powerpc/mm: support page table check

On creation and clearing of a page table mapping, instrument such calls by
invoking page_table_check_pte_set and page_table_check_pte_clear
respectively.  These calls serve as a sanity check against illegal
mappings.

Enable ARCH_SUPPORTS_PAGE_TABLE_CHECK on powerpc, except when HUGETLB_PAGE
is enabled (powerpc has some weirdness in how it implements
set_huge_pte_at(), which may require some further work).

See also:

riscv support in commit 3fee229a8e ("riscv/mm: enable
ARCH_SUPPORTS_PAGE_TABLE_CHECK")
arm64 in commit 42b2547137 ("arm64/mm: enable
ARCH_SUPPORTS_PAGE_TABLE_CHECK")
x86_64 in commit d283d422c6 ("x86: mm: add x86_64 support for page table
check")

[ajd@linux.ibm.com: rebase, add additional instrumentation, misc fixes]
Link: https://lkml.kernel.org/r/20251219-pgtable_check_v18rebase-v18-12-755bc151a50b@linux.ibm.com
Signed-off-by: Rohan McLure <rmclure@linux.ibm.com>
Signed-off-by: Andrew Donnellan <ajd@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Alexandre Ghiti <alexghiti@rivosinc.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: Donet Tom <donettom@linux.ibm.com>
Cc: Guo Weikang <guoweikang.kernel@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Kevin Brodsky <kevin.brodsky@arm.com>
Cc: Magnus Lindholm <linmag7@gmail.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Miehlbradt <nicholas@linux.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Paul Mackerras <paulus@ozlabs.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: "Ritesh Harjani (IBM)" <ritesh.list@gmail.com>
Cc: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
Cc: Thomas Huth <thuth@redhat.com>
Cc: "Vishal Moola (Oracle)" <vishal.moola@gmail.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
master
Rohan McLure 2025-12-19 04:09:44 +11:00 committed by Andrew Morton
parent 2360f523a4
commit 641d47d4c9
8 changed files with 73 additions and 18 deletions

View File

@ -172,6 +172,7 @@ config PPC
select ARCH_STACKWALK
select ARCH_SUPPORTS_ATOMIC_RMW
select ARCH_SUPPORTS_DEBUG_PAGEALLOC if PPC_BOOK3S || PPC_8xx
select ARCH_SUPPORTS_PAGE_TABLE_CHECK if !HUGETLB_PAGE
select ARCH_SUPPORTS_SCHED_MC if SMP
select ARCH_SUPPORTS_SCHED_SMT if PPC64 && SMP
select SCHED_MC if ARCH_SUPPORTS_SCHED_MC

View File

@ -202,6 +202,7 @@ void unmap_kernel_page(unsigned long va);
#ifndef __ASSEMBLER__
#include <linux/sched.h>
#include <linux/threads.h>
#include <linux/page_table_check.h>
/* Bits to mask out from a PGD to get to the PUD page */
#define PGD_MASKED_BITS 0
@ -315,7 +316,11 @@ static inline int __ptep_test_and_clear_young(struct mm_struct *mm,
static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
pte_t *ptep)
{
return __pte(pte_update(mm, addr, ptep, ~_PAGE_HASHPTE, 0, 0));
pte_t old_pte = __pte(pte_update(mm, addr, ptep, ~_PAGE_HASHPTE, 0, 0));
page_table_check_pte_clear(mm, addr, old_pte);
return old_pte;
}
#define __HAVE_ARCH_PTEP_SET_WRPROTECT

View File

@ -144,6 +144,8 @@
#define PAGE_KERNEL_ROX __pgprot(_PAGE_BASE | _PAGE_KERNEL_ROX)
#ifndef __ASSEMBLER__
#include <linux/page_table_check.h>
/*
* page table defines
*/
@ -416,8 +418,11 @@ static inline void huge_ptep_set_wrprotect(struct mm_struct *mm,
static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
unsigned long addr, pte_t *ptep)
{
unsigned long old = pte_update(mm, addr, ptep, ~0UL, 0, 0);
return __pte(old);
pte_t old_pte = __pte(pte_update(mm, addr, ptep, ~0UL, 0, 0));
page_table_check_pte_clear(mm, addr, old_pte);
return old_pte;
}
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
@ -426,11 +431,16 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
pte_t *ptep, int full)
{
if (full && radix_enabled()) {
pte_t old_pte;
/*
* We know that this is a full mm pte clear and
* hence can be sure there is no parallel set_pte.
*/
return radix__ptep_get_and_clear_full(mm, addr, ptep, full);
old_pte = radix__ptep_get_and_clear_full(mm, addr, ptep, full);
page_table_check_pte_clear(mm, addr, old_pte);
return old_pte;
}
return ptep_get_and_clear(mm, addr, ptep);
}
@ -1301,19 +1311,34 @@ extern int pudp_test_and_clear_young(struct vm_area_struct *vma,
static inline pmd_t pmdp_huge_get_and_clear(struct mm_struct *mm,
unsigned long addr, pmd_t *pmdp)
{
if (radix_enabled())
return radix__pmdp_huge_get_and_clear(mm, addr, pmdp);
return hash__pmdp_huge_get_and_clear(mm, addr, pmdp);
pmd_t old_pmd;
if (radix_enabled()) {
old_pmd = radix__pmdp_huge_get_and_clear(mm, addr, pmdp);
} else {
old_pmd = hash__pmdp_huge_get_and_clear(mm, addr, pmdp);
}
page_table_check_pmd_clear(mm, addr, old_pmd);
return old_pmd;
}
#define __HAVE_ARCH_PUDP_HUGE_GET_AND_CLEAR
static inline pud_t pudp_huge_get_and_clear(struct mm_struct *mm,
unsigned long addr, pud_t *pudp)
{
if (radix_enabled())
return radix__pudp_huge_get_and_clear(mm, addr, pudp);
BUG();
return *pudp;
pud_t old_pud;
if (radix_enabled()) {
old_pud = radix__pudp_huge_get_and_clear(mm, addr, pudp);
} else {
BUG();
}
page_table_check_pud_clear(mm, addr, old_pud);
return old_pud;
}
static inline pmd_t pmdp_collapse_flush(struct vm_area_struct *vma,

