From fbdfe8ac226139c2a7e44ae5130748a6770cd731 Mon Sep 17 00:00:00 2001 From: David Greenman Date: Tue, 24 Jan 1995 09:56:33 +0000 Subject: [PATCH] Changed buffer allocation policy (machdep.c) Moved various pmap 'bit' test/set functions back into real functions; gcc generates better code at the expense of more of it. (pmap.c) Fixed a deadlock problem with pv entry allocations (pmap.c) Added a new, optional function 'pmap_prefault' that does clustered page table preloading (pmap.c) Changed the way that page tables are held onto (trap.c). Submitted by: John Dyson --- sys/amd64/amd64/machdep.c | 13 +- sys/amd64/amd64/pmap.c | 806 ++++++++++++++++++++++---------------- sys/amd64/amd64/trap.c | 17 +- sys/i386/i386/machdep.c | 13 +- sys/i386/i386/pmap.c | 806 ++++++++++++++++++++++---------------- sys/i386/i386/trap.c | 17 +- sys/kern/subr_trap.c | 17 +- 7 files changed, 967 insertions(+), 722 deletions(-) diff --git a/sys/amd64/amd64/machdep.c b/sys/amd64/amd64/machdep.c index 9e041fc6aa1..bdbbadf120d 100644 --- a/sys/amd64/amd64/machdep.c +++ b/sys/amd64/amd64/machdep.c @@ -35,7 +35,7 @@ * SUCH DAMAGE. * * from: @(#)machdep.c 7.4 (Berkeley) 6/3/91 - * $Id: machdep.c,v 1.101 1995/01/14 13:20:08 bde Exp $ + * $Id: machdep.c,v 1.102 1995/01/17 01:15:12 bde Exp $ */ #include "npx.h" @@ -114,7 +114,7 @@ char cpu_model[sizeof("Cy486DLC") + 1]; /* * Declare these as initialized data so we can patch them. */ -int nswbuf = 128; +int nswbuf = 0; #ifdef NBUF int nbuf = NBUF; #else @@ -255,9 +255,12 @@ again: valloc(msqids, struct msqid_ds, msginfo.msgmni); #endif - if (nbuf == 0) - nbuf = min(physmem / 30, 256); - nswbuf = nbuf; + if (nbuf == 0) { + nbuf = 32; + if( physmem > 1024) + nbuf += min((physmem - 1024) / 20, 1024); + } + nswbuf = min(nbuf, 128); valloc(swbuf, struct buf, nswbuf); valloc(buf, struct buf, nbuf); diff --git a/sys/amd64/amd64/pmap.c b/sys/amd64/amd64/pmap.c index eb72d020db4..5febdd0895b 100644 --- a/sys/amd64/amd64/pmap.c +++ b/sys/amd64/amd64/pmap.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (c) 1991 Regents of the University of California. * All rights reserved. * Copyright (c) 1994 John S. Dyson @@ -39,7 +39,7 @@ * SUCH DAMAGE. * * from: @(#)pmap.c 7.7 (Berkeley) 5/12/91 - * $Id: pmap.c,v 1.41 1995/01/14 04:58:53 davidg Exp $ + * $Id: pmap.c,v 1.42 1995/01/15 09:06:23 davidg Exp $ */ /* @@ -53,10 +53,10 @@ */ /* * Major modifications by John S. Dyson primarily to support - * pageable page tables, eliminating pmap_attributes, + * pageable page tables, eliminating pmap_attributes, * discontiguous memory pages, and using more efficient string * instructions. Jan 13, 1994. Further modifications on Mar 2, 1994, - * general clean-up and efficiency mods. + * general clean-up and efficiency mods. */ /* @@ -118,7 +118,7 @@ #define pmap_pte_m(pte) ((*(int *)pte & PG_M) != 0) #define pmap_pte_u(pte) ((*(int *)pte & PG_U) != 0) #define pmap_pte_v(pte) ((*(int *)pte & PG_V) != 0) - + #define pmap_pte_set_w(pte, v) ((v)?(*(int *)pte |= PG_W):(*(int *)pte &= ~PG_W)) #define pmap_pte_set_prot(pte, v) ((*(int *)pte &= ~PG_PROT), (*(int *)pte |= (v))) @@ -127,29 +127,29 @@ * convert to a vax protection code. */ #define pte_prot(m, p) (protection_codes[p]) -int protection_codes[8]; +int protection_codes[8]; -struct pmap kernel_pmap_store; -pmap_t kernel_pmap; +struct pmap kernel_pmap_store; +pmap_t kernel_pmap; -vm_offset_t phys_avail[6]; /* 2 entries + 1 null */ -vm_offset_t avail_start; /* PA of first available physical page */ -vm_offset_t avail_end; /* PA of last available physical page */ -vm_size_t mem_size; /* memory size in bytes */ -vm_offset_t virtual_avail; /* VA of first avail page (after kernel bss)*/ -vm_offset_t virtual_end; /* VA of last avail page (end of kernel AS) */ -int i386pagesperpage; /* PAGE_SIZE / I386_PAGE_SIZE */ -boolean_t pmap_initialized = FALSE; /* Has pmap_init completed? */ -vm_offset_t vm_first_phys, vm_last_phys; +vm_offset_t phys_avail[6]; /* 2 entries + 1 null */ +vm_offset_t avail_start; /* PA of first available physical page */ +vm_offset_t avail_end; /* PA of last available physical page */ +vm_size_t mem_size; /* memory size in bytes */ +vm_offset_t virtual_avail; /* VA of first avail page (after kernel bss) */ +vm_offset_t virtual_end; /* VA of last avail page (end of kernel AS) */ +int i386pagesperpage; /* PAGE_SIZE / I386_PAGE_SIZE */ +boolean_t pmap_initialized = FALSE; /* Has pmap_init completed? */ +vm_offset_t vm_first_phys, vm_last_phys; -static inline int pmap_is_managed(); -static inline void * vm_get_pmap(); -static inline void vm_put_pmap(); -static void i386_protection_init(); -static void pmap_alloc_pv_entry(); -static inline pv_entry_t get_pv_entry(); -static inline void pmap_use_pt(); -static inline void pmap_unuse_pt(); +static inline int pmap_is_managed(); +static inline void *vm_get_pmap(); +static inline void vm_put_pmap(); +static void i386_protection_init(); +static void pmap_alloc_pv_entry(); +static inline pv_entry_t get_pv_entry(); +inline void pmap_use_pt(); +inline void pmap_unuse_pt(); int nkpt; @@ -163,12 +163,14 @@ extern int cpu_class; * All those kernel PT submaps that BSD is so fond of */ pt_entry_t *CMAP1, *CMAP2, *ptmmap; -caddr_t CADDR1, CADDR2, ptvmmap; +caddr_t CADDR1, CADDR2, ptvmmap; pt_entry_t *msgbufmap; -struct msgbuf *msgbufp; +struct msgbuf *msgbufp; + #endif -void init_pv_entries(int) ; +void +init_pv_entries(int); /* * Routine: pmap_pte @@ -178,27 +180,28 @@ void init_pv_entries(int) ; * [ what about induced faults -wfj] */ -inline pt_entry_t * -const pmap_pte(pmap, va) - register pmap_t pmap; +inline pt_entry_t * const +pmap_pte(pmap, va) + register pmap_t pmap; vm_offset_t va; { if (pmap && *pmap_pde(pmap, va)) { vm_offset_t frame = (int) pmap->pm_pdir[PTDPTDI] & PG_FRAME; + /* are we current address space or kernel? */ - if ( (pmap == kernel_pmap) || (frame == ((int) PTDpde & PG_FRAME))) + if ((pmap == kernel_pmap) || (frame == ((int) PTDpde & PG_FRAME))) return ((pt_entry_t *) vtopte(va)); /* otherwise, we are alternate address space */ else { - if ( frame != ((int) APTDpde & PG_FRAME) ) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } - return((pt_entry_t *) avtopte(va)); + return ((pt_entry_t *) avtopte(va)); } } - return(0); + return (0); } /* @@ -210,20 +213,21 @@ const pmap_pte(pmap, va) vm_offset_t pmap_extract(pmap, va) - register pmap_t pmap; + register pmap_t pmap; vm_offset_t va; { vm_offset_t pa; if (pmap && *pmap_pde(pmap, va)) { vm_offset_t frame = (int) pmap->pm_pdir[PTDPTDI] & PG_FRAME; + /* are we current address space or kernel? */ - if ( (pmap == kernel_pmap) - || (frame == ((int) PTDpde & PG_FRAME)) ) { + if ((pmap == kernel_pmap) + || (frame == ((int) PTDpde & PG_FRAME))) { pa = *(int *) vtopte(va); - /* otherwise, we are alternate address space */ + /* otherwise, we are alternate address space */ } else { - if ( frame != ((int) APTDpde & PG_FRAME)) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } @@ -232,7 +236,7 @@ pmap_extract(pmap, va) return ((pa & PG_FRAME) | (va & ~PG_FRAME)); } return 0; - + } /* @@ -262,10 +266,13 @@ pmap_pte_vm_page(pmap, pt) pmap_t pmap; vm_offset_t pt; { - pt = i386_trunc_page( pt); + vm_page_t m; + + pt = i386_trunc_page(pt); pt = (pt - UPT_MIN_ADDRESS) / NBPG; pt = ((vm_offset_t) pmap->pm_pdir[pt]) & PG_FRAME; - return PHYS_TO_VM_PAGE(pt); + m = PHYS_TO_VM_PAGE(pt); + return m; } /* @@ -278,11 +285,11 @@ pmap_use_pt(pmap, va) { vm_offset_t pt; - if (va >= VM_MAX_ADDRESS || !pmap_initialized) - return; + if ((va >= UPT_MIN_ADDRESS) || !pmap_initialized) + return; pt = (vm_offset_t) vtopte(va); - vm_page_hold( pmap_pte_vm_page(pmap, pt)); + vm_page_hold(pmap_pte_vm_page(pmap, pt)); } /* @@ -294,12 +301,20 @@ pmap_unuse_pt(pmap, va) vm_offset_t va; { vm_offset_t pt; + vm_page_t m; - if (va >= VM_MAX_ADDRESS || !pmap_initialized) - return; + if ((va >= UPT_MIN_ADDRESS) || !pmap_initialized) + return; pt = (vm_offset_t) vtopte(va); - vm_page_unhold( pmap_pte_vm_page(pmap, pt)); + m = pmap_pte_vm_page(pmap, pt); + vm_page_unhold(m); + if (pmap != kernel_pmap && + (m->hold_count == 0) && + (m->wire_count == 0) && + (va < KPT_MIN_ADDRESS)) { + vm_page_deactivate(m); + } } /* [ macro again?, should I force kstack into user map here? -wfj ] */ @@ -332,9 +347,10 @@ pmap_bootstrap(firstaddr, loadaddr) #if BSDVM_COMPAT vm_offset_t va; pt_entry_t *pte; + #endif - avail_start = firstaddr + DMAPAGES*NBPG; + avail_start = firstaddr + DMAPAGES * NBPG; virtual_avail = (vm_offset_t) KERNBASE + avail_start; virtual_end = VM_MAX_KERNEL_ADDRESS; @@ -346,13 +362,13 @@ pmap_bootstrap(firstaddr, loadaddr) i386_protection_init(); /* - * The kernel's pmap is statically allocated so we don't - * have to use pmap_create, which is unlikely to work - * correctly at this part of the boot sequence. + * The kernel's pmap is statically allocated so we don't have to use + * pmap_create, which is unlikely to work correctly at this part of + * the boot sequence. */ kernel_pmap = &kernel_pmap_store; - kernel_pmap->pm_pdir = (pd_entry_t *)(KERNBASE + IdlePTD); + kernel_pmap->pm_pdir = (pd_entry_t *) (KERNBASE + IdlePTD); simple_lock_init(&kernel_pmap->pm_lock); kernel_pmap->pm_count = 1; @@ -368,11 +384,11 @@ pmap_bootstrap(firstaddr, loadaddr) va = virtual_avail; pte = pmap_pte(kernel_pmap, va); - SYSMAP(caddr_t ,CMAP1 ,CADDR1 ,1 ) - SYSMAP(caddr_t ,CMAP2 ,CADDR2 ,1 ) - SYSMAP(caddr_t ,ptmmap ,ptvmmap ,1 ) - SYSMAP(struct msgbuf * ,msgbufmap ,msgbufp ,1 ) - virtual_avail = va; + SYSMAP(caddr_t, CMAP1, CADDR1, 1) + SYSMAP(caddr_t, CMAP2, CADDR2, 1) + SYSMAP(caddr_t, ptmmap, ptvmmap, 1) + SYSMAP(struct msgbuf *, msgbufmap, msgbufp, 1) + virtual_avail = va; #endif /* * Reserve special hunk of memory for use by bus dma as a bounce @@ -380,13 +396,14 @@ pmap_bootstrap(firstaddr, loadaddr) */ { extern vm_offset_t isaphysmem; + isaphysmem = va; virtual_avail = pmap_map(va, firstaddr, - firstaddr + DMAPAGES*NBPG, VM_PROT_ALL); + firstaddr + DMAPAGES * NBPG, VM_PROT_ALL); } - *(int *)CMAP1 = *(int *)CMAP2 = *(int *)PTD = 0; + *(int *) CMAP1 = *(int *) CMAP2 = *(int *) PTD = 0; pmap_update(); } @@ -400,30 +417,30 @@ pmap_bootstrap(firstaddr, loadaddr) */ void pmap_init(phys_start, phys_end) - vm_offset_t phys_start, phys_end; + vm_offset_t phys_start, phys_end; { - vm_offset_t addr; - vm_size_t npg, s; + vm_offset_t addr; + vm_size_t npg, s; int i; /* - * Now that kernel map has been allocated, we can mark as - * unavailable regions which we have mapped in locore. + * Now that kernel map has been allocated, we can mark as unavailable + * regions which we