On Tue, Sep 22, 2020 at 06:05:26PM +0100, Matthew Wilcox wrote: > On Tue, Sep 22, 2020 at 12:23:45PM -0400, Qian Cai wrote: > > On Fri, 2020-09-11 at 00:47 +0100, Matthew Wilcox (Oracle) wrote: > > > Size the uptodate array dynamically to support larger pages in the > > > page cache. With a 64kB page, we're only saving 8 bytes per page today, > > > but with a 2MB maximum page size, we'd have to allocate more than 4kB > > > per page. Add a few debugging assertions. > > > > > > Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx> > > > Reviewed-by: Dave Chinner <dchinner@xxxxxxxxxx> > > > > Some syscall fuzzing will trigger this on powerpc: > > > > .config: https://gitlab.com/cailca/linux-mm/-/blob/master/powerpc.config > > > > [ 8805.895344][T445431] WARNING: CPU: 61 PID: 445431 at fs/iomap/buffered-io.c:78 iomap_page_release+0x250/0x270 > > Well, I'm glad it triggered. That warning is: > WARN_ON_ONCE(bitmap_full(iop->uptodate, nr_blocks) != > PageUptodate(page)); > so there was definitely a problem of some kind. OK, I'm pretty sure the problem predated this commit, and it's simply that I added a warning (looking for something else) that caught this. I have a tree completly gunked up with debugging code now to try to understand the problem better, but my guess is that if you put this warning into a previous version, you'd see the same problem occurring (and it is a real problem, because we skip writeback of parts of the page which are !uptodate).
