a4b740db5e
Most implementations of .bdrv_open first open their file child (which is an operation that internally takes the write lock and therefore we shouldn't hold the graph lock while calling it), and afterwards many operations that require holding the graph lock, e.g. for accessing bs->file. This changes block drivers that follow this pattern to take the graph lock after opening the child node. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20231027155333.420094-24-kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
316 lines
9.4 KiB
C
316 lines
9.4 KiB
C
/*
|
|
* QEMU Block driver for CLOOP images
|
|
*
|
|
* Copyright (c) 2004 Johannes E. Schindelin
|
|
*
|
|
* Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
* of this software and associated documentation files (the "Software"), to deal
|
|
* in the Software without restriction, including without limitation the rights
|
|
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
* copies of the Software, and to permit persons to whom the Software is
|
|
* furnished to do so, subject to the following conditions:
|
|
*
|
|
* The above copyright notice and this permission notice shall be included in
|
|
* all copies or substantial portions of the Software.
|
|
*
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
|
* THE SOFTWARE.
|
|
*/
|
|
#include "qemu/osdep.h"
|
|
#include "qapi/error.h"
|
|
#include "qemu/error-report.h"
|
|
#include "block/block-io.h"
|
|
#include "block/block_int.h"
|
|
#include "qemu/module.h"
|
|
#include "qemu/bswap.h"
|
|
#include <zlib.h>
|
|
|
|
/* Maximum compressed block size */
|
|
#define MAX_BLOCK_SIZE (64 * 1024 * 1024)
|
|
|
|
typedef struct BDRVCloopState {
|
|
CoMutex lock;
|
|
uint32_t block_size;
|
|
uint32_t n_blocks;
|
|
uint64_t *offsets;
|
|
uint32_t sectors_per_block;
|
|
uint32_t current_block;
|
|
uint8_t *compressed_block;
|
|
uint8_t *uncompressed_block;
|
|
z_stream zstream;
|
|
} BDRVCloopState;
|
|
|
|
static int cloop_probe(const uint8_t *buf, int buf_size, const char *filename)
|
|
{
|
|
const char *magic_version_2_0 = "#!/bin/sh\n"
|
|
"#V2.0 Format\n"
|
|
"modprobe cloop file=$0 && mount -r -t iso9660 /dev/cloop $1\n";
|
|
int length = strlen(magic_version_2_0);
|
|
if (length > buf_size) {
|
|
length = buf_size;
|
|
}
|
|
if (!memcmp(magic_version_2_0, buf, length)) {
|
|
return 2;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int cloop_open(BlockDriverState *bs, QDict *options, int flags,
|
|
Error **errp)
|
|
{
|
|
BDRVCloopState *s = bs->opaque;
|
|
uint32_t offsets_size, max_compressed_block_size = 1, i;
|
|
int ret;
|
|
|
|
GLOBAL_STATE_CODE();
|
|
|
|
bdrv_graph_rdlock_main_loop();
|
|
ret = bdrv_apply_auto_read_only(bs, NULL, errp);
|
|
bdrv_graph_rdunlock_main_loop();
|
|
if (ret < 0) {
|
|
return ret;
|
|
}
|
|
|
|
ret = bdrv_open_file_child(NULL, options, "file", bs, errp);
|
|
if (ret < 0) {
|
|
return ret;
|
|
}
|
|
|
|
GRAPH_RDLOCK_GUARD_MAINLOOP();
|
|
|
|
/* read header */
|
|
ret = bdrv_pread(bs->file, 128, 4, &s->block_size, 0);
|
|
if (ret < 0) {
|
|
return ret;
|
|
}
|
|
s->block_size = be32_to_cpu(s->block_size);
|
|
if (s->block_size % 512) {
|
|
error_setg(errp, "block_size %" PRIu32 " must be a multiple of 512",
|
|
s->block_size);
|
|
return -EINVAL;
|
|
}
|
|
if (s->block_size == 0) {
|
|
error_setg(errp, "block_size cannot be zero");
|
|
return -EINVAL;
|
|
}
|
|
|
|
/* cloop's create_compressed_fs.c warns about block sizes beyond 256 KB but
|
|
* we can accept more. Prevent ridiculous values like 4 GB - 1 since we
|
|
* need a buffer this big.
