Fix event syntax error caused by ExtSel
  perf tools arm64: Add support for VG register
  perf unwind arm64: Decouple Libunwind register names from Perf
  perf unwind: Use dynamic register set for DWARF unwind
  perf tools arm64: Copy perf_regs.h from the kernel
  perf unwind arm64: Use perf's copy of kernel headers
  perf c2c: Use stdio interface if slang is not supported
  perf test: Add a basic offcpu profiling test
  perf record: Add cgroup support for off-cpu profiling
  perf record: Handle argument change in sched_switch
  perf record: Implement basic filtering for off-cpu
  perf record: Enable off-cpu analysis with BPF
  perf report: Do not extend sample type of bpf-output event
  perf test: Add checking for perf stat CSV output.
  perf tools: Allow system-wide events to keep their own threads
  perf tools: Allow system-wide events to keep their own CPUs
  libperf evsel: Add comments for booleans
  ...

tools/arch/arm64/include/uapi/asm/perf_regs.h

@@ -36,6 +36,11 @@ enum perf_event_arm_regs {
 	PERF_REG_ARM64_LR,
 	PERF_REG_ARM64_SP,
 	PERF_REG_ARM64_PC,
-	PERF_REG_ARM64_MAX,
+
+	/* Extended/pseudo registers */
+	PERF_REG_ARM64_VG = 46, // SVE Vector Granule
+
+	PERF_REG_ARM64_MAX = PERF_REG_ARM64_PC + 1,
+	PERF_REG_ARM64_EXTENDED_MAX = PERF_REG_ARM64_VG + 1
 };
 #endif /* _ASM_ARM64_PERF_REGS_H */

tools/arch/x86/include/asm/msr-index.h

@@ -76,6 +76,8 @@
 
 /* Abbreviated from Intel SDM name IA32_CORE_CAPABILITIES */
 #define MSR_IA32_CORE_CAPS			  0x000000cf
+#define MSR_IA32_CORE_CAPS_INTEGRITY_CAPS_BIT	  2
+#define MSR_IA32_CORE_CAPS_INTEGRITY_CAPS	  BIT(MSR_IA32_CORE_CAPS_INTEGRITY_CAPS_BIT)
 #define MSR_IA32_CORE_CAPS_SPLIT_LOCK_DETECT_BIT  5
 #define MSR_IA32_CORE_CAPS_SPLIT_LOCK_DETECT	  BIT(MSR_IA32_CORE_CAPS_SPLIT_LOCK_DETECT_BIT)
 
@@ -154,6 +156,11 @@
 #define MSR_IA32_POWER_CTL		0x000001fc
 #define MSR_IA32_POWER_CTL_BIT_EE	19
 
+/* Abbreviated from Intel SDM name IA32_INTEGRITY_CAPABILITIES */
+#define MSR_INTEGRITY_CAPS			0x000002d9
+#define MSR_INTEGRITY_CAPS_PERIODIC_BIST_BIT	4
+#define MSR_INTEGRITY_CAPS_PERIODIC_BIST	BIT(MSR_INTEGRITY_CAPS_PERIODIC_BIST_BIT)
+
 #define MSR_LBR_NHM_FROM		0x00000680
 #define MSR_LBR_NHM_TO			0x000006c0
 #define MSR_LBR_CORE_FROM		0x00000040
@@ -312,6 +319,7 @@
 
 /* Run Time Average Power Limiting (RAPL) Interface */
 
+#define MSR_VR_CURRENT_CONFIG	0x00000601
 #define MSR_RAPL_POWER_UNIT		0x00000606
 
 #define MSR_PKG_POWER_LIMIT		0x00000610
@@ -502,8 +510,10 @@
 #define MSR_AMD64_SEV			0xc0010131
 #define MSR_AMD64_SEV_ENABLED_BIT	0
 #define MSR_AMD64_SEV_ES_ENABLED_BIT	1
+#define MSR_AMD64_SEV_SNP_ENABLED_BIT	2
 #define MSR_AMD64_SEV_ENABLED		BIT_ULL(MSR_AMD64_SEV_ENABLED_BIT)
 #define MSR_AMD64_SEV_ES_ENABLED	BIT_ULL(MSR_AMD64_SEV_ES_ENABLED_BIT)
+#define MSR_AMD64_SEV_SNP_ENABLED	BIT_ULL(MSR_AMD64_SEV_SNP_ENABLED_BIT)
 
