From: GuanXuetao Date: Sat, 15 Jan 2011 10:17:56 +0000 (+0800) Subject: unicore32 core architecture: mm related: fault handling X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=56372b0b2f533c9a25bd40a0577405f6ddb7cff2;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git unicore32 core architecture: mm related: fault handling This patch implements fault handling of memory management. Signed-off-by: Guan Xuetao Reviewed-by: Arnd Bergmann --- diff --git a/arch/unicore32/include/asm/mmu.h b/arch/unicore32/include/asm/mmu.h new file mode 100644 index 000000000000..66fa341dc2c6 --- /dev/null +++ b/arch/unicore32/include/asm/mmu.h @@ -0,0 +1,17 @@ +/* + * linux/arch/unicore32/include/asm/mmu.h + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __UNICORE_MMU_H__ +#define __UNICORE_MMU_H__ + +typedef unsigned long mm_context_t; + +#endif diff --git a/arch/unicore32/include/asm/mmu_context.h b/arch/unicore32/include/asm/mmu_context.h new file mode 100644 index 000000000000..fb5e4c658f7a --- /dev/null +++ b/arch/unicore32/include/asm/mmu_context.h @@ -0,0 +1,87 @@ +/* + * linux/arch/unicore32/include/asm/mmu_context.h + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __UNICORE_MMU_CONTEXT_H__ +#define __UNICORE_MMU_CONTEXT_H__ + +#include +#include +#include + +#include +#include + +#define init_new_context(tsk, mm) 0 + +#define destroy_context(mm) do { } while (0) + +/* + * This is called when "tsk" is about to enter lazy TLB mode. + * + * mm: describes the currently active mm context + * tsk: task which is entering lazy tlb + * cpu: cpu number which is entering lazy tlb + * + * tsk->mm will be NULL + */ +static inline void +enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) +{ +} + +/* + * This is the actual mm switch as far as the scheduler + * is concerned. No registers are touched. We avoid + * calling the CPU specific function when the mm hasn't + * actually changed. + */ +static inline void +switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + unsigned int cpu = smp_processor_id(); + + if (!cpumask_test_and_set_cpu(cpu, mm_cpumask(next)) || prev != next) + cpu_switch_mm(next->pgd, next); +} + +#define deactivate_mm(tsk, mm) do { } while (0) +#define activate_mm(prev, next) switch_mm(prev, next, NULL) + +/* + * We are inserting a "fake" vma for the user-accessible vector page so + * gdb and friends can get to it through ptrace and /proc//mem. + * But we also want to remove it before the generic code gets to see it + * during process exit or the unmapping of it would cause total havoc. + * (the macro is used as remove_vma() is static to mm/mmap.c) + */ +#define arch_exit_mmap(mm) \ +do { \ + struct vm_area_struct *high_vma = find_vma(mm, 0xffff0000); \ + if (high_vma) { \ + BUG_ON(high_vma->vm_next); /* it should be last */ \ + if (high_vma->vm_prev) \ + high_vma->vm_prev->vm_next = NULL; \ + else \ + mm->mmap = NULL; \ + rb_erase(&high_vma->vm_rb, &mm->mm_rb); \ + mm->mmap_cache = NULL; \ + mm->map_count--; \ + remove_vma(high_vma); \ + } \ +} while (0) + +static inline void arch_dup_mmap(struct mm_struct *oldmm, + struct mm_struct *mm) +{ +} + +#endif diff --git a/arch/unicore32/include/asm/pgalloc.h b/arch/unicore32/include/asm/pgalloc.h new file mode 100644 index 000000000000..0213e373a895 --- /dev/null +++ b/arch/unicore32/include/asm/pgalloc.h @@ -0,0 +1,110 @@ +/* + * linux/arch/unicore32/include/asm/pgalloc.h + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __UNICORE_PGALLOC_H__ +#define __UNICORE_PGALLOC_H__ + +#include +#include +#include +#include + +#define check_pgt_cache() do { } while (0) + +#define _PAGE_USER_TABLE (PMD_TYPE_TABLE | PMD_PRESENT) +#define _PAGE_KERNEL_TABLE (PMD_TYPE_TABLE | PMD_PRESENT) + +extern pgd_t *get_pgd_slow(struct mm_struct *mm); +extern void free_pgd_slow(struct mm_struct *mm, pgd_t *pgd); + +#define pgd_alloc(mm) get_pgd_slow(mm) +#define pgd_free(mm, pgd) free_pgd_slow(mm, pgd) + +#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO) + +/* + * Allocate one PTE table. + */ +static inline pte_t * +pte_alloc_one_kernel(struct mm_struct *mm, unsigned long addr) +{ + pte_t *pte; + + pte = (pte_t *)__get_free_page(PGALLOC_GFP); + if (pte) + clean_dcache_area(pte, PTRS_PER_PTE * sizeof(pte_t)); + + return pte; +} + +static inline pgtable_t +pte_alloc_one(struct mm_struct *mm, unsigned long addr) +{ + struct page *pte; + + pte = alloc_pages(PGALLOC_GFP, 0); + if (pte) { + if (!PageHighMem(pte)) { + void *page = page_address(pte); + clean_dcache_area(page, PTRS_PER_PTE * sizeof(pte_t)); + } + pgtable_page_ctor(pte); + } + + return pte; +} + +/* + * Free one PTE table. + */ +static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) +{ + if (pte) + free_page((unsigned long)pte); +} + +static inline void pte_free(struct mm_struct *mm, pgtable_t pte) +{ + pgtable_page_dtor(pte); + __free_page(pte); +} + +static inline void __pmd_populate(pmd_t *pmdp, unsigned long pmdval) +{ + set_pmd(pmdp, __pmd(pmdval)); + flush_pmd_entry(pmdp); +} + +/* + * Populate the pmdp entry with a pointer to the pte. This pmd is part + * of the mm address space. + */ +static inline void +pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmdp, pte_t *ptep) +{ + unsigned long pte_ptr = (unsigned long)ptep; + + /* + * The pmd must be loaded with the physical + * address of the PTE table + */ + __pmd_populate(pmdp, __pa(pte_ptr) | _PAGE_KERNEL_TABLE); +} + +static inline void +pmd_populate(struct mm_struct *mm, pmd_t *pmdp, pgtable_t ptep) +{ + __pmd_populate(pmdp, + page_to_pfn(ptep) << PAGE_SHIFT | _PAGE_USER_TABLE); +} +#define pmd_pgtable(pmd) pmd_page(pmd) + +#endif diff --git a/arch/unicore32/include/asm/pgtable-hwdef.h b/arch/unicore32/include/asm/pgtable-hwdef.h new file mode 100644 index 000000000000..7314e859cca0 --- /dev/null +++ b/arch/unicore32/include/asm/pgtable-hwdef.h @@ -0,0 +1,55 @@ +/* + * linux/arch/unicore32/include/asm/pgtable-hwdef.h + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __UNICORE_PGTABLE_HWDEF_H__ +#define __UNICORE_PGTABLE_HWDEF_H__ + +/* + * Hardware page table definitions. + * + * + Level 1 descriptor (PMD) + * - common + */ +#define PMD_TYPE_MASK (3 << 0) +#define PMD_TYPE_TABLE (0 << 0) +/*#define PMD_TYPE_LARGE (1 << 0) */ +#define PMD_TYPE_INVALID (2 << 0) +#define PMD_TYPE_SECT (3 << 0) + +#define PMD_PRESENT (1 << 2) +#define PMD_YOUNG (1 << 3) + +/*#define PMD_SECT_DIRTY (1 << 4) */ +#define PMD_SECT_CACHEABLE (1 << 5) +#define PMD_SECT_EXEC (1 << 6) +#define PMD_SECT_WRITE (1 << 7) +#define PMD_SECT_READ (1 << 8) + +/* + * + Level 2 descriptor (PTE) + * - common + */ +#define PTE_TYPE_MASK (3 << 0) +#define PTE_TYPE_SMALL (0 << 0) +#define PTE_TYPE_MIDDLE (1 << 0) +#define PTE_TYPE_LARGE (2 << 0) +#define PTE_TYPE_INVALID (3 << 0) + +#define PTE_PRESENT (1 << 2) +#define PTE_FILE (1 << 3) /* only when !PRESENT */ +#define PTE_YOUNG (1 << 3) +#define PTE_DIRTY (1 << 4) +#define PTE_CACHEABLE (1 << 5) +#define PTE_EXEC (1 << 6) +#define PTE_WRITE (1 << 7) +#define PTE_READ (1 << 8) + +#endif diff --git a/arch/unicore32/include/asm/pgtable.h b/arch/unicore32/include/asm/pgtable.h new file mode 100644 index 000000000000..68b2f297ac97 --- /dev/null +++ b/arch/unicore32/include/asm/pgtable.h @@ -0,0 +1,317 @@ +/* + * linux/arch/unicore32/include/asm/pgtable.h + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __UNICORE_PGTABLE_H__ +#define __UNICORE_PGTABLE_H__ + +#include +#include + +#include +#include + +/* + * Just any arbitrary offset to the start of the vmalloc VM area: the + * current 8MB value just means that there will be a 8MB "hole" after the + * physical memory until the kernel virtual memory starts. That means that + * any out-of-bounds memory accesses will hopefully be caught. + * The vmalloc() routines leaves a hole of 4kB between each vmalloced + * area for the same reason. ;) + * + * Note that platforms may override VMALLOC_START, but they must provide + * VMALLOC_END. VMALLOC_END defines the (exclusive) limit of this space, + * which may not overlap IO space. + */ +#ifndef VMALLOC_START +#define VMALLOC_OFFSET SZ_8M +#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) \ + & ~(VMALLOC_OFFSET-1)) +#define VMALLOC_END (0xff000000UL) +#endif + +#define PTRS_PER_PTE 1024 +#define PTRS_PER_PGD 1024 + +/* + * PGDIR_SHIFT determines what a third-level page table entry can map + */ +#define PGDIR_SHIFT 22 + +#ifndef __ASSEMBLY__ +extern void __pte_error(const char *file, int line, unsigned long val); +extern void __pgd_error(const char *file, int line, unsigned long val); + +#define pte_ERROR(pte) __pte_error(__FILE__, __LINE__, pte_val(pte)) +#define pgd_ERROR(pgd) __pgd_error(__FILE__, __LINE__, pgd_val(pgd)) +#endif /* !__ASSEMBLY__ */ + +#define PGDIR_SIZE (1UL << PGDIR_SHIFT) +#define PGDIR_MASK (~(PGDIR_SIZE-1)) + +/* + * This is the lowest virtual address we can permit any user space + * mapping to be mapped at. This is particularly important for + * non-high vector CPUs. + */ +#define FIRST_USER_ADDRESS PAGE_SIZE + +#define FIRST_USER_PGD_NR 1 +#define USER_PTRS_PER_PGD ((TASK_SIZE/PGDIR_SIZE) - FIRST_USER_PGD_NR) + +/* + * section address mask and size definitions. + */ +#define SECTION_SHIFT 22 +#define SECTION_SIZE (1UL << SECTION_SHIFT) +#define SECTION_MASK (~(SECTION_SIZE-1)) + +#ifndef __ASSEMBLY__ + +/* + * The pgprot_* and protection_map entries will be fixed up in runtime + * to include the cachable bits based on memory policy, as well as any + * architecture dependent bits. + */ +#define _PTE_DEFAULT (PTE_PRESENT | PTE_YOUNG | PTE_CACHEABLE) + +extern pgprot_t pgprot_user; +extern pgprot_t pgprot_kernel; + +#define PAGE_NONE pgprot_user +#define PAGE_SHARED __pgprot(pgprot_val(pgprot_user | PTE_READ \ + | PTE_WRITE) +#define PAGE_SHARED_EXEC __pgprot(pgprot_val(pgprot_user | PTE_READ \ + | PTE_WRITE \ + | PTE_EXEC) +#define PAGE_COPY __pgprot(pgprot_val(pgprot_user | PTE_READ) +#define PAGE_COPY_EXEC __pgprot(pgprot_val(pgprot_user | PTE_READ \ + | PTE_EXEC) +#define PAGE_READONLY __pgprot(pgprot_val(pgprot_user | PTE_READ) +#define PAGE_READONLY_EXEC __pgprot(pgprot_val(pgprot_user | PTE_READ \ + | PTE_EXEC) +#define PAGE_KERNEL pgprot_kernel +#define PAGE_KERNEL_EXEC __pgprot(pgprot_val(pgprot_kernel | PTE_EXEC)) + +#define __PAGE_NONE __pgprot(_PTE_DEFAULT) +#define __PAGE_SHARED __pgprot(_PTE_DEFAULT | PTE_READ \ + | PTE_WRITE) +#define __PAGE_SHARED_EXEC __pgprot(_PTE_DEFAULT | PTE_READ \ + | PTE_WRITE \ + | PTE_EXEC) +#define __PAGE_COPY __pgprot(_PTE_DEFAULT | PTE_READ) +#define __PAGE_COPY_EXEC __pgprot(_PTE_DEFAULT | PTE_READ \ + | PTE_EXEC) +#define __PAGE_READONLY __pgprot(_PTE_DEFAULT | PTE_READ) +#define __PAGE_READONLY_EXEC __pgprot(_PTE_DEFAULT | PTE_READ \ + | PTE_EXEC) + +#endif /* __ASSEMBLY__ */ + +/* + * The table below defines the page protection levels that we insert into our + * Linux page table version. These get translated into the best that the + * architecture can perform. Note that on UniCore hardware: + * 1) We cannot do execute protection + * 2) If we could do execute protection, then read is implied + * 3) write implies read permissions + */ +#define __P000 __PAGE_NONE +#define __P001 __PAGE_READONLY +#define __P010 __PAGE_COPY +#define __P011 __PAGE_COPY +#define __P100 __PAGE_READONLY_EXEC +#define __P101 __PAGE_READONLY_EXEC +#define __P110 __PAGE_COPY_EXEC +#define __P111 __PAGE_COPY_EXEC + +#define __S000 __PAGE_NONE +#define __S001 __PAGE_READONLY +#define __S010 __PAGE_SHARED +#define __S011 __PAGE_SHARED +#define __S100 __PAGE_READONLY_EXEC +#define __S101 __PAGE_READONLY_EXEC +#define __S110 __PAGE_SHARED_EXEC +#define __S111 __PAGE_SHARED_EXEC + +#ifndef __ASSEMBLY__ +/* + * ZERO_PAGE is a global shared page that is always zero: used + * for zero-mapped memory areas etc.. + */ +extern struct page *empty_zero_page; +#define ZERO_PAGE(vaddr) (empty_zero_page) + +#define pte_pfn(pte) (pte_val(pte) >> PAGE_SHIFT) +#define pfn_pte(pfn, prot) (__pte(((pfn) << PAGE_SHIFT) \ + | pgprot_val(prot))) + +#define pte_none(pte) (!pte_val(pte)) +#define pte_clear(mm, addr, ptep) set_pte(ptep, __pte(0)) +#define pte_page(pte) (pfn_to_page(pte_pfn(pte))) +#define pte_offset_kernel(dir, addr) (pmd_page_vaddr(*(dir)) \ + + __pte_index(addr)) + +#define pte_offset_map(dir, addr) (pmd_page_vaddr(*(dir)) \ + + __pte_index(addr)) +#define pte_unmap(pte) do { } while (0) + +#define set_pte(ptep, pte) cpu_set_pte(ptep, pte) + +#define set_pte_at(mm, addr, ptep, pteval) \ + do { \ + set_pte(ptep, pteval); \ + } while (0) + +/* + * The following only work if pte_present() is true. + * Undefined behaviour if not.. + */ +#define pte_present(pte) (pte_val(pte) & PTE_PRESENT) +#define pte_write(pte) (pte_val(pte) & PTE_WRITE) +#define pte_dirty(pte) (pte_val(pte) & PTE_DIRTY) +#define pte_young(pte) (pte_val(pte) & PTE_YOUNG) +#define pte_exec(pte) (pte_val(pte) & PTE_EXEC) +#define pte_special(pte) (0) + +#define PTE_BIT_FUNC(fn, op) \ +static inline pte_t pte_##fn(pte_t pte) { pte_val(pte) op; return pte; } + +PTE_BIT_FUNC(wrprotect, &= ~PTE_WRITE); +PTE_BIT_FUNC(mkwrite, |= PTE_WRITE); +PTE_BIT_FUNC(mkclean, &= ~PTE_DIRTY); +PTE_BIT_FUNC(mkdirty, |= PTE_DIRTY); +PTE_BIT_FUNC(mkold, &= ~PTE_YOUNG); +PTE_BIT_FUNC(mkyoung, |= PTE_YOUNG); + +static inline pte_t pte_mkspecial(pte_t pte) { return pte; } + +/* + * Mark the prot value as uncacheable. + */ +#define pgprot_noncached(prot) \ + __pgprot(pgprot_val(prot) & ~PTE_CACHEABLE) +#define pgprot_writecombine(prot) \ + __pgprot(pgprot_val(prot) & ~PTE_CACHEABLE) +#define pgprot_dmacoherent(prot) \ + __pgprot(pgprot_val(prot) & ~PTE_CACHEABLE) + +#define pmd_none(pmd) (!pmd_val(pmd)) +#define pmd_present(pmd) (pmd_val(pmd) & PMD_PRESENT) +#define pmd_bad(pmd) (((pmd_val(pmd) & \ + (PMD_PRESENT | PMD_TYPE_MASK)) \ + != (PMD_PRESENT | PMD_TYPE_TABLE))) + +#define set_pmd(pmdpd, pmdval) \ + do { \ + *(pmdpd) = pmdval; \ + } while (0) + +#define pmd_clear(pmdp) \ + do { \ + set_pmd(pmdp, __pmd(0));\ + clean_pmd_entry(pmdp); \ + } while (0) + +#define pmd_page_vaddr(pmd) ((pte_t *)__va(pmd_val(pmd) & PAGE_MASK)) +#define pmd_page(pmd) pfn_to_page(__phys_to_pfn(pmd_val(pmd))) + +/* + * Conversion functions: convert a page and protection to a page entry, + * and a page entry and page directory to the page they refer to. + */ +#define mk_pte(page, prot) pfn_pte(page_to_pfn(page), prot) + +/* to find an entry in a page-table-directory */ +#define pgd_index(addr) ((addr) >> PGDIR_SHIFT) + +#define pgd_offset(mm, addr) ((mm)->pgd+pgd_index(addr)) + +/* to find an entry in a kernel page-table-directory */ +#define pgd_offset_k(addr) pgd_offset(&init_mm, addr) + +/* Find an entry in the third-level page table.. */ +#define __pte_index(addr) (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)) + +static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) +{ + const unsigned long mask = PTE_EXEC | PTE_WRITE | PTE_READ; + pte_val(pte) = (pte_val(pte) & ~mask) | (pgprot_val(newprot) & mask); + return pte; +} + +extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; + +/* + * Encode and decode a swap entry. Swap entries are stored in the Linux + * page tables as follows: + * + * 3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 + * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 + * <--------------- offset --------------> <--- type --> 0 0 0 0 0 + * + * This gives us up to 127 swap files and 32GB per swap file. Note that + * the offset field is always non-zero. + */ +#define __SWP_TYPE_SHIFT 5 +#define __SWP_TYPE_BITS 7 +#define __SWP_TYPE_MASK ((1 << __SWP_TYPE_BITS) - 1) +#define __SWP_OFFSET_SHIFT (__SWP_TYPE_BITS + __SWP_TYPE_SHIFT) + +#define __swp_type(x) (((x).val >> __SWP_TYPE_SHIFT) \ + & __SWP_TYPE_MASK) +#define __swp_offset(x) ((x).val >> __SWP_OFFSET_SHIFT) +#define __swp_entry(type, offset) ((swp_entry_t) { \ + ((type) << __SWP_TYPE_SHIFT) | \ + ((offset) << __SWP_OFFSET_SHIFT) }) + +#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) +#define __swp_entry_to_pte(swp) ((pte_t) { (swp).val }) + +/* + * It is an error for the kernel to have more swap files than we can + * encode in the PTEs. This ensures that we know when MAX_SWAPFILES + * is increased beyond what we presently support. + */ +#define MAX_SWAPFILES_CHECK() \ + BUILD_BUG_ON(MAX_SWAPFILES_SHIFT > __SWP_TYPE_BITS) + +/* + * Encode and decode a file entry. File entries are stored in the Linux + * page tables as follows: + * + * 3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 + * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 + * <----------------------- offset ----------------------> 1 0 0 0 + */ +#define pte_file(pte) (pte_val(pte) & PTE_FILE) +#define pte_to_pgoff(x) (pte_val(x) >> 4) +#define pgoff_to_pte(x) __pte(((x) << 4) | PTE_FILE) + +#define PTE_FILE_MAX_BITS 28 + +/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */ +/* FIXME: this is not correct */ +#define kern_addr_valid(addr) (1) + +#include + +/* + * remap a physical page `pfn' of size `size' with page protection `prot' + * into virtual address `from' + */ +#define io_remap_pfn_range(vma, from, pfn, size, prot) \ + remap_pfn_range(vma, from, pfn, size, prot) + +#define pgtable_cache_init() do { } while (0) + +#endif /* !__ASSEMBLY__ */ + +#endif /* __UNICORE_PGTABLE_H__ */ diff --git a/arch/unicore32/mm/alignment.c b/arch/unicore32/mm/alignment.c new file mode 100644 index 000000000000..28f576d733ee --- /dev/null +++ b/arch/unicore32/mm/alignment.c @@ -0,0 +1,523 @@ +/* + * linux/arch/unicore32/mm/alignment.c + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +/* + * TODO: + * FPU ldm/stm not handling + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#define CODING_BITS(i) (i & 0xe0000120) + +#define LDST_P_BIT(i) (i & (1 << 28)) /* Preindex */ +#define LDST_U_BIT(i) (i & (1 << 27)) /* Add offset */ +#define LDST_W_BIT(i) (i & (1 << 25)) /* Writeback */ +#define LDST_L_BIT(i) (i & (1 << 24)) /* Load */ + +#define LDST_P_EQ_U(i) ((((i) ^ ((i) >> 1)) & (1 << 27)) == 0) + +#define LDSTH_I_BIT(i) (i & (1 << 26)) /* half-word immed */ +#define LDM_S_BIT(i) (i & (1 << 26)) /* write ASR from BSR */ +#define LDM_H_BIT(i) (i & (1 << 6)) /* select r0-r15 or r16-r31 */ + +#define RN_BITS(i) ((i >> 19) & 31) /* Rn */ +#define RD_BITS(i) ((i >> 14) & 31) /* Rd */ +#define RM_BITS(i) (i & 31) /* Rm */ + +#define REGMASK_BITS(i) (((i & 0x7fe00) >> 3) | (i & 0x3f)) +#define OFFSET_BITS(i) (i & 0x03fff) + +#define SHIFT_BITS(i) ((i >> 9) & 0x1f) +#define SHIFT_TYPE(i) (i & 0xc0) +#define SHIFT_LSL 0x00 +#define SHIFT_LSR 0x40 +#define SHIFT_ASR 0x80 +#define SHIFT_RORRRX 0xc0 + +union offset_union { + unsigned long un; + signed long sn; +}; + +#define TYPE_ERROR 0 +#define TYPE_FAULT 1 +#define TYPE_LDST 2 +#define TYPE_DONE 3 +#define TYPE_SWAP 4 +#define TYPE_COLS 5 /* Coprocessor load/store */ + +#define get8_unaligned_check(val, addr, err) \ + __asm__( \ + "1: ldb.u %1, [%2], #1\n" \ + "2:\n" \ + " .pushsection .fixup,\"ax\"\n" \ + " .align 2\n" \ + "3: mov %0, #1\n" \ + " b 2b\n" \ + " .popsection\n" \ + " .pushsection __ex_table,\"a\"\n" \ + " .align 3\n" \ + " .long 1b, 3b\n" \ + " .popsection\n" \ + : "=r" (err), "=&r" (val), "=r" (addr) \ + : "0" (err), "2" (addr)) + +#define get8t_unaligned_check(val, addr, err) \ + __asm__( \ + "1: ldb.u %1, [%2], #1\n" \ + "2:\n" \ + " .pushsection .fixup,\"ax\"\n" \ + " .align 2\n" \ + "3: mov %0, #1\n" \ + " b 2b\n" \ + " .popsection\n" \ + " .pushsection __ex_table,\"a\"\n" \ + " .align 3\n" \ + " .long 1b, 3b\n" \ + " .popsection\n" \ + : "=r" (err), "=&r" (val), "=r" (addr) \ + : "0" (err), "2" (addr)) + +#define get16_unaligned_check(val, addr) \ + do { \ + unsigned int err = 0, v, a = addr; \ + get8_unaligned_check(val, a, err); \ + get8_unaligned_check(v, a, err); \ + val |= v << 8; \ + if (err) \ + goto fault; \ + } while (0) + +#define put16_unaligned_check(val, addr) \ + do { \ + unsigned int err = 0, v = val, a = addr; \ + __asm__( \ + "1: stb.u %1, [%2], #1\n" \ + " mov %1, %1 >> #8\n" \ + "2: stb.u %1, [%2]\n" \ + "3:\n" \ + " .pushsection .fixup,\"ax\"\n" \ + " .align 2\n" \ + "4: mov %0, #1\n" \ + " b 3b\n" \ + " .popsection\n" \ + " .pushsection __ex_table,\"a\"\n" \ + " .align 3\n" \ + " .long 1b, 4b\n" \ + " .long 2b, 4b\n" \ + " .popsection\n" \ + : "=r" (err), "=&r" (v), "=&r" (a) \ + : "0" (err), "1" (v), "2" (a)); \ + if (err) \ + goto fault; \ + } while (0) + +#define __put32_unaligned_check(ins, val, addr) \ + do { \ + unsigned int err = 0, v = val, a = addr; \ + __asm__( \ + "1: "ins" %1, [%2], #1\n" \ + " mov %1, %1 >> #8\n" \ + "2: "ins" %1, [%2], #1\n" \ + " mov %1, %1 >> #8\n" \ + "3: "ins" %1, [%2], #1\n" \ + " mov %1, %1 >> #8\n" \ + "4: "ins" %1, [%2]\n" \ + "5:\n" \ + " .pushsection .fixup,\"ax\"\n" \ + " .align 2\n" \ + "6: mov %0, #1\n" \ + " b 5b\n" \ + " .popsection\n" \ + " .pushsection __ex_table,\"a\"\n" \ + " .align 3\n" \ + " .long 1b, 6b\n" \ + " .long 2b, 6b\n" \ + " .long 3b, 6b\n" \ + " .long 4b, 6b\n" \ + " .popsection\n" \ + : "=r" (err), "=&r" (v), "=&r" (a) \ + : "0" (err), "1" (v), "2" (a)); \ + if (err) \ + goto fault; \ + } while (0) + +#define get32_unaligned_check(val, addr) \ + do { \ + unsigned int err = 0, v, a = addr; \ + get8_unaligned_check(val, a, err); \ + get8_unaligned_check(v, a, err); \ + val |= v << 8; \ + get8_unaligned_check(v, a, err); \ + val |= v << 16; \ + get8_unaligned_check(v, a, err); \ + val |= v << 24; \ + if (err) \ + goto fault; \ + } while (0) + +#define put32_unaligned_check(val, addr) \ + __put32_unaligned_check("stb.u", val, addr) + +#define get32t_unaligned_check(val, addr) \ + do { \ + unsigned int err = 0, v, a = addr; \ + get8t_unaligned_check(val, a, err); \ + get8t_unaligned_check(v, a, err); \ + val |= v << 8; \ + get8t_unaligned_check(v, a, err); \ + val |= v << 16; \ + get8t_unaligned_check(v, a, err); \ + val |= v << 24; \ + if (err) \ + goto fault; \ + } while (0) + +#define put32t_unaligned_check(val, addr) \ + __put32_unaligned_check("stb.u", val, addr) + +static void +do_alignment_finish_ldst(unsigned long addr, unsigned long instr, + struct pt_regs *regs, union offset_union offset) +{ + if (!LDST_U_BIT(instr)) + offset.un = -offset.un; + + if (!LDST_P_BIT(instr)) + addr += offset.un; + + if (!LDST_P_BIT(instr) || LDST_W_BIT(instr)) + regs->uregs[RN_BITS(instr)] = addr; +} + +static int +do_alignment_ldrhstrh(unsigned long addr, unsigned long instr, + struct pt_regs *regs) +{ + unsigned int rd = RD_BITS(instr); + + /* old value 0x40002120, can't judge swap instr correctly */ + if ((instr & 0x4b003fe0) == 0x40000120) + goto swp; + + if (LDST_L_BIT(instr)) { + unsigned long val; + get16_unaligned_check(val, addr); + + /* signed half-word? */ + if (instr & 0x80) + val = (signed long)((signed short)val); + + regs->uregs[rd] = val; + } else + put16_unaligned_check(regs->uregs[rd], addr); + + return TYPE_LDST; + +swp: + /* only handle swap word + * for swap byte should not active this alignment exception */ + get32_unaligned_check(regs->uregs[RD_BITS(instr)], addr); + put32_unaligned_check(regs->uregs[RM_BITS(instr)], addr); + return TYPE_SWAP; + +fault: + return TYPE_FAULT; +} + +static int +do_alignment_ldrstr(unsigned long addr, unsigned long instr, + struct pt_regs *regs) +{ + unsigned int rd = RD_BITS(instr); + + if (!LDST_P_BIT(instr) && LDST_W_BIT(instr)) + goto trans; + + if (LDST_L_BIT(instr)) + get32_unaligned_check(regs->uregs[rd], addr); + else + put32_unaligned_check(regs->uregs[rd], addr); + return TYPE_LDST; + +trans: + if (LDST_L_BIT(instr)) + get32t_unaligned_check(regs->uregs[rd], addr); + else + put32t_unaligned_check(regs->uregs[rd], addr); + return TYPE_LDST; + +fault: + return TYPE_FAULT; +} + +/* + * LDM/STM alignment handler. + * + * There are 4 variants of this instruction: + * + * B = rn pointer before instruction, A = rn pointer after instruction + * ------ increasing address -----> + * | | r0 | r1 | ... | rx | | + * PU = 01 B A + * PU = 11 B A + * PU = 00 A B + * PU = 10 A B + */ +static int +do_alignment_ldmstm(unsigned long addr, unsigned long instr, + struct pt_regs *regs) +{ + unsigned int rd, rn, pc_correction, reg_correction, nr_regs, regbits; + unsigned long eaddr, newaddr; + + if (LDM_S_BIT(instr)) + goto bad; + + pc_correction = 4; /* processor implementation defined */ + + /* count the number of registers in the mask to be transferred */ + nr_regs = hweight16(REGMASK_BITS(instr)) * 4; + + rn = RN_BITS(instr); + newaddr = eaddr = regs->uregs[rn]; + + if (!LDST_U_BIT(instr)) + nr_regs = -nr_regs; + newaddr += nr_regs; + if (!LDST_U_BIT(instr)) + eaddr = newaddr; + + if (LDST_P_EQ_U(instr)) /* U = P */ + eaddr += 4; + + /* + * This is a "hint" - we already have eaddr worked out by the + * processor for us. + */ + if (addr != eaddr) { + printk(KERN_ERR "LDMSTM: PC = %08lx, instr = %08lx, " + "addr = %08lx, eaddr = %08lx\n", + instruction_pointer(regs), instr, addr, eaddr); + show_regs(regs); + } + + if (LDM_H_BIT(instr)) + reg_correction = 0x10; + else + reg_correction = 0x00; + + for (regbits = REGMASK_BITS(instr), rd = 0; regbits; + regbits >>= 1, rd += 1) + if (regbits & 1) { + if (LDST_L_BIT(instr)) + get32_unaligned_check(regs-> + uregs[rd + reg_correction], eaddr); + else + put32_unaligned_check(regs-> + uregs[rd + reg_correction], eaddr); + eaddr += 4; + } + + if (LDST_W_BIT(instr)) + regs->uregs[rn] = newaddr; + return TYPE_DONE; + +fault: + regs->UCreg_pc -= pc_correction; + return TYPE_FAULT; + +bad: + printk(KERN_ERR "Alignment trap: not handling ldm with s-bit set\n"); + return TYPE_ERROR; +} + +static int +do_alignment(unsigned long addr, unsigned int error_code, struct pt_regs *regs) +{ + union offset_union offset; + unsigned long instr, instrptr; + int (*handler) (unsigned long addr, unsigned long instr, + struct pt_regs *regs); + unsigned int type; + + instrptr = instruction_pointer(regs); + if (instrptr >= PAGE_OFFSET) + instr = *(unsigned long *)instrptr; + else { + __asm__ __volatile__( + "ldw.u %0, [%1]\n" + : "=&r"(instr) + : "r"(instrptr)); + } + + regs->UCreg_pc += 4; + + switch (CODING_BITS(instr)) { + case 0x40000120: /* ldrh or strh */ + if (LDSTH_I_BIT(instr)) + offset.un = (instr & 0x3e00) >> 4 | (instr & 31); + else + offset.un = regs->uregs[RM_BITS(instr)]; + handler = do_alignment_ldrhstrh; + break; + + case 0x60000000: /* ldr or str immediate */ + case 0x60000100: /* ldr or str immediate */ + case 0x60000020: /* ldr or str immediate */ + case 0x60000120: /* ldr or str immediate */ + offset.un = OFFSET_BITS(instr); + handler = do_alignment_ldrstr; + break; + + case 0x40000000: /* ldr or str register */ + offset.un = regs->uregs[RM_BITS(instr)]; + { + unsigned int shiftval = SHIFT_BITS(instr); + + switch (SHIFT_TYPE(instr)) { + case SHIFT_LSL: + offset.un <<= shiftval; + break; + + case SHIFT_LSR: + offset.un >>= shiftval; + break; + + case SHIFT_ASR: + offset.sn >>= shiftval; + break; + + case SHIFT_RORRRX: + if (shiftval == 0) { + offset.un >>= 1; + if (regs->UCreg_asr & PSR_C_BIT) + offset.un |= 1 << 31; + } else + offset.un = offset.un >> shiftval | + offset.un << (32 - shiftval); + break; + } + } + handler = do_alignment_ldrstr; + break; + + case 0x80000000: /* ldm or stm */ + case 0x80000020: /* ldm or stm */ + handler = do_alignment_ldmstm; + break; + + default: + goto bad; + } + + type = handler(addr, instr, regs); + + if (type == TYPE_ERROR || type == TYPE_FAULT) + goto bad_or_fault; + + if (type == TYPE_LDST) + do_alignment_finish_ldst(addr, instr, regs, offset); + + return 0; + +bad_or_fault: + if (type == TYPE_ERROR) + goto bad; + regs->UCreg_pc -= 4; + /* + * We got a fault - fix it up, or die. + */ + do_bad_area(addr, error_code, regs); + return 0; + +bad: + /* + * Oops, we didn't handle the instruction. + * However, we must handle fpu instr firstly. + */ +#ifdef CONFIG_UNICORE_FPU_F64 + /* handle co.load/store */ +#define CODING_COLS 0xc0000000 +#define COLS_OFFSET_BITS(i) (i & 0x1FF) +#define COLS_L_BITS(i) (i & (1<<24)) +#define COLS_FN_BITS(i) ((i>>14) & 31) + if ((instr & 0xe0000000) == CODING_COLS) { + unsigned int fn = COLS_FN_BITS(instr); + unsigned long val = 0; + if (COLS_L_BITS(instr)) { + get32t_unaligned_check(val, addr); + switch (fn) { +#define ASM_MTF(n) case n: \ + __asm__ __volatile__("MTF %0, F" __stringify(n) \ + : : "r"(val)); \ + break; + ASM_MTF(0); ASM_MTF(1); ASM_MTF(2); ASM_MTF(3); + ASM_MTF(4); ASM_MTF(5); ASM_MTF(6); ASM_MTF(7); + ASM_MTF(8); ASM_MTF(9); ASM_MTF(10); ASM_MTF(11); + ASM_MTF(12); ASM_MTF(13); ASM_MTF(14); ASM_MTF(15); + ASM_MTF(16); ASM_MTF(17); ASM_MTF(18); ASM_MTF(19); + ASM_MTF(20); ASM_MTF(21); ASM_MTF(22); ASM_MTF(23); + ASM_MTF(24); ASM_MTF(25); ASM_MTF(26); ASM_MTF(27); + ASM_MTF(28); ASM_MTF(29); ASM_MTF(30); ASM_MTF(31); +#undef ASM_MTF + } + } else { + switch (fn) { +#define ASM_MFF(n) case n: \ + __asm__ __volatile__("MFF %0, F" __stringify(n) \ + : : "r"(val)); \ + break; + ASM_MFF(0); ASM_MFF(1); ASM_MFF(2); ASM_MFF(3); + ASM_MFF(4); ASM_MFF(5); ASM_MFF(6); ASM_MFF(7); + ASM_MFF(8); ASM_MFF(9); ASM_MFF(10); ASM_MFF(11); + ASM_MFF(12); ASM_MFF(13); ASM_MFF(14); ASM_MFF(15); + ASM_MFF(16); ASM_MFF(17); ASM_MFF(18); ASM_MFF(19); + ASM_MFF(20); ASM_MFF(21); ASM_MFF(22); ASM_MFF(23); + ASM_MFF(24); ASM_MFF(25); ASM_MFF(26); ASM_MFF(27); + ASM_MFF(28); ASM_MFF(29); ASM_MFF(30); ASM_MFF(31); +#undef ASM_MFF + } + put32t_unaligned_check(val, addr); + } + return TYPE_COLS; + } +fault: + return TYPE_FAULT; +#endif + printk(KERN_ERR "Alignment trap: not handling instruction " + "%08lx at [<%08lx>]\n", instr, instrptr); + return 1; +} + +/* + * This needs to be done after sysctl_init, otherwise sys/ will be + * overwritten. Actually, this shouldn't be in sys/ at all since + * it isn't a sysctl, and it doesn't contain sysctl information. + */ +static int __init alignment_init(void) +{ + hook_fault_code(1, do_alignment, SIGBUS, BUS_ADRALN, + "alignment exception"); + + return 0; +} + +fs_initcall(alignment_init); diff --git a/arch/unicore32/mm/extable.c b/arch/unicore32/mm/extable.c new file mode 100644 index 000000000000..6564180eb285 --- /dev/null +++ b/arch/unicore32/mm/extable.c @@ -0,0 +1,24 @@ +/* + * linux/arch/unicore32/mm/extable.c + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include + +int fixup_exception(struct pt_regs *regs) +{ + const struct exception_table_entry *fixup; + + fixup = search_exception_tables(instruction_pointer(regs)); + if (fixup) + regs->UCreg_pc = fixup->fixup; + + return fixup != NULL; +} diff --git a/arch/unicore32/mm/fault.c b/arch/unicore32/mm/fault.c new file mode 100644 index 000000000000..283aa4b50b7a --- /dev/null +++ b/arch/unicore32/mm/fault.c @@ -0,0 +1,479 @@ +/* + * linux/arch/unicore32/mm/fault.c + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +/* + * Fault status register encodings. We steal bit 31 for our own purposes. + */ +#define FSR_LNX_PF (1 << 31) + +static inline int fsr_fs(unsigned int fsr) +{ + /* xyabcde will be abcde+xy */ + return (fsr & 31) + ((fsr & (3 << 5)) >> 5); +} + +/* + * This is useful to dump out the page tables associated with + * 'addr' in mm 'mm'. + */ +void show_pte(struct mm_struct *mm, unsigned long addr) +{ + pgd_t *pgd; + + if (!mm) + mm = &init_mm; + + printk(KERN_ALERT "pgd = %p\n", mm->pgd); + pgd = pgd_offset(mm, addr); + printk(KERN_ALERT "[%08lx] *pgd=%08lx", addr, pgd_val(*pgd)); + + do { + pmd_t *pmd; + pte_t *pte; + + if (pgd_none(*pgd)) + break; + + if (pgd_bad(*pgd)) { + printk("(bad)"); + break; + } + + pmd = pmd_offset((pud_t *) pgd, addr); + if (PTRS_PER_PMD != 1) + printk(", *pmd=%08lx", pmd_val(*pmd)); + + if (pmd_none(*pmd)) + break; + + if (pmd_bad(*pmd)) { + printk("(bad)"); + break; + } + + /* We must not map this if we have highmem enabled */ + if (PageHighMem(pfn_to_page(pmd_val(*pmd) >> PAGE_SHIFT))) + break; + + pte = pte_offset_map(pmd, addr); + printk(", *pte=%08lx", pte_val(*pte)); + pte_unmap(pte); + } while (0); + + printk("\n"); +} + +/* + * Oops. The kernel tried to access some page that wasn't present. + */ +static void __do_kernel_fault(struct mm_struct *mm, unsigned long addr, + unsigned int fsr, struct pt_regs *regs) +{ + /* + * Are we prepared to handle this kernel fault? + */ + if (fixup_exception(regs)) + return; + + /* + * No handler, we'll have to terminate things with extreme prejudice. + */ + bust_spinlocks(1); + printk(KERN_ALERT + "Unable to handle kernel %s at virtual address %08lx\n", + (addr < PAGE_SIZE) ? "NULL pointer dereference" : + "paging request", addr); + + show_pte(mm, addr); + die("Oops", regs, fsr); + bust_spinlocks(0); + do_exit(SIGKILL); +} + +/* + * Something tried to access memory that isn't in our memory map.. + * User mode accesses just cause a SIGSEGV + */ +static void __do_user_fault(struct task_struct *tsk, unsigned long addr, + unsigned int fsr, unsigned int sig, int code, + struct pt_regs *regs) +{ + struct siginfo si; + + tsk->thread.address = addr; + tsk->thread.error_code = fsr; + tsk->thread.trap_no = 14; + si.si_signo = sig; + si.si_errno = 0; + si.si_code = code; + si.si_addr = (void __user *)addr; + force_sig_info(sig, &si, tsk); +} + +void do_bad_area(unsigned long addr, unsigned int fsr, struct pt_regs *regs) +{ + struct task_struct *tsk = current; + struct mm_struct *mm = tsk->active_mm; + + /* + * If we are in kernel mode at this point, we + * have no context to handle this fault with. + */ + if (user_mode(regs)) + __do_user_fault(tsk, addr, fsr, SIGSEGV, SEGV_MAPERR, regs); + else + __do_kernel_fault(mm, addr, fsr, regs); +} + +#define VM_FAULT_BADMAP 0x010000 +#define VM_FAULT_BADACCESS 0x020000 + +/* + * Check that the permissions on the VMA allow for the fault which occurred. + * If we encountered a write fault, we must have write permission, otherwise + * we allow any permission. + */ +static inline bool access_error(unsigned int fsr, struct vm_area_struct *vma) +{ + unsigned int mask = VM_READ | VM_WRITE | VM_EXEC; + + if (!(fsr ^ 0x12)) /* write? */ + mask = VM_WRITE; + if (fsr & FSR_LNX_PF) + mask = VM_EXEC; + + return vma->vm_flags & mask ? false : true; +} + +static int __do_pf(struct mm_struct *mm, unsigned long addr, unsigned int fsr, + struct task_struct *tsk) +{ + struct vm_area_struct *vma; + int fault; + + vma = find_vma(mm, addr); + fault = VM_FAULT_BADMAP; + if (unlikely(!vma)) + goto out; + if (unlikely(vma->vm_start > addr)) + goto check_stack; + + /* + * Ok, we have a good vm_area for this + * memory access, so we can handle it. + */ +good_area: + if (access_error(fsr, vma)) { + fault = VM_FAULT_BADACCESS; + goto out; + } + + /* + * If for any reason at all we couldn't handle the fault, make + * sure we exit gracefully rather than endlessly redo the fault. + */ + fault = handle_mm_fault(mm, vma, addr & PAGE_MASK, + (!(fsr ^ 0x12)) ? FAULT_FLAG_WRITE : 0); + if (unlikely(fault & VM_FAULT_ERROR)) + return fault; + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + return fault; + +check_stack: + if (vma->vm_flags & VM_GROWSDOWN && !expand_stack(vma, addr)) + goto good_area; +out: + return fault; +} + +static int do_pf(unsigned long addr, unsigned int fsr, struct pt_regs *regs) +{ + struct task_struct *tsk; + struct mm_struct *mm; + int fault, sig, code; + + tsk = current; + mm = tsk->mm; + + /* + * If we're in an interrupt or have no user + * context, we must not take the fault.. + */ + if (in_atomic() || !mm) + goto no_context; + + /* + * As per x86, we may deadlock here. However, since the kernel only + * validly references user space from well defined areas of the code, + * we can bug out early if this is from code which shouldn't. + */ + if (!down_read_trylock(&mm->mmap_sem)) { + if (!user_mode(regs) + && !search_exception_tables(regs->UCreg_pc)) + goto no_context; + down_read(&mm->mmap_sem); + } else { + /* + * The above down_read_trylock() might have succeeded in + * which case, we'll have missed the might_sleep() from + * down_read() + */ + might_sleep(); +#ifdef CONFIG_DEBUG_VM + if (!user_mode(regs) && + !search_exception_tables(regs->UCreg_pc)) + goto no_context; +#endif + } + + fault = __do_pf(mm, addr, fsr, tsk); + up_read(&mm->mmap_sem); + + /* + * Handle the "normal" case first - VM_FAULT_MAJOR / VM_FAULT_MINOR + */ + if (likely(!