From sacadmin Tue Jul 17 05:23:10 2007 Received: from sac.sfbay.sun.com (localhost [127.0.0.1]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HCNA3B015723; Tue, 17 Jul 2007 05:23:10 -0700 (PDT) Received: (from petede@localhost) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8/Submit) id l6HCN9ob015719; Tue, 17 Jul 2007 05:23:09 -0700 (PDT) Date: Tue, 17 Jul 2007 05:23:09 -0700 (PDT) From: Peter Dennis Message-Id: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> To: PSARC-record@sac.sfbay.sun.com Cc: pcfs-iteam@sun.com Subject: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] Status: RO Content-Length: 2526 Case: PCFS mount options "noatime"/"atime" Author: Frank Hofmann Summary: This proposal requests * to introduce a mount option pair for PCFS, "-o noatime" resp. "-o atime", which suppresses / enables creating access time updates. * to disable access time updates (use the "-o noatime" option by default) for PCFS if it is used on removeable media. The Release Binding is patch and the Interface is Committed. The Problem Flash-based removeable media can only perform a limited amount of write cycles. In addition to that, writes on such media are very slow (compared to reads). Since PCFS is frequently used on such devices ("memory sticks"), attempts should be made to minimize media write access whenever possible. Access timestamps are an easy target for that objective, because: - FAT access timestamps are day-granular - FAT access timestamps are optional as per FAT spec In addition, UFS has had the "noatime" mount option since 1998. PCFS should have it as well. The Proposed Solution PCFS will be changed to provide the "atime/noatime" mount options just as UFS does via PSARC 1998/093 (RFE 4088189). The default behaviour will be different from UFS, though: - removeable media: "noatime" will be default - fixed disks: "atime" will be default The reason for that is the abovementioned behaviour of flash devices, which are mostly irrelevant for UFS but a main usage scenario for PCFS, and the abovementioned behaviour of FAT access timestamps as per the specification. MAN PAGE CHANGES The man page for mount_pcfs(1M) will be changed to describe the new mount options "atime"/"noatime" as per above (6581003). The additional text will be: atime | noatime Enable or disable writing access timestamps on DOS formatted media. Default for fixed disks is atime, while for removeable media noatime is used. This is so that writes to flash-based media ("memory sticks") can be minim- ized, to prolong lifetime. From sacadmin Tue Jul 17 05:49:20 2007 Received: from sfbaymail2sca.sfbay.sun.com (sfbaymail2sca.SFBay.Sun.COM [129.145.155.42]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HCnKBe016013; Tue, 17 Jul 2007 05:49:20 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com (gmpes-gis-mail-2.UK.Sun.COM [129.156.42.6]) by sfbaymail2sca.sfbay.sun.com (8.13.6+Sun/8.12.10/ENSMAIL,v2.2) with ESMTP id l6HClIsT015521; Tue, 17 Jul 2007 05:47:19 -0700 (PDT) Received: from d1-emea-10.sun.com ([192.18.2.120]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6HClD0G010531; Tue, 17 Jul 2007 12:47:13 GMT Received: from conversion-daemon.d1-emea-10.sun.com by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLB00F01Q2S7900@d1-emea-10.sun.com> (original mail from Darren.Moffat@Sun.COM); Tue, 17 Jul 2007 13:47:13 +0100 (BST) Received: from [129.156.173.21] by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLB007AQQ6O4K00@d1-emea-10.sun.com>; Tue, 17 Jul 2007 13:47:12 +0100 (BST) Date: Tue, 17 Jul 2007 13:47:12 +0100 From: Darren J Moffat Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> Sender: Darren.Moffat@Sun.COM To: Peter Dennis Cc: PSARC-record@sac.sfbay.sun.com, pcfs-iteam@Sun.COM Message-id: <469CBA50.7020009@Sun.COM> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> User-Agent: Thunderbird 2.0.0.0 (X11/20070605) Status: RO Content-Length: 505 How does this proposals default of "noatime" compare to other operating systems ? Why not have it so that the atime for a given file is only ever updated a single time in a given day rather than doing a write on every access - given you say that granularity is per day anyway ? For my personal use I'm okay with the default, because I don't actually like that fact that just reading a file causes a write cycle on my CF cards, but it might not