have mapped in locore. */ addr = atdevbase; (void) vm_map_find(kernel_map, NULL, (vm_offset_t) 0, - &addr, (0x100000-0xa0000), FALSE); + &addr, (0x100000 - 0xa0000), FALSE); addr = (vm_offset_t) KERNBASE + IdlePTD; vm_object_reference(kernel_object); (void) vm_map_find(kernel_map, kernel_object, addr, - &addr, (4 + NKPDE) * NBPG, FALSE); + &addr, (4 + NKPDE) * NBPG, FALSE); /* * calculate the number of pv_entries needed */ vm_first_phys = phys_avail[0]; - for (i = 0; phys_avail[i + 1]; i += 2) ; + for (i = 0; phys_avail[i + 1]; i += 2); npg = (phys_avail[(i - 2) + 1] - vm_first_phys) / NBPG; /* @@ -454,17 +471,17 @@ pmap_init(phys_start, phys_end) */ vm_offset_t pmap_map(virt, start, end, prot) - vm_offset_t virt; - vm_offset_t start; - vm_offset_t end; - int prot; + vm_offset_t virt; + vm_offset_t start; + vm_offset_t end; + int prot; { while (start < end) { pmap_enter(kernel_pmap, virt, start, prot, FALSE); virt += PAGE_SIZE; start += PAGE_SIZE; } - return(virt); + return (virt); } /* @@ -485,7 +502,7 @@ pmap_map(virt, start, end, prot) pmap_t pmap_create(size) - vm_size_t size; + vm_size_t size; { register pmap_t pmap; @@ -493,7 +510,7 @@ pmap_create(size) * Software use map does not need a pmap */ if (size) - return(NULL); + return (NULL); pmap = (pmap_t) malloc(sizeof *pmap, M_VMPMAP, M_WAITOK); bzero(pmap, sizeof(*pmap)); @@ -511,7 +528,7 @@ vm_get_pmap() { struct pmaplist *rtval; - rtval = (struct pmaplist *)kmem_alloc(kernel_map, ctob(1)); + rtval = (struct pmaplist *) kmem_alloc(kernel_map, ctob(1)); bzero(rtval, ctob(1)); return rtval; } @@ -520,7 +537,7 @@ static inline void vm_put_pmap(up) struct pmaplist *up; { - kmem_free(kernel_map, (vm_offset_t)up, ctob(1)); + kmem_free(kernel_map, (vm_offset_t) up, ctob(1)); } /* @@ -532,17 +549,17 @@ pmap_pinit(pmap) register struct pmap *pmap; { /* - * No need to allocate page table space yet but we do need a - * valid page directory table. + * No need to allocate page table space yet but we do need a valid + * page directory table. */ pmap->pm_pdir = (pd_entry_t *) vm_get_pmap(); /* wire in kernel global address entries */ - bcopy(PTD+KPTDI, pmap->pm_pdir+KPTDI, nkpt*PTESIZE); + bcopy(PTD + KPTDI, pmap->pm_pdir + KPTDI, nkpt * PTESIZE); /* install self-referential address mapping entry */ - *(int *)(pmap->pm_pdir+PTDPTDI) = - ((int)pmap_kextract((vm_offset_t)pmap->pm_pdir)) | PG_V | PG_KW; + *(int *) (pmap->pm_pdir + PTDPTDI) = + ((int) pmap_kextract((vm_offset_t) pmap->pm_pdir)) | PG_V | PG_KW; pmap->pm_count = 1; simple_lock_init(&pmap->pm_lock); @@ -556,54 +573,47 @@ vm_page_t nkpg; vm_offset_t kernel_vm_end; void -pmap_growkernel(vm_offset_t addr) { +pmap_growkernel(vm_offset_t addr) +{ struct proc *p; struct pmap *pmap; int s; - + s = splhigh(); if (kernel_vm_end == 0) { kernel_vm_end = KERNBASE; nkpt = 0; - while(pdir_pde(PTD, kernel_vm_end)) { - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + while (pdir_pde(PTD, kernel_vm_end)) { + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); ++nkpt; } } - - addr = (addr + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); - while( kernel_vm_end < addr) { - if( pdir_pde( PTD, kernel_vm_end)) { - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + addr = (addr + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); + while (kernel_vm_end < addr) { + if (pdir_pde(PTD, kernel_vm_end)) { + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); continue; } - ++nkpt; - if( !nkpg) { - nkpg = vm_page_alloc(kernel_object, 0, TRUE); + if (!nkpg) { + nkpg = vm_page_alloc(kernel_object, 0, VM_ALLOC_SYSTEM); + if (!nkpg) + panic("pmap_growkernel: no memory to grow kernel"); + vm_page_wire(nkpg); vm_page_remove(nkpg); pmap_zero_page(VM_PAGE_TO_PHYS(nkpg)); - if( !nkpg) - panic("pmap_growkernel: no memory to grow kernel"); } - pdir_pde( PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_KW); + pdir_pde(PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_KW); nkpg = NULL; - for (p = (struct proc *)allproc; p != NULL; p = p->p_next) { - if( p->p_vmspace) { + for (p = (struct proc *) allproc; p != NULL; p = p->p_next) { + if (p->p_vmspace) { pmap = &p->p_vmspace->vm_pmap; - *pmap_pde( pmap, kernel_vm_end) = pdir_pde( PTD, kernel_vm_end); + *pmap_pde(pmap, kernel_vm_end) = pdir_pde(PTD, kernel_vm_end); } } - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); } -#if 0 - if( !nkpg) { - nkpg = vm_page_alloc(kernel_object, 0, TRUE); - vm_page_remove(nkpg); - pmap_zero_page(VM_PAGE_TO_PHYS(nkpg)); - } -#endif splx(s); } @@ -626,7 +636,7 @@ pmap_destroy(pmap) simple_unlock(&pmap->pm_lock); if (count == 0) { pmap_release(pmap); - free((caddr_t)pmap, M_VMPMAP); + free((caddr_t) pmap, M_VMPMAP); } } @@ -647,7 +657,7 @@ pmap_release(pmap) */ void pmap_reference(pmap) - pmap_t pmap; + pmap_t pmap; { if (pmap != NULL) { simple_lock(&pmap->pm_lock); @@ -673,7 +683,8 @@ inline static void free_pv_entry(pv) pv_entry_t pv; { - if (!pv) return; + if (!pv) + return; ++pv_freelistcnt; pv->pv_next = pv_freelist; pv_freelist = pv; @@ -693,19 +704,15 @@ get_pv_entry() /* * get more pv_entry pages if needed */ - while (pv_freelistcnt < PV_FREELIST_MIN || pv_freelist == 0) { + if (pv_freelistcnt < PV_FREELIST_MIN || pv_freelist == 0) { pmap_alloc_pv_entry(); } - /* * get a pv_entry off of the free list */ --pv_freelistcnt; tmp = pv_freelist; pv_freelist = tmp->pv_next; - tmp->pv_pmap = 0; - tmp->pv_va = 0; - tmp->pv_next = 0; return tmp; } @@ -722,6 +729,7 @@ pmap_alloc_pv_entry() */ if (npvvapg) { vm_page_t m; + /* * we do this to keep recursion away */ @@ -729,12 +737,14 @@ pmap_alloc_pv_entry() /* * allocate a physical page out of the vm system */ - m = vm_page_alloc(kernel_object, pvva-vm_map_min(kernel_map), TRUE); + m = vm_page_alloc(kernel_object, + pvva - vm_map_min(kernel_map), VM_ALLOC_INTERRUPT); if (m) { int newentries; int i; pv_entry_t entry; - newentries = (NBPG/sizeof (struct pv_entry)); + + newentries = (NBPG / sizeof(struct pv_entry)); /* * wire the page */ @@ -778,10 +788,10 @@ init_pv_entries(npg) { /* * allocate enough kvm space for PVSPERPAGE entries per page (lots) - * kvm space is fairly cheap, be generous!!! (the system can panic - * if this is too small.) + * kvm space is fairly cheap, be generous!!! (the system can panic if + * this is too small.) */ - npvvapg = ((npg*PVSPERPAGE) * sizeof(struct pv_entry) + NBPG - 1)/NBPG; + npvvapg = ((npg * PVSPERPAGE) * sizeof(struct pv_entry) + NBPG - 1) / NBPG; pvva = kmem_alloc_pageable(kernel_map, npvvapg * NBPG); /* * get the first batch of entries @@ -800,7 +810,7 @@ get_pt_entry(pmap) return PTmap; } /* otherwise, we are alternate address space */ - if ( frame != ((int) APTDpde & PG_FRAME)) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } @@ -821,6 +831,7 @@ pmap_remove_entry(pmap, pv, va) { pv_entry_t npv; int s; + s = splhigh(); if (pmap == pv->pv_pmap && va == pv->pv_va) { npv = pv->pv_next; @@ -840,7 +851,7 @@ pmap_remove_entry(pmap, pv, va) if (npv) { pv->pv_next = npv->pv_next; free_pv_entry(npv); - } + } } splx(s); } @@ -857,7 +868,7 @@ pmap_remove(pmap, sva, eva) register vm_offset_t sva; register vm_offset_t eva; { - register pt_entry_t *ptp,*ptq; + register pt_entry_t *ptp, *ptq; vm_offset_t pa; register pv_entry_t pv; vm_offset_t va; @@ -869,19 +880,19 @@ pmap_remove(pmap, sva, eva) ptp = get_pt_entry(pmap); -/* - * special handling of removing one page. a very - * common operation and easy to short circuit some - * code. - */ - if( (sva + NBPG) == eva) { - - if( *pmap_pde( pmap, sva) == 0) + /* + * special handling of removing one page. a very + * common operation and easy to short circuit some + * code. + */ + if ((sva + NBPG) == eva) { + + if (*pmap_pde(pmap, sva) == 0) return; ptq = ptp + i386_btop(sva); - if( !*ptq) + if (!*ptq) return; /* * Update statistics @@ -897,46 +908,41 @@ pmap_remove(pmap, sva, eva) if (pmap_is_managed(pa)) { if ((int) oldpte & PG_M) { if ((sva < USRSTACK || sva > UPT_MAX_ADDRESS) || - (sva >= USRSTACK && sva < USRSTACK+(UPAGES*NBPG))) { + (sva >= USRSTACK && sva < USRSTACK + (UPAGES * NBPG))) { if (sva < clean_sva || sva >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - pv = pa_to_pvh(pa); pmap_remove_entry(pmap, pv, sva); } - pmap_unuse_pt(pmap, sva); + pmap_unuse_pt(pmap, sva); pmap_update(); return; } - sva = i386_btop(sva); eva = i386_btop(eva); while (sva < eva) { /* - * Weed out invalid mappings. - * Note: we assume that the page directory table is - * always allocated, and in kernel virtual. + * Weed out invalid mappings. Note: we assume that the page + * directory table is always allocated, and in kernel virtual. */ - if ( *pmap_pde(pmap, i386_ptob(sva)) == 0 ) { + if (*pmap_pde(pmap, i386_ptob(sva)) == 0) { /* We can race ahead here, straight to next pde.. */ sva = ((sva + NPTEPG) & ~(NPTEPG - 1)); continue; } - ptq = ptp + sva; /* - * search for page table entries, use string operations - * that are much faster than - * explicitly scanning when page tables are not fully - * populated. + * search for page table entries, use string operations that + * are much faster than explicitly scanning when page tables + * are not fully populated. */ - if ( *ptq == 0) { + if (*ptq == 0) { vm_offset_t pdnxt = ((sva + NPTEPG) & ~(NPTEPG - 1)); vm_offset_t nscan = pdnxt - sva; int found = 0; @@ -944,12 +950,10 @@ pmap_remove(pmap, sva, eva) if ((nscan + sva) > eva) nscan = eva - sva; - asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" - :"=D"(ptq),"=a"(found) - :"c"(nscan),"0"(ptq) - :"cx"); + asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" : + "=D"(ptq), "=a"(found) : "c"(nscan), "0"(ptq) : "cx"); - if( !found) { + if (!found) { sva = pdnxt; continue; } @@ -957,47 +961,43 @@ pmap_remove(pmap, sva, eva) sva = ptq - ptp; } - /* * Update statistics */ oldpte = *ptq; - if (((int)oldpte) & PG_W) + if (((int) oldpte) & PG_W) pmap->pm_stats.wired_count--; pmap->pm_stats.resident_count--; /* - * Invalidate the PTEs. - * XXX: should cluster them up and invalidate as many - * as possible at once. + * Invalidate the PTEs. XXX: should cluster them up and + * invalidate as many as possible at once. */ *ptq = 0; va = i386_ptob(sva); /* - * Remove from the PV table (raise IPL since we - * may be called at interrupt time). + * Remove from the PV table (raise IPL since we may be called + * at interrupt time). */ - pa = ((int)oldpte) & PG_FRAME; + pa = ((int) oldpte) & PG_FRAME; if (!pmap_is_managed(pa)) { - pmap_unuse_pt(pmap, va); + pmap_unuse_pt(pmap, va); ++sva; continue; } - if ((int) oldpte & PG_M) { if ((va < USRSTACK || va > UPT_MAX_ADDRESS) || - (va >= USRSTACK && va < USRSTACK+(UPAGES*NBPG))) { + (va >= USRSTACK && va < USRSTACK + (UPAGES * NBPG))) { if (va < clean_sva || va >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - pv = pa_to_pvh(pa); pmap_remove_entry(pmap, pv, va); - pmap_unuse_pt(pmap, va); + pmap_unuse_pt(pmap, va); ++sva; } pmap_update(); @@ -1030,7 +1030,10 @@ pmap_remove_all(pa) /* * Not one of ours */ - /* XXX this makes pmap_page_protect(NONE) illegal for non-managed pages! */ + /* + * XXX this makes pmap_page_protect(NONE) illegal for non-managed + * pages! + */ if (!pmap_is_managed(pa)) return; @@ -1049,23 +1052,21 @@ pmap_remove_all(pa) if (*pte) { pmap->pm_stats.resident_count--; anyvalid++; - + /* * Update the vm_page_t clean and reference bits. */ if ((int) *pte & PG_M) { if ((va < USRSTACK || va > UPT_MAX_ADDRESS) || - (va >= USRSTACK && va < USRSTACK+(UPAGES*NBPG))) { + (va >= USRSTACK && va < USRSTACK + (UPAGES * NBPG))) { if (va < clean_sva || va >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - *pte = 0; pmap_unuse_pt(pmap, va); } - npv = pv->pv_next; if (npv) { *pv = *npv; @@ -1086,9 +1087,9 @@ pmap_remove_all(pa) */ void pmap_protect(pmap, sva, eva, prot) - register pmap_t pmap; - vm_offset_t sva, eva; - vm_prot_t prot; + register pmap_t pmap; + vm_offset_t sva, eva; + vm_prot_t prot; { register pt_entry_t *pte; register vm_offset_t va; @@ -1111,29 +1112,29 @@ pmap_protect(pmap, sva, eva, prot) va = sva; while (va < eva) { - int found=0; + int found = 0; int svap; vm_offset_t nscan; + /* - * Page table page is not allocated. - * Skip it, we don't want to force allocation - * of unnecessary PTE pages just to set the protection. + * Page table page is not allocated. Skip it, we don't want to + * force allocation of unnecessary PTE pages just to set the + * protection. */ - if (! *pmap_pde(pmap, va)) { + if (!*pmap_pde(pmap, va)) { /* XXX: avoid address wrap around */ -nextpde: - if (va >= i386_trunc_pdr((vm_offset_t)-1)) + nextpde: + if (va >= i386_trunc_pdr((vm_offset_t) - 1)) break; va = i386_round_pdr(va + PAGE_SIZE); continue; } - pte = ptp + i386_btop(va); - if( *pte == 0) { - /* - * scan for a non-empty pte - */ + if (*pte == 0) { + /* + * scan for a non-empty pte + */ svap = pte - ptp; nscan = ((svap + NPTEPG) & ~(NPTEPG - 1)) - svap; @@ -1142,11 +1143,10 @@ nextpde: found = 0; if (nscan) - asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" - :"=D"(pte),"=a"(found) - :"c"(nscan),"0"(pte):"cx"); + asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" : + "=D"(pte), "=a"(found) : "c"(nscan), "0"(pte) : "cx"); - if( !found) + if (!found) goto nextpde; pte -= 1; @@ -1154,13 +1154,12 @@ nextpde: va = i386_ptob(svap); } - anyvalid++; i386prot = pte_prot(pmap, prot); if (va < UPT_MAX_ADDRESS) { i386prot |= PG_u; - if( va >= UPT_MIN_ADDRESS) + if (va >= UPT_MIN_ADDRESS) i386prot |= PG_RW; } pmap_pte_set_prot(pte, i386prot); @@ -1193,7 +1192,7 @@ pmap_enter(pmap, va, pa, prot, wired) register pt_entry_t *pte; register pt_entry_t npte; vm_offset_t opa; - int cacheable=1; + int cacheable = 1; int ptevalid = 0; if (pmap == NULL) @@ -1211,7 +1210,6 @@ pmap_enter(pmap, va, pa, prot, wired) printf("kernel page directory invalid pdir=0x%x, va=0x%x\n", pmap->pm_pdir[PTDPTDI], va); panic("invalid kernel page directory"); } - pte = pmap_pte(pmap, va); opa = pmap_pte_pa(pte); @@ -1220,10 +1218,10 @@ pmap_enter(pmap, va, pa, prot, wired) */ if (opa == pa) { /* - * Wiring change, just update stats. - * We don't worry about wiring PT pages as they remain - * resident as long as there are valid mappings in them. - * Hence, if a user page is wired, the PT page will be also. + * Wiring change, just update stats. We don't worry about + * wiring PT pages as they remain resident as long as there + * are valid mappings in them. Hence, if a user page is wired, + * the PT page will be also. */ if (wired && !pmap_pte_w(pte)) pmap->pm_stats.wired_count++; @@ -1232,7 +1230,6 @@ pmap_enter(pmap, va, pa, prot, wired) goto validate; } - /* * Mapping has changed, invalidate old range and fall through to * handle validating new mapping. @@ -1240,11 +1237,10 @@ pmap_enter(pmap, va, pa, prot, wired) if (opa) { pmap_remove(pmap, va, va + PAGE_SIZE); } - /* - * Enter on the PV list if part of our managed memory - * Note that we raise IPL while manipulating pv_table - * since pmap_enter can be called at interrupt time. + * Enter on the PV list if part of our managed memory Note that we + * raise IPL while manipulating pv_table since pmap_enter can be + * called at interrupt time. */ if (pmap_is_managed(pa)) { register pv_entry_t pv, npv; @@ -1261,8 +1257,8 @@ pmap_enter(pmap, va, pa, prot, wired) pv->pv_next = NULL; } /* - * There is at least one other VA mapping this page. - * Place this entry after the header. + * There is at least one other VA mapping this page. Place + * this entry after the header. */ else { npv = get_pv_entry(); @@ -1271,13 +1267,12 @@ pmap_enter(pmap, va, pa, prot, wired) npv->pv_next = pv->pv_next; pv->pv_next = npv; } - splx(s); + splx(s); cacheable = 1; } else { cacheable = 0; } - pmap_use_pt(pmap, va); /* * Increment counters @@ -1290,28 +1285,26 @@ validate: /* * Now validate mapping with desired protection/wiring. */ - npte = (pt_entry_t) ( (int) (pa | pte_prot(pmap, prot) | PG_V)); + npte = (pt_entry_t) ((int) (pa | pte_prot(pmap, prot) | PG_V)); /* * for correctness: */ - if( !cacheable) + if (!cacheable) (int) npte |= PG_N; /* - * When forking (copy-on-write, etc): - * A process will turn off write permissions for any of its writable - * pages. If the data (object) is only referred to by one process, the - * processes map is modified directly as opposed to using the - * object manipulation routine. When using pmap_protect, the - * modified bits are not kept in the vm_page_t data structure. - * Therefore, when using pmap_enter in vm_fault to bring back - * writability of a page, there has been no memory of the - * modified or referenced bits except at the pte level. - * this clause supports the carryover of the modified and - * used (referenced) bits. + * When forking (copy-on-write, etc): A process will turn off write + * permissions for any of its writable pages. If the data (object) is + * only referred to by one process, the processes map is modified + * directly as opposed to using the object manipulation routine. When + * using pmap_protect, the modified bits are not kept in the vm_page_t + * data structure. Therefore, when using pmap_enter in vm_fault to + * bring back writability of a page, there has been no memory of the + * modified or referenced bits except at the pte level. this clause + * supports the carryover of the modified and used (referenced) bits. */ if (pa == opa) - (int) npte |= (int) *pte & (PG_M|PG_U); + (int) npte |= (int) *pte & (PG_M | PG_U); if (wired) @@ -1321,13 +1314,16 @@ validate: else if (va < UPT_MAX_ADDRESS) (int) npte |= PG_u | PG_RW; - if(*pte != npte) { + if (*pte != npte) { if (*pte) ptevalid++; *pte = npte; } - if (ptevalid) + if (ptevalid) { pmap_update(); + } else { + pmap_use_pt(pmap, va); + } } /* @@ -1348,11 +1344,11 @@ pmap_qenter(va, m, count) int anyvalid = 0; register pt_entry_t *pte; - for(i=0;ipv_next = NULL; } /* - * There is at least one other VA mapping this page. - * Place this entry after the header. + * There is at least one other VA mapping this page. Place this entry + * after the header. */ else { npv = get_pv_entry(); @@ -1468,9 +1465,7 @@ pmap_enter_quick(pmap, va, pa) npv->pv_next = pv->pv_next; pv->pv_next = npv; } - splx(s); - - pmap_use_pt(pmap, va); + splx(s); /* * Increment counters @@ -1480,11 +1475,14 @@ pmap_enter_quick(pmap, va, pa) /* * Now validate mapping with desired protection/wiring. */ - *pte = (pt_entry_t) ( (int) (pa | PG_V | PG_u)); + *pte = (pt_entry_t) ((int) (pa | PG_V | PG_u)); + + pmap_use_pt(pmap, va); return; } +#define MAX_INIT_PT (1024*2048) /* * pmap_object_init_pt preloads the ptes for a given object * into the specified pmap. This eliminates the blast of soft @@ -1500,66 +1498,152 @@ pmap_object_init_pt(pmap, addr, object, offset, size) { vm_offset_t tmpoff; vm_page_t p; - vm_offset_t objbytes; int bits; + int objbytes; - if (!pmap) + if (!pmap || ((size > MAX_INIT_PT) && + (object->resident_page_count > (MAX_INIT_PT / NBPG)))) { return; - + } if (!vm_object_lock_try(object)) return; /* - * if we are processing a major portion of the object, then - * scan the entire thing. + * if we are processing a major portion of the object, then scan the + * entire thing. */ - if (size > (object->size >> 1)) { + if (size > (object->size >> 2)) { objbytes = size; - p = object->memq.tqh_first; - while ((p != NULL) && (objbytes != 0)) { + + for (p = object->memq.tqh_first; + ((objbytes > 0) && (p != NULL)); + p = p->listq.tqe_next) { + tmpoff = p->offset; - if( tmpoff < offset) { - p = p->listq.tqe_next; + if (tmpoff < offset) { continue; } tmpoff -= offset; - if( tmpoff >= size) { - p = p->listq.tqe_next; + if (tmpoff >= size) { continue; } - if ((p->bmapped == 0) && (p->busy == 0) && - ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && - (p->flags & (PG_BUSY|PG_FICTITIOUS|PG_CACHE)) == 0 ) { + ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (p->flags & (PG_BUSY | PG_FICTITIOUS | PG_CACHE)) == 0) { vm_page_hold(p); - pmap_enter_quick(pmap, addr+tmpoff, VM_PAGE_TO_PHYS(p)); + pmap_enter_quick(pmap, addr + tmpoff, VM_PAGE_TO_PHYS(p)); vm_page_unhold(p); } - p = p->listq.tqe_next; objbytes -= NBPG; } } else { - /* - * else lookup the pages one-by-one. - */ + /* + * else lookup the pages one-by-one. + */ for (tmpoff = 0; tmpoff < size; tmpoff += NBPG) { p = vm_page_lookup(object, tmpoff + offset); - if (p) { - if ((p->bmapped == 0) && - (p->busy == 0) && - ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && - (p->flags & (PG_BUSY|PG_FICTITIOUS|PG_CACHE)) == 0) { - vm_page_hold(p); - pmap_enter_quick(pmap, addr+tmpoff, VM_PAGE_TO_PHYS(p)); - vm_page_unhold(p); - } + if (p && (p->bmapped == 0) && + (p->busy == 0) && + ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (p->flags & (PG_BUSY | PG_FICTITIOUS | PG_CACHE)) == 0) { + vm_page_hold(p); + pmap_enter_quick(pmap, addr + tmpoff, VM_PAGE_TO_PHYS(p)); + vm_page_unhold(p); } } } vm_object_unlock(object); } +/* + * pmap_prefault provides a quick way of clustering + * pagefaults into a processes address space. It is a "cousin" + * of pmap_object_init_pt, except it runs at page fault time instead + * of mmap time. + */ +#define PFBAK 2 +#define PFFOR 2 +#define PAGEORDER_SIZE (PFBAK+PFFOR) + +static int pmap_prefault_pageorder[] = { + -NBPG, NBPG, -2 * NBPG, 2 * NBPG +}; + +void +pmap_prefault(pmap, addra, entry, object) + pmap_t pmap; + vm_offset_t addra; + vm_map_entry_t entry; + vm_object_t object; +{ + int i; + vm_offset_t starta, enda; + vm_offset_t offset, addr; + vm_page_t m; + int pageorder_index; + + if (entry->object.vm_object != object) + return; + + if (pmap != &curproc->p_vmspace->vm_pmap) + return; + + starta = addra - PFBAK * NBPG; + if (starta < entry->start) { + starta = entry->start; + } else if (starta > addra) + starta = 0; + + enda = addra + PFFOR * NBPG; + if (enda > entry->end) + enda = entry->end; + + for (i = 0; i < PAGEORDER_SIZE; i++) { + vm_object_t lobject; + pt_entry_t *pte; + + addr = addra + pmap_prefault_pageorder[i]; + if (addr < starta || addr >= enda) + continue; + + pte = vtopte(addr); + if (*pte) + continue; + + offset = (addr - entry->start) + entry->offset; + lobject = object; + for (m = vm_page_lookup(lobject, offset); + (!m && lobject->shadow); + lobject = lobject->shadow) { + + offset += lobject->shadow_offset; + m = vm_page_lookup(lobject->shadow, offset); + } + + /* + * give-up when a page is not in memory + */ + if (m == NULL) + break; + + if ((m->bmapped == 0) && + (m->busy == 0) && + ((m->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (m->flags & (PG_CACHE | PG_BUSY | PG_FICTITIOUS)) == 0) { + /* + * test results show that the system is faster when + * pages are activated. + */ + if ((m->flags & PG_ACTIVE) == 0) + vm_page_activate(m); + vm_page_hold(m); + pmap_enter_quick(pmap, addr, VM_PAGE_TO_PHYS(m)); + vm_page_unhold(m); + } + } +} + /* * Routine: pmap_change_wiring * Function: Change the wiring attribute for a map/virtual-address @@ -1569,9 +1653,9 @@ pmap_object_init_pt(pmap, addr, object, offset, size) */ void pmap_change_wiring(pmap, va, wired) - register pmap_t pmap; - vm_offset_t va; - boolean_t wired; + register pmap_t pmap; + vm_offset_t va; + boolean_t wired; { register pt_entry_t *pte; @@ -1586,14 +1670,14 @@ pmap_change_wiring(pmap, va, wired) pmap->pm_stats.wired_count--; /* - * Wiring is not a hardware characteristic so there is no need - * to invalidate TLB. + * Wiring is not a hardware characteristic so there is no need to + * invalidate TLB. */ pmap_pte_set_w(pte, wired); /* - * When unwiring, set the modified bit in the pte -- could have - * been changed by the kernel - */ + * When unwiring, set the modified bit in the pte -- could have been + * changed by the kernel + */ if (!wired) (int) *pte |= PG_M; } @@ -1610,9 +1694,9 @@ pmap_change_wiring(pmap, va, wired) void pmap_copy(dst_pmap, src_pmap, dst_addr, len, src_addr) pmap_t dst_pmap, src_pmap; - vm_offset_t dst_addr; - vm_size_t len; - vm_offset_t src_addr; + vm_offset_t dst_addr; + vm_size_t len; + vm_offset_t src_addr; { } @@ -1624,7 +1708,7 @@ pmap_copy(dst_pmap, src_pmap, dst_addr, len, src_addr) pmap_t pmap_kernel() { - return (kernel_pmap); + return (kernel_pmap); } /* @@ -1637,13 +1721,13 @@ void pmap_zero_page(phys) vm_offset_t phys; { - if (*(int *)CMAP2) + if (*(int *) CMAP2) panic("pmap_zero_page: CMAP busy"); - *(int *)CMAP2 = PG_V | PG_KW | i386_trunc_page(phys); - bzero(CADDR2,NBPG); + *(int *) CMAP2 = PG_V | PG_KW | i386_trunc_page(phys); + bzero(CADDR2, NBPG); - *(int *)CMAP2 = 0; + *(int *) CMAP2 = 0; pmap_update(); } @@ -1658,19 +1742,19 @@ pmap_copy_page(src, dst) vm_offset_t src; vm_offset_t dst; { - if (*(int *)CMAP1 || *(int *)CMAP2) + if (*(int *) CMAP1 || *(int *) CMAP2) panic("pmap_copy_page: CMAP busy"); - *(int *)CMAP1 = PG_V | PG_KW | i386_trunc_page(src); - *(int *)CMAP2 = PG_V | PG_KW | i386_trunc_page(dst); + *(int *) CMAP1 = PG_V | PG_KW | i386_trunc_page(src); + *(int *) CMAP2 = PG_V | PG_KW | i386_trunc_page(dst); #if __GNUC__ > 1 memcpy(CADDR2, CADDR1, NBPG); #else - bcopy(CADDR1, CADDR2, NBPG); + bcopy(CADDR1, CADDR2, NBPG); #endif - *(int *)CMAP1 = 0; - *(int *)CMAP2 = 0; + *(int *) CMAP1 = 0; + *(int *) CMAP2 = 0; pmap_update(); } @@ -1691,9 +1775,9 @@ pmap_copy_page(src, dst) */ void pmap_pageable(pmap, sva, eva, pageable) - pmap_t pmap; - vm_offset_t sva, eva; - boolean_t pageable; + pmap_t pmap; + vm_offset_t sva, eva; + boolean_t pageable; { } @@ -1716,8 +1800,7 @@ pmap_page_exists(pmap, pa) s = splhigh(); /* - * Not found, check current mappings returning - * immediately if found. + * Not found, check current mappings returning immediately if found. */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { @@ -1728,7 +1811,7 @@ pmap_page_exists(pmap, pa) } } splx(s); - return(FALSE); + return (FALSE); } /* @@ -1736,7 +1819,7 @@ pmap_page_exists(pmap, pa) * note that the testbit/changebit routines are inline, * and a lot of things compile-time evaluate. */ -boolean_t +__inline boolean_t pmap_testbit(pa, bit) register vm_offset_t pa; int bit; @@ -1752,37 +1835,35 @@ pmap_testbit(pa, bit) s = splhigh(); /* - * Not found, check current mappings returning - * immediately if found. + * Not found, check current mappings returning immediately if found. */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { /* - * if the bit being tested is the modified bit, - * then mark UPAGES as always modified, and - * ptes as never modified. + * if the bit being tested is the modified bit, then + * mark UPAGES as always modified, and ptes as never + * modified. */ - if (bit & PG_U ) { + if (bit & PG_U) { if ((pv->pv_va >= clean_sva) && (pv->pv_va < clean_eva)) { continue; } - } - if (bit & PG_M ) { + } + if (bit & PG_M) { if (pv->pv_va >= USRSTACK) { if (pv->pv_va >= clean_sva && pv->pv_va < clean_eva) { continue; } - if (pv->pv_va < USRSTACK+(UPAGES*NBPG)) { + if (pv->pv_va < USRSTACK + (UPAGES * NBPG)) { splx(s); return TRUE; - } - else if (pv->pv_va < UPT_MAX_ADDRESS) { + } else if (pv->pv_va < UPT_MAX_ADDRESS) { splx(s); return FALSE; } } } - if( !pv->pv_pmap) { + if (!pv->pv_pmap) { printf("Null pmap (tb) at va: 0x%lx\n", pv->pv_va); continue; } @@ -1794,13 +1875,13 @@ pmap_testbit(pa, bit) } } splx(s); - return(FALSE); + return (FALSE); } /* * this routine is used to modify bits in ptes */ -void +__inline void pmap_changebit(pa, bit, setem) vm_offset_t pa; int bit; @@ -1818,8 +1899,8 @@ pmap_changebit(pa, bit, setem) s = splhigh(); /* - * Loop over all current mappings setting/clearing as appropos - * If setting RO do we need to clear the VAC? + * Loop over all current mappings setting/clearing as appropos If + * setting RO do we need to clear the VAC? */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { @@ -1832,8 +1913,7 @@ pmap_changebit(pa, bit, setem) if (va >= clean_sva && va < clean_eva) continue; } - - if( !pv->pv_pmap) { + if (!pv->pv_pmap) { printf("Null pmap (cb) at va: 0x%lx\n", va); continue; } @@ -1856,8 +1936,8 @@ pmap_changebit(pa, bit, setem) */ void pmap_page_protect(phys, prot) - vm_offset_t phys; - vm_prot_t prot; + vm_offset_t phys; + vm_prot_t prot; { if ((prot & VM_PROT_WRITE) == 0) { if (prot & (VM_PROT_READ | VM_PROT_EXECUTE)) @@ -1871,7 +1951,63 @@ vm_offset_t pmap_phys_address(ppn) int ppn; { - return(i386_ptob(ppn)); + return (i386_ptob(ppn)); +} + +/* + * pmap_is_referenced: + * + * Return whether or not the specified physical page was referenced + * by any physical maps. + */ +boolean_t +pmap_is_referenced(vm_offset_t pa) +{ + return pmap_testbit((pa), PG_U); +} + +/* + * pmap_is_modified: + * + * Return whether or not the specified physical page was modified + * in any physical maps. + */ +boolean_t +pmap_is_modified(vm_offset_t pa) +{ + return pmap_testbit((pa), PG_M); +} + +/* + * Clear the modify bits on the specified physical page. + */ +void +pmap_clear_modify(vm_offset_t pa) +{ + pmap_changebit((pa), PG_M, FALSE); +} + +/* + * pmap_clear_reference: + * + * Clear the reference bit on the specified physical page. + */ +void +pmap_clear_reference(vm_offset_t pa) +{ + pmap_changebit((pa), PG_U, FALSE); +} + +/* + * Routine: pmap_copy_on_write + * Function: + * Remove write privileges from all + * physical maps for this physical page. + */ +void +pmap_copy_on_write(vm_offset_t pa) +{ + pmap_changebit((pa), PG_RW, FALSE); } /* @@ -1887,10 +2023,10 @@ i386_protection_init() for (prot = 0; prot < 8; prot++) { switch (prot) { case VM_PROT_NONE | VM_PROT_NONE | VM_PROT_NONE: - /* - * Read access is also 0. There isn't any execute - * bit, so just make it readable. - */ + /* + * Read access is also 0. There isn't any execute bit, + * so just make it readable. + */ case VM_PROT_READ | VM_PROT_NONE | VM_PROT_NONE: case VM_PROT_READ | VM_PROT_NONE | VM_PROT_EXECUTE: case VM_PROT_NONE | VM_PROT_NONE | VM_PROT_EXECUTE: @@ -1937,7 +2073,7 @@ pmap_mapdev(pa, size) } pmap_update(); - return ((void *)va); + return ((void *) va); } #ifdef DEBUG @@ -1949,20 +2085,21 @@ pads(pm) unsigned va, i, j; pt_entry_t *ptep; - if (pm == kernel_pmap) return; - for (i = 0; i < 1024; i++) + if (pm == kernel_pmap) + return; + for (i = 0; i < 1024; i++) if (pm->pm_pdir[i]) - for (j = 0; j < 1024 ; j++) { - va = (i< UPT_MAX_ADDRESS) - continue; + continue; ptep = pmap_pte(pm, va); - if (pmap_pte_v(ptep)) - printf("%x:%x ", va, *(int *)ptep); - } ; - + if (pmap_pte_v(ptep)) + printf("%x:%x ", va, *(int *) ptep); + }; + } void @@ -1975,13 +2112,12 @@ pmap_pvdump(pa) for (pv = pa_to_pvh(pa); pv; pv = pv->pv_next) { #ifdef used_to_be printf(" -> pmap %x, va %x, flags %x", - pv->pv_pmap, pv->pv_va, pv->pv_flags); + pv->pv_pmap, pv->pv_va, pv->pv_flags); #endif printf(" -> pmap %x, va %x", - pv->pv_pmap, pv->pv_va); + pv->pv_pmap, pv->pv_va); pads(pv->pv_pmap); } printf(" "); } #endif - diff --git a/sys/amd64/amd64/trap.c b/sys/amd64/amd64/trap.c index 57d586a5615..a037e615207 100644 --- a/sys/amd64/amd64/trap.c +++ b/sys/amd64/amd64/trap.c @@ -35,7 +35,7 @@ * SUCH DAMAGE. * * from: @(#)trap.c 7.4 (Berkeley) 5/13/91 - * $Id: trap.c,v 1.43 1995/01/09 16:04:39 davidg Exp $ + * $Id: trap.c,v 1.44 1995/01/14 13:20:10 bde Exp $ */ /* @@ -460,23 +460,12 @@ trap_pfault(frame, usermode) /* Fault the pte only if needed: */ *(volatile char *)v += 0; - ptepg = (vm_page_t) pmap_pte_vm_page(vm_map_pmap(map), v); - vm_page_hold(ptepg); + pmap_use_pt( vm_map_pmap(map), va); /* Fault in the user page: */ rv = vm_fault(map, va, ftype, FALSE); - vm_page_unhold(ptepg); - - /* - * page table pages don't need to be kept if they - * are not held - */ - if( ptepg->hold_count == 0 && ptepg->wire_count == 0) { - pmap_page_protect( VM_PAGE_TO_PHYS(ptepg), - VM_PROT_NONE); - vm_page_free(ptepg); - } + pmap_unuse_pt( vm_map_pmap(map), va); --p->p_lock; } else { diff --git a/sys/i386/i386/machdep.c b/sys/i386/i386/machdep.c index 9e041fc6aa1..bdbbadf120d 100644 --- a/sys/i386/i386/machdep.c +++ b/sys/i386/i386/machdep.c @@ -35,7 +35,7 @@ * SUCH DAMAGE. * * from: @(#)machdep.c 7.4 (Berkeley) 6/3/91 - * $Id: machdep.c,v 1.101 1995/01/14 13:20:08 bde Exp $ + * $Id: machdep.c,v 1.102 1995/01/17 01:15:12 bde Exp $ */ #include "npx.h" @@ -114,7 +114,7 @@ char cpu_model[sizeof("Cy486DLC") + 1]; /* * Declare these as initialized data so we can patch them. */ -int nswbuf = 128; +int nswbuf = 0; #ifdef NBUF int nbuf = NBUF; #else @@ -255,9 +255,12 @@ again: valloc(msqids, struct msqid_ds, msginfo.msgmni); #endif - if (nbuf == 0) - nbuf = min(physmem / 30, 256); - nswbuf = nbuf; + if (nbuf == 0) { + nbuf = 32; + if( physmem > 1024) + nbuf += min((physmem - 1024) / 20, 1024); + } + nswbuf = min(nbuf, 128); valloc(swbuf, struct buf, nswbuf); valloc(buf, struct buf, nbuf); diff --git a/sys/i386/i386/pmap.c b/sys/i386/i386/pmap.c index eb72d020db4..5febdd0895b 100644 --- a/sys/i386/i386/pmap.c +++ b/sys/i386/i386/pmap.