linux-kernel - Re: [PATCH v2] vfs: Shortcut AT_EMPTY_PATH early for statx, and add AT_NO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Mon, 24 Jun 2024 20:19:33 +0800
From: Xi Ruoyao <xry111@...111.site>
To: Mateusz Guzik <mjguzik@...il.com>, Linus Torvalds
	 <torvalds@...ux-foundation.org>
Cc: Christian Brauner <brauner@...nel.org>, Alexander Viro
	 <viro@...iv.linux.org.uk>, Jan Kara <jack@...e.cz>, Alejandro Colomar
	 <alx@...nel.org>, Arnd Bergmann <arnd@...db.de>, Huacai Chen
	 <chenhuacai@...ngson.cn>, Xuerui Wang <kernel@...0n.name>, Jiaxun Yang
	 <jiaxun.yang@...goat.com>, Icenowy Zheng <uwu@...nowy.me>, 
	linux-fsdevel@...r.kernel.org, linux-arch@...r.kernel.org, 
	loongarch@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] vfs: Shortcut AT_EMPTY_PATH early for statx, and add
 AT_NO_PATH for statx and fstatat

On Mon, 2024-06-24 at 11:04 +0200, Mateusz Guzik wrote:
> Below is a diff which compiles but is untested. It adds AT_EMPTY_PATH +
> NULL as suggsted by Linus, but it can be adjusted for AT_NO_PATH (which
> would be my preffered option, or better yet not do that and add fstatx).
> 
> It does not do the hack to 0-check if a pointer was passed along with
> AT_EMPTY_PATH but that again is an easy addition.
> 
> Feel free to take without attribution:

I'd still like to make it Co-developed-by: or just From: you.  Could you
give a S-o-b?

And with this change AT_FDCWD with AT_EMPTY_PATH and NULL path does not
work.  For consistency it'd be better to make it work too:

diff --git a/fs/stat.c b/fs/stat.c
index b0a4db7b90df..d04f7ba46645 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -752,10 +752,14 @@ SYSCALL_DEFINE5(statx,
 	int ret;
 	struct filename *name;
 
-	if (filename == NULL && (flags & AT_EMPTY_PATH))
-		return do_statx_fd(dfd, flags, mask, buffer);
+	if (filename == NULL && (flags & AT_EMPTY_PATH)) {
+		if (dfd >= 0)
+			return do_statx_fd(dfd, flags, mask, buffer);
+		else
+			name = getname_kernel("");
+	} else
+		name = getname_flags(filename, getname_statx_lookup_flags(flags));
 
-	name = getname_flags(filename, getname_statx_lookup_flags(flags));
 	ret = do_statx(dfd, name, flags, mask, buffer);
 	putname(name);
 
And should we do it for fstatat() as well?

Let's wait for comment from Linus anyway.

-- 
Xi Ruoyao <xry111@...111.site>
School of Aerospace Science and Technology, Xidian University