[PATCH V7 0/7] perf tools: kcore improvements

[PATCH V7 0/7] perf tools: kcore improvements

[Adrian Hunter]

Hi

Here are some improvements for using kcore (version 7).  There are 3
improvements:

	- validate that kcore matches the perf.data modules
	- workaround objdump difficulties with kcore
	- add kcore to the build-id cache

Changes in V7:
	perf tools: validate kcore module addresses
		Dropped because it has been applied
	perf tools: workaround objdump difficulties with kcore
		Dropped because it has been applied
	perf tools: Add map_groups__find_ams()
		New patch
	perf tools: fix annotate_browser__callq()
		Changed to use map_groups__find_ams()
	perf tools: find kcore symbols on other maps
		Changed to use map_groups__find_ams()
	perf tools: another fix for annotate_browser__callq()
		New patch

Changes in V6:
	perf tools: make a separate function to parse /proc/modules
		Dropped because it has been applied
	perf tools: workaround objdump difficulties with kcore
		Renamed some functions and reordered parameters
	perf buildid-cache: add ability to add kcore to the cache
		Renamed some functions and reordered parameters

Changes in V5:
	perf tools: make a separate function to parse /proc/modules
		Use scnprintf not snprintf
	perf tools: validate kcore module addresses
		Fix check for mi->name not allocated
	perf buildid-cache: add ability to add kcore to the cache
		Use scnprintf not snprintf
	perf tools: add ability to find kcore in build-id cache
		Use scnprintf not snprintf

Changes in V4:
	perf tools: fix path unpopulated in machine__create_modules()
		Dropped because it has been applied
	perf buildid-cache: add ability to add kcore to the cache
		Tweaked Documentation/perf-buildid-cache.txt
	perf tools: add ability to find kcore in build-id cache
		Changed to check read access to /proc/kcore before
		skipping the buildid cache

Changes in V3:
	perf tools: workaround objdump difficulties with kcore
		change strncpy to strlcpy
	perf buildid-cache: add ability to add kcore to the cache
		change strncpy to strlcpy
	perf tools: add ability to find kcore in build-id cache
		change strncpy to strlcpy
Changes in V2:
	perf tools: fix buildid cache handling of kallsyms with kcore
		Dropped because it has been applied
	perf tools: fix path unpopulated in machine__create_modules()
		Use 'modules' pointer


Adrian Hunter (6):
      perf tools: fix annotate_browser__callq()
      perf tools: find kcore symbols on other maps
      perf tools: add copyfile_mode()
      perf buildid-cache: add ability to add kcore to the cache
      perf tools: add ability to find kcore in build-id cache
      perf tools: another fix for annotate_browser__callq()

Arnaldo Carvalho de Melo (1):
      perf tools: Add map_groups__find_ams()

 tools/perf/Documentation/perf-buildid-cache.txt |  13 +
 tools/perf/builtin-buildid-cache.c              | 148 +++++++++-
 tools/perf/ui/browsers/annotate.c               |  24 +-
 tools/perf/util/annotate.c                      |  22 +-
 tools/perf/util/map.c                           |  48 +++-
 tools/perf/util/map.h                           |   7 +
 tools/perf/util/symbol-elf.c                    | 358 ++++++++++++++++++++++++
 tools/perf/util/symbol-minimal.c                |   6 +
 tools/perf/util/symbol.c                        | 188 ++++++++++---
 tools/perf/util/symbol.h                        |   3 +
 tools/perf/util/util.c                          |  18 +-
 tools/perf/util/util.h                          |   1 +
 12 files changed, 761 insertions(+), 75 deletions(-)


Regards
Adrian

[PATCH V7 1/7] perf tools: Add map_groups__find_ams()

[Adrian Hunter]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Add a function to find a symbol using an ip that
might be on a different map.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/util/map.c | 17 +++++++++++++++++
 tools/perf/util/map.h |  4 ++++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 17ee458..9dea404 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -371,6 +371,23 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
 	return NULL;
 }
 
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter)
+{
+	if (ams->addr < ams->map->start || ams->addr > ams->map->end) {
+		if (ams->map->groups == NULL)
+			return -1;
+		ams->map = map_groups__find(ams->map->groups, ams->map->type,
+					    ams->addr);
+		if (ams->map == NULL)
+			return -1;
+	}
+
+	ams->al_addr = ams->map->map_ip(ams->map, ams->addr);
+	ams->sym = map__find_symbol(ams->map, ams->al_addr, filter);
+
+	return ams->sym ? 0 : -1;
+}
+
 size_t __map_groups__fprintf_maps(struct map_groups *mg,
 				  enum map_type type, int verbose, FILE *fp)
 {
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index 4886ca2..0359b4a 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -167,6 +167,10 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
 					       struct map **mapp,
 					       symbol_filter_t filter);
 
+struct addr_map_symbol;
+
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter);
+
 static inline
 struct symbol *map_groups__find_function_by_name(struct map_groups *mg,
 						 const char *name, struct map **mapp,
-- 
1.7.11.7

[tip:perf/core] perf symbols: Add map_groups__find_ams()

[tip-bot for Arnaldo Carvalho de Melo]

Commit-ID:  4e987712740a3634c19a6fedaf12577b26775dc5
Gitweb:     http://git.kernel.org/tip/4e987712740a3634c19a6fedaf12577b26775dc5
Author:     Arnaldo Carvalho de Melo <acme@redhat.com>
AuthorDate: Mon, 14 Oct 2013 13:43:38 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 10:29:02 -0300

perf symbols: Add map_groups__find_ams()

Add a function to find a symbol using an ip that might be on a different
map.

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/map.c | 17 +++++++++++++++++
 tools/perf/util/map.h |  4 ++++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 17ee458..9dea404 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -371,6 +371,23 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
 	return NULL;
 }
 
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter)
+{
+	if (ams->addr < ams->map->start || ams->addr > ams->map->end) {
+		if (ams->map->groups == NULL)
+			return -1;
+		ams->map = map_groups__find(ams->map->groups, ams->map->type,
+					    ams->addr);
+		if (ams->map == NULL)
+			return -1;
+	}
+
+	ams->al_addr = ams->map->map_ip(ams->map, ams->addr);
+	ams->sym = map__find_symbol(ams->map, ams->al_addr, filter);
+
+	return ams->sym ? 0 : -1;
+}
+
 size_t __map_groups__fprintf_maps(struct map_groups *mg,
 				  enum map_type type, int verbose, FILE *fp)
 {
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index 4886ca2..0359b4a 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -167,6 +167,10 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
 					       struct map **mapp,
 					       symbol_filter_t filter);
 
+struct addr_map_symbol;
+
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter);
+
 static inline
 struct symbol *map_groups__find_function_by_name(struct map_groups *mg,
 						 const char *name, struct map **mapp,

[PATCH V7 2/7] perf tools: fix annotate_browser__callq()

[Adrian Hunter]

When following a call, annotate_browser__callq()
uses the current symbol's map to look up the
target ip.  That will not work if the target ip
is on a map with a different mapping (i.e.
start - pgoff is different).

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/ui/browsers/annotate.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index 08545ae..57d3a86 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -442,35 +442,34 @@ static bool annotate_browser__callq(struct annotate_browser *browser,
 {
 	struct map_symbol *ms = browser->b.priv;
 	struct disasm_line *dl = browser->selection;
-	struct symbol *sym = ms->sym;
 	struct annotation *notes;
-	struct symbol *target;
-	u64 ip;
+	struct addr_map_symbol target = {
+		.map = ms->map,
+		.addr = dl->ops.target.addr,
+	};
 	char title[SYM_TITLE_MAX_SIZE];
 
 	if (!ins__is_call(dl->ins))
 		return false;
 
-	ip = ms->map->map_ip(ms->map, dl->ops.target.addr);
-	target = map__find_symbol(ms->map, ip, NULL);
-	if (target == NULL) {
+	if (map_groups__find_ams(&target, NULL)) {
 		ui_helpline__puts("The called function was not found.");
 		return true;
 	}
 
-	notes = symbol__annotation(target);
+	notes = symbol__annotation(target.sym);
 	pthread_mutex_lock(&notes->lock);
 
-	if (notes->src == NULL && symbol__alloc_hist(target) < 0) {
+	if (notes->src == NULL && symbol__alloc_hist(target.sym) < 0) {
 		pthread_mutex_unlock(&notes->lock);
 		ui__warning("Not enough memory for annotating '%s' symbol!\n",
-			    target->name);
+			    target.sym->name);
 		return true;
 	}
 
 	pthread_mutex_unlock(&notes->lock);
-	symbol__tui_annotate(target, ms->map, evsel, hbt);
-	sym_title(sym, ms->map, title, sizeof(title));
+	symbol__tui_annotate(target.sym, target.map, evsel, hbt);
+	sym_title(ms->sym, ms->map, title, sizeof(title));
 	ui_browser__show_title(&browser->b, title);
 	return true;
 }
-- 
1.7.11.7

[tip:perf/core] perf annotate: Fix annotate_browser__callq()

[tip-bot for Adrian Hunter]

Commit-ID:  1179e11bbb655162e83b33e17a97c45d3fe292da
Gitweb:     http://git.kernel.org/tip/1179e11bbb655162e83b33e17a97c45d3fe292da
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 13:43:39 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 10:29:27 -0300

perf annotate: Fix annotate_browser__callq()

When following a call, annotate_browser__callq() uses the current
symbol's map to look up the target ip.  That will not work if the target
ip is on a map with a different mapping (i.e.  start - pgoff is
different).

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/ui/browsers/annotate.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index 08545ae..57d3a86 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -442,35 +442,34 @@ static bool annotate_browser__callq(struct annotate_browser *browser,
 {
 	struct map_symbol *ms = browser->b.priv;
 	struct disasm_line *dl = browser->selection;
-	struct symbol *sym = ms->sym;
 	struct annotation *notes;
-	struct symbol *target;
-	u64 ip;
+	struct addr_map_symbol target = {
+		.map = ms->map,
+		.addr = dl->ops.target.addr,
+	};
 	char title[SYM_TITLE_MAX_SIZE];
 
 	if (!ins__is_call(dl->ins))
 		return false;
 
-	ip = ms->map->map_ip(ms->map, dl->ops.target.addr);
-	target = map__find_symbol(ms->map, ip, NULL);
-	if (target == NULL) {
+	if (map_groups__find_ams(&target, NULL)) {
 		ui_helpline__puts("The called function was not found.");
 		return true;
 	}
 
-	notes = symbol__annotation(target);
+	notes = symbol__annotation(target.sym);
 	pthread_mutex_lock(&notes->lock);
 
-	if (notes->src == NULL && symbol__alloc_hist(target) < 0) {
+	if (notes->src == NULL && symbol__alloc_hist(target.sym) < 0) {
 		pthread_mutex_unlock(&notes->lock);
 		ui__warning("Not enough memory for annotating '%s' symbol!\n",
-			    target->name);
+			    target.sym->name);
 		return true;
 	}
 
 	pthread_mutex_unlock(&notes->lock);
-	symbol__tui_annotate(target, ms->map, evsel, hbt);
-	sym_title(sym, ms->map, title, sizeof(title));
+	symbol__tui_annotate(target.sym, target.map, evsel, hbt);
+	sym_title(ms->sym, ms->map, title, sizeof(title));
 	ui_browser__show_title(&browser->b, title);
 	return true;
 }

[PATCH V7 3/7] perf tools: find kcore symbols on other maps

[Adrian Hunter]

Use the new map__find_other_map_symbol() to
find kcore symbols on other maps.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/util/annotate.c | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 882bb86..cf6242c 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -825,20 +825,16 @@ static int symbol__parse_objdump_line(struct symbol *sym, struct map *map,
 		dl->ops.target.offset = dl->ops.target.addr -
 					map__rip_2objdump(map, sym->start);
 
