From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Mason Subject: Re: [RFC v2 2/2] fs: btrfs: Extends btrfs/raid56 to support up to six parities Date: Mon, 6 Jan 2014 14:12:56 +0000 Message-ID: <1389017609.23310.0.camel@ret.masoncoding.com> References: <1389000716-3274-1-git-send-email-amadvance@gmail.com> <1389000716-3274-3-git-send-email-amadvance@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-7" Content-Transfer-Encoding: 8BIT Return-path: In-Reply-To: <1389000716-3274-3-git-send-email-amadvance@gmail.com> Content-Language: en-US Content-ID: <03D3FBCD996DF84691383E1810360F94@fb.com> Sender: linux-kernel-owner@vger.kernel.org To: "amadvance@gmail.com" Cc: Josef Bacik , "neilb@suse.de" , "linux-kernel@vger.kernel.org" , "linux-raid@vger.kernel.org" , "linux-btrfs@vger.kernel.org" List-Id: linux-raid.ids On Mon, 2014-01-06 at 10:31 +-0100, Andrea Mazzoleni wrote: +AD4- This patch changes btrfs/raid56.c to use the new raid interface and +AD4- extends its support to an arbitrary number of parities. +AD4- +AD4- More in details, the two faila/failb failure indexes are now replaced +AD4- with a fail+AFsAXQ- vector that keeps track of up to six failures, and now +AD4- the new raid+AF8-par() and raid+AF8-rec() functions are used to handle with +AD4- parity instead of the old xor/raid6 ones. +AD4- Neat. The faila/failb were always my least favorite part of the btrfs code +ADs-) Did you test just raid5/6 or also the higher parity counts? -chris