* [Qemu-devel] [PULL 00/26] Block patches
@ 2010-04-23 15:30 Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 01/26] qemu-config: qemu_read_config_file() reads the normal config file Kevin Wolf
` (26 more replies)
0 siblings, 27 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Hi Anthony,
this is the first part of the block patches that accumulated in my block branch
during your absence. I consider these patches ready to be merged into master.
I've not included another 13 patches that might still need discussion/review or
depend on such patches. I'm going to send another pull request for them some
time next week.
Kevin
The following changes since commit 6c557ab975fc8e5edb4167a241266c7c4657054a:
Serge Ziryukin (1):
audio/sdlaudio: remove unused variable
are available in the git repository at:
git://repo.or.cz/qemu/kevin.git for-anthony
Bruce Rogers (1):
Remove un-needed code
Christoph Hellwig (3):
block: get rid of the BDRV_O_FILE flag
block: split raw_getlength
cleanup block driver option handling in vl.c
Kevin Wolf (14):
qemu-config: qemu_read_config_file() reads the normal config file
qemu-config: Make qemu_config_parse more generic
blkdebug: Basic request passthrough
blkdebug: Inject errors
Make qemu-config available for tools
blkdebug: Add events and rules
qcow2: Trigger blkdebug events
qcow2: Fix creation of large images
Replace calls of old bdrv_open
qcow2: Return 0/-errno in write_l2_entries
qcow2: Fix error return code in qcow2_alloc_cluster_link_l2
qcow2: Return 0/-errno in write_l1_entry
qcow2: Return 0/-errno in l2_allocate
block.h: bdrv_create2 doesn't exist any more
Stefan Hajnoczi (8):
block: Do not export bdrv_first
block: Convert bdrv_first to QTAILQ
block: Convert first_drv to QLIST
qemu-img: Eliminate bdrv_new_open() code duplication
qemu-img: Fix BRDV_O_FLAGS typo
linux-aio: Fix typo in read() EINTR check
qcow2: Use QLIST_FOREACH_SAFE macro
block: Free iovec arrays allocated by multiwrite_merge()
Makefile.objs | 6 +-
block-migration.c | 63 ++++---
block.c | 119 +++++++------
block.h | 66 ++++++-
block/blkdebug.c | 475 ++++++++++++++++++++++++++++++++++++++++++++++++
block/qcow2-cluster.c | 78 +++++---
block/qcow2-refcount.c | 18 ++
block/qcow2.c | 63 +++++--
block/raw-posix.c | 65 +++++---
block/vmdk.c | 2 +-
block/vvfat.c | 5 +-
block_int.h | 9 +-
hw/qdev-properties.c | 19 ++-
hw/qdev.h | 1 -
hw/virtio-blk.c | 1 -
hw/xen_disk.c | 2 +-
linux-aio.c | 2 +-
monitor.c | 2 +-
qemu-config.c | 48 +++---
qemu-config.h | 4 +-
qemu-img.c | 93 ++--------
qemu-io.c | 28 ++--
qemu-nbd.c | 2 +-
vl.c | 82 +++------
24 files changed, 902 insertions(+), 351 deletions(-)
create mode 100644 block/blkdebug.c
^ permalink raw reply [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 01/26] qemu-config: qemu_read_config_file() reads the normal config file
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 02/26] qemu-config: Make qemu_config_parse more generic Kevin Wolf
` (25 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Introduce a new function qemu_read_config_file which reads the VM configuration
from a config file. Unlike qemu_config_parse it doesn't take a open file but a
filename and reduces code duplication as a side effect.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qemu-config.c | 15 +++++++++++++++
qemu-config.h | 2 ++
vl.c | 35 +++++++++++------------------------
3 files changed, 28 insertions(+), 24 deletions(-)
diff --git a/qemu-config.c b/qemu-config.c
index d4a2f43..8254b35 100644
--- a/qemu-config.c
+++ b/qemu-config.c
@@ -488,3 +488,18 @@ out:
loc_pop(&loc);
return res;
}
+
+int qemu_read_config_file(const char *filename)
+{
+ FILE *f = fopen(filename, "r");
+ if (f == NULL) {
+ return -errno;
+ }
+
+ if (qemu_config_parse(f, filename) != 0) {
+ return -EINVAL;
+ }
+ fclose(f);
+
+ return 0;
+}
diff --git a/qemu-config.h b/qemu-config.h
index f217c58..07a7a27 100644
--- a/qemu-config.h
+++ b/qemu-config.h
@@ -19,4 +19,6 @@ void qemu_add_globals(void);
void qemu_config_write(FILE *fp);
int qemu_config_parse(FILE *fp, const char *fname);
+int qemu_read_config_file(const char *filename);
+
#endif /* QEMU_CONFIG_H */
diff --git a/vl.c b/vl.c
index e58441b..fe2dd6f 100644
--- a/vl.c
+++ b/vl.c
@@ -2653,25 +2653,16 @@ int main(int argc, char **argv, char **envp)
}
if (defconfig) {
- const char *fname;
- FILE *fp;
+ int ret;
- fname = CONFIG_QEMU_CONFDIR "/qemu.conf";
- fp = fopen(fname, "r");
- if (fp) {
- if (qemu_config_parse(fp, fname) != 0) {
- exit(1);
- }
- fclose(fp);
+ ret = qemu_read_config_file(CONFIG_QEMU_CONFDIR "/qemu.conf");
+ if (ret == -EINVAL) {
+ exit(1);
}
- fname = arch_config_name;
- fp = fopen(fname, "r");
- if (fp) {
- if (qemu_config_parse(fp, fname) != 0) {
- exit(1);
- }
- fclose(fp);
+ ret = qemu_read_config_file(arch_config_name);
+ if (ret == -EINVAL) {
+ exit(1);
}
}
cpudef_init();
@@ -3327,16 +3318,12 @@ int main(int argc, char **argv, char **envp)
break;
case QEMU_OPTION_readconfig:
{
- FILE *fp;
- fp = fopen(optarg, "r");
- if (fp == NULL) {
- fprintf(stderr, "open %s: %s\n", optarg, strerror(errno));
+ int ret = qemu_read_config_file(optarg);
+ if (ret < 0) {
+ fprintf(stderr, "read config %s: %s\n", optarg,
+ strerror(-ret));
exit(1);
}
- if (qemu_config_parse(fp, optarg) != 0) {
- exit(1);
- }
- fclose(fp);
break;
}
case QEMU_OPTION_writeconfig:
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 02/26] qemu-config: Make qemu_config_parse more generic
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 01/26] qemu-config: qemu_read_config_file() reads the normal config file Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 03/26] blkdebug: Basic request passthrough Kevin Wolf
` (24 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
qemu_config_parse gets the option groups as a parameter now instead of
hardcoding the VM configuration groups. This way it can be used for other
configurations, too.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qemu-config.c | 18 ++++++++++++------
qemu-config.h | 2 +-
2 files changed, 13 insertions(+), 7 deletions(-)
diff --git a/qemu-config.c b/qemu-config.c
index 8254b35..9b5fe78 100644
--- a/qemu-config.c
+++ b/qemu-config.c
@@ -296,7 +296,7 @@ QemuOptsList qemu_cpudef_opts = {
},
};
-static QemuOptsList *lists[] = {
+static QemuOptsList *vm_config_groups[] = {
&qemu_drive_opts,
&qemu_chardev_opts,
&qemu_device_opts,
@@ -309,7 +309,7 @@ static QemuOptsList *lists[] = {
NULL,
};
-QemuOptsList *qemu_find_opts(const char *group)
+static QemuOptsList *find_list(QemuOptsList **lists, const char *group)
{
int i;
@@ -323,6 +323,11 @@ QemuOptsList *qemu_find_opts(const char *group)
return lists[i];
}
+QemuOptsList *qemu_find_opts(const char *group)
+{
+ return find_list(vm_config_groups, group);
+}
+
int qemu_set_option(const char *str)
{
char group[64], id[64], arg[64];
@@ -421,6 +426,7 @@ static int config_write_opts(QemuOpts *opts, void *opaque)
void qemu_config_write(FILE *fp)
{
struct ConfigWriteData data = { .fp = fp };
+ QemuOptsList **lists = vm_config_groups;
int i;
fprintf(fp, "# qemu config file\n\n");
@@ -430,7 +436,7 @@ void qemu_config_write(FILE *fp)
}
}
-int qemu_config_parse(FILE *fp, const char *fname)
+int qemu_config_parse(FILE *fp, QemuOptsList **lists, const char *fname)
{
char line[1024], group[64], id[64], arg[64], value[1024];
Location loc;
@@ -451,7 +457,7 @@ int qemu_config_parse(FILE *fp, const char *fname)
}
if (sscanf(line, "[%63s \"%63[^\"]\"]", group, id) == 2) {
/* group with id */
- list = qemu_find_opts(group);
+ list = find_list(lists, group);
if (list == NULL)
goto out;
opts = qemu_opts_create(list, id, 1);
@@ -459,7 +465,7 @@ int qemu_config_parse(FILE *fp, const char *fname)
}
if (sscanf(line, "[%63[^]]]", group) == 1) {
/* group without id */
- list = qemu_find_opts(group);
+ list = find_list(lists, group);
if (list == NULL)
goto out;
opts = qemu_opts_create(list, NULL, 0);
@@ -496,7 +502,7 @@ int qemu_read_config_file(const char *filename)
return -errno;
}
- if (qemu_config_parse(f, filename) != 0) {
+ if (qemu_config_parse(f, vm_config_groups, filename) != 0) {
return -EINVAL;
}
fclose(f);
diff --git a/qemu-config.h b/qemu-config.h
index 07a7a27..dd299c2 100644
--- a/qemu-config.h
+++ b/qemu-config.h
@@ -17,7 +17,7 @@ int qemu_global_option(const char *str);
void qemu_add_globals(void);
void qemu_config_write(FILE *fp);
-int qemu_config_parse(FILE *fp, const char *fname);
+int qemu_config_parse(FILE *fp, QemuOptsList **lists, const char *fname);
int qemu_read_config_file(const char *filename);
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 03/26] blkdebug: Basic request passthrough
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 01/26] qemu-config: qemu_read_config_file() reads the normal config file Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 02/26] qemu-config: Make qemu_config_parse more generic Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 04/26] blkdebug: Inject errors Kevin Wolf
` (23 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
This isn't doing anything interesting. It creates the blkdebug block driver as
a protocol which just passes everything through to raw.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
Makefile.objs | 2 +-
block/blkdebug.c | 104 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 105 insertions(+), 1 deletions(-)
create mode 100644 block/blkdebug.c
diff --git a/Makefile.objs b/Makefile.objs
index 2a87e2c..ed7bbda 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -14,7 +14,7 @@ block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
block-nested-y += cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat.o
block-nested-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o
-block-nested-y += parallels.o nbd.o
+block-nested-y += parallels.o nbd.o blkdebug.o
block-nested-$(CONFIG_WIN32) += raw-win32.o
block-nested-$(CONFIG_POSIX) += raw-posix.o
block-nested-$(CONFIG_CURL) += curl.o
diff --git a/block/blkdebug.c b/block/blkdebug.c
new file mode 100644
index 0000000..2c7e0dd
--- /dev/null
+++ b/block/blkdebug.c
@@ -0,0 +1,104 @@
+/*
+ * Block protocol for I/O error injection
+ *
+ * Copyright (c) 2010 Kevin Wolf <kwolf@redhat.com>
+ *
+ * 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-common.h"
+#include "block_int.h"
+#include "module.h"
+
+typedef struct BDRVBlkdebugState {
+ BlockDriverState *hd;
+} BDRVBlkdebugState;
+
+static int blkdebug_open(BlockDriverState *bs, const char *filename, int flags)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+
+ if (strncmp(filename, "blkdebug:", strlen("blkdebug:"))) {
+ return -EINVAL;
+ }
+ filename += strlen("blkdebug:");
+
+ return bdrv_file_open(&s->hd, filename, flags);
+}
+
+static BlockDriverAIOCB *blkdebug_aio_readv(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockDriverCompletionFunc *cb, void *opaque)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ BlockDriverAIOCB *acb =
+ bdrv_aio_readv(s->hd, sector_num, qiov, nb_sectors, cb, opaque);
+ return acb;
+}
+
+static BlockDriverAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockDriverCompletionFunc *cb, void *opaque)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ BlockDriverAIOCB *acb =
