Merge tag 'perf-tools-for-v5.16-2021-11-13' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull more perf tools updates from Arnaldo Carvalho de Melo:
 "Hardware tracing:

   - ARM:
      * Print the size of the buffer size consistently in hexadecimal in
        ARM Coresight.
      * Add Coresight snapshot mode support.
      * Update --switch-events docs in 'perf record'.
      * Support hardware-based PID tracing.
      * Track task context switch for cpu-mode events.

   - Vendor events:
      * Add metric events JSON file for power10 platform

  perf test:

   - Get 'perf test' unit tests closer to kunit.

   - Topology tests improvements.

   - Remove bashisms from some tests.

  perf bench:

   - Fix memory leak of perf_cpu_map__new() in the futex benchmarks.

  libbpf:

   - Add some more weak libbpf functions o allow building with the
     libbpf versions, old ones, present in distros.

  libbeauty:

   - Translate [gs]setsockopt 'level' argument integer values to
     strings.

  tools headers UAPI:

   - Sync futex_waitv, arch prctl, sound, i195_drm and msr-index files
     with the kernel sources.

  Documentation:

   - Add documentation to 'struct symbol'.

   - Synchronize the definition of enum perf_hw_id with code in
     tools/perf/design.txt"

* tag 'perf-tools-for-v5.16-2021-11-13' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (67 commits)
  perf tests: Remove bash constructs from stat_all_pmu.sh
  perf tests: Remove bash construct from record+zstd_comp_decomp.sh
  perf test: Remove bash construct from stat_bpf_counters.sh test
  perf bench futex: Fix memory leak of perf_cpu_map__new()
  tools arch x86: Sync the msr-index.h copy with the kernel sources
  tools headers UAPI: Sync drm/i915_drm.h with the kernel sources
  tools headers UAPI: Sync sound/asound.h with the kernel sources
  tools headers UAPI: Sync linux/prctl.h with the kernel sources
  tools headers UAPI: Sync arch prctl headers with the kernel sources
  perf tools: Add more weak libbpf functions
  perf bpf: Avoid memory leak from perf_env__insert_btf()
  perf symbols: Factor out annotation init/exit
  perf symbols: Bit pack to save a byte
  perf symbols: Add documentation to 'struct symbol'
  tools headers UAPI: Sync files changed by new futex_waitv syscall
  perf test bpf: Use ARRAY_CHECK() instead of ad-hoc equivalent, addressing array_size.cocci warning
  perf arm-spe: Support hardware-based PID tracing
  perf arm-spe: Save context ID in record
  perf arm-spe: Update --switch-events docs in 'perf record'
  perf arm-spe: Track task context switch for cpu-mode events
  ...

131 files changed, 2639 insertions, 1148 deletions

diff --git a/tools/arch/x86/include/asm/msr-index.h b/tools/arch/x86/include/asm/msr-index.h
index a7c413432b33..01e2650b9585 100644
--- a/tools/arch/x86/include/asm/msr-index.h
+++ b/tools/arch/x86/include/asm/msr-index.h
@@ -625,6 +625,8 @@
 
 #define MSR_IA32_BNDCFGS_RSVD		0x00000ffc
 
+#define MSR_IA32_XFD			0x000001c4
+#define MSR_IA32_XFD_ERR		0x000001c5
 #define MSR_IA32_XSS			0x00000da0
 
 #define MSR_IA32_APICBASE		0x0000001b
diff --git a/tools/arch/x86/include/uapi/asm/prctl.h b/tools/arch/x86/include/uapi/asm/prctl.h
index 5a6aac9fa41f..754a07856817 100644
--- a/tools/arch/x86/include/uapi/asm/prctl.h
+++ b/tools/arch/x86/include/uapi/asm/prctl.h
@@ -10,6 +10,10 @@
 #define ARCH_GET_CPUID		0x1011
 #define ARCH_SET_CPUID		0x1012
 
+#define ARCH_GET_XCOMP_SUPP	0x1021
+#define ARCH_GET_XCOMP_PERM	0x1022
+#define ARCH_REQ_XCOMP_PERM	0x1023
+
 #define ARCH_MAP_VDSO_X32	0x2001
 #define ARCH_MAP_VDSO_32	0x2002
 #define ARCH_MAP_VDSO_64	0x2003
diff --git a/tools/include/uapi/asm-generic/unistd.h b/tools/include/uapi/asm-generic/unistd.h
index 1c5fb86d455a..4557a8b6086f 100644
--- a/tools/include/uapi/asm-generic/unistd.h
+++ b/tools/include/uapi/asm-generic/unistd.h
@@ -880,8 +880,11 @@ __SYSCALL(__NR_memfd_secret, sys_memfd_secret)
 #define __NR_process_mrelease 448
 __SYSCALL(__NR_process_mrelease, sys_process_mrelease)
 
+#define __NR_futex_waitv 449
+__SYSCALL(__NR_futex_waitv, sys_futex_waitv)
+
 #undef __NR_syscalls
-#define __NR_syscalls 449
+#define __NR_syscalls 450
 
 /*
  * 32 bit systems traditionally used different
diff --git a/tools/include/uapi/drm/i915_drm.h b/tools/include/uapi/drm/i915_drm.h
index bde5860b3686..914ebd9290e5 100644
--- a/tools/include/uapi/drm/i915_drm.h
+++ b/tools/include/uapi/drm/i915_drm.h
@@ -1522,6 +1522,12 @@ struct drm_i915_gem_caching {
 #define I915_TILING_NONE	0
 #define I915_TILING_X		1
 #define I915_TILING_Y		2
+/*
+ * Do not add new tiling types here.  The I915_TILING_* values are for
+ * de-tiling fence registers that no longer exist on modern platforms.  Although
+ * the hardware may support new types of tiling in general (e.g., Tile4), we
+ * do not need to add them to the uapi that is specific to now-defunct ioctls.
+ */
 #define I915_TILING_LAST	I915_TILING_Y
 
 #define I915_BIT_6_SWIZZLE_NONE		0
@@ -1824,6 +1830,7 @@ struct drm_i915_gem_context_param {
  * Extensions:
  *   i915_context_engines_load_balance (I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE)
  *   i915_context_engines_bond (I915_CONTEXT_ENGINES_EXT_BOND)
+ *   i915_context_engines_parallel_submit (I915_CONTEXT_ENGINES_EXT_PARALLEL_SUBMIT)
  */
 #define I915_CONTEXT_PARAM_ENGINES	0xa
 
@@ -1846,6 +1853,55 @@ struct drm_i915_gem_context_param {
  * attempted to use it, never re-use this context param number.
  */
 #define I915_CONTEXT_PARAM_RINGSIZE	0xc
+
+/*
+ * I915_CONTEXT_PARAM_PROTECTED_CONTENT:
+ *
+ * Mark that the context makes use of protected content, which will result
+ * in the context being invalidated when the protected content session is.
+ * Given that the protected content session is killed on suspend, the device
+ * is kept awake for the lifetime of a protected context, so the user should
+ * make sure to dispose of them once done.
+ * This flag can only be set at context creation time and, when set to true,
+ * must be preceded by an explicit setting of I915_CONTEXT_PARAM_RECOVERABLE
+ * to false. This flag can't be set to true in conjunction with setting the
+ * I915_CONTEXT_PARAM_BANNABLE flag to false. Creation example:
+ *
+ * .. code-block:: C
+ *
+ *	struct drm_i915_gem_context_create_ext_setparam p_protected = {
+ *		.base = {
+ *			.name = I915_CONTEXT_CREATE_EXT_SETPARAM,
+ *		},
+ *		.param = {
+ *			.param = I915_CONTEXT_PARAM_PROTECTED_CONTENT,
+ *			.value = 1,
+ *		}
+ *	};
+ *	struct drm_i915_gem_context_create_ext_setparam p_norecover = {
+ *		.base = {
+ *			.name = I915_CONTEXT_CREATE_EXT_SETPARAM,
+ *			.next_extension = to_user_pointer(&p_protected),
+ *		},
+ *		.param = {
+ *			.param = I915_CONTEXT_PARAM_RECOVERABLE,
+ *			.value = 0,
+ *		}
+ *	};
+ *	struct drm_i915_gem_context_create_ext create = {
+ *		.flags = I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS,
+ *		.extensions = to_user_pointer(&p_norecover);
+ *	};
+ *
+ *	ctx_id = gem_context_create_ext(drm_fd, &create);
+ *
+ * In addition to the normal failure cases, setting this flag during context
+ * creation can result in the following errors:
+ *
+ * -ENODEV: feature not available
+ * -EPERM: trying to mark a recoverable or not bannable context as protected
+ */
+#define I915_CONTEXT_PARAM_PROTECTED_CONTENT    0xd
 /* Must be kept compact -- no holes and well documented */
 
 	__u64 value;
@@ -2050,6 +2106,135 @@ struct i915_context_engines_bond {
 } __attribute__((packed)) name__
 
 /**
+ * struct i915_context_engines_parallel_submit - Configure engine for
+ * parallel submission.
+ *
+ * Setup a slot in the context engine map to allow multiple BBs to be submitted
+ * in a single execbuf IOCTL. Those BBs will then be scheduled to run on the GPU
+ * in parallel. Multiple hardware contexts are created internally in the i915 to
+ * run these BBs. Once a slot is configured for N BBs only N BBs can be
+ * submitted in each execbuf IOCTL and this is implicit behavior e.g. The user
+ * doesn't tell the execbuf IOCTL there are N BBs, the execbuf IOCTL knows how
+ * many BBs there are based on the slot's configuration. The N BBs are the last
+ * N buffer objects or first N if I915_EXEC_BATCH_FIRST is set.
+ *
+ * The default placement behavior is to create implicit bonds between each
+ * context if each context maps to more than 1 physical engine (e.g. context is
+ * a virtual engine). Also we only allow contexts of same engine class and these
+ * contexts must be in logically contiguous order. Examples of the placement
+ * behavior are described below. Lastly, the default is to not allow BBs to be
+ * preempted mid-batch. Rather insert coordinated preemption points on all
+ * hardware contexts between each set of BBs. Flags could be added in the future
+ * to change both of these default behaviors.
+ *
+ * Returns -EINVAL if hardware context placement configuration is invalid or if
+ * the placement configuration isn't supported on the platform / submission
+ * interface.
+ * Returns -ENODEV if extension isn't supported on the platform / submission
+ * interface.
+ *
+ * .. code-block:: none
+ *
+ *	Examples syntax:
+ *	CS[X] = generic engine of same class, logical instance X
+ *	INVALID = I915_ENGINE_CLASS_INVALID, I915_ENGINE_CLASS_INVALID_NONE
+ *
+ *	Example 1 pseudo code:
+ *	set_engines(INVALID)
+ *	set_parallel(engine_index=0, width=2, num_siblings=1,
+ *		     engines=CS[0],CS[1])
+ *
+ *	Results in the following valid placement:
+ *	CS[0], CS[1]
+ *
+ *	Example 2 pseudo code:
+ *	set_engines(INVALID)
+ *	set_parallel(engine_index=0, width=2, num_siblings=2,
+ *		     engines=CS[0],CS[2],CS[1],CS[3])
+ *
+ *	Results in the following valid placements:
+ *	CS[0], CS[1]
+ *	CS[2], CS[3]
+ *
+ *	This can be thought of as two virtual engines, each containing two
+ *	engines thereby making a 2D array. However, there are bonds tying the
+ *	entries together and placing restrictions on how they can be scheduled.
+ *	Specifically, the scheduler can choose only vertical columns from the 2D
+ *	array. That is, CS[0] is bonded to CS[1] and CS[2] to CS[3]. So if the
+ *	scheduler wants to submit to CS[0], it must also choose CS[1] and vice
+ *	versa. Same for CS[2] requires also using CS[3].
+ *	VE[0] = CS[0], CS[2]
+ *	VE[1] = CS[1], CS[3]
+ *
+ *	Example 3 pseudo code:
+ *	set_engines(INVALID)
+ *	set_parallel(engine_index=0, width=2, num_siblings=2,
+ *		     engines=CS[0],CS[1],CS[1],CS[3])
+ *
+ *	Results in the following valid and invalid placements:
+ *	CS[0], CS[1]
+ *	CS[1], CS[3] - Not logically contiguous, return -EINVAL
+ */
+struct i915_context_engines_parallel_submit {
+	/**
+	 * @base: base user extension.
+	 */
+	struct i915_user_extension base;
+
+	/**
+	 * @engine_index: slot for parallel engine
+	 */
+	__u16 engine_index;
+
+	/**
+	 * @width: number of contexts per parallel engine or in other words the
+	 * number of batches in each submission
+	 */
+	__u16 width;
+
+	/**
+	 * @num_siblings: number of siblings per context or in other words the
+	 * number of possible placements for each submission
+	 */
+	__u16 num_siblings;
+
+	/**
+	 * @mbz16: reserved for future use; must be zero
+	 */
+	__u16 mbz16;
+
+	/**
+	 * @flags: all undefined flags must be zero, currently not defined flags
+	 */
+	__u64 flags;
+
+	/**
+	 * @mbz64: reserved for future use; must be zero
+	 */
+	__u64 mbz64[3];
+
+	/**
+	 * @engines: 2-d array of engine instances to configure parallel engine
+	 *
+	 * length = width (i) * num_siblings (j)
+	 * index = j + i * num_siblings
+	 */
+	struct i915_engine_class_instance engines[0];
+
+} __packed;
+
+#define I915_DEFINE_CONTEXT_ENGINES_PARALLEL_SUBMIT(name__, N__) struct { \
+	struct i915_user_extension base; \
+	__u16 engine_index; \
+	__u16 width; \
+	__u16 num_siblings; \
+	__u16 mbz16; \
+	__u64 flags; \
+	__u64 mbz64[3]; \
+	struct i915_engine_class_instance engines[N__]; \
+} __attribute__((packed)) name__
+
+/**
  * DOC: Context Engine Map uAPI
  *
  * Context engine map is a new way of addressing engines when submitting batch-
@@ -2108,6 +2293,7 @@ struct i915_context_param_engines {
 	__u64 extensions; /* linked chain of extension blocks, 0 terminates */
 #define I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE 0 /* see i915_context_engines_load_balance */
 #define I915_CONTEXT_ENGINES_EXT_BOND 1 /* see i915_context_engines_bond */
+#define I915_CONTEXT_ENGINES_EXT_PARALLEL_SUBMIT 2 /* see i915_context_engines_parallel_submit */
 	struct i915_engine_class_instance engines[0];
 } __attribute__((packed));
 
@@ -2726,14 +2912,20 @@ struct drm_i915_engine_info {
 
 	/** @flags: Engine flags. */
 	__u64 flags;
+#define I915_ENGINE_INFO_HAS_LOGICAL_INSTANCE		(1 << 0)
 
 	/** @capabilities: Capabilities of this engine. */
 	__u64 capabilities;
 #define I915_VIDEO_CLASS_CAPABILITY_HEVC		(1 << 0)
 #define I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC	(1 << 1)
 
+	/** @logical_instance: Logical instance of engine */
+	__u16 logical_instance;
+
 	/** @rsvd1: Reserved fields. */
-	__u64 rsvd1[4];
+	__u16 rsvd1[3];
+	/** @rsvd2: Reserved fields. */
+	__u64 rsvd2[3];
 };
 
 /**
@@ -2979,8 +3171,12 @@ struct drm_i915_gem_create_ext {
 	 *
 	 * For I915_GEM_CREATE_EXT_MEMORY_REGIONS usage see
 	 * struct drm_i915_gem_create_ext_memory_regions.
+	 *
+	 * For I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see
+	 * struct drm_i915_gem_create_ext_protected_content.
 	 */
 #define I915_GEM_CREATE_EXT_MEMORY_REGIONS 0
+#define I915_GEM_CREATE_EXT_PROTECTED_CONTENT 1
 	__u64 extensions;
 };
 
@@ -3038,6 +3234,50 @@ struct drm_i915_gem_create_ext_memory_regions {
 	__u64 regions;
 };
 
+/**
+ * struct drm_i915_gem_create_ext_protected_content - The
+ * I915_OBJECT_PARAM_PROTECTED_CONTENT extension.
+ *
+ * If this extension is provided, buffer contents are expected to be protected
+ * by PXP encryption and require decryption for scan out and processing. This
+ * is only possible on platforms that have PXP enabled, on all other scenarios
+ * using this extension will cause the ioctl to fail and return -ENODEV. The
+ * flags parameter is reserved for future expansion and must currently be set
+ * to zero.
+ *
+ * The buffer contents are considered invalid after a PXP session teardown.
+ *
+ * The encryption is guaranteed to be processed correctly only if the object
+ * is submitted with a context created using the
+ * I915_CONTEXT_PARAM_PROTECTED_CONTENT flag. This will also enable extra checks
+ * at submission time on the validity of the objects involved.
+ *
+ * Below is an example on how to create a protected object:
+ *
+ * .. code-block:: C
+ *
+ *      struct drm_i915_gem_create_ext_protected_content protected_ext = {
+ *              .base = { .name = I915_GEM_CREATE_EXT_PROTECTED_CONTENT },
+ *              .flags = 0,
+ *      };
+ *      struct drm_i915_gem_create_ext create_ext = {
+ *              .size = PAGE_SIZE,
+ *              .extensions = (uintptr_t)&protected_ext,
+ *      };
+ *
+ *      int err = ioctl(fd, DRM_IOCTL_I915_GEM_CREATE_EXT, &create_ext);
+ *      if (err) ...
+ */
+struct drm_i915_gem_create_ext_protected_content {
+	/** @base: Extension link. See struct i915_user_extension. */
+	struct i915_user_extension base;
+	/** @flags: reserved for future usage, currently MBZ */
+	__u32 flags;
+};
+
+/* ID of the protected content session managed by i915 when PXP is active */
+#define I915_PROTECTED_CONTENT_DEFAULT_SESSION 0xf
+
 #if defined(__cplusplus)
 }
 #endif
diff --git a/tools/include/uapi/linux/prctl.h b/tools/include/uapi/linux/prctl.h
index de45fcd2dcbe..bb73e9a0b24f 100644
--- a/tools/include/uapi/linux/prctl.h
+++ b/tools/include/uapi/linux/prctl.h
@@ -268,5 +268,8 @@ struct prctl_mm_map {
 # define PR_SCHED_CORE_SHARE_TO		2 /* push core_sched cookie to pid */
 # define PR_SCHED_CORE_SHARE_FROM	3 /* pull core_sched cookie to pid */
 # define PR_SCHED_CORE_MAX		4
+# define PR_SCHED_CORE_SCOPE_THREAD		0
+# define PR_SCHED_CORE_SCOPE_THREAD_GROUP	1
+# define PR_SCHED_CORE_SCOPE_PROCESS_GROUP	2
 
 #endif /* _LINUX_PRCTL_H */
diff --git a/tools/include/uapi/sound/asound.h b/tools/include/uapi/sound/asound.h
index 5859ca0a1439..5fbb79e30819 100644
--- a/tools/include/uapi/sound/asound.h
+++ b/tools/include/uapi/sound/asound.h
@@ -1002,7 +1002,7 @@ typedef int __bitwise snd_ctl_elem_iface_t;
 #define SNDRV_CTL_ELEM_ACCESS_WRITE		(1<<1)
 #define SNDRV_CTL_ELEM_ACCESS_READWRITE		(SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE)
 #define SNDRV_CTL_ELEM_ACCESS_VOLATILE		(1<<2)	/* control value may be changed without a notification */
-// (1 << 3) is unused.
+/* (1 << 3) is unused. */
 #define SNDRV_CTL_ELEM_ACCESS_TLV_READ		(1<<4)	/* TLV read is possible */
 #define SNDRV_CTL_ELEM_ACCESS_TLV_WRITE		(1<<5)	/* TLV write is possible */
 #define SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE	(SNDRV_CTL_ELEM_ACCESS_TLV_READ|SNDRV_CTL_ELEM_ACCESS_TLV_WRITE)
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 2d7df8703cf2..3cf7bac67239 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -469,7 +469,7 @@ This option sets the time out limit. The default value is 500 ms.
 
 --switch-events::
 Record context switch events i.e. events of type PERF_RECORD_SWITCH or
-PERF_RECORD_SWITCH_CPU_WIDE. In some cases (e.g. Intel PT or CoreSight)
+PERF_RECORD_SWITCH_CPU_WIDE. In some cases (e.g. Intel PT, CoreSight or Arm SPE)
 switch events will be enabled automatically, which can be suppressed by
 by the option --no-switch-events.
 
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 0777748b6da8..80522bcfafe0 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -516,17 +516,17 @@ kvm_ioctl_tbl := $(srctree)/tools/perf/trace/beauty/kvm_ioctl.sh
 $(kvm_ioctl_array): $(kvm_hdr_dir)/kvm.h $(kvm_ioctl_tbl)
 	$(Q)$(SHELL) '$(kvm_ioctl_tbl)' $(kvm_hdr_dir) > $@
 
-socket_ipproto_array := $(beauty_outdir)/socket_ipproto_array.c
-socket_ipproto_tbl := $(srctree)/tools/perf/trace/beauty/socket_ipproto.sh
+socket_arrays := $(beauty_outdir)/socket.c
+socket_tbl := $(srctree)/tools/perf/trace/beauty/socket.sh
 
-$(socket_ipproto_array): $(linux_uapi_dir)/in.h $(socket_ipproto_tbl)
-	$(Q)$(SHELL) '$(socket_ipproto_tbl)' $(linux_uapi_dir) > $@
+$(socket_arrays): $(linux_uapi_dir)/in.h $(beauty_linux_dir)/socket.h $(socket_tbl)
+	$(Q)$(SHELL) '$(socket_tbl)' $(linux_uapi_dir) $(beauty_linux_dir) > $@
 
-socket_arrays := $(beauty_outdir)/socket_arrays.c
-socket_tbl := $(srctree)/tools/perf/trace/beauty/socket.sh
+sockaddr_arrays := $(beauty_outdir)/sockaddr.c
+sockaddr_tbl := $(srctree)/tools/perf/trace/beauty/sockaddr.sh
 
-$(socket_arrays): $(beauty_linux_dir)/socket.h $(socket_tbl)
-	$(Q)$(SHELL) '$(socket_tbl)' $(beauty_linux_dir) > $@
+$(sockaddr_arrays): $(beauty_linux_dir)/socket.h $(sockaddr_tbl)
+	$(Q)$(SHELL) '$(sockaddr_tbl)' $(beauty_linux_dir) > $@
 
 vhost_virtio_ioctl_array := $(beauty_ioctl_outdir)/vhost_virtio_ioctl_array.c
 vhost_virtio_hdr_dir := $(srctree)/tools/include/uapi/linux
@@ -736,8 +736,8 @@ prepare: $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h archheaders $(drm_ioc
 	$(sndrv_ctl_ioctl_array) \
 	$(kcmp_type_array) \
 	$(kvm_ioctl_array) \
-	$(socket_ipproto_array) \
 	$(socket_arrays) \
+	$(sockaddr_arrays) \
 	$(vhost_virtio_ioctl_array) \
 	$(madvise_behavior_array) \
 	$(mmap_flags_array) \
@@ -1113,8 +1113,8 @@ clean:: $(LIBTRACEEVENT)-clean $(LIBAPI)-clean $(LIBBPF)-clean $(LIBSUBCMD)-clea
 		$(OUTPUT)$(sndrv_pcm_ioctl_array) \
 		$(OUTPUT)$(kvm_ioctl_array) \
 		$(OUTPUT)$(kcmp_type_array) \
-		$(OUTPUT)$(socket_ipproto_array) \
 		$(OUTPUT)$(socket_arrays) \
+		$(OUTPUT)$(sockaddr_arrays) \
 		$(OUTPUT)$(vhost_virtio_ioctl_array) \
 		$(OUTPUT)$(perf_ioctl_array) \
 		$(OUTPUT)$(prctl_option_array) \
diff --git a/tools/perf/arch/arm/include/arch-tests.h b/tools/perf/arch/arm/include/arch-tests.h
index c62538052404..452b3d904521 100644
--- a/tools/perf/arch/arm/include/arch-tests.h
+++ b/tools/perf/arch/arm/include/arch-tests.h
@@ -2,6 +2,6 @@
 #ifndef ARCH_TESTS_H
 #define ARCH_TESTS_H
 
-extern struct test arch_tests[];
+extern struct test_suite *arch_tests[];
 
 #endif
diff --git a/tools/perf/arch/arm/tests/arch-tests.c b/tools/perf/arch/arm/tests/arch-tests.c
index 6848101a855f..69561111cc6f 100644
--- a/tools/perf/arch/arm/tests/arch-tests.c
+++ b/tools/perf/arch/arm/tests/arch-tests.c
@@ -3,18 +3,10 @@
 #include "tests/tests.h"
 #include "arch-tests.h"
 
-struct test arch_tests[] = {
+struct test_suite *arch_tests[] = {
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
-	{
-		.desc = "DWARF unwind",
-		.func = test__dwarf_unwind,
-	},
+	&suite__dwarf_unwind,
 #endif
-	{
-		.desc = "Vectors page",
-		.func = test__vectors_page,
-	},
-	{
-		.func = NULL,
-	},
+	&suite__vectors_page,
+	NULL,
 };
diff --git a/tools/perf/arch/arm/tests/vectors-page.c b/tools/perf/arch/arm/tests/vectors-page.c
index 7ffdd79971c8..55a835837466 100644
--- a/tools/perf/arch/arm/tests/vectors-page.c
+++ b/tools/perf/arch/arm/tests/vectors-page.c
@@ -9,8 +9,7 @@
 
 #define VECTORS__MAP_NAME "[vectors]"
 
-int test__vectors_page(struct test *test __maybe_unused,
-		       int subtest __maybe_unused)
+static int test__vectors_page(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	void *start, *end;
 
@@ -22,3 +21,5 @@ int test__vectors_page(struct test *test __maybe_unused,
 
 	return TEST_OK;
 }
+
+DEFINE_SUITE("Vectors page", vectors_page);
diff --git a/tools/perf/arch/arm64/include/arch-tests.h b/tools/perf/arch/arm64/include/arch-tests.h
index c62538052404..452b3d904521 100644
--- a/tools/perf/arch/arm64/include/arch-tests.h
+++ b/tools/perf/arch/arm64/include/arch-tests.h
@@ -2,6 +2,6 @@
 #ifndef ARCH_TESTS_H
 #define ARCH_TESTS_H
 
-extern struct test arch_tests[];
+extern struct test_suite *arch_tests[];
 
 #endif
diff --git a/tools/perf/arch/arm64/tests/arch-tests.c b/tools/perf/arch/arm64/tests/arch-tests.c
index 5b1543c98022..ad16b4f8f63e 100644
--- a/tools/perf/arch/arm64/tests/arch-tests.c
+++ b/tools/perf/arch/arm64/tests/arch-tests.c
@@ -3,14 +3,9 @@
 #include "tests/tests.h"
 #include "arch-tests.h"
 
-struct test arch_tests[] = {
+struct test_suite *arch_tests[] = {
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
-	{
-		.desc = "DWARF unwind",
-		.func = test__dwarf_unwind,
-	},
+	&suite__dwarf_unwind,
 #endif
-	{
-		.func = NULL,
-	},
+	NULL,
 };
diff --git a/tools/perf/arch/arm64/util/arm-spe.c b/tools/perf/arch/arm64/util/arm-spe.c
index a4420d4df503..2100d46ccf5e 100644
--- a/tools/perf/arch/arm64/util/arm-spe.c
+++ b/tools/perf/arch/arm64/util/arm-spe.c
@@ -23,6 +23,7 @@
 #include "../../../util/auxtrace.h"
 #include "../../../util/record.h"
 #include "../../../util/arm-spe.h"
+#include <tools/libc_compat.h> // reallocarray
 
 #define KiB(x) ((x) * 1024)
 #define MiB(x) ((x) * 1024 * 1024)
@@ -31,6 +32,8 @@ struct arm_spe_recording {
 	struct auxtrace_record		itr;
 	struct perf_pmu			*arm_spe_pmu;
 	struct evlist		*evlist;
+	int			wrapped_cnt;
+	bool			*wrapped;
 };
 
 static void arm_spe_set_timestamp(struct auxtrace_record *itr,
@@ -84,6 +87,55 @@ static int arm_spe_info_fill(struct auxtrace_record *itr,
 	return 0;
 }
 
+static void
+arm_spe_snapshot_resolve_auxtrace_defaults(struct record_opts *opts,
+					   bool privileged)
+{
+	/*
+	 * The default snapshot size is the auxtrace mmap size. If neither auxtrace mmap size nor
+	 * snapshot size is specified, then the default is 4MiB for privileged users, 128KiB for
+	 * unprivileged users.
+	 *
+	 * The default auxtrace mmap size is 4MiB/page_size for privileged users, 128KiB for
+	 * unprivileged users. If an unprivileged user does not specify mmap pages, the mmap pages
+	 * will be reduced from the default 512KiB/page_size to 256KiB/page_size, otherwise the
+	 * user is likely to get an error as they exceed their mlock limmit.
+	 */
+
+	/*
+	 * No size were given to '-S' or '-m,', so go with the default
+	 */
+	if (!opts->auxtrace_snapshot_size && !opts->auxtrace_mmap_pages) {
+		if (privileged) {
+			opts->auxtrace_mmap_pages = MiB(4) / page_size;
+		} else {
+			opts->auxtrace_mmap_pages = KiB(128) / page_size;
+			if (opts->mmap_pages == UINT_MAX)
+				opts->mmap_pages = KiB(256) / page_size;
+		}
+	} else if (!opts->auxtrace_mmap_pages && !privileged && opts->mmap_pages == UINT_MAX) {
+		opts->mmap_pages = KiB(256) / page_size;
+	}
+
+	/*
+	 * '-m,xyz' was specified but no snapshot size, so make the snapshot size as big as the
+	 * auxtrace mmap area.
+	 */
+	if (!opts->auxtrace_snapshot_size)
+		opts->auxtrace_snapshot_size = opts->auxtrace_mmap_pages * (size_t)page_size;
+
+	/*
+	 * '-Sxyz' was specified but no auxtrace mmap area, so make the auxtrace mmap area big
+	 * enough to fit the requested snapshot size.
+	 */
+	if (!opts->auxtrace_mmap_pages) {
+		size_t sz = opts->auxtrace_snapshot_size;
+
+		sz = round_up(sz, page_size) / page_size;
+		opts->auxtrace_mmap_pages = roundup_pow_of_two(sz);
+	}
+}
+
 static int arm_spe_recording_options(struct auxtrace_record *itr,
 				     struct evlist *evlist,
 				     struct record_opts *opts)
@@ -115,6 +167,36 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,
 	if (!opts->full_auxtrace)
 		return 0;
 
