diff options
Diffstat (limited to 'arch/s390/include/asm/pgtable.h')
| -rw-r--r-- | arch/s390/include/asm/pgtable.h | 691 |
1 files changed, 455 insertions, 236 deletions
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 7b03037a8475..1299b56e43f6 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -17,11 +17,16 @@ #include <linux/page-flags.h> #include <linux/radix-tree.h> #include <linux/atomic.h> +#include <asm/sections.h> +#include <asm/ctlreg.h> #include <asm/bug.h> #include <asm/page.h> +#include <asm/uv.h> extern pgd_t swapper_pg_dir[]; +extern pgd_t invalid_pg_dir[]; extern void paging_init(void); +extern struct ctlreg s390_invalid_asce; enum { PG_DIRECT_MAP_4K = 0, @@ -30,7 +35,7 @@ enum { PG_DIRECT_MAP_MAX }; -extern atomic_long_t direct_pages_count[PG_DIRECT_MAP_MAX]; +extern atomic_long_t __bootdata_preserved(direct_pages_count[PG_DIRECT_MAP_MAX]); static inline void update_page_count(int level, long count) { @@ -38,14 +43,12 @@ static inline void update_page_count(int level, long count) atomic_long_add(count, &direct_pages_count[level]); } -struct seq_file; -void arch_report_meminfo(struct seq_file *m); - /* * The S390 doesn't have any external MMU info: the kernel page * tables contain all the necessary information. */ #define update_mmu_cache(vma, address, ptep) do { } while (0) +#define update_mmu_cache_range(vmf, vma, addr, ptep, nr) do { } while (0) #define update_mmu_cache_pmd(vma, address, ptep) do { } while (0) /* @@ -63,36 +66,33 @@ extern unsigned long zero_page_mask; /* TODO: s390 cannot support io_remap_pfn_range... */ -#define FIRST_USER_ADDRESS 0UL - #define pte_ERROR(e) \ - printk("%s:%d: bad pte %p.\n", __FILE__, __LINE__, (void *) pte_val(e)) + pr_err("%s:%d: bad pte %016lx.\n", __FILE__, __LINE__, pte_val(e)) #define pmd_ERROR(e) \ - printk("%s:%d: bad pmd %p.\n", __FILE__, __LINE__, (void *) pmd_val(e)) + pr_err("%s:%d: bad pmd %016lx.\n", __FILE__, __LINE__, pmd_val(e)) #define pud_ERROR(e) \ - printk("%s:%d: bad pud %p.\n", __FILE__, __LINE__, (void *) pud_val(e)) + pr_err("%s:%d: bad pud %016lx.\n", __FILE__, __LINE__, pud_val(e)) #define p4d_ERROR(e) \ - printk("%s:%d: bad p4d %p.\n", __FILE__, __LINE__, (void *) p4d_val(e)) + pr_err("%s:%d: bad p4d %016lx.\n", __FILE__, __LINE__, p4d_val(e)) #define pgd_ERROR(e) \ - printk("%s:%d: bad pgd %p.\n", __FILE__, __LINE__, (void *) pgd_val(e)) + pr_err("%s:%d: bad pgd %016lx.\n", __FILE__, __LINE__, pgd_val(e)) /* * The vmalloc and module area will always be on the topmost area of the - * kernel mapping. We reserve 128GB (64bit) for vmalloc and modules. - * On 64 bit kernels we have a 2GB area at the top of the vmalloc area where - * modules will reside. That makes sure that inter module branches always - * happen without trampolines and in addition the placement within a 2GB frame - * is branch prediction unit friendly. + * kernel mapping. 512GB are reserved for vmalloc by default. + * At the top of the vmalloc area a 2GB area is reserved where modules + * will reside. That makes sure that inter module branches always + * happen without trampolines and in addition the placement within a + * 2GB frame is branch prediction unit friendly. */ -extern unsigned long VMALLOC_START; -extern unsigned long VMALLOC_END; -#define VMALLOC_DEFAULT_SIZE ((128UL << 30) - MODULES_LEN) -extern struct page *vmemmap; - -#define VMEM_MAX_PHYS ((unsigned long) vmemmap) - -extern unsigned long MODULES_VADDR; -extern unsigned long MODULES_END; +extern unsigned long __bootdata_preserved(VMALLOC_START); +extern unsigned long __bootdata_preserved(VMALLOC_END); +#define VMALLOC_DEFAULT_SIZE ((512UL << 30) - MODULES_LEN) +extern struct page *__bootdata_preserved(vmemmap); +extern unsigned long __bootdata_preserved(vmemmap_size); + +extern unsigned long __bootdata_preserved(MODULES_VADDR); +extern unsigned long __bootdata_preserved(MODULES_END); #define MODULES_VADDR MODULES_VADDR #define MODULES_END MODULES_END #define MODULES_LEN (1UL << 31) @@ -179,11 +179,21 @@ static inline int is_module_addr(void *addr) #define _PAGE_SOFT_DIRTY 0x000 #endif +#define _PAGE_SW_BITS 0xffUL /* All SW bits */ + +#define _PAGE_SWP_EXCLUSIVE _PAGE_LARGE /* SW pte exclusive swap bit */ + /* Set of bits not changed in pte_modify */ #define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_SPECIAL | _PAGE_DIRTY | \ _PAGE_YOUNG | _PAGE_SOFT_DIRTY) /* + * Mask of bits that must not be changed with RDP. Allow only _PAGE_PROTECT + * HW bit and all SW bits. + */ +#define _PAGE_RDP_MASK ~(_PAGE_PROTECT | _PAGE_SW_BITS) + +/* * handle_pte_fault uses pte_present and pte_none to find out the pte type * WITHOUT holding the page table lock. The _PAGE_PRESENT bit is used to * distinguish present from not-present ptes. It is changed only with the page @@ -341,8 +351,6 @@ static inline int is_module_addr(void *addr) #define PTRS_PER_P4D _CRST_ENTRIES #define PTRS_PER_PGD _CRST_ENTRIES -#define MAX_PTRS_PER_P4D PTRS_PER_P4D - /* * Segment table and region3 table entry encoding * (R = read-only, I = invalid, y = young bit): @@ -422,23 +430,6 @@ static inline int is_module_addr(void *addr) * implies read permission. */ /*xwr*/ -#define __P000 PAGE_NONE -#define __P001 PAGE_RO -#define __P010 PAGE_RO -#define __P011 PAGE_RO -#define __P100 PAGE_RX -#define __P101 PAGE_RX -#define __P110 PAGE_RX -#define __P111 PAGE_RX - -#define __S000 PAGE_NONE -#define __S001 PAGE_RO -#define __S010 PAGE_RW -#define __S011 PAGE_RW -#define __S100 PAGE_RX -#define __S101 PAGE_RX -#define __S110 PAGE_RWX -#define __S111 PAGE_RWX /* * Segment entry (large page) protection definitions. @@ -492,6 +483,12 @@ static inline int is_module_addr(void *addr) _REGION3_ENTRY_YOUNG | \ _REGION_ENTRY_PROTECT | \ _REGION_ENTRY_NOEXEC) +#define REGION3_KERNEL_EXEC __pgprot(_REGION_ENTRY_TYPE_R3 | \ + _REGION3_ENTRY_LARGE | \ + _REGION3_ENTRY_READ | \ + _REGION3_ENTRY_WRITE | \ + _REGION3_ENTRY_YOUNG | \ + _REGION3_ENTRY_DIRTY) static inline bool mm_p4d_folded(struct mm_struct *mm) { @@ -520,6 +517,15 @@ static inline int mm_has_pgste(struct mm_struct *mm) return 0; } +static inline int mm_is_protected(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (unlikely(atomic_read(&mm->context.protected_count))) + return 1; +#endif + return 0; +} + static inline int mm_alloc_pgste(struct mm_struct *mm) { #ifdef CONFIG_PGSTE @@ -529,6 +535,36 @@ static inline int mm_alloc_pgste(struct mm_struct *mm) return 0; } +static inline pte_t clear_pte_bit(pte_t pte, pgprot_t prot) +{ + return __pte(pte_val(pte) & ~pgprot_val(prot)); +} + +static inline pte_t set_pte_bit(pte_t pte, pgprot_t prot) +{ + return __pte(pte_val(pte) | pgprot_val(prot)); +} + +static inline pmd_t clear_pmd_bit(pmd_t pmd, pgprot_t prot) +{ + return __pmd(pmd_val(pmd) & ~pgprot_val(prot)); +} + +static inline pmd_t set_pmd_bit(pmd_t pmd, pgprot_t prot) +{ + return __pmd(pmd_val(pmd) | pgprot_val(prot)); +} + +static inline pud_t clear_pud_bit(pud_t pud, pgprot_t prot) +{ + return __pud(pud_val(pud) & ~pgprot_val(prot)); +} + +static inline pud_t set_pud_bit(pud_t pud, pgprot_t prot) +{ + return __pud(pud_val(pud) | pgprot_val(prot)); +} + /* * In the case that a guest uses storage keys * faults should no longer be backed by zero pages @@ -545,27 +581,25 @@ static inline int mm_uses_skeys(struct mm_struct *mm) static inline void csp(unsigned int *ptr, unsigned int old, unsigned int new) { - register unsigned long reg2 asm("2") = old; - register unsigned long reg3 asm("3") = new; + union register_pair r1 = { .even = old, .odd = new, }; unsigned long address = (unsigned long)ptr | 1; asm volatile( - " csp %0,%3" - : "+d" (reg2), "+m" (*ptr) - : "d" (reg3), "d" (address) + " csp %[r1],%[address]" + : [r1] "+&d" (r1.pair), "+m" (*ptr) + : [address] "d" (address) : "cc"); } static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new) { - register unsigned long reg2 asm("2") = old; - register unsigned long reg3 asm("3") = new; + union register_pair r1 = { .even = old, .odd = new, }; unsigned long address = (unsigned long)ptr | 1; asm volatile( - " .insn rre,0xb98a0000,%0,%3" - : "+d" (reg2), "+m" (*ptr) - : "d" (reg3), "d" (address) + " cspg %[r1],%[address]" + : [r1] "+&d" (r1.pair), "+m" (*ptr) + : [address] "d" (address) : "cc"); } @@ -576,17 +610,15 @@ static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new #define CRDTE_DTT_REGION1 0x1cUL static inline void crdte(unsigned long old, unsigned long new, - unsigned long table, unsigned long dtt, + unsigned long *table, unsigned long dtt, unsigned long address, unsigned long asce) { - register unsigned long reg2 asm("2") = old; - register unsigned long reg3 asm("3") = new; - register unsigned long reg4 asm("4") = table | dtt; - register unsigned long reg5 asm("5") = address; + union register_pair r1 = { .even = old, .odd = new, }; + union register_pair r2 = { .even = __pa(table) | dtt, .odd = address, }; - asm volatile(".insn rrf,0xb98f0000,%0,%2,%4,0" - : "+d" (reg2) - : "d" (reg3), "d" (reg4), "d" (reg5), "a" (asce) + asm volatile(".insn rrf,0xb98f0000,%[r1],%[r2],%[asce],0" + : [r1] "+&d" (r1.pair) + : [r2] "d" (r2.pair), [asce] "a" (asce) : "memory", "cc"); } @@ -673,6 +705,7 @@ static inline int pud_none(pud_t pud) return pud_val(pud) == _REGION3_ENTRY_EMPTY; } +#define pud_leaf pud_large static inline int pud_large(pud_t pud) { if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) != _REGION_ENTRY_TYPE_R3) @@ -680,16 +713,7 @@ static inline int pud_large(pud_t pud) return !!