Date: Fri, 18 Sep 2015 09:39:17 +0100 From: Steven Hartland <killing@multiplay.co.uk> To: freebsd-fs@freebsd.org Subject: Re: FreeBSD 10.1 + Samsung 850 EVO + ZFS + TRIM Message-ID: <55FBCDB5.9060403@multiplay.co.uk> In-Reply-To: <CAPyBAS6XJ=ysD6GbyE=e9BKBa913NwDBpfKmg0iPcFFj0Fe3cg@mail.gmail.com> References: <CAPyBAS6XJ=ysD6GbyE=e9BKBa913NwDBpfKmg0iPcFFj0Fe3cg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
You're explicitly doing pass0 there does that mean your not using direct attach and have some sort of RAID running on the mfi? Only CAM da and ada know how to perform TRIM requests, so if your not using either then that would be correct. What's camcontrol devlist look like? Ideally you shouldn't use a RAID based controller, a HBA will give you better performance which uses mps driver. On 18/09/2015 07:35, Tim Gustafson wrote: > Hi, > > I've Google this around a bit, but I seem to be getting some > conflicting information so I'm hoping I can get this cleared up here. > > I have a shiny new Dell PowerEdge R630 with four 1TB Samsung EVO SSDs > installed and configured as a RAID 0+1 zpool: > > pool: tank > state: ONLINE > scan: scrub repaired 14.5K in 0h13m with 0 errors on Thu Sep 17 20:47:17 2015 > config: > > NAME STATE READ WRITE CKSUM > tank ONLINE 0 0 0 > mirror-0 ONLINE 0 0 0 > mfisyspd0p3 ONLINE 0 0 0 > mfisyspd1p3 ONLINE 0 0 0 > mirror-1 ONLINE 0 0 0 > mfisyspd3p3 ONLINE 0 0 0 > mfisyspd2p3 ONLINE 0 0 0 > > errors: No known data errors > > I was investigating whether or not TRIM support was working, so I did: > > root@host: sysctl -a | grep trim > vfs.zfs.vdev.trim_on_init: 1 > vfs.zfs.vdev.trim_min_active: 1 > vfs.zfs.vdev.trim_max_active: 64 > vfs.zfs.vdev.trim_max_bytes: 2147483648 > vfs.zfs.vdev.trim_max_pending: 64 > vfs.zfs.trim.enabled: 1 > vfs.zfs.trim.txg_delay: 32 > vfs.zfs.trim.timeout: 30 > vfs.zfs.trim.max_interval: 1 > kstat.zfs.misc.zio_trim.bytes: 0 > kstat.zfs.misc.zio_trim.success: 0 > kstat.zfs.misc.zio_trim.unsupported: 546 > kstat.zfs.misc.zio_trim.failed: 0 > > That seems to imply that the OS thinks that this drive does not > support TRIM, but all the documentation about the 850 drives indicates > that they do support TRIM. Indeed, camcontrol reports that TRIM is > supported, but (if I'm reading this correctly) not enabled: > > root@host: camcontrol identify /dev/pass0 > pass0: <Samsung SSD 850 EVO 1TB EMT01B6Q> ATA-9 SATA 3.x device > pass0: 150.000MB/s transfers, Command Queueing Enabled > > protocol ATA/ATAPI-9 SATA 3.x > device model Samsung SSD 850 EVO 1TB > firmware revision EMT01B6Q > serial number S21CNXAG526512L > WWN 5002538840009228 > cylinders 16383 > heads 16 > sectors/track 63 > sector size logical 512, physical 512, offset 0 > LBA supported 268435455 sectors > LBA48 supported 1953525168 sectors > PIO supported PIO4 > DMA supported WDMA2 UDMA6 > media RPM non-rotating > > Feature Support Enabled Value Vendor > read ahead yes yes > write cache yes yes > flush cache yes yes > overlap no > Tagged Command Queuing (TCQ) no no > Native Command Queuing (NCQ) yes 32 tags > NCQ Queue Management no > NCQ Streaming no > Receive & Send FPDMA Queued yes > SMART yes yes > microcode download yes yes > security yes no > power management yes yes > advanced power management no no > automatic acoustic management no no > media status notification no no > power-up in Standby no no > write-read-verify yes no 0/0x0 > unload no no > general purpose logging yes yes > free-fall no no > Data Set Management (DSM/TRIM) yes > DSM - max 512byte blocks yes 8 > DSM - deterministic read no > Host Protected Area (HPA) yes no 1953525168/1 > HPA - Security no > > So, my guess is that while the drive supports TRIM, perhaps FreeBSD > has disabled TRIM for this drive, perhaps because of the buggy TRIM > firmware issues with some drives? I thought those issues were present > in the 840 series, but not in the 850 series. Or am I not reading > these screens correctly? > > - >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?55FBCDB5.9060403>