* [PATCH 01/10] mm: numa: Do not dereference pmd outside of the lock during NUMA hinting fault
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 02/10] mm: Add p[te|md] protnone helpers for use by NUMA balancing Mel Gorman
` (8 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
A transhuge NUMA hinting fault may find the page is migrating and should
wait until migration completes. The check is race-prone because the pmd
is deferenced outside of the page lock and while the race is tiny, it'll
be larger if the PMD is cleared while marking PMDs for hinting fault.
This patch closes the race.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/migrate.h | 4 ----
mm/huge_memory.c | 3 ++-
mm/migrate.c | 6 ------
3 files changed, 2 insertions(+), 11 deletions(-)
diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index fab9b32..78baed5 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -67,7 +67,6 @@ static inline int migrate_huge_page_move_mapping(struct address_space *mapping,
#ifdef CONFIG_NUMA_BALANCING
extern bool pmd_trans_migrating(pmd_t pmd);
-extern void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd);
extern int migrate_misplaced_page(struct page *page,
struct vm_area_struct *vma, int node);
extern bool migrate_ratelimited(int node);
@@ -76,9 +75,6 @@ static inline bool pmd_trans_migrating(pmd_t pmd)
{
return false;
}
-static inline void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd)
-{
-}
static inline int migrate_misplaced_page(struct page *page,
struct vm_area_struct *vma, int node)
{
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 817a875..a2cd021 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1283,8 +1283,9 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
* check_same as the page may no longer be mapped.
*/
if (unlikely(pmd_trans_migrating(*pmdp))) {
+ page = pmd_page(*pmdp);
spin_unlock(ptl);
- wait_migrate_huge_page(vma->anon_vma, pmdp);
+ wait_on_page_locked(page);
goto out;
}
diff --git a/mm/migrate.c b/mm/migrate.c
index 344cdf6..e6a5ff1 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1685,12 +1685,6 @@ bool pmd_trans_migrating(pmd_t pmd)
return PageLocked(page);
}
-void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd)
-{
- struct page *page = pmd_page(*pmd);
- wait_on_page_locked(page);
-}
-
/*
* Attempt to migrate a misplaced page to the specified destination
* node. Caller is expected to have an elevated reference count on
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 02/10] mm: Add p[te|md] protnone helpers for use by NUMA balancing
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
2015-01-05 10:54 ` [PATCH 01/10] mm: numa: Do not dereference pmd outside of the lock during NUMA hinting fault Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 03/10] mm: Convert p[te|md]_numa users to p[te|md]_protnone_numa Mel Gorman
` (7 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
This is a preparatory patch that introduces protnone helpers for automatic
NUMA balancing.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Tested-by: Sasha Levin <sasha.levin@oracle.com>
---
arch/powerpc/include/asm/pgtable.h | 16 ++++++++++++++++
arch/x86/include/asm/pgtable.h | 16 ++++++++++++++++
include/asm-generic/pgtable.h | 20 ++++++++++++++++++++
3 files changed, 52 insertions(+)
diff --git a/arch/powerpc/include/asm/pgtable.h b/arch/powerpc/include/asm/pgtable.h
index a8805fe..7b889a3 100644
--- a/arch/powerpc/include/asm/pgtable.h
+++ b/arch/powerpc/include/asm/pgtable.h
@@ -39,6 +39,22 @@ static inline int pte_none(pte_t pte) { return (pte_val(pte) & ~_PTE_NONE_MASK)
static inline pgprot_t pte_pgprot(pte_t pte) { return __pgprot(pte_val(pte) & PAGE_PROT_BITS); }
#ifdef CONFIG_NUMA_BALANCING
+/*
+ * These work without NUMA balancing but the kernel does not care. See the
+ * comment in include/asm-generic/pgtable.h . On powerpc, this will only
+ * work for user pages and always return true for kernel pages.
+ */
+static inline int pte_protnone(pte_t pte)
+{
+ return (pte_val(pte) &
+ (_PAGE_PRESENT | _PAGE_USER)) == _PAGE_PRESENT;
+}
+
+static inline int pmd_protnone(pmd_t pmd)
+{
+ return pte_protnone(pmd_pte(pmd));
+}
+
static inline int pte_present(pte_t pte)
{
return pte_val(pte) & _PAGE_NUMA_MASK;
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index e8a5454..8b92203 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -503,6 +503,22 @@ static inline int pmd_present(pmd_t pmd)
_PAGE_NUMA);
}
+#ifdef CONFIG_NUMA_BALANCING
+/*
+ * These work without NUMA balancing but the kernel does not care. See the
+ * comment in include/asm-generic/pgtable.h
+ */
+static inline int pte_protnone(pte_t pte)
+{
+ return pte_flags(pte) & _PAGE_PROTNONE;
+}
+
+static inline int pmd_protnone(pmd_t pmd)
+{
+ return pmd_flags(pmd) & _PAGE_PROTNONE;
+}
+#endif /* CONFIG_NUMA_BALANCING */
+
static inline int pmd_none(pmd_t pmd)
{
/* Only check low word on 32-bit platforms, since it might be
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 177d597..d497d08 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -688,6 +688,26 @@ static inline int pmd_trans_unstable(pmd_t *pmd)
#endif
}
+#ifndef CONFIG_NUMA_BALANCING
+/*
+ * Technically a PTE can be PROTNONE even when not doing NUMA balancing but
+ * the only case the kernel cares is for NUMA balancing and is only ever set
+ * when the VMA is accessible. For PROT_NONE VMAs, the PTEs are not marked
+ * _PAGE_PROTNONE so implement the helper as "always no" by default. It is
+ * the responsibility of the caller to distinguish between PROT_NONE
+ * protections and NUMA hinting fault protections.