be the best default for everyone. -- Darren J Moffat From sacadmin Tue Jul 17 06:40:39 2007 Received: from sfbaymail1sca.SFBay.Sun.COM (sfbaymail1sca.SFBay.Sun.COM [129.145.154.35]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HDedF5016437; Tue, 17 Jul 2007 06:40:39 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com (gmpes-gis-mail-1.UK.Sun.COM [129.156.42.5]) by sfbaymail1sca.SFBay.Sun.COM (8.13.6+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l6HDbicx003417; Tue, 17 Jul 2007 06:37:45 -0700 (PDT) Received: from d1-emea-09.sun.com (d1-emea-09.sun.com [192.18.2.119]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6HDbcqS028224; Tue, 17 Jul 2007 13:37:39 GMT Received: from conversion-daemon.d1-emea-09.sun.com by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLB00601S561D00@d1-emea-09.sun.com> (original mail from Frank.Hofmann@Sun.COM); Tue, 17 Jul 2007 14:37:38 +0100 (BST) Received: from dhcp-elon07-138-57 ([129.156.138.57]) by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLB0010OSI90G9G@d1-emea-09.sun.com>; Tue, 17 Jul 2007 14:37:21 +0100 (BST) Date: Tue, 17 Jul 2007 14:37:04 +0100 (BST) From: Frank Hofmann Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <469CBA50.7020009@Sun.COM> Sender: Frank.Hofmann@Sun.COM X-X-Sender: frankho@dhcp-elon07-138-57 To: Darren J Moffat Cc: Peter Dennis , PSARC-record@sac.sfbay.sun.com, pcfs-iteam@Sun.COM Message-id: Organization: Sun Microsystems OP/N1 RPE MIME-version: 1.0 Content-type: TEXT/PLAIN; format=flowed; charset=US-ASCII Content-transfer-encoding: 7BIT References: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> <469CBA50.7020009@Sun.COM> Status: RO Content-Length: 1019 On Tue, 17 Jul 2007, Darren J Moffat wrote: > How does this proposals default of "noatime" compare to other operating > systems ? I don't understand that question, please clarify "other systems" in this context. I'm aware of the "atime"/"noatime" mount options for UFS and ZFS and know what they do - are these what you have in mind with "other systems" ? > > Why not have it so that the atime for a given file is only ever updated a > single time in a given day rather than doing a write on every access - given > you say that granularity is per day anyway ? Because that's different from what users on UFS/ZFS have come to expect from the behaviour of the "noatime" mount option. > > For my personal use I'm okay with the default, because I don't actually like > that fact that just reading a file causes a write cycle on my CF cards, but > it might not be the best default for everyone. That's why a distinction is made between removeable and fixed media. Thanks, FrankH. > > -- > Darren J Moffat > From sacadmin Tue Jul 17 06:56:39 2007 Received: from sfbaymail1sca.SFBay.Sun.COM (sfbaymail1sca.SFBay.Sun.COM [129.145.154.35]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HDud18016921; Tue, 17 Jul 2007 06:56:39 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com (gmpes-gis-mail-2.UK.Sun.COM [129.156.42.6]) by sfbaymail1sca.SFBay.Sun.COM (8.13.6+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l6HDsbVU008933; Tue, 17 Jul 2007 06:54:37 -0700 (PDT) Received: from d1-emea-09.sun.com ([192.18.2.119]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6HDsVkr019878; Tue, 17 Jul 2007 13:54:31 GMT Received: from conversion-daemon.d1-emea-09.sun.com by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLB00801T475800@d1-emea-09.sun.com> (original mail from Darren.Moffat@Sun.COM); Tue, 17 Jul 2007 14:54:31 +0100 (BST) Received: from [129.156.173.21] by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLB008CVTAUUT00@d1-emea-09.sun.com>; Tue, 17 Jul 2007 14:54:30 +0100 (BST) Date: Tue, 17 Jul 2007 14:54:30 +0100 From: Darren J Moffat Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: Sender: Darren.Moffat@Sun.COM To: Frank Hofmann Cc: Peter Dennis , PSARC-record@sac.sfbay.sun.com, pcfs-iteam@Sun.COM Message-id: <469CCA16.2030006@Sun.COM> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> <469CBA50.7020009@Sun.COM> User-Agent: Thunderbird 2.0.0.0 (X11/20070605) Status: RO Content-Length: 1149 Frank Hofmann wrote: > On Tue, 17 Jul 2007, Darren J Moffat wrote: > >> How does this proposals default of "noatime" compare to other >> operating systems ? > > I don't understand that question, please clarify "other systems" in this > context. I'm aware