linux-zen-desktop/tools/perf/util/annotate.c

// SPDX-License-Identifier: GPL-2.0-only
/*
 * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
 *
 * Parts came from builtin-annotate.c, see those files for further
 * copyright notes.
 */

#include <errno.h>
#include <inttypes.h>
#include <libgen.h>
#include <stdlib.h>
#include "util.h" // hex_width()
#include "ui/ui.h"
#include "sort.h"
#include "build-id.h"
#include "color.h"
#include "config.h"
#include "dso.h"
#include "env.h"
#include "map.h"
#include "maps.h"
#include "symbol.h"
#include "srcline.h"
#include "units.h"
#include "debug.h"
#include "annotate.h"
#include "evsel.h"
#include "evlist.h"
#include "bpf-event.h"
#include "bpf-utils.h"
#include "block-range.h"
#include "string2.h"
#include "util/event.h"
#include "arch/common.h"
#include "namespaces.h"
#include <regex.h>
#include <linux/bitops.h>
#include <linux/kernel.h>
#include <linux/string.h>
#include <subcmd/parse-options.h>
#include <subcmd/run-command.h>

/* FIXME: For the HE_COLORSET */
#include "ui/browser.h"

/*
 * FIXME: Using the same values as slang.h,
 * but that header may not be available everywhere
 */
#define LARROW_CHAR	((unsigned char)',')
#define RARROW_CHAR	((unsigned char)'+')
#define DARROW_CHAR	((unsigned char)'.')
#define UARROW_CHAR	((unsigned char)'-')

#include <linux/ctype.h>

struct annotation_options annotation__default_options = {
	.use_offset     = true,
	.jump_arrows    = true,
	.annotate_src	= true,
	.offset_level	= ANNOTATION__OFFSET_JUMP_TARGETS,
	.percent_type	= PERCENT_PERIOD_LOCAL,
};

static regex_t	 file_lineno;

static struct ins_ops *ins__find(struct arch *arch, const char *name);
static void ins__sort(struct arch *arch);
static int disasm_line__parse(char *line, const char **namep, char **rawp);

struct arch {
	const char	*name;
	struct ins	*instructions;
	size_t		nr_instructions;
	size_t		nr_instructions_allocated;
	struct ins_ops  *(*associate_instruction_ops)(struct arch *arch, const char *name);
	bool		sorted_instructions;
	bool		initialized;
	void		*priv;
	unsigned int	model;
	unsigned int	family;
	int		(*init)(struct arch *arch, char *cpuid);
	bool		(*ins_is_fused)(struct arch *arch, const char *ins1,
					const char *ins2);
	struct		{
		char comment_char;
		char skip_functions_char;
	} objdump;
};

static struct ins_ops call_ops;
static struct ins_ops dec_ops;
static struct ins_ops jump_ops;
static struct ins_ops mov_ops;
static struct ins_ops nop_ops;
static struct ins_ops lock_ops;
static struct ins_ops ret_ops;

static int arch__grow_instructions(struct arch *arch)
{
	struct ins *new_instructions;
	size_t new_nr_allocated;

	if (arch->nr_instructions_allocated == 0 && arch->instructions)
		goto grow_from_non_allocated_table;

	new_nr_allocated = arch->nr_instructions_allocated + 128;
	new_instructions = realloc(arch->instructions, new_nr_allocated * sizeof(struct ins));
	if (new_instructions == NULL)
		return -1;

out_update_instructions:
	arch->instructions = new_instructions;
	arch->nr_instructions_allocated = new_nr_allocated;
	return 0;

grow_from_non_allocated_table:
	new_nr_allocated = arch->nr_instructions + 128;
	new_instructions = calloc(new_nr_allocated, sizeof(struct ins));
	if (new_instructions == NULL)
		return -1;

	memcpy(new_instructions, arch->instructions, arch->nr_instructions);
	goto out_update_instructions;
}

static int arch__associate_ins_ops(struct arch* arch, const char *name, struct ins_ops *ops)
{
	struct ins *ins;

	if (arch->nr_instructions == arch->nr_instructions_allocated &&
	    arch__grow_instructions(arch))
		return -1;

	ins = &arch->instructions[arch->nr_instructions];
	ins->name = strdup(name);
	if (!ins->name)
		return -1;

	ins->ops  = ops;
	arch->nr_instructions++;

	ins__sort(arch);
	return 0;
}

#include "arch/arc/annotate/instructions.c"
#include "arch/arm/annotate/instructions.c"
#include "arch/arm64/annotate/instructions.c"
#include "arch/csky/annotate/instructions.c"
#include "arch/mips/annotate/instructions.c"
#include "arch/x86/annotate/instructions.c"
#include "arch/powerpc/annotate/instructions.c"
#include "arch/riscv64/annotate/instructions.c"
#include "arch/s390/annotate/instructions.c"
#include "arch/sparc/annotate/instructions.c"

static struct arch architectures[] = {
	{
		.name = "arc",
		.init = arc__annotate_init,
	},
	{
		.name = "arm",
		.init = arm__annotate_init,
	},
	{
		.name = "arm64",
		.init = arm64__annotate_init,
	},
	{
		.name = "csky",
		.init = csky__annotate_init,
	},
	{
		.name = "mips",
		.init = mips__annotate_init,
		.objdump = {
			.comment_char = '#',
		},
	},
	{
		.name = "x86",
		.init = x86__annotate_init,
		.instructions = x86__instructions,
		.nr_instructions = ARRAY_SIZE(x86__instructions),
		.objdump =  {
			.comment_char = '#',
		},
	},
	{
		.name = "powerpc",
		.init = powerpc__annotate_init,
	},
	{
		.name = "riscv64",
		.init = riscv64__annotate_init,
	},
	{
		.name = "s390",
		.init = s390__annotate_init,
		.objdump =  {
			.comment_char = '#',
		},
	},
	{
		.name = "sparc",
		.init = sparc__annotate_init,
		.objdump = {
			.comment_char = '#',
		},
	},
};