+ */
+static inline int pte_protnone(pte_t pte)
+{
+ return 0;
+}
+
+static inline int pmd_protnone(pmd_t pmd)
+{
+ return 0;
+}
+#endif /* CONFIG_NUMA_BALANCING */
+
#ifdef CONFIG_NUMA_BALANCING
/*
* _PAGE_NUMA distinguishes between an unmapped page table entry, an entry that
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 03/10] mm: Convert p[te|md]_numa users to p[te|md]_protnone_numa
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
2015-01-05 10:54 ` [PATCH 01/10] mm: numa: Do not dereference pmd outside of the lock during NUMA hinting fault Mel Gorman
2015-01-05 10:54 ` [PATCH 02/10] mm: Add p[te|md] protnone helpers for use by NUMA balancing Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 04/10] ppc64: Add paranoid warnings for unexpected DSISR_PROTFAULT Mel Gorman
` (6 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
Convert existing users of pte_numa and friends to the new helper. Note
that the kernel is broken after this patch is applied until the other
page table modifiers are also altered. This patch layout is to make
review easier.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Tested-by: Sasha Levin <sasha.levin@oracle.com>
---
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 2 +-
arch/powerpc/mm/fault.c | 5 -----
arch/powerpc/mm/pgtable.c | 11 ++++++++---
arch/powerpc/mm/pgtable_64.c | 3 ++-
arch/x86/mm/gup.c | 4 ++--
include/uapi/linux/mempolicy.h | 2 +-
mm/gup.c | 10 +++++-----
mm/huge_memory.c | 16 +++++++--------
mm/memory.c | 4 ++--
mm/mprotect.c | 39 ++++++++++---------------------------
mm/pgtable-generic.c | 2 +-
11 files changed, 40 insertions(+), 58 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 510bdfb..625407e 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -212,7 +212,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
/* Look up the Linux PTE for the backing page */
pte_size = psize;
pte = lookup_linux_pte_and_update(pgdir, hva, writing, &pte_size);
- if (pte_present(pte) && !pte_numa(pte)) {
+ if (pte_present(pte) && !pte_protnone(pte)) {
if (writing && !pte_write(pte))
/* make the actual HPTE be read-only */
ptel = hpte_make_readonly(ptel);
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index eb79907..b434153 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -398,8 +398,6 @@ good_area:
* processors use the same I/D cache coherency mechanism
* as embedded.
*/
- if (error_code & DSISR_PROTFAULT)
- goto bad_area;
#endif /* CONFIG_PPC_STD_MMU */
/*
@@ -423,9 +421,6 @@ good_area:
flags |= FAULT_FLAG_WRITE;
/* a read */
} else {
- /* protection fault */
- if (error_code & 0x08000000)
- goto bad_area;
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
}
diff --git a/arch/powerpc/mm/pgtable.c b/arch/powerpc/mm/pgtable.c
index c90e602..83dfcb5 100644
--- a/arch/powerpc/mm/pgtable.c
+++ b/arch/powerpc/mm/pgtable.c
@@ -172,9 +172,14 @@ static pte_t set_access_flags_filter(pte_t pte, struct vm_area_struct *vma,
void set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep,
pte_t pte)
{
-#ifdef CONFIG_DEBUG_VM
- WARN_ON(pte_val(*ptep) & _PAGE_PRESENT);
-#endif
+ /*
+ * When handling numa faults, we already have the pte marked
+ * _PAGE_PRESENT, but we can be sure that it is not in hpte.
+ * Hence we can use set_pte_at for them.
+ */
+ VM_WARN_ON((pte_val(*ptep) & (_PAGE_PRESENT | _PAGE_USER)) ==
+ (_PAGE_PRESENT | _PAGE_USER));
+
/* Note: mm->context.id might not yet have been assigned as
* this context might not have been activated yet when this
* is called.
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 4fe5f64..91bb883 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -718,7 +718,8 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
pmd_t *pmdp, pmd_t pmd)
{
#ifdef CONFIG_DEBUG_VM
- WARN_ON(pmd_val(*pmdp) & _PAGE_PRESENT);
+ WARN_ON((pmd_val(*pmdp) & (_PAGE_PRESENT | _PAGE_USER)) ==
+ (_PAGE_PRESENT | _PAGE_USER));
assert_spin_locked(&mm->page_table_lock);
WARN_ON(!pmd_trans_huge(pmd));
#endif
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index d754782..3b1d819 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -84,7 +84,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
struct page *page;
/* Similar to the PMD case, NUMA hinting must take slow path */
- if (pte_numa(pte)) {
+ if (pte_protnone(pte)) {
pte_unmap(ptep);
return 0;
}
@@ -178,7 +178,7 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
* slowpath for accounting purposes and so that they
* can be serialised against THP migration.
*/
- if (pmd_numa(pmd))
+ if (pmd_protnone(pmd))
return 0;
if (!gup_huge_pmd(pmd, addr, next, write, pages, nr))
return 0;
diff --git a/include/uapi/linux/mempolicy.h b/include/uapi/linux/mempolicy.h
index 0d11c3d..9cd8b21 100644
--- a/include/uapi/linux/mempolicy.h
+++ b/include/uapi/linux/mempolicy.h
@@ -67,7 +67,7 @@ enum mpol_rebind_step {
#define MPOL_F_LOCAL (1 << 1) /* preferred local allocation */
#define MPOL_F_REBINDING (1 << 2) /* identify policies in rebinding */
#define MPOL_F_MOF (1 << 3) /* this policy wants migrate on fault */
-#define MPOL_F_MORON (1 << 4) /* Migrate On pte_numa Reference On Node */
+#define MPOL_F_MORON (1 << 4) /* Migrate On protnone Reference On Node */
#endif /* _UAPI_LINUX_MEMPOLICY_H */
diff --git a/mm/gup.c b/mm/gup.c
index a900759..41db889 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -64,7 +64,7 @@ retry:
migration_entry_wait(mm, pmd, address);
goto retry;
}
- if ((flags & FOLL_NUMA) && pte_numa(pte))
+ if ((flags & FOLL_NUMA) && pte_protnone(pte))
goto no_page;
if ((flags & FOLL_WRITE) && !pte_write(pte)) {
pte_unmap_unlock(ptep, ptl);
@@ -193,7 +193,7 @@ struct page *follow_page_mask(struct vm_area_struct *vma,
}
return page;
}
- if ((flags & FOLL_NUMA) && pmd_numa(*pmd))
+ if ((flags & FOLL_NUMA) && pmd_protnone(*pmd))
return no_page_table(vma, flags);
if (pmd_trans_huge(*pmd)) {
if (flags & FOLL_SPLIT) {
@@ -740,10 +740,10 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
/*
* Similar to the PMD case below, NUMA hinting must take slow
- * path
+ * path using the pte_protnone check.
*/
if (!pte_present(pte) || pte_special(pte) ||
- pte_numa(pte) || (write && !pte_write(pte)))
+ pte_protnone(pte) || (write && !pte_write(pte)))
goto pte_unmap;
VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
@@ -938,7 +938,7 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
* slowpath for accounting purposes and so that they
* can be serialised against THP migration.
*/
- if (pmd_numa(pmd))
+ if (pmd_protnone(pmd))
return 0;
if (!gup_huge_pmd(pmd, pmdp, addr, next, write,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index a2cd021..f81fddf 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1222,7 +1222,7 @@ struct page *follow_trans_huge_pmd(struct vm_area_struct *vma,
return ERR_PTR(-EFAULT);
/* Full NUMA hinting faults to serialise migration in fault paths */
- if ((flags & FOLL_NUMA) && pmd_numa(*pmd))
+ if ((flags & FOLL_NUMA) && pmd_protnone(*pmd))
goto out;
page = pmd_page(*pmd);
@@ -1353,7 +1353,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
/*
* Migrate the THP to the requested node, returns with page unlocked
- * and pmd_numa cleared.
+ * and access rights restored.
