d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
baa444b77e
freebsd-dev/sys/powerpc/aim/mmu_oea.c
Peter Wemm 0b27d7104f Make PMAP_SHPGPERPROC tunable. One shouldn't need to recompile a kernel 
for this, since it is easy to run into with large systems with lots of
shared mmap space.

Obtained from:	yahoo
2001-07-27 01:08:59 +00:00

1816 lines
36 KiB
C
Raw Blame History

/*
* Copyright (C) 1995, 1996 Wolfgang Solfrank.
* Copyright (C) 1995, 1996 TooLs GmbH.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by TooLs GmbH.
* 4. The name of TooLs GmbH may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY TOOLS GMBH ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL TOOLS GMBH BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
* OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
* WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
* ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* $NetBSD: pmap.c,v 1.28 2000/03/26 20:42:36 kleink Exp $
*/
/*
* Copyright (C) 2001 Benno Rice.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY Benno Rice ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL TOOLS GMBH BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
* OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
* WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
* ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef lint
static const char rcsid[] =
"$FreeBSD$";
#endif /* not lint */
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/proc.h>
#include <sys/malloc.h>
#include <sys/msgbuf.h>
#include <sys/vmmeter.h>
#include <sys/mman.h>
#include <sys/queue.h>
#include <sys/mutex.h>
#include <vm/vm.h>
#include <vm/vm_param.h>
#include <sys/lock.h>
#include <vm/vm_kern.h>
#include <vm/vm_page.h>
#include <vm/vm_map.h>
#include <vm/vm_object.h>
#include <vm/vm_extern.h>
#include <vm/vm_pageout.h>
#include <vm/vm_pager.h>
#include <vm/vm_zone.h>
#include <sys/user.h>
#include <machine/pcb.h>
#include <machine/powerpc.h>
#include <machine/pte.h>
pte_t *ptable;
int ptab_cnt;
u_int ptab_mask;
#define HTABSIZE (ptab_cnt * 64)
#define MINPV 2048
struct pte_ovfl {
LIST_ENTRY(pte_ovfl) po_list; /* Linked list of overflow entries */
struct pte po_pte; /* PTE for this mapping */
};
LIST_HEAD(pte_ovtab, pte_ovfl) *potable; /* Overflow entries for ptable */
static struct pmap kernel_pmap_store;
pmap_t kernel_pmap;
static int npgs;
static u_int nextavail;
#ifndef MSGBUFADDR
extern vm_offset_t msgbuf_paddr;
#endif
static struct mem_region *mem, *avail;
vm_offset_t avail_start;
vm_offset_t avail_end;
vm_offset_t virtual_avail;
vm_offset_t virtual_end;
vm_offset_t kernel_vm_end;
static int pmap_pagedaemon_waken = 0;
extern unsigned int Maxmem;
#define ATTRSHFT 4
struct pv_entry *pv_table;
static vm_zone_t pvzone;
static struct vm_zone pvzone_store;
static struct vm_object pvzone_obj;
static int pv_entry_count=0, pv_entry_max=0, pv_entry_high_water=0;
static struct pv_entry *pvinit;
#if !defined(PMAP_SHPGPERPROC)
#define PMAP_SHPGPERPROC 200
#endif
struct pv_page;
struct pv_page_info {
LIST_ENTRY(pv_page) pgi_list;
struct pv_entry *pgi_freelist;
int pgi_nfree;
};
#define NPVPPG ((PAGE_SIZE - sizeof(struct pv_page_info)) / sizeof(struct pv_entry))
struct pv_page {
struct pv_page_info pvp_pgi;
struct pv_entry pvp_pv[NPVPPG];
};
LIST_HEAD(pv_page_list, pv_page) pv_page_freelist;
int pv_nfree;
int pv_pcnt;
static struct pv_entry *pmap_alloc_pv(void);
static void pmap_free_pv(struct pv_entry *);
struct po_page;
struct po_page_info {
LIST_ENTRY(po_page) pgi_list;
vm_page_t pgi_page;
LIST_HEAD(po_freelist, pte_ovfl) pgi_freelist;
int pgi_nfree;
};
#define NPOPPG ((PAGE_SIZE - sizeof(struct po_page_info)) / sizeof(struct pte_ovfl))
struct po_page {
struct po_page_info pop_pgi;
struct pte_ovfl pop_po[NPOPPG];
};
LIST_HEAD(po_page_list, po_page) po_page_freelist;
int po_nfree;
int po_pcnt;
static struct pte_ovfl *poalloc(void);
static void pofree(struct pte_ovfl *, int);
static u_int usedsr[NPMAPS / sizeof(u_int) / 8];
static int pmap_initialized;
int pte_spill(vm_offset_t);
/*
* These small routines may have to be replaced,
* if/when we support processors other that the 604.