+ bdrv_aio_writev(s->hd, sector_num, qiov, nb_sectors, cb, opaque);
+ return acb;
+}
+
+static void blkdebug_close(BlockDriverState *bs)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ bdrv_delete(s->hd);
+}
+
+static void blkdebug_flush(BlockDriverState *bs)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ bdrv_flush(s->hd);
+}
+
+static BlockDriverAIOCB *blkdebug_aio_flush(BlockDriverState *bs,
+ BlockDriverCompletionFunc *cb, void *opaque)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ return bdrv_aio_flush(s->hd, cb, opaque);
+}
+
+static BlockDriver bdrv_blkdebug = {
+ .format_name = "blkdebug",
+ .protocol_name = "blkdebug",
+
+ .instance_size = sizeof(BDRVBlkdebugState),
+
+ .bdrv_open = blkdebug_open,
+ .bdrv_close = blkdebug_close,
+ .bdrv_flush = blkdebug_flush,
+
+ .bdrv_aio_readv = blkdebug_aio_readv,
+ .bdrv_aio_writev = blkdebug_aio_writev,
+ .bdrv_aio_flush = blkdebug_aio_flush,
+};
+
+static void bdrv_blkdebug_init(void)
+{
+ bdrv_register(&bdrv_blkdebug);
+}
+
+block_init(bdrv_blkdebug_init);
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 04/26] blkdebug: Inject errors
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (2 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 03/26] blkdebug: Basic request passthrough Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 05/26] Make qemu-config available for tools Kevin Wolf
` (22 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Add a mechanism to inject errors instead of passing requests on. With no
further patches applied, you can use it by setting inject_errno in gdb.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/blkdebug.c | 81 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 81 insertions(+), 0 deletions(-)
diff --git a/block/blkdebug.c b/block/blkdebug.c
index 2c7e0dd..dad3ac6 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -26,10 +26,41 @@
#include "block_int.h"
#include "module.h"
+#include <stdbool.h>
+
+typedef struct BlkdebugVars {
+ int state;
+
+ /* If inject_errno != 0, an error is injected for requests */
+ int inject_errno;
+
+ /* Decides if all future requests fail (false) or only the next one and
+ * after the next request inject_errno is reset to 0 (true) */
+ bool inject_once;
+
+ /* Decides if aio_readv/writev fails right away (true) or returns an error
+ * return value only in the callback (false) */
+ bool inject_immediately;
+} BlkdebugVars;
+
typedef struct BDRVBlkdebugState {
BlockDriverState *hd;
+ BlkdebugVars vars;
} BDRVBlkdebugState;
+typedef struct BlkdebugAIOCB {
+ BlockDriverAIOCB common;
+ QEMUBH *bh;
+ int ret;
+} BlkdebugAIOCB;
+
+static void blkdebug_aio_cancel(BlockDriverAIOCB *blockacb);
+
+static AIOPool blkdebug_aio_pool = {
+ .aiocb_size = sizeof(BlkdebugAIOCB),
+ .cancel = blkdebug_aio_cancel,
+};
+
static int blkdebug_open(BlockDriverState *bs, const char *filename, int flags)
{
BDRVBlkdebugState *s = bs->opaque;
@@ -42,11 +73,56 @@ static int blkdebug_open(BlockDriverState *bs, const char *filename, int flags)
return bdrv_file_open(&s->hd, filename, flags);
}
+static void error_callback_bh(void *opaque)
+{
+ struct BlkdebugAIOCB *acb = opaque;
+ qemu_bh_delete(acb->bh);
+ acb->common.cb(acb->common.opaque, acb->ret);
+ qemu_aio_release(acb);
+}
+
+static void blkdebug_aio_cancel(BlockDriverAIOCB *blockacb)
+{
+ BlkdebugAIOCB *acb = (BlkdebugAIOCB*) blockacb;
+ qemu_aio_release(acb);
+}
+
+static BlockDriverAIOCB *inject_error(BlockDriverState *bs,
+ BlockDriverCompletionFunc *cb, void *opaque)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ int error = s->vars.inject_errno;
+ struct BlkdebugAIOCB *acb;
+ QEMUBH *bh;
+
+ if (s->vars.inject_once) {
+ s->vars.inject_errno = 0;
+ }
+
+ if (s->vars.inject_immediately) {
+ return NULL;
+ }
+
+ acb = qemu_aio_get(&blkdebug_aio_pool, bs, cb, opaque);
+ acb->ret = -error;
+
+ bh = qemu_bh_new(error_callback_bh, acb);
+ acb->bh = bh;
+ qemu_bh_schedule(bh);
+
+ return (BlockDriverAIOCB*) acb;
+}
+
static BlockDriverAIOCB *blkdebug_aio_readv(BlockDriverState *bs,
int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
BlockDriverCompletionFunc *cb, void *opaque)
{
BDRVBlkdebugState *s = bs->opaque;
+
+ if (s->vars.inject_errno) {
+ return inject_error(bs, cb, opaque);
+ }
+
BlockDriverAIOCB *acb =
bdrv_aio_readv(s->hd, sector_num, qiov, nb_sectors, cb, opaque);
return acb;
@@ -57,6 +133,11 @@ static BlockDriverAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
BlockDriverCompletionFunc *cb, void *opaque)
{
BDRVBlkdebugState *s = bs->opaque;
+
+ if (s->vars.inject_errno) {
+ return inject_error(bs, cb, opaque);
+ }
+
BlockDriverAIOCB *acb =
bdrv_aio_writev(s->hd, sector_num, qiov, nb_sectors, cb, opaque);
return acb;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 05/26] Make qemu-config available for tools
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (3 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 04/26] blkdebug: Inject errors Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 06/26] blkdebug: Add events and rules Kevin Wolf
` (21 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
To be able to use config files for blkdebug, we need to make these functions
available in the tools. This involves moving two functions that can only be
built in the context of the emulator.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
Makefile.objs | 4 ++--
hw/qdev-properties.c | 19 ++++++++++++++++++-
hw/qdev.h | 1 -
qemu-config.c | 17 -----------------
4 files changed, 20 insertions(+), 21 deletions(-)
diff --git a/Makefile.objs b/Makefile.objs
index ed7bbda..69d6879 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -8,7 +8,7 @@ qobject-obj-y += qerror.o
# block-obj-y is code used by both qemu system emulation and qemu-img
block-obj-y = cutils.o cache-utils.o qemu-malloc.o qemu-option.o module.o
-block-obj-y += nbd.o block.o aio.o aes.o osdep.o
+block-obj-y += nbd.o block.o aio.o aes.o osdep.o qemu-config.o
block-obj-$(CONFIG_POSIX) += posix-aio-compat.o
block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
@@ -76,7 +76,7 @@ common-obj-y += buffered_file.o migration.o migration-tcp.o qemu-sockets.o
common-obj-y += qemu-char.o savevm.o #aio.o
common-obj-y += msmouse.o ps2.o
common-obj-y += qdev.o qdev-properties.o
-common-obj-y += qemu-config.o block-migration.o
+common-obj-y += block-migration.o
common-obj-$(CONFIG_BRLAPI) += baum.o
common-obj-$(CONFIG_POSIX) += migration-exec.o migration-unix.o migration-fd.o
diff --git a/hw/qdev-properties.c b/hw/qdev-properties.c
index 157a111..9ffdba7 100644
--- a/hw/qdev-properties.c
+++ b/hw/qdev-properties.c
@@ -661,7 +661,7 @@ void qdev_prop_set_defaults(DeviceState *dev, Property *props)
static QTAILQ_HEAD(, GlobalProperty) global_props = QTAILQ_HEAD_INITIALIZER(global_props);
-void qdev_prop_register_global(GlobalProperty *prop)
+static void qdev_prop_register_global(GlobalProperty *prop)
{
QTAILQ_INSERT_TAIL(&global_props, prop, next);
}
@@ -689,3 +689,20 @@ void qdev_prop_set_globals(DeviceState *dev)
}
}
}
+
+static int qdev_add_one_global(QemuOpts *opts, void *opaque)
+{
+ GlobalProperty *g;
+
+ g = qemu_mallocz(sizeof(*g));
+ g->driver = qemu_opt_get(opts, "driver");
+ g->property = qemu_opt_get(opts, "property");
+ g->value = qemu_opt_get(opts, "value");
+ qdev_prop_register_global(g);
+ return 0;
+}
+
+void qemu_add_globals(void)
+{
+ qemu_opts_foreach(&qemu_global_opts, qdev_add_one_global, NULL, 0);
+}
diff --git a/hw/qdev.h b/hw/qdev.h
index 40373c8..d8fbc73 100644
--- a/hw/qdev.h
+++ b/hw/qdev.h
@@ -273,7 +273,6 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char *name, uint8_t *value);
void qdev_prop_set_ptr(DeviceState *dev, const char *name, void *value);
void qdev_prop_set_defaults(DeviceState *dev, Property *props);
-void qdev_prop_register_global(GlobalProperty *prop);
void qdev_prop_register_global_list(GlobalProperty *props);
void qdev_prop_set_globals(DeviceState *dev);
diff --git a/qemu-config.c b/qemu-config.c
index 9b5fe78..9f0f143 100644
--- a/qemu-config.c
+++ b/qemu-config.c
@@ -378,23 +378,6 @@ int qemu_global_option(const char *str)
return 0;
}
-static int qemu_add_one_global(QemuOpts *opts, void *opaque)
-{
- GlobalProperty *g;
-
- g = qemu_mallocz(sizeof(*g));
- g->driver = qemu_opt_get(opts, "driver");
- g->property = qemu_opt_get(opts, "property");
- g->value = qemu_opt_get(opts, "value");
- qdev_prop_register_global(g);
- return 0;
-}
-
-void qemu_add_globals(void)
-{
- qemu_opts_foreach(&qemu_global_opts, qemu_add_one_global, NULL, 0);
-}
-
struct ConfigWriteData {
QemuOptsList *list;
FILE *fp;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 06/26] blkdebug: Add events and rules
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (4 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 05/26] Make qemu-config available for tools Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 07/26] qcow2: Trigger blkdebug events Kevin Wolf
` (20 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Block drivers can trigger a blkdebug event whenever they reach a place where it
could be useful to inject an error for testing/debugging purposes.
Rules are read from a blkdebug config file and describe which action is taken
when an event is triggered. For now this is only injecting an error (with a few
options) or changing the state (which is an integer). Rules can be declared to
be active only in a specific state; this way later rules can distiguish on
which path we came to trigger their event.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 12 +++
block.h | 9 ++
block/blkdebug.c | 250 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
block_int.h | 2 +
4 files changed, 272 insertions(+), 1 deletions(-)
diff --git a/block.c b/block.c
index 0881c93..9351aa3 100644
--- a/block.c
+++ b/block.c
@@ -1535,6 +1535,18 @@ int bdrv_load_vmstate(BlockDriverState *bs, uint8_t *buf,
return drv->bdrv_load_vmstate(bs, buf, pos, size);
}
+void bdrv_debug_event(BlockDriverState *bs, BlkDebugEvent event)
+{
+ BlockDriver *drv = bs->drv;
+
+ if (!drv || !drv->bdrv_debug_event) {
+ return;
+ }
+
+ return drv->bdrv_debug_event(bs, event);
+
+}
+
/**************************************************************/
/* handling of snapshots */
diff --git a/block.h b/block.h
index edf5704..2fd8361 100644
--- a/block.h
+++ b/block.h
@@ -207,4 +207,13 @@ int bdrv_get_dirty(BlockDriverState *bs, int64_t sector);
void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector,
int nr_sectors);
int64_t bdrv_get_dirty_count(BlockDriverState *bs);
+
+
+typedef enum {
+ BLKDBG_EVENT_MAX,
+} BlkDebugEvent;
+
+#define BLKDBG_EVENT(bs, evt) bdrv_debug_event(bs, evt)
+void bdrv_debug_event(BlockDriverState *bs, BlkDebugEvent event);
+
#endif
diff --git a/block/blkdebug.c b/block/blkdebug.c
index dad3ac6..b813bfa 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -46,6 +46,7 @@ typedef struct BlkdebugVars {
typedef struct BDRVBlkdebugState {
BlockDriverState *hd;
BlkdebugVars vars;
+ QLIST_HEAD(list, BlkdebugRule) rules[BLKDBG_EVENT_MAX];
} BDRVBlkdebugState;
typedef struct BlkdebugAIOCB {