of the "atime"/"noatime" mount options for UFS and > ZFS and know what they do - are these what you have in mind with "other > systems" ? If I mount a FAT fileystem on Windows XP, Vista, MacOS X, Ubuntu, FreeBSD do they have access time updating disabled by default ? If not why should we be different ? >> Why not have it so that the atime for a given file is only ever >> updated a single time in a given day rather than doing a write on >> every access - given you say that granularity is per day anyway ? > > Because that's different from what users on UFS/ZFS have come to expect > from the behaviour of the "noatime" mount option. I wasn't suggesting that being the noatime behaviour but the atime behaviour. Since FAT only has a 1 day granularity if the atime is already at today don't write it out again - or is this how the code already works ? -- Darren J Moffat From sacadmin Tue Jul 17 07:04:16 2007 Received: from sfbaymail1sca.SFBay.Sun.COM (sfbaymail1sca.SFBay.Sun.COM [129.145.154.35]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HE4GgR017026; Tue, 17 Jul 2007 07:04:16 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com (gmpes-gis-mail-1.UK.Sun.COM [129.156.42.5]) by sfbaymail1sca.SFBay.Sun.COM (8.13.6+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l6HE2EOQ011916; Tue, 17 Jul 2007 07:02:14 -0700 (PDT) Received: from d1-emea-10.sun.com (d1-emea-10.sun.com [192.18.2.120]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6HE28v5001414; Tue, 17 Jul 2007 14:02:08 GMT Received: from conversion-daemon.d1-emea-10.sun.com by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLB00801TH66U00@d1-emea-10.sun.com> (original mail from Frank.Hofmann@Sun.COM); Tue, 17 Jul 2007 15:02:08 +0100 (BST) Received: from dhcp-elon07-138-57 ([129.156.138.57]) by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLB006RETNKSQW1@d1-emea-10.sun.com>; Tue, 17 Jul 2007 15:02:08 +0100 (BST) Date: Tue, 17 Jul 2007 15:01:51 +0100 (BST) From: Frank Hofmann Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <469CCA16.2030006@Sun.COM> Sender: Frank.Hofmann@Sun.COM X-X-Sender: frankho@dhcp-elon07-138-57 To: Darren J Moffat Cc: Frank Hofmann , Peter Dennis , PSARC-record@sac.sfbay.sun.com, pcfs-iteam@Sun.COM Message-id: Organization: Sun Microsystems OP/N1 RPE MIME-version: 1.0 Content-type: TEXT/PLAIN; format=flowed; charset=US-ASCII Content-transfer-encoding: 7BIT References: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> <469CBA50.7020009@Sun.COM> <469CCA16.2030006@Sun.COM> Status: RO Content-Length: 2013 On Tue, 17 Jul 2007, Darren J Moffat wrote: > Frank Hofmann wrote: >> On Tue, 17 Jul 2007, Darren J Moffat wrote: >> >>> How does this proposals default of "noatime" compare to other operating >>> systems ? >> >> I don't understand that question, please clarify "other systems" in this >> context. I'm aware of the "atime"/"noatime" mount options for UFS and ZFS >> and know what they do - are these what you have in mind with "other >> systems" ? > > If I mount a FAT fileystem on Windows XP, Vista, MacOS X, Ubuntu, FreeBSD do > they have access time updating disabled by default ? The FAT specification (google 'fatgen103.pdf') states on page 25: Date and Time Formats Many FAT file systems do not support Date/Time other than DIR_WrtTime and DIR_WrtDate. For this reason, DIR_CrtTimeMil, DIR_CrtTime, DIR_CrtDate, and DIR_LstAccDate are actually optional fields. DIR_WrtTime and DIR_WrtDate must be supported, however. If the other date and time fields are not supported, they should be set to 0 on file create and ignored on other file operations. Given this statement by Microsoft, the answer is yes. > > If not why should we be different ? We're not different. We're doing what makes sense for today's PCFS target market (removeable / flash-based media). And we're adding a filesystem feature to PCFS that UFS/ZFS have had for ages. > >>> Why not have it so that the atime for a given file is only ever updated a >>> single time in a given day rather than doing a write on every access - >>> given you say that granularity is per day anyway ? >> >> Because that's different from what users on UFS/ZFS have come to expect >> from the behaviour of the "noatime" mount option. > > I wasn't suggesting that being the noatime behaviour but the atime behaviour. > Since FAT only has a 1 day granularity if the atime is