Skip to contentSkip to conversation

tools lib api fs: Move cgroupsfs_find_mountpoint()

PATCH Applied Namhyung Kim 4 participants 8 messages Jan 27, 2020 – Mar 19, 2020 perf events

lore
Share & tools
lore

Namhyung Kim

author maintainer
#0 lore raw

PATCH 1/1 tools lib api fs: Move cgroupsfs_find_mountpoint()

Move it from tools/perf/util/cgroup.c as it can be used by other places. Note that cgroup filesystem is different from others since it's usually mounted separately (in v1) for each subsystem.

I just copied the code with a little modification to pass a name of subsystem.

---
 tools/lib/api/fs/Build    |  1 +
 tools/lib/api/fs/cgroup.c | 67 +++++++++++++++++++++++++++++++++++++++
 tools/lib/api/fs/fs.h     |  2 ++
 tools/perf/util/cgroup.c  | 63 ++----------------------------------
 4 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 tools/lib/api/fs/cgroup.c
tools/lib/api/fs/Build +1 -0
index f4ed9629ae85..0f75b28654de 100644
@@ -1,2 +1,3 @@
 libapi-y += fs.o
 libapi-y += tracing_path.o
+libapi-y += cgroup.o
tools/lib/api/fs/cgroup.c new file +67 -0

67 lines not on this page · raw message

tools/lib/api/fs/fs.h +2 -0
index 92d03b8396b1..936edb95e1f3 100644
@@ -28,6 +28,8 @@ FS(bpf_fs)
 #undef FS
 
 
+int cgroupfs_find_mountpoint(char *buf, size_t maxlen, const char *subsys);
+
 int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_xll(const char *filename, unsigned long long *value);
tools/perf/util/cgroup.c +2 -61

84 lines not on this page · raw message

Signature · 1 line
2.25.0.341.g760bfbb309-goog
#1 lore raw

On Mon, Jan 27, 2020 at 07:00:31PM +0900, Namhyung Kim wrote:

SNIP

Quoted 12 lines
tools/perf/util/cgroup.c +0 -8
-
-	if (strlen(path) < maxlen) {
-		strcpy(buf, path);
-		return 0;
-	}
-	return -1;
-}
-
 static int open_cgroup(const char *name)
 {
 	char path[PATH_MAX + 1];

tools/perf/util/cgroup.c static int open_cgroup(const char *name) · line 23

Quoted 5 lines
tools/perf/util/cgroup.c +1 -1
@@ -79,7 +20,7 @@ static int open_cgroup(const char *name)
 	int fd;
 
 
-	if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+	if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1, "perf_event"))

nice, but could you please follow fs API names and change the name to cgroupfs__mountpoint

I think we don't need to define the rest of the functions now, if they are not used

	#define FS(name)                                \
		const char *name##__mountpoint(void);   \
		const char *name##__mount(void);        \
		bool name##__configured(void);          \

just follow the function name

thanks,
jirka

Quoted 6 lines
tools/perf/util/cgroup.c +0 -0
 		return -1;
 
 	scnprintf(path, PATH_MAX, "%s/%s", mnt, name);
Signature · 1 line
2.25.0.341.g760bfbb309-goog
author maintainer
#2 lore raw

Hi Jiri,

Quoted 29 lines · Jiri Olsa

On Mon, Jan 27, 2020 at 07:00:31PM +0900, Namhyung Kim wrote:

SNIP

Quoted 12 lines
tools/perf/util/cgroup.c +0 -8
-
-     if (strlen(path) < maxlen) {
-             strcpy(buf, path);
-             return 0;
-     }
-     return -1;
-}
-
 static int open_cgroup(const char *name)
 {
      char path[PATH_MAX + 1];

tools/perf/util/cgroup.c static int open_cgroup(const char *name) · line 23

Quoted 5 lines
tools/perf/util/cgroup.c +1 -1
@@ -79,7 +20,7 @@ static int open_cgroup(const char *name)
      int fd;
 
 
-     if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+     if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1, "perf_event"))

nice, but could you please follow fs API names and change the name to cgroupfs__mountpoint

I think we don't need to define the rest of the functions now, if they are not used

Quoted 6 lines
        #define FS(name)                                \
                const char *name##__mountpoint(void);   \
                const char *name##__mount(void);        \
                bool name##__configured(void);          \

just follow the function name

I thought about it too, but still it has different arguments. So I left it as is... Do you really want to rename?