@@ -61,16 +62,211 @@ static AIOPool blkdebug_aio_pool = {
.cancel = blkdebug_aio_cancel,
};
+enum {
+ ACTION_INJECT_ERROR,
+ ACTION_SET_STATE,
+};
+
+typedef struct BlkdebugRule {
+ BlkDebugEvent event;
+ int action;
+ int state;
+ union {
+ struct {
+ int error;
+ int immediately;
+ int once;
+ } inject;
+ struct {
+ int new_state;
+ } set_state;
+ } options;
+ QLIST_ENTRY(BlkdebugRule) next;
+} BlkdebugRule;
+
+static QemuOptsList inject_error_opts = {
+ .name = "inject-error",
+ .head = QTAILQ_HEAD_INITIALIZER(inject_error_opts.head),
+ .desc = {
+ {
+ .name = "event",
+ .type = QEMU_OPT_STRING,
+ },
+ {
+ .name = "state",
+ .type = QEMU_OPT_NUMBER,
+ },
+ {
+ .name = "errno",
+ .type = QEMU_OPT_NUMBER,
+ },
+ {
+ .name = "once",
+ .type = QEMU_OPT_BOOL,
+ },
+ {
+ .name = "immediately",
+ .type = QEMU_OPT_BOOL,
+ },
+ { /* end of list */ }
+ },
+};
+
+static QemuOptsList set_state_opts = {
+ .name = "set-state",
+ .head = QTAILQ_HEAD_INITIALIZER(inject_error_opts.head),
+ .desc = {
+ {
+ .name = "event",
+ .type = QEMU_OPT_STRING,
+ },
+ {
+ .name = "state",
+ .type = QEMU_OPT_NUMBER,
+ },
+ {
+ .name = "new_state",
+ .type = QEMU_OPT_NUMBER,
+ },
+ { /* end of list */ }
+ },
+};
+
+static QemuOptsList *config_groups[] = {
+ &inject_error_opts,
+ &set_state_opts,
+ NULL
+};
+
+static const char *event_names[BLKDBG_EVENT_MAX] = {
+};
+
+static int get_event_by_name(const char *name, BlkDebugEvent *event)
+{
+ int i;
+
+ for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
+ if (!strcmp(event_names[i], name)) {
+ *event = i;
+ return 0;
+ }
+ }
+
+ return -1;
+}
+
+struct add_rule_data {
+ BDRVBlkdebugState *s;
+ int action;
+};
+
+static int add_rule(QemuOpts *opts, void *opaque)
+{
+ struct add_rule_data *d = opaque;
+ BDRVBlkdebugState *s = d->s;
+ const char* event_name;
+ BlkDebugEvent event;
+ struct BlkdebugRule *rule;
+
+ /* Find the right event for the rule */
+ event_name = qemu_opt_get(opts, "event");
+ if (!event_name || get_event_by_name(event_name, &event) < 0) {
+ return -1;
+ }
+
+ /* Set attributes common for all actions */
+ rule = qemu_mallocz(sizeof(*rule));
+ *rule = (struct BlkdebugRule) {
+ .event = event,
+ .action = d->action,
+ .state = qemu_opt_get_number(opts, "state", 0),
+ };
+
+ /* Parse action-specific options */
+ switch (d->action) {
+ case ACTION_INJECT_ERROR:
+ rule->options.inject.error = qemu_opt_get_number(opts, "errno", EIO);
+ rule->options.inject.once = qemu_opt_get_bool(opts, "once", 0);
+ rule->options.inject.immediately =
+ qemu_opt_get_bool(opts, "immediately", 0);
+ break;
+
+ case ACTION_SET_STATE:
+ rule->options.set_state.new_state =
+ qemu_opt_get_number(opts, "new_state", 0);
+ break;
+ };
+
+ /* Add the rule */
+ QLIST_INSERT_HEAD(&s->rules[event], rule, next);
+
+ return 0;
+}
+
+static int read_config(BDRVBlkdebugState *s, const char *filename)
+{
+ FILE *f;
+ int ret;
+ struct add_rule_data d;
+
+ f = fopen(filename, "r");
+ if (f == NULL) {
+ return -errno;
+ }
+
+ ret = qemu_config_parse(f, config_groups, filename);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ d.s = s;
+ d.action = ACTION_INJECT_ERROR;
+ qemu_opts_foreach(&inject_error_opts, add_rule, &d, 0);
+
+ d.action = ACTION_SET_STATE;
+ qemu_opts_foreach(&set_state_opts, add_rule, &d, 0);
+
+ ret = 0;
+fail:
+ fclose(f);
+ return ret;
+}
+
+/* Valid blkdebug filenames look like blkdebug:path/to/config:path/to/image */
static int blkdebug_open(BlockDriverState *bs, const char *filename, int flags)
{
BDRVBlkdebugState *s = bs->opaque;
+ int ret;
+ char *config, *c;
+ /* Parse the blkdebug: prefix */
if (strncmp(filename, "blkdebug:", strlen("blkdebug:"))) {
return -EINVAL;
}
filename += strlen("blkdebug:");
- return bdrv_file_open(&s->hd, filename, flags);
+ /* Read rules from config file */
+ c = strchr(filename, ':');
+ if (c == NULL) {
+ return -EINVAL;
+ }
+
+ config = strdup(filename);
+ config[c - filename] = '\0';
+ ret = read_config(s, config);
+ free(config);
+ if (ret < 0) {
+ return ret;
+ }
+ filename = c + 1;
+
+ /* Open the backing file */
+ ret = bdrv_file_open(&s->hd, filename, flags);
+ if (ret < 0) {
+ return ret;
+ }
+
+ return 0;
}
static void error_callback_bh(void *opaque)
@@ -146,6 +342,16 @@ static BlockDriverAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
static void blkdebug_close(BlockDriverState *bs)
{
BDRVBlkdebugState *s = bs->opaque;
+ BlkdebugRule *rule, *next;
+ int i;
+
+ for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
+ QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
+ QLIST_REMOVE(rule, next);
+ qemu_free(rule);
+ }
+ }
+
bdrv_delete(s->hd);
}
@@ -162,6 +368,46 @@ static BlockDriverAIOCB *blkdebug_aio_flush(BlockDriverState *bs,
return bdrv_aio_flush(s->hd, cb, opaque);
}
+static void process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
+ BlkdebugVars *old_vars)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ BlkdebugVars *vars = &s->vars;
+
+ /* Only process rules for the current state */
+ if (rule->state && rule->state != old_vars->state) {
+ return;
+ }
+
+ /* Take the action */
+ switch (rule->action) {
+ case ACTION_INJECT_ERROR:
+ vars->inject_errno = rule->options.inject.error;
+ vars->inject_once = rule->options.inject.once;
+ vars->inject_immediately = rule->options.inject.immediately;
+ break;
+
+ case ACTION_SET_STATE:
+ vars->state = rule->options.set_state.new_state;
+ break;
+ }
+}
+
+static void blkdebug_debug_event(BlockDriverState *bs, BlkDebugEvent event)
+{
+ BDRVBlkdebugState *s = bs->opaque;
+ struct BlkdebugRule *rule;
+ BlkdebugVars old_vars = s->vars;
+
+ if (event < 0 || event >= BLKDBG_EVENT_MAX) {
+ return;
+ }
+
+ QLIST_FOREACH(rule, &s->rules[event], next) {
+ process_rule(bs, rule, &old_vars);
+ }
+}
+
static BlockDriver bdrv_blkdebug = {
.format_name = "blkdebug",
.protocol_name = "blkdebug",
@@ -175,6 +421,8 @@ static BlockDriver bdrv_blkdebug = {
.bdrv_aio_readv = blkdebug_aio_readv,
.bdrv_aio_writev = blkdebug_aio_writev,
.bdrv_aio_flush = blkdebug_aio_flush,
+
+ .bdrv_debug_event = blkdebug_debug_event,
};
static void bdrv_blkdebug_init(void)
diff --git a/block_int.h b/block_int.h
index 71b633b..e7e1e7e 100644
--- a/block_int.h
+++ b/block_int.h
@@ -120,6 +120,8 @@ struct BlockDriver {
/* Returns number of errors in image, -errno for internal errors */
int (*bdrv_check)(BlockDriverState* bs);
+ void (*bdrv_debug_event)(BlockDriverState *bs, BlkDebugEvent event);
+
/* Set if newly created images are not guaranteed to contain only zeros */
int no_zero_init;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 07/26] qcow2: Trigger blkdebug events
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (5 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 06/26] blkdebug: Add events and rules Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 08/26] qcow2: Fix creation of large images Kevin Wolf
` (19 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
This adds blkdebug events to qcow2 to allow injecting I/O errors in specific
places.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.h | 44 ++++++++++++++++++++++++++++++++++++++++++++
block/blkdebug.c | 42 ++++++++++++++++++++++++++++++++++++++++++
block/qcow2-cluster.c | 15 +++++++++++++++
block/qcow2-refcount.c | 18 ++++++++++++++++++
block/qcow2.c | 6 ++++++
5 files changed, 125 insertions(+), 0 deletions(-)
diff --git a/block.h b/block.h
index 2fd8361..14443f1 100644
--- a/block.h
+++ b/block.h
@@ -210,6 +210,50 @@ int64_t bdrv_get_dirty_count(BlockDriverState *bs);
typedef enum {
+ BLKDBG_L1_UPDATE,
+
+ BLKDBG_L1_GROW_ALLOC_TABLE,
+ BLKDBG_L1_GROW_WRITE_TABLE,
+ BLKDBG_L1_GROW_ACTIVATE_TABLE,
+
+ BLKDBG_L2_LOAD,
+ BLKDBG_L2_UPDATE,
+ BLKDBG_L2_UPDATE_COMPRESSED,
+ BLKDBG_L2_ALLOC_COW_READ,
+ BLKDBG_L2_ALLOC_WRITE,
+
+ BLKDBG_READ,
+ BLKDBG_READ_AIO,
+ BLKDBG_READ_BACKING,
+ BLKDBG_READ_BACKING_AIO,
+ BLKDBG_READ_COMPRESSED,
+
+ BLKDBG_WRITE_AIO,
+ BLKDBG_WRITE_COMPRESSED,
+
+ BLKDBG_VMSTATE_LOAD,
+ BLKDBG_VMSTATE_SAVE,
+
+ BLKDBG_COW_READ,
+ BLKDBG_COW_WRITE,
+
+ BLKDBG_REFTABLE_LOAD,
+ BLKDBG_REFTABLE_GROW,
+
+ BLKDBG_REFBLOCK_LOAD,
+ BLKDBG_REFBLOCK_UPDATE,
+ BLKDBG_REFBLOCK_UPDATE_PART,
+ BLKDBG_REFBLOCK_ALLOC,
+ BLKDBG_REFBLOCK_ALLOC_HOOKUP,
+ BLKDBG_REFBLOCK_ALLOC_WRITE,
+ BLKDBG_REFBLOCK_ALLOC_WRITE_BLOCKS,
+ BLKDBG_REFBLOCK_ALLOC_WRITE_TABLE,
+ BLKDBG_REFBLOCK_ALLOC_SWITCH_TABLE,
+
+ BLKDBG_CLUSTER_ALLOC,
+ BLKDBG_CLUSTER_ALLOC_BYTES,
+ BLKDBG_CLUSTER_FREE,
+
BLKDBG_EVENT_MAX,
} BlkDebugEvent;
diff --git a/block/blkdebug.c b/block/blkdebug.c
index b813bfa..643c397 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -139,6 +139,48 @@ static QemuOptsList *config_groups[] = {
};
static const char *event_names[BLKDBG_EVENT_MAX] = {
+ [BLKDBG_L1_UPDATE] = "l1_update",
+ [BLKDBG_L1_GROW_ALLOC_TABLE] = "l1_grow.alloc_table",
+ [BLKDBG_L1_GROW_WRITE_TABLE] = "l1_grow.write_table",
+ [BLKDBG_L1_GROW_ACTIVATE_TABLE] = "l1_grow.activate_table",
+
+ [BLKDBG_L2_LOAD] = "l2_load",
+ [BLKDBG_L2_UPDATE] = "l2_update",
+ [BLKDBG_L2_UPDATE_COMPRESSED] = "l2_update_compressed",
+ [BLKDBG_L2_ALLOC_COW_READ] = "l2_alloc.cow_read",
+ [BLKDBG_L2_ALLOC_WRITE] = "l2_alloc.write",
+
+ [BLKDBG_READ] = "read",
+ [BLKDBG_READ_AIO] = "read_aio",
+ [BLKDBG_READ_BACKING] = "read_backing",
+ [BLKDBG_READ_BACKING_AIO] = "read_backing_aio",
+ [BLKDBG_READ_COMPRESSED] = "read_compressed",
+
+ [BLKDBG_WRITE_AIO] = "write_aio",
+ [BLKDBG_WRITE_COMPRESSED] = "write_compressed",
+
+ [BLKDBG_VMSTATE_LOAD] = "vmstate_load",
+ [BLKDBG_VMSTATE_SAVE] = "vmstate_save",
+
+ [BLKDBG_COW_READ] = "cow_read",
+ [BLKDBG_COW_WRITE] = "cow_write",
+
+ [BLKDBG_REFTABLE_LOAD] = "reftable_load",
+ [BLKDBG_REFTABLE_GROW] = "reftable_grow",
+
+ [BLKDBG_REFBLOCK_LOAD] = "refblock_load",
+ [BLKDBG_REFBLOCK_UPDATE] = "refblock_update",
+ [BLKDBG_REFBLOCK_UPDATE_PART] = "refblock_update_part",
+ [BLKDBG_REFBLOCK_ALLOC] = "refblock_alloc",
+ [BLKDBG_REFBLOCK_ALLOC_HOOKUP] = "refblock_alloc.hookup",
+ [BLKDBG_REFBLOCK_ALLOC_WRITE] = "refblock_alloc.write",
+ [BLKDBG_REFBLOCK_ALLOC_WRITE_BLOCKS] = "refblock_alloc.write_blocks",
+ [BLKDBG_REFBLOCK_ALLOC_WRITE_TABLE] = "refblock_alloc.write_table",
+ [BLKDBG_REFBLOCK_ALLOC_SWITCH_TABLE] = "refblock_alloc.switch_table",
+
+ [BLKDBG_CLUSTER_ALLOC] = "cluster_alloc",
+ [BLKDBG_CLUSTER_ALLOC_BYTES] = "cluster_alloc_bytes",
+ [BLKDBG_CLUSTER_FREE] = "cluster_free",
};
static int get_event_by_name(const char *name, BlkDebugEvent *event)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index c7057b1..8cb4b38 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -54,12 +54,14 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
memcpy(new_l1_table, s->l1_table, s->l1_size * sizeof(uint64_t));
/* write new table (align to cluster) */
+ BLKDBG_EVENT(s->hd, BLKDBG_L1_GROW_ALLOC_TABLE);
new_l1_table_offset = qcow2_alloc_clusters(bs, new_l1_size2);
if (new_l1_table_offset < 0) {
qemu_free(new_l1_table);
return new_l1_table_offset;
}
+ BLKDBG_EVENT(s->hd, BLKDBG_L1_GROW_WRITE_TABLE);
for(i = 0; i < s->l1_size; i++)
new_l1_table[i] = cpu_to_be64(new_l1_table[i]);
ret = bdrv_pwrite(s->hd, new_l1_table_offset, new_l1_table, new_l1_size2);
