diff options
Diffstat (limited to 'arch/riscv/mm')
| -rw-r--r-- | arch/riscv/mm/Makefile | 2 | ||||
| -rw-r--r-- | arch/riscv/mm/cacheflush.c | 31 | ||||
| -rw-r--r-- | arch/riscv/mm/context.c | 2 | ||||
| -rw-r--r-- | arch/riscv/mm/fault.c | 52 | ||||
| -rw-r--r-- | arch/riscv/mm/hugetlbpage.c | 78 | ||||
| -rw-r--r-- | arch/riscv/mm/init.c | 163 | ||||
| -rw-r--r-- | arch/riscv/mm/kasan_init.c | 14 | ||||
| -rw-r--r-- | arch/riscv/mm/pgtable.c | 10 | ||||
| -rw-r--r-- | arch/riscv/mm/physaddr.c | 2 | ||||
| -rw-r--r-- | arch/riscv/mm/ptdump.c | 46 | ||||
| -rw-r--r-- | arch/riscv/mm/tlbflush.c | 76 |
11 files changed, 339 insertions, 137 deletions
diff --git a/arch/riscv/mm/Makefile b/arch/riscv/mm/Makefile index cbe4d775ef56..b916a68d324a 100644 --- a/arch/riscv/mm/Makefile +++ b/arch/riscv/mm/Makefile @@ -19,7 +19,7 @@ obj-y += context.o obj-y += pmem.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o -obj-$(CONFIG_PTDUMP_CORE) += ptdump.o +obj-$(CONFIG_PTDUMP) += ptdump.o obj-$(CONFIG_KASAN) += kasan_init.o ifdef CONFIG_KASAN diff --git a/arch/riscv/mm/cacheflush.c b/arch/riscv/mm/cacheflush.c index b81672729887..4ca5aafce22e 100644 --- a/arch/riscv/mm/cacheflush.c +++ b/arch/riscv/mm/cacheflush.c @@ -24,7 +24,20 @@ void flush_icache_all(void) if (num_online_cpus() < 2) return; - else if (riscv_use_sbi_for_rfence()) + + /* + * Make sure all previous writes to the D$ are ordered before making + * the IPI. The RISC-V spec states that a hart must execute a data fence + * before triggering a remote fence.i in order to make the modification + * visable for remote harts. + * + * IPIs on RISC-V are triggered by MMIO writes to either CLINT or + * S-IMSIC, so the fence ensures previous data writes "happen before" + * the MMIO. + */ + RISCV_FENCE(w, o); + + if (riscv_use_sbi_for_rfence()) sbi_remote_fence_i(NULL); else on_each_cpu(ipi_remote_fence_i, NULL, 1); @@ -101,6 +114,9 @@ EXPORT_SYMBOL_GPL(riscv_cbom_block_size); unsigned int riscv_cboz_block_size; EXPORT_SYMBOL_GPL(riscv_cboz_block_size); +unsigned int riscv_cbop_block_size; +EXPORT_SYMBOL_GPL(riscv_cbop_block_size); + static void __init cbo_get_block_size(struct device_node *node, const char *name, u32 *block_size, unsigned long *first_hartid) @@ -125,8 +141,8 @@ static void __init cbo_get_block_size(struct device_node *node, void __init riscv_init_cbo_blocksizes(void) { - unsigned long cbom_hartid, cboz_hartid; - u32 cbom_block_size = 0, cboz_block_size = 0; + unsigned long cbom_hartid, cboz_hartid, cbop_hartid; + u32 cbom_block_size = 0, cboz_block_size = 0, cbop_block_size = 0; struct device_node *node; struct acpi_table_header *rhct; acpi_status status; @@ -138,13 +154,15 @@ void __init riscv_init_cbo_blocksizes(void) &cbom_block_size, &cbom_hartid); cbo_get_block_size(node, "riscv,cboz-block-size", &cboz_block_size, &cboz_hartid); + cbo_get_block_size(node, "riscv,cbop-block-size", + &cbop_block_size, &cbop_hartid); } } else { status = acpi_get_table(ACPI_SIG_RHCT, 0, &rhct); if (ACPI_FAILURE(status)) return; - acpi_get_cbo_block_size(rhct, &cbom_block_size, &cboz_block_size, NULL); + acpi_get_cbo_block_size(rhct, &cbom_block_size, &cboz_block_size, &cbop_block_size); acpi_put_table((struct acpi_table_header *)rhct); } @@ -153,6 +171,9 @@ void __init riscv_init_cbo_blocksizes(void) if (cboz_block_size) riscv_cboz_block_size = cboz_block_size; + + if (cbop_block_size) + riscv_cbop_block_size = cbop_block_size; } #ifdef CONFIG_SMP @@ -172,7 +193,7 @@ static void set_icache_stale_mask(void) stale_cpu = cpumask_test_cpu(cpu, mask); cpumask_setall(mask); - cpumask_assign_cpu(cpu, mask, stale_cpu); + __assign_cpu(cpu, mask, stale_cpu); put_cpu(); } #endif diff --git a/arch/riscv/mm/context.c b/arch/riscv/mm/context.c index 4abe3de23225..55c20ad1f744 100644 --- a/arch/riscv/mm/context.c +++ b/arch/riscv/mm/context.c @@ -158,7 +158,7 @@ static void set_mm_asid(struct mm_struct *mm, unsigned int cpu) * * - We get a zero back from the cmpxchg and end up waiting on the * lock. Taking the lock synchronises with the rollover and so - * we are forced to see the updated verion. + * we are forced to see the updated version. * * - We get a valid context back from the cmpxchg then we continue * using old ASID because __flush_context() would have marked ASID diff --git a/arch/riscv/mm/fault.c b/arch/riscv/mm/fault.c index a9f2b4af8f3f..0194324a0c50 100644 --- a/arch/riscv/mm/fault.c +++ b/arch/riscv/mm/fault.c @@ -22,6 +22,57 @@ #include "../kernel/head.h" +static void show_pte(unsigned long addr) +{ + pgd_t *pgdp, pgd; + p4d_t *p4dp, p4d; + pud_t *pudp, pud; + pmd_t *pmdp, pmd; + pte_t *ptep, pte; + struct mm_struct *mm = current->mm; + + if (!mm) + mm = &init_mm; + + pr_alert("Current %s pgtable: %luK pagesize, %d-bit VAs, pgdp=0x%016llx\n", + current->comm, PAGE_SIZE / SZ_1K, VA_BITS, + mm == &init_mm ? (u64)__pa_symbol(mm->pgd) : virt_to_phys(mm->pgd)); + + pgdp = pgd_offset(mm, addr); + pgd = pgdp_get(pgdp); + pr_alert("[%016lx] pgd=%016lx", addr, pgd_val(pgd)); + if (pgd_none(pgd) || pgd_bad(pgd) || pgd_leaf(pgd)) + goto out; + + p4dp = p4d_offset(pgdp, addr); + p4d = p4dp_get(p4dp); + pr_cont(", p4d=%016lx", p4d_val(p4d)); + if (p4d_none(p4d) || p4d_bad(p4d) || p4d_leaf(p4d)) + goto out; + + pudp = pud_offset(p4dp, addr); + pud = pudp_get(pudp); + pr_cont(", pud=%016lx", pud_val(pud)); + if (pud_none(pud) || pud_bad(pud) || pud_leaf(pud)) + goto out; + + pmdp = pmd_offset(pudp, addr); + pmd = pmdp_get(pmdp); + pr_cont(", pmd=%016lx", pmd_val(pmd)); + if (pmd_none(pmd) || pmd_bad(pmd) || pmd_leaf(pmd)) + goto out; + + ptep = pte_offset_map(pmdp, addr); + if (!ptep) + goto out; + + pte = ptep_get(ptep); + pr_cont(", pte=%016lx", pte_val(pte)); + pte_unmap(ptep); +out: + pr_cont("\n"); +} + static void die_kernel_fault(const char *msg, unsigned long addr, struct pt_regs *regs) { @@ -31,6 +82,7 @@ static void die_kernel_fault(const char *msg, unsigned long addr, addr); bust_spinlocks(0); + show_pte(addr); die(regs, "Oops"); make_task_dead(SIGKILL); } diff --git a/arch/riscv/mm/hugetlbpage.c b/arch/riscv/mm/hugetlbpage.c index 42314f093922..375dd96bb4a0 100644 --- a/arch/riscv/mm/hugetlbpage.c +++ b/arch/riscv/mm/hugetlbpage.c @@ -148,22 +148,25 @@ unsigned long hugetlb_mask_last_page(struct hstate *h) static pte_t get_clear_contig(struct mm_struct *mm, unsigned long addr, pte_t *ptep, - unsigned long pte_num) + unsigned long ncontig) { - pte_t orig_pte = ptep_get(ptep); - unsigned long i; - - for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++) { - pte_t pte = ptep_get_and_clear(mm, addr, ptep); - - if (pte_dirty(pte)) - orig_pte = pte_mkdirty(orig_pte); - - if (pte_young(pte)) - orig_pte = pte_mkyoung(orig_pte); + pte_t pte, tmp_pte; + bool present; + + pte = ptep_get_and_clear(mm, addr, ptep); + present = pte_present(pte); + while (--ncontig) { + ptep++; + addr += PAGE_SIZE; + tmp_pte = ptep_get_and_clear(mm, addr, ptep); + if (present) { + if (pte_dirty(tmp_pte)) + pte = pte_mkdirty(pte); + if (pte_young(tmp_pte)) + pte = pte_mkyoung(pte); + } } - - return orig_pte; + return pte; } static pte_t get_clear_contig_flush(struct mm_struct *mm, @@ -212,6 +215,26 @@ static void clear_flush(struct mm_struct *mm, flush_tlb_range(&vma, saddr, addr); } +static int num_contig_ptes_from_size(unsigned long sz, size_t *pgsize) +{ + unsigned long hugepage_shift; + + if (sz >= PGDIR_SIZE) + hugepage_shift = PGDIR_SHIFT; + else if (sz >= P4D_SIZE) + hugepage_shift = P4D_SHIFT; + else if (sz >= PUD_SIZE) + hugepage_shift = PUD_SHIFT; + else if (sz >= PMD_SIZE) + hugepage_shift = PMD_SHIFT; + else + hugepage_shift = PAGE_SHIFT; + + *pgsize = 1 << hugepage_shift; + + return sz >> hugepage_shift; +} + /* * When dealing with NAPOT mappings, the privileged specification indicates that * "if an update needs to be made, the OS generally should first mark all of the @@ -226,22 +249,10 @@ void set_huge_pte_at(struct mm_struct *mm, pte_t pte, unsigned long sz) { - unsigned long hugepage_shift, pgsize; + size_t pgsize; int i, pte_num; - if (sz >= PGDIR_SIZE) - hugepage_shift = PGDIR_SHIFT; - else if (sz >= P4D_SIZE) - hugepage_shift = P4D_SHIFT; - else if (sz >= PUD_SIZE) - hugepage_shift = PUD_SHIFT; - else if (sz >= PMD_SIZE) - hugepage_shift = PMD_SHIFT; - else - hugepage_shift = PAGE_SHIFT; - - pte_num = sz >> hugepage_shift; - pgsize = 1 << hugepage_shift; + pte_num = num_contig_ptes_from_size(sz, &pgsize); if (!pte_present(pte)) { for (i = 0; i < pte_num; i++, ptep++, addr += pgsize) @@ -293,15 +304,16 @@ int huge_ptep_set_access_flags(struct vm_area_struct *vma, pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, - pte_t *ptep) + pte_t *ptep, unsigned long sz) { + size_t pgsize; pte_t orig_pte = ptep_get(ptep); int pte_num; if (!pte_napot(orig_pte)) return ptep_get_and_clear(mm, addr, ptep); - pte_num = napot_pte_num(napot_cont_order(orig_pte)); + pte_num = num_contig_ptes_from_size(sz, &pgsize); return get_clear_contig(mm, addr, ptep, pte_num); } @@ -351,6 +363,7 @@ void huge_pte_clear(struct mm_struct *mm, pte_t *ptep, unsigned long sz) { + size_t pgsize; pte_t pte = ptep_get(ptep); int i, pte_num; @@ -359,8 +372,9 @@ void huge_pte_clear(struct mm_struct *mm, return; } - pte_num = napot_pte_num(napot_cont_order(pte)); - for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++) + pte_num = num_contig_ptes_from_size(sz, &pgsize); + + for (i = 0; i < pte_num; i++, addr += pgsize, ptep++) pte_clear(mm, addr, ptep); } diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c index fc53ce748c80..8d0374d7ce8e 100644 --- a/arch/riscv/mm/init.c +++ b/arch/riscv/mm/init.c @@ -20,19 +20,18 @@ #include <linux/dma-map-ops.h> #include <linux/crash_dump.h> #include <linux/hugetlb.h> -#ifdef CONFIG_RELOCATABLE -#include <linux/elf.h> -#endif #include <linux/kfence.h> #include <linux/execmem.h> #include <asm/fixmap.h> #include <asm/io.h> #include <asm/kasan.h> +#include <asm/module.h> #include <asm/numa.h> #include <asm/pgtable.h> #include <asm/sections.h> #include <asm/soc.h> +#include <asm/sparsemem.h> #include <asm/tlbflush.h> #include "../kernel/head.h" @@ -62,6 +61,13 @@ EXPORT_SYMBOL(pgtable_l5_enabled); phys_addr_t phys_ram_base __ro_after_init; EXPORT_SYMBOL(phys_ram_base); +#ifdef CONFIG_SPARSEMEM_VMEMMAP +#define VMEMMAP_ADDR_ALIGN (1ULL << SECTION_SIZE_BITS) + +unsigned long vmemmap_start_pfn __ro_after_init; +EXPORT_SYMBOL(vmemmap_start_pfn); +#endif + unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)] __page_aligned_bss; EXPORT_SYMBOL(empty_zero_page); @@ -163,7 +169,7 @@ static void __init print_vm_layout(void) static void print_vm_layout(void) { } #endif /* CONFIG_DEBUG_VM */ -void __init mem_init(void) +void __init arch_mm_preinit(void) { bool swiotlb = max_pfn > PFN_DOWN(dma32_phys_limit); #ifdef CONFIG_FLATMEM @@ -184,7 +190,6 @@ void __init mem_init(void) } swiotlb_init(swiotlb, SWIOTLB_VERBOSE); - memblock_free_all(); print_vm_layout(); } @@ -240,8 +245,12 @@ static void __init setup_bootmem(void) * Make sure we align the start of the memory on a PMD boundary so that * at worst, we map the linear mapping with PMD mappings. */ - if (!IS_ENABLED(CONFIG_XIP_KERNEL)) + if (!IS_ENABLED(CONFIG_XIP_KERNEL)) { phys_ram_base = memblock_start_of_DRAM() & PMD_MASK; +#ifdef CONFIG_SPARSEMEM_VMEMMAP + vmemmap_start_pfn = round_down(phys_ram_base, VMEMMAP_ADDR_ALIGN) >> PAGE_SHIFT; +#endif + } /* * In 64-bit, any use of __va/__pa before this point is wrong as we @@ -256,8 +265,12 @@ static void __init setup_bootmem(void) */ if (IS_ENABLED(CONFIG_64BIT) && IS_ENABLED(CONFIG_MMU)) { max_mapped_addr = __pa(PAGE_OFFSET) + KERN_VIRT_SIZE; - memblock_cap_memory_range(phys_ram_base, - max_mapped_addr - phys_ram_base); + if (memblock_end_of_DRAM() > max_mapped_addr) { + memblock_cap_memory_range(phys_ram_base, + max_mapped_addr - phys_ram_base); + pr_warn("Physical memory overflows the linear mapping size: region above %pa removed", + &max_mapped_addr); + } } /* @@ -279,10 +292,8 @@ static void __init setup_bootmem(void) phys_ram_end = memblock_end_of_DRAM(); min_low_pfn = PFN_UP(phys_ram_base); max_low_pfn = max_pfn = PFN_DOWN(phys_ram_end); - high_memory = (void *)(__va(PFN_PHYS(max_low_pfn))); dma32_phys_limit = min(4UL * SZ_1G, (unsigned long)PFN_PHYS(max_low_pfn)); - set_max_mapnr(max_low_pfn - ARCH_PFN_OFFSET); reserve_initrd_mem(); @@ -307,6 +318,44 @@ static void __init setup_bootmem(void) hugetlb_cma_reserve(PUD_SHIFT - PAGE_SHIFT); } +#ifdef CONFIG_RELOCATABLE +extern unsigned long __rela_dyn_start, __rela_dyn_end; + +static void __init relocate_kernel(void) +{ + Elf_Rela *rela = (Elf_Rela *)&__rela_dyn_start; + /* + * This holds the offset between the linked virtual address and the + * relocated virtual address. + */ + uintptr_t reloc_offset = kernel_map.virt_addr - KERNEL_LINK_ADDR; + /* + * This holds the offset between kernel linked virtual address and + * physical address. + */ + uintptr_t va_kernel_link_pa_offset = KERNEL_LINK_ADDR - kernel_map.phys_addr; + + for ( ; rela < (Elf_Rela *)&__rela_dyn_end; rela++) { + Elf_Addr addr = (rela->r_offset - va_kernel_link_pa_offset); + Elf_Addr relocated_addr = rela->r_addend; + + if (rela->r_info != R_RISCV_RELATIVE) + continue; + + /* + * Make sure to not relocate vdso symbols like rt_sigreturn + * which are linked from the address 0 in vmlinux since + * vdso symbol addresses are actually used as an offset from + * mm->context.vdso in VDSO_OFFSET macro. + */ + if (relocated_addr >= KERNEL_LINK_ADDR) + relocated_addr += reloc_offset; + + *(Elf_Addr *)addr = relocated_addr; + } +} +#endif /* CONFIG_RELOCATABLE */ + #ifdef CONFIG_MMU struct pt_alloc_ops pt_ops __meminitdata; @@ -393,7 +442,12 @@ static phys_addr_t __meminit alloc_pte_late(uintptr_t va) { struct ptdesc *ptdesc = pagetable_alloc(GFP_KERNEL & ~__GFP_HIGHMEM, 0); - BUG_ON(!ptdesc || !pagetable_pte_ctor(ptdesc)); + /* + * We do not know which mm the PTE page is associated to at this point. + * Passing NULL to the ctor is the safe option, though it may result + * in unnecessary work (e.g. initialising the ptlock for init_mm). + */ + BUG_ON(!ptdesc || !pagetable_pte_ctor(NULL, ptdesc)); return __pa((pte_t *)ptdesc_address(ptdesc)); } @@ -473,7 +527,8 @@ static phys_addr_t __meminit alloc_pmd_late(uintptr_t va) { struct ptdesc *ptdesc = pagetable_alloc(GFP_KERNEL & ~__GFP_HIGHMEM, 0); - BUG_ON(!ptdesc || !pagetable_pmd_ctor(ptdesc)); + /* See comment in alloc_pte_late() regarding NULL passed the ctor */ + BUG_ON(!ptdesc || !pagetable_pmd_ctor(NULL, ptdesc)); return __pa((pmd_t *)ptdesc_address(ptdesc)); } @@ -535,11 +590,11 @@ static phys_addr_t __init alloc_pud_fixmap(uintptr_t va) static phys_addr_t __meminit alloc_pud_late(uintptr_t va) { - unsigned long vaddr; + struct ptdesc *ptdesc = pagetable_alloc(GFP_KERNEL, 0); - vaddr = __get_free_page(GFP_KERNEL); - BUG_ON(!vaddr); - return __pa(vaddr); + BUG_ON(!ptdesc); + pagetable_pud_ctor(ptdesc); + return __pa((pud_t *)ptdesc_address(ptdesc)); } static p4d_t *__init get_p4d_virt_early(phys_addr_t pa) @@ -573,11 +628,11 @@ static phys_addr_t __init alloc_p4d_fixmap(uintptr_t va) static phys_addr_t __meminit alloc_p4d_late(uintptr_t va) { - unsigned long vaddr; + struct ptdesc *ptdesc = pagetable_alloc(GFP_KERNEL, 0); - vaddr = __get_free_page(GFP_KERNEL); - BUG_ON(!vaddr); - return __pa(vaddr); + BUG_ON(!ptdesc); + pagetable_p4d_ctor(ptdesc); + return __pa((p4d_t *)ptdesc_address(ptdesc)); } static void __meminit create_pud_mapping(pud_t *pudp, uintptr_t va, phys_addr_t pa, phys_addr_t sz, @@ -807,6 +862,8 @@ static __init void set_satp_mode(uintptr_t dtb_pa) uintptr_t set_satp_mode_pmd = ((unsigned long)set_satp_mode) & PMD_MASK; u64 satp_mode_cmdline = __pi_set_satp_mode_from_cmdline(dtb_pa); + kernel_map.page_offset = PAGE_OFFSET_L5; + if (satp_mode_cmdline == SATP_MODE_57) { disable_pgtable_l5(); } else if (satp_mode_cmdline == SATP_MODE_48) { @@ -877,44 +934,6 @@ retry: #error "setup_vm() is called from head.S before relocate so it should not use absolute addressing." #endif -#ifdef CONFIG_RELOCATABLE -extern unsigned long __rela_dyn_start, __rela_dyn_end; - -static void __init relocate_kernel(void) -{ - Elf64_Rela *rela = (Elf64_Rela *)&__rela_dyn_start; - /* - * This holds the offset between the linked virtual address and the - * relocated virtual address. - */ - uintptr_t reloc_offset = kernel_map.virt_addr - KERNEL_LINK_ADDR; - /* - * This holds the offset between kernel linked virtual address and - * physical address. - */ - uintptr_t va_kernel_link_pa_offset = KERNEL_LINK_ADDR - kernel_map.phys_addr; - - for ( ; rela < (Elf64_Rela *)&__rela_dyn_end; rela++) { - Elf64_Addr addr = (rela->r_offset - va_kernel_link_pa_offset); - Elf64_Addr relocated_addr = rela->r_addend; - - if (rela->r_info != R_RISCV_RELATIVE) - continue; - - /* - * Make sure to not relocate vdso symbols like rt_sigreturn - * which are linked from the address 0 in vmlinux since - * vdso symbol addresses are actually used as an offset from - * mm->context.vdso in VDSO_OFFSET macro. - */ - if (relocated_addr >= KERNEL_LINK_ADDR) - relocated_addr += reloc_offset; - - *(Elf64_Addr *)addr = relocated_addr; - } -} -#endif /* CONFIG_RELOCATABLE */ - #ifdef CONFIG_XIP_KERNEL static void __init create_kernel_page_table(pgd_t *pgdir, __always_unused bool early) @@ -1092,15 +1111,13 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) kernel_map.virt_addr = KERNEL_LINK_ADDR + kernel_map.virt_offset; #ifdef CONFIG_XIP_KERNEL -#ifdef CONFIG_64BIT - kernel_map.page_offset = PAGE_OFFSET_L3; -#else - kernel_map.page_offset = _AC(CONFIG_PAGE_OFFSET, UL); -#endif kernel_map.xiprom = (uintptr_t)CONFIG_XIP_PHYS_ADDR; kernel_map.xiprom_sz = (uintptr_t)(&_exiprom) - (uintptr_t)(&_xiprom); phys_ram_base = CONFIG_PHYS_RAM_BASE; +#ifdef CONFIG_SPARSEMEM_VMEMMAP + vmemmap_start_pfn = round_down(phys_ram_base, VMEMMAP_ADDR_ALIGN) >> PAGE_SHIFT; +#endif kernel_map.phys_addr = (uintptr_t)CONFIG_PHYS_RAM_BASE; kernel_map.size = (uintptr_t)(&_end) - (uintptr_t)(&_start); @@ -1108,7 +1125,6 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) kernel_map.va_kernel_xip_data_pa_offset = kernel_map.virt_addr - kernel_map.phys_addr + (uintptr_t)&_sdata - (uintptr_t)&_start; #else - kernel_map.page_offset = _AC(CONFIG_PAGE_OFFSET, UL); kernel_map.phys_addr = (uintptr_t)(&_start); kernel_map.size = (uintptr_t)(&_end) - kernel_map.phys_addr; kernel_map.va_kernel_pa_offset = kernel_map.virt_addr - kernel_map.phys_addr; @@ -1155,7 +1171,8 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) * makes the kernel cross over a PUD_SIZE boundary, raise a bug * since a part of the kernel would not get mapped. */ - BUG_ON(PUD_SIZE - (kernel_map.virt_addr & (PUD_SIZE - 1)) < kernel_map.size); + if (IS_ENABLED(CONFIG_64BIT)) + BUG_ON(PUD_SIZE - (kernel_map.virt_addr & (PUD_SIZE - 1)) < kernel_map.size); relocate_kernel(); #endif @@ -1359,6 +1376,12 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) { dtb_early_va = (void *)dtb_pa; dtb_early_pa = dtb_pa; + +#ifdef CONFIG_RELOCATABLE + kernel_map.virt_addr = (uintptr_t)_start; + kernel_map.phys_addr = (uintptr_t)_start; + relocate_kernel(); +#endif } static inline void setup_vm_final(void) @@ -1377,21 +1400,19 @@ static void __init arch_reserve_crashkernel(void) { unsigned long long low_size = 0; unsigned long long crash_base, crash_size; - char *cmdline = boot_command_line; bool high = false; int ret; if (!IS_ENABLED(CONFIG_CRASH_RESERVE)) return; - ret = parse_crashkernel(cmdline, memblock_phys_mem_size(), + ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(), &crash_size, &crash_base, &low_size, &high); if (ret) return; - reserve_crashkernel_generic(cmdline, crash_size, crash_base, - low_size, high); + reserve_crashkernel_generic(crash_size, crash_base, low_size, high); } void __init paging_init(void) @@ -1558,7 +1579,7 @@ static void __meminit free_pte_table(pte_t *pte_start, pmd_t *pmd) return; } - pagetable_pte_dtor(ptdesc); + pagetable_dtor(ptdesc); if (PageReserved(page)) free_reserved_page(page); else @@ -1580,7 +1601,7 @@ static void __meminit free_pmd_table(pmd_t *pmd_start, pud_t *pud, bool is_vmemm } if (!is_vmemmap) - pagetable_pmd_dtor(ptdesc); + pagetable_dtor(ptdesc); if (PageReserved(page)) free_reserved_page(page); else diff --git a/arch/riscv/mm/kasan_init.c b/arch/riscv/mm/kasan_init.c index c301c8d291d2..41c635d6aca4 100644 --- a/arch/riscv/mm/kasan_init.c +++ b/arch/riscv/mm/kasan_init.c @@ -32,7 +32,7 @@ static void __init kasan_populate_pte(pmd_t *pmd, unsigned long vaddr, unsigned pte_t *ptep, *p; if (pmd_none(pmdp_get(pmd))) { - p = memblock_alloc(PTRS_PER_PTE * sizeof(pte_t), PAGE_SIZE); + p = memblock_alloc_or_panic(PTRS_PER_PTE * sizeof(pte_t), PAGE_SIZE); set_pmd(pmd, pfn_pmd(PFN_DOWN(__pa(p)), PAGE_TABLE)); } @@ -54,7 +54,7 @@ static void __init kasan_populate_pmd(pud_t *pud, unsigned long vaddr, unsigned unsigned long next; if (pud_none(pudp_get(pud))) { - p = memblock_alloc(PTRS_PER_PMD * sizeof(pmd_t), PAGE_SIZE); + p = memblock_alloc_or_panic(PTRS_PER_PMD * sizeof(pmd_t), PAGE_SIZE); set_pud(pud, pfn_pud(PFN_DOWN(__pa(p)), PAGE_TABLE)); } @@ -85,7 +85,7 @@ static void __init kasan_populate_pud(p4d_t *p4d, unsigned long next; if (p4d_none(p4dp_get(p4d))) { - p = memblock_alloc(PTRS_PER_PUD * sizeof(pud_t), PAGE_SIZE); + p = memblock_alloc_or_panic(PTRS_PER_PUD * sizeof(pud_t), PAGE_SIZE); set_p4d(p4d, pfn_p4d(PFN_DOWN(__pa(p)), PAGE_TABLE)); } @@ -116,7 +116,7 @@ static void __init kasan_populate_p4d(pgd_t *pgd, unsigned long next; if (pgd_none(pgdp_get(pgd))) { - p = memblock_alloc(PTRS_PER_P4D * sizeof(p4d_t), PAGE_SIZE); + p = memblock_alloc_or_panic(PTRS_PER_P4D * sizeof(p4d_t), PAGE_SIZE); set_pgd(pgd, pfn_pgd(PFN_DOWN(__pa(p)), PAGE_TABLE)); } @@ -385,7 +385,7 @@ static void __init kasan_shallow_populate_pud(p4d_t *p4d, next = pud_addr_end(vaddr, end); if (pud_none(pudp_get(pud_k))) { - p = memblock_alloc(PAGE_SIZE, PAGE_SIZE); + p = memblock_alloc_or_panic(PAGE_SIZE, PAGE_SIZE); set_pud(pud_k, pfn_pud(PFN_DOWN(__pa(p)), PAGE_TABLE)); continue; } @@ -405,7 +405,7 @@ static void __init kasan_shallow_populate_p4d(pgd_t *pgd, next = p4d_addr_end(vaddr, end); if (p4d_none(p4dp_get(p4d_k))) { - p = memblock_alloc(PAGE_SIZE, PAGE_SIZE); + p = memblock_alloc_or_panic(PAGE_SIZE, PAGE_SIZE); set_p4d(p4d_k, pfn_p4d(PFN_DOWN(__pa(p)), PAGE_TABLE)); continue; } @@ -424,7 +424,7 @@ static void __init kasan_shallow_populate_pgd(unsigned long vaddr, unsigned long next = pgd_addr_end(vaddr, end); if (pgd_none(pgdp_get(pgd_k))) { - p = memblock_alloc(PAGE_SIZE, PAGE_SIZE); + p = memblock_alloc_or_panic(PAGE_SIZE, PAGE_SIZE); set_pgd(pgd_k, pfn_pgd(PFN_DOWN(__pa(p)), PAGE_TABLE)); continue; } diff --git a/arch/riscv/mm/pgtable.c b/arch/riscv/mm/pgtable.c index 4ae67324f992..8b6c0a112a8d 100644 --- a/arch/riscv/mm/pgtable.c +++ b/arch/riscv/mm/pgtable.c @@ -154,4 +154,14 @@ pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, flush_tlb_mm(vma->vm_mm); return pmd; } + +pud_t pudp_invalidate(struct vm_area_struct *vma, unsigned long address, + pud_t *pudp) +{ + VM_WARN_ON_ONCE(!pud_present(*pudp)); + pud_t old = pudp_establish(vma, address, pudp, pud_mkinvalid(*pudp)); + + flush_pud_tlb_range(vma, address, address + HPAGE_PUD_SIZE); + return old; +} #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ diff --git a/arch/riscv/mm/physaddr.c b/arch/riscv/mm/physaddr.c index 18706f457da7..559d291fac5c 100644 --- a/arch/riscv/mm/physaddr.c +++ b/arch/riscv/mm/physaddr.c @@ -12,7 +12,7 @@ phys_addr_t __virt_to_phys(unsigned long x) * Boundary checking aginst the kernel linear mapping space. */ WARN(!is_linear_mapping(x) && !is_kernel_mapping(x), - "virt_to_phys used for non-linear address: %pK (%pS)\n", + "virt_to_phys used for non-linear address: %p (%pS)\n", (void *)x, (void *)x); return __va_to_pa_nodebug(x); diff --git a/arch/riscv/mm/ptdump.c b/arch/riscv/mm/ptdump.c index 9d5f657a251b..32922550a50a 100644 --- a/arch/riscv/mm/ptdump.c +++ b/arch/riscv/mm/ptdump.c @@ -318,6 +318,38 @@ static void note_page(struct ptdump_state *pt_st, unsigned long addr, } } +static void note_page_pte(struct ptdump_state *pt_st, unsigned long addr, pte_t pte) +{ + note_page(pt_st, addr, 4, pte_val(pte)); +} + +static void note_page_pmd(struct ptdump_state *pt_st, unsigned long addr, pmd_t pmd) +{ + note_page(pt_st, addr, 3, pmd_val(pmd)); +} + +static void note_page_pud(struct ptdump_state *pt_st, unsigned long addr, pud_t pud) +{ + note_page(pt_st, addr, 2, pud_val(pud)); +} + +static void note_page_p4d(struct ptdump_state *pt_st, unsigned long addr, p4d_t p4d) +{ + note_page(pt_st, addr, 1, p4d_val(p4d)); +} + +static void note_page_pgd(struct ptdump_state *pt_st, unsigned long addr, pgd_t pgd) +{ + note_page(pt_st, addr, 0, pgd_val(pgd)); +} + +static void note_page_flush(struct ptdump_state *pt_st) +{ + pte_t pte_zero = {0}; + + note_page(pt_st, 0, -1, pte_val(pte_zero)); +} + static void ptdump_walk(struct seq_file *s, struct ptd_mm_info *pinfo) { struct pg_state st = { @@ -325,7 +357,12 @@ static void ptdump_walk(struct seq_file *s, struct ptd_mm_info *pinfo) .marker = pinfo->markers, .level = -1, .ptdump = { - .note_page = note_page, + .note_page_pte = note_page_pte, + .note_page_pmd = note_page_pmd, + .note_page_pud = note_page_pud, + .note_page_p4d = note_page_p4d, + .note_page_pgd = note_page_pgd, + .note_page_flush = note_page_flush, .range = (struct ptdump_range[]) { {pinfo->base_addr, pinfo->end}, {0, 0} @@ -347,7 +384,12 @@ bool ptdump_check_wx(void) .level = -1, .check_wx = true, .ptdump = { - .note_page = note_page, + .note_page_pte = note_page_pte, + .note_page_pmd = note_page_pmd, + .note_page_pud = note_page_pud, + .note_page_p4d = note_page_p4d, + .note_page_pgd = note_page_pgd, + .note_page_flush = note_page_flush, .range = (struct ptdump_range[]) { {KERN_VIRT_START, ULONG_MAX}, {0, 0} diff --git a/arch/riscv/mm/tlbflush.c b/arch/riscv/mm/tlbflush.c index 9b6e86ce3867..e737ba7949b1 100644 --- a/arch/riscv/mm/tlbflush.c +++ b/arch/riscv/mm/tlbflush.c @@ -4,8 +4,30 @@ #include <linux/smp.h> #include <linux/sched.h> #include <linux/hugetlb.h> +#include <linux/mmu_notifier.h> #include <asm/sbi.h> #include <asm/mmu_context.h> +#include <asm/cpufeature.h> + +#define has_svinval() riscv_has_extension_unlikely(RISCV_ISA_EXT_SVINVAL) + +static inline void local_sfence_inval_ir(void) +{ + asm volatile(SFENCE_INVAL_IR() ::: "memory"); +} + +static inline void local_sfence_w_inval(void) +{ + asm volatile(SFENCE_W_INVAL() ::: "memory"); +} + +static inline void local_sinval_vma(unsigned long vma, unsigned long asid) +{ + if (asid != FLUSH_TLB_NO_ASID) + asm volatile(SINVAL_VMA(%0, %1) : : "r" (vma), "r" (asid) : "memory"); + else + asm volatile(SINVAL_VMA(%0, zero) : : "r" (vma) : "memory"); +} /* * Flush