*/
static __inline void
tlbie(vm_offset_t ea)
{
__asm __volatile ("tlbie %0" :: "r"(ea));
}
static __inline void
tlbsync(void)
{
__asm __volatile ("sync; tlbsync; sync");
}
static __inline void
tlbia(void)
{
vm_offset_t i;
__asm __volatile ("sync");
for (i = 0; i < (vm_offset_t)0x00040000; i += 0x00001000) {
tlbie(i);
}
tlbsync();
}
static __inline int
ptesr(sr_t *sr, vm_offset_t addr)
{
return sr[(u_int)addr >> ADDR_SR_SHFT];
}
static __inline int
pteidx(sr_t sr, vm_offset_t addr)
{
int hash;
hash = (sr & SR_VSID) ^ (((u_int)addr & ADDR_PIDX) >> ADDR_PIDX_SHFT);
return hash & ptab_mask;
}
static __inline int
ptematch(pte_t *ptp, sr_t sr, vm_offset_t va, int which)
{
return ptp->pte_hi == (((sr & SR_VSID) << PTE_VSID_SHFT) |
(((u_int)va >> ADDR_API_SHFT) & PTE_API) | which);
}
static __inline struct pv_entry *
pa_to_pv(vm_offset_t pa)
{
#if 0 /* XXX */
int bank, pg;
bank = vm_physseg_find(atop(pa), &pg);
if (bank == -1)
return NULL;
return &vm_physmem[bank].pmseg.pvent[pg];
#endif
return (NULL);
}
static __inline char *
pa_to_attr(vm_offset_t pa)
{
#if 0 /* XXX */
int bank, pg;
bank = vm_physseg_find(atop(pa), &pg);
if (bank == -1)
return NULL;
return &vm_physmem[bank].pmseg.attrs[pg];
#endif
return (NULL);
}
/*
* Try to insert page table entry *pt into the ptable at idx.
*
* Note: *pt mustn't have PTE_VALID set.
* This is done here as required by Book III, 4.12.
*/
static int
pte_insert(int idx, pte_t *pt)
{
pte_t *ptp;
int i;
/*
* First try primary hash.
*/
for (ptp = ptable + idx * 8, i = 8; --i >= 0; ptp++) {
if (!(ptp->pte_hi & PTE_VALID)) {
*ptp = *pt;
ptp->pte_hi &= ~PTE_HID;
__asm __volatile ("sync");
ptp->pte_hi |= PTE_VALID;
return 1;
}
}
/*
* Then try secondary hash.
*/
idx ^= ptab_mask;
for (ptp = ptable + idx * 8, i = 8; --i >= 0; ptp++) {
if (!(ptp->pte_hi & PTE_VALID)) {
*ptp = *pt;
ptp->pte_hi |= PTE_HID;
__asm __volatile ("sync");
ptp->pte_hi |= PTE_VALID;
return 1;
}
}
return 0;
}
/*
* Spill handler.
*
* Tries to spill a page table entry from the overflow area.
* Note that this routine runs in real mode on a separate stack,
* with interrupts disabled.
*/
int
pte_spill(vm_offset_t addr)
{
int idx, i;
sr_t sr;
struct pte_ovfl *po;
pte_t ps;
pte_t *pt;
__asm ("mfsrin %0,%1" : "=r"(sr) : "r"(addr));
idx = pteidx(sr, addr);
for (po = potable[idx].lh_first; po; po = po->po_list.le_next) {
if (ptematch(&po->po_pte, sr, addr, 0)) {
/*
* Now found an entry to be spilled into the real
* ptable.
*/
if (pte_insert(idx, &po->po_pte)) {
LIST_REMOVE(po, po_list);
pofree(po, 0);
return 1;
}
/*
* Have to substitute some entry. Use the primary
* hash for this.
*
* Use low bits of timebase as random generator
*/
__asm ("mftb %0" : "=r"(i));
pt = ptable + idx * 8 + (i & 7);
pt->pte_hi &= ~PTE_VALID;
ps = *pt;
__asm __volatile ("sync");
tlbie(addr);
tlbsync();
*pt = po->po_pte;
__asm __volatile ("sync");
pt->pte_hi |= PTE_VALID;
po->po_pte = ps;
if (ps.pte_hi & PTE_HID) {
/*
* We took an entry that was on the alternate
* hash chain, so move it to it's original
* chain.
*/
po->po_pte.pte_hi &= ~PTE_HID;
LIST_REMOVE(po, po_list);
LIST_INSERT_HEAD(potable + (idx ^ ptab_mask),
po, po_list);
}
return 1;
}
}
return 0;
}
/*
* This is called during powerpc_init, before the system is really initialized.
*/
void
pmap_bootstrap(u_int kernelstart, u_int kernelend)
{
struct mem_region *mp, *mp1;
int cnt, i;
u_int s, e, sz;
/*
* Get memory.
*/
mem_regions(&mem, &avail);
for (mp = mem; mp->size; mp++)
Maxmem += btoc(mp->size);
/*
* Count the number of available entries.
*/
for (cnt = 0, mp = avail; mp->size; mp++) {
cnt++;
}
/*
* Page align all regions.
* Non-page aligned memory isn't very interesting to us.
* Also, sort the entries for ascending addresses.
*/
kernelstart &= ~PAGE_MASK;
kernelend = (kernelend + PAGE_MASK) & ~PAGE_MASK;
for (mp = avail; mp->size; mp++) {
s = mp->start;
e = mp->start + mp->size;
/*
* Check whether this region holds all of the kernel.
*/
if (s < kernelstart && e > kernelend) {
avail[cnt].start = kernelend;
avail[cnt++].size = e - kernelend;
e = kernelstart;
}
/*
* Look whether this regions starts within the kernel.
*/
if (s >= kernelstart && s < kernelend) {
if (e <= kernelend)
goto empty;
s = kernelend;
}
/*
* Now look whether this region ends within the kernel.
*/
if (e > kernelstart && e <= kernelend) {
if (s >= kernelstart)
goto empty;
e = kernelstart;
}
/*
* Now page align the start and size of the region.
*/
s = round_page(s);
e = trunc_page(e);
if (e < s) {
e = s;
}
sz = e - s;
/*
* Check whether some memory is left here.
*/
if (sz == 0) {
empty:
bcopy(mp + 1, mp,
(cnt - (mp - avail)) * sizeof *mp);
cnt--;
mp--;
continue;
}
/*
* Do an insertion sort.