-	/*
-	 * kcore has no symbols, so add the call target name if it is on the
-	 * same map.
-	 */
+	/* kcore has no symbols, so add the call target name */
 	if (dl->ins && ins__is_call(dl->ins) && !dl->ops.target.name) {
-		struct symbol *s;
-		u64 ip = dl->ops.target.addr;
-
-		if (ip >= map->start && ip <= map->end) {
-			ip = map->map_ip(map, ip);
-			s = map__find_symbol(map, ip, NULL);
-			if (s && s->start == ip)
-				dl->ops.target.name = strdup(s->name);
-		}
+		struct addr_map_symbol target = {
+			.map = map,
+			.addr = dl->ops.target.addr,
+		};
+
+		if (!map_groups__find_ams(&target, NULL) &&
+		    target.sym->start == target.al_addr)
+			dl->ops.target.name = strdup(target.sym->name);
 	}
 
 	disasm__add(&notes->src->source, dl);
-- 
1.7.11.7

[tip:perf/core] perf annotate: Find kcore symbols on other maps

[tip-bot for Adrian Hunter]

Commit-ID:  6e427ab02c8886ca6c9ecdbb318e68fe8f605469
Gitweb:     http://git.kernel.org/tip/6e427ab02c8886ca6c9ecdbb318e68fe8f605469
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 13:43:40 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 10:29:20 -0300

perf annotate: Find kcore symbols on other maps

Use the new map_groups__find_ams() method to find kcore symbols on other
maps.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/annotate.c | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 882bb86..cf6242c 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -825,20 +825,16 @@ static int symbol__parse_objdump_line(struct symbol *sym, struct map *map,
 		dl->ops.target.offset = dl->ops.target.addr -
 					map__rip_2objdump(map, sym->start);
 
-	/*
-	 * kcore has no symbols, so add the call target name if it is on the
-	 * same map.
-	 */
+	/* kcore has no symbols, so add the call target name */
 	if (dl->ins && ins__is_call(dl->ins) && !dl->ops.target.name) {
-		struct symbol *s;
-		u64 ip = dl->ops.target.addr;
-
-		if (ip >= map->start && ip <= map->end) {
-			ip = map->map_ip(map, ip);
-			s = map__find_symbol(map, ip, NULL);
-			if (s && s->start == ip)
-				dl->ops.target.name = strdup(s->name);
-		}
+		struct addr_map_symbol target = {
+			.map = map,
+			.addr = dl->ops.target.addr,
+		};
+
+		if (!map_groups__find_ams(&target, NULL) &&
+		    target.sym->start == target.al_addr)
+			dl->ops.target.name = strdup(target.sym->name);
 	}
 
 	disasm__add(&notes->src->source, dl);

[PATCH V7 4/7] perf tools: add copyfile_mode()

[Adrian Hunter]

Add a function to copy a file specifying the
permissions to use for the created file.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/util/util.c | 18 +++++++++++++-----
 tools/perf/util/util.h |  1 +
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index ab71d62..8dc8cf3 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -55,17 +55,20 @@ int mkdir_p(char *path, mode_t mode)
 	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
 }
 
-static int slow_copyfile(const char *from, const char *to)
+static int slow_copyfile(const char *from, const char *to, mode_t mode)
 {
-	int err = 0;
+	int err = -1;
 	char *line = NULL;
 	size_t n;
 	FILE *from_fp = fopen(from, "r"), *to_fp;
+	mode_t old_umask;
 
 	if (from_fp == NULL)
 		goto out;
 
+	old_umask = umask(mode ^ 0777);
 	to_fp = fopen(to, "w");
+	umask(old_umask);
 	if (to_fp == NULL)
 		goto out_fclose_from;
 
@@ -82,7 +85,7 @@ out:
 	return err;
 }
 
-int copyfile(const char *from, const char *to)
+int copyfile_mode(const char *from, const char *to, mode_t mode)
 {
 	int fromfd, tofd;
 	struct stat st;
@@ -93,13 +96,13 @@ int copyfile(const char *from, const char *to)
 		goto out;
 
 	if (st.st_size == 0) /* /proc? do it slowly... */
-		return slow_copyfile(from, to);
+		return slow_copyfile(from, to, mode);
 
 	fromfd = open(from, O_RDONLY);
 	if (fromfd < 0)
 		goto out;
 
-	tofd = creat(to, 0755);
+	tofd = creat(to, mode);
 	if (tofd < 0)
 		goto out_close_from;
 
@@ -121,6 +124,11 @@ out:
 	return err;
 }
 
+int copyfile(const char *from, const char *to)
+{
+	return copyfile_mode(from, to, 0755);
+}
+
 unsigned long convert_unit(unsigned long value, char *unit)
 {
 	*unit = ' ';
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 1f06ba4..42dfba7 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -243,6 +243,7 @@ static inline int sane_case(int x, int high)
 
 int mkdir_p(char *path, mode_t mode);
 int copyfile(const char *from, const char *to);
+int copyfile_mode(const char *from, const char *to, mode_t mode);
 
 s64 perf_atoll(const char *str);
 char **argv_split(const char *str, int *argcp);
-- 
1.7.11.7

[tip:perf/core] perf tools: Add copyfile_mode()

[tip-bot for Adrian Hunter]

Commit-ID:  9a17d7268d71674f0bbff6821f7d8e6dc0ece19a
Gitweb:     http://git.kernel.org/tip/9a17d7268d71674f0bbff6821f7d8e6dc0ece19a
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 13:43:41 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 10:29:23 -0300

perf tools: Add copyfile_mode()

Add a function to copy a file specifying the permissions to use for the
created file.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/util.c | 18 +++++++++++++-----
 tools/perf/util/util.h |  1 +
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index ab71d62..8dc8cf3 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -55,17 +55,20 @@ int mkdir_p(char *path, mode_t mode)
 	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
 }
 
-static int slow_copyfile(const char *from, const char *to)
+static int slow_copyfile(const char *from, const char *to, mode_t mode)
 {
-	int err = 0;
+	int err = -1;
 	char *line = NULL;
 	size_t n;
 	FILE *from_fp = fopen(from, "r"), *to_fp;
+	mode_t old_umask;
 
 	if (from_fp == NULL)
 		goto out;
 
+	old_umask = umask(mode ^ 0777);
 	to_fp = fopen(to, "w");
+	umask(old_umask);
 	if (to_fp == NULL)
 		goto out_fclose_from;
 
@@ -82,7 +85,7 @@ out:
 	return err;
 }
 
-int copyfile(const char *from, const char *to)
+int copyfile_mode(const char *from, const char *to, mode_t mode)
 {
 	int fromfd, tofd;
 	struct stat st;
@@ -93,13 +96,13 @@ int copyfile(const char *from, const char *to)
 		goto out;
 
 	if (st.st_size == 0) /* /proc? do it slowly... */
-		return slow_copyfile(from, to);
+		return slow_copyfile(from, to, mode);
 
 	fromfd = open(from, O_RDONLY);
 	if (fromfd < 0)
 		goto out;
 
-	tofd = creat(to, 0755);
+	tofd = creat(to, mode);
 	if (tofd < 0)
 		goto out_close_from;
 
@@ -121,6 +124,11 @@ out:
 	return err;
 }
 
+int copyfile(const char *from, const char *to)
+{
+	return copyfile_mode(from, to, 0755);
+}
+
 unsigned long convert_unit(unsigned long value, char *unit)
 {
 	*unit = ' ';
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 1f06ba4..42dfba7 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -243,6 +243,7 @@ static inline int sane_case(int x, int high)
 
 int mkdir_p(char *path, mode_t mode);
 int copyfile(const char *from, const char *to);
+int copyfile_mode(const char *from, const char *to, mode_t mode);
 
 s64 perf_atoll(const char *str);
 char **argv_split(const char *str, int *argcp);

[PATCH V7 5/7] perf buildid-cache: add ability to add kcore to the cache

[Adrian Hunter]

kcore can be used to view the running kernel object code.
However, kcore changes as modules are loaded and unloaded,
and when the kernel decides to modify its own code.
Consequently it is useful to create a copy of kcore at a
particular time.  Unlike vmlinux, kcore is not unique
for a given build-id.  And in addition, the kallsyms
and modules files are also needed.  The tool therefore
creates a directory:

	~/.debug/[kernel.kcore]/<build-id>/<YYYYmmddHHMMSShh>

which contains: kcore, kallsyms and modules.

Note that the copied kcore contains only code sections.
See the kcore_copy() function for how that is determined.

The tool will not make additional copies of kcore if there
is already one with the same modules at the same addresses.

Currently, perf tools will not look for kcore in the cache.
That is addressed in another patch.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/Documentation/perf-buildid-cache.txt |  13 +
 tools/perf/builtin-buildid-cache.c              | 148 +++++++++-
 tools/perf/util/symbol-elf.c                    | 358 ++++++++++++++++++++++++
 tools/perf/util/symbol-minimal.c                |   6 +
 tools/perf/util/symbol.c                        |  41 +++
 tools/perf/util/symbol.h                        |   3 +
 6 files changed, 568 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
index e9a8349..fd77d81 100644
--- a/tools/perf/Documentation/perf-buildid-cache.txt
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -21,6 +21,19 @@ OPTIONS
 -a::
 --add=::
         Add specified file to the cache.
+-k::
+--kcore::
+        Add specified kcore file to the cache. For the current host that is
+        /proc/kcore which requires root permissions to read. Be aware that
+        running 'perf buildid-cache' as root may update root's build-id cache
+        not the user's. Use the -v option to see where the file is created.
+        Note that the copied file contains only code sections not the whole core
+        image. Note also that files "kallsyms" and "modules" must also be in the
+        same directory and are also copied.  All 3 files are created with read
+        permissions for root only. kcore will not be added if there is already a
+        kcore in the cache (with the same build-id) that has the same modules at
+        the same addresses. Use the -v option to see if a copy of kcore is
+        actually made.
 -r::
 --remove=::
         Remove specified file from the cache.
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
index c96c8fa..8140b7b 100644
--- a/tools/perf/builtin-buildid-cache.c
+++ b/tools/perf/builtin-buildid-cache.c
@@ -6,6 +6,11 @@
  * Copyright (C) 2010, Red Hat Inc.
  * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
  */
+#include <sys/types.h>
+#include <sys/time.h>
+#include <time.h>
+#include <dirent.h>
+#include <unistd.h>
 #include "builtin.h"
 #include "perf.h"
 #include "util/cache.h"
@@ -17,6 +22,140 @@
 #include "util/session.h"
 #include "util/symbol.h"
 
+static int build_id_cache__kcore_buildid(const char *proc_dir, char *sbuildid)
+{
+	char root_dir[PATH_MAX];
+	char notes[PATH_MAX];
+	u8 build_id[BUILD_ID_SIZE];
+	char *p;
+
+	strlcpy(root_dir, proc_dir, sizeof(root_dir));
+
+	p = strrchr(root_dir, '/');
+	if (!p)
+		return -1;
+	*p = '\0';
+
+	scnprintf(notes, sizeof(notes), "%s/sys/kernel/notes", root_dir);
+
+	if (sysfs__read_build_id(notes, build_id, sizeof(build_id)))
+		return -1;
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuildid);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_dir(char *dir, size_t sz)
+{
+	struct timeval tv;
+	struct tm tm;
+	char dt[32];
+
+	if (gettimeofday(&tv, NULL) || !localtime_r(&tv.tv_sec, &tm))
+		return -1;
+
+	if (!strftime(dt, sizeof(dt), "%Y%m%d%H%M%S", &tm))
+		return -1;
+
+	scnprintf(dir, sz, "%s%02u", dt, (unsigned)tv.tv_usec / 10000);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir,
+					  size_t to_dir_sz)
+{
+	char from[PATH_MAX];
+	char to[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(to_dir);
+	if (!d)
+		return -1;
+
+	scnprintf(from, sizeof(from), "%s/modules", from_dir);
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(to, sizeof(to), "%s/%s/modules", to_dir,
+			  dent->d_name);
+		if (!compare_proc_modules(from, to)) {
+			scnprintf(to, sizeof(to), "%s/%s", to_dir,
+				  dent->d_name);
+			strlcpy(to_dir, to, to_dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static int build_id_cache__add_kcore(const char *filename, const char *debugdir)
+{
+	char dir[32], sbuildid[BUILD_ID_SIZE * 2 + 1];
+	char from_dir[PATH_MAX], to_dir[PATH_MAX];
+	char *p;
+
+	strlcpy(from_dir, filename, sizeof(from_dir));
+
+	p = strrchr(from_dir, '/');
+	if (!p || strcmp(p + 1, "kcore"))
+		return -1;
+	*p = '\0';
+
+	if (build_id_cache__kcore_buildid(from_dir, sbuildid))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s",
+		  debugdir, sbuildid);
+
+	if (!build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
+		pr_debug("same kcore found in %s\n", to_dir);
+		return 0;
+	}
+
+	if (build_id_cache__kcore_dir(dir, sizeof(dir)))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s/%s",
+		  debugdir, sbuildid, dir);
+
+	if (mkdir_p(to_dir, 0755))
+		return -1;
+
+	if (kcore_copy(from_dir, to_dir)) {
+		/* Remove YYYYmmddHHMMSShh directory */
+		if (!rmdir(to_dir)) {
+			p = strrchr(to_dir, '/');
+			if (p)
+				*p = '\0';
+			/* Try to remove buildid directory */
+			if (!rmdir(to_dir)) {
+				p = strrchr(to_dir, '/');
+				if (p)
+					*p = '\0';
+				/* Try to remove [kernel.kcore] directory */
+				rmdir(to_dir);
+			}
+		}
+		return -1;
+	}
+
+	pr_debug("kcore added to build-id cache directory %s\n", to_dir);
+
+	return 0;
+}
+
 static int build_id_cache__add_file(const char *filename, const char *debugdir)
 {
 	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
@@ -130,11 +269,14 @@ int cmd_buildid_cache(int argc, const char **argv,
 	char const *add_name_list_str = NULL,
 		   *remove_name_list_str = NULL,
 		   *missing_filename = NULL,
-		   *update_name_list_str = NULL;
+		   *update_name_list_str = NULL,
+		   *kcore_filename;
 
