From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp1040.oracle.com ([141.146.126.69]:45692 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754284AbcHARyU (ORCPT ); Mon, 1 Aug 2016 13:54:20 -0400 Date: Mon, 1 Aug 2016 10:54:07 -0700 From: "Darrick J. Wong" To: Brian Foster Cc: linux-fsdevel@vger.kernel.org, vishal.l.verma@intel.com, xfs@oss.sgi.com Subject: Re: [PATCH v2 07/47] xfs: add function pointers for get/update keys to the btree Message-ID: <20160801175407.GA8593@birch.djwong.org> References: <146907695530.25461.3225785294902719773.stgit@birch.djwong.org> <146907700604.25461.2181974283557088355.stgit@birch.djwong.org> <20160728194615.GA15746@birch.djwong.org> <20160801155736.GA64218@bfoster.bfoster> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160801155736.GA64218@bfoster.bfoster> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Mon, Aug 01, 2016 at 11:57:36AM -0400, Brian Foster wrote: > On Thu, Jul 28, 2016 at 12:46:15PM -0700, Darrick J. Wong wrote: > > Add some function pointers to bc_ops to get the btree keys for > > leaf and node blocks, and to update parent keys of a block. > > Convert the _btree_updkey calls to use our new pointer, and > > modify the tree shape changing code to call the appropriate > > get_*_keys pointer instead of _btree_copy_keys because the > > overlapping btree has to calculate high key values. > > > > Signed-off-by: Darrick J. Wong > > --- > > fs/xfs/libxfs/xfs_alloc_btree.c | 4 + > > fs/xfs/libxfs/xfs_bmap_btree.c | 4 + > > fs/xfs/libxfs/xfs_btree.c | 164 +++++++++++++++++++++++--------------- > > fs/xfs/libxfs/xfs_btree.h | 19 ++++ > > fs/xfs/libxfs/xfs_ialloc_btree.c | 8 ++ > > 5 files changed, 135 insertions(+), 64 deletions(-) > > > ... > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c > > index 8d8e362..405442d 100644 > > --- a/fs/xfs/libxfs/xfs_btree.c > > +++ b/fs/xfs/libxfs/xfs_btree.c > ... > > @@ -2146,11 +2194,10 @@ xfs_btree_lshift( > > */ > > cur->bc_ops->init_key_from_rec(&key, > > xfs_btree_rec_addr(cur, 1, right)); > > - rkp = &key; > > Sorry, I should have been more clear. I think this entire hunk (the > above key initialization and comment) is effectively dead. > > > } > > > > - /* Update the parent key values of right. */ > > - error = xfs_btree_updkey(cur, rkp, level + 1); > > + /* Update the parent keys of the right block. */ > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > > > @@ -2292,7 +2339,6 @@ xfs_btree_rshift( > > xfs_btree_log_recs(cur, rbp, 1, rrecs + 1); > > > > cur->bc_ops->init_key_from_rec(&key, rrp); > > - rkp = &key; > > Same here. key isn't used beyond this. > > I see that hch has some objections to the design here. FWIW, resolution > of that aside, and with the above fixes: Ah, right. "key" and its initializers can go away in both functions. I'll fix that and find out if Dave wants to take this (and the other fixes) for a 4.8 merge request, or just defer them until -rc2. --D > Reviewed-by: Brian Foster > > > > > ASSERT(cur->bc_ops->recs_inorder(cur, rrp, > > xfs_btree_rec_addr(cur, 2, right))); > > @@ -2321,7 +2367,8 @@ xfs_btree_rshift( > > if (error) > > goto error1; > > > > - error = xfs_btree_updkey(tcur, rkp, level + 1); > > + /* Update the parent keys of the right block. */ > > + error = cur->bc_ops->update_keys(tcur, level); > > if (error) > > goto error1; > > > > @@ -2422,6 +2469,11 @@ __xfs_btree_split( > > > > XFS_BTREE_STATS_ADD(cur, moves, rrecs); > > > > + /* Adjust numrecs for the later get_*_keys() calls. */ > > + lrecs -= rrecs; > > + xfs_btree_set_numrecs(left, lrecs); > > + xfs_btree_set_numrecs(right, xfs_btree_get_numrecs(right) + rrecs); > > + > > /* > > * Copy btree block entries from the left block over to the > > * new block, the right. Update the right block and log the > > @@ -2447,14 +2499,15 @@ __xfs_btree_split( > > } > > #endif > > > > + /* Copy the keys & pointers to the new block. */ > > xfs_btree_copy_keys(cur, rkp, lkp, rrecs); > > xfs_btree_copy_ptrs(cur, rpp, lpp, rrecs); > > > > xfs_btree_log_keys(cur, rbp, 1, rrecs); > > xfs_btree_log_ptrs(cur, rbp, 1, rrecs); > > > > - /* Grab the keys to the entries moved to the right block */ > > - xfs_btree_copy_keys(cur, key, rkp, 1); > > + /* Stash the keys of the new block for later insertion. */ > > + cur->bc_ops->get_node_keys(cur, right, key); > > } else { > > /* It's a leaf. Move records. */ > > union xfs_btree_rec *lrp; /* left record pointer */ > > @@ -2463,27 +2516,23 @@ __xfs_btree_split( > > lrp = xfs_btree_rec_addr(cur, src_index, left); > > rrp = xfs_btree_rec_addr(cur, 1, right); > > > > + /* Copy records to the new block. */ > > xfs_btree_copy_recs(cur, rrp, lrp, rrecs); > > xfs_btree_log_recs(cur, rbp, 1, rrecs); > > > > - cur->bc_ops->init_key_from_rec(key, > > - xfs_btree_rec_addr(cur, 1, right)); > > + /* Stash the keys of the new block for later insertion. */ > > + cur->bc_ops->get_leaf_keys(cur, right, key); > > } > > > > - > > /* > > * Find the left block number by looking in the buffer. > > - * Adjust numrecs, sibling pointers. > > + * Adjust sibling pointers. > > */ > > xfs_btree_get_sibling(cur, left, &rrptr, XFS_BB_RIGHTSIB); > > xfs_btree_set_sibling(cur, right, &rrptr, XFS_BB_RIGHTSIB); > > xfs_btree_set_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); > > xfs_btree_set_sibling(cur, left, &rptr, XFS_BB_RIGHTSIB); > > > > - lrecs -= rrecs; > > - xfs_btree_set_numrecs(left, lrecs); > > - xfs_btree_set_numrecs(right, xfs_btree_get_numrecs(right) + rrecs); > > - > > xfs_btree_log_block(cur, rbp, XFS_BB_ALL_BITS); > > xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); > > > > @@ -2802,6 +2851,7 @@ xfs_btree_new_root( > > bp = lbp; > > nptr = 2; > > } > > + > > /* Fill in the new block's btree header and log it. */ > > xfs_btree_init_block_cur(cur, nbp, cur->bc_nlevels, 2); > > xfs_btree_log_block(cur, nbp, XFS_BB_ALL_BITS); > > @@ -2810,19 +2860,24 @@ xfs_btree_new_root( > > > > /* Fill in the key data in the new root. */ > > if (xfs_btree_get_level(left) > 0) { > > - xfs_btree_copy_keys(cur, > > - xfs_btree_key_addr(cur, 1, new), > > - xfs_btree_key_addr(cur, 1, left), 1); > > - xfs_btree_copy_keys(cur, > > - xfs_btree_key_addr(cur, 2, new), > > - xfs_btree_key_addr(cur, 1, right), 1); > > + /* > > + * Get the keys for the left block's keys and put them directly > > + * in the parent block. Do the same for the right block. > > + */ > > + cur->bc_ops->get_node_keys(cur, left, > > + xfs_btree_key_addr(cur, 1, new)); > > + cur->bc_ops->get_node_keys(cur, right, > > + xfs_btree_key_addr(cur, 2, new)); > > } else { > > - cur->bc_ops->init_key_from_rec( > > - xfs_btree_key_addr(cur, 1, new), > > - xfs_btree_rec_addr(cur, 1, left)); > > - cur->bc_ops->init_key_from_rec( > > - xfs_btree_key_addr(cur, 2, new), > > - xfs_btree_rec_addr(cur, 1, right)); > > + /* > > + * Get the keys for the left block's records and put them > > + * directly in the parent block. Do the same for the right > > + * block. > > + */ > > + cur->bc_ops->get_leaf_keys(cur, left, > > + xfs_btree_key_addr(cur, 1, new)); > > + cur->bc_ops->get_leaf_keys(cur, right, > > + xfs_btree_key_addr(cur, 2, new)); > > } > > xfs_btree_log_keys(cur, nbp, 