Hi Matthew, On Fri, Mar 20, 2020 at 07:29:32AM -0700, Matthew Wilcox wrote: > > I'm a little disappointed to have received no feedback on these patches > from those who were involved with creating the ugly situation in the > first place. > > There are other ext4/f2fs patches in this patch series for which it > would also be nice to get reviewed-by tags. > > ----- Forwarded message from Matthew Wilcox <willy@xxxxxxxxxxxxx> ----- > > Date: Fri, 20 Mar 2020 07:22:19 -0700 > From: Matthew Wilcox <willy@xxxxxxxxxxxxx> > To: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> > Cc: "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx>, > linux-fsdevel@xxxxxxxxxxxxxxx, linux-mm@xxxxxxxxx, > linux-kernel@xxxxxxxxxxxxxxx, linux-btrfs@xxxxxxxxxxxxxxx, > linux-erofs@xxxxxxxxxxxxxxxx, linux-ext4@xxxxxxxxxxxxxxx, > linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx, cluster-devel@xxxxxxxxxx, > ocfs2-devel@xxxxxxxxxxxxxx, linux-xfs@xxxxxxxxxxxxxxx, Christoph > Hellwig <hch@xxxxxx>, William Kucharski <william.kucharski@xxxxxxxxxx> > Subject: [PATCH v9 13/25] mm: Add page_cache_readahead_unbounded > X-Mailer: git-send-email 2.21.1 > > From: "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx> > > ext4 and f2fs have duplicated the guts of the readahead code so > they can read past i_size. Instead, separate out the guts of the > readahead code so they can call it directly. > > Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx> > Reviewed-by: Christoph Hellwig <hch@xxxxxx> > Reviewed-by: William Kucharski <william.kucharski@xxxxxxxxxx> I actually did look at this patch last month and it looked fine. Thanks for doing this. To be clear, we'd be fine with maintaining filesystem-specific code that does the setup necessary to call ext4_mpage_readpages() and f2fs_mpage_readpages() properly. And that was the original intent, so that nothing would leak into the VFS. But I understand that the new readahead calling convention is different and a bit more complex, and it's annoying to have multiple callers. So if page_cache_readahead_unbounded() makes things easier for you, it's fine with me. I'll take a closer look at your v9 series today and provide Reviewed-by on this patch and others. Sorry for not doing so earlier. Thanks! - Eric
