[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200426155958.GS23230@ZenIV.linux.org.uk>
Date: Sun, 26 Apr 2020 16:59:58 +0100
From: Al Viro <viro@...iv.linux.org.uk>
To: Christoph Hellwig <hch@....de>
Cc: Kees Cook <keescook@...omium.org>,
Iurii Zaikin <yzaikin@...gle.com>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
linux-kernel@...r.kernel.org, linux-mm@...ck.org,
linux-fsdevel@...r.kernel.org, netdev@...r.kernel.org,
bpf@...r.kernel.org
Subject: Re: pass kernel pointers to the sysctl ->proc_handler method v3
On Fri, Apr 24, 2020 at 08:43:33AM +0200, Christoph Hellwig wrote:
> Hi all,
>
> this series changes the sysctl ->proc_handler methods to take kernel
> pointers. This simplifies some of the pointer handling in the methods
> (which could probably be further simplified now), and gets rid of the
> set_fs address space overrides used by bpf.
>
> Changes since v2:
> - free the buffer modified by BPF
> - move pid_max and friends to pid.h
>
> Changes since v1:
> - drop a patch merged by Greg
> - don't copy data out on a write
> - fix buffer allocation in bpf
OK, I can live with that; further work can live on top of that, anyway.
How are we going to handle that? I can put it into never-rebased branch
in vfs.git (#work.sysctl), so that people could pull that.
FWIW, I'm putting together more uaccess stuff (will probably hit -next
tonight or tomorrow); this would fit well there...
Powered by blists - more mailing lists