 	const struct option buildid_cache_options[] = {
 	OPT_STRING('a', "add", &add_name_list_str,
 		   "file list", "file(s) to add"),
+	OPT_STRING('k', "kcore", &kcore_filename,
+		   "file", "kcore file to add"),
 	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
 		    "file(s) to remove"),
 	OPT_STRING('M', "missing", &missing_filename, "file",
@@ -217,5 +359,9 @@ int cmd_buildid_cache(int argc, const char **argv,
 		}
 	}
 
+	if (kcore_filename &&
+	    build_id_cache__add_kcore(kcore_filename, debugdir))
+		pr_warning("Couldn't add %s\n", kcore_filename);
+
 	return ret;
 }
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index 499c71d..c4a65ec 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -1202,6 +1202,364 @@ static off_t kcore__write(struct kcore *kcore)
 	return elf_update(kcore->elf, ELF_C_WRITE);
 }
 
+struct phdr_data {
+	off_t offset;
+	u64 addr;
+	u64 len;
+};
+
+struct kcore_copy_info {
+	u64 stext;
+	u64 etext;
+	u64 first_symbol;
+	u64 last_symbol;
+	u64 first_module;
+	u64 last_module_symbol;
+	struct phdr_data kernel_map;
+	struct phdr_data modules_map;
+};
+
+static int kcore_copy__process_kallsyms(void *arg, const char *name, char type,
+					u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!symbol_type__is_a(type, MAP__FUNCTION))
+		return 0;
+
+	if (strchr(name, '[')) {
+		if (start > kci->last_module_symbol)
+			kci->last_module_symbol = start;
+		return 0;
+	}
+
+	if (!kci->first_symbol || start < kci->first_symbol)
+		kci->first_symbol = start;
+
+	if (!kci->last_symbol || start > kci->last_symbol)
+		kci->last_symbol = start;
+
+	if (!strcmp(name, "_stext")) {
+		kci->stext = start;
+		return 0;
+	}
+
+	if (!strcmp(name, "_etext")) {
+		kci->etext = start;
+		return 0;
+	}
+
+	return 0;
+}
+
+static int kcore_copy__parse_kallsyms(struct kcore_copy_info *kci,
+				      const char *dir)
+{
+	char kallsyms_filename[PATH_MAX];
+
+	scnprintf(kallsyms_filename, PATH_MAX, "%s/kallsyms", dir);
+
+	if (symbol__restricted_filename(kallsyms_filename, "/proc/kallsyms"))
+		return -1;
+
+	if (kallsyms__parse(kallsyms_filename, kci,
+			    kcore_copy__process_kallsyms) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__process_modules(void *arg,
+				       const char *name __maybe_unused,
+				       u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!kci->first_module || start < kci->first_module)
+		kci->first_module = start;
+
+	return 0;
+}
+
+static int kcore_copy__parse_modules(struct kcore_copy_info *kci,
+				     const char *dir)
+{
+	char modules_filename[PATH_MAX];
+
+	scnprintf(modules_filename, PATH_MAX, "%s/modules", dir);
+
+	if (symbol__restricted_filename(modules_filename, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(modules_filename, kci,
+			   kcore_copy__process_modules) < 0)
+		return -1;
+
+	return 0;
+}
+
+static void kcore_copy__map(struct phdr_data *p, u64 start, u64 end, u64 pgoff,
+			    u64 s, u64 e)
+{
+	if (p->addr || s < start || s >= end)
+		return;
+
+	p->addr = s;
+	p->offset = (s - start) + pgoff;
+	p->len = e < end ? e - s : end - s;
+}
+
+static int kcore_copy__read_map(u64 start, u64 len, u64 pgoff, void *data)
+{
+	struct kcore_copy_info *kci = data;
+	u64 end = start + len;
+
+	kcore_copy__map(&kci->kernel_map, start, end, pgoff, kci->stext,
+			kci->etext);
+
+	kcore_copy__map(&kci->modules_map, start, end, pgoff, kci->first_module,
+			kci->last_module_symbol);
+
+	return 0;
+}
+
+static int kcore_copy__read_maps(struct kcore_copy_info *kci, Elf *elf)
+{
+	if (elf_read_maps(elf, true, kcore_copy__read_map, kci) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__calc_maps(struct kcore_copy_info *kci, const char *dir,
+				 Elf *elf)
+{
+	if (kcore_copy__parse_kallsyms(kci, dir))
+		return -1;
+
+	if (kcore_copy__parse_modules(kci, dir))
+		return -1;
+
+	if (kci->stext)
+		kci->stext = round_down(kci->stext, page_size);
+	else
+		kci->stext = round_down(kci->first_symbol, page_size);
+
+	if (kci->etext) {
+		kci->etext = round_up(kci->etext, page_size);
+	} else if (kci->last_symbol) {
+		kci->etext = round_up(kci->last_symbol, page_size);
+		kci->etext += page_size;
+	}
+
+	kci->first_module = round_down(kci->first_module, page_size);
+
+	if (kci->last_module_symbol) {
+		kci->last_module_symbol = round_up(kci->last_module_symbol,
+						   page_size);
+		kci->last_module_symbol += page_size;
+	}
+
+	if (!kci->stext || !kci->etext)
+		return -1;
+
+	if (kci->first_module && !kci->last_module_symbol)
+		return -1;
+
+	return kcore_copy__read_maps(kci, elf);
+}
+
+static int kcore_copy__copy_file(const char *from_dir, const char *to_dir,
+				 const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return copyfile_mode(from_filename, to_filename, 0400);
+}
+
+static int kcore_copy__unlink(const char *dir, const char *name)
+{
+	char filename[PATH_MAX];
+
+	scnprintf(filename, PATH_MAX, "%s/%s", dir, name);
+
+	return unlink(filename);
+}
+
+static int kcore_copy__compare_fds(int from, int to)
+{
+	char buf_from[page_size];
+	char buf_to[page_size];
+	ssize_t ret;
+	size_t len, n;
+
+	while (1) {
+		/* Use read because mmap won't work on proc files */
+		ret = read(from, buf_from, page_size);
+		if (ret < 0)
+			return -1;
+
+		if (!ret)
+			return 0;
+
+		n = ret;
+		len = ret;
+		while (n) {
+			ret = read(to, buf_to, n);
+			if (ret <= 0)
+				return -1;
+			n -= ret;
+		}
+
+		if (memcmp(buf_from, buf_to, len))
+			return -1;
+	}
+}
+
+static int kcore_copy__compare_files(const char *from_filename,
+				     const char *to_filename)
+{
+	int from, to, err = -1;
+
+	from = open(from_filename, O_RDONLY);
+	if (from < 0)
+		return -1;
+
+	to = open(to_filename, O_RDONLY);
+	if (to < 0)
+		goto out_close_from;
+
+	err = kcore_copy__compare_fds(from, to);
+
+	close(to);
+out_close_from:
+	close(from);
+	return err;
+}
+
+static int kcore_copy__compare_file(const char *from_dir, const char *to_dir,
+				    const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return kcore_copy__compare_files(from_filename, to_filename);
+}
+
+/**
+ * kcore_copy - copy kallsyms, modules and kcore from one directory to another.
+ * @from_dir: from directory
+ * @to_dir: to directory
+ *
+ * This function copies kallsyms, modules and kcore files from one directory to
+ * another.  kallsyms and modules are copied entirely.  Only code segments are
+ * copied from kcore.  It is assumed that two segments suffice: one for the
+ * kernel proper and one for all the modules.  The code segments are determined
+ * from kallsyms and modules files.  The kernel map starts at _stext or the
+ * lowest function symbol, and ends at _etext or the highest function symbol.
+ * The module map starts at the lowest module address and ends at the highest
+ * module symbol.  Start addresses are rounded down to the nearest page.  End
+ * addresses are rounded up to the nearest page.  An extra page is added to the
+ * highest kernel symbol and highest module symbol to, hopefully, encompass that
+ * symbol too.  Because it contains only code sections, the resulting kcore is
+ * unusual.  One significant peculiarity is that the mapping (start -> pgoff)
+ * is not the same for the kernel map and the modules map.  That happens because
+ * the data is copied adjacently whereas the original kcore has gaps.  Finally,
+ * kallsyms and modules files are compared with their copies to check that
+ * modules have not been loaded or unloaded while the copies were taking place.
+ *
+ * Return: %0 on success, %-1 on failure.
+ */
+int kcore_copy(const char *from_dir, const char *to_dir)
+{
+	struct kcore kcore;
+	struct kcore extract;
+	size_t count = 2;
+	int index = 0, err = -1;
+	off_t offset = page_size, sz, modules_offset = 0;
+	struct kcore_copy_info kci = {0};
+	char kcore_filename[PATH_MAX];
+	char extract_filename[PATH_MAX];
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "kallsyms"))
+		return -1;
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "modules"))
+		goto out_unlink_kallsyms;
+
+	scnprintf(kcore_filename, PATH_MAX, "%s/kcore", from_dir);
+	scnprintf(extract_filename, PATH_MAX, "%s/kcore", to_dir);
+
+	if (kcore__open(&kcore, kcore_filename))
+		goto out_unlink_modules;
+
+	if (kcore_copy__calc_maps(&kci, from_dir, kcore.elf))
+		goto out_kcore_close;
+
+	if (kcore__init(&extract, extract_filename, kcore.elfclass, false))
+		goto out_kcore_close;
+
+	if (!kci.modules_map.addr)
+		count -= 1;
+
+	if (kcore__copy_hdr(&kcore, &extract, count))
+		goto out_extract_close;
+
+	if (kcore__add_phdr(&extract, index++, offset, kci.kernel_map.addr,
+			    kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (kci.modules_map.addr) {
+		modules_offset = offset + kci.kernel_map.len;
+		if (kcore__add_phdr(&extract, index, modules_offset,
+				    kci.modules_map.addr, kci.modules_map.len))
+			goto out_extract_close;
+	}
+
+	sz = kcore__write(&extract);
+	if (sz < 0 || sz > offset)
+		goto out_extract_close;
+
+	if (copy_bytes(kcore.fd, kci.kernel_map.offset, extract.fd, offset,
+		       kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (modules_offset && copy_bytes(kcore.fd, kci.modules_map.offset,
+					 extract.fd, modules_offset,
+					 kci.modules_map.len))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "modules"))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "kallsyms"))
+		goto out_extract_close;
+
+	err = 0;
+
+out_extract_close:
+	kcore__close(&extract);
+	if (err)
+		unlink(extract_filename);
+out_kcore_close:
+	kcore__close(&kcore);
+out_unlink_modules:
+	if (err)
+		kcore_copy__unlink(to_dir, "modules");
+out_unlink_kallsyms:
+	if (err)
+		kcore_copy__unlink(to_dir, "kallsyms");
+
+	return err;
+}
+
 int kcore_extract__create(struct kcore_extract *kce)
 {
 	struct kcore kcore;
diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c
index 928556d..2d2dd05 100644
--- a/tools/perf/util/symbol-minimal.c
+++ b/tools/perf/util/symbol-minimal.c
@@ -317,6 +317,12 @@ void kcore_extract__delete(struct kcore_extract *kce __maybe_unused)
 {
 }
 