*/
spin_unlock(ptl);
migrated = migrate_misplaced_transhuge_page(mm, vma,
@@ -1368,7 +1368,7 @@ clear_pmdnuma:
BUG_ON(!PageLocked(page));
pmd = pmd_mknonnuma(pmd);
set_pmd_at(mm, haddr, pmdp, pmd);
- VM_BUG_ON(pmd_numa(*pmdp));
+ VM_BUG_ON(pmd_protnone(*pmdp));
update_mmu_cache_pmd(vma, addr, pmdp);
unlock_page(page);
out_unlock:
@@ -1514,7 +1514,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
ret = 1;
if (!prot_numa) {
entry = pmdp_get_and_clear_notify(mm, addr, pmd);
- if (pmd_numa(entry))
+ if (pmd_protnone(entry))
entry = pmd_mknonnuma(entry);
entry = pmd_modify(entry, newprot);
ret = HPAGE_PMD_NR;
@@ -1530,7 +1530,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
* local vs remote hits on the zero page.
*/
if (!is_huge_zero_page(page) &&
- !pmd_numa(*pmd)) {
+ !pmd_protnone(*pmd)) {
pmdp_set_numa(mm, addr, pmd);
ret = HPAGE_PMD_NR;
}
@@ -1798,9 +1798,9 @@ static int __split_huge_page_map(struct page *page,
pte_t *pte, entry;
BUG_ON(PageCompound(page+i));
/*
- * Note that pmd_numa is not transferred deliberately
- * to avoid any possibility that pte_numa leaks to
- * a PROT_NONE VMA by accident.
+ * Note that NUMA hinting access restrictions are not
+ * transferred to avoid any possibility of altering
+ * permissions across VMAs.
*/
entry = mk_pte(page + i, vma->vm_page_prot);
entry = maybe_mkwrite(pte_mkdirty(entry), vma);
diff --git a/mm/memory.c b/mm/memory.c
index ca920d1..47aa715 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3222,7 +3222,7 @@ static int handle_pte_fault(struct mm_struct *mm,
pte, pmd, flags, entry);
}
- if (pte_numa(entry))
+ if (pte_protnone(entry))
return do_numa_page(mm, vma, address, entry, pte, pmd);
ptl = pte_lockptr(mm, pmd);
@@ -3300,7 +3300,7 @@ static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
if (pmd_trans_splitting(orig_pmd))
return 0;
- if (pmd_numa(orig_pmd))
+ if (pmd_protnone(orig_pmd))
return do_huge_pmd_numa_page(mm, vma, address,
orig_pmd, pmd);
diff --git a/mm/mprotect.c b/mm/mprotect.c
index ace9345..e93ddac 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -75,36 +75,17 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
oldpte = *pte;
if (pte_present(oldpte)) {
pte_t ptent;
- bool updated = false;
-
- if (!prot_numa) {
- ptent = ptep_modify_prot_start(mm, addr, pte);
- if (pte_numa(ptent))
- ptent = pte_mknonnuma(ptent);
- ptent = pte_modify(ptent, newprot);
- /*
- * Avoid taking write faults for pages we
- * know to be dirty.
- */
- if (dirty_accountable && pte_dirty(ptent) &&
- (pte_soft_dirty(ptent) ||
- !(vma->vm_flags & VM_SOFTDIRTY)))
- ptent = pte_mkwrite(ptent);
- ptep_modify_prot_commit(mm, addr, pte, ptent);
- updated = true;
- } else {
- struct page *page;
-
- page = vm_normal_page(vma, addr, oldpte);
- if (page && !PageKsm(page)) {
- if (!pte_numa(oldpte)) {
- ptep_set_numa(mm, addr, pte);
- updated = true;
- }
- }
+ ptent = ptep_modify_prot_start(mm, addr, pte);
+ ptent = pte_modify(ptent, newprot);
+
+ /* Avoid taking write faults for known dirty pages */
+ if (dirty_accountable && pte_dirty(ptent) &&
+ (pte_soft_dirty(ptent) ||
+ !(vma->vm_flags & VM_SOFTDIRTY))) {
+ ptent = pte_mkwrite(ptent);
}
- if (updated)
- pages++;
+ ptep_modify_prot_commit(mm, addr, pte, ptent);
+ pages++;
} else if (IS_ENABLED(CONFIG_MIGRATION) && !pte_file(oldpte)) {
swp_entry_t entry = pte_to_swp_entry(oldpte);
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index dfb79e0..4b8ad76 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -193,7 +193,7 @@ void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmdp)
{
pmd_t entry = *pmdp;
- if (pmd_numa(entry))
+ if (pmd_protnone(entry))
entry = pmd_mknonnuma(entry);
set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 04/10] ppc64: Add paranoid warnings for unexpected DSISR_PROTFAULT
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (2 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 03/10] mm: Convert p[te|md]_numa users to p[te|md]_protnone_numa Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 05/10] mm: Convert p[te|md]_mknonnuma and remaining page table manipulations Mel Gorman
` (5 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
ppc64 should not be depending on DSISR_PROTFAULT and it's unexpected
if they are triggered. This patch adds warnings just in case they
are being accidentally depended upon.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Tested-by: Sasha Levin <sasha.levin@oracle.com>
---
arch/powerpc/mm/copro_fault.c | 8 ++++++--
arch/powerpc/mm/fault.c | 20 +++++++++-----------
2 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/arch/powerpc/mm/copro_fault.c b/arch/powerpc/mm/copro_fault.c
index 5a236f0..0450d68 100644
--- a/arch/powerpc/mm/copro_fault.c
+++ b/arch/powerpc/mm/copro_fault.c
@@ -64,10 +64,14 @@ int copro_handle_mm_fault(struct mm_struct *mm, unsigned long ea,
if (!(vma->vm_flags & VM_WRITE))
goto out_unlock;
} else {
- if (dsisr & DSISR_PROTFAULT)
- goto out_unlock;
if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
goto out_unlock;
+ /*
+ * protfault should only happen due to us
+ * mapping a region readonly temporarily. PROT_NONE
+ * is also covered by the VMA check above.
+ */
+ WARN_ON_ONCE(dsisr & DSISR_PROTFAULT);
}
ret = 0;
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index b434153..1bcd378 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -389,17 +389,6 @@ good_area:
#endif /* CONFIG_8xx */
if (is_exec) {
-#ifdef CONFIG_PPC_STD_MMU
- /* Protection fault on exec go straight to failure on
- * Hash based MMUs as they either don't support per-page
- * execute permission, or if they do, it's handled already
- * at the hash level. This test would probably have to
- * be removed if we change the way this works to make hash
- * processors use the same I/D cache coherency mechanism
- * as embedded.
- */
-#endif /* CONFIG_PPC_STD_MMU */
-
/*
* Allow execution from readable areas if the MMU does not
* provide separate controls over reading and executing.
