mirror_zfs/module/zfs/lzjb.c
Richard Yao b8d06fca08 Switch KM_SLEEP to KM_PUSHPAGE
Differences between how paging is done on Solaris and Linux can cause
deadlocks if KM_SLEEP is used in any the following contexts.

  * The txg_sync thread
  * The zvol write/discard threads
  * The zpl_putpage() VFS callback

This is because KM_SLEEP will allow for direct reclaim which may result
in the VM calling back in to the filesystem or block layer to write out
pages.  If a lock is held over this operation the potential exists to
deadlock the system.  To ensure forward progress all memory allocations
in these contexts must us KM_PUSHPAGE which disables performing any I/O
to accomplish the memory allocation.

Previously, this behavior was acheived by setting PF_MEMALLOC on the
thread.  However, that resulted in unexpected side effects such as the
exhaustion of pages in ZONE_DMA.  This approach touchs more of the zfs
code, but it is more consistent with the right way to handle these cases
under Linux.

This is patch lays the ground work for being able to safely revert the
following commits which used PF_MEMALLOC:

  21ade34 Disable direct reclaim for z_wr_* threads
  cfc9a5c Fix zpl_writepage() deadlock
  eec8164 Fix ASSERTION(!dsl_pool_sync_context(tx->tx_pool))

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #726
2012-08-27 12:01:37 -07:00

129 lines
3.8 KiB
C

/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
*/
/*
* We keep our own copy of this algorithm for 3 main reasons:
* 1. If we didn't, anyone modifying common/os/compress.c would
* directly break our on disk format
* 2. Our version of lzjb does not have a number of checks that the
* common/os version needs and uses
* 3. We initialize the lempel to ensure deterministic results,
* so that identical blocks can always be deduplicated.
* In particular, we are adding the "feature" that compress() can
* take a destination buffer size and returns the compressed length, or the
* source length if compression would overflow the destination buffer.
*/
#include <sys/zfs_context.h>
#define MATCH_BITS 6
#define MATCH_MIN 3
#define MATCH_MAX ((1 << MATCH_BITS) + (MATCH_MIN - 1))
#define OFFSET_MASK ((1 << (16 - MATCH_BITS)) - 1)
#define LEMPEL_SIZE 1024
/*ARGSUSED*/
size_t
lzjb_compress(void *s_start, void *d_start, size_t s_len, size_t d_len, int n)
{
uchar_t *src = s_start;
uchar_t *dst = d_start;
uchar_t *cpy, *copymap = NULL;
int copymask = 1 << (NBBY - 1);
int mlen, offset, hash;
uint16_t *hp;
uint16_t *lempel;
lempel = kmem_zalloc(LEMPEL_SIZE * sizeof (uint16_t), KM_PUSHPAGE);
while (src < (uchar_t *)s_start + s_len) {
if ((copymask <<= 1) == (1 << NBBY)) {
if (dst >= (uchar_t *)d_start + d_len - 1 - 2 * NBBY) {
kmem_free(lempel, LEMPEL_SIZE*sizeof(uint16_t));
return (s_len);
}
copymask = 1;
copymap = dst;
*dst++ = 0;
}
if (src > (uchar_t *)s_start + s_len - MATCH_MAX) {
*dst++ = *src++;
continue;
}
hash = (src[0] << 16) + (src[1] << 8) + src[2];
hash += hash >> 9;
hash += hash >> 5;
hp = &lempel[hash & (LEMPEL_SIZE - 1)];
offset = (intptr_t)(src - *hp) & OFFSET_MASK;
*hp = (uint16_t)(uintptr_t)src;
cpy = src - offset;
if (cpy >= (uchar_t *)s_start && cpy != src &&
src[0] == cpy[0] && src[1] == cpy[1] && src[2] == cpy[2]) {
*copymap |= copymask;
for (mlen = MATCH_MIN; mlen < MATCH_MAX; mlen++)
if (src[mlen] != cpy[mlen])
break;
*dst++ = ((mlen - MATCH_MIN) << (NBBY - MATCH_BITS)) |
(offset >> NBBY);
*dst++ = (uchar_t)offset;
src += mlen;
} else {
*dst++ = *src++;
}
}
kmem_free(lempel, LEMPEL_SIZE * sizeof (uint16_t));
return (dst - (uchar_t *)d_start);
}
/*ARGSUSED*/
int
lzjb_decompress(void *s_start, void *d_start, size_t s_len, size_t d_len, int n)
{
uchar_t *src = s_start;
uchar_t *dst = d_start;
uchar_t *d_end = (uchar_t *)d_start + d_len;
uchar_t *cpy, copymap = 0;
int copymask = 1 << (NBBY - 1);
while (dst < d_end) {
if ((copymask <<= 1) == (1 << NBBY)) {
copymask = 1;
copymap = *src++;
}
if (copymap & copymask) {
int mlen = (src[0] >> (NBBY - MATCH_BITS)) + MATCH_MIN;
int offset = ((src[0] << NBBY) | src[1]) & OFFSET_MASK;
src += 2;
if ((cpy = dst - offset) < (uchar_t *)d_start)
return (-1);
while (--mlen >= 0 && dst < d_end)
*dst++ = *cpy++;
} else {
*dst++ = *src++;
}
}
return (0);
}