Commit 872fec16 authored by Hugh Dickins's avatar Hugh Dickins Committed by Linus Torvalds
Browse files

[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: default avatarHugh Dickins <hugh@veritas.com>
Signed-off-by: default avatarAndrew Morton <akpm@osdl.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
parent 46dea3d0
Loading
Loading
Loading
Loading
+1 −5
Original line number Diff line number Diff line
@@ -2,7 +2,6 @@
#include <asm/pgalloc.h>
#include <asm/cacheflush.h>

/* called with the page_table_lock held */
static inline void 
remap_area_pte(pte_t * pte, unsigned long address, unsigned long size, 
	       unsigned long phys_addr, unsigned long flags)
@@ -31,7 +30,6 @@ remap_area_pte(pte_t * pte, unsigned long address, unsigned long size,
	} while (address && (address < end));
}

/* called with the page_table_lock held */
static inline int 
remap_area_pmd(pmd_t * pmd, unsigned long address, unsigned long size, 
	       unsigned long phys_addr, unsigned long flags)
@@ -46,7 +44,7 @@ remap_area_pmd(pmd_t * pmd, unsigned long address, unsigned long size,
	if (address >= end)
		BUG();
	do {
		pte_t * pte = pte_alloc_kernel(&init_mm, pmd, address);
		pte_t * pte = pte_alloc_kernel(pmd, address);
		if (!pte)
			return -ENOMEM;
		remap_area_pte(pte, address, end - address, 
@@ -70,7 +68,6 @@ __alpha_remap_area_pages(unsigned long address, unsigned long phys_addr,
	flush_cache_all();
	if (address >= end)
		BUG();
	spin_lock(&init_mm.page_table_lock);
	do {
		pmd_t *pmd;
		pmd = pmd_alloc(&init_mm, dir, address);
@@ -84,7 +81,6 @@ __alpha_remap_area_pages(unsigned long address, unsigned long phys_addr,
		address = (address + PGDIR_SIZE) & PGDIR_MASK;
		dir++;
	} while (address && (address < end));
	spin_unlock(&init_mm.page_table_lock);
	return error;
}
+1 −5
Original line number Diff line number Diff line
@@ -397,8 +397,6 @@ static int __init consistent_init(void)
	pte_t *pte;
	int ret = 0;

	spin_lock(&init_mm.page_table_lock);

	do {
		pgd = pgd_offset(&init_mm, CONSISTENT_BASE);
		pmd = pmd_alloc(&init_mm, pgd, CONSISTENT_BASE);
@@ -409,7 +407,7 @@ static int __init consistent_init(void)
		}
		WARN_ON(!pmd_none(*pmd));

		pte = pte_alloc_kernel(&init_mm, pmd, CONSISTENT_BASE);
		pte = pte_alloc_kernel(pmd, CONSISTENT_BASE);
		if (!pte) {
			printk(KERN_ERR "%s: no pte tables\n", __func__);
			ret = -ENOMEM;
@@ -419,8 +417,6 @@ static int __init consistent_init(void)
		consistent_pte = pte;
	} while (0);

	spin_unlock(&init_mm.page_table_lock);

	return ret;
}

+1 −3
Original line number Diff line number Diff line
@@ -75,7 +75,7 @@ remap_area_pmd(pmd_t * pmd, unsigned long address, unsigned long size,

	pgprot = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY | L_PTE_WRITE | flags);
	do {
		pte_t * pte = pte_alloc_kernel(&init_mm, pmd, address);
		pte_t * pte = pte_alloc_kernel(pmd, address);
		if (!pte)
			return -ENOMEM;
		remap_area_pte(pte, address, end - address, address + phys_addr, pgprot);
@@ -97,7 +97,6 @@ remap_area_pages(unsigned long start, unsigned long phys_addr,
	phys_addr -= address;
	dir = pgd_offset(&init_mm, address);
	BUG_ON(address >= end);
	spin_lock(&init_mm.page_table_lock);
	do {
		pmd_t *pmd = pmd_alloc(&init_mm, dir, address);
		if (!pmd) {
@@ -114,7 +113,6 @@ remap_area_pages(unsigned long start, unsigned long phys_addr,
		dir++;
	} while (address && (address < end));

	spin_unlock(&init_mm.page_table_lock);
	flush_cache_vmap(start, end);
	return err;
}
+2 −1
Original line number Diff line number Diff line
@@ -92,7 +92,7 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
	if (!new_pmd)
		goto no_pmd;

	new_pte = pte_alloc_kernel(mm, new_pmd, 0);
	new_pte = pte_alloc_map(mm, new_pmd, 0);
	if (!new_pte)
		goto no_pte;

@@ -101,6 +101,7 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
	init_pte = pte_offset(init_pmd, 0);

	set_pte(new_pte, *init_pte);
	pte_unmap(new_pte);

	/*
	 * the page table entries are zeroed
+1 −3
Original line number Diff line number Diff line
@@ -52,7 +52,7 @@ static inline int remap_area_pmd(pmd_t * pmd, unsigned long address, unsigned lo
	if (address >= end)
		BUG();
	do {
		pte_t * pte = pte_alloc_kernel(&init_mm, pmd, address);
		pte_t * pte = pte_alloc_kernel(pmd, address);
		if (!pte)
			return -ENOMEM;
		remap_area_pte(pte, address, end - address, address + phys_addr, prot);
@@ -74,7 +74,6 @@ static int remap_area_pages(unsigned long address, unsigned long phys_addr,
	flush_cache_all();
	if (address >= end)
		BUG();
	spin_lock(&init_mm.page_table_lock);
	do {
		pud_t *pud;
		pmd_t *pmd;
@@ -94,7 +93,6 @@ static int remap_area_pages(unsigned long address, unsigned long phys_addr,
		address = (address + PGDIR_SIZE) & PGDIR_MASK;
		dir++;
	} while (address && (address < end));
	spin_unlock(&init_mm.page_table_lock);
	flush_tlb_all();
	return error;
}
Loading