+int kcore_copy(const char *from_dir __maybe_unused,
+	       const char *to_dir __maybe_unused)
+{
+	return -1;
+}
+
 void symbol__elf_init(void)
 {
 }
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index b2f60dd..ef2680e 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -893,6 +893,47 @@ static int read_proc_modules(const char *filename, struct rb_root *modules)
 	return 0;
 }
 
+int compare_proc_modules(const char *from, const char *to)
+{
+	struct rb_root from_modules = RB_ROOT;
+	struct rb_root to_modules = RB_ROOT;
+	struct rb_node *from_node, *to_node;
+	struct module_info *from_m, *to_m;
+	int ret = -1;
+
+	if (read_proc_modules(from, &from_modules))
+		return -1;
+
+	if (read_proc_modules(to, &to_modules))
+		goto out_delete_from;
+
+	from_node = rb_first(&from_modules);
+	to_node = rb_first(&to_modules);
+	while (from_node) {
+		if (!to_node)
+			break;
+
+		from_m = rb_entry(from_node, struct module_info, rb_node);
+		to_m = rb_entry(to_node, struct module_info, rb_node);
+
+		if (from_m->start != to_m->start ||
+		    strcmp(from_m->name, to_m->name))
+			break;
+
+		from_node = rb_next(from_node);
+		to_node = rb_next(to_node);
+	}
+
+	if (!from_node && !to_node)
+		ret = 0;
+
+	delete_modules(&to_modules);
+out_delete_from:
+	delete_modules(&from_modules);
+
+	return ret;
+}
+
 static int do_validate_kcore_modules(const char *filename, struct map *map,
 				  struct map_groups *kmaps)
 {
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index fb107e1..07de8fe 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -270,4 +270,7 @@ struct kcore_extract {
 int kcore_extract__create(struct kcore_extract *kce);
 void kcore_extract__delete(struct kcore_extract *kce);
 
+int kcore_copy(const char *from_dir, const char *to_dir);
+int compare_proc_modules(const char *from, const char *to);
+
 #endif /* __PERF_SYMBOL */
-- 
1.7.11.7

Re: [PATCH V7 5/7] perf buildid-cache: add ability to add kcore to the cache

[Arnaldo Carvalho de Melo]

Em Mon, Oct 14, 2013 at 01:43:42PM +0300, Adrian Hunter escreveu:
> +static int kcore_copy__compare_fds(int from, int to)
<SNIP>
> +	while (1) {
> +		/* Use read because mmap won't work on proc files */
> +		ret = read(from, buf_from, page_size);
> +		if (ret < 0)
> +			return -1;
> +		if (!ret)
> +			return 0;
> +		n = ret;
> +		len = ret;
> +		while (n) {
> +			ret = read(to, buf_to, n);
> +			if (ret <= 0)
> +				return -1;
> +			n -= ret;

You can't go on potentially rewriting 'buf_to' like that, consider using
readn().

> +		}
> +
> +		if (memcmp(buf_from, buf_to, len))
> +			return -1;
> +	}
> +}

Re: [PATCH V7 5/7] perf buildid-cache: add ability to add kcore to the cache

[Adrian Hunter]

kcore can be used to view the running kernel object code.
However, kcore changes as modules are loaded and unloaded,
and when the kernel decides to modify its own code.
Consequently it is useful to create a copy of kcore at a
particular time.  Unlike vmlinux, kcore is not unique
for a given build-id.  And in addition, the kallsyms
and modules files are also needed.  The tool therefore
creates a directory:

	~/.debug/[kernel.kcore]/<build-id>/<YYYYmmddHHMMSShh>

which contains: kcore, kallsyms and modules.

Note that the copied kcore contains only code sections.
See the kcore_copy() function for how that is determined.

The tool will not make additional copies of kcore if there
is already one with the same modules at the same addresses.

Currently, perf tools will not look for kcore in the cache.
That is addressed in another patch.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/Documentation/perf-buildid-cache.txt |  13 +
 tools/perf/builtin-buildid-cache.c              | 148 +++++++++-
 tools/perf/util/symbol-elf.c                    | 366 ++++++++++++++++++++++++
 tools/perf/util/symbol-minimal.c                |   6 +
 tools/perf/util/symbol.c                        |  41 +++
 tools/perf/util/symbol.h                        |   3 +
 6 files changed, 576 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
index e9a8349..fd77d81 100644
--- a/tools/perf/Documentation/perf-buildid-cache.txt
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -21,6 +21,19 @@ OPTIONS
 -a::
 --add=::
         Add specified file to the cache.
+-k::
+--kcore::
+        Add specified kcore file to the cache. For the current host that is
+        /proc/kcore which requires root permissions to read. Be aware that
+        running 'perf buildid-cache' as root may update root's build-id cache
+        not the user's. Use the -v option to see where the file is created.
+        Note that the copied file contains only code sections not the whole core
+        image. Note also that files "kallsyms" and "modules" must also be in the
+        same directory and are also copied.  All 3 files are created with read
+        permissions for root only. kcore will not be added if there is already a
+        kcore in the cache (with the same build-id) that has the same modules at
+        the same addresses. Use the -v option to see if a copy of kcore is
+        actually made.
 -r::
 --remove=::
         Remove specified file from the cache.
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
index c96c8fa..8140b7b 100644
--- a/tools/perf/builtin-buildid-cache.c
+++ b/tools/perf/builtin-buildid-cache.c
@@ -6,6 +6,11 @@
  * Copyright (C) 2010, Red Hat Inc.
  * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
  */
+#include <sys/types.h>
+#include <sys/time.h>
+#include <time.h>
+#include <dirent.h>
+#include <unistd.h>
 #include "builtin.h"
 #include "perf.h"
 #include "util/cache.h"
@@ -17,6 +22,140 @@
 #include "util/session.h"
 #include "util/symbol.h"
 
+static int build_id_cache__kcore_buildid(const char *proc_dir, char *sbuildid)
+{
+	char root_dir[PATH_MAX];
+	char notes[PATH_MAX];
+	u8 build_id[BUILD_ID_SIZE];
+	char *p;
+
+	strlcpy(root_dir, proc_dir, sizeof(root_dir));
+
+	p = strrchr(root_dir, '/');
+	if (!p)
+		return -1;
+	*p = '\0';
+
+	scnprintf(notes, sizeof(notes), "%s/sys/kernel/notes", root_dir);
+
+	if (sysfs__read_build_id(notes, build_id, sizeof(build_id)))
+		return -1;
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuildid);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_dir(char *dir, size_t sz)
+{
+	struct timeval tv;
+	struct tm tm;
+	char dt[32];
+
+	if (gettimeofday(&tv, NULL) || !localtime_r(&tv.tv_sec, &tm))
+		return -1;
+
+	if (!strftime(dt, sizeof(dt), "%Y%m%d%H%M%S", &tm))
+		return -1;
+
+	scnprintf(dir, sz, "%s%02u", dt, (unsigned)tv.tv_usec / 10000);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir,
+					  size_t to_dir_sz)
+{
+	char from[PATH_MAX];
+	char to[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(to_dir);
+	if (!d)
+		return -1;
+
+	scnprintf(from, sizeof(from), "%s/modules", from_dir);
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(to, sizeof(to), "%s/%s/modules", to_dir,
+			  dent->d_name);
+		if (!compare_proc_modules(from, to)) {
+			scnprintf(to, sizeof(to), "%s/%s", to_dir,
+				  dent->d_name);
+			strlcpy(to_dir, to, to_dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static int build_id_cache__add_kcore(const char *filename, const char *debugdir)
+{
+	char dir[32], sbuildid[BUILD_ID_SIZE * 2 + 1];
+	char from_dir[PATH_MAX], to_dir[PATH_MAX];
+	char *p;
+
+	strlcpy(from_dir, filename, sizeof(from_dir));
+
+	p = strrchr(from_dir, '/');
+	if (!p || strcmp(p + 1, "kcore"))
+		return -1;
+	*p = '\0';
+
+	if (build_id_cache__kcore_buildid(from_dir, sbuildid))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s",
+		  debugdir, sbuildid);
+
+	if (!build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
+		pr_debug("same kcore found in %s\n", to_dir);
+		return 0;
+	}
+
+	if (build_id_cache__kcore_dir(dir, sizeof(dir)))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s/%s",
+		  debugdir, sbuildid, dir);
+
+	if (mkdir_p(to_dir, 0755))
+		return -1;
+
+	if (kcore_copy(from_dir, to_dir)) {
+		/* Remove YYYYmmddHHMMSShh directory */
+		if (!rmdir(to_dir)) {
+			p = strrchr(to_dir, '/');
+			if (p)
+				*p = '\0';
+			/* Try to remove buildid directory */
+			if (!rmdir(to_dir)) {
+				p = strrchr(to_dir, '/');
+				if (p)
+					*p = '\0';
+				/* Try to remove [kernel.kcore] directory */
+				rmdir(to_dir);
+			}
+		}
+		return -1;
+	}
+
+	pr_debug("kcore added to build-id cache directory %s\n", to_dir);
+
+	return 0;
+}
+
 static int build_id_cache__add_file(const char *filename, const char *debugdir)
 {
 	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
@@ -130,11 +269,14 @@ int cmd_buildid_cache(int argc, const char **argv,
 	char const *add_name_list_str = NULL,
 		   *remove_name_list_str = NULL,
 		   *missing_filename = NULL,
-		   *update_name_list_str = NULL;
+		   *update_name_list_str = NULL,
+		   *kcore_filename;
 
 	const struct option buildid_cache_options[] = {
 	OPT_STRING('a', "add", &add_name_list_str,
 		   "file list", "file(s) to add"),
+	OPT_STRING('k', "kcore", &kcore_filename,
+		   "file", "kcore file to add"),
 	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
 		    "file(s) to remove"),
 	OPT_STRING('M', "missing", &missing_filename, "file",
@@ -217,5 +359,9 @@ int cmd_buildid_cache(int argc, const char **argv,
 		}
 	}
 
+	if (kcore_filename &&
+	    build_id_cache__add_kcore(kcore_filename, debugdir))
+		pr_warning("Couldn't add %s\n", kcore_filename);
+
 	return ret;
 }
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index 499c71d..2c6e830 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -1202,6 +1202,372 @@ static off_t kcore__write(struct kcore *kcore)
 	return elf_update(kcore->elf, ELF_C_WRITE);
 }
 