@@ -69,6 +71,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
new_l1_table[i] = be64_to_cpu(new_l1_table[i]);
/* set new table */
+ BLKDBG_EVENT(s->hd, BLKDBG_L1_GROW_ACTIVATE_TABLE);
cpu_to_be32w((uint32_t*)data, new_l1_size);
cpu_to_be64w((uint64_t*)(data + 4), new_l1_table_offset);
ret = bdrv_pwrite(s->hd, offsetof(QCowHeader, l1_size), data,sizeof(data));
@@ -170,6 +173,8 @@ static uint64_t *l2_load(BlockDriverState *bs, uint64_t l2_offset)
min_index = l2_cache_new_entry(bs);
l2_table = s->l2_cache + (min_index << s->l2_bits);
+
+ BLKDBG_EVENT(s->hd, BLKDBG_L2_LOAD);
if (bdrv_pread(s->hd, l2_offset, l2_table, s->l2_size * sizeof(uint64_t)) !=
s->l2_size * sizeof(uint64_t))
return NULL;
@@ -195,6 +200,7 @@ static int write_l1_entry(BDRVQcowState *s, int l1_index)
buf[i] = cpu_to_be64(s->l1_table[l1_start_index + i]);
}
+ BLKDBG_EVENT(s->hd, BLKDBG_L1_UPDATE);
if (bdrv_pwrite(s->hd, s->l1_table_offset + 8 * l1_start_index,
buf, sizeof(buf)) != sizeof(buf))
{
@@ -248,12 +254,14 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
memset(l2_table, 0, s->l2_size * sizeof(uint64_t));
} else {
/* if there was an old l2 table, read it from the disk */
+ BLKDBG_EVENT(s->hd, BLKDBG_L2_ALLOC_COW_READ);
if (bdrv_pread(s->hd, old_l2_offset,
l2_table, s->l2_size * sizeof(uint64_t)) !=
s->l2_size * sizeof(uint64_t))
return NULL;
}
/* write the l2 table to the file */
+ BLKDBG_EVENT(s->hd, BLKDBG_L2_ALLOC_WRITE);
if (bdrv_pwrite(s->hd, l2_offset,
l2_table, s->l2_size * sizeof(uint64_t)) !=
s->l2_size * sizeof(uint64_t))
@@ -335,6 +343,7 @@ static int qcow_read(BlockDriverState *bs, int64_t sector_num,
/* read from the base image */
n1 = qcow2_backing_read1(bs->backing_hd, sector_num, buf, n);
if (n1 > 0) {
+ BLKDBG_EVENT(s->hd, BLKDBG_READ_BACKING);
ret = bdrv_read(bs->backing_hd, sector_num, buf, n1);
if (ret < 0)
return -1;
@@ -347,6 +356,7 @@ static int qcow_read(BlockDriverState *bs, int64_t sector_num,
return -1;
memcpy(buf, s->cluster_cache + index_in_cluster * 512, 512 * n);
} else {
+ BLKDBG_EVENT(s->hd, BLKDBG_READ);
ret = bdrv_pread(s->hd, cluster_offset + index_in_cluster * 512, buf, n * 512);
if (ret != n * 512)
return -1;
@@ -371,6 +381,7 @@ static int copy_sectors(BlockDriverState *bs, uint64_t start_sect,
n = n_end - n_start;
if (n <= 0)
return 0;
+ BLKDBG_EVENT(s->hd, BLKDBG_COW_READ);
ret = qcow_read(bs, start_sect + n_start, s->cluster_data, n);
if (ret < 0)
return ret;
@@ -380,6 +391,7 @@ static int copy_sectors(BlockDriverState *bs, uint64_t start_sect,
s->cluster_data, n, 1,
&s->aes_encrypt_key);
}
+ BLKDBG_EVENT(s->hd, BLKDBG_COW_WRITE);
ret = bdrv_write(s->hd, (cluster_offset >> 9) + n_start,
s->cluster_data, n);
if (ret < 0)
@@ -592,6 +604,7 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
/* compressed clusters never have the copied flag */
+ BLKDBG_EVENT(s->hd, BLKDBG_L2_UPDATE_COMPRESSED);
l2_table[l2_index] = cpu_to_be64(cluster_offset);
if (bdrv_pwrite(s->hd,
l2_offset + l2_index * sizeof(uint64_t),
@@ -615,6 +628,7 @@ static int write_l2_entries(BDRVQcowState *s, uint64_t *l2_table,
int end_offset = (8 * (l2_index + num) + 511) & ~511;
size_t len = end_offset - start_offset;
+ BLKDBG_EVENT(s->hd, BLKDBG_L2_UPDATE);
if (bdrv_pwrite(s->hd, l2_offset + start_offset, &l2_table[l2_start_index],
len) != len)
{
@@ -866,6 +880,7 @@ int qcow2_decompress_cluster(BDRVQcowState *s, uint64_t cluster_offset)
nb_csectors = ((cluster_offset >> s->csize_shift) & s->csize_mask) + 1;
sector_offset = coffset & 511;
csize = nb_csectors * 512 - sector_offset;
+ BLKDBG_EVENT(s->hd, BLKDBG_READ_COMPRESSED);
ret = bdrv_read(s->hd, coffset >> 9, s->cluster_data, nb_csectors);
if (ret < 0) {
return -1;
diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 917fc88..47c9978 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -42,6 +42,7 @@ static int write_refcount_block(BDRVQcowState *s)
return 0;
}
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_UPDATE);
if (bdrv_pwrite(s->hd, s->refcount_block_cache_offset,
s->refcount_block_cache, size) != size)
{
@@ -63,6 +64,7 @@ int qcow2_refcount_init(BlockDriverState *bs)
refcount_table_size2 = s->refcount_table_size * sizeof(uint64_t);
s->refcount_table = qemu_malloc(refcount_table_size2);
if (s->refcount_table_size > 0) {
+ BLKDBG_EVENT(s->hd, BLKDBG_REFTABLE_LOAD);
ret = bdrv_pread(s->hd, s->refcount_table_offset,
s->refcount_table, refcount_table_size2);
if (ret != refcount_table_size2)
@@ -93,6 +95,7 @@ static int load_refcount_block(BlockDriverState *bs,
write_refcount_block(s);
}
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_LOAD);
ret = bdrv_pread(s->hd, refcount_block_offset, s->refcount_block_cache,
s->cluster_size);
if (ret != s->cluster_size)
@@ -164,6 +167,8 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
unsigned int refcount_table_index;
int ret;
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC);
+
/* Find the refcount block for the given cluster */
refcount_table_index = cluster_index >> (s->cluster_bits - REFCOUNT_SHIFT);
@@ -239,6 +244,7 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
}
/* Now the new refcount block needs to be written to disk */
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC_WRITE);
ret = bdrv_pwrite(s->hd, new_block, s->refcount_block_cache,
s->cluster_size);
if (ret < 0) {
@@ -248,6 +254,7 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
/* If the refcount table is big enough, just hook the block up there */
if (refcount_table_index < s->refcount_table_size) {
uint64_t data64 = cpu_to_be64(new_block);
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC_HOOKUP);
ret = bdrv_pwrite(s->hd,
s->refcount_table_offset + refcount_table_index * sizeof(uint64_t),
&data64, sizeof(data64));
@@ -270,6 +277,8 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
* refcount table at once without producing an inconsistent state in
* between.
*/
+ BLKDBG_EVENT(s->hd, BLKDBG_REFTABLE_GROW);
+
/* Calculate the number of refcount blocks needed so far */
uint64_t refcount_block_clusters = 1 << (s->cluster_bits - REFCOUNT_SHIFT);
uint64_t blocks_used = (s->free_cluster_index +
@@ -325,6 +334,7 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
}
/* Write refcount blocks to disk */
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC_WRITE_BLOCKS);
ret = bdrv_pwrite(s->hd, meta_offset, new_blocks,
blocks_clusters * s->cluster_size);
qemu_free(new_blocks);
@@ -337,6 +347,7 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
cpu_to_be64s(&new_table[i]);
}
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC_WRITE_TABLE);
ret = bdrv_pwrite(s->hd, table_offset, new_table,
table_size * sizeof(uint64_t));
if (ret < 0) {
@@ -351,6 +362,7 @@ static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
uint8_t data[12];
cpu_to_be64w((uint64_t*)data, table_offset);
cpu_to_be32w((uint32_t*)(data + 8), table_clusters);
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_ALLOC_SWITCH_TABLE);
ret = bdrv_pwrite(s->hd, offsetof(QCowHeader, refcount_table_offset),
data, sizeof(data));
if (ret < 0) {
@@ -400,6 +412,7 @@ static int write_refcount_block_entries(BDRVQcowState *s,
& ~(REFCOUNTS_PER_SECTOR - 1);
size = (last_index - first_index) << REFCOUNT_SHIFT;
+ BLKDBG_EVENT(s->hd, BLKDBG_REFBLOCK_UPDATE_PART);
if (bdrv_pwrite(s->hd,
refcount_block_offset + (first_index << REFCOUNT_SHIFT),
&s->refcount_block_cache[first_index], size) != size)
@@ -555,9 +568,11 @@ retry:
int64_t qcow2_alloc_clusters(BlockDriverState *bs, int64_t size)
{
+ BDRVQcowState *s = bs->opaque;
int64_t offset;
int ret;
+ BLKDBG_EVENT(s->hd, BLKDBG_CLUSTER_ALLOC);
offset = alloc_clusters_noref(bs, size);
ret = update_refcount(bs, offset, size, 1);
if (ret < 0) {
@@ -574,6 +589,7 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
int64_t offset, cluster_offset;
int free_in_cluster;
+ BLKDBG_EVENT(s->hd, BLKDBG_CLUSTER_ALLOC_BYTES);
assert(size > 0 && size <= s->cluster_size);
if (s->free_byte_offset == 0) {
s->free_byte_offset = qcow2_alloc_clusters(bs, s->cluster_size);
@@ -615,8 +631,10 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
void qcow2_free_clusters(BlockDriverState *bs,
int64_t offset, int64_t size)
{
+ BDRVQcowState *s = bs->opaque;
int ret;
+ BLKDBG_EVENT(s->hd, BLKDBG_CLUSTER_FREE);
ret = update_refcount(bs, offset, size, -1);
if (ret < 0) {
fprintf(stderr, "qcow2_free_clusters failed: %s\n", strerror(-ret));
diff --git a/block/qcow2.c b/block/qcow2.c
index 4e97eb6..80c99af 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -429,6 +429,7 @@ static void qcow_aio_read_cb(void *opaque, int ret)
acb->hd_iov.iov_base = (void *)acb->buf;
acb->hd_iov.iov_len = acb->cur_nr_sectors * 512;
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
+ BLKDBG_EVENT(s->hd, BLKDBG_READ_BACKING_AIO);
acb->hd_aiocb = bdrv_aio_readv(bs->backing_hd, acb->sector_num,
&acb->hd_qiov, acb->cur_nr_sectors,
qcow_aio_read_cb, acb);
@@ -464,6 +465,7 @@ static void qcow_aio_read_cb(void *opaque, int ret)
acb->hd_iov.iov_base = (void *)acb->buf;
acb->hd_iov.iov_len = acb->cur_nr_sectors * 512;
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
+ BLKDBG_EVENT(s->hd, BLKDBG_READ_AIO);
acb->hd_aiocb = bdrv_aio_readv(s->hd,
(acb->cluster_offset >> 9) + index_in_cluster,
&acb->hd_qiov, acb->cur_nr_sectors,
@@ -619,6 +621,7 @@ static void qcow_aio_write_cb(void *opaque, int ret)
acb->hd_iov.iov_base = (void *)src_buf;
acb->hd_iov.iov_len = acb->cur_nr_sectors * 512;
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
+ BLKDBG_EVENT(s->hd, BLKDBG_WRITE_AIO);
acb->hd_aiocb = bdrv_aio_writev(s->hd,
(acb->cluster_offset >> 9) + index_in_cluster,
&acb->hd_qiov, acb->cur_nr_sectors,
@@ -1141,6 +1144,7 @@ static int qcow_write_compressed(BlockDriverState *bs, int64_t sector_num,
if (!cluster_offset)
return -1;
cluster_offset &= s->cluster_offset_mask;
+ BLKDBG_EVENT(s->hd, BLKDBG_WRITE_COMPRESSED);
if (bdrv_pwrite(s->hd, cluster_offset, out_buf, out_len) != out_len) {
qemu_free(out_buf);
return -1;
@@ -1211,6 +1215,7 @@ static int qcow_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
int growable = bs->growable;
int ret;
+ BLKDBG_EVENT(s->hd, BLKDBG_VMSTATE_SAVE);
bs->growable = 1;
ret = bdrv_pwrite(bs, qcow_vm_state_offset(s) + pos, buf, size);
bs->growable = growable;
@@ -1225,6 +1230,7 @@ static int qcow_load_vmstate(BlockDriverState *bs, uint8_t *buf,
int growable = bs->growable;
int ret;
+ BLKDBG_EVENT(s->hd, BLKDBG_VMSTATE_LOAD);
bs->growable = 1;
ret = bdrv_pread(bs, qcow_vm_state_offset(s) + pos, buf, size);
bs->growable = growable;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 08/26] qcow2: Fix creation of large images
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (6 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 07/26] qcow2: Trigger blkdebug events Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 09/26] Replace calls of old bdrv_open Kevin Wolf
` (18 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
qcow_create2 assumes that the new image will only need one cluster for its
refcount table initially. Obviously that's not true any more when the image is
big enough (exact value depends on the cluster size).
