linux-kernel - Re: [PATCH] Fix proc_file

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, 31 Aug 2009 10:33:34 +0400
From:	Alexey Dobriyan <adobriyan@...il.com>
To:	Stefani Seibold <stefani@...bold.net>
Cc:	Christoph Hellwig <hch@...radead.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] Fix proc_file_write missing ppos update

On Sun, Aug 30, 2009 at 09:05:44PM +0200, Stefani Seibold wrote:
> Am Samstag, den 29.08.2009, 19:16 -0400 schrieb Christoph Hellwig:
> > On Sat, Aug 29, 2009 at 06:38:12PM +0200, Stefani Seibold wrote:
> > > The following fix a long standing issue in the proc_file_write function,
> > > which doesn't update the ppos file position pointer.
> > 
> > The right fix is to get rid of the last remaining
> > read_proc_t/write_proc_t instances and switch everyone to implement file
> > operations.  Alexey has been working on this on and off for a while.
> > 
> 
> Switching all users of read_proc_t/write_proc_t to file operation is a
> huge job. About 180 files must be fixed.

Less, staging stuff is on it's own.

> But the main reason not to do this is because the breakage of "out of
> tree" drivers.

Just how many times it needs to be repeated.

Out-of-tree argument means almost nothing.

> I like the current simplified proc interface. It saves a lot of code
> duplication because the basic operations will be handled inside the
> kernel and not in the driver.

It's garbage.

Arbirtrary 3 KB limit on read,
kernel memory giving trailer in ->read_proc which is impossible to get right,
was always buggy wrt module unload (quickly fixed, but it was always meant
to be temporary).

Now that SCSI pile is mostly finished, what a horrible code interface provoked.

> There is no reason for Alexey to finish his work, submit it and maybe it
> will be accepted. 

You'd better start converting to struct file_operations::write now.

> In the meantime i will see my patch as the preferred solution: a cleanup
> of the current interface.
--
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/