Date: Sun, 15 Nov 2015 19:23:16 -0800 From: John-Mark Gurney <jmg@funkthat.com> To: Alexey Dokuchaev <danfe@freebsd.org> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r290900 - head/sbin/restore Message-ID: <20151116032316.GJ65715@funkthat.com> In-Reply-To: <20151116020842.GA5488@FreeBSD.org> References: <201511160129.tAG1Tx0V080628@repo.freebsd.org> <20151116020842.GA5488@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Alexey Dokuchaev wrote this message on Mon, Nov 16, 2015 at 02:08 +0000: > On Mon, Nov 16, 2015 at 01:29:59AM +0000, John-Mark Gurney wrote: > > New Revision: 290900 > > URL: https://svnweb.freebsd.org/changeset/base/290900 > > > > Log: > > If you backup a large file that is mostly holes, previously we'd issue > > a seek for every block... For large (Exabyte sized files) this would > > issue lots of unneeded seeks, so combine them... > > > > [...] > > @@ -938,14 +953,15 @@ getfile(void (*datafill)(char *, long), > > { > > int i; > > off_t size; > > + long seekpos; > > Briefly reading through tape.c, it looks like `long' is abused here (and in > other places). IMHO we should try to fix these bugs, not add new misuses. I agree... I'll ask if Jan is interested in fixing it, but it may need to be added to the IdeasPage... It'd also be good to have a test of regression tests for restore such that changes can be verified not to break things.. -- John-Mark Gurney Voice: +1 415 225 5579 "All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20151116032316.GJ65715>