From mboxrd@z Thu Jan 1 00:00:00 1970 From: Killian De Volder Subject: Re: block size question Date: Wed, 18 May 2016 09:11:25 +0200 Message-ID: <573C159D.20806@megasoft.be> References: <573C02B3.5010501@net.home> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: Received: from relay5-d.mail.gandi.net ([217.70.183.197]:53296 "EHLO relay5-d.mail.gandi.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750925AbcERHLc (ORCPT ); Wed, 18 May 2016 03:11:32 -0400 In-Reply-To: <573C02B3.5010501@net.home> Sender: linux-bcache-owner@vger.kernel.org List-Id: linux-bcache@vger.kernel.org To: Vlad , linux-bcache@vger.kernel.org The blocksize should be equal or bigger then the blocking device. Otherwise you risk writing multiple blocks 512 for 1 1k sector, resulting in read-modify-write transaction on 4k disks. But for some reason I didn't do this on my desktop. (I might need to reformat if i'm right about this :( ) Killian De Volder On 18-05-16 07:50, Vlad wrote: > Hallo, > > What would be the optimal block size in case if physical block sizes > different for cache and backing devices? Can I use different block sizes > for cache and backing devices, like: > make-bcache --block 4k -B /dev/sdc > make-bcache --block 512 -C /dev/sda > > I have a feeling that the block size should be the same and the biggest > block size should be used, but I'd appreciate any thoughts about that. > > Thanks in advance, > Vlad. > -- > To unsubscribe from this list: send the line "unsubscribe linux-bcache" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >