Currently invalidate_inode_pages2_range() and invalidate_mapping_pages() just delete all exceptional radix tree entries they find. For DAX this is not desirable as we track cache dirtiness in these entries and when they are evicted, we may not flush caches although it is necessary. This can for example manifest when we write to the same block both via mmap and via write(2) (to different offsets) and fsync(2) then does not properly flush CPU caches when modification via write(2) was the last one. Create appropriate DAX functions to handle invalidation of DAX entries for invalidate_inode_pages2_range() and invalidate_mapping_pages() and wire them up into the corresponding mm functions.
Showing the changes to include/linux (1 of 3 files). Show all
Unified · Split
| @@ -41,6 +41,9 @@ ssize_t dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter, · 40 unmodified lines | ||
| 41 | 41 | int dax_iomap_fault(struct vm_area_struct *vma, struct vm_fault *vmf, |
| 42 | 42 | struct iomap_ops *ops); |
| 43 | 43 | int dax_delete_mapping_entry(struct address_space *mapping, pgoff_t index); |
| 44 | +int dax_invalidate_mapping_entry(struct address_space *mapping, pgoff_t index); | |
| 45 | +int dax_invalidate_mapping_entry_sync(struct address_space *mapping, | |
| 46 | + pgoff_t index); | |
| 44 | 47 | void dax_wake_mapping_entry_waiter(struct address_space *mapping, |
| 45 | 48 | pgoff_t index, void *entry, bool wake_all); |
| 46 | 49 | |