linux_dsm_epyc7002/arch/powerpc/include/asm/pgtable.h
David Gibson a4fe3ce769 powerpc/mm: Allow more flexible layouts for hugepage pagetables
Currently each available hugepage size uses a slightly different
pagetable layout: that is, the bottem level table of pointers to
hugepages is a different size, and may branch off from the normal page
tables at a different level.  Every hugepage aware path that needs to
walk the pagetables must therefore look up the hugepage size from the
slice info first, and work out the correct way to walk the pagetables
accordingly.  Future hardware is likely to add more possible hugepage
sizes, more layout options and more mess.

This patch, therefore reworks the handling of hugepage pagetables to
reduce this complexity.  In the new scheme, instead of having to
consult the slice mask, pagetable walking code can check a flag in the
PGD/PUD/PMD entries to see where to branch off to hugepage pagetables,
and the entry also contains the information (eseentially hugepage
shift) necessary to then interpret that table without recourse to the
slice mask.  This scheme can be extended neatly to handle multiple
levels of self-describing "special" hugepage pagetables, although for
now we assume only one level exists.

This approach means that only the pagetable allocation path needs to
know how the pagetables should be set out.  All other (hugepage)
pagetable walking paths can just interpret the structure as they go.

There already was a flag bit in PGD/PUD/PMD entries for hugepage
directory pointers, but it was only used for debug.  We alter that
flag bit to instead be a 0 in the MSB to indicate a hugepage pagetable
pointer (normally it would be 1 since the pointer lies in the linear
mapping).  This means that asm pagetable walking can test for (and
punt on) hugepage pointers with the same test that checks for
unpopulated page directory entries (beq becomes bge), since hugepage
pointers will always be positive, and normal pointers always negative.

While we're at it, we get rid of the confusing (and grep defeating)
#defining of hugepte_shift to be the same thing as mmu_huge_psizes.