(pud_val(pud) & _REGION3_ENTRY_LARGE); } -static inline unsigned long pud_pfn(pud_t pud) -{ - unsigned long origin_mask; - - origin_mask = _REGION_ENTRY_ORIGIN; - if (pud_large(pud)) - origin_mask = _REGION3_ENTRY_ORIGIN_LARGE; - return (pud_val(pud) & origin_mask) >> PAGE_SHIFT; -} - +#define pmd_leaf pmd_large static inline int pmd_large(pmd_t pmd) { return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0; @@ -734,27 +758,25 @@ static inline int pmd_none(pmd_t pmd) return pmd_val(pmd) == _SEGMENT_ENTRY_EMPTY; } -static inline unsigned long pmd_pfn(pmd_t pmd) -{ - unsigned long origin_mask; - - origin_mask = _SEGMENT_ENTRY_ORIGIN; - if (pmd_large(pmd)) - origin_mask = _SEGMENT_ENTRY_ORIGIN_LARGE; - return (pmd_val(pmd) & origin_mask) >> PAGE_SHIFT; -} - #define pmd_write pmd_write static inline int pmd_write(pmd_t pmd) { return (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) != 0; } +#define pud_write pud_write +static inline int pud_write(pud_t pud) +{ + return (pud_val(pud) & _REGION3_ENTRY_WRITE) != 0; +} + +#define pmd_dirty pmd_dirty static inline int pmd_dirty(pmd_t pmd) { return (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) != 0; } +#define pmd_young pmd_young static inline int pmd_young(pmd_t pmd) { return (pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG) != 0; @@ -803,6 +825,21 @@ static inline int pmd_protnone(pmd_t pmd) } #endif +static inline int pte_swp_exclusive(pte_t pte) +{ + return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; +} + +static inline pte_t pte_swp_mkexclusive(pte_t pte) +{ + return set_pte_bit(pte, __pgprot(_PAGE_SWP_EXCLUSIVE)); +} + +static inline pte_t pte_swp_clear_exclusive(pte_t pte) +{ + return clear_pte_bit(pte, __pgprot(_PAGE_SWP_EXCLUSIVE)); +} + static inline int pte_soft_dirty(pte_t pte) { return pte_val(pte) & _PAGE_SOFT_DIRTY; @@ -811,15 +848,13 @@ static inline int pte_soft_dirty(pte_t pte) static inline pte_t pte_mksoft_dirty(pte_t pte) { - pte_val(pte) |= _PAGE_SOFT_DIRTY; - return pte; + return set_pte_bit(pte, __pgprot(_PAGE_SOFT_DIRTY)); } #define pte_swp_mksoft_dirty pte_mksoft_dirty static inline pte_t pte_clear_soft_dirty(pte_t pte) { - pte_val(pte) &= ~_PAGE_SOFT_DIRTY; - return pte; + return clear_pte_bit(pte, __pgprot(_PAGE_SOFT_DIRTY)); } #define pte_swp_clear_soft_dirty pte_clear_soft_dirty @@ -830,14 +865,12 @@ static inline int pmd_soft_dirty(pmd_t pmd) static inline pmd_t pmd_mksoft_dirty(pmd_t pmd) { - pmd_val(pmd) |= _SEGMENT_ENTRY_SOFT_DIRTY; - return pmd; + return set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_SOFT_DIRTY)); } static inline pmd_t pmd_clear_soft_dirty(pmd_t pmd) { - pmd_val(pmd) &= ~_SEGMENT_ENTRY_SOFT_DIRTY; - return pmd; + return clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_SOFT_DIRTY)); } /* @@ -866,35 +899,79 @@ static inline int pte_unused(pte_t pte) } /* + * Extract the pgprot value from the given pte while at the same time making it + * usable for kernel address space mappings where fault driven dirty and + * young/old accounting is not supported, i.e _PAGE_PROTECT and _PAGE_INVALID + * must not be set. + */ +static inline pgprot_t pte_pgprot(pte_t pte) +{ + unsigned long pte_flags = pte_val(pte) & _PAGE_CHG_MASK; + + if (pte_write(pte)) + pte_flags |= pgprot_val(PAGE_KERNEL); + else + pte_flags |= pgprot_val(PAGE_KERNEL_RO); + pte_flags |= pte_val(pte) & mio_wb_bit_mask; + + return __pgprot(pte_flags); +} + +/* * pgd/pmd/pte modification functions */ +static inline void set_pgd(pgd_t *pgdp, pgd_t pgd) +{ + WRITE_ONCE(*pgdp, pgd); +} + +static inline void set_p4d(p4d_t *p4dp, p4d_t p4d) +{ + WRITE_ONCE(*p4dp, p4d); +} + +static inline void set_pud(pud_t *pudp, pud_t pud) +{ + WRITE_ONCE(*pudp, pud); +} + +static inline void set_pmd(pmd_t *pmdp, pmd_t pmd) +{ + WRITE_ONCE(*pmdp, pmd); +} + +static inline void set_pte(pte_t *ptep, pte_t pte) +{ + WRITE_ONCE(*ptep, pte); +} + static inline void pgd_clear(pgd_t *pgd) { if ((pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R1) - pgd_val(*pgd) = _REGION1_ENTRY_EMPTY; + set_pgd(pgd, __pgd(_REGION1_ENTRY_EMPTY)); } static inline void p4d_clear(p4d_t *p4d) { if ((p4d_val(*p4d) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R2) - p4d_val(*p4d) = _REGION2_ENTRY_EMPTY; + set_p4d(p4d, __p4d(_REGION2_ENTRY_EMPTY)); } static inline void pud_clear(pud_t *pud) { if ((pud_val(*pud) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) - pud_val(*pud) = _REGION3_ENTRY_EMPTY; + set_pud(pud, __pud(_REGION3_ENTRY_EMPTY)); } static inline void pmd_clear(pmd_t *pmdp) { - pmd_val(*pmdp) = _SEGMENT_ENTRY_EMPTY; + set_pmd(pmdp, __pmd(_SEGMENT_ENTRY_EMPTY)); } static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - pte_val(*ptep) = _PAGE_INVALID; + set_pte(ptep, __pte(_PAGE_INVALID)); } /* @@ -903,79 +980,74 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *pt */ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { - pte_val(pte) &= _PAGE_CHG_MASK; - pte_val(pte) |= pgprot_val(newprot); + pte = clear_pte_bit(pte, __pgprot(~_PAGE_CHG_MASK)); + pte = set_pte_bit(pte, newprot); /* * newprot for PAGE_NONE, PAGE_RO, PAGE_RX, PAGE_RW and PAGE_RWX * has the invalid bit set, clear it again for readable, young pages */ if ((pte_val(pte) & _PAGE_YOUNG) && (pte_val(pte) & _PAGE_READ)) - pte_val(pte) &= ~_PAGE_INVALID; + pte = clear_pte_bit(pte, __pgprot(_PAGE_INVALID)); /* * newprot for PAGE_RO, PAGE_RX, PAGE_RW and PAGE_RWX has the page * protection bit set, clear it again for writable, dirty pages */ if ((pte_val(pte) & _PAGE_DIRTY) && (pte_val(pte) & _PAGE_WRITE)) - pte_val(pte) &= ~_PAGE_PROTECT; + pte = clear_pte_bit(pte, __pgprot(_PAGE_PROTECT)); return pte; } static inline pte_t