@@ -414,6 +403,14 @@ good_area:
(cpu_has_feature(CPU_FTR_NOEXECUTE) ||
!(vma->vm_flags & (VM_READ | VM_WRITE))))
goto bad_area;
+#ifdef CONFIG_PPC_STD_MMU
+ /*
+ * protfault should only happen due to us
+ * mapping a region readonly temporarily. PROT_NONE
+ * is also covered by the VMA check above.
+ */
+ WARN_ON_ONCE(error_code & DSISR_PROTFAULT);
+#endif /* CONFIG_PPC_STD_MMU */
/* a write */
} else if (is_write) {
if (!(vma->vm_flags & VM_WRITE))
@@ -423,6 +420,7 @@ good_area:
} else {
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
+ WARN_ON_ONCE(error_code & DSISR_PROTFAULT);
}
/*
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 05/10] mm: Convert p[te|md]_mknonnuma and remaining page table manipulations
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (3 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 04/10] ppc64: Add paranoid warnings for unexpected DSISR_PROTFAULT Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 06/10] mm: Remove remaining references to NUMA hinting bits and helpers Mel Gorman
` (4 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
With PROT_NONE, the traditional page table manipulation functions are
sufficient.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Tested-by: Sasha Levin <sasha.levin@oracle.com>
---
include/linux/huge_mm.h | 3 +--
mm/huge_memory.c | 33 +++++++--------------------------
mm/memory.c | 10 ++++++----
mm/mempolicy.c | 2 +-
mm/migrate.c | 2 +-
mm/mprotect.c | 2 +-
mm/pgtable-generic.c | 2 --
7 files changed, 17 insertions(+), 37 deletions(-)
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index ad9051b..554bbe3 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -31,8 +31,7 @@ extern int move_huge_pmd(struct vm_area_struct *vma,
unsigned long new_addr, unsigned long old_end,
pmd_t *old_pmd, pmd_t *new_pmd);
extern int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, pgprot_t newprot,
- int prot_numa);
+ unsigned long addr, pgprot_t newprot);
enum transparent_hugepage_flag {
TRANSPARENT_HUGEPAGE_FLAG,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index f81fddf..5618e22 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1366,9 +1366,8 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
goto out;
clear_pmdnuma:
BUG_ON(!PageLocked(page));
- pmd = pmd_mknonnuma(pmd);
+ pmd = pmd_modify(pmd, vma->vm_page_prot);
set_pmd_at(mm, haddr, pmdp, pmd);
- VM_BUG_ON(pmd_protnone(*pmdp));
update_mmu_cache_pmd(vma, addr, pmdp);
unlock_page(page);
out_unlock:
@@ -1503,7 +1502,7 @@ out:
* - HPAGE_PMD_NR is protections changed and TLB flush necessary
*/
int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, pgprot_t newprot, int prot_numa)
+ unsigned long addr, pgprot_t newprot)
{
struct mm_struct *mm = vma->vm_mm;
spinlock_t *ptl;
@@ -1512,29 +1511,11 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
pmd_t entry;
ret = 1;
- if (!prot_numa) {
- entry = pmdp_get_and_clear_notify(mm, addr, pmd);
- if (pmd_protnone(entry))
- entry = pmd_mknonnuma(entry);
- entry = pmd_modify(entry, newprot);
- ret = HPAGE_PMD_NR;
- set_pmd_at(mm, addr, pmd, entry);
- BUG_ON(pmd_write(entry));
- } else {
- struct page *page = pmd_page(*pmd);
-
- /*
- * Do not trap faults against the zero page. The
- * read-only data is likely to be read-cached on the
- * local CPU cache and it is less useful to know about
- * local vs remote hits on the zero page.
- */
- if (!is_huge_zero_page(page) &&
- !pmd_protnone(*pmd)) {
- pmdp_set_numa(mm, addr, pmd);
- ret = HPAGE_PMD_NR;
- }
- }
+ entry = pmdp_get_and_clear_notify(mm, addr, pmd);
+ entry = pmd_modify(entry, newprot);
+ ret = HPAGE_PMD_NR;
+ set_pmd_at(mm, addr, pmd, entry);
+ BUG_ON(pmd_write(entry));
spin_unlock(ptl);
}
diff --git a/mm/memory.c b/mm/memory.c
index 47aa715..debe3f4 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3113,9 +3113,9 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
* validation through pte_unmap_same(). It's of NUMA type but
* the pfn may be screwed if the read is non atomic.
*
- * ptep_modify_prot_start is not called as this is clearing
- * the _PAGE_NUMA bit and it is not really expected that there
- * would be concurrent hardware modifications to the PTE.
+ * We can safely just do a "set_pte_at()", because the old
+ * page table entry is not accessible, so there would be no
+ * concurrent hardware modifications to the PTE.
*/
ptl = pte_lockptr(mm, pmd);
spin_lock(ptl);
@@ -3124,7 +3124,9 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
goto out;
}
- pte = pte_mknonnuma(pte);
+ /* Make it present again */
+ pte = pte_modify(pte, vma->vm_page_prot);
+ pte = pte_mkyoung(pte);
set_pte_at(mm, addr, ptep, pte);
update_mmu_cache(vma, addr, ptep);
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 0e0961b..4fcbf12 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -627,7 +627,7 @@ unsigned long change_prot_numa(struct vm_area_struct *vma,
{
int nr_updated;
- nr_updated = change_protection(vma, addr, end, vma->vm_page_prot, 0, 1);
+ nr_updated = change_protection(vma, addr, end, PAGE_NONE, 0, 1);
if (nr_updated)
count_vm_numa_events(NUMA_PTE_UPDATES, nr_updated);
diff --git a/mm/migrate.c b/mm/migrate.c
index e6a5ff1..8491bee 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1878,7 +1878,7 @@ out_fail:
out_dropref:
ptl = pmd_lock(mm, pmd);
if (pmd_same(*pmd, entry)) {
- entry = pmd_mknonnuma(entry);
+ entry = pmd_modify(entry, vma->vm_page_prot);
set_pmd_at(mm, mmun_start, pmd, entry);
update_mmu_cache_pmd(vma, address, &entry);
}
diff --git a/mm/mprotect.c b/mm/mprotect.c
index e93ddac..dc65c0f 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -141,7 +141,7 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
split_huge_page_pmd(vma, addr, pmd);
else {
int nr_ptes = change_huge_pmd(vma, pmd, addr,
- newprot, prot_numa);
+ newprot);
if (nr_ptes) {
if (nr_ptes == HPAGE_PMD_NR) {
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index 4b8ad76..c25f94b 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -193,8 +193,6 @@ void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmdp)
{
pmd_t entry = *pmdp;
- if (pmd_protnone(entry))
- entry = pmd_mknonnuma(entry);
set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
}
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 06/10] mm: Remove remaining references to NUMA hinting bits and helpers
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (4 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 05/10] mm: Convert p[te|md]_mknonnuma and remaining page table manipulations Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 07/10] mm: numa: Do not trap faults on the huge zero page Mel Gorman
` (3 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
This patch removes the NUMA PTE bits and associated helpers. As a side-effect
it increases the maximum possible swap space on x86-64.