Thanks
Namhyung

#3 lore raw
Quoted 37 lines · Namhyung Kim

Hi Jiri,

Quoted 34 lines · Jiri Olsa

On Mon, Jan 27, 2020 at 07:00:31PM +0900, Namhyung Kim wrote:

SNIP

Quoted 12 lines
tools/perf/util/cgroup.c +0 -8
-
-     if (strlen(path) < maxlen) {
-             strcpy(buf, path);
-             return 0;
-     }
-     return -1;
-}
-
 static int open_cgroup(const char *name)
 {
      char path[PATH_MAX + 1];

tools/perf/util/cgroup.c static int open_cgroup(const char *name) · line 23

Quoted 5 lines
tools/perf/util/cgroup.c +1 -1
@@ -79,7 +20,7 @@ static int open_cgroup(const char *name)
      int fd;
 
 
-     if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+     if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1, "perf_event"))

nice, but could you please follow fs API names and change the name to cgroupfs__mountpoint

I think we don't need to define the rest of the functions now, if they are not used

        #define FS(name)                                \
                const char *name##__mountpoint(void);   \
                const char *name##__mount(void);        \
                bool name##__configured(void);          \
Quoted 4 lines
Quoted 1 line

just follow the function name

I thought about it too, but still it has different arguments. So I left it as is... Do you really want to rename?

yea I think we should keep the same name, Arnaldo?

jirka

Quoted 4 lines

Thanks
Namhyung

author maintainer
#4 lore raw

PATCH v2 1/1 tools lib api fs: Move cgroupsfs__mountpoint()

Move it from tools/perf/util/cgroup.c as it can be used by other places. Note that cgroup filesystem is different from others since it's usually mounted separately (in v1) for each subsystem.

I just copied the code with a little modification to pass a name of subsystem and renamed it to follow other APIs.

---
* rename to cgroupfs__mountpoint()

 tools/lib/api/fs/Build    |  1 +
 tools/lib/api/fs/cgroup.c | 67 +++++++++++++++++++++++++++++++++++++++
 tools/lib/api/fs/fs.h     |  2 ++
 tools/perf/util/cgroup.c  | 63 ++----------------------------------
 4 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 tools/lib/api/fs/cgroup.c
tools/lib/api/fs/Build +1 -0
index f4ed9629ae85..0f75b28654de 100644
@@ -1,2 +1,3 @@
 libapi-y += fs.o
 libapi-y += tracing_path.o
+libapi-y += cgroup.o
tools/lib/api/fs/cgroup.c new file +67 -0

67 lines not on this page · raw message

tools/lib/api/fs/fs.h +2 -0
index 92d03b8396b1..07591ecbe39f 100644
@@ -28,6 +28,8 @@ FS(bpf_fs)
 #undef FS
 
 
+int cgroupfs__mountpoint(char *buf, size_t maxlen, const char *subsys);
+
 int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_xll(const char *filename, unsigned long long *value);
tools/perf/util/cgroup.c +2 -61

84 lines not on this page · raw message

Signature · 1 line
2.25.0.341.g760bfbb309-goog
Acked-by on v2 1/1
#5 lore raw

Re: [PATCH v2] tools lib api fs: Move cgroupsfs__mountpoint()

Quoted 9 lines · Namhyung Kim

Move it from tools/perf/util/cgroup.c as it can be used by other places. Note that cgroup filesystem is different from others since it's usually mounted separately (in v1) for each subsystem.

I just copied the code with a little modification to pass a name of subsystem and renamed it to follow other APIs.

thanks,
jirka

Quoted 197 lines

---
* rename to cgroupfs__mountpoint()