*/
npgs += btoc(sz);
for (mp1 = avail; mp1 < mp; mp1++) {
if (s < mp1->start) {
break;
}
}
if (mp1 < mp) {
bcopy(mp1, mp1 + 1, (char *)mp - (char *)mp1);
mp1->start = s;
mp1->size = sz;
} else {
mp->start = s;
mp->size = sz;
}
}
#ifdef HTABENTS
ptab_cnt = HTABENTS;
#else
ptab_cnt = (Maxmem + 1) / 2;
/* The minimum is 1024 PTEGs. */
if (ptab_cnt < 1024) {
ptab_cnt = 1024;
}
/* Round up to power of 2. */
__asm ("cntlzw %0,%1" : "=r"(i) : "r"(ptab_cnt - 1));
ptab_cnt = 1 << (32 - i);
#endif
/*
* Find suitably aligned memory for HTAB.
*/
for (mp = avail; mp->size; mp++) {
s = roundup(mp->start, HTABSIZE) - mp->start;
if (mp->size < s + HTABSIZE) {
continue;
}
ptable = (pte_t *)(mp->start + s);
if (mp->size == s + HTABSIZE) {
if (s)
mp->size = s;
else {
bcopy(mp + 1, mp,
(cnt - (mp - avail)) * sizeof *mp);
mp = avail;
}
break;
}
if (s != 0) {
bcopy(mp, mp + 1,
(cnt - (mp - avail)) * sizeof *mp);
mp++->size = s;
cnt++;
}
mp->start += s + HTABSIZE;
mp->size -= s + HTABSIZE;
break;
}
if (!mp->size) {
panic("not enough memory?");
}
npgs -= btoc(HTABSIZE);
bzero((void *)ptable, HTABSIZE);
ptab_mask = ptab_cnt - 1;
/*
* We cannot do pmap_steal_memory here,
* since we don't run with translation enabled yet.
*/
s = sizeof(struct pte_ovtab) * ptab_cnt;
sz = round_page(s);
for (mp = avail; mp->size; mp++) {
if (mp->size >= sz) {
break;
}
}
if (!mp->size) {
panic("not enough memory?");
}
npgs -= btoc(sz);
potable = (struct pte_ovtab *)mp->start;
mp->size -= sz;
mp->start += sz;
if (mp->size <= 0) {
bcopy(mp + 1, mp, (cnt - (mp - avail)) * sizeof *mp);
}
for (i = 0; i < ptab_cnt; i++) {
LIST_INIT(potable + i);
}
#ifndef MSGBUFADDR
/*
* allow for msgbuf
*/
sz = round_page(MSGBUFSIZE);
mp = NULL;
for (mp1 = avail; mp1->size; mp1++) {
if (mp1->size >= sz) {
mp = mp1;
}
}
if (mp == NULL) {
panic("not enough memory?");
}
npgs -= btoc(sz);
msgbuf_paddr = mp->start + mp->size - sz;
mp->size -= sz;
if (mp->size <= 0) {
bcopy(mp + 1, mp, (cnt - (mp - avail)) * sizeof *mp);
}
#endif
/*
* Initialize kernel pmap and hardware.
*/
kernel_pmap = &kernel_pmap_store;
{
int batu, batl;
batu = 0x80001ffe;
batl = 0x80000012;
__asm ("mtdbatu 1,%0; mtdbatl 1,%1" :: "r" (batu), "r" (batl));
}
#if NPMAPS >= KERNEL_SEGMENT / 16
usedsr[KERNEL_SEGMENT / 16 / (sizeof usedsr[0] * 8)]
|= 1 << ((KERNEL_SEGMENT / 16) % (sizeof usedsr[0] * 8));
#endif
#if 0 /* XXX */
for (i = 0; i < 16; i++) {
kernel_pmap->pm_sr[i] = EMPTY_SEGMENT;
__asm __volatile ("mtsrin %0,%1"
:: "r"(EMPTY_SEGMENT), "r"(i << ADDR_SR_SHFT));
}
#endif
for (i = 0; i < 16; i++) {
int j;
__asm __volatile ("mfsrin %0,%1"
: "=r" (j)
: "r" (i << ADDR_SR_SHFT));
kernel_pmap->pm_sr[i] = j;
}
kernel_pmap->pm_sr[KERNEL_SR] = KERNEL_SEGMENT;
__asm __volatile ("mtsr %0,%1"
:: "n"(KERNEL_SR), "r"(KERNEL_SEGMENT));
__asm __volatile ("sync; mtsdr1 %0; isync"
:: "r"((u_int)ptable | (ptab_mask >> 10)));
tlbia();
nextavail = avail->start;
avail_start = avail->start;
for (mp = avail, i = 0; mp->size; mp++) {
avail_end = mp->start + mp->size;
phys_avail[i++] = mp->start;
phys_avail[i++] = mp->start + mp->size;
}
virtual_avail = VM_MIN_KERNEL_ADDRESS;
virtual_end = VM_MAX_KERNEL_ADDRESS;
}
/*
* Initialize anything else for pmap handling.
* Called during vm_init().
*/
void
pmap_init(vm_offset_t phys_start, vm_offset_t phys_end)
{
int initial_pvs;
/*
* init the pv free list
*/
initial_pvs = vm_page_array_size;
if (initial_pvs < MINPV) {
initial_pvs = MINPV;
}
pvzone = &pvzone_store;
pvinit = (struct pv_entry *) kmem_alloc(kernel_map,
initial_pvs * sizeof(struct pv_entry));
zbootinit(pvzone, "PV ENTRY", sizeof(struct pv_entry), pvinit,
vm_page_array_size);
pmap_initialized = TRUE;
}
/*
* Initialize a preallocated and zeroed pmap structure.