One potential source of problems is races between the marking of PTEs
PROT_NONE, NUMA hinting faults and migration. It must be guaranteed that
a PTE being protected is not faulted in parallel, seen as a pte_none and
corrupting memory. The base case is safe but transhuge has problems in the
past due to an different migration mechanism and a dependance on page lock
to serialise migrations and warrants a closer look.
task_work hinting update parallel fault
------------------------ --------------
change_pmd_range
change_huge_pmd
__pmd_trans_huge_lock
pmdp_get_and_clear
__handle_mm_fault
pmd_none
do_huge_pmd_anonymous_page
read? pmd_lock blocks until hinting complete, fail !pmd_none test
write? __do_huge_pmd_anonymous_page acquires pmd_lock, checks pmd_none
pmd_modify
set_pmd_at
task_work hinting update parallel migration
------------------------ ------------------
change_pmd_range
change_huge_pmd
__pmd_trans_huge_lock
pmdp_get_and_clear
__handle_mm_fault
do_huge_pmd_numa_page
migrate_misplaced_transhuge_page
pmd_lock waits for updates to complete, recheck pmd_same
pmd_modify
set_pmd_at
Both of those are safe and the case where a transhuge page is inserted
during a protection update is unchanged. The case where two processes try
migrating at the same time is unchanged by this series so should still be
ok. I could not find a case where we are accidentally depending on the
PTE not being cleared and flushed. If one is missed, it'll manifest as
corruption problems that start triggering shortly after this series is
merged and only happen when NUMA balancing is enabled.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Tested-by: Sasha Levin <sasha.levin@oracle.com>
---
arch/powerpc/include/asm/pgtable.h | 54 +-----------
arch/powerpc/include/asm/pte-common.h | 5 --
arch/powerpc/include/asm/pte-hash64.h | 6 --
arch/x86/include/asm/pgtable.h | 22 +----
arch/x86/include/asm/pgtable_64.h | 5 --
arch/x86/include/asm/pgtable_types.h | 41 +--------
include/asm-generic/pgtable.h | 155 ----------------------------------
include/linux/swapops.h | 2 +-
8 files changed, 7 insertions(+), 283 deletions(-)
diff --git a/arch/powerpc/include/asm/pgtable.h b/arch/powerpc/include/asm/pgtable.h
index 7b889a3..0a85e33 100644
--- a/arch/powerpc/include/asm/pgtable.h
+++ b/arch/powerpc/include/asm/pgtable.h
@@ -54,64 +54,12 @@ static inline int pmd_protnone(pmd_t pmd)
{
return pte_protnone(pmd_pte(pmd));
}
-
-static inline int pte_present(pte_t pte)
-{
- return pte_val(pte) & _PAGE_NUMA_MASK;
-}
-
-#define pte_present_nonuma pte_present_nonuma
-static inline int pte_present_nonuma(pte_t pte)
-{
- return pte_val(pte) & (_PAGE_PRESENT);
-}
-
-#define ptep_set_numa ptep_set_numa
-static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr,
- pte_t *ptep)
-{
- if ((pte_val(*ptep) & _PAGE_PRESENT) == 0)
- VM_BUG_ON(1);
-
- pte_update(mm, addr, ptep, _PAGE_PRESENT, _PAGE_NUMA, 0);
- return;
-}
-
-#define pmdp_set_numa pmdp_set_numa
-static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr,
- pmd_t *pmdp)
-{
- if ((pmd_val(*pmdp) & _PAGE_PRESENT) == 0)
- VM_BUG_ON(1);
-
- pmd_hugepage_update(mm, addr, pmdp, _PAGE_PRESENT, _PAGE_NUMA);
- return;
-}
-
-/*
- * Generic NUMA pte helpers expect pteval_t and pmdval_t types to exist
- * which was inherited from x86. For the purposes of powerpc pte_basic_t and
- * pmd_t are equivalent
- */
-#define pteval_t pte_basic_t
-#define pmdval_t pmd_t
-static inline pteval_t ptenuma_flags(pte_t pte)
-{
- return pte_val(pte) & _PAGE_NUMA_MASK;
-}
-
-static inline pmdval_t pmdnuma_flags(pmd_t pmd)
-{
- return pmd_val(pmd) & _PAGE_NUMA_MASK;
-}
-
-# else
+#endif /* CONFIG_NUMA_BALANCING */
static inline int pte_present(pte_t pte)
{
return pte_val(pte) & _PAGE_PRESENT;
}
-#endif /* CONFIG_NUMA_BALANCING */
/* Conversion functions: convert a page and protection to a page entry,
* and a page entry and page directory to the page they refer to.
diff --git a/arch/powerpc/include/asm/pte-common.h b/arch/powerpc/include/asm/pte-common.h
index e040c35..8d1569c 100644
--- a/arch/powerpc/include/asm/pte-common.h
+++ b/arch/powerpc/include/asm/pte-common.h
@@ -98,11 +98,6 @@ extern unsigned long bad_call_to_PMD_PAGE_SIZE(void);
_PAGE_USER | _PAGE_ACCESSED | \
_PAGE_RW | _PAGE_HWWRITE | _PAGE_DIRTY | _PAGE_EXEC)
-#ifdef CONFIG_NUMA_BALANCING
-/* Mask of bits that distinguish present and numa ptes */
-#define _PAGE_NUMA_MASK (_PAGE_NUMA|_PAGE_PRESENT)
-#endif
-
/*
* We define 2 sets of base prot bits, one for basic pages (ie,
* cacheable kernel and user pages) and one for non cacheable
diff --git a/arch/powerpc/include/asm/pte-hash64.h b/arch/powerpc/include/asm/pte-hash64.h
index 2505d8e..55aea0c 100644
--- a/arch/powerpc/include/asm/pte-hash64.h
+++ b/arch/powerpc/include/asm/pte-hash64.h
@@ -27,12 +27,6 @@
#define _PAGE_RW 0x0200 /* software: user write access allowed */
#define _PAGE_BUSY 0x0800 /* software: PTE & hash are busy */
-/*
- * Used for tracking numa faults
- */
-#define _PAGE_NUMA 0x00000010 /* Gather numa placement stats */
-
-
/* No separate kernel read-only */
#define _PAGE_KERNEL_RW (_PAGE_RW | _PAGE_DIRTY) /* user access blocked by key */
#define _PAGE_KERNEL_RO _PAGE_KERNEL_RW
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 8b92203..b9a13e9 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -305,7 +305,7 @@ static inline pmd_t pmd_mkwrite(pmd_t pmd)
static inline pmd_t pmd_mknotpresent(pmd_t pmd)
{
- return pmd_clear_flags(pmd, _PAGE_PRESENT);
+ return pmd_clear_flags(pmd, _PAGE_PRESENT | _PAGE_PROTNONE);
}
#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
@@ -463,13 +463,6 @@ static inline int pte_same(pte_t a, pte_t b)
static inline int pte_present(pte_t a)
{
- return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE |
- _PAGE_NUMA);
-}
-
-#define pte_present_nonuma pte_present_nonuma
-static inline int pte_present_nonuma(pte_t a)
-{
return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE);
}
@@ -479,7 +472,7 @@ static inline bool pte_accessible(struct mm_struct *mm, pte_t a)
if (pte_flags(a) & _PAGE_PRESENT)
return true;
- if ((pte_flags(a) & (_PAGE_PROTNONE | _PAGE_NUMA)) &&
+ if ((pte_flags(a) & _PAGE_PROTNONE) &&
mm_tlb_flush_pending(mm))
return true;
@@ -499,8 +492,7 @@ static inline int pmd_present(pmd_t pmd)
* the _PAGE_PSE flag will remain set at all times while the
* _PAGE_PRESENT bit is clear).
