2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* linux/arch/arm26/mm/memc.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 1998-2000 Russell King
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* Page table sludge for older ARM processor architectures.
|
|
|
|
*/
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/bootmem.h>
|
|
|
|
|
|
|
|
#include <asm/pgtable.h>
|
|
|
|
#include <asm/pgalloc.h>
|
|
|
|
#include <asm/page.h>
|
|
|
|
#include <asm/memory.h>
|
|
|
|
#include <asm/hardware.h>
|
|
|
|
|
|
|
|
#include <asm/map.h>
|
|
|
|
|
|
|
|
#define MEMC_TABLE_SIZE (256*sizeof(unsigned long))
|
|
|
|
|
|
|
|
kmem_cache_t *pte_cache, *pgd_cache;
|
|
|
|
int page_nr;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Allocate space for a page table and a MEMC table.
|
|
|
|
* Note that we place the MEMC
|
|
|
|
* table before the page directory. This means we can
|
|
|
|
* easily get to both tightly-associated data structures
|
|
|
|
* with a single pointer.
|
|
|
|
*/
|
|
|
|
static inline pgd_t *alloc_pgd_table(void)
|
|
|
|
{
|
|
|
|
void *pg2k = kmem_cache_alloc(pgd_cache, GFP_KERNEL);
|
|
|
|
|
|
|
|
if (pg2k)
|
|
|
|
pg2k += MEMC_TABLE_SIZE;
|
|
|
|
|
|
|
|
return (pgd_t *)pg2k;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Free a page table. this function is the counterpart to get_pgd_slow
|
|
|
|
* below, not alloc_pgd_table above.
|
|
|
|
*/
|
|
|
|
void free_pgd_slow(pgd_t *pgd)
|
|
|
|
{
|
|
|
|
unsigned long tbl = (unsigned long)pgd;
|
|
|
|
|
|
|
|
tbl -= MEMC_TABLE_SIZE;
|
|
|
|
|
|
|
|
kmem_cache_free(pgd_cache, (void *)tbl);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Allocate a new pgd and fill it in ready for use
|
|
|
|
*
|
|
|
|
* A new tasks pgd is completely empty (all pages !present) except for:
|
|
|
|
*
|
|
|
|
* o The machine vectors at virtual address 0x0
|
|
|
|
* o The vmalloc region at the top of address space
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
#define FIRST_KERNEL_PGD_NR (FIRST_USER_PGD_NR + USER_PTRS_PER_PGD)
|
|
|
|
|
|
|
|
pgd_t *get_pgd_slow(struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
pgd_t *new_pgd, *init_pgd;
|
|
|
|
pmd_t *new_pmd, *init_pmd;
|
|
|
|
pte_t *new_pte, *init_pte;
|
|
|
|
|
|
|
|
new_pgd = alloc_pgd_table();
|
|
|
|
if (!new_pgd)
|
|
|
|
goto no_pgd;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* On ARM, first page must always be allocated since it contains
|
|
|
|
* the machine vectors.
|
|
|
|
*/
|
|
|
|
new_pmd = pmd_alloc(mm, new_pgd, 0);
|
|
|
|
if (!new_pmd)
|
|
|
|
goto no_pmd;
|
|
|
|
|
[PATCH] mm: init_mm without ptlock
First step in pushing down the page_table_lock. init_mm.page_table_lock has
been used throughout the architectures (usually for ioremap): not to serialize
kernel address space allocation (that's usually vmlist_lock), but because
pud_alloc,pmd_alloc,pte_alloc_kernel expect caller holds it.
Reverse that: don't lock or unlock init_mm.page_table_lock in any of the
architectures; instead rely on pud_alloc,pmd_alloc,pte_alloc_kernel to take
and drop it when allocating a new one, to check lest a racing task already
did. Similarly no page_table_lock in vmalloc's map_vm_area.
Some temporary ugliness in __pud_alloc and __pmd_alloc: since they also handle
user mms, which are converted only by a later patch, for now they have to lock
differently according to whether or not it's init_mm.
If sources get muddled, there's a danger that an arch source taking
init_mm.page_table_lock will be mixed with common source also taking it (or
neither take it). So break the rules and make another change, which should
break the build for such a mismatch: remove the redundant mm arg from
pte_alloc_kernel (ppc64 scrapped its distinct ioremap_mm in 2.6.13).
Exceptions: arm26 used pte_alloc_kernel on user mm, now pte_alloc_map; ia64
used pte_alloc_map on init_mm, now pte_alloc_kernel; parisc had bad args to
pmd_alloc and pte_alloc_kernel in unused USE_HPPA_IOREMAP code; ppc64
map_io_page forgot to unlock on failure; ppc mmu_mapin_ram and ppc64 im_free
took page_table_lock for no good reason.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 02:16:21 +01:00
|
|
|
new_pte = pte_alloc_map(mm, new_pmd, 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
if (!new_pte)
|
|
|
|
goto no_pte;
|
|
|
|
|
|
|
|
init_pgd = pgd_offset(&init_mm, 0);
|
|
|
|
init_pmd = pmd_offset(init_pgd, 0);
|
|
|
|
init_pte = pte_offset(init_pmd, 0);
|
|
|
|
|
|
|
|
set_pte(new_pte, *init_pte);
|
[PATCH] mm: init_mm without ptlock
First step in pushing down the page_table_lock. init_mm.page_table_lock has
been used throughout the architectures (usually for ioremap): not to serialize
kernel address space allocation (that's usually vmlist_lock), but because
pud_alloc,pmd_alloc,pte_alloc_kernel expect caller holds it.