*/
void
pmap_pinit(struct pmap *pm)
{
int i, j;
/*
* Allocate some segment registers for this pmap.
*/
pm->pm_refs = 1;
for (i = 0; i < sizeof usedsr / sizeof usedsr[0]; i++) {
if (usedsr[i] != 0xffffffff) {
j = ffs(~usedsr[i]) - 1;
usedsr[i] |= 1 << j;
pm->pm_sr[0] = (i * sizeof usedsr[0] * 8 + j) * 16;
for (i = 1; i < 16; i++) {
pm->pm_sr[i] = pm->pm_sr[i - 1] + 1;
}
return;
}
}
panic("out of segments");
}
void
pmap_pinit2(pmap_t pmap)
{
/*
* Nothing to be done.
*/
return;
}
/*
* Add a reference to the given pmap.
*/
void
pmap_reference(struct pmap *pm)
{
pm->pm_refs++;
}
/*
* Retire the given pmap from service.
* Should only be called if the map contains no valid mappings.
*/
void
pmap_destroy(struct pmap *pm)
{
if (--pm->pm_refs == 0) {
pmap_release(pm);
free((caddr_t)pm, M_VMPGDATA);
}
}
/*
* Release any resources held by the given physical map.
* Called when a pmap initialized by pmap_pinit is being released.
*/
void
pmap_release(struct pmap *pm)
{
int i, j;
if (!pm->pm_sr[0]) {
panic("pmap_release");
}
i = pm->pm_sr[0] / 16;
j = i % (sizeof usedsr[0] * 8);
i /= sizeof usedsr[0] * 8;
usedsr[i] &= ~(1 << j);
}
/*
* Copy the range specified by src_addr/len
* from the source map to the range dst_addr/len
* in the destination map.
*
* This routine is only advisory and need not do anything.
*/
void
pmap_copy(struct pmap *dst_pmap, struct pmap *src_pmap, vm_offset_t dst_addr,
vm_size_t len, vm_offset_t src_addr)
{
return;
}
/*
* Garbage collects the physical map system for
* pages which are no longer used.
* Success need not be guaranteed -- that is, there
* may well be pages which are not referenced, but
* others may be collected.
* Called by the pageout daemon when pages are scarce.
*/
void
pmap_collect(void)
{
return;
}
/*
* Fill the given physical page with zeroes.
*/
void
pmap_zero_page(vm_offset_t pa)
{
#if 0
bzero((caddr_t)pa, PAGE_SIZE);
#else
int i;
for (i = PAGE_SIZE/CACHELINESIZE; i > 0; i--) {
__asm __volatile ("dcbz 0,%0" :: "r"(pa));
pa += CACHELINESIZE;
}
#endif
}
void
pmap_zero_page_area(vm_offset_t pa, int off, int size)
{
bzero((caddr_t)pa + off, size);
}
/*
* Copy the given physical source page to its destination.
*/
void
pmap_copy_page(vm_offset_t src, vm_offset_t dst)
{
bcopy((caddr_t)src, (caddr_t)dst, PAGE_SIZE);
}
static struct pv_entry *
pmap_alloc_pv()
{
pv_entry_count++;
if (pv_entry_high_water &&
(pv_entry_count > pv_entry_high_water) &&
(pmap_pagedaemon_waken == 0)) {
pmap_pagedaemon_waken = 1;
wakeup(&vm_pages_needed);
}
return zalloc(pvzone);
}
static void
pmap_free_pv(struct pv_entry *pv)
{
pv_entry_count--;
zfree(pvzone, pv);
}
/*
* We really hope that we don't need overflow entries
* before the VM system is initialized!
*
* XXX: Should really be switched over to the zone allocator.
*/
static struct pte_ovfl *
poalloc()
{
struct po_page *pop;
struct pte_ovfl *po;
vm_page_t mem;
int i;
if (!pmap_initialized) {
panic("poalloc");
}
if (po_nfree == 0) {
/*
* Since we cannot use maps for potable allocation,
* we have to steal some memory from the VM system. XXX
*/
mem = vm_page_alloc(NULL, 0, VM_ALLOC_SYSTEM);
po_pcnt++;
pop = (struct po_page *)VM_PAGE_TO_PHYS(mem);
pop->pop_pgi.pgi_page = mem;
LIST_INIT(&pop->pop_pgi.pgi_freelist);
for (i = NPOPPG - 1, po = pop->pop_po + 1; --i >= 0; po++) {
LIST_INSERT_HEAD(&pop->pop_pgi.pgi_freelist, po,
po_list);
}
po_nfree += pop->pop_pgi.pgi_nfree = NPOPPG - 1;
LIST_INSERT_HEAD(&po_page_freelist, pop, pop_pgi.pgi_list);
po = pop->pop_po;
} else {
po_nfree--;
pop = po_page_freelist.lh_first;
if (--pop->pop_pgi.pgi_nfree <= 0) {
LIST_REMOVE(pop, pop_pgi.pgi_list);
}
po = pop->pop_pgi.pgi_freelist.lh_first;
LIST_REMOVE(po, po_list);
}
return po;
}
static void
pofree(struct pte_ovfl *po, int freepage)
{
struct po_page *pop;
pop = (struct po_page *)trunc_page((vm_offset_t)po);
switch (++pop->pop_pgi.pgi_nfree) {
case NPOPPG:
if (!freepage) {
break;
}
po_nfree -= NPOPPG - 1;
po_pcnt--;
LIST_REMOVE(pop, pop_pgi.pgi_list);
vm_page_free(pop->pop_pgi.pgi_page);
return;
case 1:
LIST_INSERT_HEAD(&po_page_freelist, pop, pop_pgi.pgi_list);
default:
break;
}
LIST_INSERT_HEAD(&pop->pop_pgi.pgi_freelist, po, po_list);
po_nfree++;
}
/*
* This returns whether this is the first mapping of a page.
