bug-gnulib
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] Use xattr (Linux) in qcopy-acl.c


From: Paul Eggert
Subject: Re: [PATCH] Use xattr (Linux) in qcopy-acl.c
Date: Wed, 4 Jan 2023 17:19:21 -0800
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.4.2

On 2023-01-04 12:54, Ondrej Valousek wrote:

LIB_HAS_ACL has value only if glibc does not know getxattr() system call. It 
has nothing to do with libattr (yes, it's bit confusing).

No kidding. This stuff is waaaayy too complicated.

Why don't we solve the linker problem with the "--as-needed" option?

--as-needed isn't portable, but Gnulib has a lib-ignore module that should be more portable. I suppose we could make the qcopy-acl module depend on the lib-ignore module, but we've never done anything like that before. Another possibility would be to implement a new variable LIB_XATTR_FALLBACK (or perhaps a better name), that acts like LIB_ACL if the xattr library is absent and the acl library is present, and is empty otherwise.

Perhaps Bruno has an opinion here. I'm not sure how well lib-ignore would work for packages that build libraries rather than apps.

It depends on a kernel ACL implementation. On Linux the ACLs make the 
permissions only more opened (hence my code would be fine).
In contrast, on Solaris/ZFS, the opposite could happen.

Ouch, in that case perhaps we should not use the new code on Solaris. Solaris is a dead end now anyway, no point trying to make it go faster.

What about OpenZFS on GNU/Linux? That's more important.

NetApp NFSv4 server even allows you to break RFC7530 the way that chmod 0 
<file> will still allow the inherited ACLs to be applied.

We should be OK there, right? We do the chmod 0 first.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]