* [PATCH v3] Btrfs-progs: check out if the swap device
@ 2013-02-14 7:53 Tsutomu Itoh
2013-02-14 12:22 ` David Sterba
2013-03-03 16:47 ` Brendan Hide
0 siblings, 2 replies; 4+ messages in thread
From: Tsutomu Itoh @ 2013-02-14 7:53 UTC (permalink / raw)
To: linux-btrfs
Currently, the following commands succeed.
# cat /proc/swaps
Filename Type Size Used Priority
/dev/sda3 partition 8388604 0 -1
/dev/sdc8 partition 9765884 0 -2
# mkfs.btrfs /dev/sdc8
WARNING! - Btrfs v0.20-rc1-165-g82ac345 IS EXPERIMENTAL
WARNING! - see http://btrfs.wiki.kernel.org before using
fs created label (null) on /dev/sdc8
nodesize 4096 leafsize 4096 sectorsize 4096 size 9.31GB
Btrfs v0.20-rc1-165-g82ac345
# btrfs fi sh /dev/sdc8
Label: none uuid: fc0bdbd0-7eed-460f-b4e9-131273b66df2
Total devices 1 FS bytes used 28.00KB
devid 1 size 9.31GB used 989.62MB path /dev/sdc8
Btrfs v0.20-rc1-165-g82ac345
#
But we should check out the swap device. Fixed it.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
---
(this patch is based on Chris's raid56-experimental branch)
v2 -> v3:
- Open again without O_EXCL so that the problem should not occur
(btrfs_register_one_device() fails if O_EXCL is on)
v1 -> v2:
- O_EXCL is specified at open(), as a last safety check
- Check when the swap device is a file is added
- Translate '\nnn'(octal) to character
Thank you Eric, David, Zach and others.
---
mkfs.c | 52 +++++++++++++++++++++++++++++++++++++++++++--
utils.c | 75 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
utils.h | 1 +
3 files changed, 126 insertions(+), 2 deletions(-)
diff --git a/mkfs.c b/mkfs.c
index 2d3c2af..2210312 100644
--- a/mkfs.c
+++ b/mkfs.c
@@ -1366,6 +1366,16 @@ int main(int ac, char **av)
if (source_dir == 0) {
file = av[optind++];
+ ret = is_swap_device(file);
+ if (ret < 0) {
+ fprintf(stderr, "error checking %s status: %s\n", file,
+ strerror(-ret));
+ exit(1);
+ }
+ if (ret == 1) {
+ fprintf(stderr, "%s is a swap device\n", file);
+ exit(1);
+ }
ret = check_mounted(file);
if (ret < 0) {
fprintf(stderr, "error checking %s mount status\n", file);
@@ -1376,9 +1386,23 @@ int main(int ac, char **av)
exit(1);
}
ac--;
+ /* check if the device is busy */
+ fd = open(file, O_RDWR|O_EXCL);
+ if (fd < 0) {
+ fprintf(stderr, "unable to open %s: %s\n", file,
+ strerror(errno));
+ exit(1);
+ }
+ close(fd);
+ /*
+ * open again without O_EXCL so that the problem should not
+ * occur by the following processing.
+ * (btrfs_register_one_device() fails if O_EXCL is on)
+ */
fd = open(file, O_RDWR);
if (fd < 0) {
- fprintf(stderr, "unable to open %s\n", file);
+ fprintf(stderr, "unable to open %s: %s\n", file,
+ strerror(errno));
exit(1);
}
first_file = file;
@@ -1461,6 +1485,16 @@ int main(int ac, char **av)
int old_mixed = mixed;
file = av[optind++];
+ ret = is_swap_device(file);
+ if (ret < 0) {
+ fprintf(stderr, "error checking %s status: %s\n", file,
+ strerror(-ret));
+ exit(1);
+ }
+ if (ret == 1) {
+ fprintf(stderr, "%s is a swap device\n", file);
+ exit(1);
+ }
ret = check_mounted(file);
if (ret < 0) {
fprintf(stderr, "error checking %s mount status\n",
@@ -1471,9 +1505,23 @@ int main(int ac, char **av)
fprintf(stderr, "%s is mounted\n", file);
exit(1);
}
+ /* check if the device is busy */
+ fd = open(file, O_RDWR|O_EXCL);
+ if (fd < 0) {
+ fprintf(stderr, "unable to open %s: %s\n", file,
+ strerror(errno));
+ exit(1);
+ }
+ close(fd);
+ /*
+ * open again without O_EXCL so that the problem should not
+ * occur by the following processing.