*/
static int
pmap_enter_pv(int pteidx, vm_offset_t va, vm_offset_t pa)
{
struct pv_entry *pv, *npv;
int s, first;
if (!pmap_initialized) {
return 0;
}
s = splimp();
pv = pa_to_pv(pa);
first = pv->pv_idx;
if (pv->pv_idx == -1) {
/*
* No entries yet, use header as the first entry.
*/
pv->pv_va = va;
pv->pv_idx = pteidx;
pv->pv_next = NULL;
} else {
/*
* There is at least one other VA mapping this page.
* Place this entry after the header.
*/
npv = pmap_alloc_pv();
npv->pv_va = va;
npv->pv_idx = pteidx;
npv->pv_next = pv->pv_next;
pv->pv_next = npv;
}
splx(s);
return first;
}
static void
pmap_remove_pv(int pteidx, vm_offset_t va, vm_offset_t pa, struct pte *pte)
{
struct pv_entry *pv, *npv;
char *attr;
/*
* First transfer reference/change bits to cache.
*/
attr = pa_to_attr(pa);
if (attr == NULL) {
return;
}
*attr |= (pte->pte_lo & (PTE_REF | PTE_CHG)) >> ATTRSHFT;
/*
* Remove from the PV table.
*/
pv = pa_to_pv(pa);
/*
* If it is the first entry on the list, it is actually
* in the header and we must copy the following entry up
* to the header. Otherwise we must search the list for
* the entry. In either case we free the now unused entry.
*/
if (pteidx == pv->pv_idx && va == pv->pv_va) {
npv = pv->pv_next;
if (npv) {
*pv = *npv;
pmap_free_pv(npv);
} else {
pv->pv_idx = -1;
}
} else {
for (; (npv = pv->pv_next); pv = npv) {
if (pteidx == npv->pv_idx && va == npv->pv_va) {
break;
}
}
if (npv) {
pv->pv_next = npv->pv_next;
pmap_free_pv(npv);
}
#ifdef DIAGNOSTIC
else {
panic("pmap_remove_pv: not on list\n");
}
#endif
}
}
/*
* Insert physical page at pa into the given pmap at virtual address va.
*/
void
pmap_enter(pmap_t pm, vm_offset_t va, vm_page_t pg, vm_prot_t prot,
boolean_t wired)
{
sr_t sr;
int idx, s;
pte_t pte;
struct pte_ovfl *po;
struct mem_region *mp;
vm_offset_t pa;
pa = VM_PAGE_TO_PHYS(pg) & ~PAGE_MASK;
/*
* Have to remove any existing mapping first.
*/
pmap_remove(pm, va, va + PAGE_SIZE);
/*
* Compute the HTAB index.
*/
idx = pteidx(sr = ptesr(pm->pm_sr, va), va);
/*
* Construct the PTE.
*
* Note: Don't set the valid bit for correct operation of tlb update.
*/
pte.pte_hi = ((sr & SR_VSID) << PTE_VSID_SHFT)
| ((va & ADDR_PIDX) >> ADDR_API_SHFT);
pte.pte_lo = (pa & PTE_RPGN) | PTE_M | PTE_I | PTE_G;
for (mp = mem; mp->size; mp++) {
if (pa >= mp->start && pa < mp->start + mp->size) {
pte.pte_lo &= ~(PTE_I | PTE_G);
break;
}
}
if (prot & VM_PROT_WRITE) {
pte.pte_lo |= PTE_RW;
} else {
pte.pte_lo |= PTE_RO;
}
/*
* Now record mapping for later back-translation.
*/
if (pmap_initialized && (pg->flags & PG_FICTITIOUS) == 0) {
if (pmap_enter_pv(idx, va, pa)) {
/*
* Flush the real memory from the cache.
*/
__syncicache((void *)pa, PAGE_SIZE);
}
}
s = splimp();
pm->pm_stats.resident_count++;
/*
* Try to insert directly into HTAB.
*/
if (pte_insert(idx, &pte)) {
splx(s);
return;
}
/*
* Have to allocate overflow entry.
*
* Note, that we must use real addresses for these.
*/
po = poalloc();
po->po_pte = pte;
LIST_INSERT_HEAD(potable + idx, po, po_list);
splx(s);
}
void
pmap_kenter(vm_offset_t va, vm_offset_t pa)
{
struct vm_page pg;
pg.phys_addr = pa;
pmap_enter(kernel_pmap, va, &pg, VM_PROT_READ|VM_PROT_WRITE, TRUE);
}
void
pmap_kremove(vm_offset_t va)
{
pmap_remove(kernel_pmap, va, va + PAGE_SIZE);
}
/*
* Remove the given range of mapping entries.
