Debian support on newer 4K Advanced format drives (rather than 512 bytes)
in [Debian]
|
Prev: Join July Global Debian meetings via VOIP - Free SW HW Culture Mtgs - BerkeleyTIP
Next: monolingual Americans (was switching to console and zapping)
From: Stan Hoeppner on 3 Jul 2010 16:20 Ron Johnson put forth on 7/3/2010 2:36 PM: >> This is unrelated. FS block size != sector size. > > It is when you use a 4KB drive!!!! Not according to man on Stable: mkfs.xfs [ -b block_size ] ... [ -s sector_size ] [ -L label ] [ -N ] device -b block_size_options This option specifies the fundamental block size of the filesystem. The valid block_size_options are: log=value or size=value and only one can be supplied. The block size is specified either as a base two logarithm value with log=, or in bytes with size=. The default value is 4096 bytes (4 KiB), the minimum is 512, and the maximum is 65536 (64 KiB). XFS on Linux currently only supports pagesize or smaller blocks. -s sector_size This option specifies the fundamental sector size of the filesystem. The sector_size is specified either as a value in bytes with size=value or as a base two logarithm value with log=value. The default sector_size is 512 bytes. The minimum value for sector size is 512; the maximum is 32768 (32 KiB). The sector_size must be a power of 2 size and cannot be made larger than the filesystem block size. -- Stan -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4C2F9915.7050500(a)hardwarefreak.com
From: Mike Viau on 4 Jul 2010 00:30 On the link http://www.ibm.com/developerworks/linux/library/l-4kb-sector-disks/index.html [snip] Tip: If you want to dual-boot between Linux and an older operating system that requires cylinder alignment, try aligning the starts of all your partitions on multiples of eight cylinders. This translates to 8-sector alignment for optimum disk performance as well as cylinder alignment for the older operating system. [/snip] It sounds like the term cylinders is used synonymously with sectors. Will this always be the case?? > On Sat, 3 Jul 2010 15:09:57 -0500 <stan(a)hardwarefreak.com> wrote: > > Ron Johnson put forth on 7/3/2010 2:36 PM: > > >> This is unrelated. FS block size != sector size. > > > > It is when you use a 4KB drive!!!! > > Not according to man on Stable: > > mkfs.xfs [ -b block_size ] ... [ -s sector_size ] [ -L label ] [ -N ] device > > -b block_size_options > > This option specifies the fundamental block size of the filesystem. The > valid block_size_options are: log=value or size=value and only one can be > supplied. The block size is specified either as a base two logarithm value > with log=, or in bytes with size=. The default value is 4096 bytes (4 KiB), > the minimum is 512, and the maximum is 65536 (64 KiB). XFS on Linux currently > only supports pagesize or smaller blocks. > > -s sector_size > This option specifies the fundamental sector size of the filesystem. The > sector_size is specified either as a value in bytes with size=value or as a > base two logarithm value with log=value. The default sector_size is 512 bytes. > The minimum value for sector size is 512; the maximum is 32768 (32 KiB).. The > sector_size must be a power of 2 size and cannot be made larger than the > filesystem block size. > > -- > Stan > > That seems like a very clean way to prepare a XFS file system that is aware of the 4096 byte sector size with the -s option. Its like away to override the 4096 byte sector size even if it gets read incorrectly as 512 by the /sys/block/sdX/queue/physical_block_size. [snip] In theory, the Linux kernel should return information on the physical sector size in the /sys/block/sdX/queue/physical_block_size pseudo-file and on the logical sector size in the /sys/block/sdX/queue/logical_block_size pseudo-file, where sdX is your device's node name (normally sda, sdb, and so on). In practice, however, the physical block size information is spurious, at least for the first generation of Western Digital Advanced Format drives. Unfortunately, this means that disk utilities cannot properly detect the presence of such disks. [/snip] -M _________________________________________________________________ Game on: Challenge friends to great games on Messenger http://go.microsoft.com/?linkid=9734387
From: Ron Johnson on 4 Jul 2010 09:00 On 07/03/2010 03:09 PM, Stan Hoeppner wrote: > Ron Johnson put forth on 7/3/2010 2:36 PM: > >>> This is unrelated. FS block size != sector size. >> >> It is when you use a 4KB drive!!!! > > Not according to man on Stable: > > mkfs.xfs [ -b block_size ] ... [ -s sector_size ] [ -L label ] [ -N ] device > > -b block_size_options > > This option specifies the fundamental block size of the filesystem. The > valid block_size_options are: log=value or size=value and only one can be > supplied. The block size is specified either as a base two logarithm value > with log=, or in bytes with size=. The default value is 4096 bytes (4 KiB), Ok, shame on me for forgetting to use the word "default". > the minimum is 512, and the maximum is 65536 (64 KiB). XFS on Linux currently > only supports pagesize or smaller blocks. > -- Seek truth from facts. -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4C30852A.8080607(a)cox.net
From: lee on 4 Jul 2010 11:40 On Sun, Jul 04, 2010 at 02:31:15PM +1000, CaT wrote: > > I wont be buying more of these if I can avoid it. I'd rather a 4k drive > that says it's a 4k drive and get on with life. Well, I wonder what the manufacturers thinking behind lieing about the sector size is. It only leads to problems --- everyone who bought a disk like that and partitions it as usual should just exchange it if permance testing shows poor performance until they get one that just works. And how do RAID controllers handle such disks? They present the disks transparently to the OS, and if they can't figure out that a 4k alignment is required, you can only return the disks when the performance is poor ... This problem has greatly contributed to my decision to buy one or two more 500GB disks (same model as the others I have) and to convert the RAID-1 to a RAID-5, rather than buying two 2TB disks to set up another RAID-1. It's also a lot cheaper: Two more disks will triple the capacity for less than half the price of one 2TB disk, and they are somewhat likely to be faster. -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/20100704153026.GE12004(a)yun.yagibdah.de
From: thib on 4 Jul 2010 21:30
lee wrote: > Well, I wonder what the manufacturers thinking behind lieing about the > sector size is. [...] XP, AFAIK. -t -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4C31347B.50104(a)stammed.net |