From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Fedin Subject: Re: Fw: [PATCH] Full NLS support for HFS (classic) filesystem Date: Tue, 30 Aug 2005 10:13:23 +0400 Message-ID: <4313F903.6010101@rambler.ru> References: <20050829134533.541465d2.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=KOI8-R; format=flowed Content-Transfer-Encoding: 7bit Cc: Roman Zippel , linux-fsdevel@vger.kernel.org Return-path: Received: from mxb.rambler.ru ([81.19.66.30]:55314 "EHLO mxb.rambler.ru") by vger.kernel.org with ESMTP id S1750938AbVH3GMg (ORCPT ); Tue, 30 Aug 2005 02:12:36 -0400 To: Andrew Morton In-Reply-To: <20050829134533.541465d2.akpm@osdl.org> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org Andrew Morton wrote: > So I have this very old patch sitting in my todo folder. Has it moved forward at all? Roman Zippel didn't agree with such an implementation and created another version which is the only proper one as he says (despite it's less functional). Well, i am waiting for it to appear in "stable" kernel in order to re-add CP10007 NLS and re-test it. You can cut out cp10007 NLS from this patch and apply it to the kernel, it is absent in Andrew's version because he is not russian. I am copying this message to Roman for reference. P.S. I still would like to take the following changes into consideration: 1. IMHO all the filesystems in the kernel should ignore unknown arguments rather than raise an error and refuse to mount. This would make autofs much more usable. For example line like: /dev/cdrom /mnt/cdrom hfsplus,hfs user,noauto,iocharset=koi8-r,codepage=10007 would work in this case. Currently it would fail because hfsplus knows nothing about "codepage". There can be more examples where ignoring unknown arguments would be useful, this is just the most common one for me. A side-effect would be fixup of smbfs which currently doesn't recognise "user,noauto". I posted a patch introducing these dummy arguments a long time ago but it was rejected because it was considered to be just a code pollution. Probably a policy of ignoring all unknown arguments would do? 2. There are two arguments doing the same job. Some filesystems understand only "nls" and some - only "iocharset". I think this should be made uniform. Kind regards.