static void ins__delete(struct ins_operands *ops)
{
	if (ops == NULL)
		return;
	zfree(&ops->source.raw);
	zfree(&ops->source.name);
	zfree(&ops->target.raw);
	zfree(&ops->target.name);
}

static int ins__raw_scnprintf(struct ins *ins, char *bf, size_t size,
			      struct ins_operands *ops, int max_ins_name)
{
	return scnprintf(bf, size, "%-*s %s", max_ins_name, ins->name, ops->raw);
}

int ins__scnprintf(struct ins *ins, char *bf, size_t size,
		   struct ins_operands *ops, int max_ins_name)
{
	if (ins->ops->scnprintf)
		return ins->ops->scnprintf(ins, bf, size, ops, max_ins_name);

	return ins__raw_scnprintf(ins, bf, size, ops, max_ins_name);
}

bool ins__is_fused(struct arch *arch, const char *ins1, const char *ins2)
{
	if (!arch || !arch->ins_is_fused)
		return false;

	return arch->ins_is_fused(arch, ins1, ins2);
}

static int call__parse(struct arch *arch, struct ins_operands *ops, struct map_symbol *ms)
{
	char *endptr, *tok, *name;
	struct map *map = ms->map;
	struct addr_map_symbol target = {
		.ms = { .map = map, },
	};

	ops->target.addr = strtoull(ops->raw, &endptr, 16);

	name = strchr(endptr, '<');
	if (name == NULL)
		goto indirect_call;

	name++;

	if (arch->objdump.skip_functions_char &&
	    strchr(name, arch->objdump.skip_functions_char))
		return -1;

	tok = strchr(name, '>');
	if (tok == NULL)
		return -1;

	*tok = '\0';
	ops->target.name = strdup(name);
	*tok = '>';

	if (ops->target.name == NULL)
		return -1;
find_target:
	target.addr = map__objdump_2mem(map, ops->target.addr);

	if (maps__find_ams(ms->maps, &target) == 0 &&
	    map__rip_2objdump(target.ms.map, map->map_ip(target.ms.map, target.addr)) == ops->target.addr)
		ops->target.sym = target.ms.sym;

	return 0;

indirect_call:
	tok = strchr(endptr, '*');
	if (tok != NULL) {
		endptr++;

		/* Indirect call can use a non-rip register and offset: callq  *0x8(%rbx).
		 * Do not parse such instruction.  */
		if (strstr(endptr, "(%r") == NULL)
			ops->target.addr = strtoull(endptr, NULL, 16);
	}
	goto find_target;
}

static int call__scnprintf(struct ins *ins, char *bf, size_t size,
			   struct ins_operands *ops, int max_ins_name)
{
	if (ops->target.sym)
		return scnprintf(bf, size, "%-*s %s", max_ins_name, ins->name, ops->target.sym->name);

	if (ops->target.addr == 0)
		return ins__raw_scnprintf(ins, bf, size, ops, max_ins_name);

	if (ops->target.name)
		return scnprintf(bf, size, "%-*s %s", max_ins_name, ins->name, ops->target.name);

	return scnprintf(bf, size, "%-*s *%" PRIx64, max_ins_name, ins->name, ops->target.addr);
}

static struct ins_ops call_ops = {
	.parse	   = call__parse,
	.scnprintf = call__scnprintf,
};

bool ins__is_call(const struct ins *ins)
{
	return ins->ops == &call_ops || ins->ops == &s390_call_ops;
}

/*
 * Prevents from matching commas in the comment section, e.g.:
 * ffff200008446e70:       b.cs    ffff2000084470f4 <generic_exec_single+0x314>  // b.hs, b.nlast
 *
 * and skip comma as part of function arguments, e.g.:
 * 1d8b4ac <linemap_lookup(line_maps const*, unsigned int)+0xcc>
 */
static inline const char *validate_comma(const char *c, struct ins_operands *ops)
{
	if (ops->raw_comment && c > ops->raw_comment)
		return NULL;

	if (ops->raw_func_start && c > ops->raw_func_start)
		return NULL;

	return c;
}

static int jump__parse(struct arch *arch, struct ins_operands *ops, struct map_symbol *ms)
{
	struct map *map = ms->map;
	struct symbol *sym = ms->sym;
	struct addr_map_symbol target = {
		.ms = { .map = map, },
	};
	const char *c = strchr(ops->raw, ',');
	u64 start, end;

	ops->raw_comment = strchr(ops->raw, arch->objdump.comment_char);
	ops->raw_func_start = strchr(ops->raw, '<');

	c = validate_comma(c, ops);

	/*
	 * Examples of lines to parse for the _cpp_lex_token@@Base
	 * function:
	 *
	 * 1159e6c: jne    115aa32 <_cpp_lex_token@@Base+0xf92>
	 * 1159e8b: jne    c469be <cpp_named_operator2name@@Base+0xa72>
	 *
	 * The first is a jump to an offset inside the same function,
	 * the second is to another function, i.e. that 0xa72 is an
	 * offset in the cpp_named_operator2name@@base function.
	 */
	/*
	 * skip over possible up to 2 operands to get to address, e.g.:
	 * tbnz	 w0, #26, ffff0000083cd190 <security_file_permission+0xd0>
	 */
	if (c++ != NULL) {
		ops->target.addr = strtoull(c, NULL, 16);
		if (!ops->target.addr) {
			c = strchr(c, ',');
			c = validate_comma(c, ops);
			if (c++ != NULL)
				ops->target.addr = strtoull(c, NULL, 16);
		}
	} else {
		ops->target.addr = strtoull(ops->raw, NULL, 16);
	}

	target.addr = map__objdump_2mem(map, ops->target.addr);
	start = map->unmap_ip(map, sym->start),
	end = map->unmap_ip(map, sym->end);

	ops->target.outside = target.addr < start || target.addr > end;