+ * (btrfs_register_one_device() fails if O_EXCL is on)
+ */
fd = open(file, O_RDWR);
if (fd < 0) {
- fprintf(stderr, "unable to open %s\n", file);
+ fprintf(stderr, "unable to open %s: %s\n", file,
+ strerror(errno));
exit(1);
}
ret = btrfs_device_already_in_root(root, fd,
diff --git a/utils.c b/utils.c
index f9ee812..891edcc 100644
--- a/utils.c
+++ b/utils.c
@@ -1386,3 +1386,78 @@ int get_fs_info(int fd, char *path, struct btrfs_ioctl_fs_info_args *fi_args,
return 0;
}
+
+#define isoctal(c) (((c) & ~7) == '0')
+
+static inline void translate(char *f, char *t)
+{
+ while (*f != '\0') {
+ if (*f == '\\' &&
+ isoctal(f[1]) && isoctal(f[2]) && isoctal(f[3])) {
+ *t++ = 64*(f[1] & 7) + 8*(f[2] & 7) + (f[3] & 7);
+ f += 4;
+ } else
+ *t++ = *f++;
+ }
+ *t = '\0';
+ return;
+}
+
+/*
+ * Checks if the swap device.
+ * Returns 1 if swap device, < 0 on error or 0 if not swap device.
+ */
+int is_swap_device(const char *file)
+{
+ FILE *f;
+ struct stat st_buf;
+ dev_t dev;
+ ino_t ino = 0;
+ char tmp[PATH_MAX];
+ char buf[PATH_MAX];
+ char *cp;
+ int ret = 0;
+
+ if (stat(file, &st_buf) < 0)
+ return -errno;
+ if (S_ISBLK(st_buf.st_mode))
+ dev = st_buf.st_rdev;
+ else if (S_ISREG(st_buf.st_mode)) {
+ dev = st_buf.st_dev;
+ ino = st_buf.st_ino;
+ } else
+ return 0;
+
+ if ((f = fopen("/proc/swaps", "r")) == NULL)
+ return 0;
+
+ /* skip the first line */
+ if (fgets(tmp, sizeof(tmp), f) == NULL)
+ goto out;
+
+ while (fgets(tmp, sizeof(tmp), f) != NULL) {
+ if ((cp = strchr(tmp, ' ')) != NULL)
+ *cp = '\0';
+ if ((cp = strchr(tmp, '\t')) != NULL)
+ *cp = '\0';
+ translate(tmp, buf);
+ if (stat(buf, &st_buf) != 0)
+ continue;
+ if (S_ISBLK(st_buf.st_mode)) {
+ if (dev == st_buf.st_rdev) {
+ ret = 1;
+ break;
+ }
+ } else if (S_ISREG(st_buf.st_mode)) {
+ if (dev == st_buf.st_dev && ino == st_buf.st_ino) {
+ ret = 1;
+ break;
+ }
+ }
+ }
+
+out:
+ fclose(f);
+
+ return ret;
+}
diff --git a/utils.h b/utils.h
index bbcaf6a..60a0fea 100644
--- a/utils.h
+++ b/utils.h
@@ -55,6 +55,7 @@ int get_fs_info(int fd, char *path, struct btrfs_ioctl_fs_info_args *fi_args,
struct btrfs_ioctl_dev_info_args **di_ret);
char *__strncpy__null(char *dest, const char *src, size_t n);
+int is_swap_device(const char *file);
/* Helper to always get proper size of the destination string */
#define strncpy_null(dest, src) __strncpy__null(dest, src, sizeof(dest))
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v3] Btrfs-progs: check out if the swap device
2013-02-14 7:53 [PATCH v3] Btrfs-progs: check out if the swap device Tsutomu Itoh
@ 2013-02-14 12:22 ` David Sterba
2013-03-03 16:47 ` Brendan Hide
1 sibling, 0 replies; 4+ messages in thread
From: David Sterba @ 2013-02-14 12:22 UTC (permalink / raw)
To: Tsutomu Itoh; +Cc: linux-btrfs
On Thu, Feb 14, 2013 at 04:53:04PM +0900, Tsutomu Itoh wrote:
> Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Worked for me here in these situations (ie. mkfs failed, filenames
specified given as simple path or with mixture of ./ and ../ leading to
the device):
* mounted /dev/sdx
* /dev/sdx part of a multi-device mounted fs
* /dev/sdx swap device
* file as a swap device
* swap files with funny names: \ \040 \134 \888 <space>
Tested-by: David Sterba <dsterba@suse.cz>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3] Btrfs-progs: check out if the swap device
2013-02-14 7:53 [PATCH v3] Btrfs-progs: check out if the swap device Tsutomu Itoh
2013-02-14 12:22 ` David Sterba
@ 2013-03-03 16:47 ` Brendan Hide
2013-03-05 1:18 ` Tsutomu Itoh
1 sibling, 1 reply; 4+ messages in thread
From: Brendan Hide @ 2013-03-03 16:47 UTC (permalink / raw)
To: Tsutomu Itoh; +Cc: linux-btrfs
On 2013/02/14 09:53 AM, Tsutomu Itoh wrote:
> + if (ret < 0) {
> + fprintf(stderr, "error checking %s status: %s\n", file,
> + strerror(-ret));
> + exit(1);
> + }
>
> ...