+	/*
+	 * we are in snapshot mode.
+	 */
+	if (opts->auxtrace_snapshot_mode) {
+		/*
+		 * Command arguments '-Sxyz' and/or '-m,xyz' are missing, so fill those in with
+		 * default values.
+		 */
+		if (!opts->auxtrace_snapshot_size || !opts->auxtrace_mmap_pages)
+			arm_spe_snapshot_resolve_auxtrace_defaults(opts, privileged);
+
+		/*
+		 * Snapshot size can't be bigger than the auxtrace area.
+		 */
+		if (opts->auxtrace_snapshot_size > opts->auxtrace_mmap_pages * (size_t)page_size) {
+			pr_err("Snapshot size %zu must not be greater than AUX area tracing mmap size %zu\n",
+			       opts->auxtrace_snapshot_size,
+			       opts->auxtrace_mmap_pages * (size_t)page_size);
+			return -EINVAL;
+		}
+
+		/*
+		 * Something went wrong somewhere - this shouldn't happen.
+		 */
+		if (!opts->auxtrace_snapshot_size || !opts->auxtrace_mmap_pages) {
+			pr_err("Failed to calculate default snapshot size and/or AUX area tracing mmap pages\n");
+			return -EINVAL;
+		}
+	}
+
 	/* We are in full trace mode but '-m,xyz' wasn't specified */
 	if (!opts->auxtrace_mmap_pages) {
 		if (privileged) {
@@ -138,6 +220,9 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,
 		}
 	}
 
+	if (opts->auxtrace_snapshot_mode)
+		pr_debug2("%sx snapshot size: %zu\n", ARM_SPE_PMU_NAME,
+			  opts->auxtrace_snapshot_size);
 
 	/*
 	 * To obtain the auxtrace buffer file descriptor, the auxtrace event
@@ -166,8 +251,199 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,
 	tracking_evsel->core.attr.sample_period = 1;
 
 	/* In per-cpu case, always need the time of mmap events etc */
-	if (!perf_cpu_map__empty(cpus))
+	if (!perf_cpu_map__empty(cpus)) {
 		evsel__set_sample_bit(tracking_evsel, TIME);
+		evsel__set_sample_bit(tracking_evsel, CPU);
+
+		/* also track task context switch */
+		if (!record_opts__no_switch_events(opts))
+			tracking_evsel->core.attr.context_switch = 1;
+	}
+
+	return 0;
+}
+
+static int arm_spe_parse_snapshot_options(struct auxtrace_record *itr __maybe_unused,
+					 struct record_opts *opts,
+					 const char *str)
+{
+	unsigned long long snapshot_size = 0;
+	char *endptr;
+
+	if (str) {
+		snapshot_size = strtoull(str, &endptr, 0);
+		if (*endptr || snapshot_size > SIZE_MAX)
+			return -1;
+	}
+
+	opts->auxtrace_snapshot_mode = true;
+	opts->auxtrace_snapshot_size = snapshot_size;
+
+	return 0;
+}
+
+static int arm_spe_snapshot_start(struct auxtrace_record *itr)
+{
+	struct arm_spe_recording *ptr =
+			container_of(itr, struct arm_spe_recording, itr);
+	struct evsel *evsel;
+
+	evlist__for_each_entry(ptr->evlist, evsel) {
+		if (evsel->core.attr.type == ptr->arm_spe_pmu->type)
+			return evsel__disable(evsel);
+	}
+	return -EINVAL;
+}
+
+static int arm_spe_snapshot_finish(struct auxtrace_record *itr)
+{
+	struct arm_spe_recording *ptr =
+			container_of(itr, struct arm_spe_recording, itr);
+	struct evsel *evsel;
+
+	evlist__for_each_entry(ptr->evlist, evsel) {
+		if (evsel->core.attr.type == ptr->arm_spe_pmu->type)
+			return evsel__enable(evsel);
+	}
+	return -EINVAL;
+}
+
+static int arm_spe_alloc_wrapped_array(struct arm_spe_recording *ptr, int idx)
+{
+	bool *wrapped;
+	int cnt = ptr->wrapped_cnt, new_cnt, i;
+
+	/*
+	 * No need to allocate, so return early.
+	 */
+	if (idx < cnt)
+		return 0;
+
+	/*
+	 * Make ptr->wrapped as big as idx.
+	 */
+	new_cnt = idx + 1;
+
+	/*
+	 * Free'ed in arm_spe_recording_free().
+	 */
+	wrapped = reallocarray(ptr->wrapped, new_cnt, sizeof(bool));
+	if (!wrapped)
+		return -ENOMEM;
+
+	/*
+	 * init new allocated values.
+	 */
+	for (i = cnt; i < new_cnt; i++)
+		wrapped[i] = false;
+
+	ptr->wrapped_cnt = new_cnt;
+	ptr->wrapped = wrapped;
+
+	return 0;
+}
+
+static bool arm_spe_buffer_has_wrapped(unsigned char *buffer,
+				      size_t buffer_size, u64 head)
+{
+	u64 i, watermark;
+	u64 *buf = (u64 *)buffer;
+	size_t buf_size = buffer_size;
+
+	/*
+	 * Defensively handle the case where head might be continually increasing - if its value is
+	 * equal or greater than the size of the ring buffer, then we can safely determine it has
+	 * wrapped around. Otherwise, continue to detect if head might have wrapped.
+	 */
+	if (head >= buffer_size)
+		return true;
+
+	/*
+	 * We want to look the very last 512 byte (chosen arbitrarily) in the ring buffer.
+	 */
+	watermark = buf_size - 512;
+
+	/*
+	 * The value of head is somewhere within the size of the ring buffer. This can be that there
+	 * hasn't been enough data to fill the ring buffer yet or the trace time was so long that
+	 * head has numerically wrapped around.  To find we need to check if we have data at the
+	 * very end of the ring buffer.  We can reliably do this because mmap'ed pages are zeroed
+	 * out and there is a fresh mapping with every new session.
+	 */
+
+	/*
+	 * head is less than 512 byte from the end of the ring buffer.
+	 */
+	if (head > watermark)
+		watermark = head;
+
+	/*
+	 * Speed things up by using 64 bit transactions (see "u64 *buf" above)
+	 */
+	watermark /= sizeof(u64);
+	buf_size /= sizeof(u64);
+
+	/*
+	 * If we find trace data at the end of the ring buffer, head has been there and has
+	 * numerically wrapped around at least once.
+	 */
+	for (i = watermark; i < buf_size; i++)
+		if (buf[i])
+			return true;
+
+	return false;
+}
+
+static int arm_spe_find_snapshot(struct auxtrace_record *itr, int idx,
+				  struct auxtrace_mmap *mm, unsigned char *data,
+				  u64 *head, u64 *old)
+{
+	int err;
+	bool wrapped;
+	struct arm_spe_recording *ptr =
+			container_of(itr, struct arm_spe_recording, itr);
+
+	/*
+	 * Allocate memory to keep track of wrapping if this is the first
+	 * time we deal with this *mm.
+	 */
+	if (idx >= ptr->wrapped_cnt) {
+		err = arm_spe_alloc_wrapped_array(ptr, idx);
+		if (err)
+			return err;
+	}
+
+	/*
+	 * Check to see if *head has wrapped around.  If it hasn't only the
+	 * amount of data between *head and *old is snapshot'ed to avoid
+	 * bloating the perf.data file with zeros.  But as soon as *head has
+	 * wrapped around the entire size of the AUX ring buffer it taken.
+	 */
+	wrapped = ptr->wrapped[idx];
+	if (!wrapped && arm_spe_buffer_has_wrapped(data, mm->len, *head)) {
+		wrapped = true;
+		ptr->wrapped[idx] = true;
+	}
+
+	pr_debug3("%s: mmap index %d old head %zu new head %zu size %zu\n",
+		  __func__, idx, (size_t)*old, (size_t)*head, mm->len);
+
+	/*
+	 * No wrap has occurred, we can just use *head and *old.
+	 */
+	if (!wrapped)
+		return 0;
+
+	/*
+	 * *head has wrapped around - adjust *head and *old to pickup the
+	 * entire content of the AUX buffer.
+	 */
+	if (*head >= mm->len) {
+		*old = *head - mm->len;
+	} else {
+		*head += mm->len;
+		*old = *head - mm->len;
+	}
 
 	return 0;
 }
@@ -186,6 +462,7 @@ static void arm_spe_recording_free(struct auxtrace_record *itr)
 	struct arm_spe_recording *sper =
 			container_of(itr, struct arm_spe_recording, itr);
 
+	free(sper->wrapped);
 	free(sper);
 }
 
@@ -207,6 +484,10 @@ struct auxtrace_record *arm_spe_recording_init(int *err,
 
 	sper->arm_spe_pmu = arm_spe_pmu;
 	sper->itr.pmu = arm_spe_pmu;
+	sper->itr.snapshot_start = arm_spe_snapshot_start;
+	sper->itr.snapshot_finish = arm_spe_snapshot_finish;
+	sper->itr.find_snapshot = arm_spe_find_snapshot;
+	sper->itr.parse_snapshot_options = arm_spe_parse_snapshot_options;
 	sper->itr.recording_options = arm_spe_recording_options;
 	sper->itr.info_priv_size = arm_spe_info_priv_size;
 	sper->itr.info_fill = arm_spe_info_fill;
diff --git a/tools/perf/arch/powerpc/include/arch-tests.h b/tools/perf/arch/powerpc/include/arch-tests.h
index c62538052404..452b3d904521 100644
--- a/tools/perf/arch/powerpc/include/arch-tests.h
+++ b/tools/perf/arch/powerpc/include/arch-tests.h
@@ -2,6 +2,6 @@
 #ifndef ARCH_TESTS_H
 #define ARCH_TESTS_H
 
-extern struct test arch_tests[];
+extern struct test_suite *arch_tests[];
 
 #endif
diff --git a/tools/perf/arch/powerpc/tests/arch-tests.c b/tools/perf/arch/powerpc/tests/arch-tests.c
index 8c3fbd4af817..eb98c57b5aeb 100644
--- a/tools/perf/arch/powerpc/tests/arch-tests.c
+++ b/tools/perf/arch/powerpc/tests/arch-tests.c
@@ -3,14 +3,10 @@
 #include "tests/tests.h"
 #include "arch-tests.h"
 
-struct test arch_tests[] = {
+
+struct test_suite *arch_tests[] = {
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
-	{
-		.desc = "Test dwarf unwind",
-		.func = test__dwarf_unwind,
-	},
+	&suite__dwarf_unwind,
 #endif
-	{
-		.func = NULL,
-	},
+	NULL,
 };
diff --git a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
index 18b5500ea8bf..fe8f8dd157b4 100644
--- a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
+++ b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
@@ -370,6 +370,7 @@
 446	common	landlock_restrict_self	sys_landlock_restrict_self
 447	common	memfd_secret		sys_memfd_secret
 448	common	process_mrelease	sys_process_mrelease
+449	common	futex_waitv		sys_futex_waitv
 
 #
 # Due to a historical design error, certain syscalls are numbered differently
diff --git a/tools/perf/arch/x86/include/arch-tests.h b/tools/perf/arch/x86/include/arch-tests.h
index 9599e7a3f1af..6a1a1b3c0827 100644
--- a/tools/perf/arch/x86/include/arch-tests.h
+++ b/tools/perf/arch/x86/include/arch-tests.h
@@ -2,15 +2,15 @@
 #ifndef ARCH_TESTS_H
 #define ARCH_TESTS_H
 
-struct test;
+struct test_suite;
 
 /* Tests */
-int test__rdpmc(struct test *test, int subtest);
-int test__insn_x86(struct test *test, int subtest);
-int test__intel_pt_pkt_decoder(struct test *test, int subtest);
-int test__bp_modify(struct test *test, int subtest);
-int test__x86_sample_parsing(struct test *test, int subtest);
+int test__rdpmc(struct test_suite *test, int subtest);
+int test__insn_x86(struct test_suite *test, int subtest);
+int test__intel_pt_pkt_decoder(struct test_suite *test, int subtest);
+int test__bp_modify(struct test_suite *test, int subtest);
+int test__x86_sample_parsing(struct test_suite *test, int subtest);
 
-extern struct test arch_tests[];
+extern struct test_suite *arch_tests[];
 
 #endif
diff --git a/tools/perf/arch/x86/tests/arch-tests.c b/tools/perf/arch/x86/tests/arch-tests.c
index 71aa67367ad6..64fb73d14d2f 100644
--- a/tools/perf/arch/x86/tests/arch-tests.c
+++ b/tools/perf/arch/x86/tests/arch-tests.c
@@ -3,39 +3,28 @@
 #include "tests/tests.h"
 #include "arch-tests.h"
 
-struct test arch_tests[] = {
-	{
-		.desc = "x86 rdpmc",
-		.func = test__rdpmc,
-	},
+DEFINE_SUITE("x86 rdpmc", rdpmc);
+#ifdef HAVE_AUXTRACE_SUPPORT
+DEFINE_SUITE("x86 instruction decoder - new instructions", insn_x86);
+DEFINE_SUITE("Intel PT packet decoder", intel_pt_pkt_decoder);
+#endif
+#if defined(__x86_64__)
+DEFINE_SUITE("x86 bp modify", bp_modify);
+#endif
+DEFINE_SUITE("x86 Sample parsing", x86_sample_parsing);
+
+struct test_suite *arch_tests[] = {
+	&suite__rdpmc,
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
-	{
-		.desc = "DWARF unwind",
-		.func = test__dwarf_unwind,
-	},
+	&suite__dwarf_unwind,
 #endif
 #ifdef HAVE_AUXTRACE_SUPPORT
-	{
-		.desc = "x86 instruction decoder - new instructions",
-		.func = test__insn_x86,
-	},
-	{
-		.desc = "Intel PT packet decoder",
-		.func = test__intel_pt_pkt_decoder,
-	},
+	&suite__insn_x86,
+	&suite__intel_pt_pkt_decoder,
 #endif
 #if defined(__x86_64__)
-	{
-		.desc = "x86 bp modify",
-		.func = test__bp_modify,
-	},
+	&suite__bp_modify,
 #endif
-	{
-		.desc = "x86 Sample parsing",
-		.func = test__x86_sample_parsing,
-	},
-	{
-		.func = NULL,
-	},
-
+	&suite__x86_sample_parsing,
+	NULL,
 };
diff --git a/tools/perf/arch/x86/tests/bp-modify.c b/tools/perf/arch/x86/tests/bp-modify.c
index dffcf9b52153..0924ccd9e36d 100644
--- a/tools/perf/arch/x86/tests/bp-modify.c
+++ b/tools/perf/arch/x86/tests/bp-modify.c
@@ -204,7 +204,7 @@ out:
 	return rip == (unsigned long) bp_1 ? TEST_OK : TEST_FAIL;
 }
 
-int test__bp_modify(struct test *test __maybe_unused,
+int test__bp_modify(struct test_suite *test __maybe_unused,
 		    int subtest __maybe_unused)
 {
 	TEST_ASSERT_VAL("modify test 1 failed\n", !bp_modify1());
diff --git a/tools/perf/arch/x86/tests/insn-x86.c b/tools/perf/arch/x86/tests/insn-x86.c
index 0262b0d8ccf5..94b490c434d0 100644
--- a/tools/perf/arch/x86/tests/insn-x86.c
+++ b/tools/perf/arch/x86/tests/insn-x86.c
@@ -173,7 +173,7 @@ static int test_data_set(struct test_data *dat_set, int x86_64)
  * verbose (-v) option to see all the instructions and whether or not they
  * decoded successfully.
  */
-int test__insn_x86(struct test *test __maybe_unused, int subtest __maybe_unused)
+int test__insn_x86(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret = 0;
 
diff --git a/tools/perf/arch/x86/tests/intel-cqm.c b/tools/perf/arch/x86/tests/intel-cqm.c
index 27dd8cf9e060..cb5b2c6c3b3b 100644
--- a/tools/perf/arch/x86/tests/intel-cqm.c
+++ b/tools/perf/arch/x86/tests/intel-cqm.c
@@ -37,7 +37,7 @@ static pid_t spawn(void)
  * the last read counter value to avoid triggering a WARN_ON_ONCE() in
  * smp_call_function_many() caused by sending IPIs from NMI context.
  */
-int test__intel_cqm_count_nmi_context(struct test *test __maybe_unused, int subtest __maybe_unused)
+int test__intel_cqm_count_nmi_context(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct evlist *evlist = NULL;
 	struct evsel *evsel = NULL;
diff --git a/tools/perf/arch/x86/tests/intel-pt-pkt-decoder-test.c b/tools/perf/arch/x86/tests/intel-pt-pkt-decoder-test.c
index c933e3dcd0a8..2fc882ab24c1 100644
--- a/tools/perf/arch/x86/tests/intel-pt-pkt-decoder-test.c
+++ b/tools/perf/arch/x86/tests/intel-pt-pkt-decoder-test.c
@@ -289,7 +289,7 @@ static int test_one(struct test_data *d)
  * This test feeds byte sequences to the Intel PT packet decoder and checks the
  * results. Changes to the packet context are also checked.
  */
-int test__intel_pt_pkt_decoder(struct test *test __maybe_unused, int subtest __maybe_unused)
+int test__intel_pt_pkt_decoder(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct test_data *d = data;
 	int ret;
diff --git a/tools/perf/arch/x86/tests/rdpmc.c b/tools/perf/arch/x86/tests/rdpmc.c
index 1ea916656a2d..498413ad9c97 100644
--- a/tools/perf/arch/x86/tests/rdpmc.c
+++ b/tools/perf/arch/x86/tests/rdpmc.c
@@ -157,7 +157,7 @@ out_close:
 	return 0;
 }
 
-int test__rdpmc(struct test *test __maybe_unused, int subtest __maybe_unused)
+int test__rdpmc(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int status = 0;
 	int wret = 0;
diff --git a/tools/perf/arch/x86/tests/sample-parsing.c b/tools/perf/arch/x86/tests/sample-parsing.c
index c92db87e4479..bfbd3662b69e 100644
--- a/tools/perf/arch/x86/tests/sample-parsing.c
+++ b/tools/perf/arch/x86/tests/sample-parsing.c
@@ -115,7 +115,7 @@ out_free:
  * For now, the PERF_SAMPLE_WEIGHT_STRUCT is the only X86 specific sample type.
  * The test only checks the PERF_SAMPLE_WEIGHT_STRUCT type.
  */
-int test__x86_sample_parsing(struct test *test __maybe_unused, int subtest __maybe_unused)
+int test__x86_sample_parsing(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	return do_test(PERF_SAMPLE_WEIGHT_STRUCT);
 }
diff --git a/tools/perf/bench/futex-lock-pi.c b/tools/perf/bench/futex-lock-pi.c
index 5d1fe9c35807..137890f78e17 100644
--- a/tools/perf/bench/futex-lock-pi.c
+++ b/tools/perf/bench/futex-lock-pi.c
@@ -233,6 +233,7 @@ int bench_futex_lock_pi(int argc, const char **argv)
 	print_summary();
 
 	free(worker);
+	perf_cpu_map__put(cpu);
 	return ret;
 err:
 	usage_with_options(bench_futex_lock_pi_usage, options);
diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c
index 97fe31fd3a23..f7a5ffebb940 100644
--- a/tools/perf/bench/futex-requeue.c
+++ b/tools/perf/bench/futex-requeue.c
@@ -294,6 +294,7 @@ int bench_futex_requeue(int argc, const char **argv)
 	print_summary();
 
 	free(worker);
+	perf_cpu_map__put(cpu);
 	return ret;
 err:
 	usage_with_options(bench_futex_requeue_usage, options);
diff --git a/tools/perf/bench/futex-wake-parallel.c b/tools/perf/bench/futex-wake-parallel.c
index e970e6b9ad53..0983f40b4b40 100644
--- a/tools/perf/bench/futex-wake-parallel.c
+++ b/tools/perf/bench/futex-wake-parallel.c
@@ -329,6 +329,7 @@ int bench_futex_wake_parallel(int argc, const char **argv)
 	print_summary();
 
 	free(blocked_worker);
+	perf_cpu_map__put(cpu);
 	return ret;
 }
 #endif /* HAVE_PTHREAD_BARRIER */
diff --git a/tools/perf/bench/futex-wake.c b/tools/perf/bench/futex-wake.c
index 77f058a47790..2226a475e782 100644
--- a/tools/perf/bench/futex-wake.c
+++ b/tools/perf/bench/futex-wake.c
@@ -222,5 +222,6 @@ int bench_futex_wake(int argc, const char **argv)
 	print_summary();
 
 	free(worker);
+	perf_cpu_map__put(cpu);
 	return ret;
 }
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 624ea12ce5ca..0b52e08e558e 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -979,6 +979,8 @@ static struct syscall_fmt syscall_fmts[] = {
 	  .arg = { [2] = { .scnprintf = SCA_GETRANDOM_FLAGS, /* flags */ }, }, },
 	{ .name	    = "getrlimit",
 	  .arg = { [0] = STRARRAY(resource, rlimit_resources), }, },
+	{ .name	    = "getsockopt",
+	  .arg = { [1] = STRARRAY(level, socket_level), }, },
 	{ .name	    = "gettid",	    .errpid = true, },
 	{ .name	    = "ioctl",
 	  .arg = {
@@ -1121,6 +1123,8 @@ static struct syscall_fmt syscall_fmts[] = {
 	  .arg = { [0] = STRARRAY(which, itimers), }, },
 	{ .name	    = "setrlimit",
 	  .arg = { [0] = STRARRAY(resource, rlimit_resources), }, },
+	{ .name	    = "setsockopt",
+	  .arg = { [1] = STRARRAY(level, socket_level), }, },
 	{ .name	    = "socket",
 	  .arg = { [0] = STRARRAY(family, socket_families),
 		   [1] = { .scnprintf = SCA_SK_TYPE, /* type */ },
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
index a42fab308ff6..aa8cfeabb743 100644
--- a/tools/perf/design.txt
+++ b/tools/perf/design.txt
@@ -106,6 +106,9 @@ enum perf_hw_id {
 	PERF_COUNT_HW_BRANCH_INSTRUCTIONS	= 4,
 	PERF_COUNT_HW_BRANCH_MISSES		= 5,
 	PERF_COUNT_HW_BUS_CYCLES		= 6,
+	PERF_COUNT_HW_STALLED_CYCLES_FRONTEND	= 7,
+	PERF_COUNT_HW_STALLED_CYCLES_BACKEND	= 8,
+	PERF_COUNT_HW_REF_CPU_CYCLES		= 9,
 };
 