*/
void
pmap_remove(struct pmap *pm, vm_offset_t va, vm_offset_t endva)
{
int idx, i, s;
sr_t sr;
pte_t *ptp;
struct pte_ovfl *po, *npo;
s = splimp();
while (va < endva) {
idx = pteidx(sr = ptesr(pm->pm_sr, va), va);
for (ptp = ptable + idx * 8, i = 8; --i >= 0; ptp++) {
if (ptematch(ptp, sr, va, PTE_VALID)) {
pmap_remove_pv(idx, va, ptp->pte_lo, ptp);
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(va);
tlbsync();
pm->pm_stats.resident_count--;
}
}
for (ptp = ptable + (idx ^ ptab_mask) * 8, i = 8; --i >= 0;
ptp++) {
if (ptematch(ptp, sr, va, PTE_VALID | PTE_HID)) {
pmap_remove_pv(idx, va, ptp->pte_lo, ptp);
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(va);
tlbsync();
pm->pm_stats.resident_count--;
}
}
for (po = potable[idx].lh_first; po; po = npo) {
npo = po->po_list.le_next;
if (ptematch(&po->po_pte, sr, va, 0)) {
pmap_remove_pv(idx, va, po->po_pte.pte_lo,
&po->po_pte);
LIST_REMOVE(po, po_list);
pofree(po, 1);
pm->pm_stats.resident_count--;
}
}
va += PAGE_SIZE;
}
splx(s);
}
static pte_t *
pte_find(struct pmap *pm, vm_offset_t va)
{
int idx, i;
sr_t sr;
pte_t *ptp;
struct pte_ovfl *po;
idx = pteidx(sr = ptesr(pm->pm_sr, va), va);
for (ptp = ptable + idx * 8, i = 8; --i >= 0; ptp++) {
if (ptematch(ptp, sr, va, PTE_VALID)) {
return ptp;
}
}
for (ptp = ptable + (idx ^ ptab_mask) * 8, i = 8; --i >= 0; ptp++) {
if (ptematch(ptp, sr, va, PTE_VALID | PTE_HID)) {
return ptp;
}
}
for (po = potable[idx].lh_first; po; po = po->po_list.le_next) {
if (ptematch(&po->po_pte, sr, va, 0)) {
return &po->po_pte;
}
}
return 0;
}
/*
* Get the physical page address for the given pmap/virtual address.
*/
vm_offset_t
pmap_extract(pmap_t pm, vm_offset_t va)
{
pte_t *ptp;
int s;
s = splimp();
if (!(ptp = pte_find(pm, va))) {
splx(s);
return (0);
}
splx(s);
return ((ptp->pte_lo & PTE_RPGN) | (va & ADDR_POFF));
}
/*
* Lower the protection on the specified range of this pmap.
*
* There are only two cases: either the protection is going to 0,
* or it is going to read-only.
*/
void
pmap_protect(struct pmap *pm, vm_offset_t sva, vm_offset_t eva, vm_prot_t prot)
{
pte_t *ptp;
int valid, s;
if (prot & VM_PROT_READ) {
s = splimp();
while (sva < eva) {
ptp = pte_find(pm, sva);
if (ptp) {
valid = ptp->pte_hi & PTE_VALID;
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(sva);
tlbsync();
ptp->pte_lo &= ~PTE_PP;
ptp->pte_lo |= PTE_RO;
__asm __volatile ("sync");
ptp->pte_hi |= valid;
}
sva += PAGE_SIZE;
}
splx(s);
return;
}
pmap_remove(pm, sva, eva);
}
boolean_t
ptemodify(vm_page_t pg, u_int mask, u_int val)
{
vm_offset_t pa;
struct pv_entry *pv;
pte_t *ptp;
struct pte_ovfl *po;
int i, s;
char *attr;
int rv;
pa = VM_PAGE_TO_PHYS(pg);
/*
* First modify bits in cache.
*/
attr = pa_to_attr(pa);
if (attr == NULL) {
return FALSE;
}
*attr &= ~mask >> ATTRSHFT;
*attr |= val >> ATTRSHFT;
pv = pa_to_pv(pa);
if (pv->pv_idx < 0) {
return FALSE;
}
rv = FALSE;
s = splimp();
for (; pv; pv = pv->pv_next) {
for (ptp = ptable + pv->pv_idx * 8, i = 8; --i >= 0; ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(pv->pv_va);
tlbsync();
rv |= ptp->pte_lo & mask;
ptp->pte_lo &= ~mask;
ptp->pte_lo |= val;
__asm __volatile ("sync");
ptp->pte_hi |= PTE_VALID;
}
}
for (ptp = ptable + (pv->pv_idx ^ ptab_mask) * 8, i = 8;
--i >= 0; ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(pv->pv_va);
tlbsync();
rv |= ptp->pte_lo & mask;
ptp->pte_lo &= ~mask;
ptp->pte_lo |= val;
__asm __volatile ("sync");
ptp->pte_hi |= PTE_VALID;
}
}
for (po = potable[pv->pv_idx].lh_first; po;
po = po->po_list.le_next) {
if ((po->po_pte.pte_lo & PTE_RPGN) == pa) {
rv |= ptp->pte_lo & mask;
po->po_pte.pte_lo &= ~mask;
po->po_pte.pte_lo |= val;
}
}
}
splx(s);
return rv != 0;
}
int
ptebits(vm_page_t pg, int bit)
{
struct pv_entry *pv;
pte_t *ptp;
struct pte_ovfl *po;
int i, s, bits;
char *attr;
vm_offset_t pa;
bits = 0;
pa = VM_PAGE_TO_PHYS(pg);
/*
* First try the cache.
*/
attr = pa_to_attr(pa);
if (attr == NULL) {
return 0;
}
bits |= (*attr << ATTRSHFT) & bit;
if (bits == bit) {
return bits;
}
pv = pa_to_pv(pa);
if (pv->pv_idx < 0) {
return 0;
}
s = splimp();
for (; pv; pv = pv->pv_next) {
for (ptp = ptable + pv->pv_idx * 8, i = 8; --i >= 0; ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
bits |= ptp->pte_lo & bit;
if (bits == bit) {
splx(s);
return bits;
}
}
}
for (ptp = ptable + (pv->pv_idx ^ ptab_mask) * 8, i = 8;
--i >= 0; ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
bits |= ptp->pte_lo & bit;
if (bits == bit) {
splx(s);
return bits;
}
}
}
for (po = potable[pv->pv_idx].lh_first; po;
po = po->po_list.le_next) {
if ((po->po_pte.pte_lo & PTE_RPGN) == pa) {
bits |= po->po_pte.pte_lo & bit;
if (bits == bit) {
splx(s);
return bits;
}
}
}
}
splx(s);
return bits;
}
/*
* Lower the protection on the specified physical page.