 tools/lib/api/fs/Build    |  1 +
 tools/lib/api/fs/cgroup.c | 67 +++++++++++++++++++++++++++++++++++++++
 tools/lib/api/fs/fs.h     |  2 ++
 tools/perf/util/cgroup.c  | 63 ++----------------------------------
 4 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 tools/lib/api/fs/cgroup.c
tools/lib/api/fs/Build +1 -0
index f4ed9629ae85..0f75b28654de 100644
@@ -1,2 +1,3 @@
 libapi-y += fs.o
 libapi-y += tracing_path.o
+libapi-y += cgroup.o
tools/lib/api/fs/cgroup.c new file +67 -0
new file mode 100644
index 000000000000..c7e1cdaa36e1
@@ -0,0 +1,67 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/stringify.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "fs.h"
+
+int cgroupfs__mountpoint(char *buf, size_t maxlen, const char *subsys)
+{
+	FILE *fp;
+	char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
+	char path_v1[PATH_MAX + 1], path_v2[PATH_MAX + 2], *path;
+	char *token, *saved_ptr = NULL;
+
+	fp = fopen("/proc/mounts", "r");
+	if (!fp)
+		return -1;
+
+	/*
+	 * in order to handle split hierarchy, we need to scan /proc/mounts
+	 * and inspect every cgroupfs mount point to find one that has
+	 * perf_event subsystem
+	 */
+	path_v1[0] = '\0';
+	path_v2[0] = '\0';
+
+	while (fscanf(fp, "%*s %"__stringify(PATH_MAX)"s %"__stringify(PATH_MAX)"s %"
+				__stringify(PATH_MAX)"s %*d %*d\n",
+				mountpoint, type, tokens) == 3) {
+
+		if (!path_v1[0] && !strcmp(type, "cgroup")) {
+
+			token = strtok_r(tokens, ",", &saved_ptr);
+
+			while (token != NULL) {
+				if (subsys && !strcmp(token, subsys)) {
+					strcpy(path_v1, mountpoint);
+					break;
+				}
+				token = strtok_r(NULL, ",", &saved_ptr);
+			}
+		}
+
+		if (!path_v2[0] && !strcmp(type, "cgroup2"))
+			strcpy(path_v2, mountpoint);
+
+		if (path_v1[0] && path_v2[0])
+			break;
+	}
+	fclose(fp);
+
+	if (path_v1[0])
+		path = path_v1;
+	else if (path_v2[0])
+		path = path_v2;
+	else
+		return -1;
+
+	if (strlen(path) < maxlen) {
+		strcpy(buf, path);
+		return 0;
+	}
+	return -1;
+}
tools/lib/api/fs/fs.h +2 -0
index 92d03b8396b1..07591ecbe39f 100644
@@ -28,6 +28,8 @@ FS(bpf_fs)
 #undef FS
 
 
+int cgroupfs__mountpoint(char *buf, size_t maxlen, const char *subsys);
+
 int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_xll(const char *filename, unsigned long long *value);
tools/perf/util/cgroup.c +2 -61
index 4881d4af3381..12e466d1ec3b 100644
@@ -3,75 +3,16 @@
 #include "evsel.h"
 #include "cgroup.h"
 #include "evlist.h"
-#include <linux/stringify.h>
 #include <linux/zalloc.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <fcntl.h>
 #include <stdlib.h>
 #include <string.h>
+#include <api/fs/fs.h>
 
 int nr_cgroups;
 
-static int
-cgroupfs_find_mountpoint(char *buf, size_t maxlen)
-{
-	FILE *fp;
-	char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
-	char path_v1[PATH_MAX + 1], path_v2[PATH_MAX + 2], *path;
-	char *token, *saved_ptr = NULL;
-
-	fp = fopen("/proc/mounts", "r");
-	if (!fp)
-		return -1;
-
-	/*
-	 * in order to handle split hierarchy, we need to scan /proc/mounts
-	 * and inspect every cgroupfs mount point to find one that has
-	 * perf_event subsystem
-	 */
-	path_v1[0] = '\0';
-	path_v2[0] = '\0';
-
-	while (fscanf(fp, "%*s %"__stringify(PATH_MAX)"s %"__stringify(PATH_MAX)"s %"
-				__stringify(PATH_MAX)"s %*d %*d\n",
-				mountpoint, type, tokens) == 3) {
-
-		if (!path_v1[0] && !strcmp(type, "cgroup")) {
-
-			token = strtok_r(tokens, ",", &saved_ptr);
-
-			while (token != NULL) {
-				if (!strcmp(token, "perf_event")) {
-					strcpy(path_v1, mountpoint);
-					break;
-				}
-				token = strtok_r(NULL, ",", &saved_ptr);
-			}
-		}
-
-		if (!path_v2[0] && !strcmp(type, "cgroup2"))
-			strcpy(path_v2, mountpoint);
-
-		if (path_v1[0] && path_v2[0])
-			break;
-	}
-	fclose(fp);
-
-	if (path_v1[0])
-		path = path_v1;
-	else if (path_v2[0])
-		path = path_v2;
-	else
-		return -1;
-
-	if (strlen(path) < maxlen) {
-		strcpy(buf, path);
-		return 0;
-	}
-	return -1;
-}
-
 static int open_cgroup(const char *name)
 {
 	char path[PATH_MAX + 1];
@@ -79,7 +20,7 @@ static int open_cgroup(const char *name)
 	int fd;
 