View File

@ -29,6 +29,8 @@ static inline pte_basic_t pte_update(struct mm_struct *mm, unsigned long addr, p
#ifndef __ASSEMBLER__
#include <linux/page_table_check.h>
extern int icache_44x_need_flush;
#ifndef pte_huge_size
@ -122,7 +124,11 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
pte_t *ptep)
{
return __pte(pte_update(mm, addr, ptep, ~0UL, 0, 0));
pte_t old_pte = __pte(pte_update(mm, addr, ptep, ~0UL, 0, 0));
page_table_check_pte_clear(mm, addr, old_pte);
return old_pte;
}
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR

View File

@ -8,6 +8,7 @@
#include <linux/sched.h>
#include <linux/mm_types.h>
#include <linux/mm.h>
#include <linux/page_table_check.h>
#include <linux/stop_machine.h>
#include <asm/sections.h>
@ -230,6 +231,9 @@ pmd_t hash__pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long addres
pmd = *pmdp;
pmd_clear(pmdp);
page_table_check_pmd_clear(vma->vm_mm, address, pmd);
/*
* Wait for all pending hash_page to finish. This is needed
* in case of subpage collapse. When we collapse normal pages

View File

@ -10,6 +10,7 @@
#include <linux/pkeys.h>
#include <linux/debugfs.h>
#include <linux/proc_fs.h>
#include <linux/page_table_check.h>
#include <asm/pgalloc.h>
#include <asm/tlb.h>
@ -127,6 +128,7 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
WARN_ON(!(pmd_leaf(pmd)));
#endif
trace_hugepage_set_pmd(addr, pmd_val(pmd));
page_table_check_pmd_set(mm, addr, pmdp, pmd);
return set_pte_at_unchecked(mm, addr, pmdp_ptep(pmdp), pmd_pte(pmd));
}
@ -144,6 +146,7 @@ void set_pud_at(struct mm_struct *mm, unsigned long addr,
WARN_ON(!(pud_leaf(pud)));
#endif
trace_hugepage_set_pud(addr, pud_val(pud));
page_table_check_pud_set(mm, addr, pudp, pud);
return set_pte_at_unchecked(mm, addr, pudp_ptep(pudp), pud_pte(pud));
}
@ -179,23 +182,27 @@ void serialize_against_pte_lookup(struct mm_struct *mm)
pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmdp)
{
unsigned long old_pmd;
pmd_t old_pmd;
VM_WARN_ON_ONCE(!pmd_present(*pmdp));
old_pmd = pmd_hugepage_update(vma->vm_mm, address, pmdp, _PAGE_PRESENT, _PAGE_INVALID);
old_pmd = __pmd(pmd_hugepage_update(vma->vm_mm, address, pmdp, _PAGE_PRESENT, _PAGE_INVALID));
flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
return __pmd(old_pmd);
page_table_check_pmd_clear(vma->vm_mm, address, old_pmd);
return old_pmd;
}
pud_t pudp_invalidate(struct vm_area_struct *vma, unsigned long address,
pud_t *pudp)
{
unsigned long old_pud;
pud_t old_pud;
VM_WARN_ON_ONCE(!pud_present(*pudp));
old_pud = pud_hugepage_update(vma->vm_mm, address, pudp, _PAGE_PRESENT, _PAGE_INVALID);
old_pud = __pud(pud_hugepage_update(vma->vm_mm, address, pudp, _PAGE_PRESENT, _PAGE_INVALID));
flush_pud_tlb_range(vma, address, address + HPAGE_PUD_SIZE);
return __pud(old_pud);
page_table_check_pud_clear(vma->vm_mm, address, old_pud);
return old_pud;
}
pmd_t pmdp_huge_get_and_clear_full(struct vm_area_struct *vma,

View File

@ -14,6 +14,7 @@
#include <linux/of.h>
#include <linux/of_fdt.h>
#include <linux/mm.h>
#include <linux/page_table_check.h>
#include <linux/hugetlb.h>
#include <linux/string_helpers.h>
#include <linux/memory.h>
@ -1474,6 +1475,8 @@ pmd_t radix__pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long addre
pmd = *pmdp;
pmd_clear(pmdp);
page_table_check_pmd_clear(vma->vm_mm, address, pmd);
radix__flush_tlb_collapsed_pmd(vma->vm_mm, address);
return pmd;

View File

@ -22,6 +22,7 @@
#include <linux/mm.h>
#include <linux/percpu.h>
#include <linux/hardirq.h>
#include <linux/page_table_check.h>
#include <linux/hugetlb.h>
#include <asm/tlbflush.h>
#include <asm/tlb.h>
@ -206,6 +207,9 @@ void set_ptes(struct mm_struct *mm, unsigned long addr, pte_t *ptep,
* and not hw_valid ptes. Hence there is no translation cache flush
* involved that need to be batched.
*/
page_table_check_ptes_set(mm, addr, ptep, pte, nr);
for (;;) {
/*