Date: Thu, 22 May 2003 20:47:33 -0700 (PDT) From: Peter Wemm <peter@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 31672 for review Message-ID: <200305230347.h4N3lXsE006960@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=31672 Change 31672 by peter@peter_hammer on 2003/05/22 20:47:07 move the recursive pml4 slot to the bottom of the negative address space, to leave all 256 positive slots for userland, for a total of 128TB of user space. I haven't increased NUPML4E though because I'm not sure how the 'unsigned pindex' stuff is going to go, and haven't investigated the vm_page_t hold_count stuff for the page table pages. It should be save to increase it from 1 to 2 though, thats 1TB of user VM space. That should be enough for a 'beta' release :-) Affected files ... .. //depot/projects/hammer/sys/amd64/include/pmap.h#9 edit .. //depot/projects/hammer/sys/amd64/include/vmparam.h#8 edit Differences ... ==== //depot/projects/hammer/sys/amd64/include/pmap.h#9 (text+ko) ==== @@ -113,7 +113,7 @@ /* * The *PDI values control the layout of virtual memory */ -#define PML4PML4I 1 +#define PML4PML4I (NPML4EPG/2) /* Index of recursive pml4 mapping */ #define KPML4I (NPML4EPG-1) #define DMPML4I (KPML4I-1) ==== //depot/projects/hammer/sys/amd64/include/vmparam.h#8 (text+ko) ==== @@ -103,7 +103,7 @@ #define UPT_MAX_ADDRESS VADDR(PML4PML4I, PML4PML4I, PML4PML4I, PML4PML4I) #define UPT_MIN_ADDRESS VADDR(PML4PML4I, 0, 0, 0) -#define VM_MAXUSER_ADDRESS UPT_MIN_ADDRESS +#define VM_MAXUSER_ADDRESS VADDR(NUPML4E, 0, 0, 0) #define USRSTACK VM_MAXUSER_ADDRESS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200305230347.h4N3lXsE006960>