(fault & + (VM_FAULT_ERROR | VM_FAULT_BADMAP | VM_FAULT_BADACCESS)))) + return 0; + + if (fault & VM_FAULT_OOM) { + /* + * We ran out of memory, call the OOM killer, and return to + * userspace (which will retry the fault, or kill us if we + * got oom-killed) + */ + pagefault_out_of_memory(); + return 0; + } + + /* + * If we are in kernel mode at this point, we + * have no context to handle this fault with. + */ + if (!user_mode(regs)) + goto no_context; + + if (fault & VM_FAULT_SIGBUS) { + /* + * We had some memory, but were unable to + * successfully fix up this page fault. + */ + sig = SIGBUS; + code = BUS_ADRERR; + } else { + /* + * Something tried to access memory that + * isn't in our memory map.. + */ + sig = SIGSEGV; + code = fault == VM_FAULT_BADACCESS ? SEGV_ACCERR : SEGV_MAPERR; + } + + __do_user_fault(tsk, addr, fsr, sig, code, regs); + return 0; + +no_context: + __do_kernel_fault(mm, addr, fsr, regs); + return 0; +} + +/* + * First Level Translation Fault Handler + * + * We enter here because the first level page table doesn't contain + * a valid entry for the address. + * + * If the address is in kernel space (>= TASK_SIZE), then we are + * probably faulting in the vmalloc() area. + * + * If the init_task's first level page tables contains the relevant + * entry, we copy the it to this task. If not, we send the process + * a signal, fixup the exception, or oops the kernel. + * + * NOTE! We MUST NOT take any locks for this case. We may be in an + * interrupt or a critical region, and should only copy the information + * from the master page table, nothing more. + */ +static int do_ifault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) +{ + unsigned int index; + pgd_t *pgd, *pgd_k; + pmd_t *pmd, *pmd_k; + + if (addr < TASK_SIZE) + return do_pf(addr, fsr, regs); + + if (user_mode(regs)) + goto bad_area; + + index = pgd_index(addr); + + pgd = cpu_get_pgd() + index; + pgd_k = init_mm.pgd + index; + + if (pgd_none(*pgd_k)) + goto bad_area; + + pmd_k = pmd_offset((pud_t *) pgd_k, addr); + pmd = pmd_offset((pud_t *) pgd, addr); + + if (pmd_none(*pmd_k)) + goto bad_area; + + set_pmd(pmd, *pmd_k); + flush_pmd_entry(pmd); + return 0; + +bad_area: + do_bad_area(addr, fsr, regs); + return 0; +} + +/* + * This abort handler always returns "fault". + */ +static int do_bad(unsigned long addr, unsigned int fsr, struct pt_regs *regs) +{ + return 1; +} + +static int do_good(unsigned long addr, unsigned int fsr, struct pt_regs *regs) +{ + unsigned int res1, res2; + + printk("dabt exception but no error!\n"); + + __asm__ __volatile__( + "mff %0,f0\n" + "mff %1,f1\n" + : "=r"(res1), "=r"(res2) + : + : "memory"); + + printk(KERN_EMERG "r0 :%08x r1 :%08x\n", res1, res2); + panic("shut up\n"); + return 0; +} + +static struct fsr_info { + int (*fn) (unsigned long addr, unsigned int fsr, struct pt_regs *regs); + int sig; + int code; + const char *name; +} fsr_info[] = { + /* + * The following are the standard Unicore-I and UniCore-II aborts. + */ + { do_good, SIGBUS, 0, "no error" }, + { do_bad, SIGBUS, BUS_ADRALN, "alignment exception" }, + { do_bad, SIGBUS, BUS_OBJERR, "external exception" }, + { do_bad, SIGBUS, 0, "burst operation" }, + { do_bad, SIGBUS, 0, "unknown 00100" }, + { do_ifault, SIGSEGV, SEGV_MAPERR, "2nd level pt non-exist"}, + { do_bad, SIGBUS, 0, "2nd lvl large pt non-exist" }, + { do_bad, SIGBUS, 0, "invalid pte" }, + { do_pf, SIGSEGV, SEGV_MAPERR, "page miss" }, + { do_bad, SIGBUS, 0, "middle page miss" }, + { do_bad, SIGBUS, 0, "large page miss" }, + { do_pf, SIGSEGV, SEGV_MAPERR, "super page (section) miss" }, + { do_bad, SIGBUS, 0, "unknown 01100" }, + { do_bad, SIGBUS, 0, "unknown 01101" }, + { do_bad, SIGBUS, 0, "unknown 01110" }, + { do_bad, SIGBUS, 0, "unknown 01111" }, + { do_bad, SIGBUS, 0, "addr: up 3G or IO" }, + { do_pf, SIGSEGV, SEGV_ACCERR, "read unreadable addr" }, + { do_pf, SIGSEGV, SEGV_ACCERR, "write unwriteable addr"}, + { do_pf, SIGSEGV, SEGV_ACCERR, "exec unexecutable addr"}, + { do_bad, SIGBUS, 0, "unknown 10100" }, + { do_bad, SIGBUS, 0, "unknown 10101" }, + { do_bad, SIGBUS, 0, "unknown 10110" }, + { do_bad, SIGBUS, 0, "unknown 10111" }, + { do_bad, SIGBUS, 0, "unknown 11000" }, + { do_bad, SIGBUS, 0, "unknown 11001" }, + { do_bad, SIGBUS, 0, "unknown 11010" }, + { do_bad, SIGBUS, 0, "unknown 11011" }, + { do_bad, SIGBUS, 0, "unknown 11100" }, + { do_bad, SIGBUS, 0, "unknown 11101" }, + { do_bad, SIGBUS, 0, "unknown 11110" }, + { do_bad, SIGBUS, 0, "unknown 11111" } +}; + +void __init hook_fault_code(int nr, + int (*fn) (unsigned long, unsigned int, struct pt_regs *), + int sig, int code, const char *name) +{ + if (nr < 0 || nr >= ARRAY_SIZE(fsr_info)) + BUG(); + + fsr_info[nr].fn = fn; + fsr_info[nr].sig = sig; + fsr_info[nr].code = code; + fsr_info[nr].name = name; +} + +/* + * Dispatch a data abort to the relevant handler. + */ +asmlinkage void do_DataAbort(unsigned long addr, unsigned int fsr, + struct pt_regs *regs) +{ + const struct fsr_info *inf = fsr_info + fsr_fs(fsr); + struct siginfo info; + + if (!inf->fn(addr, fsr & ~FSR_LNX_PF, regs)) + return; + + printk(KERN_ALERT "Unhandled fault: %s (0x%03x) at 0x%08lx\n", + inf->name, fsr, addr); + + info.si_signo = inf->sig; + info.si_errno = 0; + info.si_code = inf->code; + info.si_addr = (void __user *)addr; + uc32_notify_die("", regs, &info, fsr, 0); +} + +asmlinkage void do_PrefetchAbort(unsigned long addr, + unsigned int ifsr, struct pt_regs *regs) +{ + const struct fsr_info *inf = fsr_info + fsr_fs(ifsr); + struct siginfo info; + + if (!inf->fn(addr, ifsr | FSR_LNX_PF, regs)) + return; + + printk(KERN_ALERT "Unhandled prefetch abort: %s (0x%03x) at 0x%08lx\n", + inf->name, ifsr, addr); + + info.si_signo = inf->sig; + info.si_errno = 0; + info.si_code = inf->code; + info.si_addr = (void __user *)addr; + uc32_notify_die("", regs, &info, ifsr, 0); +} diff --git a/arch/unicore32/mm/mmu.c b/arch/unicore32/mm/mmu.c new file mode 100644 index 000000000000..7bf3d588631f --- /dev/null +++ b/arch/unicore32/mm/mmu.c @@ -0,0 +1,533 @@ +/* + * linux/arch/unicore32/mm/mmu.c + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include + +#include "mm.h" + +DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); + +/* + * empty_zero_page is a special page that is used for + * zero-initialized data and COW. + */ +struct page *empty_zero_page; +EXPORT_SYMBOL(empty_zero_page); + +/* + * The pmd table for the upper-most set of pages. + */ +pmd_t *top_pmd; + +pgprot_t pgprot_user; +EXPORT_SYMBOL(pgprot_user); + +pgprot_t pgprot_kernel; +EXPORT_SYMBOL(pgprot_kernel); + +static int __init noalign_setup(char *__unused) +{ + cr_alignment &= ~CR_A; + cr_no_alignment &= ~CR_A; + set_cr(cr_alignment); + return 1; +} +__setup("noalign", noalign_setup); + +void adjust_cr(unsigned long mask, unsigned long set) +{ + unsigned long flags; + + mask &= ~CR_A; + + set &= mask; + + local_irq_save(flags); + + cr_no_alignment = (cr_no_alignment & ~mask) | set; + cr_alignment = (cr_alignment & ~mask) | set; + + set_cr((get_cr() & ~mask) | set); + + local_irq_restore(flags); +} + +struct map_desc { + unsigned long virtual; + unsigned long pfn; + unsigned long length; + unsigned int type; +}; + +#define PROT_PTE_DEVICE (PTE_PRESENT | PTE_YOUNG | \ + PTE_DIRTY | PTE_READ | PTE_WRITE) +#define PROT_SECT_DEVICE (PMD_TYPE_SECT | PMD_PRESENT | \ + PMD_SECT_READ | PMD_SECT_WRITE) + +static struct mem_type mem_types[] = { + [MT_DEVICE] = { /* Strongly ordered */ + .prot_pte = PROT_PTE_DEVICE, + .prot_l1 = PMD_TYPE_TABLE | PMD_PRESENT, + .prot_sect = PROT_SECT_DEVICE, + }, + /* + * MT_KUSER: pte for vecpage -- cacheable, + * and sect for unigfx mmap -- noncacheable + */ + [MT_KUSER] = { + .prot_pte = PTE_PRESENT | PTE_YOUNG | PTE_DIRTY | + PTE_CACHEABLE | PTE_READ | PTE_EXEC, + .prot_l1 = PMD_TYPE_TABLE | PMD_PRESENT, + .prot_sect = PROT_SECT_DEVICE, + }, + [MT_HIGH_VECTORS] = { + .prot_pte = PTE_PRESENT | PTE_YOUNG | PTE_DIRTY | + PTE_CACHEABLE | PTE_READ | PTE_WRITE | + PTE_EXEC, + .prot_l1 = PMD_TYPE_TABLE | PMD_PRESENT, + }, + [MT_MEMORY] = { + .prot_pte = PTE_PRESENT | PTE_YOUNG | PTE_DIRTY | + PTE_WRITE | PTE_EXEC, + .prot_l1 = PMD_TYPE_TABLE | PMD_PRESENT, + .prot_sect = PMD_TYPE_SECT | PMD_PRESENT | PMD_SECT_CACHEABLE | + PMD_SECT_READ | PMD_SECT_WRITE | PMD_SECT_EXEC, + }, + [MT_ROM] = { + .prot_sect = PMD_TYPE_SECT | PMD_PRESENT | PMD_SECT_CACHEABLE | + PMD_SECT_READ, + }, +}; + +const struct mem_type *get_mem_type(unsigned int type) +{ + return type < ARRAY_SIZE(mem_types) ? &mem_types[type] : NULL; +} +EXPORT_SYMBOL(get_mem_type); + +/* + * Adjust the PMD section entries according to the CPU in use. + */ +static void __init build_mem_type_table(void) +{ + pgprot_user = __pgprot(PTE_PRESENT | PTE_YOUNG | PTE_CACHEABLE); + pgprot_kernel = __pgprot(PTE_PRESENT | PTE_YOUNG | + PTE_DIRTY | PTE_READ | PTE_WRITE | + PTE_EXEC | PTE_CACHEABLE); +} + +#define vectors_base() (vectors_high() ? 0xffff0000 : 0) + +static void __init *early_alloc(unsigned long sz) +{ + void *ptr = __va(memblock_alloc(sz, sz)); + memset(ptr, 0, sz); + return ptr; +} + +static pte_t * __init early_pte_alloc(pmd_t *pmd, unsigned long addr, + unsigned long prot) +{ + if (pmd_none(*pmd)) { + pte_t *pte = early_alloc(PTRS_PER_PTE * sizeof(pte_t)); + __pmd_populate(pmd, __pa(pte) | prot); + } + BUG_ON(pmd_bad(*pmd)); + return pte_offset_kernel(pmd, addr); +} + +static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr, + unsigned long end, unsigned long pfn, + const struct mem_type *type) +{ + pte_t *pte = early_pte_alloc(pmd, addr, type->prot_l1); + do { + set_pte(pte, pfn_pte(pfn, __pgprot(type->prot_pte))); + pfn++; + } while (pte++, addr += PAGE_SIZE, addr != end); +} + +static void __init alloc_init_section(pgd_t *pgd, unsigned long addr, + unsigned long end, unsigned long phys, + const struct mem_type *type) +{ + pmd_t *pmd = pmd_offset((pud_t *)pgd, addr); + + /* + * Try a section mapping - end, addr and phys must all be aligned + * to a section boundary. + */ + if (((addr | end | phys) & ~SECTION_MASK) == 0) { + pmd_t *p = pmd; + + do { + set_pmd(pmd, __pmd(phys | type->prot_sect)); + phys += SECTION_SIZE; + } while (pmd++, addr += SECTION_SIZE, addr != end); + + flush_pmd_entry(p); + } else { + /* + * No need to loop; pte's aren't interested in the + * individual L1 entries. + */ + alloc_init_pte(pmd, addr, end, __phys_to_pfn(phys), type); + } +} + +/* + * Create the page directory entries and any necessary + * page tables for the mapping specified by `md'. We + * are able to cope here with varying sizes and address + * offsets, and we take full advantage of sections. + */ +static void __init create_mapping(struct map_desc *md) +{ + unsigned long phys, addr, length, end; + const struct mem_type *type; + pgd_t *pgd; + + if (md->virtual != vectors_base() && md->virtual < TASK_SIZE) { + printk(KERN_WARNING "BUG: not creating mapping for " + "0x%08llx at 0x%08lx in user region\n", + __pfn_to_phys((u64)md->pfn), md->virtual); + return; + } + + if ((md->type == MT_DEVICE || md->type == MT_ROM) && + md->virtual >= PAGE_OFFSET && md->virtual < VMALLOC_END) { + printk(KERN_WARNING "BUG: mapping for 0x%08llx at 0x%08lx " + "overlaps vmalloc space\n", + __pfn_to_phys((u64)md->pfn), md->virtual); + } + + type = &mem_types[md->type]; + + addr = md->virtual & PAGE_MASK; + phys = (unsigned long)__pfn_to_phys(md->pfn); + length = PAGE_ALIGN(md->length + (md->virtual & ~PAGE_MASK)); + + if (type->prot_l1 == 0 && ((addr | phys | length) & ~SECTION_MASK)) { + printk(KERN_WARNING "BUG: map for 0x%08lx at 0x%08lx can not " + "be mapped using pages, ignoring.\n", + __pfn_to_phys(md->pfn), addr); + return; + } + + pgd = pgd_offset_k(addr); + end = addr + length; + do { + unsigned long next = pgd_addr_end(addr, end); + + alloc_init_section(pgd, addr, next, phys, type); + + phys += next - addr; + addr = next; + } while (pgd++, addr != end); +} + +static void * __initdata vmalloc_min = (void *)(VMALLOC_END - SZ_128M); + +/* + * vmalloc=size forces the vmalloc area to be exactly 'size' + * bytes. This can be used to increase (or decrease) the vmalloc + * area - the default is 128m. + */ +static int __init early_vmalloc(char *arg) +{ + unsigned long vmalloc_reserve = memparse(arg, NULL); + + if (vmalloc_reserve < SZ_16M) { + vmalloc_reserve = SZ_16M; + printk(KERN_WARNING + "vmalloc area too small, limiting to %luMB\n", + vmalloc_reserve >> 20); + } + + if (vmalloc_reserve > VMALLOC_END - (PAGE_OFFSET + SZ_32M)) { + vmalloc_reserve = VMALLOC_END - (PAGE_OFFSET + SZ_32M); + printk(KERN_WARNING + "vmalloc area is too big, limiting to %luMB\n", + vmalloc_reserve >> 20); + } + + vmalloc_min = (void *)(VMALLOC_END - vmalloc_reserve); + return 0; +} +early_param("vmalloc", early_vmalloc); + +static phys_addr_t lowmem_limit __initdata = SZ_1G; + +static void __init sanity_check_meminfo(void) +{ + int i, j; + + lowmem_limit = __pa(vmalloc_min - 1) + 1; + memblock_set_current_limit(lowmem_limit); + + for (i = 0, j = 0; i < meminfo.nr_banks; i++) { + struct membank *bank = &meminfo.bank[j]; + *bank = meminfo.bank[i]; + j++; + } + meminfo.nr_banks = j; +} + +static inline void prepare_page_table(void) +{ + unsigned long addr; + phys_addr_t end; + + /* + * Clear out all the mappings below the kernel image. + */ + for (addr = 0; addr < MODULES_VADDR; addr += PGDIR_SIZE) + pmd_clear(pmd_off_k(addr)); + + for ( ; addr < PAGE_OFFSET; addr += PGDIR_SIZE) + pmd_clear(pmd_off_k(addr)); + + /* + * Find the end of the first block of lowmem. + */ + end = memblock.memory.regions[0].base + memblock.memory.regions[0].size; + if (end >= lowmem_limit) + end = lowmem_limit; + + /* + * Clear out all the kernel space mappings, except for the first + * memory bank, up to the end of the vmalloc region. + */ + for (addr = __phys_to_virt(end); + addr < VMALLOC_END; addr += PGDIR_SIZE) + pmd_clear(pmd_off_k(addr)); +} + +/* + * Reserve the special regions of memory + */ +void __init uc32_mm_memblock_reserve(void) +{ + /* + * Reserve the page tables. These are already in use, + * and can only be in node 0. + */ + memblock_reserve(__pa(swapper_pg_dir), PTRS_PER_PGD * sizeof(pgd_t)); + +#ifdef CONFIG_PUV3_UNIGFX + /* + * These should likewise go elsewhere. They pre-reserve the + * screen/video memory region at the 48M~64M of main system memory. + */ + memblock_reserve(PKUNITY_UNIGFX_MMAP_BASE, PKUNITY_UNIGFX_MMAP_SIZE); + memblock_reserve(PKUNITY_UVC_MMAP_BASE, PKUNITY_UVC_MMAP_SIZE); +#endif +} + +/* + * Set up device the mappings. Since we clear out the page tables for all + * mappings above VMALLOC_END, we will remove any debug device mappings. + * This means you have to be careful how you debug this function, or any + * called function. This means you can't use any function or debugging + * method which may touch any device, otherwise the kernel _will_ crash. + */ +static void __init devicemaps_init(void) +{ + struct map_desc map; + unsigned long addr; + void *vectors; + + /* + * Allocate the vector page early. + */ + vectors = early_alloc(PAGE_SIZE); + + for (addr = VMALLOC_END; addr; addr += PGDIR_SIZE) + pmd_clear(pmd_off_k(addr)); + + /* + * Create a mapping for UniGFX VRAM + */ +#ifdef CONFIG_PUV3_UNIGFX + map.pfn = __phys_to_pfn(PKUNITY_UNIGFX_MMAP_BASE); + map.virtual = KUSER_UNIGFX_BASE; + map.length = PKUNITY_UNIGFX_MMAP_SIZE; + map.type = MT_KUSER; + create_mapping(&map); +#endif + + /* + * Create a mapping for the machine vectors at the high-vectors + * location (0xffff0000). If we aren't using high-vectors, also + * create a mapping at the low-vectors virtual address. + */ + map.pfn = __phys_to_pfn(virt_to_phys(vectors)); + map.virtual = VECTORS_BASE; + map.length = PAGE_SIZE; + map.type = MT_HIGH_VECTORS; + create_mapping(&map); + + /* + * Create a mapping for the kuser page at the special + * location (0xbfff0000) to the same vectors location. + */ + map.pfn = __phys_to_pfn(virt_to_phys(vectors)); + map.virtual = KUSER_VECPAGE_BASE; + map.length = PAGE_SIZE; + map.type = MT_KUSER; + create_mapping(&map); + + /* + * Finally flush the caches and tlb to ensure that we're in a + * consistent state wrt the writebuffer. This also ensures that + * any write-allocated cache lines in the vector page are written + * back. After this point, we can start to touch devices again. + */ + local_flush_tlb_all(); + flush_cache_all(); +} + +static void __init map_lowmem(void) +{ + struct memblock_region *reg; + + /* Map all the lowmem memory banks. */ + for_each_memblock(memory, reg) { + phys_addr_t start = reg->base; + phys_addr_t end = start + reg->size; + struct map_desc map; + + if (end > lowmem_limit) + end = lowmem_limit; + if (start >= end) + break; + + map.pfn = __phys_to_pfn(start); + map.virtual = __phys_to_virt(start); + map.length = end - start; + map.type = MT_MEMORY; + + create_mapping(&map); + } +} + +/* + * paging_init() sets up the page tables, initialises the zone memory + * maps, and sets up the zero page, bad page and bad page tables. + */ +void __init paging_init(void) +{ + void *zero_page; + + build_mem_type_table(); + sanity_check_meminfo(); + prepare_page_table(); + map_lowmem(); + devicemaps_init(); + + top_pmd = pmd_off_k(0xffff0000); + + /* allocate the zero page. */ + zero_page = early_alloc(PAGE_SIZE); + + bootmem_init(); + + empty_zero_page = virt_to_page(zero_page); + __flush_dcache_page(NULL, empty_zero_page); +} + +/* + * In order to soft-boot, we need to insert a 1:1 mapping in place of + * the user-mode pages. This will then ensure that we have predictable + * results when turning the mmu off + */ +void setup_mm_for_reboot(char mode) +{ + unsigned long base_pmdval; + pgd_t *pgd; + int i; + + /* + * We need to access to user-mode page tables here. For kernel threads + * we don't have any user-mode mappings so we use the context that we + * "borrowed". + */ + pgd = current->active_mm->pgd; + + base_pmdval = PMD_SECT_WRITE | PMD_SECT_READ | PMD_TYPE_SECT; + + for (i = 0; i < FIRST_USER_PGD_NR + USER_PTRS_PER_PGD; i++, pgd++) { + unsigned long pmdval = (i << PGDIR_SHIFT) | base_pmdval; + pmd_t *pmd; + + pmd = pmd_off(pgd, i << PGDIR_SHIFT); + set_pmd(pmd, __pmd(pmdval)); + flush_pmd_entry(pmd); + } + + local_flush_tlb_all(); +} + +/* + * Take care of architecture specific things when placing a new PTE into + * a page table, or changing an existing PTE. Basically, there are two + * things that we need to take care of: + * + * 1. If PG_dcache_clean is not set for the page, we need to ensure + * that any cache entries for the kernels virtual memory + * range are written back to the page. + * 2. If we have multiple shared mappings of the same space in + * an object, we need to deal with the cache aliasing issues. + * + * Note that the pte lock will be held. + */ +void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr, + pte_t *ptep) +{ + unsigned long pfn = pte_pfn(*ptep); + struct address_space *mapping; + struct page *page; + + if (!pfn_valid(pfn)) + return; + + /* + * The zero page is never written to, so never has any dirty + * cache lines, and therefore never needs to be flushed. + */ + page = pfn_to_page(pfn); + if (page == ZERO_PAGE(0)) + return; + + mapping = page_mapping(page); + if (!test_and_set_bit(PG_dcache_clean, &page->flags)) + __flush_dcache_page(mapping, page); + if (mapping) + if (vma->vm_flags & VM_EXEC) + __flush_icache_all(); +} diff --git a/arch/unicore32/mm/pgd.c b/arch/unicore32/mm/pgd.c new file mode 100644 index 000000000000..632cef7cd378 --- /dev/null +++ b/arch/unicore32/mm/pgd.c @@ -0,0 +1,102 @@ +/* + * linux/arch/unicore32/mm/pgd.c + * + * Code specific to PKUnity SoC and UniCore ISA + * + * Copyright (C) 2001-2010 GUAN Xue-tao + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include + +#include +#include +#include + +#include "mm.h" + +#define FIRST_KERNEL_PGD_NR (FIRST_USER_PGD_NR + USER_PTRS_PER_PGD) + +/* + * need to get a 4k page for level 1 + */ +pgd_t *get_pgd_slow(struct mm_struct *mm) +{ + pgd_t *new_pgd, *init_pgd; + pmd_t *new_pmd, *init_pmd; + pte_t *new_pte, *init_pte; + + new_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL, 0); + if (!new_pgd) + goto no_pgd; + + memset(new_pgd, 0, FIRST_KERNEL_PGD_NR * sizeof(pgd_t)); + + /* + * Copy over the kernel and IO PGD entries + */ + init_pgd = pgd_offset_k(0); + memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR, + (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t)); + + clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t)); + + if (!vectors_high()) { + /* + * On UniCore, first page must always be allocated since it + * contains the machine vectors. + */ + new_pmd = pmd_alloc(mm, (pud_t *)new_pgd, 0); + if (!new_pmd) + goto no_pmd; + + new_pte = pte_alloc_map(mm, new_pmd, 0); + if (!new_pte) + goto no_pte; + + init_pmd = pmd_offset((pud_t *)init_pgd, 0); + init_pte = pte_offset_map(init_pmd, 0); + set_pte(new_pte, *init_pte); + pte_unmap(init_pte); + pte_unmap(new_pte); + } + + return new_pgd; + +no_pte: + pmd_free(mm, new_pmd); +no_pmd: + free_pages((unsigned long)new_pgd, 0); +no_pgd: + return NULL; +} + +void free_pgd_slow(struct mm_struct *mm, pgd_t *pgd) +{ + pmd_t *pmd; + pgtable_t pte; + + if (!pgd) + return; + + /* pgd is always present and good */ + pmd = pmd_off(pgd, 0); + if (pmd_none(*pmd)) + goto free; + if (pmd_bad(*pmd)) { + pmd_ERROR(*pmd); + pmd_clear(pmd); + goto free; + } + + pte = pmd_pgtable(*pmd); + pmd_clear(pmd); + pte_free(mm, pte); + pmd_free(mm, pmd); +free: + free_pages((unsigned long) pgd, 0); +}