Date: Thu, 09 Dec 2010 21:21:34 -0800 From: Julian Elischer <julian@freebsd.org> To: Jeremy Chadwick <freebsd@jdc.parodius.com> Cc: Kirk McKusick <mckusick@mckusick.com>, pjd@freebsd.org, freebsd-fs@freebsd.org, Oliver Fromme <olli@lurza.secnetix.de> Subject: Re: TRIM support for UFS? Message-ID: <4D01B8DE.1080601@freebsd.org> In-Reply-To: <20101210004452.GA15765@icarus.home.lan> References: <20101208225352.GK33073@deviant.kiev.zoral.com.ua> <201012082356.oB8NuxwX040914@chez.mckusick.com> <20101209103411.GL33073@deviant.kiev.zoral.com.ua> <4D01731C.4070802@freebsd.org> <20101210004452.GA15765@icarus.home.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12/9/10 4:44 PM, Jeremy Chadwick wrote: > On Thu, Dec 09, 2010 at 04:23:56PM -0800, Julian Elischer wrote: >> On 12/9/10 2:34 AM, Kostik Belousov wrote: >> >> [patch] >> >> commit when ready.. I'm looking forwards to being able to try it out. > And how does one actually verify TRIM is working? Most SSDs I've used > do not have a SMART attribute that can help determine this. I would > rather not make a bunch of assumptions about Attributes 226, 232, or 233 > to help give some indication of it working. > > For folks to test this, there needs to be some indicator/verification > mechanism that it's working. > On our (fusion-io) cards we do not go via scsi or SATA so we have our own interfaces which can provide all this information.. (tons of sysctls in FreeBSD.. tons of /proc entries in linux)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4D01B8DE.1080601>