Merge tag 'v6.14-rc1' into perf-tools-next

To get the various fixes in the current master.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>

8 files changed, 139 insertions, 90 deletions

diff --git a/tools/perf/Documentation/perf-arm-spe.txt b/tools/perf/Documentation/perf-arm-spe.txt
index de2b0b479249..37afade4f1b2 100644
--- a/tools/perf/Documentation/perf-arm-spe.txt
+++ b/tools/perf/Documentation/perf-arm-spe.txt
@@ -150,6 +150,7 @@ arm_spe/load_filter=1,min_latency=10/'
   pct_enable=1        - collect physical timestamp instead of virtual timestamp (PMSCR.PCT) - requires privilege
   store_filter=1      - collect stores only (PMSFCR.ST)
   ts_enable=1         - enable timestamping with value of generic timer (PMSCR.TS)
+  discard=1           - enable SPE PMU events but don't collect sample data - see 'Discard mode' (PMBLIMITR.FM = DISCARD)
 
 +++*+++ Latency is the total latency from the point at which sampling started on that instruction, rather
 than only the execution latency.
@@ -220,6 +221,31 @@ Common errors
 
    Increase sampling interval (see above)
 
+PMU events
+~~~~~~~~~~
+
+SPE has events that can be counted on core PMUs. These are prefixed with
+SAMPLE_, for example SAMPLE_POP, SAMPLE_FEED, SAMPLE_COLLISION and
+SAMPLE_FEED_BR.
+
+These events will only count when an SPE event is running on the same core that
+the PMU event is opened on, otherwise they read as 0. There are various ways to
+ensure that the PMU event and SPE event are scheduled together depending on the
+way the event is opened. For example opening both events as per-process events
+on the same process, although it's not guaranteed that the PMU event is enabled
+first when context switching. For that reason it may be better to open the PMU
+event as a systemwide event and then open SPE on the process of interest.
+
+Discard mode
+~~~~~~~~~~~~
+
+SPE related (SAMPLE_* etc) core PMU events can be used without the overhead of
+collecting sample data if discard mode is supported (optional from Armv8.6).
+First run a system wide SPE session (or on the core of interest) using options
+to minimize output. Then run perf stat:
+
+  perf record -e arm_spe/discard/ -a -N -B --no-bpf-event -o - > /dev/null &
+  perf stat -e SAMPLE_FEED_LD
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index ac97632f13dc..06356217adeb 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2107,8 +2107,12 @@ static int trace__read_syscall_info(struct trace *trace, int id)
 		return PTR_ERR(sc->tp_format);
 	}
 
+	/*
+	 * The tracepoint format contains __syscall_nr field, so it's one more
+	 * than the actual number of syscall arguments.
+	 */
 	if (syscall__alloc_arg_fmts(sc, IS_ERR(sc->tp_format) ?
-					RAW_SYSCALL_ARGS_NUM : sc->tp_format->format.nr_fields))
+					RAW_SYSCALL_ARGS_NUM : sc->tp_format->format.nr_fields - 1))
 		return -ENOMEM;
 
 	sc->args = sc->tp_format->format.fields;
diff --git a/tools/perf/tests/shell/trace_btf_enum.sh b/tools/perf/tests/shell/trace_btf_enum.sh
index 5a3b8a5a9b5c..8d1e6bbeac90 100755
--- a/tools/perf/tests/shell/trace_btf_enum.sh
+++ b/tools/perf/tests/shell/trace_btf_enum.sh
@@ -26,8 +26,12 @@ check_vmlinux() {
 trace_landlock() {
   echo "Tracing syscall ${syscall}"
 
-  # test flight just to see if landlock_add_rule and libbpf are available
-  $TESTPROG
+  # test flight just to see if landlock_add_rule is available
+  if ! perf trace $TESTPROG 2>&1 | grep -q landlock
+  then
+    echo "No landlock system call found, skipping to non-syscall tracing."
+    return
+  fi
 
   if perf trace -e $syscall $TESTPROG 2>&1 | \
      grep -q -E ".*landlock_add_rule\(ruleset_fd: 11, rule_type: (LANDLOCK_RULE_PATH_BENEATH|LANDLOCK_RULE_NET_PORT), rule_attr: 0x[a-f0-9]+, flags: 45\) = -1.*"
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 0d2ea22bd9e4..31bb326b07a6 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -2100,6 +2100,57 @@ int symbol__annotate2(struct map_symbol *ms, struct evsel *evsel,
 	return 0;
 }
 