1, 2); > > > > @@ -2858,7 +2913,7 @@ xfs_btree_make_block_unfull( > > int *index, /* new tree index */ > > union xfs_btree_ptr *nptr, /* new btree ptr */ > > struct xfs_btree_cur **ncur, /* new btree cursor */ > > - union xfs_btree_key *key, /* key of new block */ > > + union xfs_btree_key *key, /* key of new block */ > > int *stat) > > { > > int error = 0; > > @@ -3086,8 +3141,8 @@ xfs_btree_insrec( > > xfs_btree_log_block(cur, bp, XFS_BB_NUMRECS); > > > > /* If we inserted at the start of a block, update the parents' keys. */ > > - if (optr == 1) { > > - error = xfs_btree_updkey(cur, key, level + 1); > > + if (xfs_btree_needs_key_update(cur, optr)) { > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > } > > @@ -3107,7 +3162,7 @@ xfs_btree_insrec( > > */ > > *ptrp = nptr; > > if (!xfs_btree_ptr_is_null(cur, &nptr)) { > > - *key = nkey; > > + xfs_btree_copy_keys(cur, key, &nkey, 1); > > *curp = ncur; > > } > > > > @@ -3386,8 +3441,6 @@ xfs_btree_delrec( > > struct xfs_buf *bp; /* buffer for block */ > > int error; /* error return value */ > > int i; /* loop counter */ > > - union xfs_btree_key key; /* storage for keyp */ > > - union xfs_btree_key *keyp = &key; /* passed to the next level */ > > union xfs_btree_ptr lptr; /* left sibling block ptr */ > > struct xfs_buf *lbp; /* left buffer pointer */ > > struct xfs_btree_block *left; /* left btree block */ > > @@ -3458,13 +3511,6 @@ xfs_btree_delrec( > > xfs_btree_log_keys(cur, bp, ptr, numrecs - 1); > > xfs_btree_log_ptrs(cur, bp, ptr, numrecs - 1); > > } > > - > > - /* > > - * If it's the first record in the block, we'll need to pass a > > - * key up to the next level (updkey). > > - */ > > - if (ptr == 1) > > - keyp = xfs_btree_key_addr(cur, 1, block); > > } else { > > /* It's a leaf. operate on records */ > > if (ptr < numrecs) { > > @@ -3473,16 +3519,6 @@ xfs_btree_delrec( > > -1, numrecs - ptr); > > xfs_btree_log_recs(cur, bp, ptr, numrecs - 1); > > } > > - > > - /* > > - * If it's the first record in the block, we'll need a key > > - * structure to pass up to the next level (updkey). > > - */ > > - if (ptr == 1) { > > - cur->bc_ops->init_key_from_rec(&key, > > - xfs_btree_rec_addr(cur, 1, block)); > > - keyp = &key; > > - } > > } > > > > /* > > @@ -3549,8 +3585,8 @@ xfs_btree_delrec( > > * If we deleted the leftmost entry in the block, update the > > * key values above us in the tree. > > */ > > - if (ptr == 1) { > > - error = xfs_btree_updkey(cur, keyp, level + 1); > > + if (xfs_btree_needs_key_update(cur, ptr)) { > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > } > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h > > index b4f3035..e097e60 100644 > > --- a/fs/xfs/libxfs/xfs_btree.h > > +++ b/fs/xfs/libxfs/xfs_btree.h > > @@ -180,6 +180,19 @@ struct xfs_btree_ops { > > union xfs_btree_rec *r1, > > union xfs_btree_rec *r2); > > #endif > > + > > + /* derive the low & high keys from the records in a leaf block */ > > + void (*get_leaf_keys)(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, > > + union xfs_btree_key *key); > > + > > + /* derive the low & high keys from the keys in a node block */ > > + void (*get_node_keys)(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, > > + union xfs_btree_key *key); > > + > > + /* update the parent keys of given btree level */ > > + int (*update_keys)(struct xfs_btree_cur *cur, int level); > > }; > > > > /* > > @@ -475,4 +488,10 @@ bool xfs_btree_sblock_verify(struct xfs_buf *bp, unsigned int max_recs); > > uint xfs_btree_compute_maxlevels(struct xfs_mount *mp, uint *limits, > > unsigned