From: Gao Xiang <hsiangkao@linux.alibaba.com>
To: Yifan Zhao <yifan.yfzhao@foxmail.com>,
linux-erofs@lists.ozlabs.org,
Robert Rose <robert.rose@mailbox.org>
Subject: Re: [PATCH v2] erofs-utils: mount: fix flag-clearing bug and missing error check in parse_flagopts
Date: Mon, 2 Mar 2026 09:58:12 +0800 [thread overview]
Message-ID: <e6dedb36-d6b9-4eee-ae36-3694971f06f7@linux.alibaba.com> (raw)
In-Reply-To: <tencent_003DF0338EAB42F1573BC0CCFBEACE321E06@qq.com>
On 2026/3/1 20:12, Yifan Zhao wrote:
> The MS_* constants in glibc's <sys/mount.h> are defined as members of
> an anonymous enum whose underlying type is unsigned int (because the
> last member, MS_NOUSER, is initialised with '1U << 31'). Therefore
> ~MS_RDONLY, ~MS_NOSUID, etc. are unsigned int values that, when stored
The problem seems glibc refines these macros as enum.
> into a 'long flags' field, undergo zero-extension, not sign-extension.
> As a result every 'clearing' entry (rw, suid, dev, exec, async, atime,
> diratime, norelatime, loud) produced a positive long, so the
> opts[i].flags < 0 guard in erofsmount_parse_flagopts() was never true
> and the corresponding flags were set rather than cleared.
>
> Fix by casting the operand to long before applying bitwise-NOT,
> ensuring the result is a negative long with the correct bit pattern.
>
> Also add the missing return-value check for erofsmount_parse_flagopts()
> in the '-o' option handler.
>
> Reported-By: rorosen <76747196+rorosen@users.noreply.github.com>
It seems the real email is:
Reported-by: Robert Rose <robert.rose@mailbox.org>
Otherwise it looks good to me, will apply:
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Thanks,
Gao Xiang
> Closes: https://github.com/NixOS/nixpkgs/issues/494653
> Signed-off-By: Yifan Zhao <yifan.yfzhao@foxmail.com>
> ---
> mount/main.c | 27 +++++++++++++++------------
> 1 file changed, 15 insertions(+), 12 deletions(-)
>
> diff --git a/mount/main.c b/mount/main.c
> index b04be5d..7c557bd 100644
> --- a/mount/main.c
> +++ b/mount/main.c
> @@ -203,15 +203,15 @@ static long erofsmount_parse_flagopts(char *s, long flags, char **more)
> } opts[] = {
> {"defaults", 0}, {"quiet", 0}, // NOPs
> {"user", 0}, {"nouser", 0}, // checked in fstab, ignored in -o
> - {"ro", MS_RDONLY}, {"rw", ~MS_RDONLY},
> - {"nosuid", MS_NOSUID}, {"suid", ~MS_NOSUID},
> - {"nodev", MS_NODEV}, {"dev", ~MS_NODEV},
> - {"noexec", MS_NOEXEC}, {"exec", ~MS_NOEXEC},
> - {"sync", MS_SYNCHRONOUS}, {"async", ~MS_SYNCHRONOUS},
> - {"noatime", MS_NOATIME}, {"atime", ~MS_NOATIME},
> - {"norelatime", ~MS_RELATIME}, {"relatime", MS_RELATIME},
> - {"nodiratime", MS_NODIRATIME}, {"diratime", ~MS_NODIRATIME},
> - {"loud", ~MS_SILENT},
> + {"ro", MS_RDONLY}, {"rw", ~(long)MS_RDONLY},
> + {"nosuid", MS_NOSUID}, {"suid", ~(long)MS_NOSUID},
> + {"nodev", MS_NODEV}, {"dev", ~(long)MS_NODEV},
> + {"noexec", MS_NOEXEC}, {"exec", ~(long)MS_NOEXEC},
> + {"sync", MS_SYNCHRONOUS}, {"async", ~(long)MS_SYNCHRONOUS},
> + {"noatime", MS_NOATIME}, {"atime", ~(long)MS_NOATIME},
> + {"norelatime", ~(long)MS_RELATIME}, {"relatime", MS_RELATIME},
> + {"nodiratime", MS_NODIRATIME}, {"diratime", ~(long)MS_NODIRATIME},
> + {"loud", ~(long)MS_SILENT},
> {"remount", MS_REMOUNT}, {"move", MS_MOVE},
> // mand dirsync rec iversion strictatime
> };
> @@ -281,6 +281,7 @@ static int erofsmount_parse_options(int argc, char **argv)
> {0, 0, 0, 0},
> };
> char *dot;
> + long ret;
> int opt;
> int i;
>
> @@ -305,9 +306,11 @@ static int erofsmount_parse_options(int argc, char **argv)
> break;
> case 'o':
> mountcfg.full_options = optarg;
> - mountcfg.flags =
> - erofsmount_parse_flagopts(optarg, mountcfg.flags,
> - &mountcfg.options);
> + ret = erofsmount_parse_flagopts(optarg, mountcfg.flags,
> + &mountcfg.options);
> + if (ret < 0)
> + return (int)ret;
> + mountcfg.flags = ret;
> break;
> case 't':
> dot = strchr(optarg, '.');
prev parent reply other threads:[~2026-03-02 1:58 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-01 12:12 [PATCH v2] erofs-utils: mount: fix flag-clearing bug and missing error check in parse_flagopts Yifan Zhao
2026-03-02 1:58 ` Gao Xiang [this message]
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=e6dedb36-d6b9-4eee-ae36-3694971f06f7@linux.alibaba.com \
--to=hsiangkao@linux.alibaba.com \
--cc=linux-erofs@lists.ozlabs.org \
--cc=robert.rose@mailbox.org \
--cc=yifan.yfzhao@foxmail.com \
/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