+struct phdr_data {
+	off_t offset;
+	u64 addr;
+	u64 len;
+};
+
+struct kcore_copy_info {
+	u64 stext;
+	u64 etext;
+	u64 first_symbol;
+	u64 last_symbol;
+	u64 first_module;
+	u64 last_module_symbol;
+	struct phdr_data kernel_map;
+	struct phdr_data modules_map;
+};
+
+static int kcore_copy__process_kallsyms(void *arg, const char *name, char type,
+					u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!symbol_type__is_a(type, MAP__FUNCTION))
+		return 0;
+
+	if (strchr(name, '[')) {
+		if (start > kci->last_module_symbol)
+			kci->last_module_symbol = start;
+		return 0;
+	}
+
+	if (!kci->first_symbol || start < kci->first_symbol)
+		kci->first_symbol = start;
+
+	if (!kci->last_symbol || start > kci->last_symbol)
+		kci->last_symbol = start;
+
+	if (!strcmp(name, "_stext")) {
+		kci->stext = start;
+		return 0;
+	}
+
+	if (!strcmp(name, "_etext")) {
+		kci->etext = start;
+		return 0;
+	}
+
+	return 0;
+}
+
+static int kcore_copy__parse_kallsyms(struct kcore_copy_info *kci,
+				      const char *dir)
+{
+	char kallsyms_filename[PATH_MAX];
+
+	scnprintf(kallsyms_filename, PATH_MAX, "%s/kallsyms", dir);
+
+	if (symbol__restricted_filename(kallsyms_filename, "/proc/kallsyms"))
+		return -1;
+
+	if (kallsyms__parse(kallsyms_filename, kci,
+			    kcore_copy__process_kallsyms) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__process_modules(void *arg,
+				       const char *name __maybe_unused,
+				       u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!kci->first_module || start < kci->first_module)
+		kci->first_module = start;
+
+	return 0;
+}
+
+static int kcore_copy__parse_modules(struct kcore_copy_info *kci,
+				     const char *dir)
+{
+	char modules_filename[PATH_MAX];
+
+	scnprintf(modules_filename, PATH_MAX, "%s/modules", dir);
+
+	if (symbol__restricted_filename(modules_filename, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(modules_filename, kci,
+			   kcore_copy__process_modules) < 0)
+		return -1;
+
+	return 0;
+}
+
+static void kcore_copy__map(struct phdr_data *p, u64 start, u64 end, u64 pgoff,
+			    u64 s, u64 e)
+{
+	if (p->addr || s < start || s >= end)
+		return;
+
+	p->addr = s;
+	p->offset = (s - start) + pgoff;
+	p->len = e < end ? e - s : end - s;
+}
+
+static int kcore_copy__read_map(u64 start, u64 len, u64 pgoff, void *data)
+{
+	struct kcore_copy_info *kci = data;
+	u64 end = start + len;
+
+	kcore_copy__map(&kci->kernel_map, start, end, pgoff, kci->stext,
+			kci->etext);
+
+	kcore_copy__map(&kci->modules_map, start, end, pgoff, kci->first_module,
+			kci->last_module_symbol);
+
+	return 0;
+}
+
+static int kcore_copy__read_maps(struct kcore_copy_info *kci, Elf *elf)
+{
+	if (elf_read_maps(elf, true, kcore_copy__read_map, kci) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__calc_maps(struct kcore_copy_info *kci, const char *dir,
+				 Elf *elf)
+{
+	if (kcore_copy__parse_kallsyms(kci, dir))
+		return -1;
+
+	if (kcore_copy__parse_modules(kci, dir))
+		return -1;
+
+	if (kci->stext)
+		kci->stext = round_down(kci->stext, page_size);
+	else
+		kci->stext = round_down(kci->first_symbol, page_size);
+
+	if (kci->etext) {
+		kci->etext = round_up(kci->etext, page_size);
+	} else if (kci->last_symbol) {
+		kci->etext = round_up(kci->last_symbol, page_size);
+		kci->etext += page_size;
+	}
+
+	kci->first_module = round_down(kci->first_module, page_size);
+
+	if (kci->last_module_symbol) {
+		kci->last_module_symbol = round_up(kci->last_module_symbol,
+						   page_size);
+		kci->last_module_symbol += page_size;
+	}
+
+	if (!kci->stext || !kci->etext)
+		return -1;
+
+	if (kci->first_module && !kci->last_module_symbol)
+		return -1;
+
+	return kcore_copy__read_maps(kci, elf);
+}
+
+static int kcore_copy__copy_file(const char *from_dir, const char *to_dir,
+				 const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return copyfile_mode(from_filename, to_filename, 0400);
+}
+
+static int kcore_copy__unlink(const char *dir, const char *name)
+{
+	char filename[PATH_MAX];
+
+	scnprintf(filename, PATH_MAX, "%s/%s", dir, name);
+
+	return unlink(filename);
+}
+
+static int kcore_copy__compare_fds(int from, int to)
+{
+	char *buf_from;
+	char *buf_to;
+	ssize_t ret;
+	size_t len;
+	int err = -1;
+
+	buf_from = malloc(page_size);
+	buf_to = malloc(page_size);
+	if (!buf_from || !buf_to)
+		goto out;
+
+	while (1) {
+		/* Use read because mmap won't work on proc files */
+		ret = read(from, buf_from, page_size);
+		if (ret < 0)
+			goto out;
+
+		if (!ret)
+			break;
+
+		len = ret;
+
+		if (readn(to, buf_to, len) != (int)len)
+			goto out;
+
+		if (memcmp(buf_from, buf_to, len))
+			goto out;
+	}
+
+	err = 0;
+out:
+	free(buf_to);
+	free(buf_from);
+	return err;
+}
+
+static int kcore_copy__compare_files(const char *from_filename,
+				     const char *to_filename)
+{
+	int from, to, err = -1;
+
+	from = open(from_filename, O_RDONLY);
+	if (from < 0)
+		return -1;
+
+	to = open(to_filename, O_RDONLY);
+	if (to < 0)
+		goto out_close_from;
+
+	err = kcore_copy__compare_fds(from, to);
+
+	close(to);
+out_close_from:
+	close(from);
+	return err;
+}
+
+static int kcore_copy__compare_file(const char *from_dir, const char *to_dir,
+				    const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return kcore_copy__compare_files(from_filename, to_filename);
+}
+
+/**
+ * kcore_copy - copy kallsyms, modules and kcore from one directory to another.
+ * @from_dir: from directory
+ * @to_dir: to directory
+ *
+ * This function copies kallsyms, modules and kcore files from one directory to
+ * another.  kallsyms and modules are copied entirely.  Only code segments are
+ * copied from kcore.  It is assumed that two segments suffice: one for the
+ * kernel proper and one for all the modules.  The code segments are determined
+ * from kallsyms and modules files.  The kernel map starts at _stext or the
+ * lowest function symbol, and ends at _etext or the highest function symbol.
+ * The module map starts at the lowest module address and ends at the highest
+ * module symbol.  Start addresses are rounded down to the nearest page.  End
+ * addresses are rounded up to the nearest page.  An extra page is added to the
+ * highest kernel symbol and highest module symbol to, hopefully, encompass that
+ * symbol too.  Because it contains only code sections, the resulting kcore is
+ * unusual.  One significant peculiarity is that the mapping (start -> pgoff)
+ * is not the same for the kernel map and the modules map.  That happens because
+ * the data is copied adjacently whereas the original kcore has gaps.  Finally,
+ * kallsyms and modules files are compared with their copies to check that
+ * modules have not been loaded or unloaded while the copies were taking place.
+ *
+ * Return: %0 on success, %-1 on failure.
+ */
+int kcore_copy(const char *from_dir, const char *to_dir)
+{
+	struct kcore kcore;
+	struct kcore extract;
+	size_t count = 2;
+	int index = 0, err = -1;
+	off_t offset = page_size, sz, modules_offset = 0;
+	struct kcore_copy_info kci = {0};
+	char kcore_filename[PATH_MAX];
+	char extract_filename[PATH_MAX];
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "kallsyms"))
+		return -1;
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "modules"))
+		goto out_unlink_kallsyms;
+
+	scnprintf(kcore_filename, PATH_MAX, "%s/kcore", from_dir);
+	scnprintf(extract_filename, PATH_MAX, "%s/kcore", to_dir);
+
+	if (kcore__open(&kcore, kcore_filename))
+		goto out_unlink_modules;
+
+	if (kcore_copy__calc_maps(&kci, from_dir, kcore.elf))
+		goto out_kcore_close;
+
+	if (kcore__init(&extract, extract_filename, kcore.elfclass, false))
+		goto out_kcore_close;
+
+	if (!kci.modules_map.addr)
+		count -= 1;
+
+	if (kcore__copy_hdr(&kcore, &extract, count))
+		goto out_extract_close;
+
+	if (kcore__add_phdr(&extract, index++, offset, kci.kernel_map.addr,
+			    kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (kci.modules_map.addr) {
+		modules_offset = offset + kci.kernel_map.len;
+		if (kcore__add_phdr(&extract, index, modules_offset,
+				    kci.modules_map.addr, kci.modules_map.len))
+			goto out_extract_close;
+	}
+
+	sz = kcore__write(&extract);
+	if (sz < 0 || sz > offset)
+		goto out_extract_close;
+
+	if (copy_bytes(kcore.fd, kci.kernel_map.offset, extract.fd, offset,
+		       kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (modules_offset && copy_bytes(kcore.fd, kci.modules_map.offset,
+					 extract.fd, modules_offset,
+					 kci.modules_map.len))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "modules"))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "kallsyms"))
+		goto out_extract_close;
+
+	err = 0;
+
+out_extract_close:
+	kcore__close(&extract);
+	if (err)
+		unlink(extract_filename);
+out_kcore_close:
+	kcore__close(&kcore);
+out_unlink_modules:
+	if (err)
+		kcore_copy__unlink(to_dir, "modules");
+out_unlink_kallsyms:
+	if (err)
+		kcore_copy__unlink(to_dir, "kallsyms");
+
+	return err;
+}
+
 int kcore_extract__create(struct kcore_extract *kce)
 {
 	struct kcore kcore;
diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c
index 928556d..2d2dd05 100644
--- a/tools/perf/util/symbol-minimal.c
+++ b/tools/perf/util/symbol-minimal.c
@@ -317,6 +317,12 @@ void kcore_extract__delete(struct kcore_extract *kce __maybe_unused)
 {
 }
 
+int kcore_copy(const char *from_dir __maybe_unused,
+	       const char *to_dir __maybe_unused)
+{
+	return -1;
+}
+
 void symbol__elf_init(void)
 {
 }
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index b2f60dd..ef2680e 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -893,6 +893,47 @@ static int read_proc_modules(const char *filename, struct rb_root *modules)
 	return 0;
 }
 
+int compare_proc_modules(const char *from, const char *to)
+{
+	struct rb_root from_modules = RB_ROOT;
+	struct rb_root to_modules = RB_ROOT;
+	struct rb_node *from_node, *to_node;
+	struct module_info *from_m, *to_m;
+	int ret = -1;
+
+	if (read_proc_modules(from, &from_modules))
+		return -1;
+
+	if (read_proc_modules(to, &to_modules))
+		goto out_delete_from;
+
+	from_node = rb_first(&from_modules);
+	to_node = rb_first(&to_modules);
+	while (from_node) {
+		if (!to_node)
+			break;
+
+		from_m = rb_entry(from_node, struct module_info, rb_node);
+		to_m = rb_entry(to_node, struct module_info, rb_node);
+
+		if (from_m->start != to_m->start ||
+		    strcmp(from_m->name, to_m->name))
+			break;
+
+		from_node = rb_next(from_node);
+		to_node = rb_next(to_node);
+	}
+
+	if (!from_node && !to_node)
+		ret = 0;
+
+	delete_modules(&to_modules);
+out_delete_from:
+	delete_modules(&from_modules);
+
+	return ret;
+}
+
 static int do_validate_kcore_modules(const char *filename, struct map *map,
 				  struct map_groups *kmaps)
 {
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index fb107e1..07de8fe 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -270,4 +270,7 @@ struct kcore_extract {
 int kcore_extract__create(struct kcore_extract *kce);
 void kcore_extract__delete(struct kcore_extract *kce);
 
+int kcore_copy(const char *from_dir, const char *to_dir);
+int compare_proc_modules(const char *from, const char *to);
+
 #endif /* __PERF_SYMBOL */
-- 
1.7.11.7

[tip:perf/core] perf buildid-cache: Add ability to add kcore to the cache

[tip-bot for Adrian Hunter]

Commit-ID:  fc1b691d7651d9496e912de7e0fc73a5be3294af
Gitweb:     http://git.kernel.org/tip/fc1b691d7651d9496e912de7e0fc73a5be3294af
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 16:57:29 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 12:20:38 -0300

perf buildid-cache: Add ability to add kcore to the cache

kcore can be used to view the running kernel object code.  However,
kcore changes as modules are loaded and unloaded, and when the kernel
decides to modify its own code.  Consequently it is useful to create a
copy of kcore at a particular time.  Unlike vmlinux, kcore is not unique
for a given build-id.  And in addition, the kallsyms and modules files
are also needed.  The tool therefore creates a directory:

	~/.debug/[kernel.kcore]/<build-id>/<YYYYmmddHHMMSShh>

which contains: kcore, kallsyms and modules.

Note that the copied kcore contains only code sections.  See the
kcore_copy() function for how that is determined.

The tool will not make additional copies of kcore if there is already
one with the same modules at the same addresses.

Currently, perf tools will not look for kcore in the cache.  That is
addressed in another patch.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/525BF849.5030405@intel.com
[ renamed 'index' to 'idx' to avoid shadowing string.h symbol in f12,
  use at least one member initializer when initializing a struct to
  zeros, also to fix the build on f12 ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-buildid-cache.txt |  13 +
 tools/perf/builtin-buildid-cache.c              | 148 +++++++++-
 tools/perf/util/symbol-elf.c                    | 366 ++++++++++++++++++++++++
 tools/perf/util/symbol-minimal.c                |   6 +
 tools/perf/util/symbol.c                        |  41 +++
 tools/perf/util/symbol.h                        |   3 +
 6 files changed, 576 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
index e9a8349..fd77d81 100644
--- a/tools/perf/Documentation/perf-buildid-cache.txt
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -21,6 +21,19 @@ OPTIONS
 -a::
 --add=::
         Add specified file to the cache.
+-k::
+--kcore::
+        Add specified kcore file to the cache. For the current host that is
+        /proc/kcore which requires root permissions to read. Be aware that
+        running 'perf buildid-cache' as root may update root's build-id cache
+        not the user's. Use the -v option to see where the file is created.
+        Note that the copied file contains only code sections not the whole core
+        image. Note also that files "kallsyms" and "modules" must also be in the
+        same directory and are also copied.  All 3 files are created with read
+        permissions for root only. kcore will not be added if there is already a
+        kcore in the cache (with the same build-id) that has the same modules at
+        the same addresses. Use the -v option to see if a copy of kcore is
+        actually made.
 -r::
 --remove=::
         Remove specified file from the cache.
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
index c96c8fa..8140b7b 100644
--- a/tools/perf/builtin-buildid-cache.c
+++ b/tools/perf/builtin-buildid-cache.c
@@ -6,6 +6,11 @@
  * Copyright (C) 2010, Red Hat Inc.
  * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
  */
+#include <sys/types.h>
+#include <sys/time.h>
+#include <time.h>
+#include <dirent.h>
+#include <unistd.h>
 #include "builtin.h"
 #include "perf.h"
 #include "util/cache.h"
@@ -17,6 +22,140 @@
 #include "util/session.h"
 #include "util/symbol.h"
 
+static int build_id_cache__kcore_buildid(const char *proc_dir, char *sbuildid)
+{
+	char root_dir[PATH_MAX];
+	char notes[PATH_MAX];
+	u8 build_id[BUILD_ID_SIZE];
+	char *p;
+
+	strlcpy(root_dir, proc_dir, sizeof(root_dir));
+
+	p = strrchr(root_dir, '/');
+	if (!p)
+		return -1;
+	*p = '\0';
+
+	scnprintf(notes, sizeof(notes), "%s/sys/kernel/notes", root_dir);
+
+	if (sysfs__read_build_id(notes, build_id, sizeof(build_id)))
+		return -1;
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuildid);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_dir(char *dir, size_t sz)
+{
+	struct timeval tv;
+	struct tm tm;
+	char dt[32];
+
+	if (gettimeofday(&tv, NULL) || !localtime_r(&tv.tv_sec, &tm))
+		return -1;
+
+	if (!strftime(dt, sizeof(dt), "%Y%m%d%H%M%S", &tm))
+		return -1;
+
+	scnprintf(dir, sz, "%s%02u", dt, (unsigned)tv.tv_usec / 10000);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir,
+					  size_t to_dir_sz)
+{
+	char from[PATH_MAX];
+	char to[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(to_dir);
+	if (!d)
+		return -1;
+
+	scnprintf(from, sizeof(from), "%s/modules", from_dir);
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(to, sizeof(to), "%s/%s/modules", to_dir,
+			  dent->d_name);
+		if (!compare_proc_modules(from, to)) {
+			scnprintf(to, sizeof(to), "%s/%s", to_dir,
+				  dent->d_name);
+			strlcpy(to_dir, to, to_dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static int build_id_cache__add_kcore(const char *filename, const char *debugdir)
+{
+	char dir[32], sbuildid[BUILD_ID_SIZE * 2 + 1];
+	char from_dir[PATH_MAX], to_dir[PATH_MAX];
+	char *p;
+
+	strlcpy(from_dir, filename, sizeof(from_dir));
+
+	p = strrchr(from_dir, '/');
+	if (!p || strcmp(p + 1, "kcore"))
+		return -1;
+	*p = '\0';
+
+	if (build_id_cache__kcore_buildid(from_dir, sbuildid))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s",
+		  debugdir, sbuildid);
+
+	if (!build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
+		pr_debug("same kcore found in %s\n", to_dir);
+		return 0;
+	}
+
+	if (build_id_cache__kcore_dir(dir, sizeof(dir)))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s/%s",
+		  debugdir, sbuildid, dir);
+
+	if (mkdir_p(to_dir, 0755))
+		return -1;
+
+	if (kcore_copy(from_dir, to_dir)) {
+		/* Remove YYYYmmddHHMMSShh directory */
+		if (!rmdir(to_dir)) {
+			p = strrchr(to_dir, '/');
+			if (p)
+				*p = '\0';
+			/* Try to remove buildid directory */
+			if (!rmdir(to_dir)) {
+				p = strrchr(to_dir, '/');
+				if (p)
+					*p = '\0';
+				/* Try to remove [kernel.kcore] directory */
+				rmdir(to_dir);
+			}
+		}
+		return -1;
+	}
+
+	pr_debug("kcore added to build-id cache directory %s\n", to_dir);
+
+	return 0;
+}
+
 static int build_id_cache__add_file(const char *filename, const char *debugdir)
 {
 	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
@@ -130,11 +269,14 @@ int cmd_buildid_cache(int argc, const char **argv,
 	char const *add_name_list_str = NULL,
 		   *remove_name_list_str = NULL,
 		   *missing_filename = NULL,
-		   *update_name_list_str = NULL;
+		   *update_name_list_str = NULL,
+		   *kcore_filename;
 
 	const struct option buildid_cache_options[] = {
 	OPT_STRING('a', "add", &add_name_list_str,
 		   "file list", "file(s) to add"),
+	OPT_STRING('k', "kcore", &kcore_filename,
+		   "file", "kcore file to add"),
 	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
 		    "file(s) to remove"),
 	OPT_STRING('M', "missing", &missing_filename, "file",
@@ -217,5 +359,9 @@ int cmd_buildid_cache(int argc, const char **argv,
 		}
 	}
 
+	if (kcore_filename &&
+	    build_id_cache__add_kcore(kcore_filename, debugdir))
+		pr_warning("Couldn't add %s\n", kcore_filename);
+
 	return ret;
 }
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index 499c71d..d6b8af3 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -1202,6 +1202,372 @@ static off_t kcore__write(struct kcore *kcore)
 	return elf_update(kcore->elf, ELF_C_WRITE);
 }
 
+struct phdr_data {
+	off_t offset;
+	u64 addr;
+	u64 len;
+};
+
+struct kcore_copy_info {
+	u64 stext;
+	u64 etext;
+	u64 first_symbol;
+	u64 last_symbol;
+	u64 first_module;
+	u64 last_module_symbol;
+	struct phdr_data kernel_map;
+	struct phdr_data modules_map;
+};
+
+static int kcore_copy__process_kallsyms(void *arg, const char *name, char type,
+					u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!symbol_type__is_a(type, MAP__FUNCTION))
+		return 0;
+
+	if (strchr(name, '[')) {
+		if (start > kci->last_module_symbol)
+			kci->last_module_symbol = start;
+		return 0;
+	}
+
+	if (!kci->first_symbol || start < kci->first_symbol)
+		kci->first_symbol = start;
+
+	if (!kci->last_symbol || start > kci->last_symbol)
+		kci->last_symbol = start;
+
+	if (!strcmp(name, "_stext")) {
+		kci->stext = start;
+		return 0;
+	}
+
+	if (!strcmp(name, "_etext")) {
+		kci->etext = start;
+		return 0;
+	}
+
+	return 0;
+}
+
+static int kcore_copy__parse_kallsyms(struct kcore_copy_info *kci,
+				      const char *dir)
+{
+	char kallsyms_filename[PATH_MAX];
+
+	scnprintf(kallsyms_filename, PATH_MAX, "%s/kallsyms", dir);
+
+	if (symbol__restricted_filename(kallsyms_filename, "/proc/kallsyms"))
+		return -1;
+
+	if (kallsyms__parse(kallsyms_filename, kci,
+			    kcore_copy__process_kallsyms) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__process_modules(void *arg,
+				       const char *name __maybe_unused,
+				       u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!kci->first_module || start < kci->first_module)
+		kci->first_module = start;
+
+	return 0;
+}
+
+static int kcore_copy__parse_modules(struct kcore_copy_info *kci,
+				     const char *dir)
+{
+	char modules_filename[PATH_MAX];
+
+	scnprintf(modules_filename, PATH_MAX, "%s/modules", dir);
+
+	if (symbol__restricted_filename(modules_filename, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(modules_filename, kci,
+			   kcore_copy__process_modules) < 0)
+		return -1;
+
+	return 0;
+}
+
+static void kcore_copy__map(struct phdr_data *p, u64 start, u64 end, u64 pgoff,
+			    u64 s, u64 e)
+{
+	if (p->addr || s < start || s >= end)
+		return;
+
+	p->addr = s;
+	p->offset = (s - start) + pgoff;
+	p->len = e < end ? e - s : end - s;
+}
+
+static int kcore_copy__read_map(u64 start, u64 len, u64 pgoff, void *data)
+{
+	struct kcore_copy_info *kci = data;
+	u64 end = start + len;
+
+	kcore_copy__map(&kci->kernel_map, start, end, pgoff, kci->stext,
+			kci->etext);
+
+	kcore_copy__map(&kci->modules_map, start, end, pgoff, kci->first_module,
+			kci->last_module_symbol);
+
+	return 0;
+}
+
+static int kcore_copy__read_maps(struct kcore_copy_info *kci, Elf *elf)
+{
+	if (elf_read_maps(elf, true, kcore_copy__read_map, kci) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__calc_maps(struct kcore_copy_info *kci, const char *dir,
+				 Elf *elf)
+{
+	if (kcore_copy__parse_kallsyms(kci, dir))
+		return -1;
+
+	if (kcore_copy__parse_modules(kci, dir))
+		return -1;
+
+	if (kci->stext)
+		kci->stext = round_down(kci->stext, page_size);
+	else
+		kci->stext = round_down(kci->first_symbol, page_size);
+
+	if (kci->etext) {
+		kci->etext = round_up(kci->etext, page_size);
+	} else if (kci->last_symbol) {
+		kci->etext = round_up(kci->last_symbol, page_size);
+		kci->etext += page_size;
+	}
+
+	kci->first_module = round_down(kci->first_module, page_size);
+
+	if (kci->last_module_symbol) {
+		kci->last_module_symbol = round_up(kci->last_module_symbol,
+						   page_size);
+		kci->last_module_symbol += page_size;
+	}
+
+	if (!kci->stext || !kci->etext)
+		return -1;
+
+	if (kci->first_module && !kci->last_module_symbol)
+		return -1;
+
+	return kcore_copy__read_maps(kci, elf);
+}
+
+static int kcore_copy__copy_file(const char *from_dir, const char *to_dir,
+				 const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return copyfile_mode(from_filename, to_filename, 0400);
+}
+
+static int kcore_copy__unlink(const char *dir, const char *name)
+{
+	char filename[PATH_MAX];
+
+	scnprintf(filename, PATH_MAX, "%s/%s", dir, name);
+
+	return unlink(filename);
+}
+
+static int kcore_copy__compare_fds(int from, int to)
+{
+	char *buf_from;
+	char *buf_to;
+	ssize_t ret;
+	size_t len;
+	int err = -1;
+
+	buf_from = malloc(page_size);
+	buf_to = malloc(page_size);
+	if (!buf_from || !buf_to)
+		goto out;
+
+	while (1) {
+		/* Use read because mmap won't work on proc files */
+		ret = read(from, buf_from, page_size);
+		if (ret < 0)
+			goto out;
+
+		if (!ret)
+			break;
+
+		len = ret;
+
+		if (readn(to, buf_to, len) != (int)len)
+			goto out;
+
+		if (memcmp(buf_from, buf_to, len))
+			goto out;
+	}
+
+	err = 0;
+out:
+	free(buf_to);
+	free(buf_from);
+	return err;
+}
+
+static int kcore_copy__compare_files(const char *from_filename,
+				     const char *to_filename)
+{
+	int from, to, err = -1;
+
+	from = open(from_filename, O_RDONLY);
+	if (from < 0)
+		return -1;
+
+	to = open(to_filename, O_RDONLY);
+	if (to < 0)
+		goto out_close_from;
+
+	err = kcore_copy__compare_fds(from, to);
+
+	close(to);
+out_close_from:
+	close(from);
+	return err;
+}
+
+static int kcore_copy__compare_file(const char *from_dir, const char *to_dir,
+				    const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return kcore_copy__compare_files(from_filename, to_filename);
+}
+
+/**
+ * kcore_copy - copy kallsyms, modules and kcore from one directory to another.
+ * @from_dir: from directory
+ * @to_dir: to directory
+ *
+ * This function copies kallsyms, modules and kcore files from one directory to
+ * another.  kallsyms and modules are copied entirely.  Only code segments are
+ * copied from kcore.  It is assumed that two segments suffice: one for the
+ * kernel proper and one for all the modules.  The code segments are determined
+ * from kallsyms and modules files.  The kernel map starts at _stext or the
+ * lowest function symbol, and ends at _etext or the highest function symbol.
+ * The module map starts at the lowest module address and ends at the highest
+ * module symbol.  Start addresses are rounded down to the nearest page.  End
+ * addresses are rounded up to the nearest page.  An extra page is added to the
+ * highest kernel symbol and highest module symbol to, hopefully, encompass that
+ * symbol too.  Because it contains only code sections, the resulting kcore is
+ * unusual.  One significant peculiarity is that the mapping (start -> pgoff)
+ * is not the same for the kernel map and the modules map.  That happens because
+ * the data is copied adjacently whereas the original kcore has gaps.  Finally,
+ * kallsyms and modules files are compared with their copies to check that
+ * modules have not been loaded or unloaded while the copies were taking place.
+ *
+ * Return: %0 on success, %-1 on failure.
+ */
+int kcore_copy(const char *from_dir, const char *to_dir)
+{
+	struct kcore kcore;
+	struct kcore extract;
+	size_t count = 2;
+	int idx = 0, err = -1;
+	off_t offset = page_size, sz, modules_offset = 0;
+	struct kcore_copy_info kci = { .stext = 0, };
+	char kcore_filename[PATH_MAX];
+	char extract_filename[PATH_MAX];
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "kallsyms"))
+		return -1;
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "modules"))
+		goto out_unlink_kallsyms;
+
+	scnprintf(kcore_filename, PATH_MAX, "%s/kcore", from_dir);
+	scnprintf(extract_filename, PATH_MAX, "%s/kcore", to_dir);
+
+	if (kcore__open(&kcore, kcore_filename))
+		goto out_unlink_modules;
+
+	if (kcore_copy__calc_maps(&kci, from_dir, kcore.elf))
+		goto out_kcore_close;
+
+	if (kcore__init(&extract, extract_filename, kcore.elfclass, false))
+		goto out_kcore_close;
+
+	if (!kci.modules_map.addr)
+		count -= 1;
+
+	if (kcore__copy_hdr(&kcore, &extract, count))
+		goto out_extract_close;
+
+	if (kcore__add_phdr(&extract, idx++, offset, kci.kernel_map.addr,
+			    kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (kci.modules_map.addr) {
+		modules_offset = offset + kci.kernel_map.len;
+		if (kcore__add_phdr(&extract, idx, modules_offset,
+				    kci.modules_map.addr, kci.modules_map.len))
+			goto out_extract_close;
+	}
+
+	sz = kcore__write(&extract);
+	if (sz < 0 || sz > offset)
+		goto out_extract_close;
+
+	if (copy_bytes(kcore.fd, kci.kernel_map.offset, extract.fd, offset,
+		       kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (modules_offset && copy_bytes(kcore.fd, kci.modules_map.offset,
+					 extract.fd, modules_offset,
+					 kci.modules_map.len))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "modules"))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "kallsyms"))
+		goto out_extract_close;
+
+	err = 0;
+
+out_extract_close:
+	kcore__close(&extract);
+	if (err)
+		unlink(extract_filename);
+out_kcore_close:
+	kcore__close(&kcore);
+out_unlink_modules:
+	if (err)
+		kcore_copy__unlink(to_dir, "modules");
+out_unlink_kallsyms:
+	if (err)
+		kcore_copy__unlink(to_dir, "kallsyms");
+
+	return err;
+}
+
 int kcore_extract__create(struct kcore_extract *kce)
 {
 	struct kcore kcore;
diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c
index 928556d..2d2dd05 100644
--- a/tools/perf/util/symbol-minimal.c
+++ b/tools/perf/util/symbol-minimal.c
@@ -317,6 +317,12 @@ void kcore_extract__delete(struct kcore_extract *kce __maybe_unused)
 {
 }
 