already at today don't > write it out again - or is this how the code already works ? This is how the new code works. FrankH. > > -- > Darren J Moffat > From sacadmin Tue Jul 17 08:08:34 2007 Received: from sfbaymail1sca.SFBay.Sun.COM (sfbaymail1sca.SFBay.Sun.COM [129.145.154.35]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6HF8YBC018289; Tue, 17 Jul 2007 08:08:34 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com (gmpes-gis-mail-2.UK.Sun.COM [129.156.42.6]) by sfbaymail1sca.SFBay.Sun.COM (8.13.6+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l6HF6W3N007441; Tue, 17 Jul 2007 08:06:33 -0700 (PDT) Received: from d1-emea-10.sun.com ([192.18.2.120]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6HF6RCa029032; Tue, 17 Jul 2007 15:06:27 GMT Received: from conversion-daemon.d1-emea-10.sun.com by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLB00201WMAXI00@d1-emea-10.sun.com> (original mail from Frank.Batschulat@Sun.COM); Tue, 17 Jul 2007 16:06:27 +0100 (BST) Received: from opteron ([84.188.238.118]) by d1-emea-10.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLB006JCWMPT2K1@d1-emea-10.sun.com>; Tue, 17 Jul 2007 16:06:26 +0100 (BST) Date: Tue, 17 Jul 2007 17:06:54 +0200 From: "Frank Batschulat (Home)" Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: Sender: Frank.Batschulat@Sun.COM To: Frank Hofmann , Darren J Moffat Cc: Peter Dennis , PSARC-record@sac.sfbay.sun.com, pcfs-iteam@Sun.COM Message-id: Organization: SUN Microsystems MIME-version: 1.0 Content-type: text/plain; charset=utf-8 Content-transfer-encoding: 7BIT References: <200707171223.l6HCN9ob015719@sac.sfbay.sun.com> <469CBA50.7020009@Sun.COM> <469CCA16.2030006@Sun.COM> User-Agent: Opera Mail/9.21 (SunOS) Status: RO Content-Length: 224 On Tue, 17 Jul 2007 16:01:51 +0200, Frank Hofmann wrote: > The FAT specification (google 'fatgen103.pdf') states on page 25: or see : http://fs.central.sun.com/ufs_toi/ufs_data/fs/pcfs/ --- frankB From peter.dennis@sun.com Wed Jul 18 02:20:13 2007 Received: from sunmail4.Singapore.Sun.COM (sunmail4.Singapore.Sun.COM [129.158.71.19]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6I9KCOQ016678 for ; Wed, 18 Jul 2007 02:20:12 -0700 (PDT) Received: from nwk-avmta-2.sfbay.sun.com (nwk-avmta-2.SFBay.Sun.COM [129.145.155.6]) by sunmail4.Singapore.Sun.COM (8.13.4+Sun/8.13.3/ENSMAIL,v2.2) with ESMTP id l6I9I6Hr010438; Wed, 18 Jul 2007 17:18:09 +0800 (SGT) Received: from pmxchannel-daemon.nwk-avmta-2.sfbay.sun.com by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JLD00B1FB677W00@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 02:18:07 -0700 (PDT) Received: from dm-uk-01.uk.sun.com ([129.156.101.115]) by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JLD007RXB652W40@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 02:18:06 -0700 (PDT) Received: from serinus.UK.Sun.COM (serinus.UK.Sun.COM [129.156.173.208]) by dm-uk-01.uk.sun.com (8.13.7+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l6I9I47m010335; Wed, 18 Jul 2007 10:18:04 +0100 (BST) Received: from [129.156.173.66] (eraser [129.156.173.66]) by serinus.UK.Sun.COM (8.13.7+Sun/8.13.7/CTE 3.0) with ESMTP id l6I9I2vF006583; Wed, 18 Jul 2007 10:18:04 +0100 (BST) Date: Wed, 18 Jul 2007 10:17:42 +0100 From: Peter Dennis - Solaris Sustaining Engineering Subject: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] To: psarc-ext@sun.com Cc: pcfs-iteam@sun.com, Frank Hofmann Message-id: <469DDAB6.4040900@sun.com> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7BIT X-PMX-Version: 5.2.0.264296 User-Agent: Thunderbird 2.0.0.0 (X11/20070619) Status: RO Content-Length: 2802 [resending this as it would appear the initial submission of it has not reached some folks] I am sponsoring this fasttrack for Frank Hofmann. The Release Binding is patch and the Interface is Committed for the new flags. Thanks pete Case: PCFS mount options "noatime"/"atime" Author: Frank Hofmann Summary: This proposal requests * to introduce a mount option pair for PCFS, "-o noatime" resp. "-o atime", which suppresses / enables creating access time updates. * to disable access time updates (use the "-o noatime" option by default) for PCFS if it is used on removeable media. The Release Binding is patch and the Interface is Committed. The Problem Flash-based removeable media can only perform a limited amount of write cycles. In addition