	/*
	 * FIXME: things like this in _cpp_lex_token (gcc's cc1 program):

		cpp_named_operator2name@@Base+0xa72

	 * Point to a place that is after the cpp_named_operator2name
	 * boundaries, i.e.  in the ELF symbol table for cc1
	 * cpp_named_operator2name is marked as being 32-bytes long, but it in
	 * fact is much larger than that, so we seem to need a symbols__find()
	 * routine that looks for >= current->start and  < next_symbol->start,
	 * possibly just for C++ objects?
	 *
	 * For now lets just make some progress by marking jumps to outside the
	 * current function as call like.
	 *
	 * Actual navigation will come next, with further understanding of how
	 * the symbol searching and disassembly should be done.
	 */
	if (maps__find_ams(ms->maps, &target) == 0 &&
	    map__rip_2objdump(target.ms.map, map->map_ip(target.ms.map, target.addr)) == ops->target.addr)
		ops->target.sym = target.ms.sym;

	if (!ops->target.outside) {
		ops->target.offset = target.addr - start;
		ops->target.offset_avail = true;
	} else {
		ops->target.offset_avail = false;
	}

	return 0;
}

static int jump__scnprintf(struct ins *ins, char *bf, size_t size,
			   struct ins_operands *ops, int max_ins_name)
{
	const char *c;

	if (!ops->target.addr || ops->target.offset < 0)
		return ins__raw_scnprintf(ins, bf, size, ops, max_ins_name);

	if (ops->target.outside && ops->target.sym != NULL)
		return scnprintf(bf, size, "%-*s %s", max_ins_name, ins->name, ops->target.sym->name);

	c = strchr(ops->raw, ',');
	c = validate_comma(c, ops);

	if (c != NULL) {
		const char *c2 = strchr(c + 1, ',');

		c2 = validate_comma(c2, ops);
		/* check for 3-op insn */
		if (c2 != NULL)
			c = c2;
		c++;

		/* mirror arch objdump's space-after-comma style */
		if (*c == ' ')
			c++;
	}

	return scnprintf(bf, size, "%-*s %.*s%" PRIx64, max_ins_name,
			 ins->name, c ? c - ops->raw : 0, ops->raw,
			 ops->target.offset);
}

static struct ins_ops jump_ops = {
	.parse	   = jump__parse,
	.scnprintf = jump__scnprintf,
};

bool ins__is_jump(const struct ins *ins)
{
	return ins->ops == &jump_ops;
}

static int comment__symbol(char *raw, char *comment, u64 *addrp, char **namep)
{
	char *endptr, *name, *t;

	if (strstr(raw, "(%rip)") == NULL)
		return 0;

	*addrp = strtoull(comment, &endptr, 16);
	if (endptr == comment)
		return 0;
	name = strchr(endptr, '<');
	if (name == NULL)
		return -1;

	name++;

	t = strchr(name, '>');
	if (t == NULL)
		return 0;

	*t = '\0';
	*namep = strdup(name);
	*t = '>';

	return 0;
}

static int lock__parse(struct arch *arch, struct ins_operands *ops, struct map_symbol *ms)
{
	ops->locked.ops = zalloc(sizeof(*ops->locked.ops));
	if (ops->locked.ops == NULL)
		return 0;

	if (disasm_line__parse(ops->raw, &ops->locked.ins.name, &ops->locked.ops->raw) < 0)
		goto out_free_ops;

	ops->locked.ins.ops = ins__find(arch, ops->locked.ins.name);

	if (ops->locked.ins.ops == NULL)
		goto out_free_ops;

	if (ops->locked.ins.ops->parse &&
	    ops->locked.ins.ops->parse(arch, ops->locked.ops, ms) < 0)
		goto out_free_ops;

	return 0;

out_free_ops:
	zfree(&ops->locked.ops);
	return 0;
}

static int lock__scnprintf(struct ins *ins, char *bf, size_t size,
			   struct ins_operands *ops, int max_ins_name)
{
	int printed;

	if (ops->locked.ins.ops == NULL)
		return ins__raw_scnprintf(ins, bf, size, ops, max_ins_name);

	printed = scnprintf(bf, size, "%-*s ", max_ins_name, ins->name);
	return printed + ins__scnprintf(&ops->locked.ins, bf + printed,
					size - printed, ops->locked.ops, max_ins_name);
}

static void lock__delete(struct ins_operands *ops)
{
	struct ins *ins = &ops->locked.ins;

	if (ins->ops && ins->ops->free)
		ins->ops->free(ops->locked.ops);
	else
		ins__delete(ops->locked.ops);

	zfree(&ops->locked.ops);
	zfree(&ops->target.raw);
	zfree(&ops->target.name);
}

static struct ins_ops lock_ops = {
	.free	   = lock__delete,
	.parse	   = lock__parse,
	.scnprintf = lock__scnprintf,
};

static int mov__parse(struct arch *arch, struct ins_operands *ops, struct map_symbol *ms __maybe_unused)
{
	char *s = strchr(ops->raw, ','), *target, *comment, prev;

	if (s == NULL)
		return -1;