+int kcore_copy(const char *from_dir __maybe_unused,
+	       const char *to_dir __maybe_unused)
+{
+	return -1;
+}
+
 void symbol__elf_init(void)
 {
 }
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 76a9e93..b66c1ee 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -893,6 +893,47 @@ static int read_proc_modules(const char *filename, struct rb_root *modules)
 	return 0;
 }
 
+int compare_proc_modules(const char *from, const char *to)
+{
+	struct rb_root from_modules = RB_ROOT;
+	struct rb_root to_modules = RB_ROOT;
+	struct rb_node *from_node, *to_node;
+	struct module_info *from_m, *to_m;
+	int ret = -1;
+
+	if (read_proc_modules(from, &from_modules))
+		return -1;
+
+	if (read_proc_modules(to, &to_modules))
+		goto out_delete_from;
+
+	from_node = rb_first(&from_modules);
+	to_node = rb_first(&to_modules);
+	while (from_node) {
+		if (!to_node)
+			break;
+
+		from_m = rb_entry(from_node, struct module_info, rb_node);
+		to_m = rb_entry(to_node, struct module_info, rb_node);
+
+		if (from_m->start != to_m->start ||
+		    strcmp(from_m->name, to_m->name))
+			break;
+
+		from_node = rb_next(from_node);
+		to_node = rb_next(to_node);
+	}
+
+	if (!from_node && !to_node)
+		ret = 0;
+
+	delete_modules(&to_modules);
+out_delete_from:
+	delete_modules(&from_modules);
+
+	return ret;
+}
+
 static int do_validate_kcore_modules(const char *filename, struct map *map,
 				  struct map_groups *kmaps)
 {
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index fb107e1..07de8fe 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -270,4 +270,7 @@ struct kcore_extract {
 int kcore_extract__create(struct kcore_extract *kce);
 void kcore_extract__delete(struct kcore_extract *kce);
 
+int kcore_copy(const char *from_dir, const char *to_dir);
+int compare_proc_modules(const char *from, const char *to);
+
 #endif /* __PERF_SYMBOL */

[PATCH V7 6/7] perf tools: add ability to find kcore in build-id cache

[Adrian Hunter]

When no vmlinux is found, tools will use kallsyms and,
if possible, kcore.  Add the ability to find kcore in
the build-id cache.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/util/symbol.c | 147 +++++++++++++++++++++++++++++++++--------------
 1 file changed, 103 insertions(+), 44 deletions(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index ef2680e..b66c1ee 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1442,6 +1442,105 @@ out:
 	return err;
 }
 
+static int find_matching_kcore(struct map *map, char *dir, size_t dir_sz)
+{
+	char kallsyms_filename[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(dir);
+	if (!d)
+		return -1;
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(kallsyms_filename, sizeof(kallsyms_filename),
+			  "%s/%s/kallsyms", dir, dent->d_name);
+		if (!validate_kcore_modules(kallsyms_filename, map)) {
+			strlcpy(dir, kallsyms_filename, dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static char *dso__find_kallsyms(struct dso *dso, struct map *map)
+{
+	u8 host_build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	bool is_host = false;
+	char path[PATH_MAX];
+
+	if (!dso->has_build_id) {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		goto proc_kallsyms;
+	}
+
+	if (sysfs__read_build_id("/sys/kernel/notes", host_build_id,
+				 sizeof(host_build_id)) == 0)
+		is_host = dso__build_id_equal(dso, host_build_id);
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+
+	/* Use /proc/kallsyms if possible */
+	if (is_host) {
+		DIR *d;
+		int fd;
+
+		/* If no cached kcore go with /proc/kallsyms */
+		scnprintf(path, sizeof(path), "%s/[kernel.kcore]/%s",
+			  buildid_dir, sbuild_id);
+		d = opendir(path);
+		if (!d)
+			goto proc_kallsyms;
+		closedir(d);
+
+		/*
+		 * Do not check the build-id cache, until we know we cannot use
+		 * /proc/kcore.
+		 */
+		fd = open("/proc/kcore", O_RDONLY);
+		if (fd != -1) {
+			close(fd);
+			/* If module maps match go with /proc/kallsyms */
+			if (!validate_kcore_modules("/proc/kallsyms", map))
+				goto proc_kallsyms;
+		}
+
+		/* Find kallsyms in build-id cache with kcore */
+		if (!find_matching_kcore(map, path, sizeof(path)))
+			return strdup(path);
+
+		goto proc_kallsyms;
+	}
+
+	scnprintf(path, sizeof(path), "%s/[kernel.kallsyms]/%s",
+		  buildid_dir, sbuild_id);
+
+	if (access(path, F_OK)) {
+		pr_err("No kallsyms or vmlinux with build-id %s was found\n",
+		       sbuild_id);
+		return NULL;
+	}
+
+	return strdup(path);
+
+proc_kallsyms:
+	return strdup("/proc/kallsyms");
+}
+
 static int dso__load_kernel_sym(struct dso *dso, struct map *map,
 				symbol_filter_t filter)
 {
@@ -1490,51 +1589,11 @@ static int dso__load_kernel_sym(struct dso *dso, struct map *map,
 	if (symbol_conf.symfs[0] != 0)
 		return -1;
 
-	/*
-	 * Say the kernel DSO was created when processing the build-id header table,
-	 * we have a build-id, so check if it is the same as the running kernel,
-	 * using it if it is.
-	 */
-	if (dso->has_build_id) {
-		u8 kallsyms_build_id[BUILD_ID_SIZE];
-		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
-
-		if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
-					 sizeof(kallsyms_build_id)) == 0) {
-			if (dso__build_id_equal(dso, kallsyms_build_id)) {
-				kallsyms_filename = "/proc/kallsyms";
-				goto do_kallsyms;
-			}
-		}
-		/*
-		 * Now look if we have it on the build-id cache in
-		 * $HOME/.debug/[kernel.kallsyms].
-		 */
-		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
-				  sbuild_id);
-
-		if (asprintf(&kallsyms_allocated_filename,
-			     "%s/.debug/[kernel.kallsyms]/%s",
-			     getenv("HOME"), sbuild_id) == -1) {
-			pr_err("Not enough memory for kallsyms file lookup\n");
-			return -1;
-		}
-
-		kallsyms_filename = kallsyms_allocated_filename;
+	kallsyms_allocated_filename = dso__find_kallsyms(dso, map);
+	if (!kallsyms_allocated_filename)
+		return -1;
 
-		if (access(kallsyms_filename, F_OK)) {
-			pr_err("No kallsyms or vmlinux with build-id %s "
-			       "was found\n", sbuild_id);
-			free(kallsyms_allocated_filename);
-			return -1;
-		}
-	} else {
-		/*
-		 * Last resort, if we don't have a build-id and couldn't find
-		 * any vmlinux file, try the running kernel kallsyms table.
-		 */
-		kallsyms_filename = "/proc/kallsyms";
-	}
+	kallsyms_filename = kallsyms_allocated_filename;
 
 do_kallsyms:
 	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);
-- 
1.7.11.7

