xfs
[Top] [All Lists]

Re: extended attributes interface

To: "David S. Miller" <davem@xxxxxxxxxx>
Subject: Re: extended attributes interface
From: Keith Owens <kaos@xxxxxxx>
Date: Thu, 31 Jan 2002 10:10:55 +1100
Cc: a.gruenbacher@xxxxxxxxxxxx, linux-xfs@xxxxxxxxxxx
In-reply-to: Your message of "Wed, 30 Jan 2002 14:46:47 -0800." <20020130.144647.21928212.davem@xxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Wed, 30 Jan 2002 14:46:47 -0800 (PST), 
"David S. Miller" <davem@xxxxxxxxxx> wrote:
>All the values that go through these syscalls seem to be
>opaque and filesystem specific.  Therefore can I ask the filesystems
>that use these things use fixed sized types such as "u32" "u16" et al.
>instead of things such as "long" or "int"?

Absolutely agree.

>The reason I ask is, unless strict sized types are used it is going
>to be a real pain in the ass to translate the types passed to these
>system calls in mixed 32-bit/64-bit environments.  This is thus going
>to be a mess on sparc64, ppc64, mips64, ia64, and probably others I
>have forgotten :-)

Hopefully not for ia64.  The ia32 compatibility mode syscall table is a
subset of the i386 table, in particular new syscalls and Linux specific
ones tend not to be in arch/ia64/ia32.  ia32 mode on ia64 is not a long
term compatibility aim, it is intended for running existing binaries
that cannot be recompiled for native ia64 mode.  I strongly recommend
that the ACL syscalls are NOT added to arch/ia64/ia32.  We have the
source, compile the ACL programs in native ia64 mode.

Sparc64 and mips64 are a problem as long as the kernel is 64 bit and
userspace is 32 bit.  I don't know if the kernel/userspace size
mismatch applies to ppc64 as well or if it is more like ia64.


<Prev in Thread] Current Thread [Next in Thread>