pte_wrprotect(pte_t pte) { - pte_val(pte) &= ~_PAGE_WRITE; - pte_val(pte) |= _PAGE_PROTECT; - return pte; + pte = clear_pte_bit(pte, __pgprot(_PAGE_WRITE)); + return set_pte_bit(pte, __pgprot(_PAGE_PROTECT)); } -static inline pte_t pte_mkwrite(pte_t pte) +static inline pte_t pte_mkwrite_novma(pte_t pte) { - pte_val(pte) |= _PAGE_WRITE; + pte = set_pte_bit(pte, __pgprot(_PAGE_WRITE)); if (pte_val(pte) & _PAGE_DIRTY) - pte_val(pte) &= ~_PAGE_PROTECT; + pte = clear_pte_bit(pte, __pgprot(_PAGE_PROTECT)); return pte; } static inline pte_t pte_mkclean(pte_t pte) { - pte_val(pte) &= ~_PAGE_DIRTY; - pte_val(pte) |= _PAGE_PROTECT; - return pte; + pte = clear_pte_bit(pte, __pgprot(_PAGE_DIRTY)); + return set_pte_bit(pte, __pgprot(_PAGE_PROTECT)); } static inline pte_t pte_mkdirty(pte_t pte) { - pte_val(pte) |= _PAGE_DIRTY | _PAGE_SOFT_DIRTY; + pte = set_pte_bit(pte, __pgprot(_PAGE_DIRTY | _PAGE_SOFT_DIRTY)); if (pte_val(pte) & _PAGE_WRITE) - pte_val(pte) &= ~_PAGE_PROTECT; + pte = clear_pte_bit(pte, __pgprot(_PAGE_PROTECT)); return pte; } static inline pte_t pte_mkold(pte_t pte) { - pte_val(pte) &= ~_PAGE_YOUNG; - pte_val(pte) |= _PAGE_INVALID; - return pte; + pte = clear_pte_bit(pte, __pgprot(_PAGE_YOUNG)); + return set_pte_bit(pte, __pgprot(_PAGE_INVALID)); } static inline pte_t pte_mkyoung(pte_t pte) { - pte_val(pte) |= _PAGE_YOUNG; + pte = set_pte_bit(pte, __pgprot(_PAGE_YOUNG)); if (pte_val(pte) & _PAGE_READ) - pte_val(pte) &= ~_PAGE_INVALID; + pte = clear_pte_bit(pte, __pgprot(_PAGE_INVALID)); return pte; } static inline pte_t pte_mkspecial(pte_t pte) { - pte_val(pte) |= _PAGE_SPECIAL; - return pte; + return set_pte_bit(pte, __pgprot(_PAGE_SPECIAL)); } #ifdef CONFIG_HUGETLB_PAGE static inline pte_t pte_mkhuge(pte_t pte) { - pte_val(pte) |= _PAGE_LARGE; - return pte; + return set_pte_bit(pte, __pgprot(_PAGE_LARGE)); } #endif @@ -985,16 +1057,29 @@ static inline pte_t pte_mkhuge(pte_t pte) #define IPTE_NODAT 0x400 #define IPTE_GUEST_ASCE 0x800 +static __always_inline void __ptep_rdp(unsigned long addr, pte_t *ptep, + unsigned long opt, unsigned long asce, + int local) +{ + unsigned long pto; + + pto = __pa(ptep) & ~(PTRS_PER_PTE * sizeof(pte_t) - 1); + asm volatile(".insn rrf,0xb98b0000,%[r1],%[r2],%[asce],%[m4]" + : "+m" (*ptep) + : [r1] "a" (pto), [r2] "a" ((addr & PAGE_MASK) | opt), + [asce] "a" (asce), [m4] "i" (local)); +} + static __always_inline void __ptep_ipte(unsigned long address, pte_t *ptep, unsigned long opt, unsigned long asce, int local) { - unsigned long pto = (unsigned long) ptep; + unsigned long pto = __pa(ptep); if (__builtin_constant_p(opt) && opt == 0) { /* Invalidation + TLB flush for the pte */ asm volatile( - " .insn rrf,0xb2210000,%[r1],%[r2],0,%[m4]" + " ipte %[r1],%[r2],0,%[m4]" : "+m" (*ptep) : [r1] "a" (pto), [r2] "a" (address), [m4] "i" (local)); return; @@ -1003,7 +1088,7 @@ static __always_inline void __ptep_ipte(unsigned long address, pte_t *ptep, /* Invalidate ptes with options + TLB flush of the ptes */ opt = opt | (asce & _ASCE_ORIGIN); asm volatile( - " .insn rrf,0xb2210000,%[r1],%[r2],%[r3],%[m4]" + " ipte %[r1],%[r2],%[r3],%[m4]" : [r2] "+a" (address), [r3] "+a" (opt) : [r1] "a" (pto), [m4] "i" (local) : "memory"); } @@ -1011,12 +1096,12 @@ static __always_inline void __ptep_ipte(unsigned long address, pte_t *ptep, static __always_inline void __ptep_ipte_range(unsigned long address, int nr, pte_t *ptep, int local) { - unsigned long pto = (unsigned long) ptep; + unsigned long pto = __pa(ptep); /* Invalidate a range of ptes + TLB flush of the ptes */ do { asm volatile( - " .insn rrf,0xb2210000,%[r1],%[r2],%[r3],%[m4]" + " ipte %[r1],%[r2],%[r3],%[m4]" : [r2] "+a" (address), [r3] "+a" (nr) : [r1] "a" (pto), [m4] "i" (local) : "memory"); } while (nr != 255); @@ -1059,7 +1144,13 @@ static inline int ptep_clear_flush_young(struct vm_area_struct *vma, static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - return ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID)); + pte_t res; + + res = ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID)); + /* At this point the reference through the mapping is still present */ + if (mm_is_protected(mm) && pte_present(res)) + uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + return res; } #define __HAVE_ARCH_PTEP_MODIFY_PROT_TRANSACTION @@ -1071,7 +1162,13 @@ void ptep_modify_prot_commit(struct vm_area_struct *, unsigned long, static inline pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) { - return ptep_xchg_direct(vma->vm_mm, addr, ptep, __pte(_PAGE_INVALID)); + pte_t res; + + res = ptep_xchg_direct(vma->vm_mm, addr, ptep, __pte(_PAGE_INVALID)); + /* At this point the reference through the mapping is still present */ + if (mm_is_protected(vma->vm_mm) && pte_present(res)) + uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + return res; } /* @@ -1086,12 +1183,31 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full) { + pte_t res; + if (full) { - pte_t pte = *ptep; - *ptep = __pte(_PAGE_INVALID); - return pte; + res = *ptep; + set_pte(ptep, __pte(_PAGE_INVALID)); + } else { + res = ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID)); } - return ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID)); + /* Nothing to do */ + if (!mm_is_protected(mm) || !pte_present(res)) + return res; + /* + * At this point the reference through the mapping is still present. + * The notifier should have destroyed all protected vCPUs at this + * point, so the destroy should be successful. + */ + if (full && !uv_destroy_owned_page(pte_val(res) & PAGE_MASK)) + return res; + /* + * If something went wrong and the page could not be destroyed, or + * if this is not a mm teardown, the slower export is used as + * fallback instead. + */ + uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + return res; } #define __HAVE_ARCH_PTEP_SET_WRPROTECT @@ -1104,6 +1220,44 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, ptep_xchg_lazy(mm, addr, ptep, pte_wrprotect(pte)); } +/* + * Check if PTEs only differ in _PAGE_PROTECT HW bit, but also allow SW PTE + * bits in the comparison. Those might change e.g. because of dirty and young + * tracking. + */ +static inline int pte_allow_rdp(pte_t old, pte_t new) +{ + /* + * Only allow changes from RO to RW + */ + if (!(pte_val(old) & _PAGE_PROTECT) || pte_val(new) & _PAGE_PROTECT) + return 0; + + return (pte_val(old) & _PAGE_RDP_MASK) == (pte_val(new) & _PAGE_RDP_MASK); +} + +static inline void flush_tlb_fix_spurious_fault(struct vm_area_struct *vma, + unsigned long address, + pte_t *ptep) +{ + /* + * RDP might not have propagated the PTE protection reset to all CPUs, + * so there could be spurious TLB protection faults. + * NOTE: This will also be called when a racing pagetable update on + * another thread already installed the correct PTE. Both cases cannot + * really be distinguished. + * Therefore, only do the local TLB flush when RDP can be used, and the + * PTE does not have _PAGE_PROTECT set, to avoid unnecessary overhead. + * A local RDP can be used to do the flush. + */ + if (MACHINE_HAS_RDP && !(pte_val(*ptep) & _PAGE_PROTECT)) + __ptep_rdp(address, ptep, 0, 0, 1); +} +#define flush_tlb_fix_spurious_fault flush_tlb_fix_spurious_fault + +void ptep_reset_dat_prot(struct mm_struct *mm, unsigned long addr, pte_t *ptep, + pte_t new); + #define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS static inline int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, @@ -1111,7 +1265,10 @@ static inline int ptep_set_access_flags(struct vm_area_struct *vma, { if (pte_same(*ptep, entry)) return 0; - ptep_xchg_direct(vma->vm_mm, addr, ptep, entry); + if (MACHINE_HAS_RDP && !mm_has_pgste(vma->vm_mm) && pte_allow_rdp(*ptep, entry)) + ptep_reset_dat_prot(vma->vm_mm, addr, ptep, entry); + else + ptep_xchg_direct(vma->vm_mm, addr, ptep, entry); return 1; } @@ -1153,21 +1310,41 @@ void gmap_pmdp_invalidate(struct mm_struct *mm, unsigned long vmaddr); void gmap_pmdp_idte_local(struct mm_struct *mm, unsigned long vmaddr); void gmap_pmdp_idte_global(struct mm_struct *mm, unsigned long vmaddr); +#define pgprot_writecombine pgprot_writecombine +pgprot_t pgprot_writecombine(pgprot_t prot); + +#define pgprot_writethrough pgprot_writethrough +pgprot_t pgprot_writethrough(pgprot_t prot); + /* - * Certain architectures need to do special things when PTEs - * within a page table are directly modified. Thus, the following - * hook is made available. + * Set multiple PTEs to consecutive pages with a single call. All PTEs + * are within the same folio, PMD and VMA. */ -static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t entry) +static inline void set_ptes(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t entry, unsigned int nr) { if (pte_present(entry)) - pte_val(entry) &= ~_PAGE_UNUSED; - if (mm_has_pgste(mm)) - ptep_set_pte_at(mm, addr, ptep, entry); - else - *ptep = entry; + entry = clear_pte_bit(entry, __pgprot(_PAGE_UNUSED)); + if (mm_has_pgste(mm)) { + for (;;) { + ptep_set_pte_at(mm, addr, ptep, entry); + if (--nr == 0) + break; + ptep++; + entry = __pte(pte_val(entry) + PAGE_SIZE); + addr += PAGE_SIZE; + } + } else { + for (;;) { + set_pte(ptep, entry); + if (--nr == 0) + break; + ptep++; + entry = __pte(pte_val(entry) + PAGE_SIZE); + } + } } +#define set_ptes set_ptes /* * Conversion functions: convert a page and protection to a page entry, @@ -1176,9 +1353,10 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot) { pte_t __pte; - pte_val(__pte) = physpage + pgprot_val(pgprot); + + __pte = __pte(physpage | pgprot_val(pgprot)); if (!MACHINE_HAS_NX) - pte_val(__pte) &= ~_PAGE_NOEXEC; + __pte = clear_pte_bit(__pte, __pgprot(_PAGE_NOEXEC)); return pte_mkyoung(__pte); } @@ -1196,12 +1374,39 @@ static inline pte_t mk_pte(struct page *page, pgprot_t pgprot) #define p4d_index(address) (((address) >> P4D_SHIFT) & (PTRS_PER_P4D-1)) #define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1)) #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)) -#define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1)) -#define pmd_deref(pmd) (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) -#define