*
* There are only two cases: either the protection is going to 0,
* or it is going to read-only.
*/
void
pmap_page_protect(vm_page_t m, vm_prot_t prot)
{
vm_offset_t pa;
vm_offset_t va;
pte_t *ptp;
struct pte_ovfl *po, *npo;
int i, s, idx;
struct pv_entry *pv;
pa = VM_PAGE_TO_PHYS(m);
pa &= ~ADDR_POFF;
if (prot & VM_PROT_READ) {
ptemodify(m, PTE_PP, PTE_RO);
return;
}
pv = pa_to_pv(pa);
if (pv == NULL) {
return;
}
s = splimp();
while (pv->pv_idx >= 0) {
idx = pv->pv_idx;
va = pv->pv_va;
for (ptp = ptable + idx * 8, i = 8; --i >= 0; ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
pmap_remove_pv(idx, va, pa, ptp);
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(va);
tlbsync();
goto next;
}
}
for (ptp = ptable + (idx ^ ptab_mask) * 8, i = 8; --i >= 0;
ptp++) {
if ((ptp->pte_hi & PTE_VALID)
&& (ptp->pte_lo & PTE_RPGN) == pa) {
pmap_remove_pv(idx, va, pa, ptp);
ptp->pte_hi &= ~PTE_VALID;
__asm __volatile ("sync");
tlbie(va);
tlbsync();
goto next;
}
}
for (po = potable[idx].lh_first; po; po = npo) {
npo = po->po_list.le_next;
if ((po->po_pte.pte_lo & PTE_RPGN) == pa) {
pmap_remove_pv(idx, va, pa, &po->po_pte);
LIST_REMOVE(po, po_list);
pofree(po, 1);
goto next;
}
}
next:
}
splx(s);
}
/*
* Activate the address space for the specified process. If the process
* is the current process, load the new MMU context.
*/
void
pmap_activate(struct proc *p)
{
struct pcb *pcb;
pmap_t pmap;
pmap_t rpm;
int psl, i, ksr, seg;
pcb = &p->p_addr->u_pcb;
pmap = p->p_vmspace->vm_map.pmap;
/*
* XXX Normally performed in cpu_fork().
*/
if (pcb->pcb_pm != pmap) {
pcb->pcb_pm = pmap;
(vm_offset_t) pcb->pcb_pmreal = pmap_extract(kernel_pmap,
(vm_offset_t)pcb->pcb_pm);
}
if (p == curproc) {
/* Disable interrupts while switching. */
psl = mfmsr();
mtmsr(psl & ~PSL_EE);
#if 0 /* XXX */
/* Store pointer to new current pmap. */
curpm = pcb->pcb_pmreal;
#endif
/* Save kernel SR. */
__asm __volatile("mfsr %0,14" : "=r"(ksr) :);
/*
* Set new segment registers. We use the pmap's real
* address to avoid accessibility problems.
*/
rpm = pcb->pcb_pmreal;
for (i = 0; i < 16; i++) {
seg = rpm->pm_sr[i];
__asm __volatile("mtsrin %0,%1"
:: "r"(seg), "r"(i << ADDR_SR_SHFT));
}
/* Restore kernel SR. */
__asm __volatile("mtsr 14,%0" :: "r"(ksr));
/* Interrupts are OK again. */
mtmsr(psl);
}
}
/*
* Add a list of wired pages to the kva
* this routine is only used for temporary
* kernel mappings that do not need to have
* page modification or references recorded.
* Note that old mappings are simply written
* over. The page *must* be wired.
*/
void
pmap_qenter(vm_offset_t va, vm_page_t *m, int count)
{
int i;
for (i = 0; i < count; i++) {
vm_offset_t tva = va + i * PAGE_SIZE;
pmap_kenter(tva, VM_PAGE_TO_PHYS(m[i]));
}
}
/*
* this routine jerks page mappings from the
* kernel -- it is meant only for temporary mappings.
*/
void
pmap_qremove(vm_offset_t va, int count)
{
vm_offset_t end_va;
end_va = va + count*PAGE_SIZE;
while (va < end_va) {
unsigned *pte;
pte = (unsigned *)vtopte(va);
*pte = 0;
tlbie(va);
va += PAGE_SIZE;
}
}
/*
* pmap_ts_referenced:
*
* Return the count of reference bits for a page, clearing all of them.
*/
int
pmap_ts_referenced(vm_page_t m)
{
/* XXX: coming soon... */
return (0);
}
/*
* this routine returns true if a physical page resides
* in the given pmap.
*/
boolean_t
pmap_page_exists(pmap_t pmap, vm_page_t m)
{
#if 0 /* XXX: This must go! */
register pv_entry_t pv;
int s;
if (!pmap_initialized || (m->flags & PG_FICTITIOUS))
return FALSE;
s = splvm();
/*
* Not found, check current mappings returning immediately if found.
*/
for (pv = pv_table; pv; pv = pv->pv_next) {
if (pv->pv_pmap == pmap) {
splx(s);
return TRUE;
}
}
splx(s);
#endif
return (FALSE);
}
/*
* Used to map a range of physical addresses into kernel
* virtual address space.