entire TLB if number of entries to be flushed is greater @@ -26,6 +48,16 @@ static void local_flush_tlb_range_threshold_asid(unsigned long start, return; } + if (has_svinval()) { + local_sfence_w_inval(); + for (i = 0; i < nr_ptes_in_range; ++i) { + local_sinval_vma(start, asid); + start += stride; + } + local_sfence_inval_ir(); + return; + } + for (i = 0; i < nr_ptes_in_range; ++i) { local_flush_tlb_page_asid(start, asid); start += stride; @@ -78,10 +110,17 @@ static void __ipi_flush_tlb_range_asid(void *info) local_flush_tlb_range_asid(d->start, d->size, d->stride, d->asid); } -static void __flush_tlb_range(const struct cpumask *cmask, unsigned long asid, +static inline unsigned long get_mm_asid(struct mm_struct *mm) +{ + return mm ? cntx2asid(atomic_long_read(&mm->context.id)) : FLUSH_TLB_NO_ASID; +} + +static void __flush_tlb_range(struct mm_struct *mm, + const struct cpumask *cmask, unsigned long start, unsigned long size, unsigned long stride) { + unsigned long asid = get_mm_asid(mm); unsigned int cpu; if (cpumask_empty(cmask)) @@ -105,30 +144,26 @@ static void __flush_tlb_range(const struct cpumask *cmask, unsigned long asid, } put_cpu(); -} -static inline unsigned long get_mm_asid(struct mm_struct *mm) -{ - return cntx2asid(atomic_long_read(&mm->context.id)); + if (mm) + mmu_notifier_arch_invalidate_secondary_tlbs(mm, start, start + size); } void flush_tlb_mm(struct mm_struct *mm) { - __flush_tlb_range(mm_cpumask(mm), get_mm_asid(mm), - 0, FLUSH_TLB_MAX_SIZE, PAGE_SIZE); + __flush_tlb_range(mm, mm_cpumask(mm), 0, FLUSH_TLB_MAX_SIZE, PAGE_SIZE); } void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned int page_size) { - __flush_tlb_range(mm_cpumask(mm), get_mm_asid(mm), - start, end - start, page_size); + __flush_tlb_range(mm, mm_cpumask(mm), start, end - start, page_size); } void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr) { - __flush_tlb_range(mm_cpumask(vma->vm_mm), get_mm_asid(vma->vm_mm), + __flush_tlb_range(vma->vm_mm, mm_cpumask(vma->vm_mm), addr, PAGE_SIZE, PAGE_SIZE); } @@ -161,13 +196,13 @@ void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, } } - __flush_tlb_range(mm_cpumask(vma->vm_mm), get_mm_asid(vma->vm_mm), + __flush_tlb_range(vma->vm_mm, mm_cpumask(vma->vm_mm), start, end - start, stride_size); } void flush_tlb_kernel_range(unsigned long start, unsigned long end) { - __flush_tlb_range(cpu_online_mask, FLUSH_TLB_NO_ASID, + __flush_tlb_range(NULL, cpu_online_mask, start, end - start, PAGE_SIZE); } @@ -175,9 +210,16 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end) void flush_pmd_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) { - __flush_tlb_range(mm_cpumask(vma->vm_mm), get_mm_asid(vma->vm_mm), + __flush_tlb_range(vma->vm_mm, mm_cpumask(vma->vm_mm), start, end - start, PMD_SIZE); } + +void flush_pud_tlb_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + __flush_tlb_range(vma->vm_mm, mm_cpumask(vma->vm_mm), + start, end - start, PUD_SIZE); +} #endif bool arch_tlbbatch_should_defer(struct mm_struct *mm) @@ -186,10 +228,10 @@ bool arch_tlbbatch_should_defer(struct mm_struct *mm) } void arch_tlbbatch_add_pending(struct arch_tlbflush_unmap_batch *batch, - struct mm_struct *mm, - unsigned long uaddr) + struct mm_struct *mm, unsigned long start, unsigned long end) { cpumask_or(&batch->cpumask, &batch->cpumask, mm_cpumask(mm)); + mmu_notifier_arch_invalidate_secondary_tlbs(mm, start, end); } void arch_flush_tlb_batched_pending(struct mm_struct *mm) @@ -199,7 +241,7 @@ void arch_flush_tlb_batched_pending(struct mm_struct *mm) void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch) { - __flush_tlb_range(&batch->cpumask, FLUSH_TLB_NO_ASID, 0, - FLUSH_TLB_MAX_SIZE, PAGE_SIZE); + __flush_tlb_range(NULL, &batch->cpumask, + 0, FLUSH_TLB_MAX_SIZE, PAGE_SIZE); cpumask_clear(&batch->cpumask); } |