Date: Sun, 21 Jun 1998 18:17:51 -0500 (EST) From: "John S. Dyson" <dyson@iquest.net> To: karpen@ocean.campus.luth.se (Mikael Karpberg) Cc: dyson@iquest.net, mark@vmunix.com, current@FreeBSD.ORG Subject: Re: Here is what I promised :-) Message-ID: <199806212317.SAA08047@dyson.iquest.net> In-Reply-To: <199806211908.VAA27557@ocean.campus.luth.se> from Mikael Karpberg at "Jun 21, 98 09:08:51 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Mikael Karpberg said: > According to John S. Dyson: > > I am slightly suspicious of certain LL performance issues, but > > those often can be mitigated by specialization of code sequences > > (micro-optimization.) There are also aspects of the kernel > > structure that are so different (but potentially much more > > cache conserving) that it is really hard to compare non-scaled LL > > and scaled LL behavior yet. > > Short question: What does LL mean? :-) > It is *my* shortcut for Low Level. It is just laziness on my part that I don't type it. Sorry!!! :-). I define: scaled LL behavior as behavior of low level benchmarks in scaled conditions. That should NOT be confused with scaled performance in general, which is a composite. It is often true that low level performance doesn't figure significantly in the performance picture. It is also true that low level performance can count significantly, depending on the application. -- John | Never try to teach a pig to sing, dyson@iquest.net | it makes one look stupid jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199806212317.SAA08047>