Linux Btrfs filesystem development
 help / color / mirror / Atom feed
From: Florian Margaine <florian.margaine@commerceguys.com>
To: Karel Zak <kzak@redhat.com>
Cc: linux-btrfs@vger.kernel.org
Subject: Re: [PATCH] btrfs-progs: Fix partitioned loop devices resolve.
Date: Thu, 12 Nov 2015 10:10:51 +0100	[thread overview]
Message-ID: <5644579B.3000102@commerceguys.com> (raw)
In-Reply-To: <20151110115018.GY19508@ws.net.home>


[-- Attachment #1.1: Type: text/plain, Size: 568 bytes --]

New patch is attached.

On 11/10/2015 12:50 PM, Karel Zak wrote:
> 
> Yep, first try /sys/... and when unsuccessful then try ioctl.
> 
> losetup example:
> https://github.com/karelzak/util-linux/blob/master/lib/loopdev.c#L686
> 
> (it's probably too complex, but the basic idea is obvious)
> 
> Maybe we need libloop.so to share all these things between various
> project :-)

That would be lovely indeed.

> 
>     Karel
> 

Regards,

-- 
Florian Margaine

Product Engineer @ Platform.sh
https://platform.sh

https://keybase.io/fmargaine

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1.2: fix-partitioned-loop-devices-resolve.patch --]
[-- Type: text/x-patch; name="fix-partitioned-loop-devices-resolve.patch", Size: 2001 bytes --]

When using partitions on a loop device, the device's name can be
/dev/loop0p1 or similar, and no relevant entry exists in the /sys
filesystem, so the current resolve_loop_device function fails.

This patch adds a fallback to using loopdev API which will get the
correct backing file even for partitioned loop devices, at the expense
of using more memory, hence using it as a fallback only.
---
 utils.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/utils.c b/utils.c
index d546bea..5e4813d 100644
--- a/utils.c
+++ b/utils.c
@@ -1170,2 +1170,2 @@ static int is_loop_device (const char* device) {
 		MAJOR(statbuf.st_rdev) == LOOP_MAJOR);
 }

+/*
+ * Takes a loop device path (e.g. /dev/loop0) and returns
+ * the associated file (e.g. /images/my_btrfs.img) using
+ * loopdev API
+ */
+static int resolve_loop_device_with_loopdev(const char* loop_dev, char* loop_file)
+{
+	int fd;
+	struct loop_info64 lo64;
+
+	if (!(fd = open(loop_dev, O_RDONLY | O_NONBLOCK)))
+		return -errno;
+	if (ioctl(fd, LOOP_GET_STATUS64, &lo64) != 0)
+		return -errno;
+
+	memcpy(loop_file, lo64.lo_file_name, strlen(lo64.lo_file_name) + 1);
+	if (close(fd) != 0)
+		return -errno;
+
+	return 0;
+}

 /* Takes a loop device path (e.g. /dev/loop0) and returns
  * the associated file (e.g. /images/my_btrfs.img) */
@@ -1185,5 +1206,10 @@ static int resolve_loop_device(const char* loop_dev, char* loop_file,
 	if (!realpath(loop_dev, real_loop_dev))
 		return -errno;
 	snprintf(p, PATH_MAX, "/sys/block/%s/loop/backing_file", strrchr(real_loop_dev, '/'));
-	if (!(f = fopen(p, "r")))
+	if (!(f = fopen(p, "r"))) {
+		if (errno == ENOENT)
+			/* It's possibly a partitioned loop device, which
+			   is resolvable with loopdev API. */
+			return resolve_loop_device_with_loopdev(loop_dev, loop_file);
 		return -errno;
+	}

 	snprintf(fmt, 20, "%%%i[^\n]", max_len-1);
 	ret = fscanf(f, fmt, loop_file);
--
2.6.1

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

  reply	other threads:[~2015-11-12  9:10 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-09 13:06 [PATCH] btrfs-progs: Fix partitioned loop devices resolve Florian Margaine
2015-11-09 14:12 ` Karel Zak
2015-11-09 14:14   ` Florian Margaine
2015-11-10  8:35   ` Florian Margaine
2015-11-10 11:50     ` Karel Zak
2015-11-12  9:10       ` Florian Margaine [this message]
2015-11-13 17:16         ` David Sterba

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5644579B.3000102@commerceguys.com \
    --to=florian.margaine@commerceguys.com \
    --cc=kzak@redhat.com \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox