Date: Thu, 19 Mar 2015 15:47:08 +0000 From: "Poul-Henning Kamp" <phk@phk.freebsd.dk> To: Alexander Motin <mav@FreeBSD.org> Cc: freebsd-fs@FreeBSD.org, john hood <cgull@glup.org> Subject: Re: MMCSD erase optimization not quite right? Message-ID: <78210.1426780028@critter.freebsd.dk> In-Reply-To: <550AA479.2020404@FreeBSD.org> References: <54E80BB6.2040501@glup.org> <54F42B6B.9080307@FreeBSD.org> <550A15E6.4060903@glup.org> <550AA479.2020404@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
-------- In message <550AA479.2020404@FreeBSD.org>, Alexander Motin writes: >> Also, UFS perhaps >> shouldn't issue BIO_DELETE immediately, because the block might be >> reused soon (I'm not sure what the right answer here is for best flash >> performance). If you reuse a block quickly, the BIO_DELETE is "wasted" in the sense that the new write would do the same job, but unless you know something about your reuse pattern that allows you to optimize this aspect, statistically the best strategy is to BIO_DELETE as soon as possible. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?78210.1426780028>