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (c) 1991 Regents of the University of California. * All rights reserved. * Copyright (c) 1994 John S. Dyson @@ -39,7 +39,7 @@ * SUCH DAMAGE. * * from: @(#)pmap.c 7.7 (Berkeley) 5/12/91 - * $Id: pmap.c,v 1.41 1995/01/14 04:58:53 davidg Exp $ + * $Id: pmap.c,v 1.42 1995/01/15 09:06:23 davidg Exp $ */ /* @@ -53,10 +53,10 @@ */ /* * Major modifications by John S. Dyson primarily to support - * pageable page tables, eliminating pmap_attributes, + * pageable page tables, eliminating pmap_attributes, * discontiguous memory pages, and using more efficient string * instructions. Jan 13, 1994. Further modifications on Mar 2, 1994, - * general clean-up and efficiency mods. + * general clean-up and efficiency mods. */ /* @@ -118,7 +118,7 @@ #define pmap_pte_m(pte) ((*(int *)pte & PG_M) != 0) #define pmap_pte_u(pte) ((*(int *)pte & PG_U) != 0) #define pmap_pte_v(pte) ((*(int *)pte & PG_V) != 0) - + #define pmap_pte_set_w(pte, v) ((v)?(*(int *)pte |= PG_W):(*(int *)pte &= ~PG_W)) #define pmap_pte_set_prot(pte, v) ((*(int *)pte &= ~PG_PROT), (*(int *)pte |= (v))) @@ -127,29 +127,29 @@ * convert to a vax protection code. */ #define pte_prot(m, p) (protection_codes[p]) -int protection_codes[8]; +int protection_codes[8]; -struct pmap kernel_pmap_store; -pmap_t kernel_pmap; +struct pmap kernel_pmap_store; +pmap_t kernel_pmap; -vm_offset_t phys_avail[6]; /* 2 entries + 1 null */ -vm_offset_t avail_start; /* PA of first available physical page */ -vm_offset_t avail_end; /* PA of last available physical page */ -vm_size_t mem_size; /* memory size in bytes */ -vm_offset_t virtual_avail; /* VA of first avail page (after kernel bss)*/ -vm_offset_t virtual_end; /* VA of last avail page (end of kernel AS) */ -int i386pagesperpage; /* PAGE_SIZE / I386_PAGE_SIZE */ -boolean_t pmap_initialized = FALSE; /* Has pmap_init completed? */ -vm_offset_t vm_first_phys, vm_last_phys; +vm_offset_t phys_avail[6]; /* 2 entries + 1 null */ +vm_offset_t avail_start; /* PA of first available physical page */ +vm_offset_t avail_end; /* PA of last available physical page */ +vm_size_t mem_size; /* memory size in bytes */ +vm_offset_t virtual_avail; /* VA of first avail page (after kernel bss) */ +vm_offset_t virtual_end; /* VA of last avail page (end of kernel AS) */ +int i386pagesperpage; /* PAGE_SIZE / I386_PAGE_SIZE */ +boolean_t pmap_initialized = FALSE; /* Has pmap_init completed? */ +vm_offset_t vm_first_phys, vm_last_phys; -static inline int pmap_is_managed(); -static inline void * vm_get_pmap(); -static inline void vm_put_pmap(); -static void i386_protection_init(); -static void pmap_alloc_pv_entry(); -static inline pv_entry_t get_pv_entry(); -static inline void pmap_use_pt(); -static inline void pmap_unuse_pt(); +static inline int pmap_is_managed(); +static inline void *vm_get_pmap(); +static inline void vm_put_pmap(); +static void i386_protection_init(); +static void pmap_alloc_pv_entry(); +static inline pv_entry_t get_pv_entry(); +inline void pmap_use_pt(); +inline void pmap_unuse_pt(); int nkpt; @@ -163,12 +163,14 @@ extern int cpu_class; * All those kernel PT submaps that BSD is so fond of */ pt_entry_t *CMAP1, *CMAP2, *ptmmap; -caddr_t CADDR1, CADDR2, ptvmmap; +caddr_t CADDR1, CADDR2, ptvmmap; pt_entry_t *msgbufmap; -struct msgbuf *msgbufp; +struct msgbuf *msgbufp; + #endif -void init_pv_entries(int) ; +void +init_pv_entries(int); /* * Routine: pmap_pte @@ -178,27 +180,28 @@ void init_pv_entries(int) ; * [ what about induced faults -wfj] */ -inline pt_entry_t * -const pmap_pte(pmap, va) - register pmap_t pmap; +inline pt_entry_t * const +pmap_pte(pmap, va) + register pmap_t pmap; vm_offset_t va; { if (pmap && *pmap_pde(pmap, va)) { vm_offset_t frame = (int) pmap->pm_pdir[PTDPTDI] & PG_FRAME; + /* are we current address space or kernel? */ - if ( (pmap == kernel_pmap) || (frame == ((int) PTDpde & PG_FRAME))) + if ((pmap == kernel_pmap) || (frame == ((int) PTDpde & PG_FRAME))) return ((pt_entry_t *) vtopte(va)); /* otherwise, we are alternate address space */ else { - if ( frame != ((int) APTDpde & PG_FRAME) ) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } - return((pt_entry_t *) avtopte(va)); + return ((pt_entry_t *) avtopte(va)); } } - return(0); + return (0); } /* @@ -210,20 +213,21 @@ const pmap_pte(pmap, va) vm_offset_t pmap_extract(pmap, va) - register pmap_t pmap; + register pmap_t pmap; vm_offset_t va; { vm_offset_t pa; if (pmap && *pmap_pde(pmap, va)) { vm_offset_t frame = (int) pmap->pm_pdir[PTDPTDI] & PG_FRAME; + /* are we current address space or kernel? */ - if ( (pmap == kernel_pmap) - || (frame == ((int) PTDpde & PG_FRAME)) ) { + if ((pmap == kernel_pmap) + || (frame == ((int) PTDpde & PG_FRAME))) { pa = *(int *) vtopte(va); - /* otherwise, we are alternate address space */ + /* otherwise, we are alternate address space */ } else { - if ( frame != ((int) APTDpde & PG_FRAME)) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } @@ -232,7 +236,7 @@ pmap_extract(pmap, va) return ((pa & PG_FRAME) | (va & ~PG_FRAME)); } return 0; - + } /* @@ -262,10 +266,13 @@ pmap_pte_vm_page(pmap, pt) pmap_t pmap; vm_offset_t pt; { - pt = i386_trunc_page( pt); + vm_page_t m; + + pt = i386_trunc_page(pt); pt = (pt - UPT_MIN_ADDRESS) / NBPG; pt = ((vm_offset_t) pmap->pm_pdir[pt]) & PG_FRAME; - return PHYS_TO_VM_PAGE(pt); + m = PHYS_TO_VM_PAGE(pt); + return m; } /* @@ -278,11 +285,11 @@ pmap_use_pt(pmap, va) { vm_offset_t pt; - if (va >= VM_MAX_ADDRESS || !pmap_initialized) - return; + if ((va >= UPT_MIN_ADDRESS) || !pmap_initialized) + return; pt = (vm_offset_t) vtopte(va); - vm_page_hold( pmap_pte_vm_page(pmap, pt)); + vm_page_hold(pmap_pte_vm_page(pmap, pt)); } /* @@ -294,12 +301,20 @@ pmap_unuse_pt(pmap, va) vm_offset_t va; { vm_offset_t pt; + vm_page_t m; - if (va >= VM_MAX_ADDRESS || !pmap_initialized) - return; + if ((va >= UPT_MIN_ADDRESS) || !pmap_initialized) + return; pt = (vm_offset_t) vtopte(va); - vm_page_unhold( pmap_pte_vm_page(pmap, pt)); + m = pmap_pte_vm_page(pmap, pt); + vm_page_unhold(m); + if (pmap != kernel_pmap && + (m->hold_count == 0) && + (m->wire_count == 0) && + (va < KPT_MIN_ADDRESS)) { + vm_page_deactivate(m); + } } /* [ macro again?, should I force kstack into user map here? -wfj ] */ @@ -332,9 +347,10 @@ pmap_bootstrap(firstaddr, loadaddr) #if BSDVM_COMPAT vm_offset_t va; pt_entry_t *pte; + #endif - avail_start = firstaddr + DMAPAGES*NBPG; + avail_start = firstaddr + DMAPAGES * NBPG; virtual_avail = (vm_offset_t) KERNBASE + avail_start; virtual_end = VM_MAX_KERNEL_ADDRESS; @@ -346,13 +362,13 @@ pmap_bootstrap(firstaddr, loadaddr) i386_protection_init(); /* - * The kernel's pmap is statically allocated so we don't - * have to use pmap_create, which is unlikely to work - * correctly at this part of the boot sequence. + * The kernel's pmap is statically allocated so we don't have to use + * pmap_create, which is unlikely to work correctly at this part of + * the boot sequence. */ kernel_pmap = &kernel_pmap_store; - kernel_pmap->pm_pdir = (pd_entry_t *)(KERNBASE + IdlePTD); + kernel_pmap->pm_pdir = (pd_entry_t *) (KERNBASE + IdlePTD); simple_lock_init(&kernel_pmap->pm_lock); kernel_pmap->pm_count = 1; @@ -368,11 +384,11 @@ pmap_bootstrap(firstaddr, loadaddr) va = virtual_avail; pte = pmap_pte(kernel_pmap, va); - SYSMAP(caddr_t ,CMAP1 ,CADDR1 ,1 ) - SYSMAP(caddr_t ,CMAP2 ,CADDR2 ,1 ) - SYSMAP(caddr_t ,ptmmap ,ptvmmap ,1 ) - SYSMAP(struct msgbuf * ,msgbufmap ,msgbufp ,1 ) - virtual_avail = va; + SYSMAP(caddr_t, CMAP1, CADDR1, 1) + SYSMAP(caddr_t, CMAP2, CADDR2, 1) + SYSMAP(caddr_t, ptmmap, ptvmmap, 1) + SYSMAP(struct msgbuf *, msgbufmap, msgbufp, 1) + virtual_avail = va; #endif /* * Reserve special hunk of memory for use by bus dma as a bounce @@ -380,13 +396,14 @@ pmap_bootstrap(firstaddr, loadaddr) */ { extern vm_offset_t isaphysmem; + isaphysmem = va; virtual_avail = pmap_map(va, firstaddr, - firstaddr + DMAPAGES*NBPG, VM_PROT_ALL); + firstaddr + DMAPAGES * NBPG, VM_PROT_ALL); } - *(int *)CMAP1 = *(int *)CMAP2 = *(int *)PTD = 0; + *(int *) CMAP1 = *(int *) CMAP2 = *(int *) PTD = 0; pmap_update(); } @@ -400,30 +417,30 @@ pmap_bootstrap(firstaddr, loadaddr) */ void pmap_init(phys_start, phys_end) - vm_offset_t phys_start, phys_end; + vm_offset_t phys_start, phys_end; { - vm_offset_t addr; - vm_size_t npg, s; + vm_offset_t addr; + vm_size_t npg, s; int i; /* - * Now that kernel map has been allocated, we can mark as - * unavailable regions which we have mapped in locore. + * Now that kernel map has been allocated, we can mark as unavailable + * regions which we have mapped in locore. */ addr = atdevbase; (void) vm_map_find(kernel_map, NULL, (vm_offset_t) 0, - &addr, (0x100000-0xa0000), FALSE); + &addr, (0x100000 - 0xa0000), FALSE); addr = (vm_offset_t) KERNBASE + IdlePTD; vm_object_reference(kernel_object); (void) vm_map_find(kernel_map, kernel_object, addr, - &addr, (4 + NKPDE) * NBPG, FALSE); + &addr, (4 + NKPDE) * NBPG, FALSE); /* * calculate the number of pv_entries needed */ vm_first_phys = phys_avail[0]; - for (i = 0; phys_avail[i + 1]; i += 2) ; + for (i = 0; phys_avail[i + 1]; i += 2); npg = (phys_avail[(i - 2) + 1] - vm_first_phys) / NBPG; /* @@ -454,17 +471,17 @@ pmap_init(phys_start, phys_end) */ vm_offset_t pmap_map(virt, start, end, prot) - vm_offset_t virt; - vm_offset_t start; - vm_offset_t end; - int prot; + vm_offset_t virt; + vm_offset_t start; + vm_offset_t end; + int prot; { while (start < end) { pmap_enter(kernel_pmap, virt, start, prot, FALSE); virt += PAGE_SIZE; start += PAGE_SIZE; } - return(virt); + return (virt); } /* @@ -485,7 +502,7 @@ pmap_map(virt, start, end, prot) pmap_t pmap_create(size) - vm_size_t size; + vm_size_t size; { register pmap_t pmap; @@ -493,7 +510,7 @@ pmap_create(size) * Software use map does not need a pmap */ if (size) - return(NULL); + return (NULL); pmap = (pmap_t) malloc(sizeof *pmap, M_VMPMAP, M_WAITOK); bzero(pmap, sizeof(*pmap)); @@ -511,7 +528,7 @@ vm_get_pmap() { struct pmaplist *rtval; - rtval = (struct pmaplist *)kmem_alloc(kernel_map, ctob(1)); + rtval = (struct pmaplist *) kmem_alloc(kernel_map, ctob(1)); bzero(rtval, ctob(1)); return rtval; } @@ -520,7 +537,7 @@ static inline void vm_put_pmap(up) struct pmaplist *up; { - kmem_free(kernel_map, (vm_offset_t)up, ctob(1)); + kmem_free(kernel_map, (vm_offset_t) up, ctob(1)); } /* @@ -532,17 +549,17 @@ pmap_pinit(pmap) register struct pmap *pmap; { /* - * No need to allocate page table space yet but we do need a - * valid page directory table. + * No need to allocate page table space yet but we do need a valid + * page directory table. */ pmap->pm_pdir = (pd_entry_t *) vm_get_pmap(); /* wire in kernel global address entries */ - bcopy(PTD+KPTDI, pmap->pm_pdir+KPTDI, nkpt*PTESIZE); + bcopy(PTD + KPTDI, pmap->pm_pdir + KPTDI, nkpt * PTESIZE); /* install self-referential address mapping entry */ - *(int *)(pmap->pm_pdir+PTDPTDI) = - ((int)pmap_kextract((vm_offset_t)pmap->pm_pdir)) | PG_V | PG_KW; + *(int *) (pmap->pm_pdir + PTDPTDI) = + ((int) pmap_kextract((vm_offset_t) pmap->pm_pdir)) | PG_V | PG_KW; pmap->pm_count = 1; simple_lock_init(&pmap->pm_lock); @@ -556,54 +573,47 @@ vm_page_t nkpg; vm_offset_t kernel_vm_end; void -pmap_growkernel(vm_offset_t addr) { +pmap_growkernel(vm_offset_t addr) +{ struct proc *p; struct pmap *pmap; int s; - + s = splhigh(); if (kernel_vm_end == 0) { kernel_vm_end = KERNBASE; nkpt = 0; - while(pdir_pde(PTD, kernel_vm_end)) { - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + while (pdir_pde(PTD, kernel_vm_end)) { + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); ++nkpt; } } - - addr = (addr + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); - while( kernel_vm_end < addr) { - if( pdir_pde( PTD, kernel_vm_end)) { - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + addr = (addr + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); + while (kernel_vm_end < addr) { + if (pdir_pde(PTD, kernel_vm_end)) { + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); continue; } - ++nkpt; - if( !nkpg) { - nkpg = vm_page_alloc(kernel_object, 0, TRUE); + if (!nkpg) { + nkpg = vm_page_alloc(kernel_object, 0, VM_ALLOC_SYSTEM); + if (!nkpg) + panic("pmap_growkernel: no memory to grow kernel"); + vm_page_wire(nkpg); vm_page_remove(nkpg); pmap_zero_page(VM_PAGE_TO_PHYS(nkpg)); - if( !nkpg) - panic("pmap_growkernel: no memory to grow kernel"); } - pdir_pde( PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_KW); + pdir_pde(PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_KW); nkpg = NULL; - for (p = (struct proc *)allproc; p != NULL; p = p->p_next) { - if( p->p_vmspace) { + for (p = (struct proc *) allproc; p != NULL; p = p->p_next) { + if (p->p_vmspace) { pmap = &p->p_vmspace->vm_pmap; - *pmap_pde( pmap, kernel_vm_end) = pdir_pde( PTD, kernel_vm_end); + *pmap_pde(pmap, kernel_vm_end) = pdir_pde(PTD, kernel_vm_end); } } - kernel_vm_end = (kernel_vm_end + NBPG*NPTEPG) & ~(NBPG*NPTEPG-1); + kernel_vm_end = (kernel_vm_end + NBPG * NPTEPG) & ~(NBPG * NPTEPG - 1); } -#if 0 - if( !nkpg) { - nkpg = vm_page_alloc(kernel_object, 0, TRUE); - vm_page_remove(nkpg); - pmap_zero_page(VM_PAGE_TO_PHYS(nkpg)); - } -#endif splx(s); } @@ -626,7 +636,7 @@ pmap_destroy(pmap) simple_unlock(&pmap->pm_lock); if (count == 0) { pmap_release(pmap); - free((caddr_t)pmap, M_VMPMAP); + free((caddr_t) pmap, M_VMPMAP); } } @@ -647,7 +657,7 @@ pmap_release(pmap) */ void pmap_reference(pmap) - pmap_t pmap; + pmap_t pmap; { if (pmap != NULL) { simple_lock(&pmap->pm_lock); @@ -673,7 +683,8 @@ inline static void free_pv_entry(pv) pv_entry_t pv; { - if (!pv) return; + if (!pv) + return; ++pv_freelistcnt; pv->pv_next = pv_freelist; pv_freelist = pv; @@ -693,19 +704,15 @@ get_pv_entry() /* * get more pv_entry pages if needed */ - while (pv_freelistcnt < PV_FREELIST_MIN || pv_freelist == 0) { + if (pv_freelistcnt < PV_FREELIST_MIN || pv_freelist == 0) { pmap_alloc_pv_entry(); } - /* * get a pv_entry off of the free list */ --pv_freelistcnt; tmp = pv_freelist; pv_freelist = tmp->pv_next; - tmp->pv_pmap = 0; - tmp->pv_va = 0; - tmp->pv_next = 0; return tmp; } @@ -722,6 +729,7 @@ pmap_alloc_pv_entry() */ if (npvvapg) { vm_page_t m; + /* * we do this to keep recursion away */ @@ -729,12 +737,14 @@ pmap_alloc_pv_entry() /* * allocate a physical page out of the vm system */ - m = vm_page_alloc(kernel_object, pvva-vm_map_min(kernel_map), TRUE); + m = vm_page_alloc(kernel_object, + pvva - vm_map_min(kernel_map), VM_ALLOC_INTERRUPT); if (m) { int newentries; int i; pv_entry_t entry; - newentries = (NBPG/sizeof (struct pv_entry)); + + newentries = (NBPG / sizeof(struct pv_entry)); /* * wire the page */ @@ -778,10 +788,10 @@ init_pv_entries(npg) { /* * allocate enough kvm space for PVSPERPAGE entries per page (lots) - * kvm space is fairly cheap, be generous!!! (the system can panic - * if this is too small.) + * kvm space is fairly cheap, be generous!!! (the system can panic if + * this is too small.) */ - npvvapg = ((npg*PVSPERPAGE) * sizeof(struct pv_entry) + NBPG - 1)/NBPG; + npvvapg = ((npg * PVSPERPAGE) * sizeof(struct pv_entry) + NBPG - 1) / NBPG; pvva = kmem_alloc_pageable(kernel_map, npvvapg * NBPG); /* * get the first batch of entries @@ -800,7 +810,7 @@ get_pt_entry(pmap) return PTmap; } /* otherwise, we are alternate address space */ - if ( frame != ((int) APTDpde & PG_FRAME)) { + if (frame != ((int) APTDpde & PG_FRAME)) { APTDpde = pmap->pm_pdir[PTDPTDI]; pmap_update(); } @@ -821,6 +831,7 @@ pmap_remove_entry(pmap, pv, va) { pv_entry_t npv; int s; + s = splhigh(); if (pmap == pv->pv_pmap && va == pv->pv_va) { npv = pv->pv_next; @@ -840,7 +851,7 @@ pmap_remove_entry(pmap, pv, va) if (npv) { pv->pv_next = npv->pv_next; free_pv_entry(npv); - } + } } splx(s); } @@ -857,7 +868,7 @@ pmap_remove(pmap, sva, eva) register vm_offset_t sva; register vm_offset_t eva; { - register pt_entry_t *ptp,*ptq; + register pt_entry_t *ptp, *ptq; vm_offset_t pa; register pv_entry_t pv; vm_offset_t va; @@ -869,19 +880,19 @@ pmap_remove(pmap, sva, eva) ptp = get_pt_entry(pmap); -/* - * special handling of removing one page. a very - * common operation and easy to short circuit some - * code. - */ - if( (sva + NBPG) == eva) { - - if( *pmap_pde( pmap, sva) == 0) + /* + * special handling of removing one page. a very + * common operation and easy to short circuit some + * code. + */ + if ((sva + NBPG) == eva) { + + if (*pmap_pde(pmap, sva) == 0) return; ptq = ptp + i386_btop(sva); - if( !*ptq) + if (!*ptq) return; /* * Update statistics @@ -897,46 +908,41 @@ pmap_remove(pmap, sva, eva) if (pmap_is_managed(pa)) { if ((int) oldpte & PG_M) { if ((sva < USRSTACK || sva > UPT_MAX_ADDRESS) || - (sva >= USRSTACK && sva < USRSTACK+(UPAGES*NBPG))) { + (sva >= USRSTACK && sva < USRSTACK + (UPAGES * NBPG))) { if (sva < clean_sva || sva >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - pv = pa_to_pvh(pa); pmap_remove_entry(pmap, pv, sva); } - pmap_unuse_pt(pmap, sva); + pmap_unuse_pt(pmap, sva); pmap_update(); return; } - sva = i386_btop(sva); eva = i386_btop(eva); while (sva < eva) { /* - * Weed out invalid mappings. - * Note: we assume that the page directory table is - * always allocated, and in kernel virtual. + * Weed out invalid mappings. Note: we assume that the page + * directory table is always allocated, and in kernel virtual. */ - if ( *pmap_pde(pmap, i386_ptob(sva)) == 0 ) { + if (*pmap_pde(pmap, i386_ptob(sva)) == 0) { /* We can race ahead here, straight to next pde.. */ sva = ((sva + NPTEPG) & ~(NPTEPG - 1)); continue; } - ptq = ptp + sva; /* - * search for page table entries, use string operations - * that are much faster than - * explicitly scanning when page tables are not fully - * populated. + * search for page table entries, use string operations that + * are much faster than explicitly scanning when page tables + * are not fully populated. */ - if ( *ptq == 0) { + if (*ptq == 0) { vm_offset_t pdnxt = ((sva + NPTEPG) & ~(NPTEPG - 1)); vm_offset_t nscan = pdnxt - sva; int found = 0; @@ -944,12 +950,10 @@ pmap_remove(pmap, sva, eva) if ((nscan + sva) > eva) nscan = eva - sva; - asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" - :"=D"(ptq),"=a"(found) - :"c"(nscan),"0"(ptq) - :"cx"); + asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" : + "=D"(ptq), "=a"(found) : "c"(nscan), "0"(ptq) : "cx"); - if( !found) { + if (!found) { sva = pdnxt; continue; } @@ -957,47 +961,43 @@ pmap_remove(pmap, sva, eva) sva = ptq - ptp; } - /* * Update statistics */ oldpte = *ptq; - if (((int)oldpte) & PG_W) + if (((int) oldpte) & PG_W) pmap->pm_stats.wired_count--; pmap->pm_stats.resident_count--; /* - * Invalidate the PTEs. - * XXX: should cluster them up and invalidate as many - * as possible at once. + * Invalidate the PTEs. XXX: should cluster them up and + * invalidate as many as possible at once. */ *ptq = 0; va = i386_ptob(sva); /* - * Remove from the PV table (raise IPL since we - * may be called at interrupt time). + * Remove from the PV table (raise IPL since we may be called + * at interrupt time). */ - pa = ((int)oldpte) & PG_FRAME; + pa = ((int) oldpte) & PG_FRAME; if (!pmap_is_managed(pa)) { - pmap_unuse_pt(pmap, va); + pmap_unuse_pt(pmap, va); ++sva; continue; } - if ((int) oldpte & PG_M) { if ((va < USRSTACK || va > UPT_MAX_ADDRESS) || - (va >= USRSTACK && va < USRSTACK+(UPAGES*NBPG))) { + (va >= USRSTACK && va < USRSTACK + (UPAGES * NBPG))) { if (va < clean_sva || va >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - pv = pa_to_pvh(pa); pmap_remove_entry(pmap, pv, va); - pmap_unuse_pt(pmap, va); + pmap_unuse_pt(pmap, va); ++sva; } pmap_update(); @@ -1030,7 +1030,10 @@ pmap_remove_all(pa) /* * Not one of ours */ - /* XXX this makes pmap_page_protect(NONE) illegal for non-managed pages! */ + /* + * XXX this makes pmap_page_protect(NONE) illegal for non-managed + * pages! + */ if (!pmap_is_managed(pa)) return; @@ -1049,23 +1052,21 @@ pmap_remove_all(pa) if (*pte) { pmap->pm_stats.resident_count--; anyvalid++; - + /* * Update the vm_page_t clean and reference bits. */ if ((int) *pte & PG_M) { if ((va < USRSTACK || va > UPT_MAX_ADDRESS) || - (va >= USRSTACK && va < USRSTACK+(UPAGES*NBPG))) { + (va >= USRSTACK && va < USRSTACK + (UPAGES * NBPG))) { if (va < clean_sva || va >= clean_eva) { PHYS_TO_VM_PAGE(pa)->dirty |= VM_PAGE_BITS_ALL; } } } - *pte = 0; pmap_unuse_pt(pmap, va); } - npv = pv->pv_next; if (npv) { *pv = *npv; @@ -1086,9 +1087,9 @@ pmap_remove_all(pa) */ void pmap_protect(pmap, sva, eva, prot) - register pmap_t pmap; - vm_offset_t sva, eva; - vm_prot_t prot; + register pmap_t pmap; + vm_offset_t sva, eva; + vm_prot_t prot; { register pt_entry_t *pte; register vm_offset_t va; @@ -1111,29 +1112,29 @@ pmap_protect(pmap, sva, eva, prot) va = sva; while (va < eva) { - int found=0; + int found = 0; int svap; vm_offset_t nscan; + /* - * Page table page is not allocated. - * Skip it, we don't want to force allocation - * of unnecessary PTE pages just to set the protection. + * Page table page is not allocated. Skip it, we don't want to + * force allocation of unnecessary PTE pages just to set the + * protection. */ - if (! *pmap_pde(pmap, va)) { + if (!