	*s = '\0';
	ops->source.raw = strdup(ops->raw);
	*s = ',';

	if (ops->source.raw == NULL)
		return -1;

	target = ++s;
	comment = strchr(s, arch->objdump.comment_char);

	if (comment != NULL)
		s = comment - 1;
	else
		s = strchr(s, '\0') - 1;

	while (s > target && isspace(s[0]))
		--s;
	s++;
	prev = *s;
	*s = '\0';

	ops->target.raw = strdup(target);
	*s = prev;

	if (ops->target.raw == NULL)
		goto out_free_source;

	if (comment == NULL)
		return 0;

	comment = skip_spaces(comment);
	comment__symbol(ops->source.raw, comment + 1, &ops->source.addr, &ops->source.name);
	comment__symbol(ops->target.raw, comment + 1, &ops->target.addr, &ops->target.name);

	return 0;

out_free_source:
	zfree(&ops->source.raw);
	return -1;
}

static int mov__scnprintf(struct ins *ins, char *bf, size_t size,
			   struct ins_operands *ops, int max_ins_name)
{
	return scnprintf(bf, size, "%-*s %s,%s", max_ins_name, ins->name,
			 ops->source.name ?: ops->source.raw,
			 ops->target.name ?: ops->target.raw);
}

static struct ins_ops mov_ops = {
	.parse	   = mov__parse,
	.scnprintf = mov__scnprintf,
};

static int dec__parse(struct arch *arch __maybe_unused, struct ins_operands *ops, struct map_symbol *ms __maybe_unused)
{
	char *target, *comment, *s, prev;

	target = s = ops->raw;

	while (s[0] != '\0' && !isspace(s[0]))
		++s;
	prev = *s;
	*s = '\0';

	ops->target.raw = strdup(target);
	*s = prev;

	if (ops->target.raw == NULL)
		return -1;

	comment = strchr(s, arch->objdump.comment_char);
	if (comment == NULL)
		return 0;

	comment = skip_spaces(comment);
	comment__symbol(ops->target.raw, comment + 1, &ops->target.addr, &ops->target.name);

	return 0;
}

static int dec__scnprintf(struct ins *ins, char *bf, size_t size,
			   struct ins_operands *ops, int max_ins_name)
{
	return scnprintf(bf, size, "%-*s %s", max_ins_name, ins->name,
			 ops->target.name ?: ops->target.raw);
}

static struct ins_ops dec_ops = {
	.parse	   = dec__parse,
	.scnprintf = dec__scnprintf,
};

static int nop__scnprintf(struct ins *ins __maybe_unused, char *bf, size_t size,
			  struct ins_operands *ops __maybe_unused, int max_ins_name)
{
	return scnprintf(bf, size, "%-*s", max_ins_name, "nop");
}

static struct ins_ops nop_ops = {
	.scnprintf = nop__scnprintf,
};

static struct ins_ops ret_ops = {
	.scnprintf = ins__raw_scnprintf,
};

bool ins__is_ret(const struct ins *ins)
{
	return ins->ops == &ret_ops;
}

bool ins__is_lock(const struct ins *ins)
{
	return ins->ops == &lock_ops;
}

static int ins__key_cmp(const void *name, const void *insp)
{
	const struct ins *ins = insp;

	return strcmp(name, ins->name);
}

static int ins__cmp(const void *a, const void *b)
{
	const struct ins *ia = a;
	const struct ins *ib = b;

	return strcmp(ia->name, ib->name);
}

static void ins__sort(struct arch *arch)
{
	const int nmemb = arch->nr_instructions;

	qsort(arch->instructions, nmemb, sizeof(struct ins), ins__cmp);
}

static struct ins_ops *__ins__find(struct arch *arch, const char *name)
{
	struct ins *ins;
	const int nmemb = arch->nr_instructions;

	if (!arch->sorted_instructions) {
		ins__sort(arch);
		arch->sorted_instructions = true;
	}

	ins = bsearch(name, arch->instructions, nmemb, sizeof(struct ins), ins__key_cmp);
	return ins ? ins->ops : NULL;
}

static struct ins_ops *ins__find(struct arch *arch, const char *name)
{
	struct ins_ops *ops = __ins__find(arch, name);

	if (!ops && arch->associate_instruction_ops)
		ops = arch->associate_instruction_ops(arch, name);

	return ops;
}

static int arch__key_cmp(const void *name, const void *archp)
{
	const struct arch *arch = archp;

	return strcmp(name, arch->name);
}

static int arch__cmp(const void *a, const void *b)
{
	const struct arch *aa = a;
	const struct arch *ab = b;

	return strcmp(aa->name, ab->name);
}

static void arch__sort(void)
{
	const int nmemb = ARRAY_SIZE(architectures);

	qsort(architectures, nmemb, sizeof(struct arch), arch__cmp);
}

static struct arch *arch__find(const char *name)
{
	const int nmemb = ARRAY_SIZE(architectures);
	static bool sorted;

	if (!sorted) {
		arch__sort();
		sorted = true;
	}

	return bsearch(name, architectures, nmemb, sizeof(struct arch), arch__key_cmp);
}

static struct annotated_source *annotated_source__new(void)
{
	struct annotated_source *src = zalloc(sizeof(*src));

	if (src != NULL)
		INIT_LIST_HEAD(&src->source);

	return src;
}

static __maybe_unused void annotated_source__delete(struct annotated_source *src)
{
	if (src == NULL)
		return;
	zfree(&src->histograms);
	zfree(&src->cycles_hist);
	free(src);
}

static int annotated_source__alloc_histograms(struct annotated_source *src,
					      size_t size, int nr_hists)
{
	size_t sizeof_sym_hist;

	/*
	 * Add buffer of one element for zero length symbol.
	 * When sample is taken from first instruction of
	 * zero length symbol, perf still resolves it and
	 * shows symbol name in perf report and allows to
	 * annotate it.
	 */
	if (size == 0)
		size = 1;

	/* Check for overflow when calculating sizeof_sym_hist */
	if (size > (SIZE_MAX - sizeof(struct sym_hist)) / sizeof(struct sym_hist_entry))
		return -1;

	sizeof_sym_hist = (sizeof(struct sym_hist) + size * sizeof(struct sym_hist_entry));

	/* Check for overflow in zalloc argument */
	if (sizeof_sym_hist > SIZE_MAX / nr_hists)
		return -1;

	src->sizeof_sym_hist = sizeof_sym_hist;
	src->nr_histograms   = nr_hists;
	src->histograms	     = calloc(nr_hists, sizeof_sym_hist) ;
	return src->histograms ? 0 : -1;
}