 These are standardized types of events that work relatively uniformly
diff --git a/tools/perf/pmu-events/arch/powerpc/power10/metrics.json b/tools/perf/pmu-events/arch/powerpc/power10/metrics.json
new file mode 100644
index 000000000000..b57526fa44f2
--- /dev/null
+++ b/tools/perf/pmu-events/arch/powerpc/power10/metrics.json
@@ -0,0 +1,676 @@
+[
+    {
+        "BriefDescription": "Percentage of cycles that are run cycles",
+        "MetricExpr": "PM_RUN_CYC / PM_CYC * 100",
+        "MetricGroup": "General",
+        "MetricName": "RUN_CYCLES_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction",
+        "MetricExpr": "PM_CYC / PM_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "CYCLES_PER_INSTRUCTION"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled for any reason",
+        "MetricExpr": "PM_DISP_STALL_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled because there was a flush",
+        "MetricExpr": "PM_DISP_STALL_FLUSH / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_FLUSH_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled because the MMU was handling a translation miss",
+        "MetricExpr": "PM_DISP_STALL_TRANSLATION / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_TRANSLATION_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled waiting to resolve an instruction ERAT miss",
+        "MetricExpr": "PM_DISP_STALL_IERAT_ONLY_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_IERAT_ONLY_MISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled waiting to resolve an instruction TLB miss",
+        "MetricExpr": "PM_DISP_STALL_ITLB_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_ITLB_MISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled due to an icache miss",
+        "MetricExpr": "PM_DISP_STALL_IC_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_IC_MISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while the instruction was fetched from the local L2",
+        "MetricExpr": "PM_DISP_STALL_IC_L2 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_IC_L2_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while the instruction was fetched from the local L3",
+        "MetricExpr": "PM_DISP_STALL_IC_L3 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_IC_L3_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while the instruction was fetched from any source beyond the local L3",
+        "MetricExpr": "PM_DISP_STALL_IC_L3MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_IC_L3MISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled due to an icache miss after a branch mispredict",
+        "MetricExpr": "PM_DISP_STALL_BR_MPRED_ICMISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_BR_MPRED_ICMISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while instruction was fetched from the local L2 after suffering a branch mispredict",
+        "MetricExpr": "PM_DISP_STALL_BR_MPRED_IC_L2 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_BR_MPRED_IC_L2_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while instruction was fetched from the local L3 after suffering a branch mispredict",
+        "MetricExpr": "PM_DISP_STALL_BR_MPRED_IC_L3 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_BR_MPRED_IC_L3_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled while instruction was fetched from any source beyond the local L3 after suffering a branch mispredict",
+        "MetricExpr": "PM_DISP_STALL_BR_MPRED_IC_L3MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_BR_MPRED_IC_L3MISS_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled due to a branch mispredict",
+        "MetricExpr": "PM_DISP_STALL_BR_MPRED / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_BR_MPRED_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch for any reason",
+        "MetricExpr": "PM_DISP_STALL_HELD_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch because of a synchronizing instruction that requires the ICT to be empty before dispatch",
+        "MetricExpr": "PM_DISP_STALL_HELD_SYNC_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISP_HELD_STALL_SYNC_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch while waiting on the scoreboard",
+        "MetricExpr": "PM_DISP_STALL_HELD_SCOREBOARD_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISP_HELD_STALL_SCOREBOARD_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch due to issue queue full",
+        "MetricExpr": "PM_DISP_STALL_HELD_ISSQ_FULL_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISP_HELD_STALL_ISSQ_FULL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch because the mapper/SRB was full",
+        "MetricExpr": "PM_DISP_STALL_HELD_RENAME_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_RENAME_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch because the STF mapper/SRB was full",
+        "MetricExpr": "PM_DISP_STALL_HELD_STF_MAPPER_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_STF_MAPPER_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch because the XVFC mapper/SRB was full",
+        "MetricExpr": "PM_DISP_STALL_HELD_XVFC_MAPPER_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_XVFC_MAPPER_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch for any other reason",
+        "MetricExpr": "PM_DISP_STALL_HELD_OTHER_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_OTHER_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction has been dispatched but not issued for any reason",
+        "MetricExpr": "PM_ISSUE_STALL / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "ISSUE_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting to be finished in one of the execution units",
+        "MetricExpr": "PM_EXEC_STALL / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "EXECUTION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction spent executing an NTC instruction that gets flushed some time after dispatch",
+        "MetricExpr": "PM_EXEC_STALL_NTC_FLUSH / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "NTC_FLUSH_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTF instruction finishes at dispatch",
+        "MetricExpr": "PM_EXEC_STALL_FIN_AT_DISP / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "FIN_AT_DISP_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is executing in the branch unit",
+        "MetricExpr": "PM_EXEC_STALL_BRU / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "BRU_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a simple fixed point instruction that is executing in the LSU",
+        "MetricExpr": "PM_EXEC_STALL_SIMPLE_FX / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "SIMPLE_FX_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is executing in the VSU",
+        "MetricExpr": "PM_EXEC_STALL_VSU / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "VSU_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting to be finished in one of the execution units",
+        "MetricExpr": "PM_EXEC_STALL_TRANSLATION / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "TRANSLATION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a load or store that suffered a translation miss",
+        "MetricExpr": "PM_EXEC_STALL_DERAT_ONLY_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DERAT_ONLY_MISS_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is recovering from a TLB miss",
+        "MetricExpr": "PM_EXEC_STALL_DERAT_DTLB_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DERAT_DTLB_MISS_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is executing in the LSU",
+        "MetricExpr": "PM_EXEC_STALL_LSU / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "LSU_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a load that is executing in the LSU",
+        "MetricExpr": "PM_EXEC_STALL_LOAD / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "LOAD_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from either the local L2 or local L3",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_L2L3 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_L2L3_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from either the local L2 or local L3, with an RC dispatch conflict",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_L2L3_CONFLICT / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_L2L3_CONFLICT_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from either the local L2 or local L3, without an RC dispatch conflict",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_L2L3_NOCONFLICT / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_L2L3_NOCONFLICT_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from a source beyond the local L2 and local L3",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_L3MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_L3MISS_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from a neighbor chiplet's L2 or L3 in the same chip",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_L21_L31 / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_L21_L31_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from L4, local memory or OpenCAPI chip",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_LMEM / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_LMEM_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from a remote chip (cache, L4, memory or OpenCAPI) in the same group",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_OFF_CHIP / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_OFF_CHIP_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is waiting for a load miss to resolve from a distant chip (cache, L4, memory or OpenCAPI chip)",
+        "MetricExpr": "PM_EXEC_STALL_DMISS_OFF_NODE / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DMISS_OFF_NODE_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is executing a TLBIEL instruction",
+        "MetricExpr": "PM_EXEC_STALL_TLBIEL / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "TLBIEL_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is finishing a load after its data has been reloaded from a data source beyond the local L1, OR when the LSU is processing an L1-hit, OR when the NTF instruction merged with another load in the LMQ",
+        "MetricExpr": "PM_EXEC_STALL_LOAD_FINISH / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "LOAD_FINISH_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a store that is executing in the LSU",
+        "MetricExpr": "PM_EXEC_STALL_STORE / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "STORE_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is in the store unit outside of handling store misses or other special store operations",
+        "MetricExpr": "PM_EXEC_STALL_STORE_PIPE / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "STORE_PIPE_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a store whose cache line was not resident in the L1 and had to wait for allocation of the missing line into the L1",
+        "MetricExpr": "PM_EXEC_STALL_STORE_MISS / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "STORE_MISS_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a TLBIE instruction waiting for a response from the L2",
+        "MetricExpr": "PM_EXEC_STALL_TLBIE / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "TLBIE_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is executing a PTESYNC instruction",
+        "MetricExpr": "PM_EXEC_STALL_PTESYNC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "PTESYNC_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction cannot complete because the thread was blocked",
+        "MetricExpr": "PM_CMPL_STALL / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction cannot complete because it was interrupted by ANY exception",
+        "MetricExpr": "PM_CMPL_STALL_EXCEPTION / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "EXCEPTION_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is stuck at finish waiting for the non-speculative finish of either a STCX instruction waiting for its result or a load waiting for non-critical sectors of data and ECC",
+        "MetricExpr": "PM_CMPL_STALL_MEM_ECC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "MEM_ECC_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a STCX instruction waiting for resolution from the nest",
+        "MetricExpr": "PM_CMPL_STALL_STCX / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "STCX_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a LWSYNC instruction waiting to complete",
+        "MetricExpr": "PM_CMPL_STALL_LWSYNC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "LWSYNC_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction is a HWSYNC instruction stuck at finish waiting for a response from the L2",
+        "MetricExpr": "PM_CMPL_STALL_HWSYNC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "HWSYNC_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction required special handling before completion",
+        "MetricExpr": "PM_CMPL_STALL_SPECIAL / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "SPECIAL_COMPLETION_STALL_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when dispatch was stalled because fetch was being held, so there was nothing in the pipeline for this thread",
+        "MetricExpr": "PM_DISP_STALL_FETCH / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_FETCH_CPI"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTC instruction was held at dispatch because of power management",
+        "MetricExpr": "PM_DISP_STALL_HELD_HALT_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "CPI",
+        "MetricName": "DISPATCHED_HELD_HALT_CPI"
+    },
+    {
+        "BriefDescription": "Percentage of flushes per completed instruction",
+        "MetricExpr": "PM_FLUSH / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Others",
+        "MetricName": "FLUSH_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of flushes due to a branch mispredict per completed instruction",
+        "MetricExpr": "PM_FLUSH_MPRED / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Others",
+        "MetricName": "BR_MPRED_FLUSH_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of branch mispredictions per completed instruction",
+        "MetricExpr": "PM_BR_MPRED_CMPL / PM_RUN_INST_CMPL",
+        "MetricGroup": "Others",
+        "MetricName": "BRANCH_MISPREDICTION_RATE"
+    },
+    {
+        "BriefDescription": "Percentage of finished loads that missed in the L1",
+        "MetricExpr": "PM_LD_MISS_L1 / PM_LD_REF_L1 * 100",
+        "MetricGroup": "Others",
+        "MetricName": "L1_LD_MISS_RATIO",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions that were loads that missed the L1",
+        "MetricExpr": "PM_LD_MISS_L1 / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Others",
+        "MetricName": "L1_LD_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions when the DPTEG required for the load/store instruction in execution was missing from the TLB",
+        "MetricExpr": "PM_DTLB_MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Others",
+        "MetricName": "DTLB_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of completed instructions dispatched per instruction completed",
+        "MetricExpr": "PM_INST_DISP / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "DISPATCH_PER_INST_CMPL"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions that were a demand load that did not hit in the L1 or L2",
+        "MetricExpr": "PM_DATA_FROM_L2MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "General",
+        "MetricName": "L2_LD_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions that were demand fetches that missed the L1 icache",
+        "MetricExpr": "PM_L1_ICACHE_MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Instruction_Misses",
+        "MetricName": "L1_INST_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions that were demand fetches that reloaded from beyond the L3 icache",
+        "MetricExpr": "PM_INST_FROM_L3MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "General",
+        "MetricName": "L3_INST_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of completed instructions per cycle",
+        "MetricExpr": "PM_INST_CMPL / PM_CYC",
+        "MetricGroup": "General",
+        "MetricName": "IPC"
+    },
+    {
+        "BriefDescription": "Average number of cycles per completed instruction group",
+        "MetricExpr": "PM_CYC / PM_1PLUS_PPC_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "CYCLES_PER_COMPLETED_INSTRUCTIONS_SET"
+    },
+    {
+        "BriefDescription": "Percentage of cycles when at least 1 instruction dispatched",
+        "MetricExpr": "PM_1PLUS_PPC_DISP / PM_RUN_CYC * 100",
+        "MetricGroup": "General",
+        "MetricName": "CYCLES_ATLEAST_ONE_INST_DISPATCHED",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of finished loads per completed instruction",
+        "MetricExpr": "PM_LD_REF_L1 / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "LOADS_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of finished stores per completed instruction",
+        "MetricExpr": "PM_ST_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "STORES_PER_INST"
+    },
+    {
+        "BriefDescription": "Percentage of demand loads that reloaded from beyond the L2 per completed instruction",
+        "MetricExpr": "PM_DATA_FROM_L2MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "dL1_Reloads",
+        "MetricName": "DL1_RELOAD_FROM_L2_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of demand loads that reloaded from beyond the L3 per completed instruction",
+        "MetricExpr": "PM_DATA_FROM_L3MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "dL1_Reloads",
+        "MetricName": "DL1_RELOAD_FROM_L3_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses with 4k page size per completed instruction",
+        "MetricExpr": "PM_DERAT_MISS_4K / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_4K_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses with 64k page size per completed instruction",
+        "MetricExpr": "PM_DERAT_MISS_64K / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_64K_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of run cycles per completed instruction",
+        "MetricExpr": "PM_RUN_CYC / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "RUN_CPI"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses per completed instruction",
+        "MetricExpr": "PM_DERAT_MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of completed instructions per run cycle",
+        "MetricExpr": "PM_RUN_INST_CMPL / PM_RUN_CYC",
+        "MetricGroup": "General",
+        "MetricName": "RUN_IPC"
+    },
+    {
+        "BriefDescription": "Average number of completed instructions per instruction group",
+        "MetricExpr": "PM_RUN_INST_CMPL / PM_1PLUS_PPC_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "AVERAGE_COMPLETED_INSTRUCTION_SET_SIZE"
+    },
+    {
+        "BriefDescription": "Average number of finished instructions per completed instructions",
+        "MetricExpr": "PM_INST_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "INST_FIN_PER_CMPL"
+    },
+    {
+        "BriefDescription": "Average cycles per completed instruction when the NTF instruction is completing and the finish was overlooked",
+        "MetricExpr": "PM_EXEC_STALL_UNKNOWN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "EXEC_STALL_UNKOWN_CPI"
+    },
+    {
+        "BriefDescription": "Percentage of finished branches that were taken",
+        "MetricExpr": "PM_BR_TAKEN_CMPL / PM_BR_FIN * 100",
+        "MetricGroup": "General",
+        "MetricName": "TAKEN_BRANCHES",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of completed instructions that were a demand load that did not hit in the L1, L2, or the L3",
+        "MetricExpr": "PM_DATA_FROM_L3MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "General",
+        "MetricName": "L3_LD_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Average number of finished branches per completed instruction",
+        "MetricExpr": "PM_BR_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "BRANCHES_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of instructions finished in the LSU per completed instruction",
+        "MetricExpr": "PM_LSU_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "LSU_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of instructions finished in the VSU per completed instruction",
+        "MetricExpr": "PM_VSU_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "VSU_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of TLBIE instructions finished in the LSU per completed instruction",
+        "MetricExpr": "PM_TLBIE_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "TLBIE_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of STCX instructions finshed per completed instruction",
+        "MetricExpr": "PM_STCX_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "STXC_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of LARX instructions finshed per completed instruction",
+        "MetricExpr": "PM_LARX_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "LARX_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of PTESYNC instructions finshed per completed instruction",
+        "MetricExpr": "PM_PTESYNC_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "PTESYNC_PER_INST"
+    },
+    {
+        "BriefDescription": "Average number of simple fixed-point instructions finshed in the store unit per completed instruction",
+        "MetricExpr": "PM_FX_LSU_FIN / PM_RUN_INST_CMPL",
+        "MetricGroup": "General",
+        "MetricName": "FX_PER_INST"
+    },
+    {
+        "BriefDescription": "Percentage of demand load misses that reloaded the L1 cache",
+        "MetricExpr": "PM_LD_DEMAND_MISS_L1 / PM_LD_MISS_L1 * 100",
+        "MetricGroup": "General",
+        "MetricName": "DL1_MISS_RELOADS",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of demand load misses that reloaded from beyond the local L2",
+        "MetricExpr": "PM_DATA_FROM_L2MISS / PM_LD_DEMAND_MISS_L1 * 100",
+        "MetricGroup": "dL1_Reloads",
+        "MetricName": "DL1_RELOAD_FROM_L2_MISS",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of demand load misses that reloaded from beyond the local L3",
+        "MetricExpr": "PM_DATA_FROM_L3MISS / PM_LD_DEMAND_MISS_L1 * 100",
+        "MetricGroup": "dL1_Reloads",
+        "MetricName": "DL1_RELOAD_FROM_L3_MISS",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of cycles stalled due to the NTC instruction waiting for a load miss to resolve from a source beyond the local L2 and local L3",
+        "MetricExpr": "DMISS_L3MISS_STALL_CPI / RUN_CPI * 100",
+        "MetricGroup": "General",
+        "MetricName": "DCACHE_MISS_CPI",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses with 2M page size per completed instruction",
+        "MetricExpr": "PM_DERAT_MISS_2M / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_2M_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses with 16M page size per completed instruction",
+        "MetricExpr": "PM_DERAT_MISS_16M / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_16M_MISS_RATE",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "DERAT miss ratio for 4K page size",
+        "MetricExpr": "PM_DERAT_MISS_4K / PM_DERAT_MISS",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_4K_MISS_RATIO"
+    },
+    {
+        "BriefDescription": "DERAT miss ratio for 2M page size",
+        "MetricExpr": "PM_DERAT_MISS_2M / PM_DERAT_MISS",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_2M_MISS_RATIO"
+    },
+    {
+        "BriefDescription": "DERAT miss ratio for 16M page size",
+        "MetricExpr": "PM_DERAT_MISS_16M / PM_DERAT_MISS",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_16M_MISS_RATIO"
+    },
+    {
+        "BriefDescription": "DERAT miss ratio for 64K page size",
+        "MetricExpr": "PM_DERAT_MISS_64K / PM_DERAT_MISS",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_64K_MISS_RATIO"
+    },
+    {
+        "BriefDescription": "Percentage of DERAT misses that resulted in TLB reloads",
+        "MetricExpr": "PM_DTLB_MISS / PM_DERAT_MISS * 100",
+        "MetricGroup": "Translation",
+        "MetricName": "DERAT_MISS_RELOAD",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of icache misses that were reloaded from beyond the local L3",
+        "MetricExpr": "PM_INST_FROM_L3MISS / PM_L1_ICACHE_MISS * 100",
+        "MetricGroup": "Instruction_Misses",
+        "MetricName": "INST_FROM_L3_MISS",
+        "ScaleUnit": "1%"
+    },
+    {
+        "BriefDescription": "Percentage of icache reloads from the beyond the L3 per completed instruction",
+        "MetricExpr": "PM_INST_FROM_L3MISS / PM_RUN_INST_CMPL * 100",
+        "MetricGroup": "Instruction_Misses",
+        "MetricName": "INST_FROM_L3_MISS_RATE",
+        "ScaleUnit": "1%"
+    }
+]
diff --git a/tools/perf/tests/api-io.c b/tools/perf/tests/api-io.c
index 2ada86ad6084..e91cf2c127f1 100644
--- a/tools/perf/tests/api-io.c
+++ b/tools/perf/tests/api-io.c
@@ -289,8 +289,8 @@ static int test_get_dec(void)
 	return ret;
 }
 
-int test__api_io(struct test *test __maybe_unused,
-		int subtest __maybe_unused)
+static int test__api_io(struct test_suite *test __maybe_unused,
+			int subtest __maybe_unused)
 {
 	int ret = 0;
 
@@ -302,3 +302,5 @@ int test__api_io(struct test *test __maybe_unused,
 		ret = TEST_FAIL;
 	return ret;
 }
+
+DEFINE_SUITE("Test api io", api_io);
diff --git a/tools/perf/tests/attr.c b/tools/perf/tests/attr.c
index 9b40a25376ae..0f73e300f207 100644
--- a/tools/perf/tests/attr.c
+++ b/tools/perf/tests/attr.c
@@ -178,7 +178,7 @@ static int run_dir(const char *d, const char *perf)
 	return system(cmd) ? TEST_FAIL : TEST_OK;
 }
 
-int test__attr(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__attr(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct stat st;
 	char path_perf[PATH_MAX];
@@ -207,3 +207,5 @@ int test__attr(struct test *test __maybe_unused, int subtest __maybe_unused)
 
 	return TEST_SKIP;
 }
+
+DEFINE_SUITE("Setup struct perf_event_attr", attr);
diff --git a/tools/perf/tests/backward-ring-buffer.c b/tools/perf/tests/backward-ring-buffer.c
index 7447a4478991..79a980b1e786 100644
--- a/tools/perf/tests/backward-ring-buffer.c
+++ b/tools/perf/tests/backward-ring-buffer.c
@@ -82,7 +82,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,
 }
 
 
-int test__backward_ring_buffer(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret = TEST_SKIP, err, sample_count = 0, comm_count = 0;
 	char pid[16], sbuf[STRERR_BUFSIZE];
@@ -167,3 +167,5 @@ out_delete_evlist:
 	evlist__delete(evlist);
 	return ret;
 }
+
+DEFINE_SUITE("Read backward ring buffer", backward_ring_buffer);
diff --git a/tools/perf/tests/bitmap.c b/tools/perf/tests/bitmap.c
index 12b805efdca0..384856347236 100644
--- a/tools/perf/tests/bitmap.c
+++ b/tools/perf/tests/bitmap.c
@@ -40,7 +40,7 @@ static int test_bitmap(const char *str)
 	return ret;
 }
 
-int test__bitmap_print(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__bitmap_print(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	TEST_ASSERT_VAL("failed to convert map", test_bitmap("1"));
 	TEST_ASSERT_VAL("failed to convert map", test_bitmap("1,5"));
@@ -51,3 +51,5 @@ int test__bitmap_print(struct test *test __maybe_unused, int subtest __maybe_unu
 	TEST_ASSERT_VAL("failed to convert map", test_bitmap("1-10,12-20,22-30,32-40"));
 	return 0;
 }
+
+DEFINE_SUITE("Print bitmap", bitmap_print);
diff --git a/tools/perf/tests/bp_account.c b/tools/perf/tests/bp_account.c
index 489b50604cf2..d1ebb5561e5b 100644
--- a/tools/perf/tests/bp_account.c
+++ b/tools/perf/tests/bp_account.c
@@ -19,6 +19,19 @@
 #include "../perf-sys.h"
 #include "cloexec.h"
 
+/*
+ * PowerPC and S390 do not support creation of instruction breakpoints using the
+ * perf_event interface.
+ *
+ * Just disable the test for these architectures until these issues are
+ * resolved.
+ */
+#if defined(__powerpc__) || defined(__s390x__)
+#define BP_ACCOUNT_IS_SUPPORTED 0
+#else
+#define BP_ACCOUNT_IS_SUPPORTED 1
+#endif
+
 static volatile long the_var;
 
 static noinline int test_function(void)
@@ -173,13 +186,18 @@ static int detect_share(int wp_cnt, int bp_cnt)
  *     we create another watchpoint to ensure
  *     the slot accounting is correct
  */
-int test__bp_accounting(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__bp_accounting(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int has_ioctl = detect_ioctl();
 	int wp_cnt = detect_cnt(false);
 	int bp_cnt = detect_cnt(true);
 	int share  = detect_share(wp_cnt, bp_cnt);
 
+	if (!BP_ACCOUNT_IS_SUPPORTED) {
+		pr_debug("Test not supported on this architecture");
+		return TEST_SKIP;
+	}
+
 	pr_debug("watchpoints count %d, breakpoints count %d, has_ioctl %d, share %d\n",
 		 wp_cnt, bp_cnt, has_ioctl, share);
 
@@ -189,18 +207,4 @@ int test__bp_accounting(struct test *test __maybe_unused, int subtest __maybe_un
 	return bp_accounting(wp_cnt, share);
 }
 
-bool test__bp_account_is_supported(void)
-{
-	/*
-	 * PowerPC and S390 do not support creation of instruction
-	 * breakpoints using the perf_event interface.
-	 *
-	 * Just disable the test for these architectures until these
-	 * issues are resolved.
-	 */
-#if defined(__powerpc__) || defined(__s390x__)
-	return false;
-#else
-	return true;
-#endif
-}
+DEFINE_SUITE("Breakpoint accounting", bp_accounting);
diff --git a/tools/perf/tests/bp_signal.c b/tools/perf/tests/bp_signal.c
index ef37353636d8..1f2908f02389 100644
--- a/tools/perf/tests/bp_signal.c
+++ b/tools/perf/tests/bp_signal.c
@@ -161,11 +161,16 @@ static long long bp_count(int fd)
 	return count;
 }
 
-int test__bp_signal(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__bp_signal(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct sigaction sa;
 	long long count1, count2, count3;
 
+	if (!BP_SIGNAL_IS_SUPPORTED) {
+		pr_debug("Test not supported on this architecture");
+		return TEST_SKIP;
+	}
+
 	/* setup SIGIO signal handler */
 	memset(&sa, 0, sizeof(struct sigaction));
 	sa.sa_sigaction = (void *) sig_handler;
@@ -285,29 +290,4 @@ int test__bp_signal(struct test *test __maybe_unused, int subtest __maybe_unused
 		TEST_OK : TEST_FAIL;
 }
 
-bool test__bp_signal_is_supported(void)
-{
-	/*
-	 * PowerPC and S390 do not support creation of instruction
-	 * breakpoints using the perf_event interface.
-	 *
-	 * ARM requires explicit rounding down of the instruction
-	 * pointer in Thumb mode, and then requires the single-step
-	 * to be handled explicitly in the overflow handler to avoid
-	 * stepping into the SIGIO handler and getting stuck on the
-	 * breakpointed instruction.
-	 *
-	 * Since arm64 has the same issue with arm for the single-step
-	 * handling, this case also gets stuck on the breakpointed
-	 * instruction.
-	 *
-	 * Just disable the test for these architectures until these
-	 * issues are resolved.
-	 */
-#if defined(__powerpc__) || defined(__s390x__) || defined(__arm__) || \
-    defined(__aarch64__)
-	return false;
-#else
-	return true;
-#endif
-}
+DEFINE_SUITE("Breakpoint overflow signal handler", bp_signal);
diff --git a/tools/perf/tests/bp_signal_overflow.c b/tools/perf/tests/bp_signal_overflow.c
index eb4dbbddf4ff..4e897c2cf26b 100644
--- a/tools/perf/tests/bp_signal_overflow.c
+++ b/tools/perf/tests/bp_signal_overflow.c
@@ -59,13 +59,18 @@ static long long bp_count(int fd)
 #define EXECUTIONS 10000
 #define THRESHOLD  100
 
-int test__bp_signal_overflow(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__bp_signal_overflow(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_event_attr pe;
 	struct sigaction sa;
 	long long count;
 	int fd, i, fails = 0;
 
+	if (!BP_SIGNAL_IS_SUPPORTED) {
+		pr_debug("Test not supported on this architecture");
+		return TEST_SKIP;
+	}
+
 	/* setup SIGIO signal handler */
 	memset(&sa, 0, sizeof(struct sigaction));
 	sa.sa_sigaction = (void *) sig_handler;
@@ -133,3 +138,5 @@ int test__bp_signal_overflow(struct test *test __maybe_unused, int subtest __may
 
 	return fails ? TEST_FAIL : TEST_OK;
 }
+
+DEFINE_SUITE("Breakpoint overflow sampling", bp_signal_overflow);
diff --git a/tools/perf/tests/bpf.c b/tools/perf/tests/bpf.c
index 2bf146e49ce8..329f77f592f4 100644
--- a/tools/perf/tests/bpf.c
+++ b/tools/perf/tests/bpf.c
@@ -62,7 +62,6 @@ static int llseek_loop(void)
 
 static struct {
 	enum test_llvm__testcase prog_id;
-	const char *desc;
 	const char *name;
 	const char *msg_compile_fail;
 	const char *msg_load_fail;
@@ -72,7 +71,6 @@ static struct {
 } bpf_testcase_table[] = {
 	{
 		.prog_id	  = LLVM_TESTCASE_BASE,
-		.desc		  = "Basic BPF filtering",
 		.name		  = "[basic_bpf_test]",
 		.msg_compile_fail = "fix 'perf test LLVM' first",
 		.msg_load_fail	  = "load bpf object failed",
@@ -81,7 +79,6 @@ static struct {
 	},
 	{
 		.prog_id	  = LLVM_TESTCASE_BASE,
-		.desc		  = "BPF pinning",
 		.name		  = "[bpf_pinning]",
 		.msg_compile_fail = "fix kbuild first",
 		.msg_load_fail	  = "check your vmlinux setting?",
@@ -92,7 +89,6 @@ static struct {
 #ifdef HAVE_BPF_PROLOGUE
 	{
 		.prog_id	  = LLVM_TESTCASE_BPF_PROLOGUE,
-		.desc		  = "BPF prologue generation",
 		.name		  = "[bpf_prologue_test]",
 		.msg_compile_fail = "fix kbuild first",
 		.msg_load_fail	  = "check your vmlinux setting?",
@@ -283,18 +279,6 @@ out:
 	return ret;
 }
 
-int test__bpf_subtest_get_nr(void)
-{
-	return (int)ARRAY_SIZE(bpf_testcase_table);
-}
-
-const char *test__bpf_subtest_get_desc(int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(bpf_testcase_table))
-		return NULL;
-	return bpf_testcase_table[i].desc;
-}
-
 static int check_env(void)
 {
 	int err;
@@ -313,7 +297,7 @@ static int check_env(void)
 	}
 
 	err = bpf_load_program(BPF_PROG_TYPE_KPROBE, insns,
-			       sizeof(insns) / sizeof(insns[0]),
+			       ARRAY_SIZE(insns),
 			       license, kver_int, NULL, 0);
 	if (err < 0) {
 		pr_err("Missing basic BPF support, skip this test: %s\n",
@@ -325,7 +309,7 @@ static int check_env(void)
 	return 0;
 }
 
-int test__bpf(struct test *test __maybe_unused, int i)
+static int test__bpf(int i)
 {
 	int err;
 
@@ -343,21 +327,60 @@ int test__bpf(struct test *test __maybe_unused, int i)
 	err = __test__bpf(i);
 	return err;
 }
+#endif
 
-#else
-int test__bpf_subtest_get_nr(void)
+static int test__basic_bpf_test(struct test_suite *test __maybe_unused,
+				int subtest __maybe_unused)
 {
-	return 0;
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__bpf(0);
+#else
+	pr_debug("Skip BPF test because BPF support is not compiled\n");
+	return TEST_SKIP;
+#endif
 }
 
-const char *test__bpf_subtest_get_desc(int i __maybe_unused)
+static int test__bpf_pinning(struct test_suite *test __maybe_unused,
+			     int subtest __maybe_unused)
 {
-	return NULL;
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__bpf(1);
+#else
+	pr_debug("Skip BPF test because BPF support is not compiled\n");
+	return TEST_SKIP;
+#endif
 }
 
-int test__bpf(struct test *test __maybe_unused, int i __maybe_unused)
+static int test__bpf_prologue_test(struct test_suite *test __maybe_unused,
+				   int subtest __maybe_unused)
 {
+#if defined(HAVE_LIBBPF_SUPPORT) && defined(HAVE_BPF_PROLOGUE)
+	return test__bpf(2);
+#else
 	pr_debug("Skip BPF test because BPF support is not compiled\n");
 	return TEST_SKIP;
+#endif
 }
+
+
+static struct test_case bpf_tests[] = {
+#ifdef HAVE_LIBBPF_SUPPORT
+	TEST_CASE("Basic BPF filtering", basic_bpf_test),
+	TEST_CASE("BPF pinning", bpf_pinning),
+#ifdef HAVE_BPF_PROLOGUE
+	TEST_CASE("BPF prologue generation", bpf_prologue_test),
+#else
+	TEST_CASE_REASON("BPF prologue generation", bpf_prologue_test, "not compiled in"),
+#endif
+#else
+	TEST_CASE_REASON("Basic BPF filtering", basic_bpf_test, "not compiled in"),
+	TEST_CASE_REASON("BPF pinning", bpf_pinning, "not compiled in"),
+	TEST_CASE_REASON("BPF prologue generation", bpf_prologue_test, "not compiled in"),
 #endif
+	{ .name = NULL, }
+};
+
+struct test_suite suite__bpf = {
+	.desc = "BPF filter",
+	.test_cases = bpf_tests,
+};
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
index da7dc5e45d0c..8cb5a1c3489e 100644
--- a/tools/perf/tests/builtin-test.c
+++ b/tools/perf/tests/builtin-test.c
@@ -30,350 +30,134 @@
 
 static bool dont_fork;
 
-struct test __weak arch_tests[] = {
-	{
-		.func = NULL,
-	},
+struct test_suite *__weak arch_tests[] = {
+	NULL,
 };
 