 #define MSR_AMD64_VIRT_SPEC_CTRL	0xc001011f
 
@@ -524,6 +534,11 @@
 #define AMD_CPPC_DES_PERF(x)		(((x) & 0xff) << 16)
 #define AMD_CPPC_ENERGY_PERF_PREF(x)	(((x) & 0xff) << 24)
 
+/* AMD Performance Counter Global Status and Control MSRs */
+#define MSR_AMD64_PERF_CNTR_GLOBAL_STATUS	0xc0000300
+#define MSR_AMD64_PERF_CNTR_GLOBAL_CTL		0xc0000301
+#define MSR_AMD64_PERF_CNTR_GLOBAL_STATUS_CLR	0xc0000302
+
 /* Fam 17h MSRs */
 #define MSR_F17H_IRPERF			0xc00000e9
 
@@ -688,6 +703,10 @@
 #define MSR_IA32_PERF_CTL		0x00000199
 #define INTEL_PERF_CTL_MASK		0xffff
 
+/* AMD Branch Sampling configuration */
+#define MSR_AMD_DBG_EXTN_CFG		0xc000010f
+#define MSR_AMD_SAMP_BR_FROM		0xc0010300
+
 #define MSR_IA32_MPERF			0x000000e7
 #define MSR_IA32_APERF			0x000000e8
 

tools/build/Makefile.feature

@@ -99,6 +99,10 @@ FEATURE_TESTS_EXTRA :=                  \
          clang                          \
          libbpf                         \
          libbpf-btf__load_from_kernel_by_id \
+         libbpf-bpf_prog_load           \
+         libbpf-bpf_object__next_program \
+         libbpf-bpf_object__next_map    \
+         libbpf-bpf_create_map		\
          libpfm4                        \
          libdebuginfod			\
          clang-bpf-co-re

tools/build/feature/Makefile

@@ -58,6 +58,11 @@ FILES=                                          \
          test-bpf.bin                           \
          test-libbpf.bin                        \
          test-libbpf-btf__load_from_kernel_by_id.bin	\
+         test-libbpf-bpf_prog_load.bin          \
+         test-libbpf-bpf_map_create.bin		\
+         test-libbpf-bpf_object__next_program.bin \
+         test-libbpf-bpf_object__next_map.bin   \
+         test-libbpf-btf__raw_data.bin          \
          test-get_cpuid.bin                     \
          test-sdt.bin                           \
          test-cxx.bin                           \
@@ -291,6 +296,21 @@ $(OUTPUT)test-libbpf.bin:
 $(OUTPUT)test-libbpf-btf__load_from_kernel_by_id.bin:
 	$(BUILD) -lbpf
 
+$(OUTPUT)test-libbpf-bpf_prog_load.bin:
+	$(BUILD) -lbpf
+
+$(OUTPUT)test-libbpf-bpf_map_create.bin:
+	$(BUILD) -lbpf
+
+$(OUTPUT)test-libbpf-bpf_object__next_program.bin:
+	$(BUILD) -lbpf
+
+$(OUTPUT)test-libbpf-bpf_object__next_map.bin:
+	$(BUILD) -lbpf
+
+$(OUTPUT)test-libbpf-btf__raw_data.bin:
+	$(BUILD) -lbpf
+
 $(OUTPUT)test-sdt.bin:
 	$(BUILD)
 

tools/build/feature/test-libbpf-bpf_map_create.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <bpf/bpf.h>
+
+int main(void)
+{
+	return bpf_map_create(0 /* map_type */, NULL /* map_name */, 0, /* key_size */,
+			      0 /* value_size */, 0 /* max_entries */, NULL /* opts */);
+}

tools/build/feature/test-libbpf-bpf_object__next_map.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <bpf/libbpf.h>
+
+int main(void)
+{
+	bpf_object__next_map(NULL /* obj */, NULL /* prev */);
+	return 0;
+}

tools/build/feature/test-libbpf-bpf_object__next_program.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <bpf/libbpf.h>
+
+int main(void)
+{
+	bpf_object__next_program(NULL /* obj */, NULL /* prev */);
+	return 0;
+}

tools/build/feature/test-libbpf-bpf_prog_load.c

@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <bpf/bpf.h>
+
+int main(void)
+{
+	return bpf_prog_load(0 /* prog_type */, NULL /* prog_name */,
+			     NULL /* license */, NULL /* insns */,
+			     0 /* insn_cnt */, NULL /* opts */);
+}