>
> + /* check if the device is busy */
> + fd = open(file, O_RDWR|O_EXCL);
> + if (fd < 0) {
> + fprintf(stderr, "unable to open %s: %s\n", file,
> + strerror(errno));
> + exit(1);
> + }
This is fine and works (as tested by David) - but I'm not sure if the
below suggestions from Zach were taken into account.
1. If the check with "open(file, O_RDWR|O_EXCL)" shows that the device
is available, there's no point in checking if it is mounted as a swap
device. A preliminary check using this could precede all other checks
which should be skipped if it shows success.
2. If there's an error checking the status (for example lets say
/proc/swaps is deprecated), we should print the informational message
but not error out.
On 2013/02/13 11:58 AM, Zach Brown wrote:
> - First always open with O_EXCL. If it succeeds then there's no reason
> to check /proc/swaps at all. (Maybe it doesn't need to try
> check_mounted() there either? Not sure if it's protecting against
> accidentally mounting mounted shared storage or not.)
>
> ...
>
> - At no point is failure of any of the /proc/swaps parsing fatal. It'd
> carry on ignoring errors until it doesnt have work to do. It'd only
> ever print the nice message when it finds a match.
--
__________
Brendan Hide
http://swiftspirit.co.za/
http://www.webafrica.co.za/?AFF1E97
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3] Btrfs-progs: check out if the swap device
2013-03-03 16:47 ` Brendan Hide
@ 2013-03-05 1:18 ` Tsutomu Itoh
0 siblings, 0 replies; 4+ messages in thread
From: Tsutomu Itoh @ 2013-03-05 1:18 UTC (permalink / raw)
To: Brendan Hide; +Cc: linux-btrfs
On 2013/03/04 1:47, Brendan Hide wrote:
> On 2013/02/14 09:53 AM, Tsutomu Itoh wrote:
>> + if (ret < 0) {
>> + fprintf(stderr, "error checking %s status: %s\n", file,
>> + strerror(-ret));
>> + exit(1);
>> + }
>>
>> ...
>>
>> + /* check if the device is busy */
>> + fd = open(file, O_RDWR|O_EXCL);
>> + if (fd < 0) {
>> + fprintf(stderr, "unable to open %s: %s\n", file,
>> + strerror(errno));
>> + exit(1);
>> + }
> This is fine and works (as tested by David) - but I'm not sure if the below
suggestions from Zach were taken into account.
>
> 1. If the check with "open(file, O_RDWR|O_EXCL)" shows that the device is
> available, there's no point in checking if it is mounted as a swap device.
> A preliminary check using this could precede all other checks which should
> be skipped if it shows success.
>
> 2. If there's an error checking the status (for example lets say /proc/swaps
> is deprecated), we should print the informational message but not error out.
If "open(file, O_RDWR|O_EXCL)" failed, we should output an appropriate message
why it failed. So, I'm testing is_swap_device() and check_mounted() first.
But is_swap_device() is not perfect, so I'm trying to open O_EXCL after all
other tests, as a last safety check.
Thanks,
Tsutomu
>
> On 2013/02/13 11:58 AM, Zach Brown wrote:
>> - First always open with O_EXCL. If it succeeds then there's no reason
>> to check /proc/swaps at all. (Maybe it doesn't need to try
>> check_mounted() there either? Not sure if it's protecting against
>> accidentally mounting mounted shared storage or not.)
>>
>> ...
>>
>> - At no point is failure of any of the /proc/swaps parsing fatal. It'd
>> carry on ignoring errors until it doesnt have work to do. It'd only
>> ever print the nice message when it finds a match.
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2013-03-05 1:19 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-02-14 7:53 [PATCH v3] Btrfs-progs: check out if the swap device Tsutomu Itoh
2013-02-14 12:22 ` David Sterba
2013-03-03 16:47 ` Brendan Hide
2013-03-05 1:18 ` Tsutomu Itoh
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.