 
-	if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+	if (cgroupfs__mountpoint(mnt, PATH_MAX + 1, "perf_event"))
 		return -1;
 
 	scnprintf(path, PATH_MAX, "%s/%s", mnt, name);
Signature · 1 line
2.25.0.341.g760bfbb309-goog
#6 lore raw

Re: [PATCH v2] tools lib api fs: Move cgroupsfs__mountpoint()

Em Tue, Jan 28, 2020 at 10:40:57AM +0100, Jiri Olsa escreveu:

Quoted 5 lines · Namhyung Kim
Quoted 4 lines

Move it from tools/perf/util/cgroup.c as it can be used by other places. Note that cgroup filesystem is different from others since it's usually mounted separately (in v1) for each subsystem.

Quoted 7 lines
Quoted 5 lines

I just copied the code with a little modification to pass a name of subsystem and renamed it to follow other APIs.

Thanks, applied to perf/core.

- Arnaldo

Quoted 201 lines

thanks,
jirka

Quoted 197 lines

---
* rename to cgroupfs__mountpoint()

 tools/lib/api/fs/Build    |  1 +
 tools/lib/api/fs/cgroup.c | 67 +++++++++++++++++++++++++++++++++++++++
 tools/lib/api/fs/fs.h     |  2 ++
 tools/perf/util/cgroup.c  | 63 ++----------------------------------
 4 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 tools/lib/api/fs/cgroup.c
tools/lib/api/fs/Build +1 -0
index f4ed9629ae85..0f75b28654de 100644
@@ -1,2 +1,3 @@
 libapi-y += fs.o
 libapi-y += tracing_path.o
+libapi-y += cgroup.o
tools/lib/api/fs/cgroup.c new file +67 -0
new file mode 100644
index 000000000000..c7e1cdaa36e1
@@ -0,0 +1,67 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/stringify.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "fs.h"
+
+int cgroupfs__mountpoint(char *buf, size_t maxlen, const char *subsys)
+{
+	FILE *fp;
+	char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
+	char path_v1[PATH_MAX + 1], path_v2[PATH_MAX + 2], *path;
+	char *token, *saved_ptr = NULL;
+
+	fp = fopen("/proc/mounts", "r");
+	if (!fp)
+		return -1;
+
+	/*
+	 * in order to handle split hierarchy, we need to scan /proc/mounts
+	 * and inspect every cgroupfs mount point to find one that has
+	 * perf_event subsystem
+	 */
+	path_v1[0] = '\0';
+	path_v2[0] = '\0';
+
+	while (fscanf(fp, "%*s %"__stringify(PATH_MAX)"s %"__stringify(PATH_MAX)"s %"
+				__stringify(PATH_MAX)"s %*d %*d\n",
+				mountpoint, type, tokens) == 3) {
+
+		if (!path_v1[0] && !strcmp(type, "cgroup")) {
+
+			token = strtok_r(tokens, ",", &saved_ptr);
+
+			while (token != NULL) {
+				if (subsys && !strcmp(token, subsys)) {
+					strcpy(path_v1, mountpoint);
+					break;
+				}
+				token = strtok_r(NULL, ",", &saved_ptr);
+			}
+		}
+
+		if (!path_v2[0] && !strcmp(type, "cgroup2"))
+			strcpy(path_v2, mountpoint);
+
+		if (path_v1[0] && path_v2[0])
+			break;
+	}
+	fclose(fp);
+
+	if (path_v1[0])
+		path = path_v1;
+	else if (path_v2[0])
+		path = path_v2;
+	else
+		return -1;
+
+	if (strlen(path) < maxlen) {
+		strcpy(buf, path);
+		return 0;
+	}
+	return -1;
+}
tools/lib/api/fs/fs.h +2 -0
index 92d03b8396b1..07591ecbe39f 100644
@@ -28,6 +28,8 @@ FS(bpf_fs)
 #undef FS
 
 
+int cgroupfs__mountpoint(char *buf, size_t maxlen, const char *subsys);
+
 int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_xll(const char *filename, unsigned long long *value);
tools/perf/util/cgroup.c +2 -61
index 4881d4af3381..12e466d1ec3b 100644
@@ -3,75 +3,16 @@
 #include "evsel.h"
 #include "cgroup.h"
 #include "evlist.h"
-#include <linux/stringify.h>
 #include <linux/zalloc.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <fcntl.h>
 #include <stdlib.h>
 #include <string.h>
+#include <api/fs/fs.h>
 
 int nr_cgroups;
 