/* The cycles histogram is lazily allocated. */
static int symbol__alloc_hist_cycles(struct symbol *sym)
{
	struct annotation *notes = symbol__annotation(sym);
	const size_t size = symbol__size(sym);

	notes->src->cycles_hist = calloc(size, sizeof(struct cyc_hist));
	if (notes->src->cycles_hist == NULL)
		return -1;
	return 0;
}

void symbol__annotate_zero_histograms(struct symbol *sym)
{
	struct annotation *notes = symbol__annotation(sym);

	mutex_lock(&notes->lock);
	if (notes->src != NULL) {
		memset(notes->src->histograms, 0,
		       notes->src->nr_histograms * notes->src->sizeof_sym_hist);
		if (notes->src->cycles_hist)
			memset(notes->src->cycles_hist, 0,
				symbol__size(sym) * sizeof(struct cyc_hist));
	}
	mutex_unlock(&notes->lock);
}

static int __symbol__account_cycles(struct cyc_hist *ch,
				    u64 start,
				    unsigned offset, unsigned cycles,
				    unsigned have_start)
{
	/*
	 * For now we can only account one basic block per
	 * final jump. But multiple could be overlapping.
	 * Always account the longest one. So when
	 * a shorter one has been already seen throw it away.
	 *
	 * We separately always account the full cycles.
	 */
	ch[offset].num_aggr++;
	ch[offset].cycles_aggr += cycles;

	if (cycles > ch[offset].cycles_max)
		ch[offset].cycles_max = cycles;

	if (ch[offset].cycles_min) {
		if (cycles && cycles < ch[offset].cycles_min)
			ch[offset].cycles_min = cycles;
	} else
		ch[offset].cycles_min = cycles;

	if (!have_start && ch[offset].have_start)
		return 0;
	if (ch[offset].num) {
		if (have_start && (!ch[offset].have_start ||
				   ch[offset].start > start)) {
			ch[offset].have_start = 0;
			ch[offset].cycles = 0;
			ch[offset].num = 0;
			if (ch[offset].reset < 0xffff)
				ch[offset].reset++;
		} else if (have_start &&
			   ch[offset].start < start)
			return 0;
	}

	if (ch[offset].num < NUM_SPARKS)
		ch[offset].cycles_spark[ch[offset].num] = cycles;

	ch[offset].have_start = have_start;
	ch[offset].start = start;
	ch[offset].cycles += cycles;
	ch[offset].num++;
	return 0;
}

static int __symbol__inc_addr_samples(struct map_symbol *ms,
				      struct annotated_source *src, int evidx, u64 addr,
				      struct perf_sample *sample)
{
	struct symbol *sym = ms->sym;
	unsigned offset;
	struct sym_hist *h;

	pr_debug3("%s: addr=%#" PRIx64 "\n", __func__, ms->map->unmap_ip(ms->map, addr));

	if ((addr < sym->start || addr >= sym->end) &&
	    (addr != sym->end || sym->start != sym->end)) {
		pr_debug("%s(%d): ERANGE! sym->name=%s, start=%#" PRIx64 ", addr=%#" PRIx64 ", end=%#" PRIx64 "\n",
		       __func__, __LINE__, sym->name, sym->start, addr, sym->end);
		return -ERANGE;
	}

	offset = addr - sym->start;
	h = annotated_source__histogram(src, evidx);
	if (h == NULL) {
		pr_debug("%s(%d): ENOMEM! sym->name=%s, start=%#" PRIx64 ", addr=%#" PRIx64 ", end=%#" PRIx64 ", func: %d\n",
			 __func__, __LINE__, sym->name, sym->start, addr, sym->end, sym->type == STT_FUNC);
		return -ENOMEM;
	}
	h->nr_samples++;
	h->addr[offset].nr_samples++;
	h->period += sample->period;
	h->addr[offset].period += sample->period;

	pr_debug3("%#" PRIx64 " %s: period++ [addr: %#" PRIx64 ", %#" PRIx64
		  ", evidx=%d] => nr_samples: %" PRIu64 ", period: %" PRIu64 "\n",
		  sym->start, sym->name, addr, addr - sym->start, evidx,
		  h->addr[offset].nr_samples, h->addr[offset].period);
	return 0;
}

static struct cyc_hist *symbol__cycles_hist(struct symbol *sym)
{
	struct annotation *notes = symbol__annotation(sym);

	if (notes->src == NULL) {
		notes->src = annotated_source__new();
		if (notes->src == NULL)
			return NULL;
		goto alloc_cycles_hist;
	}

	if (!notes->src->cycles_hist) {
alloc_cycles_hist:
		symbol__alloc_hist_cycles(sym);
	}

	return notes->src->cycles_hist;
}

struct annotated_source *symbol__hists(struct symbol *sym, int nr_hists)
{
	struct annotation *notes = symbol__annotation(sym);

	if (notes->src == NULL) {
		notes->src = annotated_source__new();
		if (notes->src == NULL)
			return NULL;
		goto alloc_histograms;
	}

	if (notes->src->histograms == NULL) {
alloc_histograms:
		annotated_source__alloc_histograms(notes->src, symbol__size(sym),
						   nr_hists);
	}

	return notes->src;
}

static int symbol__inc_addr_samples(struct map_symbol *ms,
				    struct evsel *evsel, u64 addr,
				    struct perf_sample *sample)
{
	struct symbol *sym = ms->sym;
	struct annotated_source *src;

	if (sym == NULL)
		return 0;
	src = symbol__hists(sym, evsel->evlist->core.nr_entries);
	return src ? __symbol__inc_addr_samples(ms, src, evsel->core.idx, addr, sample) : 0;
}

static int symbol__account_cycles(u64 addr, u64 start,
				  struct symbol *sym, unsigned cycles)
{
	struct cyc_hist *cycles_hist;
	unsigned offset;

	if (sym == NULL)
		return 0;
	cycles_hist = symbol__cycles_hist(sym);
	if (cycles_hist == NULL)
		return -ENOMEM;
	if (addr < sym->start || addr >= sym->end)
		return -ERANGE;

	if (start) {
		if (start < sym->start || start >= sym->end)
			return -ERANGE;
		if (start >= addr)
			start = 0;
	}
	offset = addr - sym->start;
	return __symbol__account_cycles(cycles_hist,
					start ? start - sym->start : 0,
					offset, cycles,
					!!start);
}

int addr_map_symbol__account_cycles(struct addr_map_symbol *ams,
				    struct addr_map_symbol *start,
				    unsigned cycles)
{
	u64 saddr = 0;
	int err;

	if (!cycles)
		return 0;

