From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from vps.thesusis.net (vps.thesusis.net [34.202.238.73]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C9BBF2CA4 for ; Mon, 25 Oct 2021 19:02:33 +0000 (UTC) Received: by vps.thesusis.net (Postfix, from userid 1000) id 0C83861FD2; Mon, 25 Oct 2021 15:02:27 -0400 (EDT) References: <87tuh9n9w2.fsf@vps.thesusis.net> <20211022084127.GA1026@quack2.suse.cz> User-agent: mu4e 1.7.0; emacs 27.1 From: Phillip Susi To: Jan Kara Cc: Matthew Wilcox , linux-fsdevel@vger.kernel.org, Phillip Lougher , linux-erofs@lists.ozlabs.org, linux-btrfs@vger.kernel.org, linux-ntfs-dev@lists.sourceforge.net, ntfs3@lists.linux.dev, linux-bcache@vger.kernel.org, David Howells , Hsin-Yi Wang Subject: Re: Readahead for compressed data Date: Mon, 25 Oct 2021 14:59:45 -0400 In-reply-to: <20211022084127.GA1026@quack2.suse.cz> Message-ID: <87fssprkql.fsf@vps.thesusis.net> Precedence: bulk X-Mailing-List: ntfs3@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain Jan Kara writes: > Well, one of the problems with keeping compressed data is that for mmap(2) > you have to have pages decompressed so that CPU can access them. So keeping > compressed data in the page cache would add a bunch of complexity. That > being said keeping compressed data cached somewhere else than in the page > cache may certainly me worth it and then just filling page cache on demand > from this data... True... Did that multi generational LRU cache ever get merged? I was thinking you could use that to make sure that the kernel prefers to reclaim the decompressed pages in favor of keeping the compressed ones around.