pud_deref(pud) (pud_val(pud) & _REGION_ENTRY_ORIGIN) -#define p4d_deref(pud) (p4d_val(pud) & _REGION_ENTRY_ORIGIN) -#define pgd_deref(pgd) (pgd_val(pgd) & _REGION_ENTRY_ORIGIN) +#define p4d_deref(pud) ((unsigned long)__va(p4d_val(pud) & _REGION_ENTRY_ORIGIN)) +#define pgd_deref(pgd) ((unsigned long)__va(pgd_val(pgd) & _REGION_ENTRY_ORIGIN)) + +static inline unsigned long pmd_deref(pmd_t pmd) +{ + unsigned long origin_mask; + + origin_mask = _SEGMENT_ENTRY_ORIGIN; + if (pmd_large(pmd)) + origin_mask = _SEGMENT_ENTRY_ORIGIN_LARGE; + return (unsigned long)__va(pmd_val(pmd) & origin_mask); +} + +static inline unsigned long pmd_pfn(pmd_t pmd) +{ + return __pa(pmd_deref(pmd)) >> PAGE_SHIFT; +} + +static inline unsigned long pud_deref(pud_t pud) +{ + unsigned long origin_mask; + + origin_mask = _REGION_ENTRY_ORIGIN; + if (pud_large(pud)) + origin_mask = _REGION3_ENTRY_ORIGIN_LARGE; + return (unsigned long)__va(pud_val(pud) & origin_mask); +} + +static inline unsigned long pud_pfn(pud_t pud) +{ + return __pa(pud_deref(pud)) >> PAGE_SHIFT; +} /* * The pgd_offset function *always* adds the index for the top-level @@ -1227,38 +1432,52 @@ static inline pgd_t *pgd_offset_raw(pgd_t *pgd, unsigned long address) } #define pgd_offset(mm, address) pgd_offset_raw(READ_ONCE((mm)->pgd), address) -#define pgd_offset_k(address) pgd_offset(&init_mm, address) -static inline p4d_t *p4d_offset(pgd_t *pgd, unsigned long address) +static inline p4d_t *p4d_offset_lockless(pgd_t *pgdp, pgd_t pgd, unsigned long address) { - if ((pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R1) - return (p4d_t *) pgd_deref(*pgd) + p4d_index(address); - return (p4d_t *) pgd; + if ((pgd_val(pgd) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R1) + return (p4d_t *) pgd_deref(pgd) + p4d_index(address); + return (p4d_t *) pgdp; } +#define p4d_offset_lockless p4d_offset_lockless -static inline pud_t *pud_offset(p4d_t *p4d, unsigned long address) +static inline p4d_t *p4d_offset(pgd_t *pgdp, unsigned long address) { - if ((p4d_val(*p4d) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R2) - return (pud_t *) p4d_deref(*p4d) + pud_index(address); - return (pud_t *) p4d; + return p4d_offset_lockless(pgdp, *pgdp, address); } -static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address) +static inline pud_t *pud_offset_lockless(p4d_t *p4dp, p4d_t p4d, unsigned long address) { - if ((pud_val(*pud) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R3) - return (pmd_t *) pud_deref(*pud) + pmd_index(address); - return (pmd_t *) pud; + if ((p4d_val(p4d) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R2) + return (pud_t *) p4d_deref(p4d) + pud_index(address); + return (pud_t *) p4dp; } +#define pud_offset_lockless pud_offset_lockless -static inline pte_t *pte_offset(pmd_t *pmd, unsigned long address) +static inline pud_t *pud_offset(p4d_t *p4dp, unsigned long address) { - return (pte_t *) pmd_deref(*pmd) + pte_index(address); + return pud_offset_lockless(p4dp, *p4dp, address); } +#define pud_offset pud_offset -#define pte_offset_kernel(pmd, address) pte_offset(pmd, address) -#define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address) +static inline pmd_t *pmd_offset_lockless(pud_t *pudp, pud_t pud, unsigned long address) +{ + if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) >= _REGION_ENTRY_TYPE_R3) + return (pmd_t *) pud_deref(pud) + pmd_index(address); + return (pmd_t *) pudp; +} +#define pmd_offset_lockless pmd_offset_lockless -static inline void pte_unmap(pte_t *pte) { } +static inline pmd_t *pmd_offset(pud_t *pudp, unsigned long address) +{ + return pmd_offset_lockless(pudp, *pudp, address); +} +#define pmd_offset pmd_offset + +static inline unsigned long pmd_page_vaddr(pmd_t pmd) +{ + return (unsigned long) pmd_deref(pmd); +} static inline bool gup_fast_permitted(unsigned long start, unsigned long end) { @@ -1266,7 +1485,7 @@ static inline bool gup_fast_permitted(unsigned long start, unsigned long end) } #define gup_fast_permitted gup_fast_permitted -#define pfn_pte(pfn,pgprot) mk_pte_phys(__pa((pfn) << PAGE_SHIFT),(pgprot)) +#define pfn_pte(pfn, pgprot) mk_pte_phys(((pfn) << PAGE_SHIFT), (pgprot)) #define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT) #define pte_page(x) pfn_to_page(pte_pfn(x)) @@ -1277,61 +1496,57 @@ static inline bool gup_fast_permitted(unsigned long start, unsigned long end) static inline pmd_t pmd_wrprotect(pmd_t pmd) { - pmd_val(pmd) &= ~_SEGMENT_ENTRY_WRITE; - pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; - return pmd; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_WRITE)); + return set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); } -static inline pmd_t pmd_mkwrite(pmd_t pmd) +static inline pmd_t pmd_mkwrite_novma(pmd_t pmd) { - pmd_val(pmd) |= _SEGMENT_ENTRY_WRITE; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_WRITE)); if (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) - pmd_val(pmd) &= ~_SEGMENT_ENTRY_PROTECT; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); return