From b38958d715316031fe9ea0cc6c22043072a55f49 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 20 May 2013 09:51:14 +1000 Subject: xfs: xfs_attr_shortform_allfit() does not handle attr3 format. xfstests generic/117 fails with: XFS: Assertion failed: leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC) indicating a function that does not handle the attr3 format correctly. Fix it. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 08d5457c948e..8eeb88fb3201 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -931,20 +931,22 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) */ int xfs_attr_shortform_allfit( - struct xfs_buf *bp, - struct xfs_inode *dp) + struct xfs_buf *bp, + struct xfs_inode *dp) { - xfs_attr_leafblock_t *leaf; - xfs_attr_leaf_entry_t *entry; + struct xfs_attr_leafblock *leaf; + struct xfs_attr_leaf_entry *entry; xfs_attr_leaf_name_local_t *name_loc; - int bytes, i; + struct xfs_attr3_icleaf_hdr leafhdr; + int bytes; + int i; leaf = bp->b_addr; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)); + xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); + entry = xfs_attr3_leaf_entryp(leaf); - entry = &leaf->entries[0]; bytes = sizeof(struct xfs_attr_sf_hdr); - for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) { + for (i = 0; i < leafhdr.count; entry++, i++) { if (entry->flags & XFS_ATTR_INCOMPLETE) continue; /* don't copy partial entries */ if (!(entry->flags & XFS_ATTR_LOCAL)) @@ -954,15 +956,15 @@ xfs_attr_shortform_allfit( return(0); if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX) return(0); - bytes += sizeof(struct xfs_attr_sf_entry)-1 + bytes += sizeof(struct xfs_attr_sf_entry) - 1 + name_loc->namelen + be16_to_cpu(name_loc->valuelen); } if ((dp->i_mount->m_flags & XFS_MOUNT_ATTR2) && (dp->i_d.di_format != XFS_DINODE_FMT_BTREE) && (bytes == sizeof(struct xfs_attr_sf_hdr))) - return(-1); - return(xfs_attr_shortform_bytesfit(dp, bytes)); + return -1; + return xfs_attr_shortform_bytesfit(dp, bytes); } /* -- cgit v1.2.3 From e461fcb194172b3f709e0b478d2ac1bdac7ab9a3 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 20 May 2013 09:51:16 +1000 Subject: xfs: remote attribute lookups require the value length When reading a remote attribute, to correctly calculate the length of the data buffer for CRC enable filesystems, we need to know the length of the attribute data. We get this information when we look up the attribute, but we don't store it in the args structure along with the other remote attr information we get from the lookup. Add this information to the args structure so we can use it appropriately. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 8eeb88fb3201..0bce1b348580 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -2332,9 +2332,10 @@ xfs_attr3_leaf_lookup_int( if (!xfs_attr_namesp_match(args->flags, entry->flags)) continue; args->index = probe; + args->valuelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, - be32_to_cpu(name_rmt->valuelen)); + args->valuelen); return XFS_ERROR(EEXIST); } } -- cgit v1.2.3 From 8517de2a81da830f5d90da66b4799f4040c76dc9 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 21 May 2013 18:02:05 +1000 Subject: xfs: fully initialise temp leaf in xfs_attr3_leaf_unbalance xfs_attr3_leaf_unbalance() uses a temporary buffer for recombining the entries in two leaves when the destination leaf requires compaction. The temporary buffer ends up being copied back over the original destination buffer, so the header in the temporary buffer needs to contain all the information that is in the destination buffer. To make sure the temporary buffer is fully initialised, once we've set up the temporary incore header appropriately, write is back to the temporary buffer before starting to move entries around. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 0bce1b348580..79ece72976ae 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -2181,14 +2181,24 @@ xfs_attr3_leaf_unbalance( struct xfs_attr_leafblock *tmp_leaf; struct xfs_attr3_icleaf_hdr tmphdr; - tmp_leaf = kmem_alloc(state->blocksize, KM_SLEEP); - memset(tmp_leaf, 0, state->blocksize); - memset(&tmphdr, 0, sizeof(tmphdr)); + tmp_leaf = kmem_zalloc(state->blocksize, KM_SLEEP); + + /* + * Copy the header into the temp leaf so that all the stuff + * not in the incore header is present and gets copied back in + * once we've moved all the entries. + */ + memcpy(tmp_leaf, save_leaf, xfs_attr3_leaf_hdr_size(save_leaf)); + memset(&tmphdr, 0, sizeof(tmphdr)); tmphdr.magic = savehdr.magic; tmphdr.forw = savehdr.forw; tmphdr.back = savehdr.back; tmphdr.firstused = state->blocksize; + + /* write the header to the temp buffer to initialise it */ + xfs_attr3_leaf_hdr_to_disk(tmp_leaf, &tmphdr); + if (xfs_attr3_leaf_order(save_blk->bp, &savehdr, drop_blk->bp, &drophdr)) { xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0, -- cgit v1.2.3 From d4c712bcf26a25c2b67c90e44e0b74c7993b5334 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 21 May 2013 18:02:06 +1000 Subject: xfs: fully initialise temp leaf in xfs_attr3_leaf_compact xfs_attr3_leaf_compact() uses a temporary buffer for compacting the the entries in a leaf. It copies the the original buffer into the temporary buffer, then zeros the original buffer completely. It then copies the entries back into the original buffer. However, the original buffer has not been correctly initialised, and so the movement of the entries goes horribly wrong. Make sure the zeroed destination buffer is fully initialised, and once we've set up the destination incore header appropriately, write is back to the buffer before starting to move entries around. While debugging this, the _d/_s prefixes weren't sufficient to remind me what buffer was what, so rename then all _src/_dst. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 42 ++++++++++++++++++++++++++---------------- 1 file changed, 26 insertions(+), 16 deletions(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 79ece72976ae..5b03d15b707b 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -1445,11 +1445,12 @@ xfs_attr3_leaf_add_work( STATIC void xfs_attr3_leaf_compact( struct xfs_da_args *args, - struct xfs_attr3_icleaf_hdr *ichdr_d, + struct xfs_attr3_icleaf_hdr *ichdr_dst, struct xfs_buf *bp) { - xfs_attr_leafblock_t *leaf_s, *leaf_d; - struct xfs_attr3_icleaf_hdr ichdr_s; + struct xfs_attr_leafblock *leaf_src; + struct xfs_attr_leafblock *leaf_dst; + struct xfs_attr3_icleaf_hdr ichdr_src; struct xfs_trans *trans = args->trans; struct xfs_mount *mp = trans->t_mountp; char *tmpbuffer; @@ -1457,29 +1458,38 @@ xfs_attr3_leaf_compact( trace_xfs_attr_leaf_compact(args); tmpbuffer = kmem_alloc(XFS_LBSIZE(mp), KM_SLEEP); - ASSERT(tmpbuffer != NULL); memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(mp)); memset(bp->b_addr, 0, XFS_LBSIZE(mp)); + leaf_src = (xfs_attr_leafblock_t *)tmpbuffer; + leaf_dst = bp->b_addr; /* - * Copy basic information + * Copy the on-disk header back into the destination buffer to ensure + * all the information in the header that is not part of the incore + * header structure is preserved. */ - leaf_s = (xfs_attr_leafblock_t *)tmpbuffer; - leaf_d = bp->b_addr; - ichdr_s = *ichdr_d; /* struct copy */ - ichdr_d->firstused = XFS_LBSIZE(mp); - ichdr_d->usedbytes = 0; - ichdr_d->count = 0; - ichdr_d->holes = 0; - ichdr_d->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_s); - ichdr_d->freemap[0].size = ichdr_d->firstused - ichdr_d->freemap[0].base; + memcpy(bp->b_addr, tmpbuffer, xfs_attr3_leaf_hdr_size(leaf_src)); + + /* Initialise the incore headers */ + ichdr_src = *ichdr_dst; /* struct copy */ + ichdr_dst->firstused = XFS_LBSIZE(mp); + ichdr_dst->usedbytes = 0; + ichdr_dst->count = 0; + ichdr_dst->holes = 0; + ichdr_dst->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_src); + ichdr_dst->freemap[0].size = ichdr_dst->firstused - + ichdr_dst->freemap[0].base; + + + /* write the header back to initialise the underlying buffer */ + xfs_attr3_leaf_hdr_to_disk(leaf_dst, ichdr_dst); /* * Copy all entry's in the same (sorted) order, * but allocate name/value pairs packed and in sequence. */ - xfs_attr3_leaf_moveents(leaf_s, &ichdr_s, 0, leaf_d, ichdr_d, 0, - ichdr_s.count, mp); + xfs_attr3_leaf_moveents(leaf_src, &ichdr_src, 0, leaf_dst, ichdr_dst, 0, + ichdr_src.count, mp); /* * this logs the entire buffer, but the caller must write the header * back to the buffer when it is finished modifying it. -- cgit v1.2.3 From ad1858d77771172e08016890f0eb2faedec3ecee Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 21 May 2013 18:02:08 +1000 Subject: xfs: rework remote attr CRCs Note: this changes the on-disk remote attribute format. I assert that this is OK to do as CRCs are marked experimental and the first kernel it is included in has not yet reached release yet. Further, the userspace utilities are still evolving and so anyone using this stuff right now is a developer or tester using volatile filesystems for testing this feature. Hence changing the format right now to save longer term pain is the right thing to do. The fundamental change is to move from a header per extent in the attribute to a header per filesytem block in the attribute. This means there are more header blocks and the parsing of the attribute data is slightly more complex, but it has the advantage that we always know the size of the attribute on disk based on the length of the data it contains. This is where the header-per-extent method has problems. We don't know the size of the attribute on disk without first knowing how many extents are used to hold it. And we can't tell from a mapping lookup, either, because remote attributes can be allocated contiguously with other attribute blocks and so there is no obvious way of determining the actual size of the atribute on disk short of walking and mapping buffers. The problem with this approach is that if we map a buffer incorrectly (e.g. we make the last buffer for the attribute data too long), we then get buffer cache lookup failure when we map it correctly. i.e. we get a size mismatch on lookup. This is not necessarily fatal, but it's a cache coherency problem that can lead to returning the wrong data to userspace or writing the wrong data to disk. And debug kernels will assert fail if this occurs. I found lots of niggly little problems trying to fix this issue on a 4k block size filesystem, finally getting it to pass with lots of fixes. The thing is, 1024 byte filesystems still failed, and it was getting really complex handling all the corner cases that were showing up. And there were clearly more that I hadn't found yet. It is complex, fragile code, and if we don't fix it now, it will be complex, fragile code forever more. Hence the simple fix is to add a header to each filesystem block. This gives us the same relationship between the attribute data length and the number of blocks on disk as we have without CRCs - it's a linear mapping and doesn't require us to guess anything. It is simple to implement, too - the remote block count calculated at lookup time can be used by the remote attribute set/get/remove code without modification for both CRC and non-CRC filesystems. The world becomes sane again. Because the copy-in and copy-out now need to iterate over each filesystem block, I moved them into helper functions so we separate the block mapping and buffer manupulations from the attribute data and CRC header manipulations. The code becomes much clearer as a result, and it is a lot easier to understand and debug. It also appears to be much more robust - once it worked on 4k block size filesystems, it has worked without failure on 1k block size filesystems, too. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 5b03d15b707b..d788302e506a 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -1412,7 +1412,7 @@ xfs_attr3_leaf_add_work( name_rmt->valuelen = 0; name_rmt->valueblk = 0; args->rmtblkno = 1; - args->rmtblkcnt = XFS_B_TO_FSB(mp, args->valuelen); + args->rmtblkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); } xfs_trans_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index), @@ -2354,8 +2354,9 @@ xfs_attr3_leaf_lookup_int( args->index = probe; args->valuelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); - args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, - args->valuelen); + args->rmtblkcnt = xfs_attr3_rmt_blocks( + args->dp->i_mount, + args->valuelen); return XFS_ERROR(EEXIST); } } @@ -2406,7 +2407,8 @@ xfs_attr3_leaf_getvalue( ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0); valuelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); - args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, valuelen); + args->rmtblkcnt = xfs_attr3_rmt_blocks(args->dp->i_mount, + valuelen); if (args->flags & ATTR_KERNOVAL) { args->valuelen = valuelen; return 0; @@ -2732,7 +2734,8 @@ xfs_attr3_leaf_list_int( args.valuelen = valuelen; args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS); args.rmtblkno = be32_to_cpu(name_rmt->valueblk); - args.rmtblkcnt = XFS_B_TO_FSB(args.dp->i_mount, valuelen); + args.rmtblkcnt = xfs_attr3_rmt_blocks( + args.dp->i_mount, valuelen); retval = xfs_attr_rmtval_get(&args); if (retval) return retval; -- cgit v1.2.3 From 59913f14dfe8eb772ff93eb442947451b4416329 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 3 Jun 2013 15:28:49 +1000 Subject: xfs: fix remote attribute invalidation for a leaf When invalidating an attribute leaf block block, there might be remote attributes that it points to. With the recent rework of the remote attribute format, we have to make sure we calculate the length of the attribute correctly. We aren't doing that in xfs_attr3_leaf_inactive(), so fix it. Signed-off-by: Dave Chinner Reviewed-by: Brian Foster Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index d788302e506a..31d3cd129269 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -3258,7 +3258,7 @@ xfs_attr3_leaf_inactive( name_rmt = xfs_attr3_leaf_name_remote(leaf, i); if (name_rmt->valueblk) { lp->valueblk = be32_to_cpu(name_rmt->valueblk); - lp->valuelen = XFS_B_TO_FSB(dp->i_mount, + lp->valuelen = xfs_attr3_rmt_blocks(dp->i_mount, be32_to_cpu(name_rmt->valuelen)); lp++; } -- cgit v1.2.3 From f3508bcddf8fae6ebd21000d708cf09e7e77a963 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 10 Jul 2013 07:04:00 +1000 Subject: xfs: remove local fork format handling from xfs_bmapi_write() The conversion from local format to extent format requires interpretation of the data in the fork being converted, so it cannot be done in a generic way. It is up to the caller to convert the fork format to extent format before calling into xfs_bmapi_write() so format conversion can be done correctly. The code in xfs_bmapi_write() to convert the format is used implicitly by the attribute and directory code, but they specifically zero the fork size so that the conversion does not do any allocation or manipulation. Move this conversion into the shortform to leaf functions for the dir/attr code so the conversions are explicitly controlled by all callers. Now we can remove the conversion code in xfs_bmapi_write. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_attr_leaf.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs/xfs/xfs_attr_leaf.c') diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 31d3cd129269..b800fbcafc7f 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -690,6 +690,8 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args) sf = (xfs_attr_shortform_t *)tmpbuffer; xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); + xfs_bmap_local_to_extents_empty(dp, XFS_ATTR_FORK); + bp = NULL; error = xfs_da_grow_inode(args, &blkno); if (error) { -- cgit v1.2.3