tools/build/feature/test-libbpf-btf__load_from_kernel_by_id.c

@@ -1,7 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0
-#include <bpf/libbpf.h>
+#include <bpf/btf.h>
 
 int main(void)
 {
-	return btf__load_from_kernel_by_id(20151128, NULL);
+	btf__load_from_kernel_by_id(20151128);
+	return 0;
 }

tools/build/feature/test-libbpf-btf__raw_data.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <bpf/btf.h>
+
+int main(void)
+{
+	btf__raw_data(NULL /* btf_ro */, NULL /* size */);
+	return 0;
+}

tools/lib/perf/evlist.c

@@ -23,6 +23,7 @@
 #include <perf/cpumap.h>
 #include <perf/threadmap.h>
 #include <api/fd/array.h>
+#include "internal.h"
 
 void perf_evlist__init(struct perf_evlist *evlist)
 {
@@ -39,19 +40,23 @@ static void __perf_evlist__propagate_maps(struct perf_evlist *evlist,
 	 * We already have cpus for evsel (via PMU sysfs) so
 	 * keep it, if there's no target cpu list defined.
 	 */
-	if (!evsel->own_cpus || evlist->has_user_cpus) {
-		perf_cpu_map__put(evsel->cpus);
-		evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
-	} else if (!evsel->system_wide && perf_cpu_map__empty(evlist->user_requested_cpus)) {
+	if (!evsel->own_cpus ||
+	    (!evsel->system_wide && evlist->has_user_cpus) ||
+	    (!evsel->system_wide &&
+	     !evsel->requires_cpu &&
+	     perf_cpu_map__empty(evlist->user_requested_cpus))) {
 		perf_cpu_map__put(evsel->cpus);
 		evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
 	} else if (evsel->cpus != evsel->own_cpus) {
 		perf_cpu_map__put(evsel->cpus);
 		evsel->cpus = perf_cpu_map__get(evsel->own_cpus);
 	}
 
-	perf_thread_map__put(evsel->threads);
-	evsel->threads = perf_thread_map__get(evlist->threads);
+	if (!evsel->system_wide) {
+		perf_thread_map__put(evsel->threads);
+		evsel->threads = perf_thread_map__get(evlist->threads);
+	}
+
 	evlist->all_cpus = perf_cpu_map__merge(evlist->all_cpus, evsel->cpus);
 }
 
@@ -298,7 +303,7 @@ int perf_evlist__id_add_fd(struct perf_evlist *evlist,
 
 int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)
 {
-	int nr_cpus = perf_cpu_map__nr(evlist->user_requested_cpus);
+	int nr_cpus = perf_cpu_map__nr(evlist->all_cpus);
 	int nr_threads = perf_thread_map__nr(evlist->threads);
 	int nfds = 0;
 	struct perf_evsel *evsel;
@@ -428,9 +433,9 @@ static void perf_evlist__set_mmap_first(struct perf_evlist *evlist, struct perf_
 static int
 mmap_per_evsel(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
 	       int idx, struct perf_mmap_param *mp, int cpu_idx,
-	       int thread, int *_output, int *_output_overwrite)
+	       int thread, int *_output, int *_output_overwrite, int *nr_mmaps)
 {
-	struct perf_cpu evlist_cpu = perf_cpu_map__cpu(evlist->user_requested_cpus, cpu_idx);
+	struct perf_cpu evlist_cpu = perf_cpu_map__cpu(evlist->all_cpus, cpu_idx);
 	struct perf_evsel *evsel;
 	int revent;
 
@@ -484,6 +489,8 @@ mmap_per_evsel(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
 			if (ops->mmap(map, mp, *output, evlist_cpu) < 0)
 				return -1;
 
+			*nr_mmaps += 1;
+
 			if (!idx)
 				perf_evlist__set_mmap_first(evlist, map, overwrite);
 		} else {
@@ -512,35 +519,13 @@ mmap_per_evsel(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
 	return 0;
 }
 