[tip:perf/core] perf symbols: Add ability to find kcore in build-id cache

[tip-bot for Adrian Hunter]

Commit-ID:  0544d4225c52ca31ab2a55b22ddce1392d8f45a4
Gitweb:     http://git.kernel.org/tip/0544d4225c52ca31ab2a55b22ddce1392d8f45a4
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 13:43:43 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 10:29:25 -0300

perf symbols: Add ability to find kcore in build-id cache

When no vmlinux is found, tools will use kallsyms and, if possible,
kcore.  Add the ability to find kcore in the build-id cache.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-7-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/symbol.c | 147 +++++++++++++++++++++++++++++++++--------------
 1 file changed, 103 insertions(+), 44 deletions(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index b2f60dd..76a9e93 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1401,6 +1401,105 @@ out:
 	return err;
 }
 
+static int find_matching_kcore(struct map *map, char *dir, size_t dir_sz)
+{
+	char kallsyms_filename[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(dir);
+	if (!d)
+		return -1;
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(kallsyms_filename, sizeof(kallsyms_filename),
+			  "%s/%s/kallsyms", dir, dent->d_name);
+		if (!validate_kcore_modules(kallsyms_filename, map)) {
+			strlcpy(dir, kallsyms_filename, dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static char *dso__find_kallsyms(struct dso *dso, struct map *map)
+{
+	u8 host_build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	bool is_host = false;
+	char path[PATH_MAX];
+
+	if (!dso->has_build_id) {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		goto proc_kallsyms;
+	}
+
+	if (sysfs__read_build_id("/sys/kernel/notes", host_build_id,
+				 sizeof(host_build_id)) == 0)
+		is_host = dso__build_id_equal(dso, host_build_id);
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+
+	/* Use /proc/kallsyms if possible */
+	if (is_host) {
+		DIR *d;
+		int fd;
+
+		/* If no cached kcore go with /proc/kallsyms */
+		scnprintf(path, sizeof(path), "%s/[kernel.kcore]/%s",
+			  buildid_dir, sbuild_id);
+		d = opendir(path);
+		if (!d)
+			goto proc_kallsyms;
+		closedir(d);
+
+		/*
+		 * Do not check the build-id cache, until we know we cannot use
+		 * /proc/kcore.
+		 */
+		fd = open("/proc/kcore", O_RDONLY);
+		if (fd != -1) {
+			close(fd);
+			/* If module maps match go with /proc/kallsyms */
+			if (!validate_kcore_modules("/proc/kallsyms", map))
+				goto proc_kallsyms;
+		}
+
+		/* Find kallsyms in build-id cache with kcore */
+		if (!find_matching_kcore(map, path, sizeof(path)))
+			return strdup(path);
+
+		goto proc_kallsyms;
+	}
+
+	scnprintf(path, sizeof(path), "%s/[kernel.kallsyms]/%s",
+		  buildid_dir, sbuild_id);
+
+	if (access(path, F_OK)) {
+		pr_err("No kallsyms or vmlinux with build-id %s was found\n",
+		       sbuild_id);
+		return NULL;
+	}
+
+	return strdup(path);
+
+proc_kallsyms:
+	return strdup("/proc/kallsyms");
+}
+
 static int dso__load_kernel_sym(struct dso *dso, struct map *map,
 				symbol_filter_t filter)
 {
@@ -1449,51 +1548,11 @@ static int dso__load_kernel_sym(struct dso *dso, struct map *map,
 	if (symbol_conf.symfs[0] != 0)
 		return -1;
 
-	/*
-	 * Say the kernel DSO was created when processing the build-id header table,
-	 * we have a build-id, so check if it is the same as the running kernel,
-	 * using it if it is.
-	 */
-	if (dso->has_build_id) {
-		u8 kallsyms_build_id[BUILD_ID_SIZE];
-		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
-
-		if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
-					 sizeof(kallsyms_build_id)) == 0) {
-			if (dso__build_id_equal(dso, kallsyms_build_id)) {
-				kallsyms_filename = "/proc/kallsyms";
-				goto do_kallsyms;
-			}
-		}
-		/*
-		 * Now look if we have it on the build-id cache in
-		 * $HOME/.debug/[kernel.kallsyms].
-		 */
-		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
-				  sbuild_id);
-
-		if (asprintf(&kallsyms_allocated_filename,
-			     "%s/.debug/[kernel.kallsyms]/%s",
-			     getenv("HOME"), sbuild_id) == -1) {
-			pr_err("Not enough memory for kallsyms file lookup\n");
-			return -1;
-		}
-
-		kallsyms_filename = kallsyms_allocated_filename;
+	kallsyms_allocated_filename = dso__find_kallsyms(dso, map);
+	if (!kallsyms_allocated_filename)
+		return -1;
 
-		if (access(kallsyms_filename, F_OK)) {
-			pr_err("No kallsyms or vmlinux with build-id %s "
-			       "was found\n", sbuild_id);
-			free(kallsyms_allocated_filename);
-			return -1;
-		}
-	} else {
-		/*
-		 * Last resort, if we don't have a build-id and couldn't find
-		 * any vmlinux file, try the running kernel kallsyms table.
-		 */
-		kallsyms_filename = "/proc/kallsyms";
-	}
+	kallsyms_filename = kallsyms_allocated_filename;
 
 do_kallsyms:
 	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);

[PATCH V7 7/7] perf tools: another fix for annotate_browser__callq()

[Adrian Hunter]

The target address is provided by objdump and is not
necessary a memory address.  Add a helper to get the
correct address.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/ui/browsers/annotate.c |  7 +++++--
 tools/perf/util/map.c             | 31 ++++++++++++++++++++++++++++++-
 tools/perf/util/map.h             |  3 +++
 3 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index 57d3a86..f0697a3 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -445,14 +445,17 @@ static bool annotate_browser__callq(struct annotate_browser *browser,
 	struct annotation *notes;
 	struct addr_map_symbol target = {
 		.map = ms->map,
-		.addr = dl->ops.target.addr,
+		.addr = map__objdump_2mem(ms->map, dl->ops.target.addr),
 	};
 	char title[SYM_TITLE_MAX_SIZE];
 
 	if (!ins__is_call(dl->ins))
 		return false;
 
-	if (map_groups__find_ams(&target, NULL)) {
+	if (map_groups__find_ams(&target, NULL) ||
+	    map__rip_2objdump(target.map, target.map->map_ip(target.map,
+							     target.addr)) !=
+	    dl->ops.target.addr) {
 		ui_helpline__puts("The called function was not found.");
 		return true;
 	}
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 9dea404..ef5bc91 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -252,10 +252,16 @@ size_t map__fprintf_dsoname(struct map *map, FILE *fp)
 	return fprintf(fp, "%s", dsoname);
 }
 
-/*
+/**
+ * map__rip_2objdump - convert symbol start address to objdump address.
+ * @map: memory map
+ * @rip: symbol start address
+ *
  * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
  * map->dso->adjust_symbols==1 for ET_EXEC-like cases except ET_REL which is
  * relative to section start.
+ *
+ * Return: Address suitable for passing to "objdump --start-address="
  */
 u64 map__rip_2objdump(struct map *map, u64 rip)
 {
@@ -268,6 +274,29 @@ u64 map__rip_2objdump(struct map *map, u64 rip)
 	return map->unmap_ip(map, rip);
 }
 
+/**
+ * map__objdump_2mem - convert objdump address to a memory address.
+ * @map: memory map
+ * @ip: objdump address
+ *
+ * Closely related to map__rip_2objdump(), this function takes an address from
+ * objdump and converts it to a memory address.  Note this assumes that @map
+ * contains the address.  To be sure the result is valid, check it forwards
+ * e.g. map__rip_2objdump(map->map_ip(map, map__objdump_2mem(map, ip))) == ip
+ *
+ * Return: Memory address.
+ */
+u64 map__objdump_2mem(struct map *map, u64 ip)
+{
+	if (!map->dso->adjust_symbols)
+		return map->unmap_ip(map, ip);
+
+	if (map->dso->rel)
+		return map->unmap_ip(map, ip + map->pgoff);
+
+	return ip;
+}
+
 void map_groups__init(struct map_groups *mg)
 {
 	int i;
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index 0359b4a..e4e259c 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -84,6 +84,9 @@ static inline u64 identity__map_ip(struct map *map __maybe_unused, u64 ip)
 /* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
 u64 map__rip_2objdump(struct map *map, u64 rip);
 
+/* objdump address -> memory address */
+u64 map__objdump_2mem(struct map *map, u64 ip);
+
 struct symbol;
 
 typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
-- 
1.7.11.7

[tip:perf/core] perf annotate: Another fix for annotate_browser__callq()

[tip-bot for Adrian Hunter]

Commit-ID:  1d5077bdd9a10c4297cded139989bb9ee2998a6c
Gitweb:     http://git.kernel.org/tip/1d5077bdd9a10c4297cded139989bb9ee2998a6c
Author:     Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Mon, 14 Oct 2013 13:43:44 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Oct 2013 12:21:18 -0300

perf annotate: Another fix for annotate_browser__callq()

The target address is provided by objdump and is not necessary a memory
address.  Add a helper to get the correct address.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1381747424-3557-8-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/ui/browsers/annotate.c |  7 +++++--
 tools/perf/util/map.c             | 31 ++++++++++++++++++++++++++++++-
 tools/perf/util/map.h             |  3 +++
 3 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index 57d3a86..f0697a3 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -445,14 +445,17 @@ static bool annotate_browser__callq(struct annotate_browser *browser,
 	struct annotation *notes;
 	struct addr_map_symbol target = {
 		.map = ms->map,
-		.addr = dl->ops.target.addr,
+		.addr = map__objdump_2mem(ms->map, dl->ops.target.addr),
 	};
 	char title[SYM_TITLE_MAX_SIZE];
 
 	if (!ins__is_call(dl->ins))
 		return false;
 
-	if (map_groups__find_ams(&target, NULL)) {
+	if (map_groups__find_ams(&target, NULL) ||
+	    map__rip_2objdump(target.map, target.map->map_ip(target.map,
+							     target.addr)) !=
+	    dl->ops.target.addr) {
 		ui_helpline__puts("The called function was not found.");
 		return true;
 	}
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 9dea404..ef5bc91 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -252,10 +252,16 @@ size_t map__fprintf_dsoname(struct map *map, FILE *fp)
 	return fprintf(fp, "%s", dsoname);
 }
 
-/*
+/**
+ * map__rip_2objdump - convert symbol start address to objdump address.
+ * @map: memory map
+ * @rip: symbol start address
+ *
  * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
  * map->dso->adjust_symbols==1 for ET_EXEC-like cases except ET_REL which is
  * relative to section start.
+ *
+ * Return: Address suitable for passing to "objdump --start-address="
  */
 u64 map__rip_2objdump(struct map *map, u64 rip)
 {
@@ -268,6 +274,29 @@ u64 map__rip_2objdump(struct map *map, u64 rip)
 	return map->unmap_ip(map, rip);
 }
 
+/**
+ * map__objdump_2mem - convert objdump address to a memory address.
+ * @map: memory map
+ * @ip: objdump address
+ *
+ * Closely related to map__rip_2objdump(), this function takes an address from
+ * objdump and converts it to a memory address.  Note this assumes that @map
+ * contains the address.  To be sure the result is valid, check it forwards
+ * e.g. map__rip_2objdump(map->map_ip(map, map__objdump_2mem(map, ip))) == ip
+ *
+ * Return: Memory address.
+ */
+u64 map__objdump_2mem(struct map *map, u64 ip)
+{
+	if (!map->dso->adjust_symbols)
+		return map->unmap_ip(map, ip);
+
+	if (map->dso->rel)
+		return map->unmap_ip(map, ip + map->pgoff);
+
+	return ip;
+}
+
 void map_groups__init(struct map_groups *mg)
 {
 	int i;
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index 0359b4a..e4e259c 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -84,6 +84,9 @@ static inline u64 identity__map_ip(struct map *map __maybe_unused, u64 ip)
 /* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
 u64 map__rip_2objdump(struct map *map, u64 rip);
 
+/* objdump address -> memory address */
+u64 map__objdump_2mem(struct map *map, u64 ip);
+
 struct symbol;
 
 typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);