*/
- return pmd_flags(pmd) & (_PAGE_PRESENT | _PAGE_PROTNONE | _PAGE_PSE |
- _PAGE_NUMA);
+ return pmd_flags(pmd) & (_PAGE_PRESENT | _PAGE_PROTNONE | _PAGE_PSE);
}
#ifdef CONFIG_NUMA_BALANCING
@@ -575,11 +567,6 @@ static inline pte_t *pte_offset_kernel(pmd_t *pmd, unsigned long address)
static inline int pmd_bad(pmd_t pmd)
{
-#ifdef CONFIG_NUMA_BALANCING
- /* pmd_numa check */
- if ((pmd_flags(pmd) & (_PAGE_NUMA|_PAGE_PRESENT)) == _PAGE_NUMA)
- return 0;
-#endif
return (pmd_flags(pmd) & ~_PAGE_USER) != _KERNPG_TABLE;
}
@@ -898,19 +885,16 @@ static inline void update_mmu_cache_pmd(struct vm_area_struct *vma,
#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
- VM_BUG_ON(pte_present_nonuma(pte));
return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
static inline int pte_swp_soft_dirty(pte_t pte)
{
- VM_BUG_ON(pte_present_nonuma(pte));
return pte_flags(pte) & _PAGE_SWP_SOFT_DIRTY;
}
static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
{
- VM_BUG_ON(pte_present_nonuma(pte));
return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
#endif
diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
index 4572b2f..06ffca8 100644
--- a/arch/x86/include/asm/pgtable_64.h
+++ b/arch/x86/include/asm/pgtable_64.h
@@ -146,12 +146,7 @@ static inline int pgd_large(pgd_t pgd) { return 0; }
/* Encode and de-code a swap entry */
#define SWP_TYPE_BITS (_PAGE_BIT_FILE - _PAGE_BIT_PRESENT - 1)
-#ifdef CONFIG_NUMA_BALANCING
-/* Automatic NUMA balancing needs to be distinguishable from swap entries */
-#define SWP_OFFSET_SHIFT (_PAGE_BIT_PROTNONE + 2)
-#else
#define SWP_OFFSET_SHIFT (_PAGE_BIT_PROTNONE + 1)
-#endif
#define MAX_SWAPFILES_CHECK() BUILD_BUG_ON(MAX_SWAPFILES_SHIFT > SWP_TYPE_BITS)
diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
index 25bcd4a..620bed5 100644
--- a/arch/x86/include/asm/pgtable_types.h
+++ b/arch/x86/include/asm/pgtable_types.h
@@ -27,14 +27,6 @@
#define _PAGE_BIT_SOFT_DIRTY _PAGE_BIT_SOFTW3 /* software dirty tracking */
#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */
-/*
- * Swap offsets on configurations that allow automatic NUMA balancing use the
- * bits after _PAGE_BIT_GLOBAL. To uniquely distinguish NUMA hinting PTEs from
- * swap entries, we use the first bit after _PAGE_BIT_GLOBAL and shrink the
- * maximum possible swap space from 16TB to 8TB.
- */
-#define _PAGE_BIT_NUMA (_PAGE_BIT_GLOBAL+1)
-
/* If _PAGE_BIT_PRESENT is clear, we use these: */
/* - if the user mapped it with PROT_NONE; pte_present gives true */
#define _PAGE_BIT_PROTNONE _PAGE_BIT_GLOBAL
@@ -78,21 +70,6 @@
#endif
/*
- * _PAGE_NUMA distinguishes between a numa hinting minor fault and a page
- * that is not present. The hinting fault gathers numa placement statistics
- * (see pte_numa()). The bit is always zero when the PTE is not present.
- *
- * The bit picked must be always zero when the pmd is present and not
- * present, so that we don't lose information when we set it while
- * atomically clearing the present bit.
- */
-#ifdef CONFIG_NUMA_BALANCING
-#define _PAGE_NUMA (_AT(pteval_t, 1) << _PAGE_BIT_NUMA)
-#else
-#define _PAGE_NUMA (_AT(pteval_t, 0))
-#endif
-
-/*
* Tracking soft dirty bit when a page goes to a swap is tricky.
* We need a bit which can be stored in pte _and_ not conflict
* with swap entry format. On x86 bits 6 and 7 are *not* involved
@@ -125,8 +102,8 @@
/* Set of bits not changed in pte_modify */
#define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \
_PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY | \
- _PAGE_SOFT_DIRTY | _PAGE_NUMA)
-#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE | _PAGE_NUMA)
+ _PAGE_SOFT_DIRTY)
+#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)
/*
* The cache modes defined here are used to translate between pure SW usage
@@ -327,20 +304,6 @@ static inline pteval_t pte_flags(pte_t pte)
return native_pte_val(pte) & PTE_FLAGS_MASK;
}
-#ifdef CONFIG_NUMA_BALANCING
-/* Set of bits that distinguishes present, prot_none and numa ptes */
-#define _PAGE_NUMA_MASK (_PAGE_NUMA|_PAGE_PROTNONE|_PAGE_PRESENT)
-static inline pteval_t ptenuma_flags(pte_t pte)
-{
- return pte_flags(pte) & _PAGE_NUMA_MASK;
-}
-
-static inline pmdval_t pmdnuma_flags(pmd_t pmd)
-{
- return pmd_flags(pmd) & _PAGE_NUMA_MASK;
-}
-#endif /* CONFIG_NUMA_BALANCING */
-
#define pgprot_val(x) ((x).pgprot)
#define __pgprot(x) ((pgprot_t) { (x) } )
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index d497d08..d2ce339 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -244,10 +244,6 @@ static inline int pmd_same(pmd_t pmd_a, pmd_t pmd_b)
# define pte_accessible(mm, pte) ((void)(pte), 1)
#endif
-#ifndef pte_present_nonuma
-#define pte_present_nonuma(pte) pte_present(pte)
-#endif
-
#ifndef flush_tlb_fix_spurious_fault
#define flush_tlb_fix_spurious_fault(vma, address) flush_tlb_page(vma, address)
#endif
@@ -708,157 +704,6 @@ static inline int pmd_protnone(pmd_t pmd)
}
#endif /* CONFIG_NUMA_BALANCING */
-#ifdef CONFIG_NUMA_BALANCING
-/*
- * _PAGE_NUMA distinguishes between an unmapped page table entry, an entry that
- * is protected for PROT_NONE and a NUMA hinting fault entry. If the
- * architecture defines __PAGE_PROTNONE then it should take that into account
- * but those that do not can rely on the fact that the NUMA hinting scanner
- * skips inaccessible VMAs.