*pmap_pde(pmap, va)) { /* XXX: avoid address wrap around */ -nextpde: - if (va >= i386_trunc_pdr((vm_offset_t)-1)) + nextpde: + if (va >= i386_trunc_pdr((vm_offset_t) - 1)) break; va = i386_round_pdr(va + PAGE_SIZE); continue; } - pte = ptp + i386_btop(va); - if( *pte == 0) { - /* - * scan for a non-empty pte - */ + if (*pte == 0) { + /* + * scan for a non-empty pte + */ svap = pte - ptp; nscan = ((svap + NPTEPG) & ~(NPTEPG - 1)) - svap; @@ -1142,11 +1143,10 @@ nextpde: found = 0; if (nscan) - asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" - :"=D"(pte),"=a"(found) - :"c"(nscan),"0"(pte):"cx"); + asm("xorl %%eax,%%eax;cld;repe;scasl;jz 1f;incl %%eax;1:;" : + "=D"(pte), "=a"(found) : "c"(nscan), "0"(pte) : "cx"); - if( !found) + if (!found) goto nextpde; pte -= 1; @@ -1154,13 +1154,12 @@ nextpde: va = i386_ptob(svap); } - anyvalid++; i386prot = pte_prot(pmap, prot); if (va < UPT_MAX_ADDRESS) { i386prot |= PG_u; - if( va >= UPT_MIN_ADDRESS) + if (va >= UPT_MIN_ADDRESS) i386prot |= PG_RW; } pmap_pte_set_prot(pte, i386prot); @@ -1193,7 +1192,7 @@ pmap_enter(pmap, va, pa, prot, wired) register pt_entry_t *pte; register pt_entry_t npte; vm_offset_t opa; - int cacheable=1; + int cacheable = 1; int ptevalid = 0; if (pmap == NULL) @@ -1211,7 +1210,6 @@ pmap_enter(pmap, va, pa, prot, wired) printf("kernel page directory invalid pdir=0x%x, va=0x%x\n", pmap->pm_pdir[PTDPTDI], va); panic("invalid kernel page directory"); } - pte = pmap_pte(pmap, va); opa = pmap_pte_pa(pte); @@ -1220,10 +1218,10 @@ pmap_enter(pmap, va, pa, prot, wired) */ if (opa == pa) { /* - * Wiring change, just update stats. - * We don't worry about wiring PT pages as they remain - * resident as long as there are valid mappings in them. - * Hence, if a user page is wired, the PT page will be also. + * Wiring change, just update stats. We don't worry about + * wiring PT pages as they remain resident as long as there + * are valid mappings in them. Hence, if a user page is wired, + * the PT page will be also. */ if (wired && !pmap_pte_w(pte)) pmap->pm_stats.wired_count++; @@ -1232,7 +1230,6 @@ pmap_enter(pmap, va, pa, prot, wired) goto validate; } - /* * Mapping has changed, invalidate old range and fall through to * handle validating new mapping. @@ -1240,11 +1237,10 @@ pmap_enter(pmap, va, pa, prot, wired) if (opa) { pmap_remove(pmap, va, va + PAGE_SIZE); } - /* - * Enter on the PV list if part of our managed memory - * Note that we raise IPL while manipulating pv_table - * since pmap_enter can be called at interrupt time. + * Enter on the PV list if part of our managed memory Note that we + * raise IPL while manipulating pv_table since pmap_enter can be + * called at interrupt time. */ if (pmap_is_managed(pa)) { register pv_entry_t pv, npv; @@ -1261,8 +1257,8 @@ pmap_enter(pmap, va, pa, prot, wired) pv->pv_next = NULL; } /* - * There is at least one other VA mapping this page. - * Place this entry after the header. + * There is at least one other VA mapping this page. Place + * this entry after the header. */ else { npv = get_pv_entry(); @@ -1271,13 +1267,12 @@ pmap_enter(pmap, va, pa, prot, wired) npv->pv_next = pv->pv_next; pv->pv_next = npv; } - splx(s); + splx(s); cacheable = 1; } else { cacheable = 0; } - pmap_use_pt(pmap, va); /* * Increment counters @@ -1290,28 +1285,26 @@ validate: /* * Now validate mapping with desired protection/wiring. */ - npte = (pt_entry_t) ( (int) (pa | pte_prot(pmap, prot) | PG_V)); + npte = (pt_entry_t) ((int) (pa | pte_prot(pmap, prot) | PG_V)); /* * for correctness: */ - if( !cacheable) + if (!cacheable) (int) npte |= PG_N; /* - * When forking (copy-on-write, etc): - * A process will turn off write permissions for any of its writable - * pages. If the data (object) is only referred to by one process, the - * processes map is modified directly as opposed to using the - * object manipulation routine. When using pmap_protect, the - * modified bits are not kept in the vm_page_t data structure. - * Therefore, when using pmap_enter in vm_fault to bring back - * writability of a page, there has been no memory of the - * modified or referenced bits except at the pte level. - * this clause supports the carryover of the modified and - * used (referenced) bits. + * When forking (copy-on-write, etc): A process will turn off write + * permissions for any of its writable pages. If the data (object) is + * only referred to by one process, the processes map is modified + * directly as opposed to using the object manipulation routine. When + * using pmap_protect, the modified bits are not kept in the vm_page_t + * data structure. Therefore, when using pmap_enter in vm_fault to + * bring back writability of a page, there has been no memory of the + * modified or referenced bits except at the pte level. this clause + * supports the carryover of the modified and used (referenced) bits. */ if (pa == opa) - (int) npte |= (int) *pte & (PG_M|PG_U); + (int) npte |= (int) *pte & (PG_M | PG_U); if (wired) @@ -1321,13 +1314,16 @@ validate: else if (va < UPT_MAX_ADDRESS) (int) npte |= PG_u | PG_RW; - if(*pte != npte) { + if (*pte != npte) { if (*pte) ptevalid++; *pte = npte; } - if (ptevalid) + if (ptevalid) { pmap_update(); + } else { + pmap_use_pt(pmap, va); + } } /* @@ -1348,11 +1344,11 @@ pmap_qenter(va, m, count) int anyvalid = 0; register pt_entry_t *pte; - for(i=0;ipv_next = NULL; } /* - * There is at least one other VA mapping this page. - * Place this entry after the header. + * There is at least one other VA mapping this page. Place this entry + * after the header. */ else { npv = get_pv_entry(); @@ -1468,9 +1465,7 @@ pmap_enter_quick(pmap, va, pa) npv->pv_next = pv->pv_next; pv->pv_next = npv; } - splx(s); - - pmap_use_pt(pmap, va); + splx(s); /* * Increment counters @@ -1480,11 +1475,14 @@ pmap_enter_quick(pmap, va, pa) /* * Now validate mapping with desired protection/wiring. */ - *pte = (pt_entry_t) ( (int) (pa | PG_V | PG_u)); + *pte = (pt_entry_t) ((int) (pa | PG_V | PG_u)); + + pmap_use_pt(pmap, va); return; } +#define MAX_INIT_PT (1024*2048) /* * pmap_object_init_pt preloads the ptes for a given object * into the specified pmap. This eliminates the blast of soft @@ -1500,66 +1498,152 @@ pmap_object_init_pt(pmap, addr, object, offset, size) { vm_offset_t tmpoff; vm_page_t p; - vm_offset_t objbytes; int bits; + int objbytes; - if (!pmap) + if (!pmap || ((size > MAX_INIT_PT) && + (object->resident_page_count > (MAX_INIT_PT / NBPG)))) { return; - + } if (!vm_object_lock_try(object)) return; /* - * if we are processing a major portion of the object, then - * scan the entire thing. + * if we are processing a major portion of the object, then scan the + * entire thing. */ - if (size > (object->size >> 1)) { + if (size > (object->size >> 2)) { objbytes = size; - p = object->memq.tqh_first; - while ((p != NULL) && (objbytes != 0)) { + + for (p = object->memq.tqh_first; + ((objbytes > 0) && (p != NULL)); + p = p->listq.tqe_next) { + tmpoff = p->offset; - if( tmpoff < offset) { - p = p->listq.tqe_next; + if (tmpoff < offset) { continue; } tmpoff -= offset; - if( tmpoff >= size) { - p = p->listq.tqe_next; + if (tmpoff >= size) { continue; } - if ((p->bmapped == 0) && (p->busy == 0) && - ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && - (p->flags & (PG_BUSY|PG_FICTITIOUS|PG_CACHE)) == 0 ) { + ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (p->flags & (PG_BUSY | PG_FICTITIOUS | PG_CACHE)) == 0) { vm_page_hold(p); - pmap_enter_quick(pmap, addr+tmpoff, VM_PAGE_TO_PHYS(p)); + pmap_enter_quick(pmap, addr + tmpoff, VM_PAGE_TO_PHYS(p)); vm_page_unhold(p); } - p = p->listq.tqe_next; objbytes -= NBPG; } } else { - /* - * else lookup the pages one-by-one. - */ + /* + * else lookup the pages one-by-one. + */ for (tmpoff = 0; tmpoff < size; tmpoff += NBPG) { p = vm_page_lookup(object, tmpoff + offset); - if (p) { - if ((p->bmapped == 0) && - (p->busy == 0) && - ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && - (p->flags & (PG_BUSY|PG_FICTITIOUS|PG_CACHE)) == 0) { - vm_page_hold(p); - pmap_enter_quick(pmap, addr+tmpoff, VM_PAGE_TO_PHYS(p)); - vm_page_unhold(p); - } + if (p && (p->bmapped == 0) && + (p->busy == 0) && + ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (p->flags & (PG_BUSY | PG_FICTITIOUS | PG_CACHE)) == 0) { + vm_page_hold(p); + pmap_enter_quick(pmap, addr + tmpoff, VM_PAGE_TO_PHYS(p)); + vm_page_unhold(p); } } } vm_object_unlock(object); } +/* + * pmap_prefault provides a quick way of clustering + * pagefaults into a processes address space. It is a "cousin" + * of pmap_object_init_pt, except it runs at page fault time instead + * of mmap time. + */ +#define PFBAK 2 +#define PFFOR 2 +#define PAGEORDER_SIZE (PFBAK+PFFOR) + +static int pmap_prefault_pageorder[] = { + -NBPG, NBPG, -2 * NBPG, 2 * NBPG +}; + +void +pmap_prefault(pmap, addra, entry, object) + pmap_t pmap; + vm_offset_t addra; + vm_map_entry_t entry; + vm_object_t object; +{ + int i; + vm_offset_t starta, enda; + vm_offset_t offset, addr; + vm_page_t m; + int pageorder_index; + + if (entry->object.vm_object != object) + return; + + if (pmap != &curproc->p_vmspace->vm_pmap) + return; + + starta = addra - PFBAK * NBPG; + if (starta < entry->start) { + starta = entry->start; + } else if (starta > addra) + starta = 0; + + enda = addra + PFFOR * NBPG; + if (enda > entry->end) + enda = entry->end; + + for (i = 0; i < PAGEORDER_SIZE; i++) { + vm_object_t lobject; + pt_entry_t *pte; + + addr = addra + pmap_prefault_pageorder[i]; + if (addr < starta || addr >= enda) + continue; + + pte = vtopte(addr); + if (*pte) + continue; + + offset = (addr - entry->start) + entry->offset; + lobject = object; + for (m = vm_page_lookup(lobject, offset); + (!m && lobject->shadow); + lobject = lobject->shadow) { + + offset += lobject->shadow_offset; + m = vm_page_lookup(lobject->shadow, offset); + } + + /* + * give-up when a page is not in memory + */ + if (m == NULL) + break; + + if ((m->bmapped == 0) && + (m->busy == 0) && + ((m->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL) && + (m->flags & (PG_CACHE | PG_BUSY | PG_FICTITIOUS)) == 0) { + /* + * test results show that the system is faster when + * pages are activated. + */ + if ((m->flags & PG_ACTIVE) == 0) + vm_page_activate(m); + vm_page_hold(m); + pmap_enter_quick(pmap, addr, VM_PAGE_TO_PHYS(m)); + vm_page_unhold(m); + } + } +} + /* * Routine: pmap_change_wiring * Function: Change the wiring attribute for a map/virtual-address @@ -1569,9 +1653,9 @@ pmap_object_init_pt(pmap, addr, object, offset, size) */ void pmap_change_wiring(pmap, va, wired) - register pmap_t pmap; - vm_offset_t va; - boolean_t wired; + register pmap_t pmap; + vm_offset_t va; + boolean_t wired; { register pt_entry_t *pte; @@ -1586,14 +1670,14 @@ pmap_change_wiring(pmap, va, wired) pmap->pm_stats.wired_count--; /* - * Wiring is not a hardware characteristic so there is no need - * to invalidate TLB. + * Wiring is not a hardware characteristic so there is no need to + * invalidate TLB. */ pmap_pte_set_w(pte, wired); /* - * When unwiring, set the modified bit in the pte -- could have - * been changed by the kernel - */ + * When unwiring, set the modified bit in the pte -- could have been + * changed by the kernel + */ if (!wired) (int) *pte |= PG_M; } @@ -1610,9 +1694,9 @@ pmap_change_wiring(pmap, va, wired) void pmap_copy(dst_pmap, src_pmap, dst_addr, len, src_addr) pmap_t dst_pmap, src_pmap; - vm_offset_t dst_addr; - vm_size_t len; - vm_offset_t src_addr; + vm_offset_t dst_addr; + vm_size_t len; + vm_offset_t src_addr; { } @@ -1624,7 +1708,7 @@ pmap_copy(dst_pmap, src_pmap, dst_addr, len, src_addr) pmap_t pmap_kernel() { - return (kernel_pmap); + return (kernel_pmap); } /* @@ -1637,13 +1721,13 @@ void pmap_zero_page(phys) vm_offset_t phys; { - if (*(int *)CMAP2) + if (*(int *) CMAP2) panic("pmap_zero_page: CMAP busy"); - *(int *)CMAP2 = PG_V | PG_KW | i386_trunc_page(phys); - bzero(CADDR2,NBPG); + *(int *) CMAP2 = PG_V | PG_KW | i386_trunc_page(phys); + bzero(CADDR2, NBPG); - *(int *)CMAP2 = 0; + *(int *) CMAP2 = 0; pmap_update(); } @@ -1658,19 +1742,19 @@ pmap_copy_page(src, dst) vm_offset_t src; vm_offset_t dst; { - if (*(int *)CMAP1 || *(int *)CMAP2) + if (*(int *) CMAP1 || *(int *) CMAP2) panic("pmap_copy_page: CMAP busy"); - *(int *)CMAP1 = PG_V | PG_KW | i386_trunc_page(src); - *(int *)CMAP2 = PG_V | PG_KW | i386_trunc_page(dst); + *(int *) CMAP1 = PG_V | PG_KW | i386_trunc_page(src); + *(int *) CMAP2 = PG_V | PG_KW | i386_trunc_page(dst); #if __GNUC__ > 1 memcpy(CADDR2, CADDR1, NBPG); #else - bcopy(CADDR1, CADDR2, NBPG); + bcopy(CADDR1, CADDR2, NBPG); #endif - *(int *)CMAP1 = 0; - *(int *)CMAP2 = 0; + *(int *) CMAP1 = 0; + *(int *) CMAP2 = 0; pmap_update(); } @@ -1691,9 +1775,9 @@ pmap_copy_page(src, dst) */ void pmap_pageable(pmap, sva, eva, pageable) - pmap_t pmap; - vm_offset_t sva, eva; - boolean_t pageable; + pmap_t pmap; + vm_offset_t sva, eva; + boolean_t pageable; { } @@ -1716,8 +1800,7 @@ pmap_page_exists(pmap, pa) s = splhigh(); /* - * Not found, check current mappings returning - * immediately if found. + * Not found, check current mappings returning immediately if found. */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { @@ -1728,7 +1811,7 @@ pmap_page_exists(pmap, pa) } } splx(s); - return(FALSE); + return (FALSE); } /* @@ -1736,7 +1819,7 @@ pmap_page_exists(pmap, pa) * note that the testbit/changebit routines are inline, * and a lot of things compile-time evaluate. */ -boolean_t +__inline boolean_t pmap_testbit(pa, bit) register vm_offset_t pa; int bit; @@ -1752,37 +1835,35 @@ pmap_testbit(pa, bit) s = splhigh(); /* - * Not found, check current mappings returning - * immediately if found. + * Not found, check current mappings returning immediately if found. */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { /* - * if the bit being tested is the modified bit, - * then mark UPAGES as always modified, and - * ptes as never modified. + * if the bit being tested is the modified bit, then + * mark UPAGES as always modified, and ptes as never + * modified. */ - if (bit & PG_U ) { + if (bit & PG_U) { if ((pv->pv_va >= clean_sva) && (pv->pv_va < clean_eva)) { continue; } - } - if (bit & PG_M ) { + } + if (bit & PG_M) { if (pv->pv_va >= USRSTACK) { if (pv->pv_va >= clean_sva && pv->pv_va < clean_eva) { continue; } - if (pv->pv_va < USRSTACK+(UPAGES*NBPG)) { + if (pv->pv_va < USRSTACK + (UPAGES * NBPG)) { splx(s); return TRUE; - } - else if (pv->pv_va < UPT_MAX_ADDRESS) { + } else if (pv->pv_va < UPT_MAX_ADDRESS) { splx(s); return FALSE; } } } - if( !pv->pv_pmap) { + if (!pv->pv_pmap) { printf("Null pmap (tb) at va: 0x%lx\n", pv->pv_va); continue; } @@ -1794,13 +1875,13 @@ pmap_testbit(pa, bit) } } splx(s); - return(FALSE); + return (FALSE); } /* * this routine is used to modify bits in ptes */ -void +__inline void pmap_changebit(pa, bit, setem) vm_offset_t pa; int bit; @@ -1818,8 +1899,8 @@ pmap_changebit(pa, bit, setem) s = splhigh(); /* - * Loop over all current mappings setting/clearing as appropos - * If setting RO do we need to clear the VAC? + * Loop over all current mappings setting/clearing as appropos If + * setting RO do we need to clear the VAC? */ if (pv->pv_pmap != NULL) { for (; pv; pv = pv->pv_next) { @@ -1832,8 +1913,7 @@ pmap_changebit(pa, bit, setem) if (va >= clean_sva && va < clean_eva) continue; } - - if( !pv->pv_pmap) { + if (!pv->pv_pmap) { printf("Null pmap (cb) at va: 0x%lx\n", va); continue; } @@ -1856,8 +1936,8 @@ pmap_changebit(pa, bit, setem) */ void pmap_page_protect(phys, prot) - vm_offset_t phys; - vm_prot_t prot; + vm_offset_t phys; + vm_prot_t prot; { if ((prot & VM_PROT_WRITE) == 0) { if (prot & (VM_PROT_READ | VM_PROT_EXECUTE)) @@ -1871,7 +1951,63 @@ vm_offset_t pmap_phys_address(ppn) int ppn; { - return(i386_ptob(ppn)); + return (i386_ptob(ppn)); +} + +/* + * pmap_is_referenced: + * + * Return whether or not the specified physical page was referenced + * by any physical maps. + */ +boolean_t +pmap_is_referenced(vm_offset_t pa) +{ + return pmap_testbit((pa), PG_U); +} + +/* + * pmap_is_modified: + * + * Return whether or not the specified physical page was modified + * in any physical maps. + */ +boolean_t +pmap_is_modified(vm_offset_t pa) +{ + return pmap_testbit((pa), PG_M); +} + +/* + * Clear the modify bits on the specified physical page. + */ +void +pmap_clear_modify(vm_offset_t pa) +{ + pmap_changebit((pa), PG_M, FALSE); +} + +/* + * pmap_clear_reference: + * + * Clear the reference bit on the specified physical page. + */ +void +pmap_clear_reference(vm_offset_t pa) +{ + pmap_changebit((pa), PG_U, FALSE); +} + +/* + * Routine: pmap_copy_on_write + * Function: + * Remove write privileges from all + * physical maps for this physical page. + */ +void +pmap_copy_on_write(vm_offset_t pa) +{ + pmap_changebit((pa), PG_RW, FALSE); } /* @@ -1887,10 +2023,10 @@ i386_protection_init() for (prot = 0; prot < 8; prot++) { switch (prot) { case VM_PROT_NONE | VM_PROT_NONE | VM_PROT_NONE: - /* - * Read access is also 0. There isn't any execute - * bit, so just make it readable. - */ + /* + * Read access is also 0. There isn't any execute bit, + * so just make it readable. + */ case VM_PROT_READ | VM_PROT_NONE | VM_PROT_NONE: case VM_PROT_READ | VM_PROT_NONE | VM_PROT_EXECUTE: case VM_PROT_NONE | VM_PROT_NONE | VM_PROT_EXECUTE: @@ -1937,7 +2073,7 @@ pmap_mapdev(pa, size) } pmap_update(); - return ((void *)va); + return ((void *) va); } #ifdef DEBUG @@ -1949,20 +2085,21 @@ pads(pm) unsigned va, i, j; pt_entry_t *ptep; - if (pm == kernel_pmap) return; - for (i = 0; i < 1024; i++) + if (pm == kernel_pmap) + return; + for (i = 0; i < 1024; i++) if (pm->pm_pdir[i]) - for (j = 0; j < 1024 ; j++) { - va = (i< UPT_MAX_ADDRESS) - continue; + continue; ptep = pmap_pte(pm, va); - if (pmap_pte_v(ptep)) - printf("%x:%x ", va, *(int *)ptep); - } ; - + if (pmap_pte_v(ptep)) + printf("%x:%x ", va, *(int *) ptep); + }; + } void @@ -1975,13 +2112,12 @@ pmap_pvdump(pa) for (pv = pa_to_pvh(pa); pv; pv = pv->pv_next) { #ifdef used_to_be printf(" -> pmap %x, va %x, flags %x", - pv->pv_pmap, pv->pv_va, pv->pv_flags); + pv->pv_pmap, pv->pv_va, pv->pv_flags); #endif printf(" -> pmap %x, va %x", - pv->pv_pmap, pv->pv_va); + pv->pv_pmap, pv->pv_va); pads(pv->pv_pmap); } printf(" "); } #endif - diff --git a/sys/i386/i386/trap.c b/sys/i386/i386/trap.c index 57d586a5615..a037e615207 100644 --- a/sys/i386/i386/trap.c +++ b/sys/i386/i386/trap.c @@ -35,7 +35,7 @@ * SUCH DAMAGE. * * from: @(#)trap.c 7.4 (Berkeley) 5/13/91 - * $Id: trap.c,v 1.43 1995/01/09 16:04:39 davidg Exp $ + * $Id: trap.c,v 1.44 1995/01/14 13:20:10 bde Exp $ */ /* @@ -460,23 +460,12 @@ trap_pfault(frame, usermode) /* Fault the pte only if needed: */ *(volatile char *)v += 0; - ptepg = (vm_page_t) pmap_pte_vm_page(vm_map_pmap(map), v); - vm_page_hold(ptepg); + pmap_use_pt( vm_map_pmap(map), va); /* Fault in the user page: */ rv = vm_fault(map, va, ftype, FALSE); - vm_page_unhold(ptepg); - - /* - * page table pages don't need to be kept if they - * are not held - */ - if( ptepg->hold_count == 0 && ptepg->wire_count == 0) { - pmap_page_protect( VM_PAGE_TO_PHYS(ptepg), - VM_PROT_NONE); - vm_page_free(ptepg); - } + pmap_unuse_pt( vm_map_pmap(map), va); --p->p_lock; } else { diff --git a/sys/kern/subr_trap.c b/sys/kern/subr_trap.c index 57d586a5615..a037e615207 100644 --- a/sys/kern/subr_trap.c +++ b/sys/kern/subr_trap.c @@ -35,7 +35,7 @@ * SUCH DAMAGE. * * from: @(#)trap.c 7.4 (Berkeley) 5/13/91 - * $Id: trap.c,v 1.43 1995/01/09 16:04:39 davidg Exp $ + * $Id: trap.c,v 1.44 1995/01/14 13:20:10 bde Exp $ */ /* @@ -460,23 +460,12 @@ trap_pfault(frame, usermode) /* Fault the pte only if needed: */ *(volatile char *)v += 0; - ptepg = (vm_page_t) pmap_pte_vm_page(vm_map_pmap(map), v); - vm_page_hold(ptepg); + pmap_use_pt( vm_map_pmap(map), va); /* Fault in the user page: */ rv = vm_fault(map, va, ftype, FALSE); - vm_page_unhold(ptepg); - - /* - * page table pages don't need to be kept if they - * are not held - */ - if( ptepg->hold_count == 0 && ptepg->wire_count == 0) { - pmap_page_protect( VM_PAGE_TO_PHYS(ptepg), - VM_PROT_NONE); - vm_page_free(ptepg); - } + pmap_unuse_pt( vm_map_pmap(map), va); --p->p_lock; } else {