Date: Thu, 03 Jul 2003 16:49:03 -0500 From: "Alan L. Cox" <alc@imimic.com> To: Marcel Moolenaar <marcel@xcllnt.net> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/alpha/alpha pmap.c src/sys/amd64/amd64pmap.c src/sys/i386/i386 pmap.c src/sys/ia64/ia64 pmap.c src/sys/powerpc/powerpc pmap.c src/sys/sparc64/sparc64 pmap.c src/sys/vm pmap.h vm_map.c Message-ID: <3F04A4CF.EAA71410@imimic.com> References: <200307032018.h63KI2TL097241@repoman.freebsd.org> <20030703203510.GA23160@dhcp01.pn.xcllnt.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Marcel Moolenaar wrote: > > On Thu, Jul 03, 2003 at 01:18:02PM -0700, Alan Cox wrote: > > alc 2003/07/03 13:18:02 PDT > > > > This commit moves the vnode-backed case into the (relatively) new > > function vm_map_pmap_enter(). On amd64 and i386, this commit only > > amounts to code rearrangement. On alpha and ia64, the new machine > > independent (MI) implementation of the vnode case is smaller and more > > efficient than their pmap-based implementations. > > Top! > > (dutch slang: means great or excellent :-) > Thanks, there is one piece of code that I actually intend to push down into the MD code: the sfbufs allocator that is used to provide a kernel virtual address for sendfile() and the zero-copy sockets code. This will benefit the 64-bit architectures in that they can then use their direct virtual-to-physical mappings, avoiding page-by-page mapping and TLB shootdown overheads. Regards, Alan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F04A4CF.EAA71410>