Date: Wed, 14 Feb 2001 11:05:40 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Matt Dillon <dillon@earth.backplane.com> Cc: freebsd-emulation@FreeBSD.ORG, vsilyaev@mindspring.com Subject: Re: What's changed recently with vmware/linuxemu/file I/O Message-ID: <14986.44244.784506.737009@grasshopper.cs.duke.edu> In-Reply-To: <200102131717.f1DHHa373170@earth.backplane.com> References: <14980.8856.555504.633075@grasshopper.cs.duke.edu> <Pine.BSF.4.21.0102101303001.20995-100000@besplex.bde.org> <14980.48507.507487.690557@grasshopper.cs.duke.edu> <200102100616.f1A6GCf21887@earth.backplane.com> <3A896B57.DF61F266@Lustig.COM> <200102131717.f1DHHa373170@earth.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Matt Dillon writes: > > : > :Matt Dillon wrote: > :> As far as I know vmware is not wiring pages down. > :> > : > :Actually, when vmware starts the number of wired pages, according to > :systat, jumps significantly. > : > :barry > > That doesn't mean VMWare is wiring them down, it simply means that > the pages have been mapped and accessed by the process. Pages referenced > by page tables are always wired in the system, but the wiring is > temporary. > > -Matt I think that VMware is wiring them down. I'm sure I have heard Vladimir Silyaev mention it & host_lock_ppn() in VMware's kernel module is apparently calling vm_page_wire(). Hmm.. it should probably also be setting the PG_NOSYNC bit itself, so as to avoid most of this mess.. Speaking of PG_NOSYNC, did you happen to see my message earlier in this thread (when it was on -hackers) regarding MAP_NOSYNC not working properly when the first fault is a read fault? Thanks, Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-emulation" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14986.44244.784506.737009>