- *
- * pte/pmd_present() returns true if pte/pmd_numa returns true. Page
- * fault triggers on those regions if pte/pmd_numa returns true
- * (because _PAGE_PRESENT is not set).
- */
-#ifndef pte_numa
-static inline int pte_numa(pte_t pte)
-{
- return ptenuma_flags(pte) == _PAGE_NUMA;
-}
-#endif
-
-#ifndef pmd_numa
-static inline int pmd_numa(pmd_t pmd)
-{
- return pmdnuma_flags(pmd) == _PAGE_NUMA;
-}
-#endif
-
-/*
- * pte/pmd_mknuma sets the _PAGE_ACCESSED bitflag automatically
- * because they're called by the NUMA hinting minor page fault. If we
- * wouldn't set the _PAGE_ACCESSED bitflag here, the TLB miss handler
- * would be forced to set it later while filling the TLB after we
- * return to userland. That would trigger a second write to memory
- * that we optimize away by setting _PAGE_ACCESSED here.
- */
-#ifndef pte_mknonnuma
-static inline pte_t pte_mknonnuma(pte_t pte)
-{
- pteval_t val = pte_val(pte);
-
- val &= ~_PAGE_NUMA;
- val |= (_PAGE_PRESENT|_PAGE_ACCESSED);
- return __pte(val);
-}
-#endif
-
-#ifndef pmd_mknonnuma
-static inline pmd_t pmd_mknonnuma(pmd_t pmd)
-{
- pmdval_t val = pmd_val(pmd);
-
- val &= ~_PAGE_NUMA;
- val |= (_PAGE_PRESENT|_PAGE_ACCESSED);
-
- return __pmd(val);
-}
-#endif
-
-#ifndef pte_mknuma
-static inline pte_t pte_mknuma(pte_t pte)
-{
- pteval_t val = pte_val(pte);
-
- VM_BUG_ON(!(val & _PAGE_PRESENT));
-
- val &= ~_PAGE_PRESENT;
- val |= _PAGE_NUMA;
-
- return __pte(val);
-}
-#endif
-
-#ifndef ptep_set_numa
-static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr,
- pte_t *ptep)
-{
- pte_t ptent = *ptep;
-
- ptent = pte_mknuma(ptent);
- set_pte_at(mm, addr, ptep, ptent);
- return;
-}
-#endif
-
-#ifndef pmd_mknuma
-static inline pmd_t pmd_mknuma(pmd_t pmd)
-{
- pmdval_t val = pmd_val(pmd);
-
- val &= ~_PAGE_PRESENT;
- val |= _PAGE_NUMA;
-
- return __pmd(val);
-}
-#endif
-
-#ifndef pmdp_set_numa
-static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr,
- pmd_t *pmdp)
-{
- pmd_t pmd = *pmdp;
-
- pmd = pmd_mknuma(pmd);
- set_pmd_at(mm, addr, pmdp, pmd);
- return;
-}
-#endif
-#else
-static inline int pmd_numa(pmd_t pmd)
-{
- return 0;
-}
-
-static inline int pte_numa(pte_t pte)
-{
- return 0;
-}
-
-static inline pte_t pte_mknonnuma(pte_t pte)
-{
- return pte;
-}
-
-static inline pmd_t pmd_mknonnuma(pmd_t pmd)
-{
- return pmd;
-}
-
-static inline pte_t pte_mknuma(pte_t pte)
-{
- return pte;
-}
-
-static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr,
- pte_t *ptep)
-{
- return;
-}
-
-
-static inline pmd_t pmd_mknuma(pmd_t pmd)
-{
- return pmd;
-}
-
-static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr,
- pmd_t *pmdp)
-{
- return ;
-}
-#endif /* CONFIG_NUMA_BALANCING */
-
#endif /* CONFIG_MMU */
#endif /* !__ASSEMBLY__ */
diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index 6adfb7b..2b1fa56 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -54,7 +54,7 @@ static inline pgoff_t swp_offset(swp_entry_t entry)
/* check whether a pte points to a swap entry */
static inline int is_swap_pte(pte_t pte)
{
- return !pte_none(pte) && !pte_present_nonuma(pte) && !pte_file(pte);
+ return !pte_none(pte) && !pte_file(pte);
}
#endif
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 07/10] mm: numa: Do not trap faults on the huge zero page
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (5 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 06/10] mm: Remove remaining references to NUMA hinting bits and helpers Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 08/10] x86: mm: Restore original pte_special check Mel Gorman
` (2 subsequent siblings)
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
Faults on the huge zero page are pointless and there is a BUG_ON
to catch them during fault time. This patch reintroduces a check
that avoids marking the zero page PAGE_NONE.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/huge_mm.h | 3 ++-
mm/huge_memory.c | 13 ++++++++++++-
mm/memory.c | 1 -
mm/mprotect.c | 15 ++++++++++++++-
4 files changed, 28 insertions(+), 4 deletions(-)
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index 554bbe3..ad9051b 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -31,7 +31,8 @@ extern int move_huge_pmd(struct vm_area_struct *vma,
unsigned long new_addr, unsigned long old_end,
pmd_t *old_pmd, pmd_t *new_pmd);
extern int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, pgprot_t newprot);
+ unsigned long addr, pgprot_t newprot,
+ int prot_numa);
enum transparent_hugepage_flag {
TRANSPARENT_HUGEPAGE_FLAG,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 5618e22..ad2a3ee 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1502,7 +1502,7 @@ out:
* - HPAGE_PMD_NR is protections changed and TLB flush necessary
*/
int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, pgprot_t newprot)
+ unsigned long addr, pgprot_t newprot, int prot_numa)
{
struct mm_struct *mm = vma->vm_mm;
spinlock_t *ptl;
@@ -1510,6 +1510,17 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
pmd_t entry;
+
+ /*
+ * Avoid trapping faults against the zero page. The read-only
+ * data is likely to be read-cached on the local CPU and
+ * local/remote hits to the zero page are not interesting.