-static struct test generic_tests[] = {
-	{
-		.desc = "vmlinux symtab matches kallsyms",
-		.func = test__vmlinux_matches_kallsyms,
-	},
-	{
-		.desc = "Detect openat syscall event",
-		.func = test__openat_syscall_event,
-	},
-	{
-		.desc = "Detect openat syscall event on all cpus",
-		.func = test__openat_syscall_event_on_all_cpus,
-	},
-	{
-		.desc = "Read samples using the mmap interface",
-		.func = test__basic_mmap,
-	},
-	{
-		.desc = "Test data source output",
-		.func = test__mem,
-	},
-	{
-		.desc = "Parse event definition strings",
-		.func = test__parse_events,
-	},
-	{
-		.desc = "Simple expression parser",
-		.func = test__expr,
-	},
-	{
-		.desc = "PERF_RECORD_* events & perf_sample fields",
-		.func = test__PERF_RECORD,
-	},
-	{
-		.desc = "Parse perf pmu format",
-		.func = test__pmu,
-	},
-	{
-		.desc = "PMU events",
-		.func = test__pmu_events,
-		.subtest = {
-			.skip_if_fail	= false,
-			.get_nr		= test__pmu_events_subtest_get_nr,
-			.get_desc	= test__pmu_events_subtest_get_desc,
-			.skip_reason	= test__pmu_events_subtest_skip_reason,
-		},
-
-	},
-	{
-		.desc = "DSO data read",
-		.func = test__dso_data,
-	},
-	{
-		.desc = "DSO data cache",
-		.func = test__dso_data_cache,
-	},
-	{
-		.desc = "DSO data reopen",
-		.func = test__dso_data_reopen,
-	},
-	{
-		.desc = "Roundtrip evsel->name",
-		.func = test__perf_evsel__roundtrip_name_test,
-	},
-	{
-		.desc = "Parse sched tracepoints fields",
-		.func = test__perf_evsel__tp_sched_test,
-	},
-	{
-		.desc = "syscalls:sys_enter_openat event fields",
-		.func = test__syscall_openat_tp_fields,
-	},
-	{
-		.desc = "Setup struct perf_event_attr",
-		.func = test__attr,
-	},
-	{
-		.desc = "Match and link multiple hists",
-		.func = test__hists_link,
-	},
-	{
-		.desc = "'import perf' in python",
-		.func = test__python_use,
-	},
-	{
-		.desc = "Breakpoint overflow signal handler",
-		.func = test__bp_signal,
-		.is_supported = test__bp_signal_is_supported,
-	},
-	{
-		.desc = "Breakpoint overflow sampling",
-		.func = test__bp_signal_overflow,
-		.is_supported = test__bp_signal_is_supported,
-	},
-	{
-		.desc = "Breakpoint accounting",
-		.func = test__bp_accounting,
-		.is_supported = test__bp_account_is_supported,
-	},
-	{
-		.desc = "Watchpoint",
-		.func = test__wp,
-		.is_supported = test__wp_is_supported,
-		.subtest = {
-			.skip_if_fail	= false,
-			.get_nr		= test__wp_subtest_get_nr,
-			.get_desc	= test__wp_subtest_get_desc,
-			.skip_reason    = test__wp_subtest_skip_reason,
-		},
-	},
-	{
-		.desc = "Number of exit events of a simple workload",
-		.func = test__task_exit,
-	},
-	{
-		.desc = "Software clock events period values",
-		.func = test__sw_clock_freq,
-	},
-	{
-		.desc = "Object code reading",
-		.func = test__code_reading,
-	},
-	{
-		.desc = "Sample parsing",
-		.func = test__sample_parsing,
-	},
-	{
-		.desc = "Use a dummy software event to keep tracking",
-		.func = test__keep_tracking,
-	},
-	{
-		.desc = "Parse with no sample_id_all bit set",
-		.func = test__parse_no_sample_id_all,
-	},
-	{
-		.desc = "Filter hist entries",
-		.func = test__hists_filter,
-	},
-	{
-		.desc = "Lookup mmap thread",
-		.func = test__mmap_thread_lookup,
-	},
-	{
-		.desc = "Share thread maps",
-		.func = test__thread_maps_share,
-	},
-	{
-		.desc = "Sort output of hist entries",
-		.func = test__hists_output,
-	},
-	{
-		.desc = "Cumulate child hist entries",
-		.func = test__hists_cumulate,
-	},
-	{
-		.desc = "Track with sched_switch",
-		.func = test__switch_tracking,
-	},
-	{
-		.desc = "Filter fds with revents mask in a fdarray",
-		.func = test__fdarray__filter,
-	},
-	{
-		.desc = "Add fd to a fdarray, making it autogrow",
-		.func = test__fdarray__add,
-	},
-	{
-		.desc = "kmod_path__parse",
-		.func = test__kmod_path__parse,
-	},
-	{
-		.desc = "Thread map",
-		.func = test__thread_map,
-	},
-	{
-		.desc = "LLVM search and compile",
-		.func = test__llvm,
-		.subtest = {
-			.skip_if_fail	= true,
-			.get_nr		= test__llvm_subtest_get_nr,
-			.get_desc	= test__llvm_subtest_get_desc,
-		},
-	},
-	{
-		.desc = "Session topology",
-		.func = test__session_topology,
-	},
-	{
-		.desc = "BPF filter",
-		.func = test__bpf,
-		.subtest = {
-			.skip_if_fail	= true,
-			.get_nr		= test__bpf_subtest_get_nr,
-			.get_desc	= test__bpf_subtest_get_desc,
-		},
-	},
-	{
-		.desc = "Synthesize thread map",
-		.func = test__thread_map_synthesize,
-	},
-	{
-		.desc = "Remove thread map",
-		.func = test__thread_map_remove,
-	},
-	{
-		.desc = "Synthesize cpu map",
-		.func = test__cpu_map_synthesize,
-	},
-	{
-		.desc = "Synthesize stat config",
-		.func = test__synthesize_stat_config,
-	},
-	{
-		.desc = "Synthesize stat",
-		.func = test__synthesize_stat,
-	},
-	{
-		.desc = "Synthesize stat round",
-		.func = test__synthesize_stat_round,
-	},
-	{
-		.desc = "Synthesize attr update",
-		.func = test__event_update,
-	},
-	{
-		.desc = "Event times",
-		.func = test__event_times,
-	},
-	{
-		.desc = "Read backward ring buffer",
-		.func = test__backward_ring_buffer,
-	},
-	{
-		.desc = "Print cpu map",
-		.func = test__cpu_map_print,
-	},
-	{
-		.desc = "Merge cpu map",
-		.func = test__cpu_map_merge,
-	},
-
-	{
-		.desc = "Probe SDT events",
-		.func = test__sdt_event,
-	},
-	{
-		.desc = "is_printable_array",
-		.func = test__is_printable_array,
-	},
-	{
-		.desc = "Print bitmap",
-		.func = test__bitmap_print,
-	},
-	{
-		.desc = "perf hooks",
-		.func = test__perf_hooks,
-	},
-	{
-		.desc = "builtin clang support",
-		.func = test__clang,
-		.subtest = {
-			.skip_if_fail	= true,
-			.get_nr		= test__clang_subtest_get_nr,
-			.get_desc	= test__clang_subtest_get_desc,
-		}
-	},
-	{
-		.desc = "unit_number__scnprintf",
-		.func = test__unit_number__scnprint,
-	},
-	{
-		.desc = "mem2node",
-		.func = test__mem2node,
-	},
-	{
-		.desc = "time utils",
-		.func = test__time_utils,
-	},
-	{
-		.desc = "Test jit_write_elf",
-		.func = test__jit_write_elf,
-	},
-	{
-		.desc = "Test libpfm4 support",
-		.func = test__pfm,
-		.subtest = {
-			.skip_if_fail	= true,
-			.get_nr		= test__pfm_subtest_get_nr,
-			.get_desc	= test__pfm_subtest_get_desc,
-		}
-	},
-	{
-		.desc = "Test api io",
-		.func = test__api_io,
-	},
-	{
-		.desc = "maps__merge_in",
-		.func = test__maps__merge_in,
-	},
-	{
-		.desc = "Demangle Java",
-		.func = test__demangle_java,
-	},
-	{
-		.desc = "Demangle OCaml",
-		.func = test__demangle_ocaml,
-	},
-	{
-		.desc = "Parse and process metrics",
-		.func = test__parse_metric,
-	},
-	{
-		.desc = "PE file support",
-		.func = test__pe_file_parsing,
-	},
-	{
-		.desc = "Event expansion for cgroups",
-		.func = test__expand_cgroup_events,
-	},
-	{
-		.desc = "Convert perf time to TSC",
-		.func = test__perf_time_to_tsc,
-		.is_supported = test__tsc_is_supported,
-	},
-	{
-		.desc = "dlfilter C API",
-		.func = test__dlfilter,
-	},
-	{
-		.func = NULL,
-	},
+static struct test_suite *generic_tests[] = {
+	&suite__vmlinux_matches_kallsyms,
+	&suite__openat_syscall_event,
+	&suite__openat_syscall_event_on_all_cpus,
+	&suite__basic_mmap,
+	&suite__mem,
+	&suite__parse_events,
+	&suite__expr,
+	&suite__PERF_RECORD,
+	&suite__pmu,
+	&suite__pmu_events,
+	&suite__dso_data,
+	&suite__dso_data_cache,
+	&suite__dso_data_reopen,
+	&suite__perf_evsel__roundtrip_name_test,
+	&suite__perf_evsel__tp_sched_test,
+	&suite__syscall_openat_tp_fields,
+	&suite__attr,
+	&suite__hists_link,
+	&suite__python_use,
+	&suite__bp_signal,
+	&suite__bp_signal_overflow,
+	&suite__bp_accounting,
+	&suite__wp,
+	&suite__task_exit,
+	&suite__sw_clock_freq,
+	&suite__code_reading,
+	&suite__sample_parsing,
+	&suite__keep_tracking,
+	&suite__parse_no_sample_id_all,
+	&suite__hists_filter,
+	&suite__mmap_thread_lookup,
+	&suite__thread_maps_share,
+	&suite__hists_output,
+	&suite__hists_cumulate,
+	&suite__switch_tracking,
+	&suite__fdarray__filter,
+	&suite__fdarray__add,
+	&suite__kmod_path__parse,
+	&suite__thread_map,
+	&suite__llvm,
+	&suite__session_topology,
+	&suite__bpf,
+	&suite__thread_map_synthesize,
+	&suite__thread_map_remove,
+	&suite__cpu_map_synthesize,
+	&suite__synthesize_stat_config,
+	&suite__synthesize_stat,
+	&suite__synthesize_stat_round,
+	&suite__event_update,
+	&suite__event_times,
+	&suite__backward_ring_buffer,
+	&suite__cpu_map_print,
+	&suite__cpu_map_merge,
+	&suite__sdt_event,
+	&suite__is_printable_array,
+	&suite__bitmap_print,
+	&suite__perf_hooks,
+	&suite__clang,
+	&suite__unit_number__scnprint,
+	&suite__mem2node,
+	&suite__time_utils,
+	&suite__jit_write_elf,
+	&suite__pfm,
+	&suite__api_io,
+	&suite__maps__merge_in,
+	&suite__demangle_java,
+	&suite__demangle_ocaml,
+	&suite__parse_metric,
+	&suite__pe_file_parsing,
+	&suite__expand_cgroup_events,
+	&suite__perf_time_to_tsc,
+	&suite__dlfilter,
+	NULL,
 };
 
-static struct test *tests[] = {
+static struct test_suite **tests[] = {
 	generic_tests,
 	arch_tests,
 };
 
+static int num_subtests(const struct test_suite *t)
+{
+	int num;
+
+	if (!t->test_cases)
+		return 0;
+
+	num = 0;
+	while (t->test_cases[num].name)
+		num++;
+
+	return num;
+}
+
+static bool has_subtests(const struct test_suite *t)
+{
+	return num_subtests(t) > 1;
+}
+
+static const char *skip_reason(const struct test_suite *t, int subtest)
+{
+	if (t->test_cases && subtest >= 0)
+		return t->test_cases[subtest].skip_reason;
+
+	return NULL;
+}
+
+static const char *test_description(const struct test_suite *t, int subtest)
+{
+	if (t->test_cases && subtest >= 0)
+		return t->test_cases[subtest].desc;
+
+	return t->desc;
+}
+
+static test_fnptr test_function(const struct test_suite *t, int subtest)
+{
+	if (subtest <= 0)
+		return t->test_cases[0].run_case;
+
+	return t->test_cases[subtest].run_case;
+}
+
 static bool perf_test__matches(const char *desc, int curr, int argc, const char *argv[])
 {
 	int i;
@@ -398,7 +182,7 @@ static bool perf_test__matches(const char *desc, int curr, int argc, const char
 	return false;
 }
 
-static int run_test(struct test *test, int subtest)
+static int run_test(struct test_suite *test, int subtest)
 {
 	int status, err = -1, child = dont_fork ? 0 : fork();
 	char sbuf[STRERR_BUFSIZE];
@@ -430,7 +214,7 @@ static int run_test(struct test *test, int subtest)
 			}
 		}
 
-		err = test->func(test, subtest);
+		err = test_function(test, subtest)(test, subtest);
 		if (!dont_fork)
 			exit(err);
 	}
@@ -450,24 +234,19 @@ static int run_test(struct test *test, int subtest)
 	return err;
 }
 
-#define for_each_test(j, t)	 				\
+#define for_each_test(j, k, t)			\
 	for (j = 0; j < ARRAY_SIZE(tests); j++)	\
-		for (t = &tests[j][0]; t->func; t++)
+		for (k = 0, t = tests[j][k]; tests[j][k]; k++, t = tests[j][k])
 
-static int test_and_print(struct test *t, bool force_skip, int subtest)
+static int test_and_print(struct test_suite *t, int subtest)
 {
 	int err;
 
-	if (!force_skip) {
-		pr_debug("\n--- start ---\n");
-		err = run_test(t, subtest);
-		pr_debug("---- end ----\n");
-	} else {
-		pr_debug("\n--- force skipped ---\n");
-		err = TEST_SKIP;
-	}
+	pr_debug("\n--- start ---\n");
+	err = run_test(t, subtest);
+	pr_debug("---- end ----\n");
 
-	if (!t->subtest.get_nr)
+	if (!has_subtests(t))
 		pr_debug("%s:", t->desc);
 	else
 		pr_debug("%s subtest %d:", t->desc, subtest + 1);
@@ -477,11 +256,10 @@ static int test_and_print(struct test *t, bool force_skip, int subtest)
 		pr_info(" Ok\n");
 		break;
 	case TEST_SKIP: {
-		const char *skip_reason = NULL;
-		if (t->subtest.skip_reason)
-			skip_reason = t->subtest.skip_reason(subtest);
-		if (skip_reason)
-			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (%s)\n", skip_reason);
+		const char *reason = skip_reason(t, subtest);
+
+		if (reason)
+			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (%s)\n", reason);
 		else
 			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip\n");
 	}
@@ -580,7 +358,7 @@ struct shell_test {
 	const char *file;
 };
 
-static int shell_test__run(struct test *test, int subdir __maybe_unused)
+static int shell_test__run(struct test_suite *test, int subdir __maybe_unused)
 {
 	int err;
 	char script[PATH_MAX];
@@ -622,24 +400,34 @@ static int run_shell_tests(int argc, const char *argv[], int i, int width,
 	for_each_shell_test(entlist, n_dirs, st.dir, ent) {
 		int curr = i++;
 		char desc[256];
-		struct test test = {
-			.desc = shell_test__description(desc, sizeof(desc), st.dir, ent->d_name),
-			.func = shell_test__run,
+		struct test_case test_cases[] = {
+			{
+				.desc = shell_test__description(desc,
+								sizeof(desc),
+								st.dir,
+								ent->d_name),
+				.run_case = shell_test__run,
+			},
+			{ .name = NULL, }
+		};
+		struct test_suite test_suite = {
+			.desc = test_cases[0].desc,
+			.test_cases = test_cases,
 			.priv = &st,
 		};
 
-		if (!perf_test__matches(test.desc, curr, argc, argv))
+		if (!perf_test__matches(test_suite.desc, curr, argc, argv))
 			continue;
 
 		st.file = ent->d_name;
-		pr_info("%2d: %-*s:", i, width, test.desc);
+		pr_info("%2d: %-*s:", i, width, test_suite.desc);
 
 		if (intlist__find(skiplist, i)) {
 			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (user override)\n");
 			continue;
 		}
 
-		test_and_print(&test, false, -1);
+		test_and_print(&test_suite, 0);
 	}
 
 	for (e = 0; e < n_dirs; e++)
@@ -650,33 +438,31 @@ static int run_shell_tests(int argc, const char *argv[], int i, int width,
 
 static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist)
 {
-	struct test *t;
-	unsigned int j;
+	struct test_suite *t;
+	unsigned int j, k;
 	int i = 0;
 	int width = shell_tests__max_desc_width();
 
-	for_each_test(j, t) {
-		int len = strlen(t->desc);
+	for_each_test(j, k, t) {
+		int len = strlen(test_description(t, -1));
 
 		if (width < len)
 			width = len;
 	}
 
-	for_each_test(j, t) {
-		int curr = i++, err;
+	for_each_test(j, k, t) {
+		int curr = i++;
 		int subi;
 
-		if (!perf_test__matches(t->desc, curr, argc, argv)) {
+		if (!perf_test__matches(test_description(t, -1), curr, argc, argv)) {
 			bool skip = true;
 			int subn;
 
-			if (!t->subtest.get_nr)
-				continue;
-
-			subn = t->subtest.get_nr();
+			subn = num_subtests(t);
 
 			for (subi = 0; subi < subn; subi++) {
-				if (perf_test__matches(t->subtest.get_desc(subi), curr, argc, argv))
+				if (perf_test__matches(test_description(t, subi),
+							curr, argc, argv))
 					skip = false;
 			}
 
@@ -684,22 +470,17 @@ static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist)
 				continue;
 		}
 
-		if (t->is_supported && !t->is_supported()) {
-			pr_debug("%2d: %-*s: Disabled\n", i, width, t->desc);
-			continue;
-		}
-
-		pr_info("%2d: %-*s:", i, width, t->desc);
+		pr_info("%2d: %-*s:", i, width, test_description(t, -1));
 
 		if (intlist__find(skiplist, i)) {
 			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (user override)\n");
 			continue;
 		}
 
-		if (!t->subtest.get_nr) {
-			test_and_print(t, false, -1);
+		if (!has_subtests(t)) {
+			test_and_print(t, -1);
 		} else {
-			int subn = t->subtest.get_nr();
+			int subn = num_subtests(t);
 			/*
 			 * minus 2 to align with normal testcases.
 			 * For subtest we print additional '.x' in number.
@@ -709,7 +490,6 @@ static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist)
 			 * 35.1: Basic BPF llvm compiling test                          : Ok
 			 */
 			int subw = width > 2 ? width - 2 : width;
-			bool skip = false;
 
 			if (subn <= 0) {
 				color_fprintf(stderr, PERF_COLOR_YELLOW,
@@ -719,21 +499,20 @@ static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist)
 			pr_info("\n");
 
 			for (subi = 0; subi < subn; subi++) {
-				int len = strlen(t->subtest.get_desc(subi));
+				int len = strlen(test_description(t, subi));
 
 				if (subw < len)
 					subw = len;
 			}
 
 			for (subi = 0; subi < subn; subi++) {
-				if (!perf_test__matches(t->subtest.get_desc(subi), curr, argc, argv))
+				if (!perf_test__matches(test_description(t, subi),
+							curr, argc, argv))
 					continue;
 
 				pr_info("%2d.%1d: %-*s:", i, subi + 1, subw,
-					t->subtest.get_desc(subi));
-				err = test_and_print(t, skip, subi);
-				if (err != TEST_OK && t->subtest.skip_if_fail)
-					skip = true;
+					test_description(t, subi));
+				test_and_print(t, subi);
 			}
 		}
 	}
@@ -759,7 +538,7 @@ static int perf_test__list_shell(int argc, const char **argv, int i)
 	for_each_shell_test(entlist, n_dirs, path, ent) {
 		int curr = i++;
 		char bf[256];
-		struct test t = {
+		struct test_suite t = {
 			.desc = shell_test__description(bf, sizeof(bf), path, ent->d_name),
 		};
 
@@ -778,26 +557,25 @@ static int perf_test__list_shell(int argc, const char **argv, int i)
 
 static int perf_test__list(int argc, const char **argv)
 {
-	unsigned int j;
-	struct test *t;
+	unsigned int j, k;
+	struct test_suite *t;
 	int i = 0;
 
-	for_each_test(j, t) {
+	for_each_test(j, k, t) {
 		int curr = i++;
 
-		if (!perf_test__matches(t->desc, curr, argc, argv) ||
-		    (t->is_supported && !t->is_supported()))
+		if (!perf_test__matches(test_description(t, -1), curr, argc, argv))
 			continue;
 
-		pr_info("%2d: %s\n", i, t->desc);
+		pr_info("%2d: %s\n", i, test_description(t, -1));
 
-		if (t->subtest.get_nr) {
-			int subn = t->subtest.get_nr();
+		if (has_subtests(t)) {
+			int subn = num_subtests(t);
 			int subi;
 
 			for (subi = 0; subi < subn; subi++)
 				pr_info("%2d:%1d: %s\n", i, subi + 1,
-					t->subtest.get_desc(subi));
+					test_description(t, subi));
 		}
 	}
 
diff --git a/tools/perf/tests/clang.c b/tools/perf/tests/clang.c
index 2577d3ed1531..a7111005d5b9 100644
--- a/tools/perf/tests/clang.c
+++ b/tools/perf/tests/clang.c
@@ -3,44 +3,30 @@
 #include "c++/clang-c.h"
 #include <linux/kernel.h>
 
-static struct {
-	int (*func)(void);
-	const char *desc;
-} clang_testcase_table[] = {
-#ifdef HAVE_LIBCLANGLLVM_SUPPORT
-	{
-		.func = test__clang_to_IR,
-		.desc = "builtin clang compile C source to IR",
-	},
-	{
-		.func = test__clang_to_obj,
-		.desc = "builtin clang compile C source to ELF object",
-	},
-#endif
-};
-
-int test__clang_subtest_get_nr(void)
-{
-	return (int)ARRAY_SIZE(clang_testcase_table);
-}
-
-const char *test__clang_subtest_get_desc(int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(clang_testcase_table))
-		return NULL;
-	return clang_testcase_table[i].desc;
-}
-
 #ifndef HAVE_LIBCLANGLLVM_SUPPORT
-int test__clang(struct test *test __maybe_unused, int i __maybe_unused)
+static int test__clang_to_IR(struct test_suite *test __maybe_unused,
+			     int subtest __maybe_unused)
 {
 	return TEST_SKIP;
 }
-#else
-int test__clang(struct test *test __maybe_unused, int i)
+
+static int test__clang_to_obj(struct test_suite *test __maybe_unused,
+			      int subtest __maybe_unused)
 {
-	if (i < 0 || i >= (int)ARRAY_SIZE(clang_testcase_table))
-		return TEST_FAIL;
-	return clang_testcase_table[i].func();
+	return TEST_SKIP;
 }
 #endif
+
+static struct test_case clang_tests[] = {
+	TEST_CASE_REASON("builtin clang compile C source to IR", clang_to_IR,
+			 "not compiled in"),
+	TEST_CASE_REASON("builtin clang compile C source to ELF object",
+			 clang_to_obj,
+			 "not compiled in"),
+	{ .name = NULL, }
+};
+
+struct test_suite suite__clang = {
+	.desc = "builtin clang support",
+	.test_cases = clang_tests,
+};
diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
index f439bd49da19..5610767b407f 100644
--- a/tools/perf/tests/code-reading.c
+++ b/tools/perf/tests/code-reading.c
@@ -716,7 +716,7 @@ out_err:
 	return err;
 }
 
-int test__code_reading(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__code_reading(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret;
 
@@ -743,3 +743,5 @@ int test__code_reading(struct test *test __maybe_unused, int subtest __maybe_unu
 		return -1;
 	};
 }
+
+DEFINE_SUITE("Object code reading", code_reading);
diff --git a/tools/perf/tests/cpumap.c b/tools/perf/tests/cpumap.c
index 0472b110fe65..89a155092f85 100644
--- a/tools/perf/tests/cpumap.c
+++ b/tools/perf/tests/cpumap.c
@@ -75,7 +75,7 @@ static int process_event_cpus(struct perf_tool *tool __maybe_unused,
 }
 
 
-int test__cpu_map_synthesize(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__cpu_map_synthesize(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_cpu_map *cpus;
 
@@ -111,7 +111,7 @@ static int cpu_map_print(const char *str)
 	return !strcmp(buf, str);
 }
 
-int test__cpu_map_print(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__cpu_map_print(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1"));
 	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1,5"));
@@ -123,7 +123,7 @@ int test__cpu_map_print(struct test *test __maybe_unused, int subtest __maybe_un
 	return 0;
 }
 
-int test__cpu_map_merge(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__cpu_map_merge(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_cpu_map *a = perf_cpu_map__new("4,2,1");
 	struct perf_cpu_map *b = perf_cpu_map__new("4,5,7");
@@ -137,3 +137,7 @@ int test__cpu_map_merge(struct test *test __maybe_unused, int subtest __maybe_un
 	perf_cpu_map__put(c);
 	return 0;
 }
+
+DEFINE_SUITE("Synthesize cpu map", cpu_map_synthesize);
+DEFINE_SUITE("Print cpu map", cpu_map_print);
+DEFINE_SUITE("Merge cpu map", cpu_map_merge);
diff --git a/tools/perf/tests/demangle-java-test.c b/tools/perf/tests/demangle-java-test.c
index 8f3b90832fb0..44d1be303b67 100644
--- a/tools/perf/tests/demangle-java-test.c
+++ b/tools/perf/tests/demangle-java-test.c
@@ -7,7 +7,7 @@
 #include "debug.h"
 #include "demangle-java.h"
 
-int test__demangle_java(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__demangle_java(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret = TEST_OK;
 	char *buf = NULL;
@@ -40,3 +40,5 @@ int test__demangle_java(struct test *test __maybe_unused, int subtest __maybe_un
 
 	return ret;
 }
+
+DEFINE_SUITE("Demangle Java", demangle_java);
diff --git a/tools/perf/tests/demangle-ocaml-test.c b/tools/perf/tests/demangle-ocaml-test.c
index 0043be812355..90a4285e2ad5 100644
--- a/tools/perf/tests/demangle-ocaml-test.c
+++ b/tools/perf/tests/demangle-ocaml-test.c
@@ -7,7 +7,7 @@
 #include "debug.h"
 #include "demangle-ocaml.h"
 
-int test__demangle_ocaml(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__demangle_ocaml(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret = TEST_OK;
 	char *buf = NULL;
@@ -41,3 +41,5 @@ int test__demangle_ocaml(struct test *test __maybe_unused, int subtest __maybe_u
 
 	return ret;
 }
+
+DEFINE_SUITE("Demangle OCaml", demangle_ocaml);
diff --git a/tools/perf/tests/dlfilter-test.c b/tools/perf/tests/dlfilter-test.c
index bc03b5df6828..84352d55347d 100644
--- a/tools/perf/tests/dlfilter-test.c
+++ b/tools/perf/tests/dlfilter-test.c
@@ -398,7 +398,7 @@ static void test_data__free(struct test_data *td)
 	}
 }
 
-int test__dlfilter(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__dlfilter(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct test_data td = {.fd = -1};
 	int pid = getpid();
@@ -414,3 +414,5 @@ int test__dlfilter(struct test *test __maybe_unused, int subtest __maybe_unused)
 	test_data__free(&td);
 	return err;
 }
+
+DEFINE_SUITE("dlfilter C API", dlfilter);
diff --git a/tools/perf/tests/dso-data.c b/tools/perf/tests/dso-data.c
index 43e1b01e5afc..3419a4ab5590 100644
--- a/tools/perf/tests/dso-data.c
+++ b/tools/perf/tests/dso-data.c
@@ -113,7 +113,7 @@ static int dso__data_fd(struct dso *dso, struct machine *machine)
 	return fd;
 }
 
-int test__dso_data(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__dso_data(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct machine machine;
 	struct dso *dso;
@@ -248,7 +248,7 @@ static int set_fd_limit(int n)
 	return setrlimit(RLIMIT_NOFILE, &rlim);
 }
 
-int test__dso_data_cache(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__dso_data_cache(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct machine machine;
 	long nr_end, nr = open_files_cnt();
@@ -318,7 +318,7 @@ static long new_limit(int count)
 	return ret;
 }
 
-int test__dso_data_reopen(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__dso_data_reopen(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct machine machine;
 	long nr_end, nr = open_files_cnt(), lim = new_limit(3);
@@ -393,3 +393,7 @@ int test__dso_data_reopen(struct test *test __maybe_unused, int subtest __maybe_
 	TEST_ASSERT_VAL("failed leaking files", nr == nr_end);
 	return 0;
 }
+
+DEFINE_SUITE("DSO data read", dso_data);
+DEFINE_SUITE("DSO data cache", dso_data_cache);
+DEFINE_SUITE("DSO data reopen", dso_data_reopen);
diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf-unwind.c
index c756284b3b13..2dab2d262060 100644
--- a/tools/perf/tests/dwarf-unwind.c
+++ b/tools/perf/tests/dwarf-unwind.c
@@ -195,7 +195,8 @@ NO_TAIL_CALL_ATTRIBUTE noinline int test_dwarf_unwind__krava_1(struct thread *th
 	return ret;
 }
 
-int test__dwarf_unwind(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__dwarf_unwind(struct test_suite *test __maybe_unused,
+			      int subtest __maybe_unused)
 {
 	struct machine *machine;
 	struct thread *thread;
@@ -237,3 +238,5 @@ int test__dwarf_unwind(struct test *test __maybe_unused, int subtest __maybe_unu
 	machine__delete(machine);
 	return err;
 }
+
+DEFINE_SUITE("Test dwarf unwind", dwarf_unwind);
diff --git a/tools/perf/tests/event-times.c b/tools/perf/tests/event-times.c
index 04ce4401f775..7606eb3df92f 100644
--- a/tools/perf/tests/event-times.c
+++ b/tools/perf/tests/event-times.c
@@ -216,7 +216,7 @@ out_err:
  * and checks that enabled and running times
  * match.
  */
-int test__event_times(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__event_times(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err, ret = 0;
 
@@ -239,3 +239,5 @@ int test__event_times(struct test *test __maybe_unused, int subtest __maybe_unus
 #undef _T
 	return ret;
 }
+
+DEFINE_SUITE("Event times", event_times);
diff --git a/tools/perf/tests/event_update.c b/tools/perf/tests/event_update.c
index 44a50527f9d9..fbb68deba59f 100644
--- a/tools/perf/tests/event_update.c
+++ b/tools/perf/tests/event_update.c
@@ -83,7 +83,7 @@ static int process_event_cpus(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int test__event_update(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__event_update(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct evsel *evsel;
 	struct event_name tmp;
@@ -123,3 +123,5 @@ int test__event_update(struct test *test __maybe_unused, int subtest __maybe_unu
 	evlist__delete(evlist);
 	return 0;
 }
+
+DEFINE_SUITE("Synthesize attr update", event_update);
diff --git a/tools/perf/tests/evsel-roundtrip-name.c b/tools/perf/tests/evsel-roundtrip-name.c
index 4e09f0a312af..fdbf17642e45 100644
--- a/tools/perf/tests/evsel-roundtrip-name.c
+++ b/tools/perf/tests/evsel-roundtrip-name.c
@@ -99,7 +99,8 @@ out_delete_evlist:
 #define perf_evsel__name_array_test(names, distance) \
 	__perf_evsel__name_array_test(names, ARRAY_SIZE(names), distance)
 
-int test__perf_evsel__roundtrip_name_test(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__perf_evsel__roundtrip_name_test(struct test_suite *test __maybe_unused,
+						 int subtest __maybe_unused)
 {
 	int err = 0, ret = 0;
 
@@ -120,3 +121,5 @@ int test__perf_evsel__roundtrip_name_test(struct test *test __maybe_unused, int
 
 	return ret;
 }
+
+DEFINE_SUITE("Roundtrip evsel->name", perf_evsel__roundtrip_name_test);
diff --git a/tools/perf/tests/evsel-tp-sched.c b/tools/perf/tests/evsel-tp-sched.c
index f9e34bd26cf3..cf4da3d748c2 100644
--- a/tools/perf/tests/evsel-tp-sched.c
+++ b/tools/perf/tests/evsel-tp-sched.c
@@ -32,7 +32,8 @@ static int evsel__test_field(struct evsel *evsel, const char *name, int size, bo
 	return ret;
 }
 
-int test__perf_evsel__tp_sched_test(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__perf_evsel__tp_sched_test(struct test_suite *test __maybe_unused,
+					   int subtest __maybe_unused)
 {
 	struct evsel *evsel = evsel__newtp("sched", "sched_switch");
 	int ret = 0;
@@ -87,3 +88,5 @@ int test__perf_evsel__tp_sched_test(struct test *test __maybe_unused, int subtes
 	evsel__delete(evsel);
 	return ret;
 }
+
+DEFINE_SUITE("Parse sched tracepoints fields", perf_evsel__tp_sched_test);
diff --git a/tools/perf/tests/expand-cgroup.c b/tools/perf/tests/expand-cgroup.c
index 80cff8a3558c..dfefe5b60eb2 100644
--- a/tools/perf/tests/expand-cgroup.c
+++ b/tools/perf/tests/expand-cgroup.c
@@ -221,8 +221,8 @@ out:
 	return ret;
 }
 
-int test__expand_cgroup_events(struct test *test __maybe_unused,
-			       int subtest __maybe_unused)
+static int test__expand_cgroup_events(struct test_suite *test __maybe_unused,
+				      int subtest __maybe_unused)
 {
 	int ret;
 
@@ -240,3 +240,5 @@ int test__expand_cgroup_events(struct test *test __maybe_unused,
 
 	return ret;
 }
+
+DEFINE_SUITE("Event expansion for cgroups", expand_cgroup_events);
diff --git a/tools/perf/tests/expr.c b/tools/perf/tests/expr.c
index 077783223ce0..c895de481fe1 100644
--- a/tools/perf/tests/expr.c
+++ b/tools/perf/tests/expr.c
@@ -62,11 +62,11 @@ static int test(struct expr_parse_ctx *ctx, const char *e, double val2)
 	return 0;
 }
 
