Revision 9230a0b65b47fe6856c4468ec0175c4987e5bede authored by Dave Chinner on 20 November 2018, 06:50:08 UTC, committed by Darrick J. Wong on 21 November 2018, 18:10:53 UTC
Long saga. There have been days spent following this through dead end
after dead end in multi-GB event traces. This morning, after writing
a trace-cmd wrapper that enabled me to be more selective about XFS
trace points, I discovered that I could get just enough essential
tracepoints enabled that there was a 50:50 chance the fsx config
would fail at ~115k ops. If it didn't fail at op 115547, I stopped
fsx at op 115548 anyway.

That gave me two traces - one where the problem manifested, and one
where it didn't. After refining the traces to have the necessary
information, I found that in the failing case there was a real
extent in the COW fork compared to an unwritten extent in the
working case.

Walking back through the two traces to the point where the CWO fork
extents actually diverged, I found that the bad case had an extra
unwritten extent in it. This is likely because the bug it led me to
had triggered multiple times in those 115k ops, leaving stray
COW extents around. What I saw was a COW delalloc conversion to an
unwritten extent (as they should always be through
xfs_iomap_write_allocate()) resulted in a /written extent/:

xfs_writepage:        dev 259:0 ino 0x83 pgoff 0x17000 size 0x79a00 offset 0 length 0
xfs_iext_remove:      dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/2 offset 32 block 152 count 20 flag 1 caller xfs_bmap_add_extent_delay_real
xfs_bmap_pre_update:  dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 4503599627239429 count 31 flag 0 caller xfs_bmap_add_extent_delay_real
xfs_bmap_post_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 121 count 51 flag 0 caller xfs_bmap_add_ex

Basically, Cow fork before:

	0 1            32          52
	+H+DDDDDDDDDDDD+UUUUUUUUUUU+
	   PREV		RIGHT

COW delalloc conversion allocates:

	  1	       32
	  +uuuuuuuuuuuu+
	  NEW

And the result according to the xfs_bmap_post_update trace was:

	0 1            32          52
	+H+wwwwwwwwwwwwwwwwwwwwwwww+
	   PREV

Which is clearly wrong - it should be a merged unwritten extent,
not an unwritten extent.

That lead me to look at the LEFT_FILLING|RIGHT_FILLING|RIGHT_CONTIG
case in xfs_bmap_add_extent_delay_real(), and sure enough, there's
the bug.

It takes the old delalloc extent (PREV) and adds the length of the
RIGHT extent to it, takes the start block from NEW, removes the
RIGHT extent and then updates PREV with the new extent.

What it fails to do is update PREV.br_state. For delalloc, this is
always XFS_EXT_NORM, while in this case we are converting the
delayed allocation to unwritten, so it needs to be updated to
XFS_EXT_UNWRITTEN. This LF|RF|RC case does not do this, and so
the resultant extent is always written.

And that's the bug I've been chasing for a week - a bmap btree bug,
not a reflink/dedupe/copy_file_range bug, but a BMBT bug introduced
with the recent in core extent tree scalability enhancements.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
1 parent 2c30717
Raw File
netlabel_calipso.h
/*
 * NetLabel CALIPSO Support
 *
 * This file defines the CALIPSO functions for the NetLabel system.  The
 * NetLabel system manages static and dynamic label mappings for network
 * protocols such as CIPSO and RIPSO.
 *
 * Authors: Paul Moore <paul@paul-moore.com>
 *          Huw Davies <huw@codeweavers.com>
 *
 */

/* (c) Copyright Hewlett-Packard Development Company, L.P., 2006
 * (c) Copyright Huw Davies <huw@codeweavers.com>, 2015
 *
 * This program is free software;  you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY;  without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
 * the GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program;  if not, see <http://www.gnu.org/licenses/>.
 *
 */