-static int
-cgroupfs_find_mountpoint(char *buf, size_t maxlen)
-{
-	FILE *fp;
-	char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
-	char path_v1[PATH_MAX + 1], path_v2[PATH_MAX + 2], *path;
-	char *token, *saved_ptr = NULL;
-
-	fp = fopen("/proc/mounts", "r");
-	if (!fp)
-		return -1;
-
-	/*
-	 * in order to handle split hierarchy, we need to scan /proc/mounts
-	 * and inspect every cgroupfs mount point to find one that has
-	 * perf_event subsystem
-	 */
-	path_v1[0] = '\0';
-	path_v2[0] = '\0';
-
-	while (fscanf(fp, "%*s %"__stringify(PATH_MAX)"s %"__stringify(PATH_MAX)"s %"
-				__stringify(PATH_MAX)"s %*d %*d\n",
-				mountpoint, type, tokens) == 3) {
-
-		if (!path_v1[0] && !strcmp(type, "cgroup")) {
-
-			token = strtok_r(tokens, ",", &saved_ptr);
-
-			while (token != NULL) {
-				if (!strcmp(token, "perf_event")) {
-					strcpy(path_v1, mountpoint);
-					break;
-				}
-				token = strtok_r(NULL, ",", &saved_ptr);
-			}
-		}
-
-		if (!path_v2[0] && !strcmp(type, "cgroup2"))
-			strcpy(path_v2, mountpoint);
-
-		if (path_v1[0] && path_v2[0])
-			break;
-	}
-	fclose(fp);
-
-	if (path_v1[0])
-		path = path_v1;
-	else if (path_v2[0])
-		path = path_v2;
-	else
-		return -1;
-
-	if (strlen(path) < maxlen) {
-		strcpy(buf, path);
-		return 0;
-	}
-	return -1;
-}
-
 static int open_cgroup(const char *name)
 {
 	char path[PATH_MAX + 1];
@@ -79,7 +20,7 @@ static int open_cgroup(const char *name)
 	int fd;
 
 
-	if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+	if (cgroupfs__mountpoint(mnt, PATH_MAX + 1, "perf_event"))
 		return -1;
 
 	scnprintf(path, PATH_MAX, "%s/%s", mnt, name);
Signature · 1 line
2.25.0.341.g760bfbb309-goog
Signature · 1 line
- Arnaldo
merge bot Applied
#7 lore raw
In reply to Namhyung Kim
The following commit has been merged into the perf/core branch of tip:

The following commit has been merged into the perf/core branch of tip:

Commit-ID:     7982a898515064ba180d958bfc89ed22d13905ee
Gitweb:        https://git.kernel.org/tip/7982a898515064ba180d958bfc89ed22d13905ee
Author:        Namhyung Kim <[email protected]>
AuthorDate:    Mon, 27 Jan 2020 19:00:31 +09:00
Committer:     Arnaldo Carvalho de Melo <[email protected]>
CommitterDate: Wed, 04 Mar 2020 10:34:09 -03:00

tools lib api fs: Move cgroupsfs_find_mountpoint()

Move it from tools/perf/util/cgroup.c as it can be used by other places. Note that cgroup filesystem is different from others since it's usually mounted separately (in v1) for each subsystem.

I just copied the code with a little modification to pass a name of subsystem.

---
 tools/lib/api/fs/Build    |  1 +-
 tools/lib/api/fs/cgroup.c | 67 ++++++++++++++++++++++++++++++++++++++-
 tools/lib/api/fs/fs.h     |  2 +-
 tools/perf/util/cgroup.c  | 63 +-----------------------------------
 4 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 tools/lib/api/fs/cgroup.c
tools/lib/api/fs/Build +1 -0
index f4ed962..0f75b28 100644
@@ -1,2 +1,3 @@
 libapi-y += fs.o
 libapi-y += tracing_path.o
+libapi-y += cgroup.o
tools/lib/api/fs/cgroup.c new file +67 -0

67 lines not on this page · raw message

tools/lib/api/fs/fs.h +2 -0
index 92d03b8..936edb9 100644
@@ -28,6 +28,8 @@ FS(bpf_fs)
 #undef FS
 
 
+int cgroupfs_find_mountpoint(char *buf, size_t maxlen, const char *subsys);
+
 int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_xll(const char *filename, unsigned long long *value);
tools/perf/util/cgroup.c +2 -61

84 lines not on this page · raw message