to that, writes on such media are very slow (compared to reads). Since PCFS is frequently used on such devices ("memory sticks"), attempts should be made to minimize media write access whenever possible. Access timestamps are an easy target for that objective, because: - FAT access timestamps are day-granular - FAT access timestamps are optional as per FAT spec In addition, UFS has had the "noatime" mount option since 1998. PCFS should have it as well. The Proposed Solution PCFS will be changed to provide the "atime/noatime" mount options just as UFS does via PSARC 1998/093 (RFE 4088189). The default behaviour will be different from UFS, though: - removeable media: "noatime" will be default - fixed disks: "atime" will be default The reason for that is the abovementioned behaviour of flash devices, which are mostly irrelevant for UFS but a main usage scenario for PCFS, and the abovementioned behaviour of FAT access timestamps as per the specification. MAN PAGE CHANGES The man page for mount_pcfs(1M) will be changed to describe the new mount options "atime"/"noatime" as per above (6581003). The additional text will be: atime | noatime Enable or disable writing access timestamps on DOS formatted media. Default for fixed disks is atime, while for removeable media noatime is used. This is so that writes to flash-based media ("memory sticks") can be minim- ized, to prolong lifetime. From Joerg.Schilling@fokus.fraunhofer.de Wed Jul 18 02:26:52 2007 Received: from sunmail4.Singapore.Sun.COM (sunmail4.Singapore.Sun.COM [129.158.71.19]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6I9QpB5016711 for ; Wed, 18 Jul 2007 02:26:52 -0700 (PDT) Received: from nwk-avmta-1.SFBay.Sun.COM (nwk-avmta-1.SFBay.Sun.COM [129.146.11.74]) by sunmail4.Singapore.Sun.COM (8.13.4+Sun/8.13.3/ENSMAIL,v2.2) with ESMTP id l6I9OiS0012432; Wed, 18 Jul 2007 17:24:45 +0800 (SGT) Received: from pmxchannel-daemon.nwk-avmta-1.sfbay.Sun.COM by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JLD00201BH84900@nwk-avmta-1.sfbay.Sun.COM>; Wed, 18 Jul 2007 02:24:44 -0700 (PDT) Received: from brmea-mail-2.sun.com ([192.18.98.43]) by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JLD00HJYBH7Q7D0@nwk-avmta-1.sfbay.Sun.COM>; Wed, 18 Jul 2007 02:24:44 -0700 (PDT) Received: from relay41i.sun.com ([192.5.209.70]) by brmea-mail-2.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6I9ITeP026370; Wed, 18 Jul 2007 09:24:43 +0000 (GMT) Received: from mms48es.sun.com ([160.41.221.231] [160.41.221.231]) by relay41i.sun.com with ESMTP id BT-MMP-1053527; Wed, 18 Jul 2007 09:24:43 +0000 (Z) Received: from relay44i.sun.com ([192.5.209.118] [192.5.209.118]) by mms48es.sun.com with ESMTP id BT-MMP-1617215; Wed, 18 Jul 2007 09:24:43 +0000 (Z) Received: from mailgw1.fraunhofer.de ([153.96.1.18] [153.96.1.18]) by relay4i.sun.com with ESMTP id BT-MMP-8337573; Wed, 18 Jul 2007 09:24:42 +0000 (Z) Received: from mailgw1.fraunhofer.de (localhost [127.0.0.1]) by mailgw1.fraunhofer.de (8.13.5+/8.13.4) with ESMTP id l6I9OfP0009495; Wed, 18 Jul 2007 11:24:41 +0200 (CEST) Received: from pluto.fokus.fraunhofer.de (pluto.fokus.fraunhofer.de [195.37.77.164]) by mailgw1.fraunhofer.de (8.13.5+/8.13.4) with ESMTP id l6I9OeKA009440 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 18 Jul 2007 11:24:41 +0200 (CEST) Received: from EXCHSRV.fokus.fraunhofer.de (bohr [10.147.9.231]) by pluto.fokus.fraunhofer.de (8.13.7/8.13.7) with SMTP id l6I9Oe5Z029889; Wed, 18 Jul 2007 11:24:40 +0200 (MEST) Received: from burner ([10.147.65.166]) by EXCHSRV.fokus.fraunhofer.de with Microsoft SMTPSVC(6.0.3790.3959); Wed, 18 Jul 2007 11:24:40 +0200 Date: Wed, 18 Jul 2007 11:22:16 +0200 From: Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <469DDAB6.4040900@sun.com> To: psarc-ext@sun.com, peter.dennis@sun.com Cc: pcfs-iteam@sun.com, Frank.Hofmann@sun.com Message-id: <469ddbc8.6N6s0B8KDY4S3tiJ%Joerg.Schilling@fokus.fraunhofer.de> MIME-version: 1.0 Content-type: text/plain; charset=ISO8859-1 Content-transfer-encoding: 8BIT Received-SPF: None (mailgw1.fraunhofer.de: domain of Joerg.Schilling@fokus.fraunhofer.de does not designate permitted sender hosts) receiver=mailgw1.fraunhofer.de; client-ip=195.37.77.164; envelope-from=; helo=pluto.fokus.fraunhofer.de; x-agent=smf-spf; X-PMX-Version: 5.2.0.264296 X-Fraunhofer-Email-Policy: accepted References: <469DDAB6.4040900@sun.com> User-Agent: nail 11.22 3/20/05 X-OriginalArrivalTime: 18 Jul 2007 09:24:40.0828 (UTC) FILETIME=[780693C0:01C7C91D] Status: RO Content-Length: 728 Peter Dennis - Solaris Sustaining Engineering wrote: > [resending this as it would appear the initial submission of it has not > reached some folks] > > I am sponsoring this fasttrack for Frank Hofmann. > > The Release Binding is patch and the Interface is Committed for the new > flags. > > Thanks > pete > > > Case: PCFS mount options "noatime"/"atime" > > Author: Frank Hofmann +1 Jörg -- EMail:joerg@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin js@cs.tu-berlin.de (uni) schilling@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From don.cragun@Sun.COM Wed Jul 18 07:13:53 2007 Received: from sunmail5.uk.sun.com (sunmail5.UK.Sun.COM [129.156.85.165]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6IEDqaQ021666 for ; Wed, 18 Jul 2007 07:13:53 -0700 (PDT) Received: from nwk-avmta-2.sfbay.sun.com (nwk-avmta-2.SFBay.Sun.COM [129.145.155.6]) by sunmail5.uk.sun.com (8.13.7+Sun/8.13.7/ENSMAIL,v2.2) with ESMTP id l6IEBmj1028652; Wed, 18 Jul 2007 15:11:48 +0100 (BST) Received: from pmxchannel-daemon.nwk-avmta-2.sfbay.sun.com by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JLD0010LOROHO00@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 07:11:48 -0700 (PDT) Received: from spartan.SFBay.Sun.COM ([129.146.226.64]) by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JLD00M0YORNHQ30@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 07:11:47 -0700 (PDT) Received: from spartan.SFBay.Sun.COM (spartan.SFBay.Sun.COM [129.146.226.64]) by spartan.SFBay.Sun.COM (8.13.6+Sun/8.13.6) with SMTP id l6IEBlT6012378; Wed, 18 Jul 2007 07:11:47 -0700 (PDT) Date: Wed, 18 Jul 2007 07:11:47 -0700 (PDT) From: Don Cragun Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] To: psarc-ext@Sun.COM, Frank.Hofmann@Sun.COM Cc: pcfs-iteam@Sun.COM Reply-to: Don Cragun Message-id: <200707181411.l6IEBlT6012378@spartan.SFBay.Sun.COM> MIME-version: 1.0 X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.5.5 SunOS 5.9 sun4u sparc Content-type: TEXT/plain; charset=us-ascii Content-transfer-encoding: 7BIT Content-MD5: y4xsZUw3NUcOR6fbm33bZw== X-PMX-Version: 5.2.0.264296 Status: RO Content-Length: 1351 >Date: Wed, 18 Jul 2007 10:17:42 +0100 >From: Peter Dennis - Solaris Sustaining Engineering > >The Problem > Flash-based removeable media can only perform a limited amount > of write cycles. In addition to that, writes on such media are > very slow (compared to reads). Since PCFS is frequently used > on such devices ("memory sticks"), attempts should be made to > minimize media write access whenever possible. Access timestamps > are an easy target for that objective, because: > - FAT access timestamps are day-granular > - FAT access timestamps are optional as per FAT spec > > In addition, UFS has had the "noatime" mount option since 1998. > PCFS should have it as well. I wholeheartedly agree that the noatime mount option should be available on PCFS. It isn't obvious to me, however, that it should be the default. Note that since FAT access timestamps have 1 day resolution, the filesystem code is free to discard a write of an atime that has been marked for update if the new value matches the current value on the filesystem. Noting once a day that a file has been accessed shouldn't grossly affect reliability of removeable media. Do Windows OSes have a noatime mount option? If so, is it on by default? - Don From Frank.Hofmann@sun.com Wed Jul 18 07:49:38 2007 Received: from sunmail5.uk.sun.com (sunmail5.UK.Sun.COM [129.156.85.165]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6IEnc0E022522 for ; Wed, 18 Jul 2007 07:49:38 -0700 (PDT) Received: from nwk-avmta-1.SFBay.Sun.COM (nwk-avmta-1.SFBay.Sun.COM [129.146.11.74]) by sunmail5.uk.sun.com (8.13.7+Sun/8.13.7/ENSMAIL,v2.2) with ESMTP id l6IElHLM010882; Wed, 18 Jul 2007 15:47:32 +0100 (BST) Received: from pmxchannel-daemon.nwk-avmta-1.sfbay.Sun.COM by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JLD0090PQF7M800@nwk-avmta-1.sfbay.Sun.COM>; Wed, 18 Jul 2007 07:47:31 -0700 (PDT) Received: from gmp-ea-fw-1.sun.com ([129.156.42.6]) by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JLD001Z9QF55670@nwk-avmta-1.sfbay.Sun.COM>; Wed, 18 Jul 2007 07:47:30 -0700 (PDT) Received: from d1-emea-09.sun.com ([192.18.2.119]) by gmp-ea-fw-1.sun.com (8.13.6+Sun/8.12.9) with ESMTP id l6IElShD028292; Wed, 18 Jul 2007 14:47:28 +0000 (GMT) Received: from conversion-daemon.d1-emea-09.sun.com by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) id <0JLD00J01Q9F9800@d1-emea-09.sun.com> (original mail from Frank.Hofmann@Sun.COM); Wed, 18 Jul 2007 15:47:28 +0100 (BST) Received: from dhcp-elon07-138-78 ([129.156.138.78]) by d1-emea-09.sun.com (Sun Java System Messaging Server 6.2-6.01 (built Apr 3 2006)) with ESMTPSA id <0JLD001MKQF40G0M@d1-emea-09.sun.com>; Wed, 18 Jul 2007 15:47:28 +0100 (BST) Date: Wed, 18 Jul 2007 15:47:11 +0100 (BST) From: Frank Hofmann Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <200707181411.l6IEBlT6012378@spartan.SFBay.Sun.COM> Sender: Frank.Hofmann@sun.com X-X-Sender: frankho@dhcp-elon07-138-78 To: Don Cragun Cc: psarc-ext@sun.com, Frank.Hofmann@sun.com, pcfs-iteam@sun.com Message-id: Organization: Sun Microsystems OP/N1 RPE MIME-version: 1.0 Content-type: TEXT/PLAIN; format=flowed; charset=US-ASCII Content-transfer-encoding: 7BIT X-PMX-Version: 5.2.0.264296 References: <200707181411.l6IEBlT6012378@spartan.SFBay.Sun.COM> Status: RO Content-Length: 1791 On Wed, 18 Jul 2007, Don Cragun wrote: >> Date: Wed, 18 Jul 2007 10:17:42 +0100 >> From: Peter Dennis - Solaris Sustaining Engineering >> >> The Problem >> Flash-based removeable media can only perform a limited amount >> of write cycles. In addition to that, writes on such media are >> very slow (compared to reads). Since PCFS is frequently used >> on such devices ("memory sticks"), attempts should be made to >> minimize media write access whenever possible. Access timestamps >> are an easy target for that objective, because: >> - FAT access timestamps are day-granular >> - FAT access timestamps are optional as per FAT spec >> >> In addition, UFS has had the "noatime" mount option since 1998. >> PCFS should have it as well. > > I wholeheartedly agree that the noatime mount option should be > available on PCFS. It isn't obvious to me, however, that it should be > the default. Note that since FAT access timestamps have 1 day > resolution, the filesystem code is free to discard a write of an atime > that has been marked for update if the new value matches the current > value on the filesystem. Noting once a day that a file has been > accessed shouldn't grossly affect reliability of removeable media. > > Do Windows OSes have a noatime mount option? If so, is it on by > default? It's the (nonconfigurable) default on MacOS, on your mobile phone's memory card, on your digial media player and not to forget your digital camera. Those do not record access time stamps, at least I have not found any that do. I have made no detailed checks how Windows deals with removeable media as far as timestamp updates are concerned. FrankH. > > - Don > > From Garrett.Damore@sun.com Wed Jul 18 08:14:29 2007 Received: from sunmail4.Singapore.Sun.COM (sunmail4.Singapore.Sun.COM [129.158.71.19]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l6IFESI4023458 for ; Wed, 18 Jul 2007 08:14:29 -0700 (PDT) Received: from nwk-avmta-2.sfbay.sun.com (nwk-avmta-2.SFBay.Sun.COM [129.145.155.6]) by sunmail4.Singapore.Sun.COM (8.13.4+Sun/8.13.3/ENSMAIL,v2.2) with ESMTP id l6IFCMoY024222; Wed, 18 Jul 2007 23:12:23 +0800 (SGT) Received: from pmxchannel-daemon.nwk-avmta-2.sfbay.sun.com by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JLD0030BRKLC800@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 08:12:21 -0700 (PDT) Received: from sca-es-mail-2.sun.com ([192.18.43.133]) by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JLD00MIXRKKHV90@nwk-avmta-2.sfbay.sun.com>; Wed, 18 Jul 2007 08:12:20 -0700 (PDT) Received: from fe-sfbay-10.sun.com ([192.18.43.129]) by sca-es-mail-2.sun.com (8.13.7+Sun/8.12.9) with ESMTP id l6IFCKMA028948; Wed, 18 Jul 2007 08:12:20 -0700 (PDT) Received: from conversion-daemon.fe-sfbay-10.sun.com by fe-sfbay-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0JLD00901RCKQ600@fe-sfbay-10.sun.com> (original mail from Garrett.Damore@Sun.COM); Wed, 18 Jul 2007 08:12:20 -0700 (PDT) Received: from [192.168.251.21] ([76.174.83.55]) by fe-sfbay-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0JLD0005NRKJ5Q50@fe-sfbay-10.sun.com>; Wed, 18 Jul 2007 08:12:20 -0700 (PDT) Date: Wed, 18 Jul 2007 08:09:08 -0700 From: "Garrett