#ifndef _NETLABEL_CALIPSO
#define _NETLABEL_CALIPSO

#include <net/netlabel.h>
#include <net/calipso.h>

/* The following NetLabel payloads are supported by the CALIPSO subsystem.
 *
 * o ADD:
 *   Sent by an application to add a new DOI mapping table.
 *
 *   Required attributes:
 *
 *     NLBL_CALIPSO_A_DOI
 *     NLBL_CALIPSO_A_MTYPE
 *
 *   If using CALIPSO_MAP_PASS no additional attributes are required.
 *
 * o REMOVE:
 *   Sent by an application to remove a specific DOI mapping table from the
 *   CALIPSO system.
 *
 *   Required attributes:
 *
 *     NLBL_CALIPSO_A_DOI
 *
 * o LIST:
 *   Sent by an application to list the details of a DOI definition.  On
 *   success the kernel should send a response using the following format.
 *
 *   Required attributes:
 *
 *     NLBL_CALIPSO_A_DOI
 *
 *   The valid response message format depends on the type of the DOI mapping,
 *   the defined formats are shown below.
 *
 *   Required attributes:
 *
 *     NLBL_CALIPSO_A_MTYPE
 *
 *   If using CALIPSO_MAP_PASS no additional attributes are required.
 *
 * o LISTALL:
 *   This message is sent by an application to list the valid DOIs on the
 *   system.  When sent by an application there is no payload and the
 *   NLM_F_DUMP flag should be set.  The kernel should respond with a series of
 *   the following messages.
 *
 *   Required attributes:
 *
 *    NLBL_CALIPSO_A_DOI
 *    NLBL_CALIPSO_A_MTYPE
 *
 */

/* NetLabel CALIPSO commands */
enum {
	NLBL_CALIPSO_C_UNSPEC,
	NLBL_CALIPSO_C_ADD,
	NLBL_CALIPSO_C_REMOVE,
	NLBL_CALIPSO_C_LIST,
	NLBL_CALIPSO_C_LISTALL,
	__NLBL_CALIPSO_C_MAX,
};

/* NetLabel CALIPSO attributes */
enum {
	NLBL_CALIPSO_A_UNSPEC,
	NLBL_CALIPSO_A_DOI,
	/* (NLA_U32)
	 * the DOI value */
	NLBL_CALIPSO_A_MTYPE,
	/* (NLA_U32)
	 * the mapping table type (defined in the calipso.h header as
	 * CALIPSO_MAP_*) */
	__NLBL_CALIPSO_A_MAX,
};

#define NLBL_CALIPSO_A_MAX (__NLBL_CALIPSO_A_MAX - 1)

/* NetLabel protocol functions */
#if IS_ENABLED(CONFIG_IPV6)
int netlbl_calipso_genl_init(void);
#else
static inline int netlbl_calipso_genl_init(void)
{
	return 0;
}
#endif

int calipso_doi_add(struct calipso_doi *doi_def,
		    struct netlbl_audit *audit_info);
void calipso_doi_free(struct calipso_doi *doi_def);
int calipso_doi_remove(u32 doi, struct netlbl_audit *audit_info);
struct calipso_doi *calipso_doi_getdef(u32 doi);
void calipso_doi_putdef(struct calipso_doi *doi_def);
int calipso_doi_walk(u32 *skip_cnt,
		     int (*callback)(struct calipso_doi *doi_def, void *arg),
		     void *cb_arg);
int calipso_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr);
int calipso_sock_setattr(struct sock *sk,
			 const struct calipso_doi *doi_def,
			 const struct netlbl_lsm_secattr *secattr);
void calipso_sock_delattr(struct sock *sk);
int calipso_req_setattr(struct request_sock *req,
			const struct calipso_doi *doi_def,
			const struct netlbl_lsm_secattr *secattr);
void calipso_req_delattr(struct request_sock *req);
unsigned char *calipso_optptr(const struct sk_buff *skb);
int calipso_getattr(const unsigned char *calipso,
		    struct netlbl_lsm_secattr *secattr);
int calipso_skbuff_setattr(struct sk_buff *skb,
			   const struct calipso_doi *doi_def,
			   const struct netlbl_lsm_secattr *secattr);
int calipso_skbuff_delattr(struct sk_buff *skb);
void calipso_cache_invalidate(void);
int calipso_cache_add(const unsigned char *calipso_ptr,
		      const struct netlbl_lsm_secattr *secattr);

#endif
back to top