	/*
	 * Only set start when IPC can be computed. We can only
	 * compute it when the basic block is completely in a single
	 * function.
	 * Special case the case when the jump is elsewhere, but
	 * it starts on the function start.
	 */
	if (start &&
		(start->ms.sym == ams->ms.sym ||
		 (ams->ms.sym &&
		   start->addr == ams->ms.sym->start + ams->ms.map->start)))
		saddr = start->al_addr;
	if (saddr == 0)
		pr_debug2("BB with bad start: addr %"PRIx64" start %"PRIx64" sym %"PRIx64" saddr %"PRIx64"\n",
			ams->addr,
			start ? start->addr : 0,
			ams->ms.sym ? ams->ms.sym->start + ams->ms.map->start : 0,
			saddr);
	err = symbol__account_cycles(ams->al_addr, saddr, ams->ms.sym, cycles);
	if (err)
		pr_debug2("account_cycles failed %d\n", err);
	return err;
}

static unsigned annotation__count_insn(struct annotation *notes, u64 start, u64 end)
{
	unsigned n_insn = 0;
	u64 offset;

	for (offset = start; offset <= end; offset++) {
		if (notes->offsets[offset])
			n_insn++;
	}
	return n_insn;
}

static void annotation__count_and_fill(struct annotation *notes, u64 start, u64 end, struct cyc_hist *ch)
{
	unsigned n_insn;
	unsigned int cover_insn = 0;
	u64 offset;

	n_insn = annotation__count_insn(notes, start, end);
	if (n_insn && ch->num && ch->cycles) {
		float ipc = n_insn / ((double)ch->cycles / (double)ch->num);

		/* Hide data when there are too many overlaps. */
		if (ch->reset >= 0x7fff)
			return;

		for (offset = start; offset <= end; offset++) {
			struct annotation_line *al = notes->offsets[offset];

			if (al && al->ipc == 0.0) {
				al->ipc = ipc;
				cover_insn++;
			}
		}

		if (cover_insn) {
			notes->hit_cycles += ch->cycles;
			notes->hit_insn += n_insn * ch->num;
			notes->cover_insn += cover_insn;
		}
	}
}

void annotation__compute_ipc(struct annotation *notes, size_t size)
{
	s64 offset;

	if (!notes->src || !notes->src->cycles_hist)
		return;

	notes->total_insn = annotation__count_insn(notes, 0, size - 1);
	notes->hit_cycles = 0;
	notes->hit_insn = 0;
	notes->cover_insn = 0;

	mutex_lock(&notes->lock);
	for (offset = size - 1; offset >= 0; --offset) {
		struct cyc_hist *ch;

		ch = &notes->src->cycles_hist[offset];
		if (ch && ch->cycles) {
			struct annotation_line *al;

			if (ch->have_start)
				annotation__count_and_fill(notes, ch->start, offset, ch);
			al = notes->offsets[offset];
			if (al && ch->num_aggr) {
				al->cycles = ch->cycles_aggr / ch->num_aggr;
				al->cycles_max = ch->cycles_max;
				al->cycles_min = ch->cycles_min;
			}
			notes->have_cycles = true;
		}
	}
	mutex_unlock(&notes->lock);
}

int addr_map_symbol__inc_samples(struct addr_map_symbol *ams, struct perf_sample *sample,
				 struct evsel *evsel)
{
	return symbol__inc_addr_samples(&ams->ms, evsel, ams->al_addr, sample);
}

int hist_entry__inc_addr_samples(struct hist_entry *he, struct perf_sample *sample,
				 struct evsel *evsel, u64 ip)
{
	return symbol__inc_addr_samples(&he->ms, evsel, ip, sample);
}

static void disasm_line__init_ins(struct disasm_line *dl, struct arch *arch, struct map_symbol *ms)
{
	dl->ins.ops = ins__find(arch, dl->ins.name);

	if (!dl->ins.ops)
		return;

	if (dl->ins.ops->parse && dl->ins.ops->parse(arch, &dl->ops, ms) < 0)
		dl->ins.ops = NULL;
}

static int disasm_line__parse(char *line, const char **namep, char **rawp)
{
	char tmp, *name = skip_spaces(line);

	if (name[0] == '\0')
		return -1;

	*rawp = name + 1;

	while ((*rawp)[0] != '\0' && !isspace((*rawp)[0]))
		++*rawp;

	tmp = (*rawp)[0];
	(*rawp)[0] = '\0';
	*namep = strdup(name);

	if (*namep == NULL)
		goto out;

