[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200622181338.GA21350@casper.infradead.org>
Date: Mon, 22 Jun 2020 19:13:38 +0100
From: Matthew Wilcox <willy@...radead.org>
To: Andreas Gruenbacher <agruenba@...hat.com>
Cc: Dave Chinner <david@...morbit.com>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>,
Linux-MM <linux-mm@...ck.org>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [RFC] Bypass filesystems for reading cached pages
On Mon, Jun 22, 2020 at 04:35:05PM +0200, Andreas Gruenbacher wrote:
> I'm fine with not moving that functionality into the VFS. The problem
> I have in gfs2 is that taking glocks is really expensive. Part of that
> overhead is accidental, but we definitely won't be able to fix it in
> the short term. So something like the IOCB_CACHED flag that prevents
> generic_file_read_iter from issuing readahead I/O would save the day
> for us. Does that idea stand a chance?
For the short-term fix, is switching to a trylock in gfs2_readahead()
acceptable?
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 72c9560f4467..6ccd478c81ff 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -600,7 +600,7 @@ static void gfs2_readahead(struct readahead_control *rac)
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_holder gh;
- gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+ gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_TRY, &gh);
if (gfs2_glock_nq(&gh))
goto out_uninit;
if (!gfs2_is_stuffed(ip))
Powered by blists - more mailing lists