6bd95d70db
on for a while: - fine grained TLB shootdown for SMP on i386 - ranged TLB shootdowns.. eg: specify a range of pages to shoot down with a single IPI, since the IPI is very expensive. Adjust some callers that used to trigger this inside tight loops to do a ranged shootdown at the end instead. - PG_G support for SMP on i386 (options ENABLE_PG_G) - defer PG_G activation till after we decide what we are going to do with PSE and the 4MB pages at the start of the kernel. This should solve some rumored strangeness about stale PG_G entries getting stuck underneath the 4MB pages. - add some instrumentation for the fine TLB shootdown - convert some asm instruction wrappers from functions to inlines. gcc seems to do a fair bit better with this. - [temporarily!] pessimize the tlb shootdown IPI handlers. I will fix this again shortly. This has been working fairly well for me for a while, but I have tweaked it again prior to commit since my last major testing round. The only outstanding problem that I know of is PG_G related, which is why there is an option for it (not on by default for SMP). I have seen a world speedups by a few percent (as much as 4 or 5% in one case) but I have *not* accurately measured this - I am a bit sceptical of these numbers.
277 lines
8.8 KiB
C
277 lines
8.8 KiB
C
/*
|
|
* Copyright (c) 1991 Regents of the University of California.
|
|
* All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to Berkeley by
|
|
* the Systems Programming Group of the University of Utah Computer
|
|
* Science Department and William Jolitz of UUNET Technologies Inc.
|
|
*
|
|
* 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 the University of
|
|
* California, Berkeley and its contributors.
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 THE REGENTS OR CONTRIBUTORS 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.
|
|
*
|
|
* Derived from hp300 version by Mike Hibler, this version by William
|
|
* Jolitz uses a recursive map [a pde points to the page directory] to
|
|
* map the page tables using the pagetables themselves. This is done to
|
|
* reduce the impact on kernel virtual memory for lots of sparse address
|
|
* space, and to reduce the cost of memory to each process.
|
|
*
|
|
* from: hp300: @(#)pmap.h 7.2 (Berkeley) 12/16/90
|
|
* from: @(#)pmap.h 7.4 (Berkeley) 5/12/91
|
|
* $FreeBSD$
|
|
*/
|
|
|
|
#ifndef _MACHINE_PMAP_H_
|
|
#define _MACHINE_PMAP_H_
|
|
|
|
/*
|
|
* Page-directory and page-table entires follow this format, with a few
|
|
* of the fields not present here and there, depending on a lot of things.
|
|
*/
|
|
/* ---- Intel Nomenclature ---- */
|
|
#define PG_V 0x001 /* P Valid */
|
|
#define PG_RW 0x002 /* R/W Read/Write */
|
|
#define PG_U 0x004 /* U/S User/Supervisor */
|
|
#define PG_NC_PWT 0x008 /* PWT Write through */
|
|
#define PG_NC_PCD 0x010 /* PCD Cache disable */
|
|
#define PG_A 0x020 /* A Accessed */
|
|
#define PG_M 0x040 /* D Dirty */
|
|
#define PG_PS 0x080 /* PS Page size (0=4k,1=4M) */
|
|
#define PG_G 0x100 /* G Global */
|
|
#define PG_AVAIL1 0x200 /* / Available for system */
|
|
#define PG_AVAIL2 0x400 /* < programmers use */
|
|
#define PG_AVAIL3 0x800 /* \ */
|
|
|
|
|
|
/* Our various interpretations of the above */
|
|
#define PG_W PG_AVAIL1 /* "Wired" pseudoflag */
|
|
#define PG_MANAGED PG_AVAIL2
|
|
#define PG_FRAME (~PAGE_MASK)
|
|
#define PG_PROT (PG_RW|PG_U) /* all protection bits . */
|
|
#define PG_N (PG_NC_PWT|PG_NC_PCD) /* Non-cacheable */
|
|
|
|
/*
|
|
* Page Protection Exception bits
|
|
*/
|
|
|
|
#define PGEX_P 0x01 /* Protection violation vs. not present */
|
|
#define PGEX_W 0x02 /* during a Write cycle */
|
|
#define PGEX_U 0x04 /* access from User mode (UPL) */
|
|
|
|
/*
|
|
* Size of Kernel address space. This is the number of page table pages
|
|
* (4MB each) to use for the kernel. 256 pages == 1 Gigabyte.
|
|
* This **MUST** be a multiple of 4 (eg: 252, 256, 260, etc).
|
|
*/
|
|
#ifndef KVA_PAGES
|
|
#define KVA_PAGES 256
|
|
#endif
|
|
|
|
/*
|
|
* Pte related macros
|
|
*/
|
|
#define VADDR(pdi, pti) ((vm_offset_t)(((pdi)<<PDRSHIFT)|((pti)<<PAGE_SHIFT)))
|
|
|
|
#ifndef NKPT
|
|
#define NKPT 30 /* actual number of kernel page tables */
|
|
#endif
|
|
#ifndef NKPDE
|
|
#ifdef SMP
|
|
#define NKPDE (KVA_PAGES - 2) /* addressable number of page tables/pde's */
|
|
#else
|
|
#define NKPDE (KVA_PAGES - 1) /* addressable number of page tables/pde's */
|
|
#endif
|
|
#endif
|
|
|
|
/*
|
|
* The *PTDI values control the layout of virtual memory
|
|
*
|
|
* XXX This works for now, but I am not real happy with it, I'll fix it
|
|
* right after I fix locore.s and the magic 28K hole
|
|
*
|
|
* SMP_PRIVPAGES: The per-cpu address space is 0xff80000 -> 0xffbfffff
|
|
*/
|
|
#define APTDPTDI (NPDEPG-1) /* alt ptd entry that points to APTD */
|
|
#ifdef SMP
|
|
#define MPPTDI (APTDPTDI-1) /* per cpu ptd entry */
|
|
#define KPTDI (MPPTDI-NKPDE) /* start of kernel virtual pde's */
|
|
#else
|
|
#define KPTDI (APTDPTDI-NKPDE)/* start of kernel virtual pde's */
|
|
#endif /* SMP */
|
|
#define PTDPTDI (KPTDI-1) /* ptd entry that points to ptd! */
|
|
#define UMAXPTDI (PTDPTDI-1) /* ptd entry for user space end */
|
|
#define UMAXPTEOFF (NPTEPG) /* pte entry for user space end */
|
|
|
|
/*
|
|
* XXX doesn't really belong here I guess...
|
|
*/
|
|
#define ISA_HOLE_START 0xa0000
|
|
#define ISA_HOLE_LENGTH (0x100000-ISA_HOLE_START)
|
|
|
|
#ifndef LOCORE
|
|
|
|
#include <sys/queue.h>
|
|
|
|
typedef u_int32_t pd_entry_t;
|
|
typedef u_int32_t pt_entry_t;
|
|
|
|
#define PDESIZE sizeof(pd_entry_t) /* for assembly files */
|
|
#define PTESIZE sizeof(pt_entry_t) /* for assembly files */
|
|
|
|
/*
|
|
* Address of current and alternate address space page table maps
|
|
* and directories.
|
|
*/
|
|
#ifdef _KERNEL
|
|
extern pt_entry_t PTmap[], APTmap[];
|
|
extern pd_entry_t PTD[], APTD[];
|
|
extern pd_entry_t PTDpde, APTDpde;
|
|
|
|
extern pd_entry_t *IdlePTD; /* physical address of "Idle" state directory */
|
|
#endif
|
|
|
|
#ifdef _KERNEL
|
|
/*
|
|
* virtual address to page table entry and
|
|
* to physical address. Likewise for alternate address space.
|
|
* Note: these work recursively, thus vtopte of a pte will give
|
|
* the corresponding pde that in turn maps it.
|
|
*/
|
|
#define vtopte(va) (PTmap + i386_btop(va))
|
|
|
|
#define avtopte(va) (APTmap + i386_btop(va))
|
|
|
|
/*
|
|
* Routine: pmap_kextract
|
|
* Function:
|
|
* Extract the physical page address associated
|
|
* kernel virtual address.