long len); > > > > +void xfs_btree_get_leaf_keys(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, union xfs_btree_key *key); > > +void xfs_btree_get_node_keys(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, union xfs_btree_key *key); > > +int xfs_btree_update_keys(struct xfs_btree_cur *cur, int level); > > + > > #endif /* __XFS_BTREE_H__ */ > > diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c > > index 88da2ad..a48f448 100644 > > --- a/fs/xfs/libxfs/xfs_ialloc_btree.c > > +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c > > @@ -314,6 +314,10 @@ static const struct xfs_btree_ops xfs_inobt_ops = { > > .keys_inorder = xfs_inobt_keys_inorder, > > .recs_inorder = xfs_inobt_recs_inorder, > > #endif > > + > > + .get_leaf_keys = xfs_btree_get_leaf_keys, > > + .get_node_keys = xfs_btree_get_node_keys, > > + .update_keys = xfs_btree_update_keys, > > }; > > > > static const struct xfs_btree_ops xfs_finobt_ops = { > > @@ -335,6 +339,10 @@ static const struct xfs_btree_ops xfs_finobt_ops = { > > .keys_inorder = xfs_inobt_keys_inorder, > > .recs_inorder = xfs_inobt_recs_inorder, > > #endif > > + > > + .get_leaf_keys = xfs_btree_get_leaf_keys, > > + .get_node_keys = xfs_btree_get_node_keys, > > + .update_keys = xfs_btree_update_keys, > > }; > > > > /* > > > > _______________________________________________ > > xfs mailing list > > xfs@oss.sgi.com > > http://oss.sgi.com/mailman/listinfo/xfs > > _______________________________________________ > xfs mailing list > xfs@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/xfs From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay2.corp.sgi.com [137.38.102.29]) by oss.sgi.com (Postfix) with ESMTP id EAA017D24 for ; Mon, 1 Aug 2016 12:54:16 -0500 (CDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by relay2.corp.sgi.com (Postfix) with ESMTP id 7F5AF304051 for ; Mon, 1 Aug 2016 10:54:16 -0700 (PDT) Received: from aserp1040.oracle.com (aserp1040.oracle.com [141.146.126.69]) by cuda.sgi.com with ESMTP id wBeBZBhfHTjSYfpO (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO) for ; Mon, 01 Aug 2016 10:54:13 -0700 (PDT) Date: Mon, 1 Aug 2016 10:54:07 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH v2 07/47] xfs: add function pointers for get/update keys to the btree Message-ID: <20160801175407.GA8593@birch.djwong.org> References: <146907695530.25461.3225785294902719773.stgit@birch.djwong.org> <146907700604.25461.2181974283557088355.stgit@birch.djwong.org> <20160728194615.GA15746@birch.djwong.org> <20160801155736.GA64218@bfoster.bfoster> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20160801155736.GA64218@bfoster.bfoster> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: Brian Foster Cc: linux-fsdevel@vger.kernel.org, vishal.l.verma@intel.com, xfs@oss.sgi.com On Mon, Aug 01, 2016 at 11:57:36AM -0400, Brian Foster wrote: > On Thu, Jul 28, 2016 at 12:46:15PM -0700, Darrick J. Wong wrote: > > Add some function pointers to bc_ops to get the btree keys for > > leaf and node blocks, and to update parent keys of a block. > > Convert the _btree_updkey calls to use our new pointer, and > > modify the tree shape changing code to call the appropriate > > get_*_keys pointer instead of _btree_copy_keys because the > > overlapping btree has to calculate high key values. > > > > Signed-off-by: Darrick J. Wong > > --- > > fs/xfs/libxfs/xfs_alloc_btree.c | 4 + > > fs/xfs/libxfs/xfs_bmap_btree.c | 4 + > > fs/xfs/libxfs/xfs_btree.c | 164 +++++++++++++++++++++++--------------- > > fs/xfs/libxfs/xfs_btree.h | 19 ++++ > > fs/xfs/libxfs/xfs_ialloc_btree.c | 8 ++ > > 5 files changed, 135 insertions(+), 64 deletions(-) > > > ... > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c > > index 8d8e362..405442d 100644 > > --- a/fs/xfs/libxfs/xfs_btree.c > > +++ b/fs/xfs/libxfs/xfs_btree.c > ... > > @@ -2146,11 +2194,10 @@ xfs_btree_lshift( > > */ > > cur->bc_ops->init_key_from_rec(&key, > > xfs_btree_rec_addr(cur, 1, right)); > > - rkp = &key; > > Sorry, I should have been more clear. I think this entire hunk (the > above key initialization and comment) is effectively dead. > > > } > > > > - /* Update the parent key values of right. */ > > - error = xfs_btree_updkey(cur, rkp, level + 1); > > + /* Update the parent keys of the right block. */ > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > > > @@ -2292,7 +2339,6 @@ xfs_btree_rshift( > > xfs_btree_log_recs(cur, rbp, 1, rrecs + 1); > > > > cur->bc_ops->init_key_from_rec(&key, rrp); > > - rkp = &key; > > Same here. key isn't used beyond this. > > I see that hch has some objections to the design here. FWIW, resolution > of that aside, and with the above fixes: Ah, right. "key" and its initializers can go away in both functions. I'll fix that and find out if Dave wants to take this (and the other fixes) for a 4.8 merge request, or just defer them until -rc2. --D > Reviewed-by: Brian Foster > > > > > ASSERT(cur->bc_ops->recs_inorder(cur, rrp, > > xfs_btree_rec_addr(cur, 2, right))); > > @@ -2321,7 +2367,8 @@ xfs_btree_rshift( > > if (error) > > goto error1; > > > > - error = xfs_btree_updkey(tcur, rkp, level + 1); > > + /* Update the parent keys of the right block. */ > > + error = cur->bc_ops->update_keys(tcur, level); > > if (error) > > goto error1; > > > > @@ -2422,6 +2469,11 @@ __xfs_btree_split( > > > > XFS_BTREE_STATS_ADD(cur, moves, rrecs); > > > > + /* Adjust numrecs for the later get_*_keys() calls. */ > > + lrecs -= rrecs; > > + xfs_btree_set_numrecs(left, lrecs); > > + xfs_btree_set_numrecs(right, xfs_btree_get_numrecs(right) + rrecs); > > + > > /* > > * Copy btree block entries from the left block over to the > > * new block, the right. Update the right block and log the > > @@ -2447,14 +2499,15 @@ __xfs_btree_split( > > } > > #endif > > > > + /* Copy the keys & pointers to the new block. */ > > xfs_btree_copy_keys(cur, rkp, lkp, rrecs); > > xfs_btree_copy_ptrs(cur, rpp, lpp, rrecs); > > > > xfs_btree_log_keys(cur, rbp, 1, rrecs); > > xfs_btree_log_ptrs(cur, rbp, 1, rrecs); > > > > - /* Grab the keys to the entries moved to the right block */ > > - xfs_btree_copy_keys(cur, key, rkp, 1); > > + /* Stash the keys of the new block for later insertion. */ > > + cur->bc_ops->get_node_keys(cur, right, key); > > } else { > > /* It's a leaf. Move records. */ > > union xfs_btree_rec *lrp; /* left record pointer */ > > @@ -2463,27 +2516,23 @@ __xfs_btree_split( > > lrp = xfs_btree_rec_addr(cur, src_index, left); > > rrp = xfs_btree_rec_addr(cur, 1, right); > > > > + /* Copy records to the new block. */ > > xfs_btree_copy_recs(cur, rrp, lrp, rrecs); > > xfs_btree_log_recs(cur, rbp, 1, rrecs); > > > > - cur->bc_ops->init_key_from_rec(key, > > - xfs_btree_rec_addr(cur, 1, right)); > > + /* Stash the keys of the new block for later insertion. */ > > + cur->bc_ops->get_leaf_keys(cur, right, key); > > } > > > > - > > /* > > * Find the left block number by looking in the buffer. > > - * Adjust numrecs, sibling pointers. > > + * Adjust sibling pointers. > > */ > > xfs_btree_get_sibling(cur, left, &rrptr, XFS_BB_RIGHTSIB); > > xfs_btree_set_sibling(cur, right, &rrptr, XFS_BB_RIGHTSIB); > > xfs_btree_set_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); > > xfs_btree_set_sibling(cur, left, &rptr, XFS_BB_RIGHTSIB); > > > > - lrecs -= rrecs; > > - xfs_btree_set_numrecs(left, lrecs); > > - xfs_btree_set_numrecs(right, xfs_btree_get_numrecs(right) + rrecs); > > - > > xfs_btree_log_block(cur, rbp, XFS_BB_ALL_BITS); > > xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); > > > > @@ -2802,6 +2851,7 @@ xfs_btree_new_root( > > bp = lbp; > > nptr = 2; > > } > > + > > /* Fill in the new block's btree header and log it. */ > > xfs_btree_init_block_cur(cur, nbp, cur->bc_nlevels, 2); > > xfs_btree_log_block(cur, nbp, XFS_BB_ALL_BITS); > > @@ -2810,19 +2860,24 @@ xfs_btree_new_root( > > > > /* Fill in the key data in the new root. */ > > if (xfs_btree_get_level(left) > 0) { > > - xfs_btree_copy_keys(cur, > > - xfs_btree_key_addr(cur, 1, new), > > - xfs_btree_key_addr(cur, 1, left), 1); > > - xfs_btree_copy_keys(cur, > > - xfs_btree_key_addr(cur, 2, new), > > - xfs_btree_key_addr(cur, 1, right), 1); > > + /* > > + * Get the keys for the left block's keys and put them directly > > + * in the parent block. Do the same for the right block. > > + */ > > + cur->bc_ops->get_node_keys(cur, left, > > + xfs_btree_key_addr(cur, 1, new)); > > + cur->bc_ops->get_node_keys(cur, right, > > + xfs_btree_key_addr(cur, 2, new)); > > } else { > > - cur->bc_ops->init_key_from_rec( > > - xfs_btree_key_addr(cur, 1, new), > > - xfs_btree_rec_addr(cur, 1, left)); > > - cur->bc_ops->init_key_from_rec( > > - xfs_btree_key_addr(cur, 2, new), > > - xfs_btree_rec_addr(cur, 1, right)); > > + /* > > + * Get the keys for the left block's records and put them > > + * directly in the parent block. Do the same for the right > > + * block. > > + */ > > + cur->bc_ops->get_leaf_keys(cur, left, > > + xfs_btree_key_addr(cur, 1, new)); > > + cur->bc_ops->get_leaf_keys(cur, right, > > + xfs_btree_key_addr(cur, 2, new)); > > } > > xfs_btree_log_keys(cur, nbp, 1, 2); > > > > @@ -2858,7 +2913,7 @@ xfs_btree_make_block_unfull( > > int *index, /* new tree index */ > > union xfs_btree_ptr *nptr, /* new btree ptr */ > > struct xfs_btree_cur **ncur, /* new btree cursor */ > > - union xfs_btree_key *key, /* key of new block */ > > + union xfs_btree_key *key, /* key of new block */ > > int *stat) > > { > > int error = 0; > > @@ -3086,8 +3141,8 @@ xfs_btree_insrec( > > xfs_btree_log_block(cur, bp, XFS_BB_NUMRECS); > > > > /* If we inserted at the start of a block, update the parents' keys. */ > > - if (optr == 1) { > > - error = xfs_btree_updkey(cur, key, level + 1); > > + if (xfs_btree_needs_key_update(cur, optr)) { > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > } > > @@ -3107,7 +3162,7 @@ xfs_btree_insrec( > > */ > > *ptrp = nptr; > > if (!xfs_btree_ptr_is_null(cur, &nptr)) { > > - *key = nkey; > > + xfs_btree_copy_keys(cur, key, &nkey, 1); > > *curp = ncur; > > } > > > > @@ -3386,8 +3441,6 @@ xfs_btree_delrec( > > struct xfs_buf *bp; /* buffer for block */ > > int error; /* error return value */ > > int i; /* loop counter */ > > - union xfs_btree_key key; /* storage for keyp */ > > - union xfs_btree_key *keyp = &key; /* passed to the next level */ > > union xfs_btree_ptr lptr; /* left sibling block ptr */ > > struct xfs_buf *lbp; /* left buffer pointer */ > > struct xfs_btree_block *left; /* left btree block */ > > @@ -3458,13 +3511,6 @@ xfs_btree_delrec( > > xfs_btree_log_keys(cur, bp, ptr, numrecs - 1); > > xfs_btree_log_ptrs(cur, bp, ptr, numrecs - 1); > > } > > - > > - /* > > - * If it's the first record in the block, we'll need to pass a > > - * key up to the next level (updkey). > > - */ > > - if (ptr == 1) > > - keyp = xfs_btree_key_addr(cur, 1, block); > > } else { > > /* It's a leaf. operate on records */ > > if (ptr < numrecs) { > > @@ -3473,16 +3519,6 @@ xfs_btree_delrec( > > -1, numrecs - ptr); > > xfs_btree_log_recs(cur, bp, ptr, numrecs - 1); > > } > > - > > - /* > > - * If it's the first record in the block, we'll need a key > > - * structure to pass up to the next level (updkey). > > - */ > > - if (ptr == 1) { > > - cur->bc_ops->init_key_from_rec(&key, > > - xfs_btree_rec_addr(cur, 1, block)); > > - keyp = &key; > > - } > > } > > > > /* > > @@ -3549,8 +3585,8 @@ xfs_btree_delrec( > > * If we deleted the leftmost entry in the block, update the > > * key values above us in the tree. > > */ > > - if (ptr == 1) { > > - error = xfs_btree_updkey(cur, keyp, level + 1); > > + if (xfs_btree_needs_key_update(cur, ptr)) { > > + error = cur->bc_ops->update_keys(cur, level); > > if (error) > > goto error0; > > } > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h > > index b4f3035..e097e60 100644 > > --- a/fs/xfs/libxfs/xfs_btree.h > > +++ b/fs/xfs/libxfs/xfs_btree.h > > @@ -180,6 +180,19 @@ struct xfs_btree_ops { > > union xfs_btree_rec *r1, > > union xfs_btree_rec *r2); > > #endif > > + > > + /* derive the low & high keys from the records in a leaf block */ > > + void (*get_leaf_keys)(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, > > + union xfs_btree_key *key); > > + > > + /* derive the low & high keys from the keys in a node block */ > > + void (*get_node_keys)(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, > > + union xfs_btree_key *key); > > + > > + /* update the parent keys of given btree level */ > > + int (*update_keys)(struct xfs_btree_cur *cur, int level); > > }; > > > > /* > > @@ -475,4 +488,10 @@ bool xfs_btree_sblock_verify(struct xfs_buf *bp, unsigned int max_recs); > > uint xfs_btree_compute_maxlevels(struct xfs_mount *mp, uint *limits, > > unsigned long len); > > > > +void xfs_btree_get_leaf_keys(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, union xfs_btree_key *key); > > +void xfs_btree_get_node_keys(struct xfs_btree_cur *cur, > > + struct xfs_btree_block *block, union xfs_btree_key *key); > > +int xfs_btree_update_keys(struct xfs_btree_cur *cur, int level); > > + > > #endif /* __XFS_BTREE_H__ */ > > diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c > > index 88da2ad..a48f448 100644 > > --- a/fs/xfs/libxfs/xfs_ialloc_btree.c > > +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c > > @@ -314,6 +314,10 @@ static const struct xfs_btree_ops xfs_inobt_ops = { > > .keys_inorder = xfs_inobt_keys_inorder, > > .recs_inorder = xfs_inobt_recs_inorder, > > #endif > > + > > + .get_leaf_keys = xfs_btree_get_leaf_keys, > > + .get_node_keys = xfs_btree_get_node_keys, > > + .update_keys = xfs_btree_update_keys, > > }; > > > > static const struct xfs_btree_ops xfs_finobt_ops = { > > @@ -335,6 +339,10 @@ static const struct xfs_btree_ops xfs_finobt_ops = { > > .keys_inorder = xfs_inobt_keys_inorder, > > .recs_inorder = xfs_inobt_recs_inorder, > > #endif > > + > > + .get_leaf_keys = xfs_btree_get_leaf_keys, > > + .get_node_keys = xfs_btree_get_node_keys, > > + .update_keys = xfs_btree_update_keys, > > }; > > > > /* > > > > _______________________________________________ > > xfs mailing list > > xfs@oss.sgi.com > > http://oss.sgi.com/mailman/listinfo/xfs > > _______________________________________________ > xfs mailing list > xfs@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/xfs _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs