From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vladislav Bolkhovitin Subject: Re: [Scst-devel] Discussion about SCST sysfs layout and implementation. Date: Fri, 17 Apr 2009 21:43:14 +0400 Message-ID: <49E8BFB2.50608@vlnb.net> References: <49E7307A.8000205@vlnb.net> <49E77795.7080204@linux.vnet.ibm.com> <49E85F1C.7040106@vlnb.net> <49E8832F.6070302@linux.vnet.ibm.com> <49E88E3F.8030107@vlnb.net> <49E891C8.1050007@emulex.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from moutng.kundenserver.de ([212.227.126.177]:60376 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758709AbZDQRnD (ORCPT ); Fri, 17 Apr 2009 13:43:03 -0400 In-Reply-To: <49E891C8.1050007@emulex.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: James Smart Cc: Daniel Debonzi , "scst-devel@lists.sourceforge.net" , "linux-kernel@vger.kernel.org" , "linux-scsi@vger.kernel.org" James Smart, on 04/17/2009 06:27 PM wrote: > Vladislav Bolkhovitin wrote: >> Let's go with root in /sys/class/scsi_tgt. In future, if somebody >> objects, we can easily change it. > > This maps too closely to the scsi midlayer names, especially the real > target devices initiators see - which are not yet in sysfs but there has > been discussion of in the past. I would prefer that you used > /sys/class/scst_tgt. And yes, as part of the class level support in the > midlayer/kernel, they are generally symlinks to the actual kobject in > the /sys/devices that the class entity corresponds to. > > IMHO - "scsi_tgt" also seems a bit presumptuous to propose if SCST is > not, in it's entirety, going to be the adapter target interface pulled > in upstream. Thus, another reason I prefer the "scst_tgt" name, as it > makes it very specific to SCST itself, and lessens any co-existence > conflicts if there are any in the future. James, Thank you for the suggestion. If nobody objects, we will go with /sys/class/scst_tgt. Vlad