-int test__expr(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__expr(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	struct expr_id_data *val_ptr;
 	const char *p;
-	double val;
+	double val, num_cpus, num_cores, num_dies, num_packages;
 	int ret;
 	struct expr_parse_ctx *ctx;
 
@@ -134,6 +134,16 @@ int test__expr(struct test *t __maybe_unused, int subtest __maybe_unused)
 	TEST_ASSERT_VAL("find ids", hashmap__find(ctx->ids, "EVENT2,param=3@",
 						    (void **)&val_ptr));
 
+	expr__ctx_clear(ctx);
+	TEST_ASSERT_VAL("find ids",
+			expr__find_ids("dash\\-event1 - dash\\-event2",
+				       NULL, ctx) == 0);
+	TEST_ASSERT_VAL("find ids", hashmap__size(ctx->ids) == 2);
+	TEST_ASSERT_VAL("find ids", hashmap__find(ctx->ids, "dash-event1",
+						    (void **)&val_ptr));
+	TEST_ASSERT_VAL("find ids", hashmap__find(ctx->ids, "dash-event2",
+						    (void **)&val_ptr));
+
 	/* Only EVENT1 or EVENT2 need be measured depending on the value of smt_on. */
 	expr__ctx_clear(ctx);
 	TEST_ASSERT_VAL("find ids",
@@ -151,7 +161,31 @@ int test__expr(struct test *t __maybe_unused, int subtest __maybe_unused)
 			NULL, ctx) == 0);
 	TEST_ASSERT_VAL("find ids", hashmap__size(ctx->ids) == 0);
 
+	/* Test toplogy constants appear well ordered. */
+	expr__ctx_clear(ctx);
+	TEST_ASSERT_VAL("#num_cpus", expr__parse(&num_cpus, ctx, "#num_cpus") == 0);
+	TEST_ASSERT_VAL("#num_cores", expr__parse(&num_cores, ctx, "#num_cores") == 0);
+	TEST_ASSERT_VAL("#num_cpus >= #num_cores", num_cpus >= num_cores);
+	TEST_ASSERT_VAL("#num_dies", expr__parse(&num_dies, ctx, "#num_dies") == 0);
+	TEST_ASSERT_VAL("#num_cores >= #num_dies", num_cores >= num_dies);
+	TEST_ASSERT_VAL("#num_packages", expr__parse(&num_packages, ctx, "#num_packages") == 0);
+	TEST_ASSERT_VAL("#num_dies >= #num_packages", num_dies >= num_packages);
+
+	/*
+	 * Source count returns the number of events aggregating in a leader
+	 * event including the leader. Check parsing yields an id.
+	 */
+	expr__ctx_clear(ctx);
+	TEST_ASSERT_VAL("source count",
+			expr__find_ids("source_count(EVENT1)",
+			NULL, ctx) == 0);
+	TEST_ASSERT_VAL("source count", hashmap__size(ctx->ids) == 1);
+	TEST_ASSERT_VAL("source count", hashmap__find(ctx->ids, "EVENT1",
+							(void **)&val_ptr));
+
 	expr__ctx_free(ctx);
 
 	return 0;
 }
+
+DEFINE_SUITE("Simple expression parser", expr);
diff --git a/tools/perf/tests/fdarray.c b/tools/perf/tests/fdarray.c
index d9eca8e86a6b..40983c3574b1 100644
--- a/tools/perf/tests/fdarray.c
+++ b/tools/perf/tests/fdarray.c
@@ -28,7 +28,7 @@ static int fdarray__fprintf_prefix(struct fdarray *fda, const char *prefix, FILE
 	return printed + fdarray__fprintf(fda, fp);
 }
 
-int test__fdarray__filter(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__fdarray__filter(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int nr_fds, err = TEST_FAIL;
 	struct fdarray *fda = fdarray__new(5, 5);
@@ -89,7 +89,7 @@ out:
 	return err;
 }
 
-int test__fdarray__add(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__fdarray__add(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = TEST_FAIL;
 	struct fdarray *fda = fdarray__new(2, 2);
@@ -158,3 +158,6 @@ out_delete:
 out:
 	return err;
 }
+
+DEFINE_SUITE("Filter fds with revents mask in a fdarray", fdarray__filter);
+DEFINE_SUITE("Add fd to a fdarray, making it autogrow", fdarray__add);
diff --git a/tools/perf/tests/genelf.c b/tools/perf/tests/genelf.c
index f797f9823e89..95f3be1b683a 100644
--- a/tools/perf/tests/genelf.c
+++ b/tools/perf/tests/genelf.c
@@ -16,8 +16,8 @@
 
 #define TEMPL "/tmp/perf-test-XXXXXX"
 
-int test__jit_write_elf(struct test *test __maybe_unused,
-			int subtest __maybe_unused)
+static int test__jit_write_elf(struct test_suite *test __maybe_unused,
+			       int subtest __maybe_unused)
 {
 #ifdef HAVE_JITDUMP
 	static unsigned char x86_code[] = {
@@ -49,3 +49,5 @@ int test__jit_write_elf(struct test *test __maybe_unused,
 	return TEST_SKIP;
 #endif
 }
+
+DEFINE_SUITE("Test jit_write_elf", jit_write_elf);
diff --git a/tools/perf/tests/hists_cumulate.c b/tools/perf/tests/hists_cumulate.c
index 890cb1f5bf53..17f4fcd6bdce 100644
--- a/tools/perf/tests/hists_cumulate.c
+++ b/tools/perf/tests/hists_cumulate.c
@@ -689,7 +689,7 @@ out:
 	return err;
 }
 
-int test__hists_cumulate(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__hists_cumulate(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = TEST_FAIL;
 	struct machines machines;
@@ -736,3 +736,5 @@ out:
 
 	return err;
 }
+
+DEFINE_SUITE("Cumulate child hist entries", hists_cumulate);
diff --git a/tools/perf/tests/hists_filter.c b/tools/perf/tests/hists_filter.c
index ca6120cd1d90..08cbeb9e39ae 100644
--- a/tools/perf/tests/hists_filter.c
+++ b/tools/perf/tests/hists_filter.c
@@ -101,7 +101,7 @@ out:
 	return TEST_FAIL;
 }
 
-int test__hists_filter(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__hists_filter(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = TEST_FAIL;
 	struct machines machines;
@@ -325,3 +325,5 @@ out:
 
 	return err;
 }
+
+DEFINE_SUITE("Filter hist entries", hists_filter);
diff --git a/tools/perf/tests/hists_link.c b/tools/perf/tests/hists_link.c
index a024d3f3a412..c575e13a850d 100644
--- a/tools/perf/tests/hists_link.c
+++ b/tools/perf/tests/hists_link.c
@@ -264,7 +264,7 @@ static int validate_link(struct hists *leader, struct hists *other)
 	return __validate_link(leader, 0) || __validate_link(other, 1);
 }
 
-int test__hists_link(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__hists_link(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = -1;
 	struct hists *hists, *first_hists;
@@ -339,3 +339,5 @@ out:
 
 	return err;
 }
+
+DEFINE_SUITE("Match and link multiple hists", hists_link);
diff --git a/tools/perf/tests/hists_output.c b/tools/perf/tests/hists_output.c
index 8973f35df604..0bde4a768c15 100644
--- a/tools/perf/tests/hists_output.c
+++ b/tools/perf/tests/hists_output.c
@@ -575,7 +575,7 @@ out:
 	return err;
 }
 
-int test__hists_output(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__hists_output(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = TEST_FAIL;
 	struct machines machines;
@@ -623,3 +623,5 @@ out:
 
 	return err;
 }
+
+DEFINE_SUITE("Sort output of hist entries", hists_output);
diff --git a/tools/perf/tests/is_printable_array.c b/tools/perf/tests/is_printable_array.c
index 9c7b3baca4fe..f72de2457ff1 100644
--- a/tools/perf/tests/is_printable_array.c
+++ b/tools/perf/tests/is_printable_array.c
@@ -5,7 +5,7 @@
 #include "debug.h"
 #include "print_binary.h"
 
-int test__is_printable_array(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__is_printable_array(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	char buf1[] = { 'k', 'r', 4, 'v', 'a', 0 };
 	char buf2[] = { 'k', 'r', 'a', 'v', 4, 0 };
@@ -36,3 +36,5 @@ int test__is_printable_array(struct test *test __maybe_unused, int subtest __may
 
 	return TEST_OK;
 }
+
+DEFINE_SUITE("is_printable_array", is_printable_array);
diff --git a/tools/perf/tests/keep-tracking.c b/tools/perf/tests/keep-tracking.c
index a0438b0f0805..dd2067312452 100644
--- a/tools/perf/tests/keep-tracking.c
+++ b/tools/perf/tests/keep-tracking.c
@@ -61,7 +61,7 @@ static int find_comm(struct evlist *evlist, const char *comm)
  * when an event is disabled but a dummy software event is not disabled.  If the
  * test passes %0 is returned, otherwise %-1 is returned.
  */
-int test__keep_tracking(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__keep_tracking(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct record_opts opts = {
 		.mmap_pages	     = UINT_MAX,
@@ -160,3 +160,5 @@ out_err:
 
 	return err;
 }
+
+DEFINE_SUITE("Use a dummy software event to keep tracking", keep_tracking);
diff --git a/tools/perf/tests/kmod-path.c b/tools/perf/tests/kmod-path.c
index e483210b176b..dfe1bd5dabaa 100644
--- a/tools/perf/tests/kmod-path.c
+++ b/tools/perf/tests/kmod-path.c
@@ -47,7 +47,7 @@ static int test_is_kernel_module(const char *path, int cpumode, bool expect)
 #define M(path, c, e) \
 	TEST_ASSERT_VAL("failed", !test_is_kernel_module(path, c, e))
 
-int test__kmod_path__parse(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__kmod_path__parse(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	/* path                alloc_name  kmod  comp   name   */
 	T("/xxxx/xxxx/x-x.ko", true      , true, 0    , "[x_x]");
@@ -159,3 +159,5 @@ int test__kmod_path__parse(struct test *t __maybe_unused, int subtest __maybe_un
 
 	return 0;
 }
+
+DEFINE_SUITE("kmod_path__parse", kmod_path__parse);
diff --git a/tools/perf/tests/llvm.c b/tools/perf/tests/llvm.c
index 33e43cce9064..8ac0a3a457ef 100644
--- a/tools/perf/tests/llvm.c
+++ b/tools/perf/tests/llvm.c
@@ -124,7 +124,7 @@ out:
 	return ret;
 }
 
-int test__llvm(struct test *test __maybe_unused, int subtest)
+static int test__llvm(int subtest)
 {
 	int ret;
 	void *obj_buf = NULL;
@@ -148,32 +148,72 @@ int test__llvm(struct test *test __maybe_unused, int subtest)
 
 	return ret;
 }
+#endif //HAVE_LIBBPF_SUPPORT
 
-int test__llvm_subtest_get_nr(void)
+static int test__llvm__bpf_base_prog(struct test_suite *test __maybe_unused,
+				     int subtest __maybe_unused)
 {
-	return __LLVM_TESTCASE_MAX;
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__llvm(LLVM_TESTCASE_BASE);
+#else
+	pr_debug("Skip LLVM test because BPF support is not compiled\n");
+	return TEST_SKIP;
+#endif
 }
 
-const char *test__llvm_subtest_get_desc(int subtest)
-{
-	if ((subtest < 0) || (subtest >= __LLVM_TESTCASE_MAX))
-		return NULL;
-
-	return bpf_source_table[subtest].desc;
-}
-#else //HAVE_LIBBPF_SUPPORT
-int test__llvm(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__llvm__bpf_test_kbuild_prog(struct test_suite *test __maybe_unused,
+					    int subtest __maybe_unused)
 {
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__llvm(LLVM_TESTCASE_KBUILD);
+#else
+	pr_debug("Skip LLVM test because BPF support is not compiled\n");
 	return TEST_SKIP;
+#endif
 }
 
-int test__llvm_subtest_get_nr(void)
+static int test__llvm__bpf_test_prologue_prog(struct test_suite *test __maybe_unused,
+					      int subtest __maybe_unused)
 {
-	return 0;
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__llvm(LLVM_TESTCASE_BPF_PROLOGUE);
+#else
+	pr_debug("Skip LLVM test because BPF support is not compiled\n");
+	return TEST_SKIP;
+#endif
 }
 
-const char *test__llvm_subtest_get_desc(int subtest __maybe_unused)
+static int test__llvm__bpf_test_relocation(struct test_suite *test __maybe_unused,
+					   int subtest __maybe_unused)
 {
-	return NULL;
+#ifdef HAVE_LIBBPF_SUPPORT
+	return test__llvm(LLVM_TESTCASE_BPF_RELOCATION);
+#else
+	pr_debug("Skip LLVM test because BPF support is not compiled\n");
+	return TEST_SKIP;
+#endif
 }
-#endif // HAVE_LIBBPF_SUPPORT
+
+
+static struct test_case llvm_tests[] = {
+#ifdef HAVE_LIBBPF_SUPPORT
+	TEST_CASE("Basic BPF llvm compile", llvm__bpf_base_prog),
+	TEST_CASE("kbuild searching", llvm__bpf_test_kbuild_prog),
+	TEST_CASE("Compile source for BPF prologue generation",
+		  llvm__bpf_test_prologue_prog),
+	TEST_CASE("Compile source for BPF relocation", llvm__bpf_test_relocation),
+#else
+	TEST_CASE_REASON("Basic BPF llvm compile", llvm__bpf_base_prog, "not compiled in"),
+	TEST_CASE_REASON("kbuild searching", llvm__bpf_test_kbuild_prog, "not compiled in"),
+	TEST_CASE_REASON("Compile source for BPF prologue generation",
+			llvm__bpf_test_prologue_prog, "not compiled in"),
+	TEST_CASE_REASON("Compile source for BPF relocation",
+			llvm__bpf_test_relocation, "not compiled in"),
+#endif
+	{ .name = NULL, }
+};
+
+struct test_suite suite__llvm = {
+	.desc = "LLVM search and compile",
+	.test_cases = llvm_tests,
+};
diff --git a/tools/perf/tests/maps.c b/tools/perf/tests/maps.c
index 1ac72919fa35..e308a3296cef 100644
--- a/tools/perf/tests/maps.c
+++ b/tools/perf/tests/maps.c
@@ -33,7 +33,7 @@ static int check_maps(struct map_def *merged, unsigned int size, struct maps *ma
 	return TEST_OK;
 }
 
-int test__maps__merge_in(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__maps__merge_in(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	struct maps maps;
 	unsigned int i;
@@ -120,3 +120,5 @@ int test__maps__merge_in(struct test *t __maybe_unused, int subtest __maybe_unus
 	maps__exit(&maps);
 	return TEST_OK;
 }
+
+DEFINE_SUITE("maps__merge_in", maps__merge_in);
diff --git a/tools/perf/tests/mem.c b/tools/perf/tests/mem.c
index 673a11a6cd1b..56014ec7d49d 100644
--- a/tools/perf/tests/mem.c
+++ b/tools/perf/tests/mem.c
@@ -23,7 +23,7 @@ static int check(union perf_mem_data_src data_src,
 	return 0;
 }
 
-int test__mem(struct test *text __maybe_unused, int subtest __maybe_unused)
+static int test__mem(struct test_suite *text __maybe_unused, int subtest __maybe_unused)
 {
 	int ret = 0;
 	union perf_mem_data_src src;
@@ -56,3 +56,5 @@ int test__mem(struct test *text __maybe_unused, int subtest __maybe_unused)
 
 	return ret;
 }
+
+DEFINE_SUITE("Test data source output", mem);
diff --git a/tools/perf/tests/mem2node.c b/tools/perf/tests/mem2node.c
index e4d0d58b97f8..b17b86391383 100644
--- a/tools/perf/tests/mem2node.c
+++ b/tools/perf/tests/mem2node.c
@@ -43,7 +43,7 @@ static unsigned long *get_bitmap(const char *str, int nbits)
 	return bm && map ? bm : NULL;
 }
 
-int test__mem2node(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__mem2node(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	struct mem2node map;
 	struct memory_node nodes[3];
@@ -77,3 +77,5 @@ int test__mem2node(struct test *t __maybe_unused, int subtest __maybe_unused)
 	mem2node__exit(&map);
 	return 0;
 }
+
+DEFINE_SUITE("mem2node", mem2node);
diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c
index d38757db2dc2..90b2feda31ac 100644
--- a/tools/perf/tests/mmap-basic.c
+++ b/tools/perf/tests/mmap-basic.c
@@ -29,7 +29,7 @@
  * Then it checks if the number of syscalls reported as perf events by
  * the kernel corresponds to the number of syscalls made.
  */
-int test__basic_mmap(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = -1;
 	union perf_event *event;
@@ -164,3 +164,5 @@ out_free_threads:
 	perf_thread_map__put(threads);
 	return err;
 }
+
+DEFINE_SUITE("Read samples using the mmap interface", basic_mmap);
diff --git a/tools/perf/tests/mmap-thread-lookup.c b/tools/perf/tests/mmap-thread-lookup.c
index 6f2da7a72f67..a4301fc7b770 100644
--- a/tools/perf/tests/mmap-thread-lookup.c
+++ b/tools/perf/tests/mmap-thread-lookup.c
@@ -224,7 +224,7 @@ static int mmap_events(synth_cb synth)
  *
  * by using all thread objects.
  */
-int test__mmap_thread_lookup(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__mmap_thread_lookup(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	/* perf_event__synthesize_threads synthesize */
 	TEST_ASSERT_VAL("failed with sythesizing all",
@@ -236,3 +236,5 @@ int test__mmap_thread_lookup(struct test *test __maybe_unused, int subtest __may
 
 	return 0;
 }
+
+DEFINE_SUITE("Lookup mmap thread", mmap_thread_lookup);
diff --git a/tools/perf/tests/openat-syscall-all-cpus.c b/tools/perf/tests/openat-syscall-all-cpus.c
index f7dd6c463f04..cd3dd463783f 100644
--- a/tools/perf/tests/openat-syscall-all-cpus.c
+++ b/tools/perf/tests/openat-syscall-all-cpus.c
@@ -19,7 +19,8 @@
 #include "stat.h"
 #include "util/counts.h"
 
-int test__openat_syscall_event_on_all_cpus(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __maybe_unused,
+						  int subtest __maybe_unused)
 {
 	int err = -1, fd, cpu;
 	struct perf_cpu_map *cpus;
@@ -127,3 +128,5 @@ out_thread_map_delete:
 	perf_thread_map__put(threads);
 	return err;
 }
+
+DEFINE_SUITE("Detect openat syscall event on all cpus", openat_syscall_event_on_all_cpus);
diff --git a/tools/perf/tests/openat-syscall-tp-fields.c b/tools/perf/tests/openat-syscall-tp-fields.c
index 5e4af2f0f14a..a7b2800652e4 100644
--- a/tools/perf/tests/openat-syscall-tp-fields.c
+++ b/tools/perf/tests/openat-syscall-tp-fields.c
@@ -22,7 +22,8 @@
 #define AT_FDCWD       -100
 #endif
 
-int test__syscall_openat_tp_fields(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused,
+					  int subtest __maybe_unused)
 {
 	struct record_opts opts = {
 		.target = {
@@ -142,3 +143,5 @@ out_delete_evlist:
 out:
 	return err;
 }
+
+DEFINE_SUITE("syscalls:sys_enter_openat event fields", syscall_openat_tp_fields);
diff --git a/tools/perf/tests/openat-syscall.c b/tools/perf/tests/openat-syscall.c
index 85a8f0fe7aea..7f4c13c4b14d 100644
--- a/tools/perf/tests/openat-syscall.c
+++ b/tools/perf/tests/openat-syscall.c
@@ -13,7 +13,8 @@
 #include "tests.h"
 #include "util/counts.h"
 
-int test__openat_syscall_event(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__openat_syscall_event(struct test_suite *test __maybe_unused,
+				      int subtest __maybe_unused)
 {
 	int err = -1, fd;
 	struct evsel *evsel;
@@ -66,3 +67,5 @@ out_thread_map_delete:
 	perf_thread_map__put(threads);
 	return err;
 }
+
+DEFINE_SUITE("Detect openat syscall event", openat_syscall_event);
diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c
index 6af94639b14a..a508f1dbcb2a 100644
--- a/tools/perf/tests/parse-events.c
+++ b/tools/perf/tests/parse-events.c
@@ -2276,7 +2276,7 @@ static int test_pmu_events_alias(char *event, char *alias)
 	return test_event(&e);
 }
 
-int test__parse_events(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__parse_events(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret1, ret2 = 0;
 	char *event, *alias;
@@ -2319,3 +2319,5 @@ do {							\
 
 	return ret2;
 }
+
+DEFINE_SUITE("Parse event definition strings", parse_events);
diff --git a/tools/perf/tests/parse-metric.c b/tools/perf/tests/parse-metric.c
index dfc797ecc750..574b7e4efd3a 100644
--- a/tools/perf/tests/parse-metric.c
+++ b/tools/perf/tests/parse-metric.c
@@ -369,7 +369,7 @@ static int test_metric_group(void)
 	return 0;
 }
 
-int test__parse_metric(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__parse_metric(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	TEST_ASSERT_VAL("IPC failed", test_ipc() == 0);
 	TEST_ASSERT_VAL("frontend failed", test_frontend() == 0);
@@ -383,3 +383,5 @@ int test__parse_metric(struct test *test __maybe_unused, int subtest __maybe_unu
 	}
 	return 0;
 }
+
+DEFINE_SUITE("Parse and process metrics", parse_metric);
diff --git a/tools/perf/tests/parse-no-sample-id-all.c b/tools/perf/tests/parse-no-sample-id-all.c
index 471273676701..d62e31595ab2 100644
--- a/tools/perf/tests/parse-no-sample-id-all.c
+++ b/tools/perf/tests/parse-no-sample-id-all.c
@@ -67,7 +67,8 @@ struct test_attr_event {
  *
  * Return: %0 on success, %-1 if the test fails.
  */
-int test__parse_no_sample_id_all(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__parse_no_sample_id_all(struct test_suite *test __maybe_unused,
+					int subtest __maybe_unused)
 {
 	int err;
 
@@ -103,3 +104,5 @@ int test__parse_no_sample_id_all(struct test *test __maybe_unused, int subtest _
 
 	return 0;
 }
+
+DEFINE_SUITE("Parse with no sample_id_all bit set", parse_no_sample_id_all);
diff --git a/tools/perf/tests/pe-file-parsing.c b/tools/perf/tests/pe-file-parsing.c
index 58b90c42eb38..c09a9fae1689 100644
--- a/tools/perf/tests/pe-file-parsing.c
+++ b/tools/perf/tests/pe-file-parsing.c
@@ -68,7 +68,7 @@ static int run_dir(const char *d)
 	return TEST_OK;
 }
 
-int test__pe_file_parsing(struct test *test __maybe_unused,
+static int test__pe_file_parsing(struct test_suite *test __maybe_unused,
 			  int subtest __maybe_unused)
 {
 	struct stat st;
@@ -89,10 +89,12 @@ int test__pe_file_parsing(struct test *test __maybe_unused,
 
 #else
 
-int test__pe_file_parsing(struct test *test __maybe_unused,
+static int test__pe_file_parsing(struct test_suite *test __maybe_unused,
 			  int subtest __maybe_unused)
 {
 	return TEST_SKIP;
 }
 
 #endif
+
+DEFINE_SUITE("PE file support", pe_file_parsing);
diff --git a/tools/perf/tests/perf-hooks.c b/tools/perf/tests/perf-hooks.c
index dd865e0bea12..78cdeb89645e 100644
--- a/tools/perf/tests/perf-hooks.c
+++ b/tools/perf/tests/perf-hooks.c
@@ -26,7 +26,7 @@ static void the_hook(void *_hook_flags)
 	raise(SIGSEGV);
 }
 
-int test__perf_hooks(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__perf_hooks(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int hook_flags = 0;
 
@@ -45,3 +45,5 @@ int test__perf_hooks(struct test *test __maybe_unused, int subtest __maybe_unuse
 		return TEST_FAIL;
 	return TEST_OK;
 }
+
+DEFINE_SUITE("perf hooks", perf_hooks);
diff --git a/tools/perf/tests/perf-record.c b/tools/perf/tests/perf-record.c
index 0df471bf1590..6354465067b8 100644
--- a/tools/perf/tests/perf-record.c
+++ b/tools/perf/tests/perf-record.c
@@ -41,7 +41,7 @@ realloc:
 	return cpu;
 }
 
-int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct record_opts opts = {
 		.target = {
@@ -332,3 +332,5 @@ out_delete_evlist:
 out:
 	return (err < 0 || errs > 0) ? -1 : 0;
 }
+
+DEFINE_SUITE("PERF_RECORD_* events & perf_sample fields", PERF_RECORD);
diff --git a/tools/perf/tests/perf-time-to-tsc.c b/tools/perf/tests/perf-time-to-tsc.c
index 7c56bc1f4cff..d12d0ad81801 100644
--- a/tools/perf/tests/perf-time-to-tsc.c
+++ b/tools/perf/tests/perf-time-to-tsc.c
@@ -23,6 +23,16 @@
 #include "pmu.h"
 #include "pmu-hybrid.h"
 
+/*
+ * Except x86_64/i386 and Arm64, other archs don't support TSC in perf.  Just
+ * enable the test for x86_64/i386 and Arm64 archs.
+ */
+#if defined(__x86_64__) || defined(__i386__) || defined(__aarch64__)
+#define TSC_IS_SUPPORTED 1
+#else
+#define TSC_IS_SUPPORTED 0
+#endif
+
 #define CHECK__(x) {				\
 	while ((x) < 0) {			\
 		pr_debug(#x " failed!\n");	\
@@ -45,7 +55,7 @@
  * %0 is returned, otherwise %-1 is returned.  If TSC conversion is not
  * supported then then the test passes but " (not supported)" is printed.
  */
-int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct record_opts opts = {
 		.mmap_pages	     = UINT_MAX,
@@ -69,6 +79,11 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybe
 	u64 test_time, comm1_time = 0, comm2_time = 0;
 	struct mmap *md;
 
+	if (!TSC_IS_SUPPORTED) {
+		pr_debug("Test not supported on this architecture");
+		return TEST_SKIP;
+	}
+
 	threads = thread_map__new(-1, getpid(), UINT_MAX);
 	CHECK_NOT_NULL__(threads);
 
@@ -185,15 +200,4 @@ out_err:
 	return err;
 }
 
-bool test__tsc_is_supported(void)
-{
-	/*
-	 * Except x86_64/i386 and Arm64, other archs don't support TSC in perf.
-	 * Just enable the test for x86_64/i386 and Arm64 archs.
-	 */
-#if defined(__x86_64__) || defined(__i386__) || defined(__aarch64__)
-	return true;
-#else
-	return false;
-#endif
-}
+DEFINE_SUITE("Convert perf time to TSC", perf_time_to_tsc);
diff --git a/tools/perf/tests/pfm.c b/tools/perf/tests/pfm.c
index e8fd0da0762b..71b76deb1f92 100644
--- a/tools/perf/tests/pfm.c
+++ b/tools/perf/tests/pfm.c
@@ -12,27 +12,6 @@
 #include <linux/kernel.h>
 
 #ifdef HAVE_LIBPFM
-static int test__pfm_events(void);
-static int test__pfm_group(void);
-#endif
-
-static const struct {
-	int (*func)(void);
-	const char *desc;
-} pfm_testcase_table[] = {
-#ifdef HAVE_LIBPFM
-	{
-		.func = test__pfm_events,
-		.desc = "test of individual --pfm-events",
-	},
-	{
-		.func = test__pfm_group,
-		.desc = "test groups of --pfm-events",
-	},
-#endif
-};
-
-#ifdef HAVE_LIBPFM
 static int count_pfm_events(struct perf_evlist *evlist)
 {
 	struct perf_evsel *evsel;
@@ -44,7 +23,8 @@ static int count_pfm_events(struct perf_evlist *evlist)
 	return count;
 }
 
-static int test__pfm_events(void)
+static int test__pfm_events(struct test_suite *test __maybe_unused,
+			    int subtest __maybe_unused)
 {
 	struct evlist *evlist;
 	struct option opt;
@@ -104,7 +84,8 @@ static int test__pfm_events(void)
 	return 0;
 }
 
-static int test__pfm_group(void)
+static int test__pfm_group(struct test_suite *test __maybe_unused,
+			   int subtest __maybe_unused)
 {
 	struct evlist *evlist;
 	struct option opt;
@@ -187,27 +168,27 @@ static int test__pfm_group(void)
 	}
 	return 0;
 }
-#endif
-
-const char *test__pfm_subtest_get_desc(int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(pfm_testcase_table))
-		return NULL;
-	return pfm_testcase_table[i].desc;
-}
-
-int test__pfm_subtest_get_nr(void)
+#else
+static int test__pfm_events(struct test_suite *test __maybe_unused,
+			    int subtest __maybe_unused)
 {
-	return (int)ARRAY_SIZE(pfm_testcase_table);
+	return TEST_SKIP;
 }
 
-int test__pfm(struct test *test __maybe_unused, int i __maybe_unused)
+static int test__pfm_group(struct test_suite *test __maybe_unused,
+			   int subtest __maybe_unused)
 {
-#ifdef HAVE_LIBPFM
-	if (i < 0 || i >= (int)ARRAY_SIZE(pfm_testcase_table))
-		return TEST_FAIL;
-	return pfm_testcase_table[i].func();
-#else
 	return TEST_SKIP;
-#endif
 }
+#endif
+
+static struct test_case pfm_tests[] = {
+	TEST_CASE_REASON("test of individual --pfm-events", pfm_events, "not compiled in"),
+	TEST_CASE_REASON("test groups of --pfm-events", pfm_group, "not compiled in"),
+	{ .name = NULL, }
+};
+
+struct test_suite suite__pfm = {
+	.desc = "Test libpfm4 support",
+	.test_cases = pfm_tests,
+};
diff --git a/tools/perf/tests/pmu-events.c b/tools/perf/tests/pmu-events.c
index 9ae894c406d8..df1c9a3cc05b 100644
--- a/tools/perf/tests/pmu-events.c
+++ b/tools/perf/tests/pmu-events.c
@@ -418,7 +418,8 @@ static int compare_alias_to_test_event(struct perf_pmu_alias *alias,
 }
 
