On Thu, Jan 21, 2010 at 10:09:25PM +0000, Sad Clouds wrote: > On Thursday 21 January 2010 21:39:00 Quentin Garnier wrote: > > I'd gently suggest at some point to visit, for instance, > > http://www.netbsd.org/gallery/consultants.html. Given how little you > > tell about the grand scheme behind all your questions, it might be > > better for you to work that way. > > > Am I annoying people on this list by asking too many simple questions? Nah. It's like watching a neighbour setting up scaffoldings. I do mean the suggestion though. It's not about giving money to somebody. I am just assuming that the reason you won't say anything about the bigger picture is because you can't, hence my pointing to people you can exchange information with on contracted terms. Believe it or not, knowing the big picture helps a lot to answer questions. I wouldn't call them simple, by the way. And naive doesn't really cover it either. They have that awkward feeling of being generic yet meant for a very specific purpose. > > I'm not quite sure from where you get the idea that calling mmap on a > > file makes the kernel load data into memory. > > > If you know something better than another person, why not explain it, instead > of ridiculing. By the way I do know that calling mmap on an object and > accessing it will dynamically bring in data in pages, as they are accessed. > In > my previous comments I was referring to the fact that in my view there was no > simple way of dynamically putting hard restrictions on the amount of data > cached by kernel for a particular file. Well, you're the one who said you couldn't use madvise(2). And when Steve asked what the bottleneck was in your application, you didn't answer. -- Quentin Garnier - cube%cubidou.net@localhost - cube%NetBSD.org@localhost "See the look on my face from staying too long in one place [...] every time the morning breaks I know I'm closer to falling" KT Tunstall, Saving My Face, Drastic Fantastic, 2007.
Attachment:
pgpmhQENpwelN.pgp
Description: PGP signature