-static int
-mmap_per_thread(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
-		struct perf_mmap_param *mp)
-{
-	int thread;
-	int nr_threads = perf_thread_map__nr(evlist->threads);
-
-	for (thread = 0; thread < nr_threads; thread++) {
-		int output = -1;
-		int output_overwrite = -1;
-
-		if (mmap_per_evsel(evlist, ops, thread, mp, 0, thread,
-				   &output, &output_overwrite))
-			goto out_unmap;
-	}
-
-	return 0;
-
-out_unmap:
-	perf_evlist__munmap(evlist);
-	return -1;
-}
-
 static int
 mmap_per_cpu(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
 	     struct perf_mmap_param *mp)
 {
 	int nr_threads = perf_thread_map__nr(evlist->threads);
-	int nr_cpus    = perf_cpu_map__nr(evlist->user_requested_cpus);
+	int nr_cpus    = perf_cpu_map__nr(evlist->all_cpus);
+	int nr_mmaps = 0;
 	int cpu, thread;
 
 	for (cpu = 0; cpu < nr_cpus; cpu++) {
@@ -549,11 +534,14 @@ mmap_per_cpu(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops,
 
 		for (thread = 0; thread < nr_threads; thread++) {
 			if (mmap_per_evsel(evlist, ops, cpu, mp, cpu,
-					   thread, &output, &output_overwrite))
+					   thread, &output, &output_overwrite, &nr_mmaps))
 				goto out_unmap;
 		}
 	}
 
+	if (nr_mmaps != evlist->nr_mmaps)
+		pr_err("Miscounted nr_mmaps %d vs %d\n", nr_mmaps, evlist->nr_mmaps);
+
 	return 0;
 
 out_unmap:
@@ -565,9 +553,14 @@ static int perf_evlist__nr_mmaps(struct perf_evlist *evlist)
 {
 	int nr_mmaps;
 
-	nr_mmaps = perf_cpu_map__nr(evlist->user_requested_cpus);
-	if (perf_cpu_map__empty(evlist->user_requested_cpus))
-		nr_mmaps = perf_thread_map__nr(evlist->threads);
+	/* One for each CPU */
+	nr_mmaps = perf_cpu_map__nr(evlist->all_cpus);
+	if (perf_cpu_map__empty(evlist->all_cpus)) {
+		/* Plus one for each thread */
+		nr_mmaps += perf_thread_map__nr(evlist->threads);
+		/* Minus the per-thread CPU (-1) */
+		nr_mmaps -= 1;
+	}
 
 	return nr_mmaps;
 }
@@ -577,7 +570,6 @@ int perf_evlist__mmap_ops(struct perf_evlist *evlist,
 			  struct perf_mmap_param *mp)
 {
 	struct perf_evsel *evsel;
-	const struct perf_cpu_map *cpus = evlist->user_requested_cpus;
 
 	if (!ops || !ops->get || !ops->mmap)
 		return -EINVAL;
@@ -596,9 +588,6 @@ int perf_evlist__mmap_ops(struct perf_evlist *evlist,
 	if (evlist->pollfd.entries == NULL && perf_evlist__alloc_pollfd(evlist) < 0)
 		return -ENOMEM;
 
-	if (perf_cpu_map__empty(cpus))
-		return mmap_per_thread(evlist, ops, mp);
-
 	return mmap_per_cpu(evlist, ops, mp);
 }
 

tools/lib/perf/include/internal/evsel.h

@@ -49,7 +49,18 @@ struct perf_evsel {
 
 	/* parse modifier helper */
 	int			 nr_members;
+	/*
+	 * system_wide is for events that need to be on every CPU, irrespective
+	 * of user requested CPUs or threads. Map propagation will set cpus to
+	 * this event's own_cpus, whereby they will contribute to evlist
+	 * all_cpus.
+	 */
 	bool			 system_wide;
+	/*
+	 * Some events, for example uncore events, require a CPU.
+	 * i.e. it cannot be the 'any CPU' value of -1.
+	 */
+	bool			 requires_cpu;
 	int			 idx;
 };
 

tools/perf/Documentation/perf-record.txt

@@ -758,6 +758,16 @@ include::intel-hybrid.txt[]
 	If the URLs is not specified, the value of DEBUGINFOD_URLS
 	system environment variable is used.
 
+--off-cpu::
+	Enable off-cpu profiling with BPF.  The BPF program will collect
+	task scheduling information with (user) stacktrace and save them
+	as sample data of a software event named "offcpu-time".  The
+	sample period will have the time the task slept in nanoseconds.
+
+	Note that BPF can collect stack traces using frame pointer ("fp")
+	only, as of now.  So the applications built without the frame
+	pointer might see bogus addresses.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-intel-pt[1]