From: Daeho Jeong Subject: RE: Re: [PATCH] ext4: hand over jobs handling discard commands on commit complete phase to kworkers Date: Wed, 17 May 2017 00:45:29 +0000 Message-ID: <20170517004529epcms1p297077c5039bda981024d5a10db5f676e@epcms1p2> References: <20170516115306.GA15058@infradead.org> <1494920262-10128-1-git-send-email-daeho.jeong@samsung.com> Reply-To: daeho.jeong@samsung.com Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8BIT Cc: "jack@suse.com" , "tytso@mit.edu" , "linux-ext4@vger.kernel.org" , "linux-block@vger.kernel.org" To: Christoph Hellwig , Daeho Jeong Return-path: In-Reply-To: <20170516115306.GA15058@infradead.org> Sender: linux-block-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org >> Now, when we mount ext4 filesystem with '-o discard' option, we have to >> issue all the discard commands for the blocks to be deallocated and >> wait for the completion of the commands on the commit complete phase. > No, you don't. You can execute them asynchronously, similar to what > XFS already does. Thank you, I will check XFS codes issuing discard commands asynchronously. But, even if the discard commands can be handled in parallel by doing this, we still need to wait for all the discard requests completion on the journaling commit phase. I want to modify this part, too.