+const char * const perf_disassembler__strs[] = {
+	[PERF_DISASM_UNKNOWN]  = "unknown",
+	[PERF_DISASM_LLVM]     = "llvm",
+	[PERF_DISASM_CAPSTONE] = "capstone",
+	[PERF_DISASM_OBJDUMP]  = "objdump",
+};
+
+
+static void annotation_options__add_disassembler(struct annotation_options *options,
+						 enum perf_disassembler dis)
+{
+	for (u8 i = 0; i < ARRAY_SIZE(options->disassemblers); i++) {
+		if (options->disassemblers[i] == dis) {
+			/* Disassembler is already present then don't add again. */
+			return;
+		}
+		if (options->disassemblers[i] == PERF_DISASM_UNKNOWN) {
+			/* Found a free slot. */
+			options->disassemblers[i] = dis;
+			return;
+		}
+	}
+	pr_err("Failed to add disassembler %d\n", dis);
+}
+
+static int annotation_options__add_disassemblers_str(struct annotation_options *options,
+						const char *str)
+{
+	while (str && *str != '\0') {
+		const char *comma = strchr(str, ',');
+		int len = comma ? comma - str : (int)strlen(str);
+		bool match = false;
+
+		for (u8 i = 0; i < ARRAY_SIZE(perf_disassembler__strs); i++) {
+			const char *dis_str = perf_disassembler__strs[i];
+
+			if (len == (int)strlen(dis_str) && !strncmp(str, dis_str, len)) {
+				annotation_options__add_disassembler(options, i);
+				match = true;
+				break;
+			}
+		}
+		if (!match) {
+			pr_err("Invalid disassembler '%.*s'\n", len, str);
+			return -1;
+		}
+		str = comma ? comma + 1 : NULL;
+	}
+	return 0;
+}
+
 static int annotation__config(const char *var, const char *value, void *data)
 {
 	struct annotation_options *opt = data;
@@ -2115,11 +2166,10 @@ static int annotation__config(const char *var, const char *value, void *data)
 		else if (opt->offset_level < ANNOTATION__MIN_OFFSET_LEVEL)
 			opt->offset_level = ANNOTATION__MIN_OFFSET_LEVEL;
 	} else if (!strcmp(var, "annotate.disassemblers")) {
-		opt->disassemblers_str = strdup(value);
-		if (!opt->disassemblers_str) {
-			pr_err("Not enough memory for annotate.disassemblers\n");
-			return -1;
-		}
+		int err = annotation_options__add_disassemblers_str(opt, value);
+
+		if (err)
+			return err;
 	} else if (!strcmp(var, "annotate.hide_src_code")) {
 		opt->hide_src_code = perf_config_bool("hide_src_code", value);
 	} else if (!strcmp(var, "annotate.jump_arrows")) {
@@ -2185,9 +2235,25 @@ void annotation_options__exit(void)
 	zfree(&annotate_opts.objdump_path);
 }
 
+static void annotation_options__default_init_disassemblers(struct annotation_options *options)
+{
+	if (options->disassemblers[0] != PERF_DISASM_UNKNOWN) {
+		/* Already initialized. */
+		return;
+	}
+#ifdef HAVE_LIBLLVM_SUPPORT
+	annotation_options__add_disassembler(options, PERF_DISASM_LLVM);
+#endif
+#ifdef HAVE_LIBCAPSTONE_SUPPORT
+	annotation_options__add_disassembler(options, PERF_DISASM_CAPSTONE);
+#endif
+	annotation_options__add_disassembler(options, PERF_DISASM_OBJDUMP);
+}
+
 void annotation_config__init(void)
 {
 	perf_config(annotation__config, &annotate_opts);
+	annotation_options__default_init_disassemblers(&annotate_opts);
 }
 
 static unsigned int parse_percent_type(char *str1, char *str2)
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
index 0ba5846dad4d..98db1b88daf4 100644
--- a/tools/perf/util/annotate.h
+++ b/tools/perf/util/annotate.h
@@ -34,8 +34,13 @@ struct annotated_data_type;
 #define ANNOTATION__BR_CNTR_WIDTH 30
 #define ANNOTATION_DUMMY_LEN	256
 