This patch calculates the refcount table size dynamically.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2.c | 43 +++++++++++++++++++++++++++++++++----------
1 files changed, 33 insertions(+), 10 deletions(-)
diff --git a/block/qcow2.c b/block/qcow2.c
index 80c99af..67affa6 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -850,10 +850,11 @@ static int qcow_create2(const char *filename, int64_t total_size,
{
int fd, header_size, backing_filename_len, l1_size, i, shift, l2_bits;
- int ref_clusters, backing_format_len = 0;
+ int ref_clusters, reftable_clusters, backing_format_len = 0;
int rounded_ext_bf_len = 0;
QCowHeader header;
uint64_t tmp, offset;
+ uint64_t old_ref_clusters;
QCowCreateState s1, *s = &s1;
QCowExtension ext_bf = {0, 0};
int ret;
@@ -912,17 +913,37 @@ static int qcow_create2(const char *filename, int64_t total_size,
header.l1_size = cpu_to_be32(l1_size);
offset += align_offset(l1_size * sizeof(uint64_t), s->cluster_size);
- s->refcount_table = qemu_mallocz(s->cluster_size);
+ /* count how many refcount blocks needed */
+
+#define NUM_CLUSTERS(bytes) \
+ (((bytes) + (s->cluster_size) - 1) / (s->cluster_size))
+
+ ref_clusters = NUM_CLUSTERS(NUM_CLUSTERS(offset) * sizeof(uint16_t));
+
+ do {
+ uint64_t image_clusters;
+ old_ref_clusters = ref_clusters;
+
+ /* Number of clusters used for the refcount table */
+ reftable_clusters = NUM_CLUSTERS(ref_clusters * sizeof(uint64_t));
+
+ /* Number of clusters that the whole image will have */
+ image_clusters = NUM_CLUSTERS(offset) + ref_clusters
+ + reftable_clusters;
+
+ /* Number of refcount blocks needed for the image */
+ ref_clusters = NUM_CLUSTERS(image_clusters * sizeof(uint16_t));
+
+ } while (ref_clusters != old_ref_clusters);
+
+ s->refcount_table = qemu_mallocz(reftable_clusters * s->cluster_size);
s->refcount_table_offset = offset;
header.refcount_table_offset = cpu_to_be64(offset);
- header.refcount_table_clusters = cpu_to_be32(1);
- offset += s->cluster_size;
+ header.refcount_table_clusters = cpu_to_be32(reftable_clusters);
+ offset += (reftable_clusters * s->cluster_size);
s->refcount_block_offset = offset;
- /* count how many refcount blocks needed */
- tmp = offset >> s->cluster_bits;
- ref_clusters = (tmp >> (s->cluster_bits - REFCOUNT_SHIFT)) + 1;
for (i=0; i < ref_clusters; i++) {
s->refcount_table[i] = cpu_to_be64(offset);
offset += s->cluster_size;
@@ -934,7 +955,8 @@ static int qcow_create2(const char *filename, int64_t total_size,
qcow2_create_refcount_update(s, 0, header_size);
qcow2_create_refcount_update(s, s->l1_table_offset,
l1_size * sizeof(uint64_t));
- qcow2_create_refcount_update(s, s->refcount_table_offset, s->cluster_size);
+ qcow2_create_refcount_update(s, s->refcount_table_offset,
+ reftable_clusters * s->cluster_size);
qcow2_create_refcount_update(s, s->refcount_block_offset,
ref_clusters * s->cluster_size);
@@ -986,8 +1008,9 @@ static int qcow_create2(const char *filename, int64_t total_size,
}
}
lseek(fd, s->refcount_table_offset, SEEK_SET);
- ret = qemu_write_full(fd, s->refcount_table, s->cluster_size);
- if (ret != s->cluster_size) {
+ ret = qemu_write_full(fd, s->refcount_table,
+ reftable_clusters * s->cluster_size);
+ if (ret != reftable_clusters * s->cluster_size) {
ret = -errno;
goto exit;
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 09/26] Replace calls of old bdrv_open
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (7 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 08/26] qcow2: Fix creation of large images Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 10/26] block: get rid of the BDRV_O_FILE flag Kevin Wolf
` (17 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
What is known today as bdrv_open2 becomes the new bdrv_open. All remaining
callers of the old function are converted to the new one. In some places they
even know the right format, so they should have used bdrv_open2 from the
beginning.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 24 +++++++++---------------
block.h | 5 ++---
block/qcow2.c | 4 ++--
block/vmdk.c | 2 +-
block/vvfat.c | 5 ++++-
hw/xen_disk.c | 2 +-
monitor.c | 2 +-
qemu-img.c | 16 ++++++++--------
qemu-io.c | 2 +-
qemu-nbd.c | 2 +-
vl.c | 2 +-
11 files changed, 31 insertions(+), 35 deletions(-)
diff --git a/block.c b/block.c
index 9351aa3..e74264d 100644
--- a/block.c
+++ b/block.c
@@ -338,7 +338,7 @@ int bdrv_file_open(BlockDriverState **pbs, const char *filename, int flags)
int ret;
bs = bdrv_new("");
- ret = bdrv_open2(bs, filename, flags | BDRV_O_FILE, NULL);
+ ret = bdrv_open(bs, filename, flags | BDRV_O_FILE, NULL);
if (ret < 0) {
bdrv_delete(bs);
return ret;
@@ -348,13 +348,8 @@ int bdrv_file_open(BlockDriverState **pbs, const char *filename, int flags)
return 0;
}
-int bdrv_open(BlockDriverState *bs, const char *filename, int flags)
-{
- return bdrv_open2(bs, filename, flags, NULL);
-}
-
-int bdrv_open2(BlockDriverState *bs, const char *filename, int flags,
- BlockDriver *drv)
+int bdrv_open(BlockDriverState *bs, const char *filename, int flags,
+ BlockDriver *drv)
{
int ret, open_flags;
char tmp_filename[PATH_MAX];
@@ -379,7 +374,7 @@ int bdrv_open2(BlockDriverState *bs, const char *filename, int flags,
/* if there is a backing file, use it */
bs1 = bdrv_new("");
- ret = bdrv_open2(bs1, filename, 0, drv);
+ ret = bdrv_open(bs1, filename, 0, drv);
if (ret < 0) {
bdrv_delete(bs1);
return ret;
@@ -491,9 +486,8 @@ int bdrv_open2(BlockDriverState *bs, const char *filename, int flags,
/* backing files always opened read-only */
open_flags &= ~BDRV_O_RDWR;
-
- ret = bdrv_open2(bs->backing_hd, backing_filename, open_flags,
- back_drv);
+
+ ret = bdrv_open(bs->backing_hd, backing_filename, open_flags, back_drv);
if (ret < 0) {
bdrv_close(bs);
return ret;
@@ -605,12 +599,12 @@ int bdrv_commit(BlockDriverState *bs)
bdrv_delete(bs->backing_hd);
bs->backing_hd = NULL;
bs_rw = bdrv_new("");
- rw_ret = bdrv_open2(bs_rw, filename, open_flags | BDRV_O_RDWR, NULL);
+ rw_ret = bdrv_open(bs_rw, filename, open_flags | BDRV_O_RDWR, NULL);
if (rw_ret < 0) {
bdrv_delete(bs_rw);
/* try to re-open read-only */
bs_ro = bdrv_new("");
- ret = bdrv_open2(bs_ro, filename, open_flags & ~BDRV_O_RDWR, NULL);
+ ret = bdrv_open(bs_ro, filename, open_flags & ~BDRV_O_RDWR, NULL);
if (ret < 0) {
bdrv_delete(bs_ro);
/* drive not functional anymore */
@@ -662,7 +656,7 @@ ro_cleanup:
bdrv_delete(bs->backing_hd);
bs->backing_hd = NULL;
bs_ro = bdrv_new("");
- ret = bdrv_open2(bs_ro, filename, open_flags & ~BDRV_O_RDWR, NULL);
+ ret = bdrv_open(bs_ro, filename, open_flags & ~BDRV_O_RDWR, NULL);
if (ret < 0) {
bdrv_delete(bs_ro);
/* drive not functional anymore */
diff --git a/block.h b/block.h
index 14443f1..c5900c8 100644
--- a/block.h
+++ b/block.h
@@ -68,9 +68,8 @@ int bdrv_create2(BlockDriver *drv,
BlockDriverState *bdrv_new(const char *device_name);
void bdrv_delete(BlockDriverState *bs);
int bdrv_file_open(BlockDriverState **pbs, const char *filename, int flags);
-int bdrv_open(BlockDriverState *bs, const char *filename, int flags);
-int bdrv_open2(BlockDriverState *bs, const char *filename, int flags,
- BlockDriver *drv);
+int bdrv_open(BlockDriverState *bs, const char *filename, int flags,
+ BlockDriver *drv);
void bdrv_close(BlockDriverState *bs);
int bdrv_check(BlockDriverState *bs);
int bdrv_read(BlockDriverState *bs, int64_t sector_num,
diff --git a/block/qcow2.c b/block/qcow2.c
index 67affa6..30ded6a 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -52,7 +52,7 @@ typedef struct {
#define QCOW_EXT_MAGIC_END 0
#define QCOW_EXT_MAGIC_BACKING_FORMAT 0xE2792ACA
-
+static BlockDriver bdrv_qcow2;
static int qcow_probe(const uint8_t *buf, int buf_size, const char *filename)
{
@@ -1033,7 +1033,7 @@ exit:
if (ret == 0 && prealloc) {
BlockDriverState *bs;
bs = bdrv_new("");
- bdrv_open(bs, filename, BDRV_O_CACHE_WB | BDRV_O_RDWR);
+ bdrv_open(bs, filename, BDRV_O_CACHE_WB | BDRV_O_RDWR, &bdrv_qcow2);
preallocate(bs);
bdrv_close(bs);
}
diff --git a/block/vmdk.c b/block/vmdk.c
index 007fca4..6fdea1d 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -390,7 +390,7 @@ static int vmdk_parent_open(BlockDriverState *bs, const char * filename)
return -1;
}
parent_open = 1;
- if (bdrv_open(bs->backing_hd, parent_img_name, 0) < 0)
+ if (bdrv_open(bs->backing_hd, parent_img_name, 0, NULL) < 0)
goto failure;
parent_open = 0;
}
diff --git a/block/vvfat.c b/block/vvfat.c
index 36f6ab4..0701df4 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -2795,8 +2795,11 @@ static int enable_write_target(BDRVVVFATState *s)
if (bdrv_create(bdrv_qcow, s->qcow_filename, options) < 0)
return -1;
s->qcow = bdrv_new("");
- if (s->qcow == NULL || bdrv_open(s->qcow, s->qcow_filename, BDRV_O_RDWR) < 0)
+ if (s->qcow == NULL ||
+ bdrv_open(s->qcow, s->qcow_filename, BDRV_O_RDWR, bdrv_qcow) < 0)
+ {
return -1;
+ }
#ifndef _WIN32
unlink(s->qcow_filename);
diff --git a/hw/xen_disk.c b/hw/xen_disk.c
index beadf90..95017a1 100644
--- a/hw/xen_disk.c
+++ b/hw/xen_disk.c
@@ -629,7 +629,7 @@ static int blk_init(struct XenDevice *xendev)
xen_be_printf(&blkdev->xendev, 2, "create new bdrv (xenbus setup)\n");
blkdev->bs = bdrv_new(blkdev->dev);
if (blkdev->bs) {
- if (bdrv_open2(blkdev->bs, blkdev->filename, qflags,
+ if (bdrv_open(blkdev->bs, blkdev->filename, qflags,
bdrv_find_whitelisted_format(blkdev->fileproto))
!= 0) {
bdrv_delete(blkdev->bs);
diff --git a/monitor.c b/monitor.c
index ba60f97..c25d551 100644
--- a/monitor.c
+++ b/monitor.c
@@ -1099,7 +1099,7 @@ static int do_change_block(Monitor *mon, const char *device,
if (eject_device(mon, bs, 0) < 0) {
return -1;
}
- if (bdrv_open2(bs, filename, BDRV_O_RDWR, drv) < 0) {
+ if (bdrv_open(bs, filename, BDRV_O_RDWR, drv) < 0) {
qerror_report(QERR_OPEN_FILE_FAILED, filename);
return -1;
}
diff --git a/qemu-img.c b/qemu-img.c
index 9b28664..18e43a0 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -210,7 +210,7 @@ static BlockDriverState *bdrv_new_open(const char *filename,
if (!readonly) {
flags |= BDRV_O_RDWR;
}
- if (bdrv_open2(bs, filename, flags, drv) < 0) {
+ if (bdrv_open(bs, filename, flags, drv) < 0) {
error("Could not open '%s'", filename);
}
if (bdrv_is_encrypted(bs)) {
@@ -415,7 +415,7 @@ static int img_check(int argc, char **argv)
} else {
drv = NULL;
}
- if (bdrv_open2(bs, filename, BRDV_O_FLAGS, drv) < 0) {
+ if (bdrv_open(bs, filename, BRDV_O_FLAGS, drv) < 0) {
error("Could not open '%s'", filename);
}
ret = bdrv_check(bs);
@@ -474,7 +474,7 @@ static int img_commit(int argc, char **argv)
} else {
drv = NULL;
}
- if (bdrv_open2(bs, filename, BRDV_O_FLAGS | BDRV_O_RDWR, drv) < 0) {
+ if (bdrv_open(bs, filename, BRDV_O_FLAGS | BDRV_O_RDWR, drv) < 0) {
error("Could not open '%s'", filename);
}
ret = bdrv_commit(bs);
@@ -926,7 +926,7 @@ static int img_info(int argc, char **argv)
} else {
drv = NULL;
}
- if (bdrv_open2(bs, filename, BRDV_O_FLAGS | BDRV_O_NO_BACKING, drv) < 0) {
+ if (bdrv_open(bs, filename, BRDV_O_FLAGS | BDRV_O_NO_BACKING, drv) < 0) {
error("Could not open '%s'", filename);
}
bdrv_get_format(bs, fmt_name, sizeof(fmt_name));
@@ -1032,7 +1032,7 @@ static int img_snapshot(int argc, char **argv)
if (!bs)
error("Not enough memory");
- if (bdrv_open2(bs, filename, bdrv_oflags, NULL) < 0) {
+ if (bdrv_open(bs, filename, bdrv_oflags, NULL) < 0) {
error("Could not open '%s'", filename);
}
@@ -1137,7 +1137,7 @@ static int img_rebase(int argc, char **argv)
}
flags = BRDV_O_FLAGS | BDRV_O_RDWR | (unsafe ? BDRV_O_NO_BACKING : 0);
- if (bdrv_open2(bs, filename, flags, drv) < 0) {
+ if (bdrv_open(bs, filename, flags, drv) < 0) {
error("Could not open '%s'", filename);
}
@@ -1169,7 +1169,7 @@ static int img_rebase(int argc, char **argv)
bs_old_backing = bdrv_new("old_backing");
bdrv_get_backing_filename(bs, backing_name, sizeof(backing_name));
- if (bdrv_open2(bs_old_backing, backing_name, BRDV_O_FLAGS,
+ if (bdrv_open(bs_old_backing, backing_name, BRDV_O_FLAGS,
old_backing_drv))
{
error("Could not open old backing file '%s'", backing_name);
@@ -1177,7 +1177,7 @@ static int img_rebase(int argc, char **argv)
}
bs_new_backing = bdrv_new("new_backing");
- if (bdrv_open2(bs_new_backing, out_baseimg, BRDV_O_FLAGS | BDRV_O_RDWR,
+ if (bdrv_open(bs_new_backing, out_baseimg, BRDV_O_FLAGS | BDRV_O_RDWR,
new_backing_drv))
{
error("Could not open new backing file '%s'", out_baseimg);
diff --git a/qemu-io.c b/qemu-io.c
index 6b83714..ffb5817 100644
--- a/qemu-io.c
+++ b/qemu-io.c
@@ -1284,7 +1284,7 @@ static int openfile(char *name, int flags, int growable)
flags |= BDRV_O_FILE;
}
- if (bdrv_open(bs, name, flags) < 0) {
+ if (bdrv_open(bs, name, flags, NULL) < 0) {
fprintf(stderr, "%s: can't open device %s\n", progname, name);
bs = NULL;
return 1;
diff --git a/qemu-nbd.c b/qemu-nbd.c
index 6d854d3..25aa913 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -333,7 +333,7 @@ int main(int argc, char **argv)
if (bs == NULL)
return 1;
- if (bdrv_open(bs, argv[optind], flags) < 0)
+ if (bdrv_open(bs, argv[optind], flags, NULL) < 0)
return 1;
fd_size = bs->total_sectors * 512;
diff --git a/vl.c b/vl.c
index fe2dd6f..0c76d33 100644
--- a/vl.c
+++ b/vl.c
@@ -1129,7 +1129,7 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
bdrv_flags |= ro ? 0 : BDRV_O_RDWR;
- if (bdrv_open2(dinfo->bdrv, file, bdrv_flags, drv) < 0) {
+ if (bdrv_open(dinfo->bdrv, file, bdrv_flags, drv) < 0) {
fprintf(stderr, "qemu: could not open disk image %s: %s\n",
file, strerror(errno));
return NULL;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 10/26] block: get rid of the BDRV_O_FILE flag
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (8 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 09/26] Replace calls of old bdrv_open Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 11/26] block: split raw_getlength Kevin Wolf
` (16 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Christoph Hellwig <hch@lst.de>
BDRV_O_FILE is only used to communicate between bdrv_file_open and bdrv_open.
It affects two things: first bdrv_open only searches for protocols using
find_protocol instead of all image formats and host drivers. We can easily
move that to the caller and pass the found driver to bdrv_open. Second
it is used to not force a read-write open of a snapshot file. But we never
use bdrv_file_open to open snapshots and this behaviour doesn't make sense
to start with.
qemu-io abused the BDRV_O_FILE for it's growable option, switch it to
using bdrv_file_open to make sure we only open files as growable were
we can actually support that.
This patch requires Kevin's "[PATCH] Replace calls of old bdrv_open" to
be applied first.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 19 +++++++++++--------
block.h | 4 ----
qemu-io.c | 28 ++++++++++++++--------------
3 files changed, 25 insertions(+), 26 deletions(-)
diff --git a/block.c b/block.c
index e74264d..ed4c819 100644
--- a/block.c
+++ b/block.c
@@ -335,10 +335,16 @@ static BlockDriver *find_image_format(const char *filename)
int bdrv_file_open(BlockDriverState **pbs, const char *filename, int flags)
{
BlockDriverState *bs;
+ BlockDriver *drv;
int ret;
+ drv = find_protocol(filename);
+ if (!drv) {
+ return -ENOENT;
+ }
+
bs = bdrv_new("");
- ret = bdrv_open(bs, filename, flags | BDRV_O_FILE, NULL);
+ ret = bdrv_open(bs, filename, flags, drv);
if (ret < 0) {
bdrv_delete(bs);
return ret;
@@ -416,9 +422,8 @@ int bdrv_open(BlockDriverState *bs, const char *filename, int flags,
}
pstrcpy(bs->filename, sizeof(bs->filename), filename);
- if (flags & BDRV_O_FILE) {
- drv = find_protocol(filename);
- } else if (!drv) {
+
+ if (!drv) {
drv = find_hdev_driver(filename);
if (!drv) {
drv = find_image_format(filename);
@@ -450,14 +455,12 @@ int bdrv_open(BlockDriverState *bs, const char *filename, int flags,
* Clear flags that are internal to the block layer before opening the
* image.
*/
- open_flags = flags & ~(BDRV_O_FILE | BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING);
+ open_flags = flags & ~(BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING);
/*
* Snapshots should be writeable.
- *
- * XXX(hch): and what is the point of a snapshot during a read-only open?
*/
- if (!(flags & BDRV_O_FILE) && bs->is_temporary) {
+ if (bs->is_temporary) {
open_flags |= BDRV_O_RDWR;
}
diff --git a/block.h b/block.h
index c5900c8..4a57dd5 100644
--- a/block.h
+++ b/block.h
@@ -29,10 +29,6 @@ typedef struct QEMUSnapshotInfo {
#define BDRV_O_RDWR 0x0002
#define BDRV_O_SNAPSHOT 0x0008 /* open the file read only and save writes in a snapshot */
-#define BDRV_O_FILE 0x0010 /* open as a raw file (do not try to
- use a disk image format on top of
- it (default for
- bdrv_file_open()) */
#define BDRV_O_NOCACHE 0x0020 /* do not use the host page cache */
#define BDRV_O_CACHE_WB 0x0040 /* use write-back caching */
#define BDRV_O_NATIVE_AIO 0x0080 /* use native AIO instead of the thread pool */
diff --git a/qemu-io.c b/qemu-io.c
index ffb5817..8517b90 100644
--- a/qemu-io.c
+++ b/qemu-io.c
@@ -1276,23 +1276,23 @@ static int openfile(char *name, int flags, int growable)
return 1;
}
- bs = bdrv_new("hda");
- if (!bs)
- return 1;
-
if (growable) {
- flags |= BDRV_O_FILE;
- }
-
- if (bdrv_open(bs, name, flags, NULL) < 0) {
- fprintf(stderr, "%s: can't open device %s\n", progname, name);
- bs = NULL;
- return 1;
+ if (bdrv_file_open(&bs, name, flags)) {
+ fprintf(stderr, "%s: can't open device %s\n", progname, name);
+ return 1;
+ }
+ } else {
+ bs = bdrv_new("hda");
+ if (!bs)
+ return 1;
+
+ if (bdrv_open(bs, name, flags, NULL) < 0) {
+ fprintf(stderr, "%s: can't open device %s\n", progname, name);
+ bs = NULL;
+ return 1;
+ }
}
- if (growable) {
- bs->growable = 1;
- }
return 0;
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 11/26] block: split raw_getlength
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (9 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 10/26] block: get rid of the BDRV_O_FILE flag Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 12/26] qcow2: Return 0/-errno in write_l2_entries Kevin Wolf
` (15 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Christoph Hellwig <hch@lst.de>
Split up the raw_getlength into separate generic, solaris and BSD
versions to reduce the ifdef maze a bit. The BSD variant still
is a complete maze, but to clean it up properly we'd need some
people using the BSD variants to figure out what code is used
for what variant.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/raw-posix.c | 65 ++++++++++++++++++++++++++++++++++------------------
1 files changed, 42 insertions(+), 23 deletions(-)
diff --git a/block/raw-posix.c b/block/raw-posix.c
index 6521ca4..4cda9c1 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -627,29 +627,48 @@ static int64_t raw_getlength(BlockDriverState *bs)
} else
return st.st_size;
}
-#else /* !__OpenBSD__ */
-static int64_t raw_getlength(BlockDriverState *bs)
+#elif defined(__sun__)
+static int64_t raw_getlength(BlockDriverState *bs)
+{
+ BDRVRawState *s = bs->opaque;
+ struct dk_minfo minfo;
+ int ret;
+
+ ret = fd_open(bs);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /*
+ * Use the DKIOCGMEDIAINFO ioctl to read the size.
+ */
+ ret = ioctl(s->fd, DKIOCGMEDIAINFO, &minfo);
+ if (ret != -1) {
+ return minfo.dki_lbsize * minfo.dki_capacity;
+ }
+
+ /*
+ * There are reports that lseek on some devices fails, but
+ * irc discussion said that contingency on contingency was overkill.
+ */
+ return lseek(s->fd, 0, SEEK_END);
+}
+#elif defined(CONFIG_BSD)
+static int64_t raw_getlength(BlockDriverState *bs)
{
BDRVRawState *s = bs->opaque;
int fd = s->fd;
int64_t size;
-#ifdef CONFIG_BSD
struct stat sb;
#if defined (__FreeBSD__) || defined(__FreeBSD_kernel__)
int reopened = 0;
#endif
-#endif
-#ifdef __sun__
- struct dk_minfo minfo;
- int rv;
-#endif
int ret;
ret = fd_open(bs);
if (ret < 0)
return ret;
-#ifdef CONFIG_BSD
#if defined (__FreeBSD__) || defined(__FreeBSD_kernel__)
again:
#endif
@@ -684,24 +703,24 @@ again:
}
}
#endif
- } else
-#endif
-#ifdef __sun__
- /*
- * use the DKIOCGMEDIAINFO ioctl to read the size.
- */
- rv = ioctl ( fd, DKIOCGMEDIAINFO, &minfo );
- if ( rv != -1 ) {
- size = minfo.dki_lbsize * minfo.dki_capacity;
- } else /* there are reports that lseek on some devices
- fails, but irc discussion said that contingency
- on contingency was overkill */
-#endif
- {
+ } else {
size = lseek(fd, 0, SEEK_END);
}
return size;
}
+#else
+static int64_t raw_getlength(BlockDriverState *bs)
+{
+ BDRVRawState *s = bs->opaque;
+ int ret;
+
+ ret = fd_open(bs);
+ if (ret < 0) {
+ return ret;
+ }
+
+ return lseek(s->fd, 0, SEEK_END);
+}
#endif
static int raw_create(const char *filename, QEMUOptionParameter *options)
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 12/26] qcow2: Return 0/-errno in write_l2_entries
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (10 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 11/26] block: split raw_getlength Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 13/26] qcow2: Fix error return code in qcow2_alloc_cluster_link_l2 Kevin Wolf
` (14 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Change write_l2_entries to return the real error code instead of -1.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2-cluster.c | 9 +++++----
1 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 8cb4b38..2f37acd 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -627,12 +627,13 @@ static int write_l2_entries(BDRVQcowState *s, uint64_t *l2_table,
int start_offset = (8 * l2_index) & ~511;
int end_offset = (8 * (l2_index + num) + 511) & ~511;
size_t len = end_offset - start_offset;
+ int ret;
BLKDBG_EVENT(s->hd, BLKDBG_L2_UPDATE);
- if (bdrv_pwrite(s->hd, l2_offset + start_offset, &l2_table[l2_start_index],
- len) != len)
- {
- return -1;
+ ret = bdrv_pwrite(s->hd, l2_offset + start_offset,
+ &l2_table[l2_start_index], len);
+ if (ret < 0) {
+ return ret;
}
return 0;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 13/26] qcow2: Fix error return code in qcow2_alloc_cluster_link_l2
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (11 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 12/26] qcow2: Return 0/-errno in write_l2_entries Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 14/26] qcow2: Return 0/-errno in write_l1_entry Kevin Wolf
` (13 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Fix qcow2_alloc_cluster_link_l2 to return the real error code like it does in
all other error cases.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2-cluster.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 2f37acd..d2774d1 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -687,8 +687,8 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
(i << s->cluster_bits)) | QCOW_OFLAG_COPIED);
}
- if (write_l2_entries(s, l2_table, l2_offset, l2_index, m->nb_clusters) < 0) {
- ret = -1;
+ ret = write_l2_entries(s, l2_table, l2_offset, l2_index, m->nb_clusters);
+ if (ret < 0) {
goto err;
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 14/26] qcow2: Return 0/-errno in write_l1_entry
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (12 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 13/26] qcow2: Fix error return code in qcow2_alloc_cluster_link_l2 Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 15/26] qcow2: Return 0/-errno in l2_allocate Kevin Wolf
` (12 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Change write_l1_entry to return the real error code instead of -1.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2-cluster.c | 10 +++++-----
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index d2774d1..d5c52a9 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -193,7 +193,7 @@ static int write_l1_entry(BDRVQcowState *s, int l1_index)
{
uint64_t buf[L1_ENTRIES_PER_SECTOR];
int l1_start_index;
- int i;
+ int i, ret;
l1_start_index = l1_index & ~(L1_ENTRIES_PER_SECTOR - 1);
for (i = 0; i < L1_ENTRIES_PER_SECTOR; i++) {
@@ -201,10 +201,10 @@ static int write_l1_entry(BDRVQcowState *s, int l1_index)
}
BLKDBG_EVENT(s->hd, BLKDBG_L1_UPDATE);
- if (bdrv_pwrite(s->hd, s->l1_table_offset + 8 * l1_start_index,
- buf, sizeof(buf)) != sizeof(buf))
- {
- return -1;
+ ret = bdrv_pwrite(s->hd, s->l1_table_offset + 8 * l1_start_index,
+ buf, sizeof(buf));
+ if (ret < 0) {
+ return ret;
}
return 0;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 15/26] qcow2: Return 0/-errno in l2_allocate
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (13 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 14/26] qcow2: Return 0/-errno in write_l1_entry Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 16/26] cleanup block driver option handling in vl.c Kevin Wolf
` (11 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
Returning NULL on error doesn't allow distinguishing between different errors.
