Every arch has the same code to ensure atomic operations and a check for !HIGHMEM page. Remove the duplicate code by defining a core kmap_atomic() which only calls the arch specific kmap_atomic_high() when the page is high memory. [akpm@linux-foundation.org: coding style fixes] Signed-off-by: Ira Weiny <ira.weiny@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Andy Lutomirski <luto@kernel.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Christian König <christian.koenig@amd.com> Cc: Chris Zankel <chris@zankel.net> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Helge Deller <deller@gmx.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/20200507150004.1423069-7-ira.weiny@intel.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Showing the changes to include/linux (1 of 21 files). Show all
Unified · Split
| @@ -32,6 +32,7 @@ static inline void invalidate_kernel_vmap_range(void *vaddr, int size) · 31 unmodified lines | ||
| 32 | 32 | #include <asm/kmap_types.h> |
| 33 | 33 | |
| 34 | 34 | #ifdef CONFIG_HIGHMEM |
| 35 | +extern void *kmap_atomic_high(struct page *page); | |
| 35 | 36 | #include <asm/highmem.h> |
| 36 | 37 | |
| 37 | 38 | #ifndef ARCH_HAS_KMAP_FLUSH_TLB |
| @@ -62,6 +63,28 @@ static inline void kunmap(struct page *page) · 24 unmodified lines | ||
| 62 | 63 | kunmap_high(page); |
| 63 | 64 | } |
| 64 | 65 | |
| 66 | +/* | |
| 67 | + * kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because | |
| 68 | + * no global lock is needed and because the kmap code must perform a global TLB | |
| 69 | + * invalidation when the kmap pool wraps. | |
| 70 | + * | |
| 71 | + * However when holding an atomic kmap is is not legal to sleep, so atomic | |
| 72 | + * kmaps are appropriate for short, tight code paths only. | |
| 73 | + * | |
| 74 | + * The use of kmap_atomic/kunmap_atomic is discouraged - kmap/kunmap | |
| 75 | + * gives a more generic (and caching) interface. But kmap_atomic can | |
| 76 | + * be used in IRQ contexts, so in some (very limited) cases we need | |
| 77 | + * it. | |
| 78 | + */ | |
| 79 | +static inline void *kmap_atomic(struct page *page) | |
| 80 | +{ | |
| 81 | + preempt_disable(); | |
| 82 | + pagefault_disable(); | |
| 83 | + if (!PageHighMem(page)) | |
| 84 | + return page_address(page); | |
| 85 | + return kmap_atomic_high(page); | |
| 86 | +} | |
| 87 | + | |
| 65 | 88 | /* declarations for linux/mm/highmem.c */ |
| 66 | 89 | unsigned int nr_free_highpages(void); |
| 67 | 90 | extern atomic_long_t _totalhigh_pages; |