Ceri Davies wrote: > On Thu, Nov 06, 2008 at 07:45:13PM +0100, Roland Mainz wrote: > > Ceri Davies wrote: > > > On Tue, Nov 04, 2008 at 12:26:17PM -0600, Nicolas Williams wrote: > > > > On Tue, Nov 04, 2008 at 06:56:24PM +0100, Roland Mainz wrote: > > [snip] > > > > I think a KSH93 version of smf_include.sh that uses more built-ins would > > > > be great, but then we'd have to manage code duplication until the day > > > > arrives that we feel fine mandating KSH93 for SMF method scripts. I > > > > wouldn't mind that code duplication, but others might. > > > > > > I think the performance improvement that ksh93 could bring would make a > > > very interesting RFE. > > > > I don't have the time to do this _alone_ in a seperate RFE - the extra > > overhead eats-up too much time currently. I'll be more efficient (at > > least for me) to cleanup the script code _now_ (starting with fixing the > > warnings+errors listed by "shlint" and then do some profiling). > > It's obvious from the three emails that you sent me that you really care > about this; I'm not throwing stones at you. > > What I am saying is that this project has already been ARC'd in its > current form and having it grow to cause incompatibility issues with > other include scripts will force it to go back through that process and > that's unfair overhead just because we think ksh is shiny.
Erm... I didn't suggest to cause incompatibilties - the "ipf_include.sh" is a _new_ file and all the new consumers (e.g. introduced by this putback) which use it can be switched to ksh93 by replacing the first like from "#!/sbin/sh" to "#!/usr/bin/ksh93" (ksh93 is sufficiently backwards-compatible to the original Bourne shell that this works out-of-the-box). This doesn't need to ARC'ed (and we can ask John Plocher for confirmation if you want) and there is sufficent precedent in OS/Net. > > > I don't really think it's this project's job. > > > > What about the startup time regression ? IMO six seconds are too much. > > That's six seconds on an Ultra 2. How much is it on something people > will actually use? That was an Ultra5/333MHz and it has some impact since there are machines like the UltraSPARC T1/T2 which have low single-thread performance. ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 3992797 (;O/ \/ \O;)