Change the interface to return an integer for -errno.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2-cluster.c | 40 +++++++++++++++++++++++-----------------
1 files changed, 23 insertions(+), 17 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index d5c52a9..639e05e 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -220,13 +220,14 @@ static int write_l1_entry(BDRVQcowState *s, int l1_index)
*
*/
-static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
+static int l2_allocate(BlockDriverState *bs, int l1_index, uint64_t **table)
{
BDRVQcowState *s = bs->opaque;
int min_index;
uint64_t old_l2_offset;
uint64_t *l2_table;
int64_t l2_offset;
+ int ret;
old_l2_offset = s->l1_table[l1_index];
@@ -234,14 +235,15 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
l2_offset = qcow2_alloc_clusters(bs, s->l2_size * sizeof(uint64_t));
if (l2_offset < 0) {
- return NULL;
+ return l2_offset;
}
/* update the L1 entry */
s->l1_table[l1_index] = l2_offset | QCOW_OFLAG_COPIED;
- if (write_l1_entry(s, l1_index) < 0) {
- return NULL;
+ ret = write_l1_entry(s, l1_index);
+ if (ret < 0) {
+ return ret;
}
/* allocate a new entry in the l2 cache */
@@ -255,24 +257,27 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
} else {
/* if there was an old l2 table, read it from the disk */
BLKDBG_EVENT(s->hd, BLKDBG_L2_ALLOC_COW_READ);
- if (bdrv_pread(s->hd, old_l2_offset,
- l2_table, s->l2_size * sizeof(uint64_t)) !=
- s->l2_size * sizeof(uint64_t))
- return NULL;
+ ret = bdrv_pread(s->hd, old_l2_offset, l2_table,
+ s->l2_size * sizeof(uint64_t));
+ if (ret < 0) {
+ return ret;
+ }
}
/* write the l2 table to the file */
BLKDBG_EVENT(s->hd, BLKDBG_L2_ALLOC_WRITE);
- if (bdrv_pwrite(s->hd, l2_offset,
- l2_table, s->l2_size * sizeof(uint64_t)) !=
- s->l2_size * sizeof(uint64_t))
- return NULL;
+ ret = bdrv_pwrite(s->hd, l2_offset, l2_table,
+ s->l2_size * sizeof(uint64_t));
+ if (ret < 0) {
+ return ret;
+ }
/* update the l2 cache entry */
s->l2_cache_offsets[min_index] = l2_offset;
s->l2_cache_counts[min_index] = 1;
- return l2_table;
+ *table = l2_table;
+ return 0;
}
static int count_contiguous_clusters(uint64_t nb_clusters, int cluster_size,
@@ -510,7 +515,8 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
{
BDRVQcowState *s = bs->opaque;
unsigned int l1_index, l2_index;
- uint64_t l2_offset, *l2_table;
+ uint64_t l2_offset;
+ uint64_t *l2_table = NULL;
int ret;
/* seek the the l2 offset in the l1 table */
@@ -536,9 +542,9 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
} else {
if (l2_offset)
qcow2_free_clusters(bs, l2_offset, s->l2_size * sizeof(uint64_t));
- l2_table = l2_allocate(bs, l1_index);
- if (l2_table == NULL) {
- return -EIO;
+ ret = l2_allocate(bs, l1_index, &l2_table);
+ if (ret < 0) {
+ return ret;
}
l2_offset = s->l1_table[l1_index] & ~QCOW_OFLAG_COPIED;
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 16/26] cleanup block driver option handling in vl.c
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (14 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 15/26] qcow2: Return 0/-errno in l2_allocate Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 17/26] block: Do not export bdrv_first Kevin Wolf
` (10 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Christoph Hellwig <hch@lst.de>
Assign directly to the bdrv_flags variable instead of using
magic numbers before translating to the BDRV_O_* options.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
vl.c | 45 ++++++++++++++++-----------------------------
1 files changed, 16 insertions(+), 29 deletions(-)
diff --git a/vl.c b/vl.c
index 0c76d33..a5a0f41 100644
--- a/vl.c
+++ b/vl.c
@@ -783,10 +783,8 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
QEMUMachine *machine = opaque;
int max_devs;
int index;
- int cache;
- int aio = 0;
int ro = 0;
- int bdrv_flags;
+ int bdrv_flags = 0;
int on_read_error, on_write_error;
const char *devaddr;
DriveInfo *dinfo;
@@ -795,7 +793,6 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
*fatal_error = 1;
translation = BIOS_ATA_TRANSLATION_AUTO;
- cache = 1;
if (machine && machine->use_scsi) {
type = IF_SCSI;
@@ -909,13 +906,13 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
}
if ((buf = qemu_opt_get(opts, "cache")) != NULL) {
- if (!strcmp(buf, "off") || !strcmp(buf, "none"))
- cache = 0;
- else if (!strcmp(buf, "writethrough"))
- cache = 1;
- else if (!strcmp(buf, "writeback"))
- cache = 2;
- else {
+ if (!strcmp(buf, "off") || !strcmp(buf, "none")) {
+ bdrv_flags |= BDRV_O_NOCACHE;
+ } else if (!strcmp(buf, "writeback")) {
+ bdrv_flags |= BDRV_O_CACHE_WB;
+ } else if (!strcmp(buf, "writethrough")) {
+ /* this is the default */
+ } else {
fprintf(stderr, "qemu: invalid cache option\n");
return NULL;
}
@@ -923,11 +920,11 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
#ifdef CONFIG_LINUX_AIO
if ((buf = qemu_opt_get(opts, "aio")) != NULL) {
- if (!strcmp(buf, "threads"))
- aio = 0;
- else if (!strcmp(buf, "native"))
- aio = 1;
- else {
+ if (!strcmp(buf, "native")) {
+ bdrv_flags |= BDRV_O_NATIVE_AIO;
+ } else if (!strcmp(buf, "threads")) {
+ /* this is the default */
+ } else {
fprintf(stderr, "qemu: invalid aio option\n");
return NULL;
}
@@ -1101,20 +1098,10 @@ DriveInfo *drive_init(QemuOpts *opts, void *opaque,
*fatal_error = 0;
return NULL;
}
- bdrv_flags = 0;
if (snapshot) {
- bdrv_flags |= BDRV_O_SNAPSHOT;
- cache = 2; /* always use write-back with snapshot */
- }
- if (cache == 0) /* no caching */
- bdrv_flags |= BDRV_O_NOCACHE;
- else if (cache == 2) /* write-back */
- bdrv_flags |= BDRV_O_CACHE_WB;
-
- if (aio == 1) {
- bdrv_flags |= BDRV_O_NATIVE_AIO;
- } else {
- bdrv_flags &= ~BDRV_O_NATIVE_AIO;
+ /* always use write-back with snapshot */
+ bdrv_flags &= ~BDRV_O_CACHE_MASK;
+ bdrv_flags |= (BDRV_O_SNAPSHOT|BDRV_O_CACHE_WB);
}
if (media == MEDIA_CDROM) {
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 17/26] block: Do not export bdrv_first
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (15 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 16/26] cleanup block driver option handling in vl.c Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 18/26] block: Convert bdrv_first to QTAILQ Kevin Wolf
` (9 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
The bdrv_first linked list of BlockDriverStates is currently extern so
that block migration can iterate the list. However, since there is
already a bdrv_iterate() function there is no need to expose bdrv_first.
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block-migration.c | 63 +++++++++++++++++++++++++++-------------------------
block.c | 2 +-
block_int.h | 2 -
3 files changed, 34 insertions(+), 33 deletions(-)
diff --git a/block-migration.c b/block-migration.c
index 92349a2..7d04d6d 100644
--- a/block-migration.c
+++ b/block-migration.c
@@ -230,12 +230,42 @@ static void set_dirty_tracking(int enable)
}
}
-static void init_blk_migration(Monitor *mon, QEMUFile *f)
+static void init_blk_migration_it(void *opaque, BlockDriverState *bs)
{
+ Monitor *mon = opaque;
BlkMigDevState *bmds;
- BlockDriverState *bs;
int64_t sectors;
+ if (bs->type == BDRV_TYPE_HD) {
+ sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
+ if (sectors == 0) {
+ return;
+ }
+
+ bmds = qemu_mallocz(sizeof(BlkMigDevState));
+ bmds->bs = bs;
+ bmds->bulk_completed = 0;
+ bmds->total_sectors = sectors;
+ bmds->completed_sectors = 0;
+ bmds->shared_base = block_mig_state.shared_base;
+
+ block_mig_state.total_sector_sum += sectors;
+
+ if (bmds->shared_base) {
+ monitor_printf(mon, "Start migration for %s with shared base "
+ "image\n",
+ bs->device_name);
+ } else {
+ monitor_printf(mon, "Start full migration for %s\n",
+ bs->device_name);
+ }
+
+ QSIMPLEQ_INSERT_TAIL(&block_mig_state.bmds_list, bmds, entry);
+ }
+}
+
+static void init_blk_migration(Monitor *mon, QEMUFile *f)
+{
block_mig_state.submitted = 0;
block_mig_state.read_done = 0;
block_mig_state.transferred = 0;
@@ -245,34 +275,7 @@ static void init_blk_migration(Monitor *mon, QEMUFile *f)
block_mig_state.total_time = 0;
block_mig_state.reads = 0;
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
- if (bs->type == BDRV_TYPE_HD) {
- sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
- if (sectors == 0) {
- continue;
- }
-
- bmds = qemu_mallocz(sizeof(BlkMigDevState));
- bmds->bs = bs;
- bmds->bulk_completed = 0;
- bmds->total_sectors = sectors;
- bmds->completed_sectors = 0;
- bmds->shared_base = block_mig_state.shared_base;
-
- block_mig_state.total_sector_sum += sectors;
-
- if (bmds->shared_base) {
- monitor_printf(mon, "Start migration for %s with shared base "
- "image\n",
- bs->device_name);
- } else {
- monitor_printf(mon, "Start full migration for %s\n",
- bs->device_name);
- }
-
- QSIMPLEQ_INSERT_TAIL(&block_mig_state.bmds_list, bmds, entry);
- }
- }
+ bdrv_iterate(init_blk_migration_it, mon);
}
static int blk_mig_save_bulked_block(Monitor *mon, QEMUFile *f)
diff --git a/block.c b/block.c
index ed4c819..61da183 100644
--- a/block.c
+++ b/block.c
@@ -55,7 +55,7 @@ static int bdrv_read_em(BlockDriverState *bs, int64_t sector_num,
static int bdrv_write_em(BlockDriverState *bs, int64_t sector_num,
const uint8_t *buf, int nb_sectors);
-BlockDriverState *bdrv_first;
+static BlockDriverState *bdrv_first;
static BlockDriver *first_drv;
diff --git a/block_int.h b/block_int.h
index e7e1e7e..d5a808d 100644
--- a/block_int.h
+++ b/block_int.h
@@ -200,8 +200,6 @@ void qemu_aio_release(void *p);
void *qemu_blockalign(BlockDriverState *bs, size_t size);
-extern BlockDriverState *bdrv_first;
-
#ifdef _WIN32
int is_windows_drive(const char *filename);
#endif
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 18/26] block: Convert bdrv_first to QTAILQ
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (16 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 17/26] block: Do not export bdrv_first Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 19/26] Remove un-needed code Kevin Wolf
` (8 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 41 +++++++++++++++++++----------------------
block_int.h | 3 ++-
2 files changed, 21 insertions(+), 23 deletions(-)
diff --git a/block.c b/block.c
index 61da183..5d087de 100644
--- a/block.c
+++ b/block.c
@@ -55,7 +55,8 @@ static int bdrv_read_em(BlockDriverState *bs, int64_t sector_num,
static int bdrv_write_em(BlockDriverState *bs, int64_t sector_num,
const uint8_t *buf, int nb_sectors);
-static BlockDriverState *bdrv_first;
+static QTAILQ_HEAD(, BlockDriverState) bdrv_states =
+ QTAILQ_HEAD_INITIALIZER(bdrv_states);
static BlockDriver *first_drv;
@@ -148,16 +149,12 @@ void bdrv_register(BlockDriver *bdrv)
/* create a new block device (by default it is empty) */
BlockDriverState *bdrv_new(const char *device_name)
{
- BlockDriverState **pbs, *bs;
+ BlockDriverState *bs;
bs = qemu_mallocz(sizeof(BlockDriverState));
pstrcpy(bs->device_name, sizeof(bs->device_name), device_name);
if (device_name[0] != '\0') {
- /* insert at the end */
- pbs = &bdrv_first;
- while (*pbs != NULL)
- pbs = &(*pbs)->next;