|
|
*/
|
|
if (s->block_size > MAX_BLOCK_SIZE) {
|
|
error_setg(errp, "block_size %" PRIu32 " must be %u MB or less",
|
|
s->block_size,
|
|
MAX_BLOCK_SIZE / (1024 * 1024));
|
|
return -EINVAL;
|
|
}
|
|
|
|
ret = bdrv_pread(bs->file, 128 + 4, 4, &s->n_blocks, 0);
|
|
if (ret < 0) {
|
|
return ret;
|
|
}
|
|
s->n_blocks = be32_to_cpu(s->n_blocks);
|
|
|
|
/* read offsets */
|
|
if (s->n_blocks > (UINT32_MAX - 1) / sizeof(uint64_t)) {
|
|
/* Prevent integer overflow */
|
|
error_setg(errp, "n_blocks %" PRIu32 " must be %zu or less",
|
|
s->n_blocks,
|
|
(UINT32_MAX - 1) / sizeof(uint64_t));
|
|
return -EINVAL;
|
|
}
|
|
offsets_size = (s->n_blocks + 1) * sizeof(uint64_t);
|
|
if (offsets_size > 512 * 1024 * 1024) {
|
|
/* Prevent ridiculous offsets_size which causes memory allocation to
|
|
* fail or overflows bdrv_pread() size. In practice the 512 MB
|
|
* offsets[] limit supports 16 TB images at 256 KB block size.
|
|
*/
|
|
error_setg(errp, "image requires too many offsets, "
|
|
"try increasing block size");
|
|
return -EINVAL;
|
|
}
|
|
|
|
s->offsets = g_try_malloc(offsets_size);
|
|
if (s->offsets == NULL) {
|
|
error_setg(errp, "Could not allocate offsets table");
|
|
return -ENOMEM;
|
|
}
|
|
|
|
ret = bdrv_pread(bs->file, 128 + 4 + 4, offsets_size, s->offsets, 0);
|
|
if (ret < 0) {
|
|
goto fail;
|
|
}
|
|
|
|
for (i = 0; i < s->n_blocks + 1; i++) {
|
|
uint64_t size;
|
|
|
|
s->offsets[i] = be64_to_cpu(s->offsets[i]);
|
|
if (i == 0) {
|
|
continue;
|
|
}
|
|
|
|
if (s->offsets[i] < s->offsets[i - 1]) {
|
|
error_setg(errp, "offsets not monotonically increasing at "
|
|
"index %" PRIu32 ", image file is corrupt", i);
|
|
ret = -EINVAL;
|
|
goto fail;
|
|
}
|
|
|
|
size = s->offsets[i] - s->offsets[i - 1];
|
|
|
|
/* Compressed blocks should be smaller than the uncompressed block size
|
|
* but maybe compression performed poorly so the compressed block is
|
|
* actually bigger. Clamp down on unrealistic values to prevent
|
|
* ridiculous s->compressed_block allocation.
|
|
*/
|
|
if (size > 2 * MAX_BLOCK_SIZE) {
|
|
error_setg(errp, "invalid compressed block size at index %" PRIu32
|
|
", image file is corrupt", i);
|
|
ret = -EINVAL;
|
|
goto fail;
|
|
}
|
|
|
|
if (size > max_compressed_block_size) {
|
|
max_compressed_block_size = size;
|
|
}
|
|
}
|
|
|
|
/* initialize zlib engine */
|
|
s->compressed_block = g_try_malloc(max_compressed_block_size + 1);
|
|
if (s->compressed_block == NULL) {
|
|
error_setg(errp, "Could not allocate compressed_block");
|
|
ret = -ENOMEM;
|
|
goto fail;
|
|
}
|
|
|
|
s->uncompressed_block = g_try_malloc(s->block_size);
|
|
if (s->uncompressed_block == NULL) {
|
|
error_setg(errp, "Could not allocate uncompressed_block");
|
|
ret = -ENOMEM;
|
|
goto fail;
|
|
}
|
|
|
|
if (inflateInit(&s->zstream) != Z_OK) {
|
|
ret = -EINVAL;
|
|
goto fail;
|
|
}
|
|
s->current_block = s->n_blocks;
|
|
|
|
s->sectors_per_block = s->block_size/512;
|
|