 /* Verify generated events from pmu-events.c are as expected */
-static int test_pmu_event_table(void)
+static int test__pmu_event_table(struct test_suite *test __maybe_unused,
+				 int subtest __maybe_unused)
 {
 	const struct pmu_event *sys_event_tables = __test_pmu_get_sys_events_table();
 	const struct pmu_events_map *map = __test_pmu_get_events_map();
@@ -705,7 +706,8 @@ static struct perf_pmu_test_pmu test_pmus[] = {
 };
 
 /* Test that aliases generated are as expected */
-static int test_aliases(void)
+static int test__aliases(struct test_suite *test __maybe_unused,
+			int subtest __maybe_unused)
 {
 	struct perf_pmu *pmu = NULL;
 	unsigned long i;
@@ -892,7 +894,8 @@ out_err:
 
 }
 
-static int test_parsing(void)
+static int test__parsing(struct test_suite *test __maybe_unused,
+			 int subtest __maybe_unused)
 {
 	const struct pmu_events_map *cpus_map = pmu_events_map__find();
 	const struct pmu_events_map *map;
@@ -1034,7 +1037,8 @@ out:
  * or all defined cpus via the 'fake_pmu'
  * in parse_events.
  */
-static int test_parsing_fake(void)
+static int test__parsing_fake(struct test_suite *test __maybe_unused,
+			      int subtest __maybe_unused)
 {
 	const struct pmu_events_map *map;
 	const struct pmu_event *pe;
@@ -1068,55 +1072,16 @@ static int test_parsing_fake(void)
 	return 0;
 }
 
-static const struct {
-	int (*func)(void);
-	const char *desc;
-} pmu_events_testcase_table[] = {
-	{
-		.func = test_pmu_event_table,
-		.desc = "PMU event table sanity",
-	},
-	{
-		.func = test_aliases,
-		.desc = "PMU event map aliases",
-	},
-	{
-		.func = test_parsing,
-		.desc = "Parsing of PMU event table metrics",
-	},
-	{
-		.func = test_parsing_fake,
-		.desc = "Parsing of PMU event table metrics with fake PMUs",
-	},
+static struct test_case pmu_events_tests[] = {
+	TEST_CASE("PMU event table sanity", pmu_event_table),
+	TEST_CASE("PMU event map aliases", aliases),
+	TEST_CASE_REASON("Parsing of PMU event table metrics", parsing,
+			 "some metrics failed"),
+	TEST_CASE("Parsing of PMU event table metrics with fake PMUs", parsing_fake),
+	{ .name = NULL, }
 };
 
-const char *test__pmu_events_subtest_get_desc(int subtest)
-{
-	if (subtest < 0 ||
-	    subtest >= (int)ARRAY_SIZE(pmu_events_testcase_table))
-		return NULL;
-	return pmu_events_testcase_table[subtest].desc;
-}
-
-const char *test__pmu_events_subtest_skip_reason(int subtest)
-{
-	if (subtest < 0 ||
-	    subtest >= (int)ARRAY_SIZE(pmu_events_testcase_table))
-		return NULL;
-	if (pmu_events_testcase_table[subtest].func != test_parsing)
-		return NULL;
-	return "some metrics failed";
-}
-
-int test__pmu_events_subtest_get_nr(void)
-{
-	return (int)ARRAY_SIZE(pmu_events_testcase_table);
-}
-
-int test__pmu_events(struct test *test __maybe_unused, int subtest)
-{
-	if (subtest < 0 ||
-	    subtest >= (int)ARRAY_SIZE(pmu_events_testcase_table))
-		return TEST_FAIL;
-	return pmu_events_testcase_table[subtest].func();
-}
+struct test_suite suite__pmu_events = {
+	.desc = "PMU events",
+	.test_cases = pmu_events_tests,
+};
diff --git a/tools/perf/tests/pmu.c b/tools/perf/tests/pmu.c
index 714e6830a758..8507bd615e97 100644
--- a/tools/perf/tests/pmu.c
+++ b/tools/perf/tests/pmu.c
@@ -137,7 +137,7 @@ static struct list_head *test_terms_list(void)
 	return &terms;
 }
 
-int test__pmu(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__pmu(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	char *format = test_format_dir_get();
 	LIST_HEAD(formats);
@@ -177,3 +177,5 @@ int test__pmu(struct test *test __maybe_unused, int subtest __maybe_unused)
 	test_format_dir_put(format);
 	return ret;
 }
+
+DEFINE_SUITE("Parse perf pmu format", pmu);
diff --git a/tools/perf/tests/python-use.c b/tools/perf/tests/python-use.c
index 98c6d474aa6f..6b990ee38575 100644
--- a/tools/perf/tests/python-use.c
+++ b/tools/perf/tests/python-use.c
@@ -9,7 +9,7 @@
 #include "tests.h"
 #include "util/debug.h"
 
-int test__python_use(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__python_use(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	char *cmd;
 	int ret;
@@ -23,3 +23,5 @@ int test__python_use(struct test *test __maybe_unused, int subtest __maybe_unuse
 	free(cmd);
 	return ret;
 }
+
+DEFINE_SUITE("'import perf' in python", python_use);
diff --git a/tools/perf/tests/sample-parsing.c b/tools/perf/tests/sample-parsing.c
index c83a11514129..b669d22f2b13 100644
--- a/tools/perf/tests/sample-parsing.c
+++ b/tools/perf/tests/sample-parsing.c
@@ -368,7 +368,7 @@ out_free:
  * checks sample format bits separately and together.  If the test passes %0 is
  * returned, otherwise %-1 is returned.
  */
-int test__sample_parsing(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__sample_parsing(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	const u64 rf[] = {4, 5, 6, 7, 12, 13, 14, 15};
 	u64 sample_type;
@@ -426,3 +426,5 @@ int test__sample_parsing(struct test *test __maybe_unused, int subtest __maybe_u
 
 	return 0;
 }
+
+DEFINE_SUITE("Sample parsing", sample_parsing);
diff --git a/tools/perf/tests/sdt.c b/tools/perf/tests/sdt.c
index ed76c693f65e..919712899251 100644
--- a/tools/perf/tests/sdt.c
+++ b/tools/perf/tests/sdt.c
@@ -76,7 +76,7 @@ static int search_cached_probe(const char *target,
 	return ret;
 }
 
-int test__sdt_event(struct test *test __maybe_unused, int subtests __maybe_unused)
+static int test__sdt_event(struct test_suite *test __maybe_unused, int subtests __maybe_unused)
 {
 	int ret = TEST_FAIL;
 	char __tempdir[] = "./test-buildid-XXXXXX";
@@ -114,9 +114,11 @@ error:
 	return ret;
 }
 #else
-int test__sdt_event(struct test *test __maybe_unused, int subtests __maybe_unused)
+static int test__sdt_event(struct test_suite *test __maybe_unused, int subtests __maybe_unused)
 {
 	pr_debug("Skip SDT event test because SDT support is not compiled\n");
 	return TEST_SKIP;
 }
 #endif
+
+DEFINE_SUITE("Probe SDT events", sdt_event);
diff --git a/tools/perf/tests/shell/record+zstd_comp_decomp.sh b/tools/perf/tests/shell/record+zstd_comp_decomp.sh
index 8a168cf8bacc..49bd875d5122 100755
--- a/tools/perf/tests/shell/record+zstd_comp_decomp.sh
+++ b/tools/perf/tests/shell/record+zstd_comp_decomp.sh
@@ -12,7 +12,7 @@ skip_if_no_z_record() {
 
 collect_z_record() {
 	echo "Collecting compressed record file:"
-	[[ "$(uname -m)" != s390x ]] && gflag='-g'
+	[ "$(uname -m)" != s390x ] && gflag='-g'
 	$perf_tool record -o $trace_file $gflag -z -F 5000 -- \
 		dd count=500 if=/dev/urandom of=/dev/null
 }
diff --git a/tools/perf/tests/shell/stat_all_pmu.sh b/tools/perf/tests/shell/stat_all_pmu.sh
index 2de7fd0394fd..b30dba455f36 100755
--- a/tools/perf/tests/shell/stat_all_pmu.sh
+++ b/tools/perf/tests/shell/stat_all_pmu.sh
@@ -7,11 +7,11 @@ set -e
 for p in $(perf list --raw-dump pmu); do
   echo "Testing $p"
   result=$(perf stat -e "$p" true 2>&1)
-  if [[ ! "$result" =~ "$p" ]] && [[ ! "$result" =~ "<not supported>" ]]; then
+  if ! echo "$result" | grep -q "$p" && ! echo "$result" | grep -q "<not supported>" ; then
     # We failed to see the event and it is supported. Possibly the workload was
     # too small so retry with something longer.
     result=$(perf stat -e "$p" perf bench internals synthesize 2>&1)
-    if [[ ! "$result" =~ "$p" ]]; then
+    if ! echo "$result" | grep -q "$p" ; then
       echo "Event '$p' not printed in:"
       echo "$result"
       exit 1
diff --git a/tools/perf/tests/shell/stat_bpf_counters.sh b/tools/perf/tests/shell/stat_bpf_counters.sh
index 2aed20dc2262..13473aeba489 100755
--- a/tools/perf/tests/shell/stat_bpf_counters.sh
+++ b/tools/perf/tests/shell/stat_bpf_counters.sh
@@ -23,7 +23,7 @@ compare_number()
 
 # skip if --bpf-counters is not supported
 if ! perf stat --bpf-counters true > /dev/null 2>&1; then
-	if [ "$1" == "-v" ]; then
+	if [ "$1" = "-v" ]; then
 		echo "Skipping: --bpf-counters not supported"
 		perf --no-pager stat --bpf-counters true || true
 	fi
diff --git a/tools/perf/tests/shell/test_arm_spe.sh b/tools/perf/tests/shell/test_arm_spe.sh
new file mode 100755
index 000000000000..e59044edc406
--- /dev/null
+++ b/tools/perf/tests/shell/test_arm_spe.sh
@@ -0,0 +1,89 @@
+#!/bin/sh
+# Check Arm SPE trace data recording and synthesized samples
+
+# Uses the 'perf record' to record trace data of Arm SPE events;
+# then verify if any SPE event samples are generated by SPE with
+# 'perf script' and 'perf report' commands.
+
+# SPDX-License-Identifier: GPL-2.0
+# German Gomez <german.gomez@arm.com>, 2021
+
+skip_if_no_arm_spe_event() {
+	perf list | egrep -q 'arm_spe_[0-9]+//' && return 0
+
+	# arm_spe event doesn't exist
+	return 2
+}
+
+skip_if_no_arm_spe_event || exit 2
+
+perfdata=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
+glb_err=0
+
+cleanup_files()
+{
+	rm -f ${perfdata}
+	exit $glb_err
+}
+
+trap cleanup_files exit term int
+
+arm_spe_report() {
+	if [ $2 != 0 ]; then
+		echo "$1: FAIL"
+		glb_err=$2
+	else
+		echo "$1: PASS"
+	fi
+}
+
+perf_script_samples() {
+	echo "Looking at perf.data file for dumping samples:"
+
+	# from arm-spe.c/arm_spe_synth_events()
+	events="(ld1-miss|ld1-access|llc-miss|lld-access|tlb-miss|tlb-access|branch-miss|remote-access|memory)"
+
+	# Below is an example of the samples dumping:
+	#	dd  3048 [002]          1    l1d-access:      ffffaa64999c __GI___libc_write+0x3c (/lib/aarch64-linux-gnu/libc-2.27.so)
+	#	dd  3048 [002]          1    tlb-access:      ffffaa64999c __GI___libc_write+0x3c (/lib/aarch64-linux-gnu/libc-2.27.so)
+	#	dd  3048 [002]          1        memory:      ffffaa64999c __GI___libc_write+0x3c (/lib/aarch64-linux-gnu/libc-2.27.so)
+	perf script -F,-time -i ${perfdata} 2>&1 | \
+		egrep " +$1 +[0-9]+ .* +${events}:(.*:)? +" > /dev/null 2>&1
+}
+
+perf_report_samples() {
+	echo "Looking at perf.data file for reporting samples:"
+
+	# Below is an example of the samples reporting:
+	#   73.04%    73.04%  dd    libc-2.27.so      [.] _dl_addr
+	#    7.71%     7.71%  dd    libc-2.27.so      [.] getenv
+	#    2.59%     2.59%  dd    ld-2.27.so        [.] strcmp
+	perf report --stdio -i ${perfdata} 2>&1 | \
+		egrep " +[0-9]+\.[0-9]+% +[0-9]+\.[0-9]+% +$1 " > /dev/null 2>&1
+}
+
+arm_spe_snapshot_test() {
+	echo "Recording trace with snapshot mode $perfdata"
+	perf record -o ${perfdata} -e arm_spe// -S \
+		-- dd if=/dev/zero of=/dev/null > /dev/null 2>&1 &
+	PERFPID=$!
+
+	# Wait for perf program
+	sleep 1
+
+	# Send signal to snapshot trace data
+	kill -USR2 $PERFPID
+
+	# Stop perf program
+	kill $PERFPID
+	wait $PERFPID
+
+	perf_script_samples dd &&
+	perf_report_samples dd
+
+	err=$?
+	arm_spe_report "SPE snapshot testing" $err
+}
+
+arm_spe_snapshot_test
+exit $glb_err
diff --git a/tools/perf/tests/stat.c b/tools/perf/tests/stat.c
index c1911501c39c..2eb096b5e6da 100644
--- a/tools/perf/tests/stat.c
+++ b/tools/perf/tests/stat.c
@@ -47,7 +47,8 @@ static int process_stat_config_event(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int test__synthesize_stat_config(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__synthesize_stat_config(struct test_suite *test __maybe_unused,
+					int subtest __maybe_unused)
 {
 	struct perf_stat_config stat_config = {
 		.aggr_mode	= AGGR_CORE,
@@ -77,7 +78,7 @@ static int process_stat_event(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int test__synthesize_stat(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__synthesize_stat(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_counts_values count;
 
@@ -103,7 +104,7 @@ static int process_stat_round_event(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int test__synthesize_stat_round(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__synthesize_stat_round(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	TEST_ASSERT_VAL("failed to synthesize stat_config",
 		!perf_event__synthesize_stat_round(NULL, 0xdeadbeef, PERF_STAT_ROUND_TYPE__INTERVAL,
@@ -111,3 +112,7 @@ int test__synthesize_stat_round(struct test *test __maybe_unused, int subtest __
 
 	return 0;
 }
+
+DEFINE_SUITE("Synthesize stat config", synthesize_stat_config);
+DEFINE_SUITE("Synthesize stat", synthesize_stat);
+DEFINE_SUITE("Synthesize stat round", synthesize_stat_round);
diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c
index 74988846be1d..9cd6fec375ee 100644
--- a/tools/perf/tests/sw-clock.c
+++ b/tools/perf/tests/sw-clock.c
@@ -133,7 +133,7 @@ out_delete_evlist:
 	return err;
 }
 
-int test__sw_clock_freq(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__sw_clock_freq(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret;
 
@@ -143,3 +143,5 @@ int test__sw_clock_freq(struct test *test __maybe_unused, int subtest __maybe_un
 
 	return ret;
 }
+
+DEFINE_SUITE("Software clock events period values", sw_clock_freq);
diff --git a/tools/perf/tests/switch-tracking.c b/tools/perf/tests/switch-tracking.c
index 62c0ec21aaa8..0c0c2328bf4e 100644
--- a/tools/perf/tests/switch-tracking.c
+++ b/tools/perf/tests/switch-tracking.c
@@ -321,7 +321,7 @@ out_free_nodes:
  * evsel->core.system_wide and evsel->tracking flags (respectively) with other events
  * sometimes enabled or disabled.
  */
-int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__switch_tracking(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	const char *sched_switch = "sched:sched_switch";
 	struct switch_tracking switch_tracking = { .tids = NULL, };
@@ -588,3 +588,5 @@ out_err:
 	err = -1;
 	goto out;
 }
+
+DEFINE_SUITE("Track with sched_switch", switch_tracking);
diff --git a/tools/perf/tests/task-exit.c b/tools/perf/tests/task-exit.c
index 4c2969db59b0..25f075fa9125 100644
--- a/tools/perf/tests/task-exit.c
+++ b/tools/perf/tests/task-exit.c
@@ -39,7 +39,7 @@ static void workload_exec_failed_signal(int signo __maybe_unused,
  * if the number of exit event reported by the kernel is 1 or not
  * in order to check the kernel returns correct number of event.
  */
-int test__task_exit(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__task_exit(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int err = -1;
 	union perf_event *event;
@@ -151,3 +151,5 @@ out_delete_evlist:
 	evlist__delete(evlist);
 	return err;
 }
+
+DEFINE_SUITE("Number of exit events of a simple workload", task_exit);
diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h
index fe1306f58495..8f65098110fc 100644
--- a/tools/perf/tests/tests.h
+++ b/tools/perf/tests/tests.h
@@ -27,112 +27,146 @@ enum {
 	TEST_SKIP = -2,
 };
 
-struct test {
+struct test_suite;
+
+typedef int (*test_fnptr)(struct test_suite *, int);
+
+struct test_case {
+	const char *name;
+	const char *desc;
+	const char *skip_reason;
+	test_fnptr run_case;
+};
+
+struct test_suite {
 	const char *desc;
-	int (*func)(struct test *test, int subtest);
-	struct {
-		bool skip_if_fail;
-		int (*get_nr)(void);
-		const char *(*get_desc)(int subtest);
-		const char *(*skip_reason)(int subtest);
-	} subtest;
-	bool (*is_supported)(void);
+	struct test_case *test_cases;
 	void *priv;
 };
 
+#define DECLARE_SUITE(name) \
+	extern struct test_suite suite__##name;
+
+#define TEST_CASE(description, _name)			\
+	{						\
+		.name = #_name,				\
+		.desc = description,			\
+		.run_case = test__##_name,		\
+	}
+
+#define TEST_CASE_REASON(description, _name, _reason)	\
+	{						\
+		.name = #_name,				\
+		.desc = description,			\
+		.run_case = test__##_name,		\
+		.skip_reason = _reason,			\
+	}
+
+#define DEFINE_SUITE(description, _name)		\
+	struct test_case tests__##_name[] = {           \
+		TEST_CASE(description, _name),		\
+		{	.name = NULL, }			\
+	};						\
+	struct test_suite suite__##_name = {		\
+		.desc = description,			\
+		.test_cases = tests__##_name,		\
+	}
+
 /* Tests */
-int test__vmlinux_matches_kallsyms(struct test *test, int subtest);
-int test__openat_syscall_event(struct test *test, int subtest);
-int test__openat_syscall_event_on_all_cpus(struct test *test, int subtest);
-int test__basic_mmap(struct test *test, int subtest);
-int test__PERF_RECORD(struct test *test, int subtest);
-int test__perf_evsel__roundtrip_name_test(struct test *test, int subtest);
-int test__perf_evsel__tp_sched_test(struct test *test, int subtest);
-int test__syscall_openat_tp_fields(struct test *test, int subtest);
-int test__pmu(struct test *test, int subtest);
-int test__pmu_events(struct test *test, int subtest);
-const char *test__pmu_events_subtest_get_desc(int subtest);
-const char *test__pmu_events_subtest_skip_reason(int subtest);
-int test__pmu_events_subtest_get_nr(void);
-int test__attr(struct test *test, int subtest);
-int test__dso_data(struct test *test, int subtest);
-int test__dso_data_cache(struct test *test, int subtest);
-int test__dso_data_reopen(struct test *test, int subtest);
-int test__parse_events(struct test *test, int subtest);
-int test__hists_link(struct test *test, int subtest);
-int test__python_use(struct test *test, int subtest);
-int test__bp_signal(struct test *test, int subtest);
-int test__bp_signal_overflow(struct test *test, int subtest);
-int test__bp_accounting(struct test *test, int subtest);
-int test__wp(struct test *test, int subtest);
-const char *test__wp_subtest_get_desc(int subtest);
-const char *test__wp_subtest_skip_reason(int subtest);
-int test__wp_subtest_get_nr(void);
-int test__task_exit(struct test *test, int subtest);
-int test__mem(struct test *test, int subtest);
-int test__sw_clock_freq(struct test *test, int subtest);
-int test__code_reading(struct test *test, int subtest);
-int test__sample_parsing(struct test *test, int subtest);
-int test__keep_tracking(struct test *test, int subtest);
-int test__parse_no_sample_id_all(struct test *test, int subtest);
-int test__dwarf_unwind(struct test *test, int subtest);
-int test__expr(struct test *test, int subtest);
-int test__hists_filter(struct test *test, int subtest);
-int test__mmap_thread_lookup(struct test *test, int subtest);
-int test__thread_maps_share(struct test *test, int subtest);
-int test__hists_output(struct test *test, int subtest);
-int test__hists_cumulate(struct test *test, int subtest);
-int test__switch_tracking(struct test *test, int subtest);
-int test__fdarray__filter(struct test *test, int subtest);
-int test__fdarray__add(struct test *test, int subtest);
-int test__kmod_path__parse(struct test *test, int subtest);
-int test__thread_map(struct test *test, int subtest);
-int test__llvm(struct test *test, int subtest);
-const char *test__llvm_subtest_get_desc(int subtest);
-int test__llvm_subtest_get_nr(void);
-int test__bpf(struct test *test, int subtest);
-const char *test__bpf_subtest_get_desc(int subtest);
-int test__bpf_subtest_get_nr(void);
-int test__session_topology(struct test *test, int subtest);
-int test__thread_map_synthesize(struct test *test, int subtest);
-int test__thread_map_remove(struct test *test, int subtest);
-int test__cpu_map_synthesize(struct test *test, int subtest);
-int test__synthesize_stat_config(struct test *test, int subtest);
-int test__synthesize_stat(struct test *test, int subtest);
-int test__synthesize_stat_round(struct test *test, int subtest);
-int test__event_update(struct test *test, int subtest);
-int test__event_times(struct test *test, int subtest);
-int test__backward_ring_buffer(struct test *test, int subtest);
-int test__cpu_map_print(struct test *test, int subtest);
-int test__cpu_map_merge(struct test *test, int subtest);
-int test__sdt_event(struct test *test, int subtest);
-int test__is_printable_array(struct test *test, int subtest);
-int test__bitmap_print(struct test *test, int subtest);
-int test__perf_hooks(struct test *test, int subtest);
-int test__clang(struct test *test, int subtest);
-const char *test__clang_subtest_get_desc(int subtest);
-int test__clang_subtest_get_nr(void);
-int test__unit_number__scnprint(struct test *test, int subtest);
-int test__mem2node(struct test *t, int subtest);
-int test__maps__merge_in(struct test *t, int subtest);
-int test__time_utils(struct test *t, int subtest);
-int test__jit_write_elf(struct test *test, int subtest);
-int test__api_io(struct test *test, int subtest);
-int test__demangle_java(struct test *test, int subtest);
-int test__demangle_ocaml(struct test *test, int subtest);
-int test__pfm(struct test *test, int subtest);
-const char *test__pfm_subtest_get_desc(int subtest);
-int test__pfm_subtest_get_nr(void);
-int test__parse_metric(struct test *test, int subtest);
-int test__pe_file_parsing(struct test *test, int subtest);
-int test__expand_cgroup_events(struct test *test, int subtest);
-int test__perf_time_to_tsc(struct test *test, int subtest);
-int test__dlfilter(struct test *test, int subtest);
-
-bool test__bp_signal_is_supported(void);
-bool test__bp_account_is_supported(void);
-bool test__wp_is_supported(void);
-bool test__tsc_is_supported(void);
+DECLARE_SUITE(vmlinux_matches_kallsyms);
+DECLARE_SUITE(openat_syscall_event);
+DECLARE_SUITE(openat_syscall_event_on_all_cpus);
+DECLARE_SUITE(basic_mmap);
+DECLARE_SUITE(PERF_RECORD);
+DECLARE_SUITE(perf_evsel__roundtrip_name_test);
+DECLARE_SUITE(perf_evsel__tp_sched_test);
+DECLARE_SUITE(syscall_openat_tp_fields);
+DECLARE_SUITE(pmu);
+DECLARE_SUITE(pmu_events);
+DECLARE_SUITE(attr);
+DECLARE_SUITE(dso_data);
+DECLARE_SUITE(dso_data_cache);
+DECLARE_SUITE(dso_data_reopen);
+DECLARE_SUITE(parse_events);
+DECLARE_SUITE(hists_link);
+DECLARE_SUITE(python_use);
+DECLARE_SUITE(bp_signal);
+DECLARE_SUITE(bp_signal_overflow);
+DECLARE_SUITE(bp_accounting);
+DECLARE_SUITE(wp);
+DECLARE_SUITE(task_exit);
+DECLARE_SUITE(mem);
+DECLARE_SUITE(sw_clock_freq);
+DECLARE_SUITE(code_reading);
+DECLARE_SUITE(sample_parsing);
+DECLARE_SUITE(keep_tracking);
+DECLARE_SUITE(parse_no_sample_id_all);
+DECLARE_SUITE(dwarf_unwind);
+DECLARE_SUITE(expr);
+DECLARE_SUITE(hists_filter);
+DECLARE_SUITE(mmap_thread_lookup);
+DECLARE_SUITE(thread_maps_share);
+DECLARE_SUITE(hists_output);
+DECLARE_SUITE(hists_cumulate);
+DECLARE_SUITE(switch_tracking);
+DECLARE_SUITE(fdarray__filter);
+DECLARE_SUITE(fdarray__add);
+DECLARE_SUITE(kmod_path__parse);
+DECLARE_SUITE(thread_map);
+DECLARE_SUITE(llvm);
+DECLARE_SUITE(bpf);
+DECLARE_SUITE(session_topology);
+DECLARE_SUITE(thread_map_synthesize);
+DECLARE_SUITE(thread_map_remove);
+DECLARE_SUITE(cpu_map_synthesize);
+DECLARE_SUITE(synthesize_stat_config);
+DECLARE_SUITE(synthesize_stat);
+DECLARE_SUITE(synthesize_stat_round);
+DECLARE_SUITE(event_update);
+DECLARE_SUITE(event_times);
+DECLARE_SUITE(backward_ring_buffer);
+DECLARE_SUITE(cpu_map_print);
+DECLARE_SUITE(cpu_map_merge);
+DECLARE_SUITE(sdt_event);
+DECLARE_SUITE(is_printable_array);
+DECLARE_SUITE(bitmap_print);
+DECLARE_SUITE(perf_hooks);
+DECLARE_SUITE(clang);
+DECLARE_SUITE(unit_number__scnprint);
+DECLARE_SUITE(mem2node);
+DECLARE_SUITE(maps__merge_in);
+DECLARE_SUITE(time_utils);
+DECLARE_SUITE(jit_write_elf);
+DECLARE_SUITE(api_io);
+DECLARE_SUITE(demangle_java);
+DECLARE_SUITE(demangle_ocaml);
+DECLARE_SUITE(pfm);
+DECLARE_SUITE(parse_metric);
+DECLARE_SUITE(pe_file_parsing);
+DECLARE_SUITE(expand_cgroup_events);
+DECLARE_SUITE(perf_time_to_tsc);
+DECLARE_SUITE(dlfilter);
+
+/*
+ * PowerPC and S390 do not support creation of instruction breakpoints using the
+ * perf_event interface.
+ *
+ * ARM requires explicit rounding down of the instruction pointer in Thumb mode,
+ * and then requires the single-step to be handled explicitly in the overflow
+ * handler to avoid stepping into the SIGIO handler and getting stuck on the
+ * breakpointed instruction.
+ *
+ * Since arm64 has the same issue with arm for the single-step handling, this
+ * case also gets stuck on the breakpointed instruction.
+ *
+ * Just disable the test for these architectures until these issues are
+ * resolved.
+ */
+#if defined(__powerpc__) || defined(__s390x__) || defined(__arm__) || defined(__aarch64__)
+#define BP_SIGNAL_IS_SUPPORTED 0
+#else
+#define BP_SIGNAL_IS_SUPPORTED 1
+#endif
 
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
 struct thread;
@@ -142,7 +176,7 @@ int test__arch_unwind_sample(struct perf_sample *sample,
 #endif
 
 #if defined(__arm__)
-int test__vectors_page(struct test *test, int subtest);
+DECLARE_SUITE(vectors_page);
 #endif
 
 #endif /* TESTS_H */
diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c
index d1e208b4a571..e413c1387fcb 100644
--- a/tools/perf/tests/thread-map.c
+++ b/tools/perf/tests/thread-map.c
@@ -19,7 +19,7 @@ struct machine;
 #define NAME	(const char *) "perf"
 #define NAMEUL	(unsigned long) NAME
 
-int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__thread_map(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_thread_map *map;
 
@@ -86,7 +86,7 @@ static int process_event(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int test__thread_map_synthesize(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__thread_map_synthesize(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_thread_map *threads;
 
@@ -106,7 +106,7 @@ int test__thread_map_synthesize(struct test *test __maybe_unused, int subtest __
 	return 0;
 }
 
-int test__thread_map_remove(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__thread_map_remove(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct perf_thread_map *threads;
 	char *str;
@@ -145,3 +145,7 @@ int test__thread_map_remove(struct test *test __maybe_unused, int subtest __mayb
 	perf_thread_map__put(threads);
 	return 0;
 }
+
+DEFINE_SUITE("Thread map", thread_map);
+DEFINE_SUITE("Synthesize thread map", thread_map_synthesize);
+DEFINE_SUITE("Remove thread map", thread_map_remove);
diff --git a/tools/perf/tests/thread-maps-share.c b/tools/perf/tests/thread-maps-share.c
index 9371484973f2..84edd82c519e 100644
--- a/tools/perf/tests/thread-maps-share.c
+++ b/tools/perf/tests/thread-maps-share.c
@@ -4,7 +4,7 @@
 #include "thread.h"
 #include "debug.h"
 
