From: Kevin Wolf <kwolf@redhat.com>
To: qemu-block@nongnu.org
Cc: kwolf@redhat.com, peter.maydell@linaro.org, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 4/7] file-posix: specify expected filetypes
Date: Thu, 12 Jul 2018 18:31:49 +0200 [thread overview]
Message-ID: <20180712163152.12521-5-kwolf@redhat.com> (raw)
In-Reply-To: <20180712163152.12521-1-kwolf@redhat.com>
From: John Snow <jsnow@redhat.com>
Adjust each caller of raw_open_common to specify if they are expecting
host and character devices or not. Tighten expectations of file types upon
open in the common code and refuse types that are not expected.
This has two effects:
(1) Character and block devices are now considered deprecated for the
'file' driver, which expects only S_IFREG, and
(2) no file-posix driver (file, host_cdrom, or host_device) can open
directories now.
I don't think there's a legitimate reason to open directories as if
they were files. This prevents QEMU from opening and attempting to probe
a directory inode, which can break in exciting ways. One of those ways
is lseek on ext4/xfs, which will return 0x7fffffffffffffff as the file
size instead of EISDIR. This can coax QEMU into responding with a
confusing "file too big" instead of "Hey, that's not a file".
See: https://bugs.launchpad.net/qemu/+bug/1739304/
Signed-off-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/file-posix.c | 39 +++++++++++++++++++++++++++++++--------
qemu-doc.texi | 6 ++++++
2 files changed, 37 insertions(+), 8 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 28824aae65..60af4b3d51 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -438,7 +438,8 @@ static QemuOptsList raw_runtime_opts = {
};
static int raw_open_common(BlockDriverState *bs, QDict *options,
- int bdrv_flags, int open_flags, Error **errp)
+ int bdrv_flags, int open_flags,
+ bool device, Error **errp)
{
BDRVRawState *s = bs->opaque;
QemuOpts *opts;
@@ -585,10 +586,32 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
error_setg_errno(errp, errno, "Could not stat file");
goto fail;
}
- if (S_ISREG(st.st_mode)) {
- s->discard_zeroes = true;
- s->has_fallocate = true;
+
+ if (!device) {
+ if (S_ISBLK(st.st_mode)) {
+ warn_report("Opening a block device as a file using the '%s' "
+ "driver is deprecated", bs->drv->format_name);
+ } else if (S_ISCHR(st.st_mode)) {
+ warn_report("Opening a character device as a file using the '%s' "
+ "driver is deprecated", bs->drv->format_name);
+ } else if (!S_ISREG(st.st_mode)) {
+ error_setg(errp, "A regular file was expected by the '%s' driver, "
+ "but something else was given", bs->drv->format_name);
+ ret = -EINVAL;
+ goto fail;
+ } else {
+ s->discard_zeroes = true;
+ s->has_fallocate = true;
+ }
+ } else {
+ if (!(S_ISCHR(st.st_mode) || S_ISBLK(st.st_mode))) {
+ error_setg(errp, "'%s' driver expects either "
+ "a character or block device", bs->drv->format_name);
+ ret = -EINVAL;
+ goto fail;
+ }
}
+
if (S_ISBLK(st.st_mode)) {
#ifdef BLKDISCARDZEROES
unsigned int arg;
@@ -641,7 +664,7 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
BDRVRawState *s = bs->opaque;
s->type = FTYPE_FILE;
- return raw_open_common(bs, options, flags, 0, errp);
+ return raw_open_common(bs, options, flags, 0, false, errp);
}
typedef enum {
@@ -2939,7 +2962,7 @@ hdev_open_Mac_error:
s->type = FTYPE_FILE;
- ret = raw_open_common(bs, options, flags, 0, &local_err);
+ ret = raw_open_common(bs, options, flags, 0, true, &local_err);
if (ret < 0) {
error_propagate(errp, local_err);
#if defined(__APPLE__) && defined(__MACH__)
@@ -3170,7 +3193,7 @@ static int cdrom_open(BlockDriverState *bs, QDict *options, int flags,
s->type = FTYPE_CD;
/* open will not fail even if no CD is inserted, so add O_NONBLOCK */
- return raw_open_common(bs, options, flags, O_NONBLOCK, errp);
+ return raw_open_common(bs, options, flags, O_NONBLOCK, true, errp);
}
static int cdrom_probe_device(const char *filename)
@@ -3284,7 +3307,7 @@ static int cdrom_open(BlockDriverState *bs, QDict *options, int flags,
s->type = FTYPE_CD;
- ret = raw_open_common(bs, options, flags, 0, &local_err);
+ ret = raw_open_common(bs, options, flags, 0, true, &local_err);
if (ret) {
error_propagate(errp, local_err);
return ret;
diff --git a/qemu-doc.texi b/qemu-doc.texi
index 0fbf8b108b..1047c407e7 100644
--- a/qemu-doc.texi
+++ b/qemu-doc.texi
@@ -2969,6 +2969,12 @@ replacement since it is not needed anymore.
The @option{-enable-hax} option has been replaced by @option{-accel hax}.
Both options have been introduced in QEMU version 2.9.0.
+@subsection -drive file=json:@{...@{'driver':'file'@}@} (since 3.0)
+
+The 'file' driver for drives is no longer appropriate for character or host
+devices and will only accept regular files (S_IFREG). The correct driver
+for these file types is 'host_cdrom' or 'host_device' as appropriate.
+
@section QEMU Machine Protocol (QMP) commands
@subsection block-dirty-bitmap-add "autoload" parameter (since 2.12.0)
--
2.13.6
next prev parent reply other threads:[~2018-07-12 16:32 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-12 16:31 [Qemu-devel] [PULL 0/7] Block layer fixes for 3.0-rc1 Kevin Wolf
2018-07-12 16:31 ` [Qemu-devel] [PULL 1/7] iotests: 153: Fix dead code Kevin Wolf
2018-07-12 16:31 ` [Qemu-devel] [PULL 2/7] iotests: nbd: Stop qemu-nbd before remaking image Kevin Wolf
2018-07-12 16:31 ` [Qemu-devel] [PULL 3/7] qemu-img: Document copy offloading implications with -S and -c Kevin Wolf
2018-07-12 16:31 ` Kevin Wolf [this message]
2018-07-12 16:31 ` [Qemu-devel] [PULL 5/7] iotests: add test 226 for file driver types Kevin Wolf
2018-07-12 16:31 ` [Qemu-devel] [PULL 6/7] scsi-disk: Block Device Characteristics emulation fix Kevin Wolf
2018-07-12 16:31 ` [Qemu-devel] [PULL 7/7] qemu-img: align result of is_allocated_sectors Kevin Wolf
2018-07-12 17:24 ` [Qemu-devel] [PULL 0/7] Block layer fixes for 3.0-rc1 Peter Maydell
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=20180712163152.12521-5-kwolf@redhat.com \
--to=kwolf@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).