perf_counter tools: Add 'perf annotate' feature

Add new perf sub-command to display annotated source code:

 $ perf annotate decode_tree_entry

------------------------------------------------
 Percent |	Source code & Disassembly of /home/mingo/git/git
------------------------------------------------
         :
         :	/home/mingo/git/git:     file format elf64-x86-64
         :
         :
         :	Disassembly of section .text:
         :
         :	00000000004a0da0 <decode_tree_entry>:
         :		*modep = mode;
         :		return str;
         :	}
         :
         :	static void decode_tree_entry(struct tree_desc *desc, const char *buf, unsigned long size)
         :	{
    3.82 :	  4a0da0:	41 54                	push   %r12
         :		const char *path;
         :		unsigned int mode, len;
         :
         :		if (size < 24 || buf[size - 21])
    0.17 :	  4a0da2:	48 83 fa 17          	cmp    $0x17,%rdx
         :		*modep = mode;
         :		return str;
         :	}
         :
         :	static void decode_tree_entry(struct tree_desc *desc, const char *buf, unsigned long size)
         :	{
    0.00 :	  4a0da6:	49 89 fc             	mov    %rdi,%r12
    0.00 :	  4a0da9:	55                   	push   %rbp
    3.37 :	  4a0daa:	53                   	push   %rbx
         :		const char *path;
         :		unsigned int mode, len;
         :
         :		if (size < 24 || buf[size - 21])
    0.08 :	  4a0dab:	76 73                	jbe    4a0e20 <decode_tree_entry+0x80>
    0.00 :	  4a0dad:	80 7c 16 eb 00       	cmpb   $0x0,-0x15(%rsi,%rdx,1)
    3.48 :	  4a0db2:	75 6c                	jne    4a0e20 <decode_tree_entry+0x80>
         :	static const char *get_mode(const char *str, unsigned int *modep)
         :	{
         :		unsigned char c;
         :		unsigned int mode = 0;
         :
         :		if (*str == ' ')
    1.94 :	  4a0db4:	0f b6 06             	movzbl (%rsi),%eax
    0.39 :	  4a0db7:	3c 20                	cmp    $0x20,%al
    0.00 :	  4a0db9:	74 65                	je     4a0e20 <decode_tree_entry+0x80>
         :			return NULL;
         :
         :		while ((c = *str++) != ' ') {
    0.06 :	  4a0dbb:	89 c2                	mov    %eax,%edx
         :			if (c < '0' || c > '7')
    1.99 :	  4a0dbd:	31 ed                	xor    %ebp,%ebp
         :		unsigned int mode = 0;
         :
         :		if (*str == ' ')
         :			return NULL;
         :
         :		while ((c = *str++) != ' ') {
    1.74 :	  4a0dbf:	48 8d 5e 01          	lea    0x1(%rsi),%rbx
         :			if (c < '0' || c > '7')
    0.00 :	  4a0dc3:	8d 42 d0             	lea    -0x30(%rdx),%eax
    0.17 :	  4a0dc6:	3c 07                	cmp    $0x7,%al
    0.00 :	  4a0dc8:	76 0d                	jbe    4a0dd7 <decode_tree_entry+0x37>
    0.00 :	  4a0dca:	eb 54                	jmp    4a0e20 <decode_tree_entry+0x80>
    0.00 :	  4a0dcc:	0f 1f 40 00          	nopl   0x0(%rax)
   16.57 :	  4a0dd0:	8d 42 d0             	lea    -0x30(%rdx),%eax
    0.14 :	  4a0dd3:	3c 07                	cmp    $0x7,%al
    0.00 :	  4a0dd5:	77 49                	ja     4a0e20 <decode_tree_entry+0x80>
         :				return NULL;
         :			mode = (mode << 3) + (c - '0');
    3.12 :	  4a0dd7:	0f b6 c2             	movzbl %dl,%eax
         :		unsigned int mode = 0;
         :
         :		if (*str == ' ')
         :			return NULL;
         :
         :		while ((c = *str++) != ' ') {
    0.00 :	  4a0dda:	0f b6 13             	movzbl (%rbx),%edx
   16.74 :	  4a0ddd:	48 83 c3 01          	add    $0x1,%rbx
         :			if (c < '0' || c > '7')
         :				return NULL;
         :			mode = (mode << 3) + (c - '0');

The first column is the percentage of samples that arrived on that
particular line - relative to the total cost of the function.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

2 files changed, 39 insertions, 16 deletions

diff --git a/Documentation/perf_counter/util/symbol.c b/Documentation/perf_counter/util/symbol.c
index a06bbfba835..23f4f7b3b83 100644
--- a/Documentation/perf_counter/util/symbol.c
+++ b/Documentation/perf_counter/util/symbol.c
@@ -7,21 +7,36 @@
 #include <gelf.h>
 #include <elf.h>
 