+ */
+ if (prot_numa && is_huge_zero_pmd(*pmd)) {
+ spin_unlock(ptl);
+ return 0;
+ }
+
ret = 1;
entry = pmdp_get_and_clear_notify(mm, addr, pmd);
entry = pmd_modify(entry, newprot);
diff --git a/mm/memory.c b/mm/memory.c
index debe3f4..3c50046 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3135,7 +3135,6 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
pte_unmap_unlock(ptep, ptl);
return 0;
}
- BUG_ON(is_zero_pfn(page_to_pfn(page)));
/*
* Avoid grouping on DSO/COW pages in specific and RO pages
diff --git a/mm/mprotect.c b/mm/mprotect.c
index dc65c0f..33dfafb 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -75,6 +75,19 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
oldpte = *pte;
if (pte_present(oldpte)) {
pte_t ptent;
+
+ /*
+ * Avoid trapping faults against the zero or KSM
+ * pages. See similar comment in change_huge_pmd.
+ */
+ if (prot_numa) {
+ struct page *page;
+
+ page = vm_normal_page(vma, addr, oldpte);
+ if (!page || PageKsm(page))
+ continue;
+ }
+
ptent = ptep_modify_prot_start(mm, addr, pte);
ptent = pte_modify(ptent, newprot);
@@ -141,7 +154,7 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
split_huge_page_pmd(vma, addr, pmd);
else {
int nr_ptes = change_huge_pmd(vma, pmd, addr,
- newprot);
+ newprot, prot_numa);
if (nr_ptes) {
if (nr_ptes == HPAGE_PMD_NR) {
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 08/10] x86: mm: Restore original pte_special check
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (6 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 07/10] mm: numa: Do not trap faults on the huge zero page Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 09/10] mm: numa: Add paranoid check around pte_protnone_numa Mel Gorman
2015-01-05 10:54 ` [PATCH 10/10] mm: numa: Avoid unnecessary TLB flushes when setting NUMA hinting entries Mel Gorman
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
Commit b38af4721f59 ("x86,mm: fix pte_special versus pte_numa") adjusted
the pte_special check to take into account that a special pte had SPECIAL
and neither PRESENT nor PROTNONE. Now that NUMA hinting PTEs are no
longer modifying _PAGE_PRESENT it should be safe to restore the original
pte_special behaviour.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
arch/x86/include/asm/pgtable.h | 8 +-------
1 file changed, 1 insertion(+), 7 deletions(-)
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index b9a13e9..4673d6e 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -137,13 +137,7 @@ static inline int pte_exec(pte_t pte)
static inline int pte_special(pte_t pte)
{
- /*
- * See CONFIG_NUMA_BALANCING pte_numa in include/asm-generic/pgtable.h.
- * On x86 we have _PAGE_BIT_NUMA == _PAGE_BIT_GLOBAL+1 ==
- * __PAGE_BIT_SOFTW1 == _PAGE_BIT_SPECIAL.
- */
- return (pte_flags(pte) & _PAGE_SPECIAL) &&
- (pte_flags(pte) & (_PAGE_PRESENT|_PAGE_PROTNONE));
+ return pte_flags(pte) & _PAGE_SPECIAL;
}
static inline unsigned long pte_pfn(pte_t pte)
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 09/10] mm: numa: Add paranoid check around pte_protnone_numa
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (7 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 08/10] x86: mm: Restore original pte_special check Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
2015-01-05 10:54 ` [PATCH 10/10] mm: numa: Avoid unnecessary TLB flushes when setting NUMA hinting entries Mel Gorman
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
pte_protnone_numa is only safe to use after VMA checks for PROT_NONE are
complete. Treating a real PROT_NONE PTE as a NUMA hinting fault is going
to result in strangeness so add a check for it. BUG_ON looks like overkill
but if this is hit then it's a serious bug that could result in corruption
so do not even try recovering. It would have been more comprehensive to
check VMA flags in pte_protnone_numa but it would have made the API ugly
just for a debugging check.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
mm/huge_memory.c | 3 +++
mm/memory.c | 3 +++
2 files changed, 6 insertions(+)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index ad2a3ee..8546654 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1273,6 +1273,9 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
bool migrated = false;
int flags = 0;
+ /* A PROT_NONE fault should not end up here */
+ BUG_ON(!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)));
+
ptl = pmd_lock(mm, pmdp);
if (unlikely(!pmd_same(pmd, *pmdp)))
goto out_unlock;
diff --git a/mm/memory.c b/mm/memory.c
index 3c50046..9df2d09 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3108,6 +3108,9 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
bool migrated = false;
int flags = 0;
+ /* A PROT_NONE fault should not end up here */
+ BUG_ON(!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)));
+
/*
* The "pte" at this point cannot be used safely without
* validation through pte_unmap_same(). It's of NUMA type but
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH 10/10] mm: numa: Avoid unnecessary TLB flushes when setting NUMA hinting entries
2015-01-05 10:54 [PATCH 0/10] Replace _PAGE_NUMA with PAGE_NONE protections v5 Mel Gorman
` (8 preceding siblings ...)
2015-01-05 10:54 ` [PATCH 09/10] mm: numa: Add paranoid check around pte_protnone_numa Mel Gorman
@ 2015-01-05 10:54 ` Mel Gorman
9 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2015-01-05 10:54 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Hugh Dickins, Linux Kernel, Linux-MM, Ingo Molnar,
Aneesh Kumar, Sasha Levin, LinuxPPC-dev, Kirill Shutemov,
Mel Gorman
If a PTE or PMD is already marked NUMA when scanning to mark entries
for NUMA hinting then it is not necessary to update the entry and
incur a TLB flush penalty. Avoid the avoidhead where possible.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
mm/huge_memory.c | 14 ++++++++------
mm/mprotect.c | 4 ++++
2 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 8546654..f2bf521 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1524,12 +1524,14 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
return 0;
}
- ret = 1;
- entry = pmdp_get_and_clear_notify(mm, addr, pmd);
- entry = pmd_modify(entry, newprot);
- ret = HPAGE_PMD_NR;
- set_pmd_at(mm, addr, pmd, entry);
- BUG_ON(pmd_write(entry));
+ if (!prot_numa || !pmd_protnone(*pmd)) {
+ ret = 1;
+ entry = pmdp_get_and_clear_notify(mm, addr, pmd);
+ entry = pmd_modify(entry, newprot);
+ ret = HPAGE_PMD_NR;
+ set_pmd_at(mm, addr, pmd, entry);
+ BUG_ON(pmd_write(entry));
+ }
spin_unlock(ptl);
}
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 33dfafb..109e7aa 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -86,6 +86,10 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
page = vm_normal_page(vma, addr, oldpte);
if (!page || PageKsm(page))
continue;
+
+ /* Avoid TLB flush if possible */
+ if (pte_protnone(oldpte))
+ continue;
}
ptent = ptep_modify_prot_start(mm, addr, pte);
--
2.1.2
^ permalink raw reply related [flat|nested] 14+ messages in thread