|
|
*/
|
|
static __inline vm_offset_t
|
|
pmap_kextract(vm_offset_t va)
|
|
{
|
|
vm_offset_t pa;
|
|
if ((pa = (vm_offset_t) PTD[va >> PDRSHIFT]) & PG_PS) {
|
|
pa = (pa & ~(NBPDR - 1)) | (va & (NBPDR - 1));
|
|
} else {
|
|
pa = *(vm_offset_t *)vtopte(va);
|
|
pa = (pa & PG_FRAME) | (va & PAGE_MASK);
|
|
}
|
|
return pa;
|
|
}
|
|
|
|
#if 0
|
|
#define vtophys(va) (((vm_offset_t) (*vtopte(va))&PG_FRAME) | ((vm_offset_t)(va) & PAGE_MASK))
|
|
#else
|
|
#define vtophys(va) pmap_kextract(((vm_offset_t) (va)))
|
|
#endif
|
|
|
|
#define avtophys(va) (((vm_offset_t) (*avtopte(va))&PG_FRAME) | ((vm_offset_t)(va) & PAGE_MASK))
|
|
|
|
#endif
|
|
|
|
/*
|
|
* Pmap stuff
|
|
*/
|
|
struct pv_entry;
|
|
|
|
struct md_page {
|
|
int pv_list_count;
|
|
TAILQ_HEAD(,pv_entry) pv_list;
|
|
};
|
|
|
|
struct pmap {
|
|
pd_entry_t *pm_pdir; /* KVA of page directory */
|
|
vm_object_t pm_pteobj; /* Container for pte's */
|
|
TAILQ_HEAD(,pv_entry) pm_pvlist; /* list of mappings in pmap */
|
|
int pm_count; /* reference count */
|
|
int pm_active; /* active on cpus */
|
|
struct pmap_statistics pm_stats; /* pmap statistics */
|
|
struct vm_page *pm_ptphint; /* pmap ptp hint */
|
|
LIST_ENTRY(pmap) pm_list; /* List of all pmaps */
|
|
};
|
|
|
|
#define pmap_resident_count(pmap) (pmap)->pm_stats.resident_count
|
|
|
|
typedef struct pmap *pmap_t;
|
|
|
|
#ifdef _KERNEL
|
|
extern pmap_t kernel_pmap;
|
|
#endif
|
|
|
|
/*
|
|
* For each vm_page_t, there is a list of all currently valid virtual
|
|
* mappings of that page. An entry is a pv_entry_t, the list is pv_table.
|
|
*/
|
|
typedef struct pv_entry {
|
|
pmap_t pv_pmap; /* pmap where mapping lies */
|
|
vm_offset_t pv_va; /* virtual address for mapping */
|
|
TAILQ_ENTRY(pv_entry) pv_list;
|
|
TAILQ_ENTRY(pv_entry) pv_plist;
|
|
vm_page_t pv_ptem; /* VM page for pte */
|
|
} *pv_entry_t;
|
|
|
|
#define PV_ENTRY_NULL ((pv_entry_t) 0)
|
|
|
|
#define PV_CI 0x01 /* all entries must be cache inhibited */
|
|
#define PV_PTPAGE 0x02 /* entry maps a page table page */
|
|
|
|
#ifdef _KERNEL
|
|
|
|
#define NPPROVMTRR 8
|
|
#define PPRO_VMTRRphysBase0 0x200
|
|
#define PPRO_VMTRRphysMask0 0x201
|
|
struct ppro_vmtrr {
|
|
u_int64_t base, mask;
|
|
};
|
|
extern struct ppro_vmtrr PPro_vmtrr[NPPROVMTRR];
|
|
|
|
extern caddr_t CADDR1;
|
|
extern pt_entry_t *CMAP1;
|
|
extern vm_offset_t avail_end;
|
|
extern vm_offset_t avail_start;
|
|
extern vm_offset_t clean_eva;
|
|
extern vm_offset_t clean_sva;
|
|
extern vm_offset_t phys_avail[];
|
|
extern char *ptvmmap; /* poor name! */
|
|
extern vm_offset_t virtual_avail;
|
|
extern vm_offset_t virtual_end;
|
|
|
|
void pmap_bootstrap __P(( vm_offset_t, vm_offset_t));
|
|
pmap_t pmap_kernel __P((void));
|
|
void *pmap_mapdev __P((vm_offset_t, vm_size_t));
|
|
void pmap_unmapdev __P((vm_offset_t, vm_size_t));
|
|
pt_entry_t *pmap_pte __P((pmap_t, vm_offset_t)) __pure2;
|
|
vm_page_t pmap_use_pt __P((pmap_t, vm_offset_t));
|
|
void pmap_set_opt __P((void));
|
|
|
|
#endif /* _KERNEL */
|
|
|
|
#endif /* !LOCORE */
|
|
|
|
#endif /* !_MACHINE_PMAP_H_ */
|