	(*rawp)[0] = tmp;
	*rawp = strim(*rawp);

	return 0;

out:
	return -1;
}

struct annotate_args {
	struct arch		  *arch;
	struct map_symbol	  ms;
	struct evsel		  *evsel;
	struct annotation_options *options;
	s64			  offset;
	char			  *line;
	int			  line_nr;
	char			  *fileloc;
};

static void annotation_line__init(struct annotation_line *al,
				  struct annotate_args *args,
				  int nr)
{
	al->offset = args->offset;
	al->line = strdup(args->line);
	al->line_nr = args->line_nr;
	al->fileloc = args->fileloc;
	al->data_nr = nr;
}

static void annotation_line__exit(struct annotation_line *al)
{
	free_srcline(al->path);
	zfree(&al->line);
}

static size_t disasm_line_size(int nr)
{
	struct annotation_line *al;

	return (sizeof(struct disasm_line) + (sizeof(al->data[0]) * nr));
}

/*
 * Allocating the disasm annotation line data with
 * following structure:
 *
 *    -------------------------------------------
 *    struct disasm_line | struct annotation_line
 *    -------------------------------------------
 *
 * We have 'struct annotation_line' member as last member
 * of 'struct disasm_line' to have an easy access.
 */
static struct disasm_line *disasm_line__new(struct annotate_args *args)
{
	struct disasm_line *dl = NULL;
	int nr = 1;

	if (evsel__is_group_event(args->evsel))
		nr = args->evsel->core.nr_members;

	dl = zalloc(disasm_line_size(nr));
	if (!dl)
		return NULL;

	annotation_line__init(&dl->al, args, nr);
	if (dl->al.line == NULL)
		goto out_delete;

	if (args->offset != -1) {
		if (disasm_line__parse(dl->al.line, &dl->ins.name, &dl->ops.raw) < 0)
			goto out_free_line;

		disasm_line__init_ins(dl, args->arch, &args->ms);
	}

	return dl;

out_free_line:
	zfree(&dl->al.line);
out_delete:
	free(dl);
	return NULL;
}

void disasm_line__free(struct disasm_line *dl)
{
	if (dl->ins.ops && dl->ins.ops->free)
		dl->ins.ops->free(&dl->ops);
	else
		ins__delete(&dl->ops);
	zfree(&dl->ins.name);
	annotation_line__exit(&dl->al);
	free(dl);
}

int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw, int max_ins_name)
{
	if (raw || !dl->ins.ops)
		return scnprintf(bf, size, "%-*s %s", max_ins_name, dl->ins.name, dl->ops.raw);

	return ins__scnprintf(&dl->ins, bf, size, &dl->ops, max_ins_name);
}

void annotation__init(struct annotation *notes)
{
	mutex_init(&notes->lock);
}

void annotation__exit(struct annotation *notes)
{
	annotated_source__delete(notes->src);
	mutex_destroy(&notes->lock);
}

static void annotation_line__add(struct annotation_line *al, struct list_head *head)
{
	list_add_tail(&al->node, head);
}

struct annotation_line *
annotation_line__next(struct annotation_line *pos, struct list_head *head)
{
	list_for_each_entry_continue(pos, head, node)
		if (pos->offset >= 0)
			return pos;

	return NULL;
}

static const char *annotate__address_color(struct block_range *br)
{
	double cov = block_range__coverage(br);

	if (cov >= 0) {
		/* mark red for >75% coverage */
		if (cov > 0.75)
			return PERF_COLOR_RED;

		/* mark dull for <1% coverage */
		if (cov < 0.01)
			return PERF_COLOR_NORMAL;
	}

	return PERF_COLOR_MAGENTA;
}

static const char *annotate__asm_color(struct block_range *br)
{
	double cov = block_range__coverage(br);

	if (cov >= 0) {
		/* mark dull for <1% coverage */
		if (cov < 0.01)
			return PERF_COLOR_NORMAL;
	}

	return PERF_COLOR_BLUE;
}

static void annotate__branch_printf(struct block_range *br, u64 addr)
{
	bool emit_comment = true;

	if (!br)
		return;

#if 1
	if (br->is_target && br->start == addr) {
		struct block_range *branch = br;
		double p;

		/*
		 * Find matching branch to our target.
		 */
		while (!branch->is_branch)
			branch = block_range__next(branch);

		p = 100 *(double)br->entry / branch->coverage;

		if (p > 0.1) {
			if (emit_comment) {
				emit_comment = false;
				printf("\t#");
			}

			/*
			 * The percentage of coverage joined at this target in relation
			 * to the next branch.
			 */
			printf(" +%.2f%%", p);
		}
	}
#endif
	if (br->is_branch && br->end == addr) {
		double p = 100*(double)br->taken / br->coverage;

		if (p > 0.1) {
			if (emit_comment) {
				emit_comment = false;
				printf("\t#");
			}

			/*
			 * The percentage of coverage leaving at this branch, and
			 * its prediction ratio.
			 */
			printf(" -%.2f%% (p:%.2f%%)", p, 100*(double)br->pred  / br->taken);
		}
	}
}

static int disasm_line__print(struct disasm_line *dl, u64 start, int addr_fmt_width)
{
	s64 offset = dl->al.offset;
	const u64 addr = start + offset;
	struct block_range *br;

	br = block_range__find(addr);
	color_fprintf(stdout, annotate__address_color(br), "  %*" PRIx64 ":", addr_fmt_width, addr);
	color_fprintf(stdout, annotate__asm_color(br), "%s", dl->al.line);
	annotate__branch_printf(br, addr);
	return 0;
}

static int
annotation_line__print(struct annotation_line *al, struct symbol *sym, u64 start,
		       struct evsel *evsel, u64 len, int min_pcnt, int printed,
		       int max_lines, struct annotation_line *queue, int addr_fmt_width,
		       int percent_type)
{
	struct disasm_line *dl = container_of(al, struct disasm_line, al);
	static const char *prev_line;

	if (al->offset != -1) {
		double max_percent = 0.0;
		int i, nr_percent = 1;
		const char *color;
		struct annotation *notes = symbol__annotation(sym);

		for (i = 0; i < al->data_nr; i++) {
			double percent;

			percent = annotation_data__percent(&al->data[i],
							   percent_type);

			if (percent > max_percent)
				max_percent = percent;
		}

		if (al->data_nr > nr_percent)
			nr_percent = al->data_nr;