-// llvm, capstone, objdump
-#define MAX_DISASSEMBLERS 3
+enum perf_disassembler {
+	PERF_DISASM_UNKNOWN = 0,
+	PERF_DISASM_LLVM,
+	PERF_DISASM_CAPSTONE,
+	PERF_DISASM_OBJDUMP,
+};
+#define MAX_DISASSEMBLERS (PERF_DISASM_OBJDUMP + 1)
 
 struct annotation_options {
 	bool hide_src_code,
@@ -52,14 +57,12 @@ struct annotation_options {
 	     annotate_src,
 	     full_addr;
 	u8   offset_level;
-	u8   nr_disassemblers;
+	u8   disassemblers[MAX_DISASSEMBLERS];
 	int  min_pcnt;
 	int  max_lines;
 	int  context;
 	char *objdump_path;
 	char *disassembler_style;
-	const char *disassemblers_str;
-	const char *disassemblers[MAX_DISASSEMBLERS];
 	const char *prefix;
 	const char *prefix_strip;
 	unsigned int percent_type;
@@ -134,6 +137,8 @@ struct disasm_line {
 	struct annotation_line	 al;
 };
 
+extern const char * const perf_disassembler__strs[];
+
 void annotation_line__add(struct annotation_line *al, struct list_head *head);
 
 static inline double annotation_data__percent(struct annotation_data *data,
diff --git a/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c b/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c
index 4a62ed593e84..e4352881e3fa 100644
--- a/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c
+++ b/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c
@@ -431,9 +431,9 @@ static bool pid_filter__has(struct pids_filtered *pids, pid_t pid)
 static int augment_sys_enter(void *ctx, struct syscall_enter_args *args)
 {
 	bool augmented, do_output = false;
-	int zero = 0, size, aug_size, index,
-	    value_size = sizeof(struct augmented_arg) - offsetof(struct augmented_arg, value);
+	int zero = 0, index, value_size = sizeof(struct augmented_arg) - offsetof(struct augmented_arg, value);
 	u64 output = 0; /* has to be u64, otherwise it won't pass the verifier */
+	s64 aug_size, size;
 	unsigned int nr, *beauty_map;
 	struct beauty_payload_enter *payload;
 	void *arg, *payload_offset;
@@ -484,14 +484,11 @@ static int augment_sys_enter(void *ctx, struct syscall_enter_args *args)
 		} else if (size > 0 && size <= value_size) { /* struct */
 			if (!bpf_probe_read_user(((struct augmented_arg *)payload_offset)->value, size, arg))
 				augmented = true;
-		} else if (size < 0 && size >= -6) { /* buffer */
+		} else if ((int)size < 0 && size >= -6) { /* buffer */
 			index = -(size + 1);
 			barrier_var(index); // Prevent clang (noticed with v18) from removing the &= 7 trick.
 			index &= 7;	    // Satisfy the bounds checking with the verifier in some kernels.
-			aug_size = args->args[index];
-
-			if (aug_size > TRACE_AUG_MAX_BUF)
-				aug_size = TRACE_AUG_MAX_BUF;
+			aug_size = args->args[index] > TRACE_AUG_MAX_BUF ? TRACE_AUG_MAX_BUF : args->args[index];
 
 			if (aug_size > 0) {
 				if (!bpf_probe_read_user(((struct augmented_arg *)payload_offset)->value, aug_size, arg))
diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
index 27094211edd8..5c329ad614e9 100644
--- a/tools/perf/util/cpumap.c
+++ b/tools/perf/util/cpumap.c
@@ -293,7 +293,7 @@ struct aggr_cpu_id aggr_cpu_id__die(struct perf_cpu cpu, void *data)
 
 	die = cpu__get_die_id(cpu);
 	/* There is no die_id on legacy system. */
-	if (die == -1)
+	if (die < 0)
 		die = 0;
 
 	/*
@@ -322,7 +322,7 @@ struct aggr_cpu_id aggr_cpu_id__cluster(struct perf_cpu cpu, void *data)
 	struct aggr_cpu_id id;
 
 	/* There is no cluster_id on legacy system. */
-	if (cluster == -1)
+	if (cluster < 0)
 		cluster = 0;
 
 	id = aggr_cpu_id__die(cpu, data);
diff --git a/tools/perf/util/disasm.c b/tools/perf/util/disasm.c
index b7de4d9fd004..50c5c206b70e 100644
--- a/tools/perf/util/disasm.c
+++ b/tools/perf/util/disasm.c
@@ -2216,56 +2216,6 @@ out_free_command:
 	return err;
 }
 
-static int annotation_options__init_disassemblers(struct annotation_options *options)
-{
-	char *disassembler;
-
-	if (options->disassemblers_str == NULL) {
-		const char *default_disassemblers_str =
-#ifdef HAVE_LIBLLVM_SUPPORT
-				"llvm,"
-#endif
-#ifdef HAVE_LIBCAPSTONE_SUPPORT
-				"capstone,"
-#endif
-				"objdump";
-
-		options->disassemblers_str = strdup(default_disassemblers_str);
-		if (!options->disassemblers_str)
-			goto out_enomem;
-	}
-
-	disassembler = strdup(options->disassemblers_str);
-	if (disassembler == NULL)
-		goto out_enomem;
-
-	while (1) {
-		char *comma = strchr(disassembler, ',');
-
-		if (comma != NULL)
-			*comma = '\0';
-
-		options->disassemblers[options->nr_disassemblers++] = strim(disassembler);
-
-		if (comma == NULL)
-			break;
-
-		disassembler = comma + 1;
-
-		if (options->nr_disassemblers >= MAX_DISASSEMBLERS) {
-			pr_debug("annotate.disassemblers can have at most %d entries, ignoring \"%s\"\n",
-				 MAX_DISASSEMBLERS, disassembler);
-			break;
-		}
-	}
-
-	return 0;
-
-out_enomem:
-	pr_err("Not enough memory for annotate.disassemblers\n");
-	return -1;
-}
-
 int symbol__disassemble(struct symbol *sym, struct annotate_args *args)
 {
 	struct annotation_options *options = args->options;
@@ -2274,7 +2224,6 @@ int symbol__disassemble(struct symbol *sym, struct annotate_args *args)
 	char symfs_filename[PATH_MAX];
 	bool delete_extract = false;
 	struct kcore_extract kce;
-	const char *disassembler;
 	bool decomp = false;
 	int err = dso__disassemble_filename(dso, symfs_filename, sizeof(symfs_filename));
 
@@ -2334,28 +2283,26 @@ int symbol__disassemble(struct symbol *sym, struct annotate_args *args)
 		}
 	}
 
-	err = annotation_options__init_disassemblers(options);
-	if (err)
-		goto out_remove_tmp;
-
 	err = -1;
+	for (u8 i = 0; i < ARRAY_SIZE(options->disassemblers) && err != 0; i++) {
+		enum perf_disassembler dis = options->disassemblers[i];
 
-	for (int i = 0; i < options->nr_disassemblers && err != 0; ++i) {
-		disassembler = options->disassemblers[i];
-
-		if (!strcmp(disassembler, "llvm"))
+		switch (dis) {
+		case PERF_DISASM_LLVM:
 			err = symbol__disassemble_llvm(symfs_filename, sym, args);
-		else if (!strcmp(disassembler, "capstone"))
+			break;
+		case PERF_DISASM_CAPSTONE:
 			err = symbol__disassemble_capstone(symfs_filename, sym, args);
-		else if (!strcmp(disassembler, "objdump"))
+			break;
+		case PERF_DISASM_OBJDUMP:
 			err = symbol__disassemble_objdump(symfs_filename, sym, args);
-		else
-			pr_debug("Unknown disassembler %s, skipping...\n", disassembler);
-	}
-
-	if (err == 0) {
-		pr_debug("Disassembled with %s\nannotate.disassemblers=%s\n",
-			 disassembler, options->disassemblers_str);
+			break;
+		case PERF_DISASM_UNKNOWN: /* End of disassemblers. */
+		default:
+			goto out_remove_tmp;
+		}
+		if (err == 0)
+			pr_debug("Disassembled with %s\n", perf_disassembler__strs[dis]);
 	}
 out_remove_tmp:
 	if (decomp)