From: Nicholas Piggin <npiggin@xxxxxxxxx> Subject: mm/vmalloc: remove map_kernel_range Patch series "mm/vmalloc: cleanup after hugepage series", v2. Christoph pointed out some overdue cleanups required after the huge vmalloc series, and I had another failure error message improvement as well. This patch (of 5): This is a shim around vmap_pages_range, get rid of it. Move the main API comment from the _noflush variant to the normal variant, and make _noflush internal to mm/. Link: https://lkml.kernel.org/r/20210322021806.892164-1-npiggin@xxxxxxxxx Link: https://lkml.kernel.org/r/20210322021806.892164-2-npiggin@xxxxxxxxx Signed-off-by: Nicholas Piggin <npiggin@xxxxxxxxx> Reviewed-by: Christoph Hellwig <hch@xxxxxx> Cc: Uladzislau Rezki <urezki@xxxxxxxxx> Cc: Cédric Le Goater <clg@xxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- Documentation/core-api/cachetlb.rst | 2 include/linux/vmalloc.h | 11 ---- mm/internal.h | 6 ++ mm/percpu-vm.c | 5 +- mm/vmalloc.c | 65 +++++++++++--------------- 5 files changed, 38 insertions(+), 51 deletions(-) --- a/Documentation/core-api/cachetlb.rst~mm-vmalloc-remove-map_kernel_range +++ a/Documentation/core-api/cachetlb.rst @@ -213,7 +213,7 @@ Here are the routines, one by one: there will be no entries in the cache for the kernel address space for virtual addresses in the range 'start' to 'end-1'. - The first of these two routines is invoked after map_kernel_range() + The first of these two routines is invoked after vmap_range() has installed the page table entries. The second is invoked before unmap_kernel_range() deletes the page table entries. --- a/include/linux/vmalloc.h~mm-vmalloc-remove-map_kernel_range +++ a/include/linux/vmalloc.h @@ -212,10 +212,6 @@ static inline bool is_vm_area_hugepages( int vmap_range(unsigned long addr, unsigned long end, phys_addr_t phys_addr, pgprot_t prot, unsigned int max_page_shift); -extern int map_kernel_range_noflush(unsigned long start, unsigned long size, - pgprot_t prot, struct page **pages); -int map_kernel_range(unsigned long start, unsigned long size, pgprot_t prot, - struct page **pages); extern void unmap_kernel_range_noflush(unsigned long addr, unsigned long size); extern void unmap_kernel_range(unsigned long addr, unsigned long size); static inline void set_vm_flush_reset_perms(void *addr) @@ -227,13 +223,6 @@ static inline void set_vm_flush_reset_pe } #else -static inline int -map_kernel_range_noflush(unsigned long start, unsigned long size, - pgprot_t prot, struct page **pages) -{ - return size >> PAGE_SHIFT; -} -#define map_kernel_range map_kernel_range_noflush static inline void unmap_kernel_range_noflush(unsigned long addr, unsigned long size) { --- a/mm/internal.h~mm-vmalloc-remove-map_kernel_range +++ a/mm/internal.h @@ -637,4 +637,10 @@ struct migration_target_control { gfp_t gfp_mask; }; +/* + * mm/vmalloc.c + */ +int vmap_pages_range_noflush(unsigned long addr, unsigned long end, + pgprot_t prot, struct page **pages, unsigned int page_shift); + #endif /* __MM_INTERNAL_H */ --- a/mm/percpu-vm.c~mm-vmalloc-remove-map_kernel_range +++ a/mm/percpu-vm.c @@ -8,6 +8,7 @@ * Chunks are mapped into vmalloc areas and populated page by page. * This is the default chunk allocator. */ +#include "internal.h" static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk, unsigned int cpu, int page_idx) @@ -192,8 +193,8 @@ static void pcpu_post_unmap_tlb_flush(st static int __pcpu_map_pages(unsigned long addr, struct page **pages, int nr_pages) { - return map_kernel_range_noflush(addr, nr_pages << PAGE_SHIFT, - PAGE_KERNEL, pages); + return vmap_pages_range_noflush(addr, addr + (nr_pages << PAGE_SHIFT), + PAGE_KERNEL, pages, PAGE_SHIFT); } /** --- a/mm/vmalloc.c~mm-vmalloc-remove-map_kernel_range +++ a/mm/vmalloc.c @@ -523,7 +523,16 @@ static int vmap_small_pages_range_noflus return 0; } -static int vmap_pages_range_noflush(unsigned long addr, unsigned long end, +/* + * vmap_pages_range_noflush is similar to vmap_pages_range, but does not + * flush caches. + * + * The caller is responsible for calling flush_cache_vmap() after this + * function returns successfully and before the addresses are accessed. + * + * This is an internal function only. Do not use outside mm/. + */ +int vmap_pages_range_noflush(unsigned long addr, unsigned long end, pgprot_t prot, struct page **pages, unsigned int page_shift) { unsigned int i, nr = (end - addr) >> PAGE_SHIFT; @@ -549,48 +558,26 @@ static int vmap_pages_range_noflush(unsi return 0; } -static int vmap_pages_range(unsigned long addr, unsigned long end, - pgprot_t prot, struct page **pages, unsigned int page_shift) -{ - int err; - - err = vmap_pages_range_noflush(addr, end, prot, pages, page_shift); - flush_cache_vmap(addr, end); - return err; -} - /** - * map_kernel_range_noflush - map kernel VM area with the specified pages + * vmap_pages_range - map pages to a kernel virtual address * @addr: start of the VM area to map - * @size: size of the VM area to map + * @end: end of the VM area to map (non-inclusive) * @prot: page protection flags to use - * @pages: pages to map - * - * Map PFN_UP(@size) pages at @addr. The VM area @addr and @size specify should - * have been allocated using get_vm_area() and its friends. - * - * NOTE: - * This function does NOT do any cache flushing. The caller is responsible for - * calling flush_cache_vmap() on to-be-mapped areas before calling this - * function. + * @pages: pages to map (always PAGE_SIZE pages) + * @page_shift: maximum shift that the pages may be mapped with, @pages must + * be aligned and contiguous up to at least this shift. * * RETURNS: * 0 on success, -errno on failure. */ -int map_kernel_range_noflush(unsigned long addr, unsigned long size, - pgprot_t prot, struct page **pages) -{ - return vmap_pages_range_noflush(addr, addr + size, prot, pages, PAGE_SHIFT); -} - -int map_kernel_range(unsigned long start, unsigned long size, pgprot_t prot, - struct page **pages) +static int vmap_pages_range(unsigned long addr, unsigned long end, + pgprot_t prot, struct page **pages, unsigned int page_shift) { - int ret; + int err; - ret = map_kernel_range_noflush(start, size, prot, pages); - flush_cache_vmap(start, start + size); - return ret; + err = vmap_pages_range_noflush(addr, end, prot, pages, page_shift); + flush_cache_vmap(addr, end); + return err; } int is_vmalloc_or_module_addr(const void *x) @@ -2156,10 +2143,12 @@ void *vm_map_ram(struct page **pages, un kasan_unpoison_vmalloc(mem, size); - if (map_kernel_range(addr, size, PAGE_KERNEL, pages) < 0) { + if (vmap_pages_range(addr, addr + size, PAGE_KERNEL, + pages, PAGE_SHIFT) < 0) { vm_unmap_ram(mem, count); return NULL; } + return mem; } EXPORT_SYMBOL(vm_map_ram); @@ -2703,6 +2692,7 @@ void *vmap(struct page **pages, unsigned unsigned long flags, pgprot_t prot) { struct vm_struct *area; + unsigned long addr; unsigned long size; /* In bytes */ might_sleep(); @@ -2715,8 +2705,9 @@ void *vmap(struct page **pages, unsigned if (!area) return NULL; - if (map_kernel_range((unsigned long)area->addr, size, pgprot_nx(prot), - pages) < 0) { + addr = (unsigned long)area->addr; + if (vmap_pages_range(addr, addr + size, pgprot_nx(prot), + pages, PAGE_SHIFT) < 0) { vunmap(area->addr); return NULL; } _
