Date: Sun, 10 Jan 1999 01:10:13 +1000 From: Stephen McKay <syssgm@dtir.qld.gov.au> To: Dmitrij Tejblum <dima@tejblum.dnttm.rssi.ru> Cc: cvs-committers@FreeBSD.ORG, syssgm@dtir.qld.gov.au Subject: Re: cvs commit: src/sys/i386/i386 pmap.c Message-ID: <199901091510.BAA24350@nymph.dtir.qld.gov.au> In-Reply-To: <199901091430.RAA00524@tejblum.dnttm.rssi.ru> from Dmitrij Tejblum at "Sat, 09 Jan 1999 17:30:03 %2B0300" References: <199901091430.RAA00524@tejblum.dnttm.rssi.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday, 9th January 1999, Dmitrij Tejblum wrote: >Stephen McKay wrote: >Ahh! That is because I made the patch month or two ago, and "simplified" >it before commit, when already forgot it ;-/ My original patch appears >to be more correct: That's why I never fix anything at the last minute any more. Either it goes in as tested, or I abort the mission. >(btw, I run it on 4 production servers all the >time, and didn't notice degradation...) You must have lots of ram. I think it should cause really suboptimal swapping. >> There's no need for all that removing and inserting anyway. Zap it and >> the bugs will depart for other dark parts of the code. > >Well, I think since the code was added in March 1998 or so as an optimization, >it should make some sense, right? Having pmap_ts_referenced() is an optimization. I think all the inserting and deleting are unnecessary, and just slow it down. For most cases, the list will be back the way it was. In cases where > 4 references are found, the list will be cycled somewhat and that *might* make the next call *slightly* faster, but at the cost of much pissing about. I say toast the inserts and deletes, and the bugs will magically vanish. Stephen. PS My test box has just hung in "thrd_sleep" while swapping like crazy (on a several day old kernel) doing a parallel make world. And I thought I'd have an easy time night... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199901091510.BAA24350>