bs->total_sectors = s->n_blocks * s->sectors_per_block;
|
|
qemu_co_mutex_init(&s->lock);
|
|
return 0;
|
|
|
|
fail:
|
|
g_free(s->offsets);
|
|
g_free(s->compressed_block);
|
|
g_free(s->uncompressed_block);
|
|
return ret;
|
|
}
|
|
|
|
static void cloop_refresh_limits(BlockDriverState *bs, Error **errp)
|
|
{
|
|
bs->bl.request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O */
|
|
}
|
|
|
|
static int coroutine_fn GRAPH_RDLOCK
|
|
cloop_read_block(BlockDriverState *bs, int block_num)
|
|
{
|
|
BDRVCloopState *s = bs->opaque;
|
|
|
|
if (s->current_block != block_num) {
|
|
int ret;
|
|
uint32_t bytes = s->offsets[block_num + 1] - s->offsets[block_num];
|
|
|
|
ret = bdrv_co_pread(bs->file, s->offsets[block_num], bytes,
|
|
s->compressed_block, 0);
|
|
if (ret < 0) {
|
|
return -1;
|
|
}
|
|
|
|
s->zstream.next_in = s->compressed_block;
|
|
s->zstream.avail_in = bytes;
|
|
s->zstream.next_out = s->uncompressed_block;
|
|
s->zstream.avail_out = s->block_size;
|
|
ret = inflateReset(&s->zstream);
|
|
if (ret != Z_OK) {
|
|
return -1;
|
|
}
|
|
ret = inflate(&s->zstream, Z_FINISH);
|
|
if (ret != Z_STREAM_END || s->zstream.total_out != s->block_size) {
|
|
return -1;
|
|
}
|
|
|
|
s->current_block = block_num;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int coroutine_fn GRAPH_RDLOCK
|
|
cloop_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes,
|
|
QEMUIOVector *qiov, BdrvRequestFlags flags)
|
|
{
|
|
BDRVCloopState *s = bs->opaque;
|
|
uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
|
|
int nb_sectors = bytes >> BDRV_SECTOR_BITS;
|
|
int ret, i;
|
|
|
|
assert(QEMU_IS_ALIGNED(offset, BDRV_SECTOR_SIZE));
|
|
assert(QEMU_IS_ALIGNED(bytes, BDRV_SECTOR_SIZE));
|
|
|
|
qemu_co_mutex_lock(&s->lock);
|
|
|
|
for (i = 0; i < nb_sectors; i++) {
|
|
void *data;
|
|
uint32_t sector_offset_in_block =
|
|
((sector_num + i) % s->sectors_per_block),
|
|
block_num = (sector_num + i) / s->sectors_per_block;
|
|
if (cloop_read_block(bs, block_num) != 0) {
|
|
ret = -EIO;
|
|
goto fail;
|
|
}
|
|
|
|
data = s->uncompressed_block + sector_offset_in_block * 512;
|
|
qemu_iovec_from_buf(qiov, i * 512, data, 512);
|
|
}
|
|
|
|
ret = 0;
|
|
fail:
|
|
qemu_co_mutex_unlock(&s->lock);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static void cloop_close(BlockDriverState *bs)
|
|
{
|
|
BDRVCloopState *s = bs->opaque;
|
|
g_free(s->offsets);
|
|
g_free(s->compressed_block);
|
|
g_free(s->uncompressed_block);
|
|
inflateEnd(&s->zstream);
|
|
}
|
|
|
|
static BlockDriver bdrv_cloop = {
|
|
.format_name = "cloop",
|
|
.instance_size = sizeof(BDRVCloopState),
|
|
.bdrv_probe = cloop_probe,
|
|
.bdrv_open = cloop_open,
|
|
.bdrv_child_perm = bdrv_default_perms,
|
|
.bdrv_refresh_limits = cloop_refresh_limits,
|
|
.bdrv_co_preadv = cloop_co_preadv,
|
|
.bdrv_close = cloop_close,
|
|
.is_format = true,
|
|
};
|
|
|
|
static void bdrv_cloop_init(void)
|
|
{
|
|
bdrv_register(&bdrv_cloop);
|
|
}
|
|
|
|
block_init(bdrv_cloop_init);
|