Re: Major btrfs fiemap slowdown on file with many extents once in cache (RCU stalls?) (Was: [PATCH 1/3] filemap: Correct the conditions for marking a folio as accessed)

[Dominique MARTINET]

Theodore Ts'o wrote on Thu, Jun 30, 2022 at 10:37:58AM -0400:
> On Thu, Jun 30, 2022 at 04:29:05PM +0900, Dominique MARTINET wrote:
> > Hi Willy, linux-btrfs@vger,
> > 
> > Matthew Wilcox (Oracle) wrote on Sun, Jun 19, 2022 at 04:11:41PM +0100:
> > > We had an off-by-one error which meant that we never marked the first page
> > > in a read as accessed.  This was visible as a slowdown when re-reading
> > > a file as pages were being evicted from cache too soon.  In reviewing
> > > this code, we noticed a second bug where a multi-page folio would be
> > > marked as accessed multiple times when doing reads that were less than
> > > the size of the folio.
> > 
> > when debugging an unrelated issue (short reads on btrfs with io_uring
> > and O_DIRECT[1]), I noticed that my horrible big file copy speeds fell
> > down from ~2GB/s (there's compression and lots of zeroes) to ~100MB/s
> > the second time I was copying it with cp.
> > 
> > I've taken a moment to bisect this and came down to this patch.
> 
> I think you may have forgotten to include the commit-id that was the
> results of your bisect.... ?

Sorry, this is the patch I replied to and it was recent enough that I
assumed it'd still be in mailboxes, but you're right it's better with a
commit id. This is was merged as 5ccc944dce3d ("filemap: Correct the
conditions for marking a folio as accessed")


Thanks,
-- 
Dominique