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-10-30 17:20:58 +11:00

221 lines
7.8 KiB
C

#ifndef _ASM_POWERPC_PGTABLE_H
#define _ASM_POWERPC_PGTABLE_H
#ifdef __KERNEL__
#ifndef __ASSEMBLY__
#include <asm/processor.h> /* For TASK_SIZE */
#include <asm/mmu.h>
#include <asm/page.h>
struct mm_struct;
#ifdef CONFIG_DEBUG_VM
extern void assert_pte_locked(struct mm_struct *mm, unsigned long addr);
#else /* CONFIG_DEBUG_VM */
static inline void assert_pte_locked(struct mm_struct *mm, unsigned long addr)
{
}
#endif /* !CONFIG_DEBUG_VM */
#endif /* !__ASSEMBLY__ */
#if defined(CONFIG_PPC64)
# include <asm/pgtable-ppc64.h>
#else
# include <asm/pgtable-ppc32.h>
#endif
#ifndef __ASSEMBLY__
/* Generic accessors to PTE bits */
static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW; }
static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY; }
static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED; }
static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE; }
static inline int pte_special(pte_t pte) { return pte_val(pte) & _PAGE_SPECIAL; }
static inline int pte_present(pte_t pte) { return pte_val(pte) & _PAGE_PRESENT; }
static inline int pte_none(pte_t pte) { return (pte_val(pte) & ~_PTE_NONE_MASK) == 0; }
static inline pgprot_t pte_pgprot(pte_t pte) { return __pgprot(pte_val(pte) & PAGE_PROT_BITS); }
/* Conversion functions: convert a page and protection to a page entry,
* and a page entry and page directory to the page they refer to.
*
* Even if PTEs can be unsigned long long, a PFN is always an unsigned
* long for now.
*/
static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot) {
return __pte(((pte_basic_t)(pfn) << PTE_RPN_SHIFT) |
pgprot_val(pgprot)); }
static inline unsigned long pte_pfn(pte_t pte) {
return pte_val(pte) >> PTE_RPN_SHIFT; }
/* Keep these as a macros to avoid include dependency mess */
#define pte_page(x) pfn_to_page(pte_pfn(x))
#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
/* Generic modifiers for PTE bits */
static inline pte_t pte_wrprotect(pte_t pte) {
pte_val(pte) &= ~(_PAGE_RW | _PAGE_HWWRITE); return pte; }
static inline pte_t pte_mkclean(pte_t pte) {
pte_val(pte) &= ~(_PAGE_DIRTY | _PAGE_HWWRITE); return pte; }
static inline pte_t pte_mkold(pte_t pte) {
pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
static inline pte_t pte_mkwrite(pte_t pte) {
pte_val(pte) |= _PAGE_RW; return pte; }
static inline pte_t pte_mkdirty(pte_t pte) {
pte_val(pte) |= _PAGE_DIRTY; return pte; }
static inline pte_t pte_mkyoung(pte_t pte) {
pte_val(pte) |= _PAGE_ACCESSED; return pte; }
static inline pte_t pte_mkspecial(pte_t pte) {
pte_val(pte) |= _PAGE_SPECIAL; return pte; }
static inline pte_t pte_mkhuge(pte_t pte) {
return pte; }
static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
{
pte_val(pte) = (pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot);
return pte;
}
/* Insert a PTE, top-level function is out of line. It uses an inline
* low level function in the respective pgtable-* files
*/
extern void set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep,
pte_t pte);
/* This low level function performs the actual PTE insertion
* Setting the PTE depends on the MMU type and other factors. It's
* an horrible mess that I'm not going to try to clean up now but
* I'm keeping it in one place rather than spread around
*/
static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
pte_t *ptep, pte_t pte, int percpu)
{
#if defined(CONFIG_PPC_STD_MMU_32) && defined(CONFIG_SMP) && !defined(CONFIG_PTE_64BIT)
/* First case is 32-bit Hash MMU in SMP mode with 32-bit PTEs. We use the
* helper pte_update() which does an atomic update. We need to do that
* because a concurrent invalidation can clear _PAGE_HASHPTE. If it's a
* per-CPU PTE such as a kmap_atomic, we do a simple update preserving
* the hash bits instead (ie, same as the non-SMP case)
*/
if (percpu)
*ptep = __pte((pte_val(*ptep) & _PAGE_HASHPTE)
| (pte_val(pte) & ~_PAGE_HASHPTE));
else
pte_update(ptep, ~_PAGE_HASHPTE, pte_val(pte));
#elif defined(CONFIG_PPC32) && defined(CONFIG_PTE_64BIT)
/* Second case is 32-bit with 64-bit PTE. In this case, we
* can just store as long as we do the two halves in the right order
* with a barrier in between. This is possible because we take care,
* in the hash code, to pre-invalidate if the PTE was already hashed,
* which synchronizes us with any concurrent invalidation.
* In the percpu case, we also fallback to the simple update preserving
* the hash bits
*/
if (percpu) {
*ptep = __pte((pte_val(*ptep) & _PAGE_HASHPTE)
| (pte_val(pte) & ~_PAGE_HASHPTE));
return;
}
#if _PAGE_HASHPTE != 0
if (pte_val(*ptep) & _PAGE_HASHPTE)
flush_hash_entry(mm, ptep, addr);
#endif
__asm__ __volatile__("\
stw%U0%X0 %2,%0\n\
eieio\n\
stw%U0%X0 %L2,%1"
: "=m" (*ptep), "=m" (*((unsigned char *)ptep+4))
: "r" (pte) : "memory");
#elif defined(CONFIG_PPC_STD_MMU_32)
/* Third case is 32-bit hash table in UP mode, we need to preserve
* the _PAGE_HASHPTE bit since we may not have invalidated the previous
* translation in the hash yet (done in a subsequent flush_tlb_xxx())
* and see we need to keep track that this PTE needs invalidating
*/
*ptep = __pte((pte_val(*ptep) & _PAGE_HASHPTE)
| (pte_val(pte) & ~_PAGE_HASHPTE));
#else
/* Anything else just stores the PTE normally. That covers all 64-bit
* cases, and 32-bit non-hash with 32-bit PTEs.
*/
*ptep = pte;
#endif
}
#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
extern int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long address,
pte_t *ptep, pte_t entry, int dirty);
/*
* Macro to mark a page protection value as "uncacheable".
*/
#define _PAGE_CACHE_CTL (_PAGE_COHERENT | _PAGE_GUARDED | _PAGE_NO_CACHE | \
_PAGE_WRITETHRU)
#define pgprot_noncached(prot) (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
_PAGE_NO_CACHE | _PAGE_GUARDED))
#define pgprot_noncached_wc(prot) (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
_PAGE_NO_CACHE))
#define pgprot_cached(prot) (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
_PAGE_COHERENT))
#define pgprot_cached_wthru(prot) (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
_PAGE_COHERENT | _PAGE_WRITETHRU))
struct file;
extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
unsigned long size, pgprot_t vma_prot);
#define __HAVE_PHYS_MEM_ACCESS_PROT
/*
* ZERO_PAGE is a global shared page that is always zero: used
* for zero-mapped memory areas etc..
*/
extern unsigned long empty_zero_page[];
#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
extern pgd_t swapper_pg_dir[];
extern void paging_init(void);
/*
* kern_addr_valid is intended to indicate whether an address is a valid
* kernel address. Most 32-bit archs define it as always true (like this)
* but most 64-bit archs actually perform a test. What should we do here?
*/
#define kern_addr_valid(addr) (1)
#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
remap_pfn_range(vma, vaddr, pfn, size, prot)
#include <asm-generic/pgtable.h>
/*
* This gets called at the end of handling a page fault, when
* the kernel has put a new PTE into the page table for the process.
* We use it to ensure coherency between the i-cache and d-cache
* for the page which has just been mapped in.
* On machines which use an MMU hash table, we use this to put a
* corresponding HPTE into the hash table ahead of time, instead of
* waiting for the inevitable extra hash-table miss exception.
*/
extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t);
extern int gup_hugepd(hugepd_t *hugepd, unsigned pdshift, unsigned long addr,
unsigned long end, int write, struct page **pages, int *nr);
#endif /* __ASSEMBLY__ */
#endif /* __KERNEL__ */
#endif /* _ASM_POWERPC_PGTABLE_H */