Message-ID: <10783.1279556132@redhat.com>
Date:	Mon, 19 Jul 2010 17:15:32 +0100
From:	David Howells <dhowells@...hat.com>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	dhowells@...hat.com, viro@...iv.linux.org.uk,
	linux-fsdevel@...r.kernel.org, linux-nfs@...r.kernel.org,
	linux-cifs@...r.kernel.org, linux-kernel@...r.kernel.org,
	samba-technical@...ts.samba.org, linux-ext4@...r.kernel.org
Subject: Re: [PATCH 02/18] xstat: Add a pair of system calls to make extended file stats available [ver #6]

Linus Torvalds <torvalds@...ux-foundation.org> wrote:

>  - that whole xstat buffer handling is just a mess. I think you
> already fixed the "xstat_parameters" crud and just made it a simple
> unsigned long and a direct argument,

I was thinking more of an unsigned int argument, since it can't have more than
32 flags in it if it is also to work on 32-bit arches.

> but the "buffer+buflen" thing is still disgusting.
>
>    Why not just leave a few empty fields at the end, and make the rule
> be: "We don't just add random crap, so don't expect it to grow widely
> in the future".

Because it gets allocated on the kernel stack.  It's already 160 bytes, and
expanding it will eat more kernel stack space.  Now, I can offset that by: (a)
embedding it in struct kstat so that we allocate less stack space in xstat()
overall, and (b) allocating kstat/xstat structs with kmalloc() rather than on
the stack in all the stat syscalls.

>  - you use "long long" all over the place. Don't do that. If you want
> a fixed size, say so, and use "u64/s64". That's the _real_ fixed size,
> and "long long" just _happens_ to be the same size on all current
> architectures.

I was following struct stat/stat64 in arch/x86/include/asm/stat.h which do the
same.  Also, if this is going to be seen by userspace, isn't it better to use
uint32_t and suchlike?

>  - why create that new kind of xstat() that realistically absolutely
> nobody will use outside of some very special cases, and that has no
> real advantages for 99.9% of all people?

The new information is useful for some cases.  Samba for example.  At least
two of the fields I'm adding are also made available through BSD's stat()
call, and will automatically be used for some things by autoconf magic if they
become available.

I'm still trying to get a handle on what people think will be truly useful.  I
can see things *could* be useful, particularly to GUI file managers and ls,
but not everyone is of the same opinion.

Perhaps you or others can offer answers to the following questions as these
might help:

 (1) Should I offer information that's effectively free to come by, but could
     be got through:

     (a) An extra statfs() call - such as whether a file is remote, whether
     	 it's some kernel special file?  Or what the volume label is for this
     	 file?

     (b) An extra getxattr() call - such as a file's security label.

     (c) An extra ioctl() call - such as FS_IOC_GETFLAGS.

 (2) Should I offer information that's appropriate to non-UNIX filesystems
     such as FAT, NTFS or CIFS.  Some of this may map onto other fields, such
     as FS_IOC_GETFLAGS.

 (3) Should I offer information about which results that I've returned are
     actually useful, as opposed to being fabricated on the spot?  Such as
     UID/GID in FAT or blocks in UBIFS.  This may be of use to df or a GUI.
     For instance, a GUI, seeing that UID/GID aren't useful, could ask the
     filesystem to provide information about what it considers to be valid
     ownership information.

>    You could make it a "atomic stat+open" by replacing the useless
> "size" return value with a "fd" return value, add a flag saying "we're
> also interested in opening it" (in the same result set flags), and
> instead of that stupid "buflen" input, give the "mode" input that open
> needs.

Which would be used by even fewer people, I suspect.  However, it's certainly
an interesting idea.  I suspect it doesn't gain much over open()+fstat()
though, given that you'd still have to do most of the work of fstat() after
doing the open() thing anyway.

Also, I'm not sure how much use the atomicity is, given that the file may have
changed state between the gathering of the stat data and userspace getting to
do anything with it.

> >        ssize_t ret = fxstat(unsigned fd,
> 
> Quite frankly, my gut feel is that once you do "xstat(dfd, filename,
> ...)" then it's damn stupid to do a separate "fxstat()", when you
> might as well say that "xtstat(dfd, NULL, ...)" is the same as
> "fxstat(fd, ...)"

This has been suggested and denounced as stupid already.  That said, I agree
with you.

> Now, the difference between adding one or two system calls may not be
> huge, but just from a cleanliness angle, I really don't see the point
> of having another fstat variant when the extended xstat() already very
> naturally supports the thing. And let's face it, using a NULL path
> pointer just makes sense if you don't have a path. You already passed
> it a target file descriptor in the dfd.

Agreed.

David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists