2005-11-19 12:17:32 +03:00
|
|
|
#ifndef _ASM_POWERPC_PGTABLE_H
|
|
|
|
|
#define _ASM_POWERPC_PGTABLE_H
|
2005-12-17 00:43:46 +03:00
|
|
|
#ifdef __KERNEL__
|
2005-11-19 12:17:32 +03:00
|
|
|
|
2007-06-13 08:52:56 +04:00
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
|
#include <asm/processor.h> /* For TASK_SIZE */
|
|
|
|
|
#include <asm/mmu.h>
|
|
|
|
|
#include <asm/page.h>
|
powerpc/mm: Rework I$/D$ coherency (v3)
This patch reworks the way we do I and D cache coherency on PowerPC.
The "old" way was split in 3 different parts depending on the processor type:
- Hash with per-page exec support (64-bit and >= POWER4 only) does it
at hashing time, by preventing exec on unclean pages and cleaning pages
on exec faults.
- Everything without per-page exec support (32-bit hash, 8xx, and
64-bit < POWER4) does it for all page going to user space in update_mmu_cache().
- Embedded with per-page exec support does it from do_page_fault() on
exec faults, in a way similar to what the hash code does.
That leads to confusion, and bugs. For example, the method using update_mmu_cache()
is racy on SMP where another processor can see the new PTE and hash it in before
we have cleaned the cache, and then blow trying to execute. This is hard to hit but
I think it has bitten us in the past.
Also, it's inefficient for embedded where we always end up having to do at least
one more page fault.
This reworks the whole thing by moving the cache sync into two main call sites,
though we keep different behaviours depending on the HW capability. The call
sites are set_pte_at() which is now made out of line, and ptep_set_access_flags()
which joins the former in pgtable.c
The base idea for Embedded with per-page exec support, is that we now do the
flush at set_pte_at() time when coming from an exec fault, which allows us
to avoid the double fault problem completely (we can even improve the situation
more by implementing TLB preload in update_mmu_cache() but that's for later).
If for some reason we didn't do it there and we try to execute, we'll hit
the page fault, which will do a minor fault, which will hit ptep_set_access_flags()
to do things like update _PAGE_ACCESSED or _PAGE_DIRTY if needed, we just make
this guys also perform the I/D cache sync for exec faults now. This second path
is the catch all for things that weren't cleaned at set_pte_at() time.
For cpus without per-pag exec support, we always do the sync at set_pte_at(),
thus guaranteeing that when the PTE is visible to other processors, the cache
is clean.
For the 64-bit hash with per-page exec support case, we keep the old mechanism
for now. I'll look into changing it later, once I've reworked a bit how we
use _PAGE_EXEC.
This is also a first step for adding _PAGE_EXEC support for embedded platforms
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-02-10 19:02:37 +03:00
|
|
|
|
2007-06-13 08:52:56 +04:00
|
|
|
struct mm_struct;
|
powerpc/mm: Rework I$/D$ coherency (v3)
This patch reworks the way we do I and D cache coherency on PowerPC.
The "old" way was split in 3 different parts depending on the processor type:
- Hash with per-page exec support (64-bit and >= POWER4 only) does it
at hashing time, by preventing exec on unclean pages and cleaning pages
on exec faults.
- Everything without per-page exec support (32-bit hash, 8xx, and
64-bit < POWER4) does it for all page going to user space in update_mmu_cache().
- Embedded with per-page exec support does it from do_page_fault() on
exec faults, in a way similar to what the hash code does.
That leads to confusion, and bugs. For example, the method using update_mmu_cache()
is racy on SMP where another processor can see the new PTE and hash it in before
we have cleaned the cache, and then blow trying to execute. This is hard to hit but
I think it has bitten us in the past.
Also, it's inefficient for embedded where we always end up having to do at least
one more page fault.
This reworks the whole thing by moving the cache sync into two main call sites,
though we keep different behaviours depending on the HW capability. The call
sites are set_pte_at() which is now made out of line, and ptep_set_access_flags()
which joins the former in pgtable.c
The base idea for Embedded with per-page exec support, is that we now do the
flush at set_pte_at() time when coming from an exec fault, which allows us
to avoid the double fault problem completely (we can even improve the situation
more by implementing TLB preload in update_mmu_cache() but that's for later).
If for some reason we didn't do it there and we try to execute, we'll hit
the page fault, which will do a minor fault, which will hit ptep_set_access_flags()
to do things like update _PAGE_ACCESSED or _PAGE_DIRTY if needed, we just make
this guys also perform the I/D cache sync for exec faults now. This second path
is the catch all for things that weren't cleaned at set_pte_at() time.