Reverse that: don't lock or unlock init_mm.page_table_lock in any of the
architectures; instead rely on pud_alloc,pmd_alloc,pte_alloc_kernel to take
and drop it when allocating a new one, to check lest a racing task already
did. Similarly no page_table_lock in vmalloc's map_vm_area.
Some temporary ugliness in __pud_alloc and __pmd_alloc: since they also handle
user mms, which are converted only by a later patch, for now they have to lock
differently according to whether or not it's init_mm.
If sources get muddled, there's a danger that an arch source taking
init_mm.page_table_lock will be mixed with common source also taking it (or
neither take it). So break the rules and make another change, which should
break the build for such a mismatch: remove the redundant mm arg from
pte_alloc_kernel (ppc64 scrapped its distinct ioremap_mm in 2.6.13).
Exceptions: arm26 used pte_alloc_kernel on user mm, now pte_alloc_map; ia64
used pte_alloc_map on init_mm, now pte_alloc_kernel; parisc had bad args to
pmd_alloc and pte_alloc_kernel in unused USE_HPPA_IOREMAP code; ppc64
map_io_page forgot to unlock on failure; ppc mmu_mapin_ram and ppc64 im_free
took page_table_lock for no good reason.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 02:16:21 +01:00
|
|
|
pte_unmap(new_pte);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* the page table entries are zeroed
|
|
|
|
* when the table is created. (see the cache_ctor functions below)
|
|
|
|
* Now we need to plonk the kernel (vmalloc) area at the end of
|
|
|
|
* the address space. We copy this from the init thread, just like
|
|
|
|
* the init_pte we copied above...
|
|
|
|
*/
|
|
|
|
memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
|
|
|
|
(PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
|
|
|
|
|
|
|
|
/* update MEMC tables */
|
|
|
|
cpu_memc_update_all(new_pgd);
|
|
|
|
return new_pgd;
|
|
|
|
|
|
|
|
no_pte:
|
|
|
|
pmd_free(new_pmd);
|
|
|
|
no_pmd:
|
|
|
|
free_pgd_slow(new_pgd);
|
|
|
|
no_pgd:
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* No special code is required here.
|
|
|
|
*/
|
|
|
|
void setup_mm_for_reboot(char mode)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This contains the code to setup the memory map on an ARM2/ARM250/ARM3
|
|
|
|
* o swapper_pg_dir = 0x0207d000
|
|
|
|
* o kernel proper starts at 0x0208000
|
|
|
|
* o create (allocate) a pte to contain the machine vectors
|
|
|
|
* o populate the pte (points to 0x02078000) (FIXME - is it zeroed?)
|
|
|
|
* o populate the init tasks page directory (pgd) with the new pte
|
|
|
|
* o zero the rest of the init tasks pgdir (FIXME - what about vmalloc?!)
|
|
|
|
*/
|
|
|
|
void __init memtable_init(struct meminfo *mi)
|
|
|
|
{
|
|
|
|
pte_t *pte;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
page_nr = max_low_pfn;
|
|
|
|
|
|
|
|
pte = alloc_bootmem_low_pages(PTRS_PER_PTE * sizeof(pte_t));
|
|
|
|
pte[0] = mk_pte_phys(PAGE_OFFSET + SCREEN_SIZE, PAGE_READONLY);
|
|
|
|
pmd_populate(&init_mm, pmd_offset(swapper_pg_dir, 0), pte);
|
|
|
|
|
|
|
|
for (i = 1; i < PTRS_PER_PGD; i++)
|
|
|
|
pgd_val(swapper_pg_dir[i]) = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void __init iotable_init(struct map_desc *io_desc)
|
|
|
|
{
|
|
|
|
/* nothing to do */
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We never have holes in the memmap
|
|
|
|
*/
|
|
|
|
void __init create_memmap_holes(struct meminfo *mi)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pte_cache_ctor(void *pte, kmem_cache_t *cache, unsigned long flags)
|
|
|
|
{
|
|
|
|
memzero(pte, sizeof(pte_t) * PTRS_PER_PTE);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pgd_cache_ctor(void *pgd, kmem_cache_t *cache, unsigned long flags)
|
|
|
|
{
|
|
|
|
memzero(pgd + MEMC_TABLE_SIZE, USER_PTRS_PER_PGD * sizeof(pgd_t));
|
|
|
|
}
|
|
|
|
|
|
|
|
void __init pgtable_cache_init(void)
|
|
|
|
{
|
|
|
|
pte_cache = kmem_cache_create("pte-cache",
|
|
|
|
sizeof(pte_t) * PTRS_PER_PTE,
|
|
|
|
0, 0, pte_cache_ctor, NULL);
|
|
|
|
if (!pte_cache)
|
|
|
|
BUG();
|
|
|
|
|
|
|
|
pgd_cache = kmem_cache_create("pgd-cache", MEMC_TABLE_SIZE +
|
|
|
|
sizeof(pgd_t) * PTRS_PER_PGD,
|
|
|
|
0, 0, pgd_cache_ctor, NULL);
|
|
|
|
if (!pgd_cache)
|
|
|
|
BUG();
|
|
|
|
}
|