D'Amore" Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: Sender: Garrett.Damore@sun.com To: Frank Hofmann Cc: Don Cragun , psarc-ext@sun.com, pcfs-iteam@sun.com Message-id: <469E2D14.6020104@sun.com> Organization: Sun Microsystems MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT X-PMX-Version: 5.2.0.264296 References: <200707181411.l6IEBlT6012378@spartan.SFBay.Sun.COM> User-Agent: Thunderbird 2.0.0.0 (X11/20070618) Status: RO Content-Length: 2122 Frank Hofmann wrote: > On Wed, 18 Jul 2007, Don Cragun wrote: > >>> Date: Wed, 18 Jul 2007 10:17:42 +0100 >>> From: Peter Dennis - Solaris Sustaining Engineering >>> >>> >>> The Problem >>> Flash-based removeable media can only perform a limited amount >>> of write cycles. In addition to that, writes on such media are >>> very slow (compared to reads). Since PCFS is frequently used >>> on such devices ("memory sticks"), attempts should be made to >>> minimize media write access whenever possible. Access >>> timestamps >>> are an easy target for that objective, because: >>> - FAT access timestamps are day-granular >>> - FAT access timestamps are optional as per FAT spec >>> >>> In addition, UFS has had the "noatime" mount option since 1998. >>> PCFS should have it as well. >> >> I wholeheartedly agree that the noatime mount option should be >> available on PCFS. It isn't obvious to me, however, that it should be >> the default. Note that since FAT access timestamps have 1 day >> resolution, the filesystem code is free to discard a write of an atime >> that has been marked for update if the new value matches the current >> value on the filesystem. Noting once a day that a file has been >> accessed shouldn't grossly affect reliability of removeable media. >> >> Do Windows OSes have a noatime mount option? If so, is it on by >> default? > > It's the (nonconfigurable) default on MacOS, on your mobile phone's > memory card, on your digial media player and not to forget your > digital camera. Those do not record access time stamps, at least I > have not found any that do. > > I have made no detailed checks how Windows deals with removeable media > as far as timestamp updates are concerned. I dunno about removable media, but I imagine that Windows does use this to track how recently used certain items are. If you don't access items on your desktop for long enough, Windows offers to "clean them up". I presume this works via the atime attribute. -- Garrett From peter.dennis@sun.com Thu Aug 16 02:29:50 2007 Received: from sunmail5.uk.sun.com (sunmail5.UK.Sun.COM [129.156.85.165]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id l7G9Tnaf012864 for ; Thu, 16 Aug 2007 02:29:49 -0700 (PDT) Received: from nwk-avmta-2.sfbay.sun.com (nwk-avmta-2.SFBay.Sun.COM [129.145.155.6]) by sunmail5.uk.sun.com (8.13.7+Sun/8.13.7/ENSMAIL,v2.2) with ESMTP id l7G9RK0q029781; Thu, 16 Aug 2007 10:27:20 +0100 (BST) Received: from pmxchannel-daemon.nwk-avmta-2.sfbay.sun.com by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0JMV00F050XJCV00@nwk-avmta-2.sfbay.sun.com>; Thu, 16 Aug 2007 02:27:19 -0700 (PDT) Received: from dm-uk-01.uk.sun.com ([129.156.101.115]) by nwk-avmta-2.sfbay.sun.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0JMV00EU10XIMV00@nwk-avmta-2.sfbay.sun.com>; Thu, 16 Aug 2007 02:27:19 -0700 (PDT) Received: from serinus.UK.Sun.COM (serinus.UK.Sun.COM [129.156.173.208]) by dm-uk-01.uk.sun.com (8.13.7+Sun/8.13.6/ENSMAIL,v2.2) with ESMTP id l7G9RIli004053; Thu, 16 Aug 2007 10:27:18 +0100 (BST) Received: from [129.156.173.66] (eraser [129.156.173.66]) by serinus.UK.Sun.COM (8.13.7+Sun/8.13.7/CTE 3.0) with ESMTP id l7G9RHOf025560; Thu, 16 Aug 2007 10:27:17 +0100 (BST) Date: Thu, 16 Aug 2007 10:26:52 +0100 From: Peter Dennis - Solaris Sustaining Engineering Subject: Re: PCFS mount options "noatime"/"atime" [PSARC/2007/415 FastTrack timeout 07/24/2007] In-reply-to: <469E2D14.6020104@sun.com> To: Frank Hofmann Cc: psarc-ext@sun.com, pcfs-iteam@sun.com Message-id: <46C4185C.8090202@sun.com> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7BIT X-PMX-Version: 5.2.0.264296 References: <200707181411.l6IEBlT6012378@spartan.SFBay.Sun.COM> <469E2D14.6020104@sun.com> User-Agent: Thunderbird 2.0.0.4 (X11/20070724) Status: RO Content-Length: 117 This case was approved during the meeting on 01-Aug-2007. (apologies for not sending out the notification earlier).