		if (max_percent < min_pcnt)
			return -1;

		if (max_lines && printed >= max_lines)
			return 1;

		if (queue != NULL) {
			list_for_each_entry_from(queue, &notes->src->source, node) {
				if (queue == al)
					break;
				annotation_line__print(queue, sym, start, evsel, len,
						       0, 0, 1, NULL, addr_fmt_width,
						       percent_type);
			}
		}

		color = get_percent_color(max_percent);

		for (i = 0; i < nr_percent; i++) {
			struct annotation_data *data = &al->data[i];
			double percent;

			percent = annotation_data__percent(data, percent_type);
			color = get_percent_color(percent);

			if (symbol_conf.show_total_period)
				color_fprintf(stdout, color, " %11" PRIu64,
					      data->he.period);
			else if (symbol_conf.show_nr_samples)
				color_fprintf(stdout, color, " %7" PRIu64,
					      data->he.nr_samples);
			else
				color_fprintf(stdout, color, " %7.2f", percent);
		}

		printf(" : ");

		disasm_line__print(dl, start, addr_fmt_width);

		/*
		 * Also color the filename and line if needed, with
		 * the same color than the percentage. Don't print it
		 * twice for close colored addr with the same filename:line
		 */
		if (al->path) {
			if (!prev_line || strcmp(prev_line, al->path)) {
				color_fprintf(stdout, color, " // %s", al->path);
				prev_line = al->path;
			}
		}

		printf("\n");
	} else if (max_lines && printed >= max_lines)
		return 1;
	else {
		int width = symbol_conf.show_total_period ? 12 : 8;

		if (queue)
			return -1;

		if (evsel__is_group_event(evsel))
			width *= evsel->core.nr_members;

		if (!*al->line)
			printf(" %*s:\n", width, " ");
		else
			printf(" %*s: %-*d %s\n", width, " ", addr_fmt_width, al->line_nr, al->line);
	}

	return 0;
}

/*
 * symbol__parse_objdump_line() parses objdump output (with -d --no-show-raw)
 * which looks like following
 *
 *  0000000000415500 <_init>:
 *    415500:       sub    $0x8,%rsp
 *    415504:       mov    0x2f5ad5(%rip),%rax        # 70afe0 <_DYNAMIC+0x2f8>
 *    41550b:       test   %rax,%rax
 *    41550e:       je     415515 <_init+0x15>
 *    415510:       callq  416e70 <__gmon_start__@plt>
 *    415515:       add    $0x8,%rsp
 *    415519:       retq
 *
 * it will be parsed and saved into struct disasm_line as
 *  <offset>       <name>  <ops.raw>
 *
 * The offset will be a relative offset from the start of the symbol and -1
 * means that it's not a disassembly line so should be treated differently.
 * The ops.raw part will be parsed further according to type of the instruction.
 */
static int symbol__parse_objdump_line(struct symbol *sym,
				      struct annotate_args *args,
				      char *parsed_line, int *line_nr, char **fileloc)
{
	struct map *map = args->ms.map;
	struct annotation *notes = symbol__annotation(sym);
	struct disasm_line *dl;
	char *tmp;
	s64 line_ip, offset = -1;
	regmatch_t match[2];

	/* /filename:linenr ? Save line number and ignore. */
	if (regexec(&file_lineno, parsed_line, 2, match, 0) == 0) {
		*line_nr = atoi(parsed_line + match[1].rm_so);
		*fileloc = strdup(parsed_line);
		return 0;
	}

	/* Process hex address followed by ':'. */
	line_ip = strtoull(parsed_line, &tmp, 16);
	if (parsed_line != tmp && tmp[0] == ':' && tmp[1] != '\0') {
		u64 start = map__rip_2objdump(map, sym->start),
		    end = map__rip_2objdump(map, sym->end);

		offset = line_ip - start;
		if ((u64)line_ip < start || (u64)line_ip >= end)
			offset = -1;
		else
			parsed_line = tmp + 1;
	}

	args->offset  = offset;
	args->line    = parsed_line;
	args->line_nr = *line_nr;
	args->fileloc = *fileloc;
	args->ms.sym  = sym;

	dl = disasm_line__new(args);
	(*line_nr)++;

	if (dl == NULL)
		return -1;

	if (!disasm_line__has_local_offset(dl)) {
		dl->ops.target.offset = dl->ops.target.addr -
					map__rip_2objdump(map, sym->start);
		dl->ops.target.offset_avail = true;
	}

	/* kcore has no symbols, so add the call target symbol */
	if (dl->ins.ops && ins__is_call(&dl->ins) && !dl->ops.target.sym) {
		struct addr_map_symbol target = {
			.addr = dl->ops.target.addr,
			.ms = { .map = map, },
		};

		if (!maps__find_ams(args->ms.maps, &target) &&
		    target.ms.sym->start == target.al_addr)
			dl->ops.target.sym = target.ms.sym;
	}

	annotation_line__add(&dl->al, &notes->src->source);

	return 0;
}

static __attribute__((constructor)) void symbol__init_regexpr(void)
{
	regcomp(&file_lineno, "^/[^:]+:([0-9]+)", REG_EXTENDED);
}

static void delete_last_nop(struct symbol *sym)
{
	struct annotation *notes = symbol__annotation(sym);
	struct list_head *list = &notes->src->source;
	struct disasm_line *dl;

	while (!list_empty(list)) {
		dl = list_entry(list->prev, struct disasm_line, al.node);

		if (dl->ins.ops) {
			if (dl->ins.ops != &nop_ops)
				return;
		} else {
			if (!strstr(dl->al.line, " nop ") &&
			    !strstr(dl->al.line, " nopl ") &&
			    !strstr(dl->al.line, " nopw "))
				return;
		}

		list_del_init(&dl->al.node);
		disasm_line__free(dl);
	}
}