For cpus without per-pag exec support, we always do the sync at set_pte_at(),
thus guaranteeing that when the PTE is visible to other processors, the cache
is clean.
For the 64-bit hash with per-page exec support case, we keep the old mechanism
for now. I'll look into changing it later, once I've reworked a bit how we
use _PAGE_EXEC.
This is also a first step for adding _PAGE_EXEC support for embedded platforms
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-02-10 19:02:37 +03:00
|
|
|
|
|
|
|
|
#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 */
|
|
|
|
|
|
2007-06-13 08:52:56 +04:00
|
|
|
#endif /* !__ASSEMBLY__ */
|
|
|
|
|
|
2007-04-30 10:30:56 +04:00
|
|
|
#if defined(CONFIG_PPC64)
|
|
|
|
|
# include <asm/pgtable-ppc64.h>
|
2005-11-19 12:17:32 +03:00
|
|
|
#else
|
2007-04-30 10:30:56 +04:00
|
|
|
# include <asm/pgtable-ppc32.h>
|
2005-08-05 13:39:06 +04:00
|
|
|
#endif
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
|
#ifndef __ASSEMBLY__
|
2008-12-18 22:13:51 +03:00
|
|
|
|
2009-03-19 22:34:09 +03:00
|
|
|
/* 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;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
powerpc/mm: Rework I$/D$ coherency (v3)
This patch reworks the way we do I and D cache coherency on PowerPC.
The "old" way was split in 3 different parts depending on the processor type:
- Hash with per-page exec support (64-bit and >= POWER4 only) does it
at hashing time, by preventing exec on unclean pages and cleaning pages
on exec faults.
- Everything without per-page exec support (32-bit hash, 8xx, and
64-bit < POWER4) does it for all page going to user space in update_mmu_cache().
- Embedded with per-page exec support does it from do_page_fault() on
exec faults, in a way similar to what the hash code does.
That leads to confusion, and bugs. For example, the method using update_mmu_cache()
is racy on SMP where another processor can see the new PTE and hash it in before
we have cleaned the cache, and then blow trying to execute. This is hard to hit but
I think it has bitten us in the past.
Also, it's inefficient for embedded where we always end up having to do at least
one more page fault.
This reworks the whole thing by moving the cache sync into two main call sites,
though we keep different behaviours depending on the HW capability. The call
sites are set_pte_at() which is now made out of line, and ptep_set_access_flags()
which joins the former in pgtable.c
The base idea for Embedded with per-page exec support, is that we now do the
flush at set_pte_at() time when coming from an exec fault, which allows us
to avoid the double fault problem completely (we can even improve the situation
more by implementing TLB preload in update_mmu_cache() but that's for later).
If for some reason we didn't do it there and we try to execute, we'll hit
the page fault, which will do a minor fault, which will hit ptep_set_access_flags()
to do things like update _PAGE_ACCESSED or _PAGE_DIRTY if needed, we just make
this guys also perform the I/D cache sync for exec faults now. This second path
is the catch all for things that weren't cleaned at set_pte_at() time.
For cpus without per-pag exec support, we always do the sync at set_pte_at(),
thus guaranteeing that when the PTE is visible to other processors, the cache
is clean.
For the 64-bit hash with per-page exec support case, we keep the old mechanism
for now. I'll look into changing it later, once I've reworked a bit how we
use _PAGE_EXEC.
This is also a first step for adding _PAGE_EXEC support for embedded platforms
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-02-10 19:02:37 +03:00
|
|
|
/* 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) && defined(CONFIG_SMP)
|
|
|
|
|
/* Second case is 32-bit with 64-bit PTE in SMP mode. 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 64-bit PTEs in UP mode
|
|
|
|
|
*/
|
|
|
|
|
*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);
|
|
|
|
|
|
2008-12-18 22:13:51 +03:00
|
|
|
/*
|
|
|
|
|
* 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
|
|
|
|
|
|
2007-06-13 08:52:56 +04:00
|
|
|
/*
|
|
|
|
|
* 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)
|
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <asm-generic/pgtable.h>
|
2008-07-25 10:21:11 +04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* 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);
|
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
#endif /* __ASSEMBLY__ */
|
|
|
|
|
|
2005-12-17 00:43:46 +03:00
|
|
|
#endif /* __KERNEL__ */
|
2005-11-19 12:17:32 +03:00
|
|
|
#endif /* _ASM_POWERPC_PGTABLE_H */
|
