On Fri, 26 Feb 2021 14:00:11 +0000 "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx> wrote: > After splitting generic_file_buffered_read() into smaller parts, it > turns out we can reuse one of the parts in filemap_fault(). This fixes > an oversight -- waiting for the I/O to complete is now interruptible > by a fatal signal. And it saves us a few bytes of text in an unlikely > path. We also handle AOP_TRUNCATED_PAGE which the present code fails to do. Should this be in the changelog? Did we handle AOP_TRUNCATED_PAGE in the pre-splitup code, or is this new?