*
* For now, VM is already on, we only need to map the
* specified memory.
*/
vm_offset_t
pmap_map(vm_offset_t *virt, vm_offset_t start, vm_offset_t end, int prot)
{
vm_offset_t sva, va;
sva = *virt;
va = sva;
while (start < end) {
pmap_kenter(va, start);
va += PAGE_SIZE;
start += PAGE_SIZE;
}
*virt = va;
return (sva);
}
vm_offset_t
pmap_addr_hint(vm_object_t obj, vm_offset_t addr, vm_size_t size)
{
return (addr);
}
int
pmap_mincore(pmap_t pmap, vm_offset_t addr)
{
/* XXX: coming soon... */
return (0);
}
void
pmap_object_init_pt(pmap_t pmap, vm_offset_t addr, vm_object_t object,
vm_pindex_t pindex, vm_size_t size, int limit)
{
/* XXX: coming soon... */
return;
}
void
pmap_growkernel(vm_offset_t addr)
{
/* XXX: coming soon... */
return;
}
/*
* Initialize the address space (zone) for the pv_entries. Set a
* high water mark so that the system can recover from excessive
* numbers of pv entries.
*/
void
pmap_init2()
{
int shpgperproc = PMAP_SHPGPERPROC;
TUNABLE_INT_FETCH("vm.pmap.shpgperproc", &shpgperproc);
pv_entry_max = shpgperproc * maxproc + vm_page_array_size;
pv_entry_high_water = 9 * (pv_entry_max / 10);
zinitna(pvzone, &pvzone_obj, NULL, 0, pv_entry_max, ZONE_INTERRUPT, 1);
}
void
pmap_swapin_proc(struct proc *p)
{
/* XXX: coming soon... */
return;
}
void
pmap_swapout_proc(struct proc *p)
{
/* XXX: coming soon... */
return;
}
void
pmap_pageable(pmap_t pmap, vm_offset_t sva, vm_offset_t eva, boolean_t pageable)
{
return;
}
void
pmap_change_wiring(pmap_t pmap, vm_offset_t va, boolean_t wired)
{
/* XXX: coming soon... */
return;
}
void
pmap_prefault(pmap_t pmap, vm_offset_t addra, vm_map_entry_t entry)
{
/* XXX: coming soon... */
return;
}
void
pmap_remove_pages(pmap_t pmap, vm_offset_t sva, vm_offset_t eva)
{
/* XXX: coming soon... */
return;
}
void
pmap_pinit0(pmap_t pmap)
{
/* XXX: coming soon... */
return;
}
void
pmap_dispose_proc(struct proc *p)
{
/* XXX: coming soon... */
return;
}
vm_offset_t
pmap_steal_memory(vm_size_t size)
{
vm_size_t bank_size;
vm_offset_t pa;
size = round_page(size);
bank_size = phys_avail[1] - phys_avail[0];
while (size > bank_size) {
int i;
for (i = 0; phys_avail[i+2]; i+= 2) {
phys_avail[i] = phys_avail[i+2];
phys_avail[i+1] = phys_avail[i+3];
}
phys_avail[i] = 0;
phys_avail[i+1] = 0;
if (!phys_avail[0])
panic("pmap_steal_memory: out of memory");
bank_size = phys_avail[1] - phys_avail[0];
}
pa = phys_avail[0];
phys_avail[0] += size;
bzero((caddr_t) pa, size);
return pa;
}
/*
* Create the UPAGES for a new process.
* This routine directly affects the fork perf for a process.
*/
void
pmap_new_proc(struct proc *p)
{
int i;
vm_object_t upobj;
vm_page_t m;
struct user *up;
pte_t pte;
sr_t sr;
int idx;
/*
* allocate object for the upages
*/
if ((upobj = p->p_upages_obj) == NULL) {
upobj = vm_object_allocate( OBJT_DEFAULT, UPAGES);
p->p_upages_obj = upobj;
}
/* get a kernel virtual address for the UPAGES for this proc */
if ((up = p->p_addr) == NULL) {
up = (struct user *) kmem_alloc_nofault(kernel_map,
UPAGES * PAGE_SIZE);
if (up == NULL)
panic("pmap_new_proc: u_map allocation failed");
p->p_addr = up;
}
for(i=0;i<UPAGES;i++) {
vm_offset_t va;
/*
* Get a kernel stack page
*/
m = vm_page_grab(upobj, i, VM_ALLOC_NORMAL | VM_ALLOC_RETRY);
/*
* Wire the page
*/
m->wire_count++;
cnt.v_wire_count++;
/*
* Enter the page into the kernel address space.
*/
va = (vm_offset_t)(up + i * PAGE_SIZE);
idx = pteidx(sr = ptesr(kernel_pmap->pm_sr, va), va);
pte.pte_hi = ((sr & SR_VSID) << PTE_VSID_SHFT)
| ((va & ADDR_PIDX) >> ADDR_API_SHFT);
pte.pte_lo = (VM_PAGE_TO_PHYS(m) & PTE_RPGN) | PTE_M | PTE_I |
PTE_G | PTE_RW;
if (!pte_insert(idx, &pte)) {
struct pte_ovfl *po;
po = poalloc();
po->po_pte = pte;
LIST_INSERT_HEAD(potable + idx, po, po_list);
}
tlbie(va);
vm_page_wakeup(m);
vm_page_flag_clear(m, PG_ZERO);
vm_page_flag_set(m, PG_MAPPED | PG_WRITEABLE);
m->valid = VM_PAGE_BITS_ALL;
}
}
Reference in New Issue View Git Blame Copy Permalink