From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bombadil.infradead.org ([198.137.202.133]:59458 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727453AbfEJRbK (ORCPT ); Fri, 10 May 2019 13:31:10 -0400 Date: Fri, 10 May 2019 10:31:10 -0700 From: Christoph Hellwig Subject: Re: [PATCH 3/6] xfs: use locality optimized cntbt lookups for near mode allocations Message-ID: <20190510173110.GC18992@infradead.org> References: <20190509165839.44329-1-bfoster@redhat.com> <20190509165839.44329-4-bfoster@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190509165839.44329-4-bfoster@redhat.com> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Brian Foster Cc: linux-xfs@vger.kernel.org Still digesting the algorithmic changes, but a few nitpicks below: > /* > + * BLock allocation algorithm and data structures. I think the upper case L is a typo. > +struct xfs_alloc_btree_cur { > + struct xfs_btree_cur *cur; /* cursor */ > + bool active; /* cursor active flag */ > +}; Can't we move the active flag inside the btree_cur, more specically into enum xfs_btree_cur_private? Or maybe we should byte the bullet and make xfs_btree_cur a structure embedded into the type specific structures and use container_of. But I certainly don't want to burden that on you and this series..