+const char *sym_hist_filter;
+
 static struct symbol *symbol__new(uint64_t start, uint64_t len,
-				  const char *name, unsigned int priv_size)
+				  const char *name, unsigned int priv_size,
+				  uint64_t obj_start, int verbose)
 {
 	size_t namelen = strlen(name) + 1;
-	struct symbol *self = malloc(priv_size + sizeof(*self) + namelen);
+	struct symbol *self = calloc(1, priv_size + sizeof(*self) + namelen);
 
-	if (self != NULL) {
-		if (priv_size) {
-			memset(self, 0, priv_size);
-			self = ((void *)self) + priv_size;
-		}
-		self->start = start;
-		self->end   = start + len - 1;
-		memcpy(self->name, name, namelen);
+	if (!self)
+		return NULL;
+
+	if (verbose >= 2)
+		printf("new symbol: %016Lx [%08lx]: %s, hist: %p, obj_start: %p\n",
+			(__u64)start, len, name, self->hist, (void *)obj_start);
+
+	self->obj_start= obj_start;
+	self->hist = NULL;
+	self->hist_sum = 0;
+
+	if (sym_hist_filter && !strcmp(name, sym_hist_filter))
+		self->hist = calloc(sizeof(__u64), len);
+
+	if (priv_size) {
+		memset(self, 0, priv_size);
+		self = ((void *)self) + priv_size;
 	}
+	self->start = start;
+	self->end   = start + len - 1;
+	memcpy(self->name, name, namelen);
 
 	return self;
 }
@@ -166,7 +181,7 @@ static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int verb
 		 * Well fix up the end later, when we have all sorted.
 		 */
 		sym = symbol__new(start, 0xdead, line + len + 2,
-				  self->sym_priv_size);
+				  self->sym_priv_size, 0, verbose);
 
 		if (sym == NULL)
 			goto out_delete_line;
@@ -272,7 +287,7 @@ static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
 static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
 				       GElf_Ehdr *ehdr, Elf_Scn *scn_dynsym,
 				       GElf_Shdr *shdr_dynsym,
-				       size_t dynsym_idx)
+				       size_t dynsym_idx, int verbose)
 {
 	uint32_t nr_rel_entries, idx;
 	GElf_Sym sym;
@@ -335,7 +350,7 @@ static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
 				 "%s@plt", elf_sym__name(&sym, symstrs));
 
 			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-					sympltname, self->sym_priv_size);
+					sympltname, self->sym_priv_size, 0, verbose);
 			if (!f)
 				return -1;
 
@@ -353,7 +368,7 @@ static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
 				 "%s@plt", elf_sym__name(&sym, symstrs));
 
 			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-					sympltname, self->sym_priv_size);
+					sympltname, self->sym_priv_size, 0, verbose);
 			if (!f)
 				return -1;
 
@@ -410,7 +425,7 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 	if (sec_dynsym != NULL) {
 		nr = dso__synthesize_plt_symbols(self, elf, &ehdr,
 						 sec_dynsym, &shdr,
-						 dynsym_idx);
+						 dynsym_idx, verbose);
 		if (nr < 0)
 			goto out_elf_end;
 	}
@@ -444,6 +459,7 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 
 	elf_symtab__for_each_symbol(syms, nr_syms, index, sym) {
 		struct symbol *f;
+		uint64_t obj_start;
 
 		if (!elf_sym__is_function(&sym))
 			continue;
@@ -453,11 +469,13 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 			goto out_elf_end;
 
 		gelf_getshdr(sec, &shdr);
+		obj_start = sym.st_value;
+
 		sym.st_value -= shdr.sh_addr - shdr.sh_offset;
 
 		f = symbol__new(sym.st_value, sym.st_size,
 				elf_sym__name(&sym, symstrs),
-				self->sym_priv_size);
+				self->sym_priv_size, obj_start, verbose);
 		if (!f)
 			goto out_elf_end;
 
diff --git a/Documentation/perf_counter/util/symbol.h b/Documentation/perf_counter/util/symbol.h
index e23cc312668..4839d68f14f 100644
--- a/Documentation/perf_counter/util/symbol.h
+++ b/Documentation/perf_counter/util/symbol.h
@@ -9,6 +9,9 @@ struct symbol {
 	struct rb_node	rb_node;
 	__u64		start;
 	__u64		end;
+	__u64		obj_start;
+	__u64		hist_sum;
+	__u64		*hist;
 	char		name[0];
 };
 
@@ -20,6 +23,8 @@ struct dso {
 	char		 name[0];
 };
 
+const char *sym_hist_filter;
+
 typedef int (*symbol_filter_t)(struct dso *self, struct symbol *sym);
 
 struct dso *dso__new(const char *name, unsigned int sym_priv_size);