pmd; } static inline pmd_t pmd_mkclean(pmd_t pmd) { - pmd_val(pmd) &= ~_SEGMENT_ENTRY_DIRTY; - pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; - return pmd; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_DIRTY)); + return set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); } static inline pmd_t pmd_mkdirty(pmd_t pmd) { - pmd_val(pmd) |= _SEGMENT_ENTRY_DIRTY | _SEGMENT_ENTRY_SOFT_DIRTY; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_DIRTY | _SEGMENT_ENTRY_SOFT_DIRTY)); if (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) - pmd_val(pmd) &= ~_SEGMENT_ENTRY_PROTECT; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); return pmd; } static inline pud_t pud_wrprotect(pud_t pud) { - pud_val(pud) &= ~_REGION3_ENTRY_WRITE; - pud_val(pud) |= _REGION_ENTRY_PROTECT; - return pud; + pud = clear_pud_bit(pud, __pgprot(_REGION3_ENTRY_WRITE)); + return set_pud_bit(pud, __pgprot(_REGION_ENTRY_PROTECT)); } static inline pud_t pud_mkwrite(pud_t pud) { - pud_val(pud) |= _REGION3_ENTRY_WRITE; + pud = set_pud_bit(pud, __pgprot(_REGION3_ENTRY_WRITE)); if (pud_val(pud) & _REGION3_ENTRY_DIRTY) - pud_val(pud) &= ~_REGION_ENTRY_PROTECT; + pud = clear_pud_bit(pud, __pgprot(_REGION_ENTRY_PROTECT)); return pud; } static inline pud_t pud_mkclean(pud_t pud) { - pud_val(pud) &= ~_REGION3_ENTRY_DIRTY; - pud_val(pud) |= _REGION_ENTRY_PROTECT; - return pud; + pud = clear_pud_bit(pud, __pgprot(_REGION3_ENTRY_DIRTY)); + return set_pud_bit(pud, __pgprot(_REGION_ENTRY_PROTECT)); } static inline pud_t pud_mkdirty(pud_t pud) { - pud_val(pud) |= _REGION3_ENTRY_DIRTY | _REGION3_ENTRY_SOFT_DIRTY; + pud = set_pud_bit(pud, __pgprot(_REGION3_ENTRY_DIRTY | _REGION3_ENTRY_SOFT_DIRTY)); if (pud_val(pud) & _REGION3_ENTRY_WRITE) - pud_val(pud) &= ~_REGION_ENTRY_PROTECT; + pud = clear_pud_bit(pud, __pgprot(_REGION_ENTRY_PROTECT)); return pud; } @@ -1355,37 +1570,39 @@ static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot) static inline pmd_t pmd_mkyoung(pmd_t pmd) { - pmd_val(pmd) |= _SEGMENT_ENTRY_YOUNG; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_YOUNG)); if (pmd_val(pmd) & _SEGMENT_ENTRY_READ) - pmd_val(pmd) &= ~_SEGMENT_ENTRY_INVALID; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_INVALID)); return pmd; } static inline pmd_t pmd_mkold(pmd_t pmd) { - pmd_val(pmd) &= ~_SEGMENT_ENTRY_YOUNG; - pmd_val(pmd) |= _SEGMENT_ENTRY_INVALID; - return pmd; + pmd = clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_YOUNG)); + return set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_INVALID)); } static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) { - pmd_val(pmd) &= _SEGMENT_ENTRY_ORIGIN_LARGE | - _SEGMENT_ENTRY_DIRTY | _SEGMENT_ENTRY_YOUNG | - _SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_SOFT_DIRTY; - pmd_val(pmd) |= massage_pgprot_pmd(newprot); + unsigned long mask; + + mask = _SEGMENT_ENTRY_ORIGIN_LARGE; + mask |= _SEGMENT_ENTRY_DIRTY; + mask |= _SEGMENT_ENTRY_YOUNG; + mask |= _SEGMENT_ENTRY_LARGE; + mask |= _SEGMENT_ENTRY_SOFT_DIRTY; + pmd = __pmd(pmd_val(pmd) & mask); + pmd = set_pmd_bit(pmd, __pgprot(massage_pgprot_pmd(newprot))); if (!(pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY)) - pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); if (!(pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG)) - pmd_val(pmd) |= _SEGMENT_ENTRY_INVALID; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_INVALID)); return pmd; } static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot) { - pmd_t __pmd; - pmd_val(__pmd) = physpage + massage_pgprot_pmd(pgprot); - return __pmd; + return __pmd(physpage + massage_pgprot_pmd(pgprot)); } #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLB_PAGE */ @@ -1409,11 +1626,11 @@ static __always_inline void __pmdp_idte(unsigned long addr, pmd_t *pmdp, { unsigned long sto; - sto = (unsigned long) pmdp - pmd_index(addr) * sizeof(pmd_t); + sto = __pa(pmdp) - pmd_index(addr) * sizeof(pmd_t); if (__builtin_constant_p(opt) && opt == 0) { /* flush without guest asce */ asm volatile( - " .insn rrf,0xb98e0000,%[r1],%[r2],0,%[m4]" + " idte %[r1],0,%[r2],%[m4]" : "+m" (*pmdp) : [r1] "a" (sto), [r2] "a" ((addr & HPAGE_MASK)), [m4] "i" (local) @@ -1421,7 +1638,7 @@ static __always_inline void __pmdp_idte(unsigned long addr, pmd_t *pmdp, } else { /* flush with guest asce */ asm volatile( - " .insn rrf,0xb98e0000,%[r1],%[r2],%[r3],%[m4]" + " idte %[r1],%[r3],%[r2],%[m4]" : "+m" (*pmdp) : [r1] "a" (sto), [r2] "a" ((addr & HPAGE_MASK) | opt), [r3] "a" (asce), [m4] "i" (local) @@ -1435,12 +1652,12 @@ static __always_inline void __pudp_idte(unsigned long addr, pud_t *pudp, { unsigned long r3o; - r3o = (unsigned long) pudp - pud_index(addr) * sizeof(pud_t); + r3o = __pa(pudp) - pud_index(addr) * sizeof(pud_t); r3o |= _ASCE_TYPE_REGION3; if (__builtin_constant_p(opt) && opt == 0) { /* flush without guest asce */ asm volatile( - " .insn rrf,0xb98e0000,%[r1],%[r2],0,%[m4]" + " idte %[r1],0,%[r2],%[m4]" : "+m" (*pudp) : [r1] "a" (r3o), [r2] "a" ((addr & PUD_MASK)), [m4] "i" (local) @@ -1448,7 +1665,7 @@ static __always_inline void __pudp_idte(unsigned long addr, pud_t *pudp, } else { /* flush with guest asce */ asm volatile( - " .insn rrf,0xb98e0000,%[r1],%[r2],%[r3],%[m4]" + " idte %[r1],%[r3],%[r2],%[m4]" : "+m" (*pudp) : [r1] "a" (r3o), [r2] "a" ((addr & PUD_MASK) | opt), [r3] "a" (asce), [m4] "i" (local) @@ -1507,16 +1724,15 @@ static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, pmd_t *pmdp, pmd_t entry) { if (!MACHINE_HAS_NX) - pmd_val(entry) &= ~_SEGMENT_ENTRY_NOEXEC; - *pmdp = entry; + entry = clear_pmd_bit(entry, __pgprot(_SEGMENT_ENTRY_NOEXEC)); + set_pmd(pmdp, entry); } static inline pmd_t pmd_mkhuge(pmd_t pmd) { - pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE; - pmd_val(pmd) |= _SEGMENT_ENTRY_YOUNG; - pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; - return pmd; + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_LARGE)); + pmd = set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_YOUNG)); + return set_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_PROTECT)); } #define __HAVE_ARCH_PMDP_HUGE_GET_AND_CLEAR @@ -1527,16 +1743,16 @@ static inline pmd_t pmdp_huge_get_and_clear(struct mm_struct *mm, } #define __HAVE_ARCH_PMDP_HUGE_GET_AND_CLEAR_FULL -static inline pmd_t pmdp_huge_get_and_clear_full(struct mm_struct *mm, +static inline pmd_t pmdp_huge_get_and_clear_full(struct vm_area_struct *vma, unsigned long addr, pmd_t *pmdp, int full) { if (full) { pmd_t pmd = *pmdp; - *pmdp = __pmd(_SEGMENT_ENTRY_EMPTY); + set_pmd(pmdp, __pmd(_SEGMENT_ENTRY_EMPTY)); return pmd; } - return pmdp_xchg_lazy(mm, addr, pmdp, __pmd(_SEGMENT_ENTRY_EMPTY)); + return pmdp_xchg_lazy(vma->vm_mm, addr, pmdp, __pmd(_SEGMENT_ENTRY_EMPTY)); } #define __HAVE_ARCH_PMDP_HUGE_CLEAR_FLUSH @@ -1573,7 +1789,7 @@ static inline pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, } #define pmdp_collapse_flush pmdp_collapse_flush -#define pfn_pmd(pfn, pgprot) mk_pmd_phys(__pa((pfn) << PAGE_SHIFT), (pgprot)) +#define pfn_pmd(pfn, pgprot) mk_pmd_phys(((pfn) << PAGE_SHIFT), (pgprot)) #define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) static inline int pmd_trans_huge(pmd_t pmd) @@ -1591,18 +1807,18 @@ static inline int has_transparent_hugepage(void) /* * 64 bit swap entry format: * A page-table entry has some bits we have to treat in a special way. - * Bits 52 and bit 55 have to be zero, otherwise a specification - * exception will occur instead of a page translation exception. The - * specification exception has the bad habit not to store necessary - * information in the lowcore. - * Bits 54 and 63 are used to indicate the page type. + * Bits 54 and 63 are used to indicate the page type. Bit 53 marks the pte + * as invalid. * A swap pte is indicated by bit pattern (pte & 0x201) == 0x200 - * This leaves the bits 0-51 and bits 56-62 to store type and offset. - * We use the 5 bits from 57-61 for the type and the 52 bits from 0-51 - * for the offset. - * | offset |01100|type |00| + * | offset |E11XX|type |S0| * |0000000000111111111122222222223333333333444444444455|55555|55566|66| * |0123456789012345678901234567890123456789012345678901|23456|78901|23| + * + * Bits 0-51 store the offset. + * Bit 52 (E) is used to remember PG_anon_exclusive. + * Bits 57-61 store the type. + * Bit 62 (S) is used for softdirty tracking. + * Bits 55 and 56 (X) are unused. */ #define __SWP_OFFSET_MASK ((1UL << 52) - 1) @@ -1612,12 +1828,12 @@ static inline int has_transparent_hugepage(void) static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) { - pte_t pte; + unsigned long pteval; - pte_val(pte) = _PAGE_INVALID | _PAGE_PROTECT; - pte_val(pte) |= (offset & __SWP_OFFSET_MASK) << __SWP_OFFSET_SHIFT; - pte_val(pte) |= (type & __SWP_TYPE_MASK) << __SWP_TYPE_SHIFT; - return pte; + pteval = _PAGE_INVALID | _PAGE_PROTECT; + pteval |= (offset & __SWP_OFFSET_MASK) << __SWP_OFFSET_SHIFT; + pteval |= (type & __SWP_TYPE_MASK) << __SWP_TYPE_SHIFT; + return __pte(pteval); } static inline unsigned long __swp_type(swp_entry_t entry) @@ -1638,10 +1854,12 @@ static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -#define kern_addr_valid(addr) (1) - extern int vmem_add_mapping(unsigned long start, unsigned long size); -extern int vmem_remove_mapping(unsigned long start, unsigned long size); +extern void vmem_remove_mapping(unsigned long start, unsigned long size); +extern int __vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot, bool alloc); +extern int vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot); +extern void vmem_unmap_4k_page(unsigned long addr); +extern pte_t *vmem_get_alloc_pte(unsigned long addr, bool alloc); extern int s390_enable_sie(void); extern int s390_enable_skey(void); extern void s390_reset_cmma(struct mm_struct *mm); @@ -1650,6 +1868,7 @@ extern void s390_reset_cmma(struct mm_struct *mm); #define HAVE_ARCH_UNMAPPED_AREA #define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN -#include <asm-generic/pgtable.h> +#define pmd_pgtable(pmd) \ + ((pgtable_t)__va(pmd_val(pmd) & -sizeof(pte_t)*PTRS_PER_PTE)) #endif /* _S390_PAGE_H */ |