-int test__thread_maps_share(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__thread_maps_share(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	struct machines machines;
 	struct machine *machine;
@@ -96,3 +96,5 @@ int test__thread_maps_share(struct test *test __maybe_unused, int subtest __mayb
 	machines__exit(&machines);
 	return 0;
 }
+
+DEFINE_SUITE("Share thread maps", thread_maps_share);
diff --git a/tools/perf/tests/time-utils-test.c b/tools/perf/tests/time-utils-test.c
index fe57ca3b6e54..38df10373c1e 100644
--- a/tools/perf/tests/time-utils-test.c
+++ b/tools/perf/tests/time-utils-test.c
@@ -131,7 +131,7 @@ out:
 	return pass;
 }
 
-int test__time_utils(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__time_utils(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	bool pass = true;
 
@@ -249,3 +249,5 @@ int test__time_utils(struct test *t __maybe_unused, int subtest __maybe_unused)
 
 	return pass ? 0 : TEST_FAIL;
 }
+
+DEFINE_SUITE("time utils", time_utils);
diff --git a/tools/perf/tests/topology.c b/tools/perf/tests/topology.c
index 4574c46260d9..869986139146 100644
--- a/tools/perf/tests/topology.c
+++ b/tools/perf/tests/topology.c
@@ -175,7 +175,7 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 	return 0;
 }
 
-int test__session_topology(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__session_topology(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	char path[PATH_MAX];
 	struct perf_cpu_map *map;
@@ -201,3 +201,5 @@ free_path:
 	unlink(path);
 	return ret;
 }
+
+DEFINE_SUITE("Session topology", session_topology);
diff --git a/tools/perf/tests/unit_number__scnprintf.c b/tools/perf/tests/unit_number__scnprintf.c
index 3721757435da..88bcada1c78f 100644
--- a/tools/perf/tests/unit_number__scnprintf.c
+++ b/tools/perf/tests/unit_number__scnprintf.c
@@ -7,7 +7,7 @@
 #include "units.h"
 #include "debug.h"
 
-int test__unit_number__scnprint(struct test *t __maybe_unused, int subtest __maybe_unused)
+static int test__unit_number__scnprint(struct test_suite *t __maybe_unused, int subtest __maybe_unused)
 {
 	struct {
 		u64		 n;
@@ -38,3 +38,5 @@ int test__unit_number__scnprint(struct test *t __maybe_unused, int subtest __may
 
 	return TEST_OK;
 }
+
+DEFINE_SUITE("unit_number__scnprintf", unit_number__scnprint);
diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c
index 4f884aabc7f4..e80df13c0420 100644
--- a/tools/perf/tests/vmlinux-kallsyms.c
+++ b/tools/perf/tests/vmlinux-kallsyms.c
@@ -111,7 +111,8 @@ static bool is_ignored_symbol(const char *name, char type)
 	return false;
 }
 
-int test__vmlinux_matches_kallsyms(struct test *test __maybe_unused, int subtest __maybe_unused)
+static int test__vmlinux_matches_kallsyms(struct test_suite *test __maybe_unused,
+					int subtest __maybe_unused)
 {
 	int err = -1;
 	struct rb_node *nd;
@@ -352,3 +353,5 @@ out:
 	machine__exit(&vmlinux);
 	return err;
 }
+
+DEFINE_SUITE("vmlinux symtab matches kallsyms", vmlinux_matches_kallsyms);
diff --git a/tools/perf/tests/wp.c b/tools/perf/tests/wp.c
index 9387fa76faa5..820d942b30c3 100644
--- a/tools/perf/tests/wp.c
+++ b/tools/perf/tests/wp.c
@@ -48,6 +48,7 @@ static void get__perf_event_attr(struct perf_event_attr *attr, int wp_type,
 	attr->exclude_hv     = 1;
 }
 
+#ifndef __s390x__
 static int __event(int wp_type, void *wp_addr, unsigned long wp_len)
 {
 	int fd;
@@ -61,9 +62,14 @@ static int __event(int wp_type, void *wp_addr, unsigned long wp_len)
 
 	return fd;
 }
+#endif
 
-static int wp_ro_test(void)
+static int test__wp_ro(struct test_suite *test __maybe_unused,
+		       int subtest __maybe_unused)
 {
+#if defined(__s390x__) || defined(__x86_64__) || defined(__i386__)
+	return TEST_SKIP;
+#else
 	int fd;
 	unsigned long tmp, tmp1 = rand();
 
@@ -79,10 +85,15 @@ static int wp_ro_test(void)
 
 	close(fd);
 	return 0;
+#endif
 }
 
-static int wp_wo_test(void)
+static int test__wp_wo(struct test_suite *test __maybe_unused,
+		       int subtest __maybe_unused)
 {
+#if defined(__s390x__)
+	return TEST_SKIP;
+#else
 	int fd;
 	unsigned long tmp, tmp1 = rand();
 
@@ -98,10 +109,15 @@ static int wp_wo_test(void)
 
 	close(fd);
 	return 0;
+#endif
 }
 
-static int wp_rw_test(void)
+static int test__wp_rw(struct test_suite *test __maybe_unused,
+		       int subtest __maybe_unused)
 {
+#if defined(__s390x__)
+	return TEST_SKIP;
+#else
 	int fd;
 	unsigned long tmp, tmp1 = rand();
 
@@ -118,10 +134,15 @@ static int wp_rw_test(void)
 
 	close(fd);
 	return 0;
+#endif
 }
 
-static int wp_modify_test(void)
+static int test__wp_modify(struct test_suite *test __maybe_unused,
+			   int subtest __maybe_unused)
 {
+#if defined(__s390x__)
+	return TEST_SKIP;
+#else
 	int fd, ret;
 	unsigned long tmp = rand();
 	struct perf_event_attr new_attr;
@@ -163,93 +184,18 @@ static int wp_modify_test(void)
 
 	close(fd);
 	return 0;
-}
-
-static bool wp_ro_supported(void)
-{
-#if defined (__x86_64__) || defined (__i386__)
-	return false;
-#else
-	return true;
-#endif
-}
-
-static const char *wp_ro_skip_msg(void)
-{
-#if defined (__x86_64__) || defined (__i386__)
-	return "missing hardware support";
-#else
-	return NULL;
 #endif
 }
 
-static struct {
-	const char *desc;
-	int (*target_func)(void);
-	bool (*is_supported)(void);
-	const char *(*skip_msg)(void);
-} wp_testcase_table[] = {
-	{
-		.desc = "Read Only Watchpoint",
-		.target_func = &wp_ro_test,
-		.is_supported = &wp_ro_supported,
-		.skip_msg = &wp_ro_skip_msg,
-	},
-	{
-		.desc = "Write Only Watchpoint",
-		.target_func = &wp_wo_test,
-	},
-	{
-		.desc = "Read / Write Watchpoint",
-		.target_func = &wp_rw_test,
-	},
-	{
-		.desc = "Modify Watchpoint",
-		.target_func = &wp_modify_test,
-	},
+static struct test_case wp_tests[] = {
+	TEST_CASE_REASON("Read Only Watchpoint", wp_ro, "missing hardware support"),
+	TEST_CASE_REASON("Write Only Watchpoint", wp_wo, "missing hardware support"),
+	TEST_CASE_REASON("Read / Write Watchpoint", wp_rw, "missing hardware support"),
+	TEST_CASE_REASON("Modify Watchpoint", wp_modify, "missing hardware support"),
+	{ .name = NULL, }
 };
 
-int test__wp_subtest_get_nr(void)
-{
-	return (int)ARRAY_SIZE(wp_testcase_table);
-}
-
-const char *test__wp_subtest_get_desc(int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(wp_testcase_table))
-		return NULL;
-	return wp_testcase_table[i].desc;
-}
-
-const char *test__wp_subtest_skip_reason(int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(wp_testcase_table))
-		return NULL;
-	if (!wp_testcase_table[i].skip_msg)
-		return NULL;
-	return wp_testcase_table[i].skip_msg();
-}
-
-int test__wp(struct test *test __maybe_unused, int i)
-{
-	if (i < 0 || i >= (int)ARRAY_SIZE(wp_testcase_table))
-		return TEST_FAIL;
-
-	if (wp_testcase_table[i].is_supported &&
-	    !wp_testcase_table[i].is_supported())
-		return TEST_SKIP;
-
-	return !wp_testcase_table[i].target_func() ? TEST_OK : TEST_FAIL;
-}
-
-/* The s390 so far does not have support for
- * instruction breakpoint using the perf_event_open() system call.
- */
-bool test__wp_is_supported(void)
-{
-#if defined(__s390x__)
-	return false;
-#else
-	return true;
-#endif
-}
+struct test_suite suite__wp = {
+	.desc = "Watchpoint",
+	.test_cases = wp_tests,
+};
diff --git a/tools/perf/trace/beauty/beauty.h b/tools/perf/trace/beauty/beauty.h
index d6dfe68a7612..f527a46ab4e7 100644
--- a/tools/perf/trace/beauty/beauty.h
+++ b/tools/perf/trace/beauty/beauty.h
@@ -62,6 +62,8 @@ size_t pid__scnprintf_fd(struct trace *trace, pid_t pid, int fd, char *bf, size_
 
 extern struct strarray strarray__socket_families;
 
+extern struct strarray strarray__socket_level;
+
 /**
  * augmented_arg: extra payload for syscall pointer arguments
  
@@ -230,6 +232,9 @@ size_t syscall_arg__scnprintf_sockaddr(char *bf, size_t size, struct syscall_arg
 size_t syscall_arg__scnprintf_socket_protocol(char *bf, size_t size, struct syscall_arg *arg);
 #define SCA_SK_PROTO syscall_arg__scnprintf_socket_protocol
 
+size_t syscall_arg__scnprintf_socket_level(char *bf, size_t size, struct syscall_arg *arg);
+#define SCA_SK_LEVEL syscall_arg__scnprintf_socket_level
+
 size_t syscall_arg__scnprintf_statx_flags(char *bf, size_t size, struct syscall_arg *arg);
 #define SCA_STATX_FLAGS syscall_arg__scnprintf_statx_flags
 
diff --git a/tools/perf/trace/beauty/sockaddr.c b/tools/perf/trace/beauty/sockaddr.c
index cd110634ab09..2e0e867c0c1b 100644
--- a/tools/perf/trace/beauty/sockaddr.c
+++ b/tools/perf/trace/beauty/sockaddr.c
@@ -7,7 +7,7 @@
 #include <sys/un.h>
 #include <arpa/inet.h>
 
-#include "trace/beauty/generated/socket_arrays.c"
+#include "trace/beauty/generated/sockaddr.c"
 DEFINE_STRARRAY(socket_families, "PF_");
 
 static size_t af_inet__scnprintf(struct sockaddr *sa, char *bf, size_t size)
diff --git a/tools/perf/trace/beauty/sockaddr.sh b/tools/perf/trace/beauty/sockaddr.sh
new file mode 100755
index 000000000000..3820e5c82293
--- /dev/null
+++ b/tools/perf/trace/beauty/sockaddr.sh
@@ -0,0 +1,24 @@
+#!/bin/sh
+# SPDX-License-Identifier: LGPL-2.1
+
+# This one uses a copy from the kernel sources headers that is in a
+# place used just for these tools/perf/beauty/ usage, we shouldn't not
+# put it in tools/include/linux otherwise they would be used in the
+# normal compiler building process and would drag needless stuff from the
+# kernel.
+
+# When what these scripts need is already in tools/include/ then use it,
+# otherwise grab and check the copy from the kernel sources just for these
+# string table building scripts.
+
+[ $# -eq 1 ] && header_dir=$1 || header_dir=tools/perf/trace/beauty/include/linux/
+
+printf "static const char *socket_families[] = {\n"
+# #define AF_LOCAL	1	/* POSIX name for AF_UNIX	*/
+regex='^#define[[:space:]]+AF_(\w+)[[:space:]]+([[:digit:]]+).*'
+
+egrep $regex ${header_dir}/socket.h | \
+	sed -r "s/$regex/\2 \1/g"	| \
+	xargs printf "\t[%s] = \"%s\",\n" | \
+	egrep -v "\"(UNIX|MAX)\""
+printf "};\n"
diff --git a/tools/perf/trace/beauty/socket.c b/tools/perf/trace/beauty/socket.c
index f23a3dda2902..b0870c7b48e5 100644
--- a/tools/perf/trace/beauty/socket.c
+++ b/tools/perf/trace/beauty/socket.c
@@ -9,9 +9,10 @@
 #include <sys/types.h>
 #include <sys/socket.h>
 
+#include "trace/beauty/generated/socket.c"
+
 static size_t socket__scnprintf_ipproto(int protocol, char *bf, size_t size, bool show_prefix)
 {
-#include "trace/beauty/generated/socket_ipproto_array.c"
 	static DEFINE_STRARRAY(socket_ipproto, "IPPROTO_");
 
 	return strarray__scnprintf(&strarray__socket_ipproto, bf, size, "%d", show_prefix, protocol);
@@ -26,3 +27,21 @@ size_t syscall_arg__scnprintf_socket_protocol(char *bf, size_t size, struct sysc
 
 	return syscall_arg__scnprintf_int(bf, size, arg);
 }
+
+static size_t socket__scnprintf_level(int level, char *bf, size_t size, bool show_prefix)
+{
+#if defined(__alpha__) || defined(__hppa__) || defined(__mips__) || defined(__sparc__)
+	const int sol_socket = 0xffff;
+#else
+	const int sol_socket = 1;
+#endif
+	if (level == sol_socket)
+		return scnprintf(bf, size, "%sSOCKET", show_prefix ? "SOL_" : "");
+
+	return strarray__scnprintf(&strarray__socket_level, bf, size, "%d", show_prefix, level);
+}
+
+size_t syscall_arg__scnprintf_socket_level(char *bf, size_t size, struct syscall_arg *arg)
+{
+	return socket__scnprintf_level(arg->val, bf, size, arg->show_string_prefix);
+}
diff --git a/tools/perf/trace/beauty/socket.sh b/tools/perf/trace/beauty/socket.sh
index 3820e5c82293..76330acb27e5 100755
--- a/tools/perf/trace/beauty/socket.sh
+++ b/tools/perf/trace/beauty/socket.sh
@@ -1,24 +1,28 @@
 #!/bin/sh
 # SPDX-License-Identifier: LGPL-2.1
 
-# This one uses a copy from the kernel sources headers that is in a
-# place used just for these tools/perf/beauty/ usage, we shouldn't not
-# put it in tools/include/linux otherwise they would be used in the
-# normal compiler building process and would drag needless stuff from the
-# kernel.
+if [ $# -gt 0 ] ; then
+	uapi_header_dir=$1
+	beauty_header_dir=$2
+else
+	uapi_header_dir=tools/include/uapi/linux/
+	beauty_header_dir=tools/perf/trace/beauty/include/linux/
+fi
 
-# When what these scripts need is already in tools/include/ then use it,
-# otherwise grab and check the copy from the kernel sources just for these
-# string table building scripts.
+printf "static const char *socket_ipproto[] = {\n"
+ipproto_regex='^[[:space:]]+IPPROTO_(\w+)[[:space:]]+=[[:space:]]+([[:digit:]]+),.*'
 
-[ $# -eq 1 ] && header_dir=$1 || header_dir=tools/perf/trace/beauty/include/linux/
+egrep $ipproto_regex ${uapi_header_dir}/in.h | \
+	sed -r "s/$ipproto_regex/\2 \1/g"	| \
+	sort -n | xargs printf "\t[%s] = \"%s\",\n"
+printf "};\n\n"
 
-printf "static const char *socket_families[] = {\n"
-# #define AF_LOCAL	1	/* POSIX name for AF_UNIX	*/
-regex='^#define[[:space:]]+AF_(\w+)[[:space:]]+([[:digit:]]+).*'
+printf "static const char *socket_level[] = {\n"
+socket_level_regex='^#define[[:space:]]+SOL_(\w+)[[:space:]]+([[:digit:]]+)([[:space:]]+\/.*)?'
 
-egrep $regex ${header_dir}/socket.h | \
-	sed -r "s/$regex/\2 \1/g"	| \
-	xargs printf "\t[%s] = \"%s\",\n" | \
-	egrep -v "\"(UNIX|MAX)\""
-printf "};\n"
+egrep $socket_level_regex ${beauty_header_dir}/socket.h | \
+	sed -r "s/$socket_level_regex/\2 \1/g"	| \
+	sort -n | xargs printf "\t[%s] = \"%s\",\n"
+printf "};\n\n"
+
+printf 'DEFINE_STRARRAY(socket_level, "SOL_");\n'
diff --git a/tools/perf/trace/beauty/socket_ipproto.sh b/tools/perf/trace/beauty/socket_ipproto.sh
deleted file mode 100755
index de0f2f29017f..000000000000
--- a/tools/perf/trace/beauty/socket_ipproto.sh
+++ /dev/null
@@ -1,12 +0,0 @@
-#!/bin/sh
-# SPDX-License-Identifier: LGPL-2.1
-
-[ $# -eq 1 ] && header_dir=$1 || header_dir=tools/include/uapi/linux/
-
-printf "static const char *socket_ipproto[] = {\n"
-regex='^[[:space:]]+IPPROTO_(\w+)[[:space:]]+=[[:space:]]+([[:digit:]]+),.*'
-
-egrep $regex ${header_dir}/in.h | \
-	sed -r "s/$regex/\2 \1/g"	| \
-	sort | xargs printf "\t[%s] = \"%s\",\n"
-printf "};\n"
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 8511af55fc3a..01900689dc00 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -1255,6 +1255,17 @@ int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool r
 	return ins__scnprintf(&dl->ins, bf, size, &dl->ops, max_ins_name);
 }
 
+void annotation__init(struct annotation *notes)
+{
+	pthread_mutex_init(&notes->lock, NULL);
+}
+
+void annotation__exit(struct annotation *notes)
+{
+	annotated_source__delete(notes->src);
+	pthread_mutex_destroy(&notes->lock);
+}
+
 static void annotation_line__add(struct annotation_line *al, struct list_head *head)
 {
 	list_add_tail(&al->node, head);
@@ -3132,7 +3143,7 @@ int symbol__annotate2(struct map_symbol *ms, struct evsel *evsel,
 	notes->nr_events = nr_pcnt;
 
 	annotation__update_column_widths(notes);
-	sym->annotate2 = true;
+	sym->annotate2 = 1;
 
 	return 0;
 
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
index 3757416bcf46..986f2bbe4870 100644
--- a/tools/perf/util/annotate.h
+++ b/tools/perf/util/annotate.h
@@ -299,6 +299,9 @@ struct annotation {
 	struct annotated_source *src;
 };
 
+void annotation__init(struct annotation *notes);
+void annotation__exit(struct annotation *notes);
+
 static inline int annotation__cycles_width(struct annotation *notes)
 {
 	if (notes->have_cycles && notes->options->show_minmax_cycle)
diff --git a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c
index 32fe41835fa6..3fc528c9270c 100644
--- a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c
+++ b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c
@@ -151,6 +151,7 @@ static int arm_spe_read_record(struct arm_spe_decoder *decoder)
 	u64 payload, ip;
 
 	memset(&decoder->record, 0x0, sizeof(decoder->record));
+	decoder->record.context_id = (u64)-1;
 
 	while (1) {
 		err = arm_spe_get_next_packet(decoder);
@@ -180,6 +181,7 @@ static int arm_spe_read_record(struct arm_spe_decoder *decoder)
 		case ARM_SPE_COUNTER:
 			break;
 		case ARM_SPE_CONTEXT:
+			decoder->record.context_id = payload;
 			break;
 		case ARM_SPE_OP_TYPE:
 			if (idx == SPE_OP_PKT_HDR_CLASS_LD_ST_ATOMIC) {
diff --git a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.h b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.h
index 59bdb7309674..46a8556a9e95 100644
--- a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.h
+++ b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.h
@@ -38,6 +38,7 @@ struct arm_spe_record {
 	u64 timestamp;
 	u64 virt_addr;
 	u64 phys_addr;
+	u64 context_id;
 };
 
 struct arm_spe_insn;
diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c
index 58b7069c5a5f..4748bcfe61de 100644
--- a/tools/perf/util/arm-spe.c
+++ b/tools/perf/util/arm-spe.c
@@ -71,6 +71,7 @@ struct arm_spe {
 	u64				kernel_start;
 
 	unsigned long			num_events;
+	u8				use_ctx_pkt_for_pid;
 };
 
 struct arm_spe_queue {
@@ -100,7 +101,7 @@ static void arm_spe_dump(struct arm_spe *spe __maybe_unused,
 	const char *color = PERF_COLOR_BLUE;
 
 	color_fprintf(stdout, color,
-		      ". ... ARM SPE data: size %zu bytes\n",
+		      ". ... ARM SPE data: size %#zx bytes\n",
 		      len);
 
 	while (len) {
@@ -226,6 +227,44 @@ static inline u8 arm_spe_cpumode(struct arm_spe *spe, u64 ip)
 		PERF_RECORD_MISC_USER;
 }
 
+static void arm_spe_set_pid_tid_cpu(struct arm_spe *spe,
+				    struct auxtrace_queue *queue)
+{
+	struct arm_spe_queue *speq = queue->priv;
+	pid_t tid;
+
+	tid = machine__get_current_tid(spe->machine, speq->cpu);
+	if (tid != -1) {
+		speq->tid = tid;
+		thread__zput(speq->thread);
+	} else
+		speq->tid = queue->tid;
+
+	if ((!speq->thread) && (speq->tid != -1)) {
+		speq->thread = machine__find_thread(spe->machine, -1,
+						    speq->tid);
+	}
+
+	if (speq->thread) {
+		speq->pid = speq->thread->pid_;
+		if (queue->cpu == -1)
+			speq->cpu = speq->thread->cpu;
+	}
+}
+
+static int arm_spe_set_tid(struct arm_spe_queue *speq, pid_t tid)
+{
+	struct arm_spe *spe = speq->spe;
+	int err = machine__set_current_tid(spe->machine, speq->cpu, -1, tid);
+
+	if (err)
+		return err;
+
+	arm_spe_set_pid_tid_cpu(spe, &spe->queues.queue_array[speq->queue_nr]);
+
+	return 0;
+}
+
 static void arm_spe_prep_sample(struct arm_spe *spe,
 				struct arm_spe_queue *speq,
 				union perf_event *event,
@@ -460,6 +499,19 @@ static int arm_spe_run_decoder(struct arm_spe_queue *speq, u64 *timestamp)
 		 * can correlate samples between Arm SPE trace data and other
 		 * perf events with correct time ordering.
 		 */
+
+		/*
+		 * Update pid/tid info.
+		 */
+		record = &speq->decoder->record;
+		if (!spe->timeless_decoding && record->context_id != (u64)-1) {
+			ret = arm_spe_set_tid(speq, record->context_id);
+			if (ret)
+				return ret;
+
+			spe->use_ctx_pkt_for_pid = true;
+		}
+
 		ret = arm_spe_sample(speq);
 		if (ret)
 			return ret;
@@ -586,31 +638,6 @@ static bool arm_spe__is_timeless_decoding(struct arm_spe *spe)
 	return timeless_decoding;
 }
 
-static void arm_spe_set_pid_tid_cpu(struct arm_spe *spe,
-				    struct auxtrace_queue *queue)
-{
-	struct arm_spe_queue *speq = queue->priv;
-	pid_t tid;
-
-	tid = machine__get_current_tid(spe->machine, speq->cpu);
-	if (tid != -1) {
-		speq->tid = tid;
-		thread__zput(speq->thread);
-	} else
-		speq->tid = queue->tid;
-
-	if ((!speq->thread) && (speq->tid != -1)) {
-		speq->thread = machine__find_thread(spe->machine, -1,
-						    speq->tid);
-	}
-
-	if (speq->thread) {
-		speq->pid = speq->thread->pid_;
-		if (queue->cpu == -1)
-			speq->cpu = speq->thread->cpu;
-	}
-}
-
 static int arm_spe_process_queues(struct arm_spe *spe, u64 timestamp)
 {
 	unsigned int queue_nr;
@@ -641,7 +668,12 @@ static int arm_spe_process_queues(struct arm_spe *spe, u64 timestamp)
 			ts = timestamp;
 		}
 
-		arm_spe_set_pid_tid_cpu(spe, queue);
+		/*
+		 * A previous context-switch event has set pid/tid in the machine's context, so
+		 * here we need to update the pid/tid in the thread and SPE queue.
+		 */
+		if (!spe->use_ctx_pkt_for_pid)
+			arm_spe_set_pid_tid_cpu(spe, queue);
 
 		ret = arm_spe_run_decoder(speq, &ts);
 		if (ret < 0) {
@@ -681,6 +713,25 @@ static int arm_spe_process_timeless_queues(struct arm_spe *spe, pid_t tid,
 	return 0;
 }
 
+static int arm_spe_context_switch(struct arm_spe *spe, union perf_event *event,
+				  struct perf_sample *sample)
+{
+	pid_t pid, tid;
+	int cpu;
+
+	if (!(event->header.misc & PERF_RECORD_MISC_SWITCH_OUT))
+		return 0;
+
+	pid = event->context_switch.next_prev_pid;
+	tid = event->context_switch.next_prev_tid;
+	cpu = sample->cpu;
+
+	if (tid == -1)
+		pr_warning("context_switch event has no tid\n");
+
+	return machine__set_current_tid(spe->machine, cpu, pid, tid);
+}
+
 static int arm_spe_process_event(struct perf_session *session,
 				 union perf_event *event,
 				 struct perf_sample *sample,
@@ -718,6 +769,13 @@ static int arm_spe_process_event(struct perf_session *session,
 		}
 	} else if (timestamp) {
 		err = arm_spe_process_queues(spe, timestamp);
+		if (err)
+			return err;
+
+		if (!spe->use_ctx_pkt_for_pid &&
+		    (event->header.type == PERF_RECORD_SWITCH_CPU_WIDE ||
+		    event->header.type == PERF_RECORD_SWITCH))
+			err = arm_spe_context_switch(spe, event, sample);
 	}
 
 	return err;
@@ -783,7 +841,15 @@ static int arm_spe_flush(struct perf_session *session __maybe_unused,
 		return arm_spe_process_timeless_queues(spe, -1,
 				MAX_TIMESTAMP - 1);
 
-	return arm_spe_process_queues(spe, MAX_TIMESTAMP);
+	ret = arm_spe_process_queues(spe, MAX_TIMESTAMP);
+	if (ret)
+		return ret;
+
+	if (!spe->use_ctx_pkt_for_pid)
+		ui__warning("Arm SPE CONTEXT packets not found in the traces.\n"
+			    "Matching of TIDs to SPE events could be inaccurate.\n");
+
+	return 0;
 }
 
 static void arm_spe_free_queue(void *priv)
diff --git a/tools/perf/util/bpf-event.c b/tools/perf/util/bpf-event.c
index 4d3b4cdce176..a517eaa51eb3 100644
--- a/tools/perf/util/bpf-event.c
+++ b/tools/perf/util/bpf-event.c
@@ -33,6 +33,33 @@ struct btf * __weak btf__load_from_kernel_by_id(__u32 id)
        return err ? ERR_PTR(err) : btf;
 }
 
+struct bpf_program * __weak
+bpf_object__next_program(const struct bpf_object *obj, struct bpf_program *prev)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+	return bpf_program__next(prev, obj);
+#pragma GCC diagnostic pop
+}
+
+struct bpf_map * __weak
+bpf_object__next_map(const struct bpf_object *obj, const struct bpf_map *prev)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+	return bpf_map__next(prev, obj);
+#pragma GCC diagnostic pop
+}
+
+const void * __weak
+btf__raw_data(const struct btf *btf_ro, __u32 *size)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+	return btf__get_raw_data(btf_ro, size);
+#pragma GCC diagnostic pop
+}
+
 static int snprintf_hex(char *buf, size_t size, unsigned char *data, size_t len)
 {
 	int ret = 0;
@@ -119,7 +146,11 @@ static int perf_env__fetch_btf(struct perf_env *env,
 	node->data_size = data_size;
 	memcpy(node->data, data, data_size);
 
-	perf_env__insert_btf(env, node);
+	if (!perf_env__insert_btf(env, node)) {
+		/* Insertion failed because of a duplicate. */
+		free(node);
+		return -1;
+	}
 	return 0;
 }
 
diff --git a/tools/perf/util/c++/clang-c.h b/tools/perf/util/c++/clang-c.h
index 2df8a45bd088..d3731a876b6c 100644
--- a/tools/perf/util/c++/clang-c.h
+++ b/tools/perf/util/c++/clang-c.h
@@ -12,8 +12,9 @@ extern "C" {
 extern void perf_clang__init(void);
 extern void perf_clang__cleanup(void);
 
-extern int test__clang_to_IR(void);
-extern int test__clang_to_obj(void);
+struct test_suite;
+extern int test__clang_to_IR(struct test_suite *test, int subtest);
+extern int test__clang_to_obj(struct test_suite *test, int subtest);
 
 extern int perf_clang__compile_bpf(const char *filename,
 				   void **p_obj_buf,
@@ -26,9 +27,6 @@ extern int perf_clang__compile_bpf(const char *filename,
 static inline void perf_clang__init(void) { }
 static inline void perf_clang__cleanup(void) { }
 
-static inline int test__clang_to_IR(void) { return -1; }
-static inline int test__clang_to_obj(void) { return -1;}
-
 static inline int
 perf_clang__compile_bpf(const char *filename __maybe_unused,
 			void **p_obj_buf __maybe_unused,
diff --git a/tools/perf/util/c++/clang-test.cpp b/tools/perf/util/c++/clang-test.cpp
index 21b23605f78b..a4683ca53697 100644
--- a/tools/perf/util/c++/clang-test.cpp
+++ b/tools/perf/util/c++/clang-test.cpp
@@ -35,7 +35,8 @@ __test__clang_to_IR(void)
 }
 
 extern "C" {
-int test__clang_to_IR(void)
+int test__clang_to_IR(struct test_suite *test __maybe_unused,
+                      int subtest __maybe_unused)
 {
 	perf_clang_scope _scope;
 
@@ -48,7 +49,8 @@ int test__clang_to_IR(void)
 	return -1;
 }
 
-int test__clang_to_obj(void)
+int test__clang_to_obj(struct test_suite *test __maybe_unused,
+                       int subtest __maybe_unused)
 {
 	perf_clang_scope _scope;
 
diff --git a/tools/perf/util/cputopo.c b/tools/perf/util/cputopo.c
index ec77e2a7b3ca..51b429c86f98 100644
--- a/tools/perf/util/cputopo.c
+++ b/tools/perf/util/cputopo.c
@@ -14,14 +14,16 @@
 #include "env.h"
 #include "pmu-hybrid.h"
 
-#define CORE_SIB_FMT \
+#define PACKAGE_CPUS_FMT \
+	"%s/devices/system/cpu/cpu%d/topology/package_cpus_list"
+#define PACKAGE_CPUS_FMT_OLD \
 	"%s/devices/system/cpu/cpu%d/topology/core_siblings_list"
-#define DIE_SIB_FMT \
+#define DIE_CPUS_FMT \
 	"%s/devices/system/cpu/cpu%d/topology/die_cpus_list"
-#define THRD_SIB_FMT \
-	"%s/devices/system/cpu/cpu%d/topology/thread_siblings_list"
-#define THRD_SIB_FMT_NEW \
+#define CORE_CPUS_FMT \
 	"%s/devices/system/cpu/cpu%d/topology/core_cpus_list"
+#define CORE_CPUS_FMT_OLD \
+	"%s/devices/system/cpu/cpu%d/topology/thread_siblings_list"
 #define NODE_ONLINE_FMT \
 	"%s/devices/system/node/online"
 #define NODE_MEMINFO_FMT \
@@ -39,8 +41,12 @@ static int build_cpu_topology(struct cpu_topology *tp, int cpu)
 	u32 i = 0;
 	int ret = -1;
 
-	scnprintf(filename, MAXPATHLEN, CORE_SIB_FMT,
+	scnprintf(filename, MAXPATHLEN, PACKAGE_CPUS_FMT,
 		  sysfs__mountpoint(), cpu);
+	if (access(filename, F_OK) == -1) {
+		scnprintf(filename, MAXPATHLEN, PACKAGE_CPUS_FMT_OLD,
+			sysfs__mountpoint(), cpu);
+	}
 	fp = fopen(filename, "r");
 	if (!fp)
 		goto try_dies;
@@ -54,23 +60,23 @@ static int build_cpu_topology(struct cpu_topology *tp, int cpu)
 	if (p)
 		*p = '\0';
 
-	for (i = 0; i < tp->core_sib; i++) {
-		if (!strcmp(buf, tp->core_siblings[i]))
+	for (i = 0; i < tp->package_cpus_lists; i++) {
+		if (!strcmp(buf, tp->package_cpus_list[i]))
 			break;
 	}
-	if (i == tp->core_sib) {
-		tp->core_siblings[i] = buf;
-		tp->core_sib++;
+	if (i == tp->package_cpus_lists) {
+		tp->package_cpus_list[i] = buf;
+		tp->package_cpus_lists++;
 		buf = NULL;
 		len = 0;
 	}
 	ret = 0;
 
 try_dies:
-	if (!tp->die_siblings)
+	if (!tp->die_cpus_list)
 		goto try_threads;
 
-	scnprintf(filename, MAXPATHLEN, DIE_SIB_FMT,
+	scnprintf(filename, MAXPATHLEN, DIE_CPUS_FMT,
 		  sysfs__mountpoint(), cpu);
 	fp = fopen(filename, "r");
 	if (!fp)
@@ -85,23 +91,23 @@ try_dies:
 	if (p)
 		*p = '\0';
 
-	for (i = 0; i < tp->die_sib; i++) {
-		if (!strcmp(buf, tp->die_siblings[i]))
+	for (i = 0; i < tp->die_cpus_lists; i++) {
+		if (!strcmp(buf, tp->die_cpus_list[i]))
 			break;
 	}
-	if (i == tp->die_sib) {
-		tp->die_siblings[i] = buf;
-		tp->die_sib++;
+	if (i == tp->die_cpus_lists) {
+		tp->die_cpus_list[i] = buf;
+		tp->die_cpus_lists++;
 		buf = NULL;
 		len = 0;
 	}
 	ret = 0;
 
 try_threads:
-	scnprintf(filename, MAXPATHLEN, THRD_SIB_FMT_NEW,
+	scnprintf(filename, MAXPATHLEN, CORE_CPUS_FMT,
 		  sysfs__mountpoint(), cpu);
 	if (access(filename, F_OK) == -1) {
-		scnprintf(filename, MAXPATHLEN, THRD_SIB_FMT,
+		scnprintf(filename, MAXPATHLEN, CORE_CPUS_FMT_OLD,
 			  sysfs__mountpoint(), cpu);
 	}
 	fp = fopen(filename, "r");
@@ -115,13 +121,13 @@ try_threads:
 	if (p)
 		*p = '\0';
 
-	for (i = 0; i < tp->thread_sib; i++) {
-		if (!strcmp(buf, tp->thread_siblings[i]))
+	for (i = 0; i < tp->core_cpus_lists; i++) {
+		if (!strcmp(buf, tp->core_cpus_list[i]))
 			break;
 	}
-	if (i == tp->thread_sib) {
-		tp->thread_siblings[i] = buf;
-		tp->thread_sib++;
+	if (i == tp->core_cpus_lists) {
+		tp->core_cpus_list[i] = buf;
+		tp->core_cpus_lists++;
 		buf = NULL;
 	}
 	ret = 0;
@@ -139,16 +145,14 @@ void cpu_topology__delete(struct cpu_topology *tp)
 	if (!tp)
 		return;
 
-	for (i = 0 ; i < tp->core_sib; i++)
-		zfree(&tp->core_siblings[i]);
+	for (i = 0 ; i < tp->package_cpus_lists; i++)
+		zfree(&tp->package_cpus_list[i]);
 
-	if (tp->die_sib) {
-		for (i = 0 ; i < tp->die_sib; i++)
-			zfree(&tp->die_siblings[i]);
-	}
+	for (i = 0 ; i < tp->die_cpus_lists; i++)
+		zfree(&tp->die_cpus_list[i]);
 
-	for (i = 0 ; i < tp->thread_sib; i++)
-		zfree(&tp->thread_siblings[i]);
+	for (i = 0 ; i < tp->core_cpus_lists; i++)
+		zfree(&tp->core_cpus_list[i]);
 
 	free(tp);
 }
@@ -164,7 +168,7 @@ static bool has_die_topology(void)
 	if (strncmp(uts.machine, "x86_64", 6))
 		return false;
 
-	scnprintf(filename, MAXPATHLEN, DIE_SIB_FMT,
+	scnprintf(filename, MAXPATHLEN, DIE_CPUS_FMT,
 		  sysfs__mountpoint(), 0);
 	if (access(filename, F_OK) == -1)
 		return false;
@@ -205,13 +209,13 @@ struct cpu_topology *cpu_topology__new(void)
 
 	tp = addr;
 	addr += sizeof(*tp);
-	tp->core_siblings = addr;
+	tp->package_cpus_list = addr;
 	addr += sz;
 	if (has_die) {
-		tp->die_siblings = addr;
+		tp->die_cpus_list = addr;
 		addr += sz;
 	}
-	tp->thread_siblings = addr;
+	tp->core_cpus_list = addr;
 
 	for (i = 0; i < nr; i++) {
 		if (!cpu_map__has(map, i))
diff --git a/tools/perf/util/cputopo.h b/tools/perf/util/cputopo.h
index d9af97177068..854e18f9041e 100644
--- a/tools/perf/util/cputopo.h
+++ b/tools/perf/util/cputopo.h
@@ -5,12 +5,33 @@
 #include <linux/types.h>
 
 struct cpu_topology {
-	u32	  core_sib;
-	u32	  die_sib;
-	u32	  thread_sib;
-	char	**core_siblings;
-	char	**die_siblings;
-	char	**thread_siblings;
+	/* The number of unique package_cpus_lists below. */
+	u32	  package_cpus_lists;
+	/* The number of unique die_cpu_lists below. */
+	u32	  die_cpus_lists;
+	/* The number of unique core_cpu_lists below. */
+	u32	  core_cpus_lists;
+	/*
+	 * An array of strings where each string is unique and read from
+	 * /sys/devices/system/cpu/cpuX/topology/package_cpus_list. From the ABI
+	 * each of these is a human-readable list of CPUs sharing the same
+	 * physical_package_id. The format is like 0-3, 8-11, 14,17.
+	 */
+	const char **package_cpus_list;
+	/*
+	 * An array of string where each string is unique and from
+	 * /sys/devices/system/cpu/cpuX/topology/die_cpus_list. From the ABI
+	 * each of these is a human-readable list of CPUs within the same die.
+	 * The format is like 0-3, 8-11, 14,17.
+	 */
+	const char **die_cpus_list;
+	/*
+	 * An array of string where each string is unique and from
+	 * /sys/devices/system/cpu/cpuX/topology/core_cpus_list. From the ABI
+	 * each of these is a human-readable list of CPUs within the same
+	 * core. The format is like 0-3, 8-11, 14,17.
+	 */
+	const char **core_cpus_list;
 };
 
 struct numa_topology_node {
diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c
index f323adb1af85..4f672f7d008c 100644
--- a/tools/perf/util/cs-etm.c
+++ b/tools/perf/util/cs-etm.c
@@ -537,7 +537,7 @@ static void cs_etm__dump_event(struct cs_etm_queue *etmq,
 
 	fprintf(stdout, "\n");
 	color_fprintf(stdout, color,
-		     ". ... CoreSight %s Trace data: size %zu bytes\n",
+		     ". ... CoreSight %s Trace data: size %#zx bytes\n",
 		     cs_etm_decoder__get_name(etmq->decoder), buffer->size);
 
 	do {
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
index 17f1dd0680b4..b9904896eb97 100644
--- a/tools/perf/util/env.c
+++ b/tools/perf/util/env.c
@@ -75,12 +75,13 @@ out:
 	return node;
 }
 
-void perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node)
+bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node)
 {
 	struct rb_node *parent = NULL;
 	__u32 btf_id = btf_node->id;
 	struct btf_node *node;
 	struct rb_node **p;
+	bool ret = true;
 
 	down_write(&env->bpf_progs.lock);
 	p = &env->bpf_progs.btfs.rb_node;
@@ -94,6 +95,7 @@ void perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node)
 			p = &(*p)->rb_right;
 		} else {
 			pr_debug("duplicated btf %u\n", btf_id);
+			ret = false;
 			goto out;
 		}
 	}
@@ -103,6 +105,7 @@ void perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node)
 	env->bpf_progs.btfs_cnt++;
 out:
 	up_write(&env->bpf_progs.lock);
+	return ret;
 }
 
 struct btf_node *perf_env__find_btf(struct perf_env *env, __u32 btf_id)
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
index 1383876f72b3..163e5ec503a2 100644
--- a/tools/perf/util/env.h
+++ b/tools/perf/util/env.h
@@ -167,7 +167,7 @@ void perf_env__insert_bpf_prog_info(struct perf_env *env,
 				    struct bpf_prog_info_node *info_node);
 struct bpf_prog_info_node *perf_env__find_bpf_prog_info(struct perf_env *env,
 							__u32 prog_id);
-void perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node);
+bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node);
 struct btf_node *perf_env__find_btf(struct perf_env *env, __u32 btf_id);
 
 int perf_env__numa_node(struct perf_env *env, int cpu);
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index ec967fb8d7d9..a59fb2ecb84e 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -3037,3 +3037,15 @@ void evsel__set_leader(struct evsel *evsel, struct evsel *leader)
 {
 	evsel->core.leader = &leader->core;
 }
+
+int evsel__source_count(const struct evsel *evsel)
+{
+	struct evsel *pos;
+	int count = 0;
+
+	evlist__for_each_entry(evsel->evlist, pos) {
+		if (pos->metric_leader == evsel)
+			count++;
+	}
+	return count;
+}
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 3ea687141afa..29d49a8c1e92 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -489,6 +489,7 @@ struct evsel *evsel__leader(struct evsel *evsel);
 bool evsel__has_leader(struct evsel *evsel, struct evsel *leader);
 bool evsel__is_leader(struct evsel *evsel);
 void evsel__set_leader(struct evsel *evsel, struct evsel *leader);
+int evsel__source_count(const struct evsel *evsel);
 
 /*
  * Macro to swap the bit-field postition and size.
diff --git a/tools/perf/util/expr.c b/tools/perf/util/expr.c
index 77c6ad81a923..1d532b9fed29 100644
--- a/tools/perf/util/expr.c
+++ b/tools/perf/util/expr.c
@@ -5,13 +5,17 @@
 #include <stdlib.h>
 #include <string.h>
 #include "metricgroup.h"
+#include "cpumap.h"
+#include "cputopo.h"
 #include "debug.h"
 #include "expr.h"
 #include "expr-bison.h"
 #include "expr-flex.h"
+#include "smt.h"
 #include <linux/kernel.h>
 #include <linux/zalloc.h>
 #include <ctype.h>
+#include <math.h>
 
 #ifdef PARSER_DEBUG
 extern int expr_debug;
@@ -19,7 +23,10 @@ extern int expr_debug;
 
 struct expr_id_data {
 	union {
-		double val;
+		struct {
+			double val;
+			int source_count;
+		} val;
 		struct {
 			double val;
 			const char *metric_name;
@@ -137,6 +144,13 @@ int expr__add_id(struct expr_parse_ctx *ctx, const char *id)
 /* Caller must make sure id is allocated */
 int expr__add_id_val(struct expr_parse_ctx *ctx, const char *id, double val)
 {
+	return expr__add_id_val_source_count(ctx, id, val, /*source_count=*/1);
+}
+
+/* Caller must make sure id is allocated */
+int expr__add_id_val_source_count(struct expr_parse_ctx *ctx, const char *id,
+				  double val, int source_count)
+{
 	struct expr_id_data *data_ptr = NULL, *old_data = NULL;
 	char *old_key = NULL;
 	int ret;
@@ -144,7 +158,8 @@ int expr__add_id_val(struct expr_parse_ctx *ctx, const char *id, double val)
 	data_ptr = malloc(sizeof(*data_ptr));
 	if (!data_ptr)
 		return -ENOMEM;
-	data_ptr->val = val;
+	data_ptr->val.val = val;
+	data_ptr->val.source_count = source_count;
 	data_ptr->kind = EXPR_ID_DATA__VALUE;
 
 	ret = hashmap__set(ctx->ids, id, data_ptr,
@@ -240,7 +255,7 @@ int expr__resolve_id(struct expr_parse_ctx *ctx, const char *id,
 
 	switch (data->kind) {
 	case EXPR_ID_DATA__VALUE:
-		pr_debug2("lookup(%s): val %f\n", id, data->val);
+		pr_debug2("lookup(%s): val %f\n", id, data->val.val);
 		break;
 	case EXPR_ID_DATA__REF:
 		pr_debug2("lookup(%s): ref metric name %s\n", id,
@@ -251,7 +266,7 @@ int expr__resolve_id(struct expr_parse_ctx *ctx, const char *id,
 			pr_debug("%s failed to count\n", id);
 			return -1;
 		}
-		pr_debug("processing metric: %s EXIT: %f\n", id, data->val);
+		pr_debug("processing metric: %s EXIT: %f\n", id, data->ref.val);
 		break;
 	case EXPR_ID_DATA__REF_VALUE:
 		pr_debug2("lookup(%s): ref val %f metric name %s\n", id,
@@ -366,7 +381,47 @@ int expr__find_ids(const char *expr, const char *one,
 double expr_id_data__value(const struct expr_id_data *data)
 {
 	if (data->kind == EXPR_ID_DATA__VALUE)
-		return data->val;
+		return data->val.val;
 	assert(data->kind == EXPR_ID_DATA__REF_VALUE);
 	return data->ref.val;
 }
+
+double expr_id_data__source_count(const struct expr_id_data *data)
+{
+	assert(data->kind == EXPR_ID_DATA__VALUE);
+	return data->val.source_count;
+}
+
+double expr__get_literal(const char *literal)
+{
+	static struct cpu_topology *topology;
+
+	if (!strcmp("#smt_on", literal))
+		return smt_on() > 0 ? 1.0 : 0.0;
+
+	if (!strcmp("#num_cpus", literal))
+		return cpu__max_present_cpu();
+
+	/*
+	 * Assume that topology strings are consistent, such as CPUs "0-1"
+	 * wouldn't be listed as "0,1", and so after deduplication the number of
+	 * these strings gives an indication of the number of packages, dies,
+	 * etc.
+	 */
+	if (!topology) {
+		topology = cpu_topology__new();
+		if (!topology) {
+			pr_err("Error creating CPU topology");
+			return NAN;
+		}
+	}
+	if (!strcmp("#num_packages", literal))
+		return topology->package_cpus_lists;
+	if (!strcmp("#num_dies", literal))
+		return topology->die_cpus_lists;
+	if (!strcmp("#num_cores", literal))
+		return topology->core_cpus_lists;
+
+	pr_err("Unrecognized literal '%s'", literal);
+	return NAN;
+}
diff --git a/tools/perf/util/expr.h b/tools/perf/util/expr.h
index cf81f9166dbb..bd2116983bbb 100644
--- a/tools/perf/util/expr.h
+++ b/tools/perf/util/expr.h
@@ -40,6 +40,8 @@ void expr__ctx_free(struct expr_parse_ctx *ctx);
 void expr__del_id(struct expr_parse_ctx *ctx, const char *id);
 int expr__add_id(struct expr_parse_ctx *ctx, const char *id);
 int expr__add_id_val(struct expr_parse_ctx *ctx, const char *id, double val);
+int expr__add_id_val_source_count(struct expr_parse_ctx *ctx, const char *id,
+				double val, int source_count);
 int expr__add_ref(struct expr_parse_ctx *ctx, struct metric_ref *ref);
 int expr__get_id(struct expr_parse_ctx *ctx, const char *id,
 		 struct expr_id_data **data);
@@ -55,5 +57,7 @@ int expr__find_ids(const char *expr, const char *one,
 		   struct expr_parse_ctx *ids);
 
 double expr_id_data__value(const struct expr_id_data *data);
+double expr_id_data__source_count(const struct expr_id_data *data);
+double expr__get_literal(const char *literal);
 
 #endif
diff --git a/tools/perf/util/expr.l b/tools/perf/util/expr.l
index bd20f33418ba..0a13eb20c814 100644
--- a/tools/perf/util/expr.l
+++ b/tools/perf/util/expr.l
@@ -6,6 +6,7 @@
 #include <linux/compiler.h>
 #include "expr.h"
 #include "expr-bison.h"
+#include <math.h>
 
 char *expr_get_text(yyscan_t yyscanner);
 YYSTYPE *expr_get_lval(yyscan_t yyscanner);
@@ -77,6 +78,17 @@ static int str(yyscan_t scanner, int token, int runtime)
 	yylval->str = normalize(yylval->str, runtime);
 	return token;
 }
+
+static int literal(yyscan_t scanner)
+{
+	YYSTYPE *yylval = expr_get_lval(scanner);
+
+	yylval->num = expr__get_literal(expr_get_text(scanner));
+	if (isnan(yylval->num))
+		return EXPR_ERROR;
+
+	return LITERAL;
+}
 %}
 
 number		([0-9]+\.?[0-9]*|[0-9]*\.?[0-9]+)
@@ -85,6 +97,7 @@ sch		[-,=]
 spec		\\{sch}
 sym		[0-9a-zA-Z_\.:@?]+
 symbol		({spec}|{sym})+
+literal		#[0-9a-zA-Z_\.\-]+
 
 %%
 	struct expr_scanner_ctx *sctx = expr_get_extra(yyscanner);
@@ -94,7 +107,8 @@ max		{ return MAX; }
 min		{ return MIN; }
 if		{ return IF; }
 else		{ return ELSE; }
-#smt_on		{ return SMT_ON; }
+source_count	{ return SOURCE_COUNT; }
+{literal}	{ return literal(yyscanner); }
 {number}	{ return value(yyscanner); }
 {symbol}	{ return str(yyscanner, ID, sctx->runtime); }
 "|"		{ return '|'; }
diff --git a/tools/perf/util/expr.y b/tools/perf/util/expr.y
index f969dfa525bd..a30b825adb7b 100644
--- a/tools/perf/util/expr.y
+++ b/tools/perf/util/expr.y
@@ -3,8 +3,8 @@
 #define YYDEBUG 1
 #include <assert.h>
 #include <math.h>
+#include <stdlib.h>
 #include "util/debug.h"
-#include "smt.h"
 #define IN_EXPR_Y 1
 #include "expr.h"
 %}
@@ -37,7 +37,7 @@
 	} ids;
 }
 
-%token ID NUMBER MIN MAX IF ELSE SMT_ON D_RATIO EXPR_ERROR
+%token ID NUMBER MIN MAX IF ELSE LITERAL D_RATIO SOURCE_COUNT EXPR_ERROR
 %left MIN MAX IF
 %left '|'
 %left '^'
@@ -46,7 +46,7 @@
 %left '-' '+'
 %left '*' '/' '%'
 %left NEG NOT
-%type <num> NUMBER
+%type <num> NUMBER LITERAL
 %type <str> ID
 %destructor { free ($$); } <str>
 %type <ids> expr if_expr
@@ -83,6 +83,41 @@ static struct ids union_expr(struct ids ids1, struct ids ids2)
 	return result;
 }
 
+static struct ids handle_id(struct expr_parse_ctx *ctx, char *id,
+			    bool compute_ids, bool source_count)
+{
+	struct ids result;
+
+	if (!compute_ids) {
+		/*
+		 * Compute the event's value from ID. If the ID isn't known then
+		 * it isn't used to compute the formula so set to NAN.
+		 */
+		struct expr_id_data *data;
+
+		result.val = NAN;
+		if (expr__resolve_id(ctx, id, &data) == 0) {
+			result.val = source_count
+				? expr_id_data__source_count(data)
+				: expr_id_data__value(data);
+		}
+		result.ids = NULL;
+		free(id);
+	} else {
+		/*
+		 * Set the value to BOTTOM to show that any value is possible
+		 * when the event is computed. Create a set of just the ID.
+		 */
+		result.val = BOTTOM;
+		result.ids = ids__new();
+		if (!result.ids || ids__insert(result.ids, id)) {
+			pr_err("Error creating IDs for '%s'", id);
+			free(id);
+		}
+	}
+	return result;
+}
+
 /*
  * If we're not computing ids or $1 and $3 are constants, compute the new
  * constant value using OP. Its invariant that there are no ids.  If computing
@@ -168,32 +203,8 @@ expr: NUMBER
 	$$.val = $1;
 	$$.ids = NULL;
 }
-| ID
-{
-	if (!compute_ids) {
-		/*
-		 * Compute the event's value from ID. If the ID isn't known then
-		 * it isn't used to compute the formula so set to NAN.
-		 */
-		struct expr_id_data *data;
-
-		$$.val = NAN;
-		if (expr__resolve_id(ctx, $1, &data) == 0)
-			$$.val = expr_id_data__value(data);
-
-		$$.ids = NULL;
-		free($1);
-	} else {
-		/*
-		 * Set the value to BOTTOM to show that any value is possible
-		 * when the event is computed. Create a set of just the ID.
-		 */
-		$$.val = BOTTOM;
-		$$.ids = ids__new();
-		if (!$$.ids || ids__insert($$.ids, $1))
-			YYABORT;
-	}
-}
+| ID				{ $$ = handle_id(ctx, $1, compute_ids, /*source_count=*/false); }
+| SOURCE_COUNT '(' ID ')'	{ $$ = handle_id(ctx, $3, compute_ids, /*source_count=*/true); }
 | expr '|' expr { BINARY_LONG_OP($$, |, $1, $3); }
 | expr '&' expr { BINARY_LONG_OP($$, &, $1, $3); }
 | expr '^' expr { BINARY_LONG_OP($$, ^, $1, $3); }
@@ -280,9 +291,9 @@ expr: NUMBER
 		$$ = union_expr($3, $5);
 	}
 }
-| SMT_ON
+| LITERAL
 {
-	$$.val = smt_on() > 0 ? 1.0 : 0.0;
+	$$.val = $1;
 	$$.ids = NULL;
 }
 ;
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 56511db8fa03..fda8d14c891f 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -583,21 +583,21 @@ static int write_cpu_topology(struct feat_fd *ff,
 	if (!tp)
 		return -1;
 
-	ret = do_write(ff, &tp->core_sib, sizeof(tp->core_sib));
+	ret = do_write(ff, &tp->package_cpus_lists, sizeof(tp->package_cpus_lists));
 	if (ret < 0)
 		goto done;
 
-	for (i = 0; i < tp->core_sib; i++) {
-		ret = do_write_string(ff, tp->core_siblings[i]);
+	for (i = 0; i < tp->package_cpus_lists; i++) {
+		ret = do_write_string(ff, tp->package_cpus_list[i]);
 		if (ret < 0)
 			goto done;
 	}
-	ret = do_write(ff, &tp->thread_sib, sizeof(tp->thread_sib));
+	ret = do_write(ff, &tp->core_cpus_lists, sizeof(tp->core_cpus_lists));
 	if (ret < 0)
 		goto done;
 
-	for (i = 0; i < tp->thread_sib; i++) {
-		ret = do_write_string(ff, tp->thread_siblings[i]);
+	for (i = 0; i < tp->core_cpus_lists; i++) {
+		ret = do_write_string(ff, tp->core_cpus_list[i]);
 		if (ret < 0)
 			break;
 	}
@@ -617,15 +617,15 @@ static int write_cpu_topology(struct feat_fd *ff,
 			return ret;
 	}
 
-	if (!tp->die_sib)
+	if (!tp->die_cpus_lists)
 		goto done;
 
-	ret = do_write(ff, &tp->die_sib, sizeof(tp->die_sib));
+	ret = do_write(ff, &tp->die_cpus_lists, sizeof(tp->die_cpus_lists));
 	if (ret < 0)
 		goto done;
 
-	for (i = 0; i < tp->die_sib; i++) {
-		ret = do_write_string(ff, tp->die_siblings[i]);
+	for (i = 0; i < tp->die_cpus_lists; i++) {
+		ret = do_write_string(ff, tp->die_cpus_list[i]);
 		if (ret < 0)
 			goto done;
 	}
diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
index e4fb02b05130..5c7308efa768 100644
--- a/tools/perf/util/stat-shadow.c
+++ b/tools/perf/util/stat-shadow.c
@@ -829,10 +829,12 @@ static int prepare_metric(struct evsel **metric_events,
 		struct saved_value *v;
 		struct stats *stats;
 		u64 metric_total = 0;
+		int source_count;
 
 		if (!strcmp(metric_events[i]->name, "duration_time")) {
 			stats = &walltime_nsecs_stats;
 			scale = 1e-9;
+			source_count = 1;
 		} else {
 			v = saved_value_lookup(metric_events[i], cpu, false,
 					       STAT_NONE, 0, st,
@@ -841,6 +843,7 @@ static int prepare_metric(struct evsel **metric_events,
 				break;
 			stats = &v->stats;
 			scale = 1.0;
+			source_count = evsel__source_count(metric_events[i]);
 
 			if (v->metric_other)
 				metric_total = v->metric_total;
@@ -849,7 +852,9 @@ static int prepare_metric(struct evsel **metric_events,
 		if (!n)
 			return -ENOMEM;
 
-		expr__add_id_val(pctx, n, metric_total ? : avg_stats(stats) * scale);
+		expr__add_id_val_source_count(pctx, n,
+					metric_total ? : avg_stats(stats) * scale,
+					source_count);
 	}
 
 	for (j = 0; metric_refs && metric_refs[j].metric_name; j++) {
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index aa1b7c12fd61..b2ed3140a1fa 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -274,7 +274,7 @@ struct symbol *symbol__new(u64 start, u64 len, u8 binding, u8 type, const char *
 	if (symbol_conf.priv_size) {
 		if (symbol_conf.init_annotation) {
 			struct annotation *notes = (void *)sym;
-			pthread_mutex_init(&notes->lock, NULL);
+			annotation__init(notes);
 		}
 		sym = ((void *)sym) + symbol_conf.priv_size;
 	}
@@ -294,6 +294,13 @@ struct symbol *symbol__new(u64 start, u64 len, u8 binding, u8 type, const char *
 
 void symbol__delete(struct symbol *sym)
 {
+	if (symbol_conf.priv_size) {
+		if (symbol_conf.init_annotation) {
+			struct annotation *notes = symbol__annotation(sym);
+
+			annotation__exit(notes);
+		}
+	}
 	free(((void *)sym) - symbol_conf.priv_size);
 }
 
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 166196686f2e..fbf866d82dcc 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -40,22 +40,33 @@ Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
 			     GElf_Shdr *shp, const char *name, size_t *idx);
 #endif
 
-/** struct symbol - symtab entry
- *
- * @ignore - resolvable but tools ignore it (e.g. idle routines)
+/**
+ * A symtab entry. When allocated this may be preceded by an annotation (see
+ * symbol__annotation), a browser_index (see symbol__browser_index) and rb_node
+ * to sort by name (see struct symbol_name_rb_node).
  */
 struct symbol {
 	struct rb_node	rb_node;
+	/** Range of symbol [start, end). */
 	u64		start;
 	u64		end;
+	/** Length of the string name. */
 	u16		namelen;
+	/** ELF symbol type as defined for st_info. E.g STT_OBJECT or STT_FUNC. */
 	u8		type:4;
+	/** ELF binding type as defined for st_info. E.g. STB_WEAK or STB_GLOBAL. */
 	u8		binding:4;
+	/** Set true for kernel symbols of idle routines. */
 	u8		idle:1;
+	/** Resolvable but tools ignore it (e.g. idle routines). */
 	u8		ignore:1;
+	/** Symbol for an inlined function. */
 	u8		inlined:1;
+	/** Has symbol__annotate2 been performed. */
+	u8		annotate2:1;
+	/** Architecture specific. Unused except on PPC where it holds st_other. */
 	u8		arch_sym;
-	bool		annotate2;
+	/** The name of length namelen associated with the symbol. */
 	char		name[];
 };