From: "Darrick J. Wong" <djwong@kernel.org>
To: liuh <liuhuan01@kylinos.cn>
Cc: aalbersh@redhat.com, linux-xfs@vger.kernel.org
Subject: Re: [PATCH v5] mkfs: simplify setup_proto file status checks
Date: Tue, 7 Jul 2026 22:12:32 -0700 [thread overview]
Message-ID: <20260708051232.GE9392@frogsfrogsfrogs> (raw)
In-Reply-To: <20260708032301.11458-2-liuhuan01@kylinos.cn>
On Wed, Jul 08, 2026 at 11:23:02AM +0800, liuh wrote:
> setup_proto() calls filesize() before validating the source type with
> fstat(), even though both operations ultimately require the same file
> status information.
>
> Perform a single fstat() immediately after opening the source path and
> reuse the resulting metadata for both source type validation and file
> size retrieval. This simplifies the setup logic and eliminates a
> redundant metadata lookup.
>
> Signed-off-by: liuh <liuhuan01@kylinos.cn>
> ---
> mkfs/proto.c | 23 ++++++++++++++++++++---
> 1 file changed, 20 insertions(+), 3 deletions(-)
>
> diff --git a/mkfs/proto.c b/mkfs/proto.c
> index a460aebd..ff867afe 100644
> --- a/mkfs/proto.c
> +++ b/mkfs/proto.c
> @@ -82,14 +82,17 @@ setup_proto(
> return result;
> }
>
> - if ((fd = open(fname, O_RDONLY)) < 0 || (size = filesize(fd)) < 0) {
> + if ((fd = open(fname, O_RDONLY)) < 0) {
> fprintf(stderr, _("%s: failed to open %s: %s\n"),
> progname, fname, strerror(errno));
> goto out_fail;
> }
>
> - if (fstat(fd, &statbuf) < 0)
> - fail(_("invalid or unreadable source path"), errno);
> + if (fstat(fd, &statbuf) < 0) {
> + fprintf(stderr, _("%s: failed to fstat %s: %s\n"),
> + progname, fname, strerror(errno));
> + goto out_fail;
> + }
>
> /*
> * Handle directory inputs.
> @@ -101,6 +104,20 @@ setup_proto(
> return result;
> }
>
> + if (!S_ISREG(statbuf.st_mode)) {
> + fprintf(stderr, _("%s: %s not a regular file\n"),
> + progname, fname);
> + goto out_fail;
> + }
> +
> + if (statbuf.st_size < 0) {
> + fprintf(stderr, _("%s: %s invalid file size\n"),
> + progname, fname);
> + goto out_fail;
> + }
> +
> + size = statbuf.st_size;
Hmm. What type is @size, a long? I imagine you need to fail if
statbuf.st_size >= LONG_MAX too, right? Not that I expect there are
very many 32-bit systems (such that sizeof(statbuf.st_size) >
sizeof(long)) but we might end up there again with 128-bit off_t some
day.
--D
> +
> /*
> * Else this is a protofile, let's handle traditionally.
> */
> --
> 2.43.0
>
>
prev parent reply other threads:[~2026-07-08 5:12 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-08 3:23 [PATCH v5] mkfs: simplify setup_proto file status checks liuh
2026-07-08 5:12 ` Darrick J. Wong [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=20260708051232.GE9392@frogsfrogsfrogs \
--to=djwong@kernel.org \
--cc=aalbersh@redhat.com \
--cc=linux-xfs@vger.kernel.org \
--cc=liuhuan01@kylinos.cn \
/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