- *pbs = bs;
+ QTAILQ_INSERT_TAIL(&bdrv_states, bs, list);
}
return bs;
}
@@ -545,13 +542,10 @@ void bdrv_close(BlockDriverState *bs)
void bdrv_delete(BlockDriverState *bs)
{
- BlockDriverState **pbs;
-
- pbs = &bdrv_first;
- while (*pbs != bs && *pbs != NULL)
- pbs = &(*pbs)->next;
- if (*pbs == bs)
- *pbs = bs->next;
+ /* remove from list, if necessary */
+ if (bs->device_name[0] != '\0') {
+ QTAILQ_REMOVE(&bdrv_states, bs, list);
+ }
bdrv_close(bs);
qemu_free(bs);
@@ -1172,9 +1166,10 @@ BlockDriverState *bdrv_find(const char *name)
{
BlockDriverState *bs;
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
- if (!strcmp(name, bs->device_name))
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
+ if (!strcmp(name, bs->device_name)) {
return bs;
+ }
}
return NULL;
}
@@ -1183,7 +1178,7 @@ void bdrv_iterate(void (*it)(void *opaque, BlockDriverState *bs), void *opaque)
{
BlockDriverState *bs;
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
it(opaque, bs);
}
}
@@ -1203,10 +1198,12 @@ void bdrv_flush_all(void)
{
BlockDriverState *bs;
- for (bs = bdrv_first; bs != NULL; bs = bs->next)
- if (bs->drv && !bdrv_is_read_only(bs) &&
- (!bdrv_is_removable(bs) || bdrv_is_inserted(bs)))
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
+ if (bs->drv && !bdrv_is_read_only(bs) &&
+ (!bdrv_is_removable(bs) || bdrv_is_inserted(bs))) {
bdrv_flush(bs);
+ }
+ }
}
/*
@@ -1340,7 +1337,7 @@ void bdrv_info(Monitor *mon, QObject **ret_data)
bs_list = qlist_new();
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
QObject *bs_obj;
const char *type = "unknown";
@@ -1445,7 +1442,7 @@ void bdrv_info_stats(Monitor *mon, QObject **ret_data)
devices = qlist_new();
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
obj = qobject_from_jsonf("{ 'device': %s, 'stats': {"
"'rd_bytes': %" PRId64 ","
"'wr_bytes': %" PRId64 ","
diff --git a/block_int.h b/block_int.h
index d5a808d..466a38c 100644
--- a/block_int.h
+++ b/block_int.h
@@ -26,6 +26,7 @@
#include "block.h"
#include "qemu-option.h"
+#include "qemu-queue.h"
#define BLOCK_FLAG_ENCRYPT 1
#define BLOCK_FLAG_COMPRESS 2
@@ -180,7 +181,7 @@ struct BlockDriverState {
char device_name[32];
unsigned long *dirty_bitmap;
int64_t dirty_count;
- BlockDriverState *next;
+ QTAILQ_ENTRY(BlockDriverState) list;
void *private;
};
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 19/26] Remove un-needed code
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (17 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 18/26] block: Convert bdrv_first to QTAILQ Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 20/26] block.h: bdrv_create2 doesn't exist any more Kevin Wolf
` (7 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Bruce Rogers <brogers@novell.com>
The bdrv_set_geometry_hint call below is not needed - it's just setting
what was just read.
Signed-off-by: Bruce Rogers <brogers@novell.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
hw/virtio-blk.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/hw/virtio-blk.c b/hw/virtio-blk.c
index 01d77b8..b05d15e 100644
--- a/hw/virtio-blk.c
+++ b/hw/virtio-blk.c
@@ -494,7 +494,6 @@ VirtIODevice *virtio_blk_init(DeviceState *dev, BlockConf *conf)
s->rq = NULL;
s->sector_mask = (s->conf->logical_block_size / 512) - 1;
bdrv_guess_geometry(s->bs, &cylinders, &heads, &secs);
- bdrv_set_geometry_hint(s->bs, cylinders, heads, secs);
s->vq = virtio_add_queue(&s->vdev, 128, virtio_blk_handle_output);
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 20/26] block.h: bdrv_create2 doesn't exist any more
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (18 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 19/26] Remove un-needed code Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 21/26] block: Convert first_drv to QLIST Kevin Wolf
` (6 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
The bdrv_create2 implementation has disappeared long ago. Remove its
prototype from the header file, too.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.h | 4 ----
1 files changed, 0 insertions(+), 4 deletions(-)
diff --git a/block.h b/block.h
index 4a57dd5..05ad572 100644
--- a/block.h
+++ b/block.h
@@ -57,10 +57,6 @@ BlockDriver *bdrv_find_format(const char *format_name);
BlockDriver *bdrv_find_whitelisted_format(const char *format_name);
int bdrv_create(BlockDriver *drv, const char* filename,
QEMUOptionParameter *options);
-int bdrv_create2(BlockDriver *drv,
- const char *filename, int64_t size_in_sectors,
- const char *backing_file, const char *backing_format,
- int flags);
BlockDriverState *bdrv_new(const char *device_name);
void bdrv_delete(BlockDriverState *bs);
int bdrv_file_open(BlockDriverState **pbs, const char *filename, int flags);
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 21/26] block: Convert first_drv to QLIST
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (19 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 20/26] block.h: bdrv_create2 doesn't exist any more Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 22/26] qemu-img: Eliminate bdrv_new_open() code duplication Kevin Wolf
` (5 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 22 ++++++++++++----------
block_int.h | 2 +-
2 files changed, 13 insertions(+), 11 deletions(-)
diff --git a/block.c b/block.c
index 5d087de..12cf434 100644
--- a/block.c
+++ b/block.c
@@ -58,7 +58,8 @@ static int bdrv_write_em(BlockDriverState *bs, int64_t sector_num,
static QTAILQ_HEAD(, BlockDriverState) bdrv_states =
QTAILQ_HEAD_INITIALIZER(bdrv_states);
-static BlockDriver *first_drv;
+static QLIST_HEAD(, BlockDriver) bdrv_drivers =
+ QLIST_HEAD_INITIALIZER(bdrv_drivers);
/* If non-zero, use only whitelisted block drivers */
static int use_bdrv_whitelist;
@@ -142,8 +143,7 @@ void bdrv_register(BlockDriver *bdrv)
if (!bdrv->bdrv_aio_flush)
bdrv->bdrv_aio_flush = bdrv_aio_flush_em;
- bdrv->next = first_drv;
- first_drv = bdrv;
+ QLIST_INSERT_HEAD(&bdrv_drivers, bdrv, list);
}
/* create a new block device (by default it is empty) */
@@ -162,9 +162,10 @@ BlockDriverState *bdrv_new(const char *device_name)
BlockDriver *bdrv_find_format(const char *format_name)
{
BlockDriver *drv1;
- for(drv1 = first_drv; drv1 != NULL; drv1 = drv1->next) {
- if (!strcmp(drv1->format_name, format_name))
+ QLIST_FOREACH(drv1, &bdrv_drivers, list) {
+ if (!strcmp(drv1->format_name, format_name)) {
return drv1;
+ }
}
return NULL;
}
@@ -265,10 +266,11 @@ static BlockDriver *find_protocol(const char *filename)
len = sizeof(protocol) - 1;
memcpy(protocol, filename, len);
protocol[len] = '\0';
- for(drv1 = first_drv; drv1 != NULL; drv1 = drv1->next) {
+ QLIST_FOREACH(drv1, &bdrv_drivers, list) {
if (drv1->protocol_name &&
- !strcmp(drv1->protocol_name, protocol))
+ !strcmp(drv1->protocol_name, protocol)) {
return drv1;
+ }
}
return NULL;
}
@@ -282,7 +284,7 @@ static BlockDriver *find_hdev_driver(const char *filename)
int score_max = 0, score;
BlockDriver *drv = NULL, *d;
- for (d = first_drv; d; d = d->next) {
+ QLIST_FOREACH(d, &bdrv_drivers, list) {
if (d->bdrv_probe_device) {
score = d->bdrv_probe_device(filename);
if (score > score_max) {
@@ -317,7 +319,7 @@ static BlockDriver *find_image_format(const char *filename)
}
score_max = 0;
- for(drv1 = first_drv; drv1 != NULL; drv1 = drv1->next) {
+ QLIST_FOREACH(drv1, &bdrv_drivers, list) {
if (drv1->bdrv_probe) {
score = drv1->bdrv_probe(buf, ret, filename);
if (score > score_max) {
@@ -1157,7 +1159,7 @@ void bdrv_iterate_format(void (*it)(void *opaque, const char *name),
{
BlockDriver *drv;
- for (drv = first_drv; drv != NULL; drv = drv->next) {
+ QLIST_FOREACH(drv, &bdrv_drivers, list) {
it(opaque, drv->format_name);
}
}
diff --git a/block_int.h b/block_int.h
index 466a38c..d4067ff 100644
--- a/block_int.h
+++ b/block_int.h
@@ -126,7 +126,7 @@ struct BlockDriver {
/* Set if newly created images are not guaranteed to contain only zeros */
int no_zero_init;
- struct BlockDriver *next;
+ QLIST_ENTRY(BlockDriver) list;
};
struct BlockDriverState {
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 22/26] qemu-img: Eliminate bdrv_new_open() code duplication
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (20 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 21/26] block: Convert first_drv to QLIST Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 23/26] qemu-img: Fix BRDV_O_FLAGS typo Kevin Wolf
` (4 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Several commands have code to create a BlockDriverState and open a file.
The bdrv_new_open() function can be used to perform these steps. This
patch converts the qemu-img commands to actually use bdrv_new_open().
Replaced the bdrv_new_open() 'readonly' argument with bdrv_open()-style
flags to support generic flags like BDRV_O_NO_BACKING.
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qemu-img.c | 83 +++++++----------------------------------------------------
1 files changed, 10 insertions(+), 73 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
index 18e43a0..b30effa 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -190,12 +190,11 @@ static int read_password(char *buf, int buf_size)
static BlockDriverState *bdrv_new_open(const char *filename,
const char *fmt,
- int readonly)
+ int flags)
{
BlockDriverState *bs;
BlockDriver *drv;
char password[256];
- int flags = BRDV_O_FLAGS;
bs = bdrv_new("");
if (!bs)
@@ -207,9 +206,6 @@ static BlockDriverState *bdrv_new_open(const char *filename,
} else {
drv = NULL;
}
- if (!readonly) {
- flags |= BDRV_O_RDWR;
- }
if (bdrv_open(bs, filename, flags, drv) < 0) {
error("Could not open '%s'", filename);
}
@@ -349,7 +345,7 @@ static int img_create(int argc, char **argv)
}
}
- bs = bdrv_new_open(backing_file->value.s, fmt, 1);
+ bs = bdrv_new_open(backing_file->value.s, fmt, BRDV_O_FLAGS);
bdrv_get_geometry(bs, &size);
size *= 512;
bdrv_delete(bs);
@@ -384,7 +380,6 @@ static int img_check(int argc, char **argv)
{
int c, ret;
const char *filename, *fmt;
- BlockDriver *drv;
BlockDriverState *bs;
fmt = NULL;
@@ -405,19 +400,7 @@ static int img_check(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new("");
- if (!bs)
- error("Not enough memory");
- if (fmt) {
- drv = bdrv_find_format(fmt);
- if (!drv)
- error("Unknown file format '%s'", fmt);
- } else {
- drv = NULL;
- }
- if (bdrv_open(bs, filename, BRDV_O_FLAGS, drv) < 0) {
- error("Could not open '%s'", filename);
- }
+ bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS);
ret = bdrv_check(bs);
switch(ret) {
case 0:
@@ -443,7 +426,6 @@ static int img_commit(int argc, char **argv)
{
int c, ret;
const char *filename, *fmt;
- BlockDriver *drv;
BlockDriverState *bs;
fmt = NULL;
@@ -464,19 +446,7 @@ static int img_commit(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new("");
- if (!bs)
- error("Not enough memory");
- if (fmt) {
- drv = bdrv_find_format(fmt);
- if (!drv)
- error("Unknown file format '%s'", fmt);
- } else {
- drv = NULL;
- }
- if (bdrv_open(bs, filename, BRDV_O_FLAGS | BDRV_O_RDWR, drv) < 0) {
- error("Could not open '%s'", filename);
- }
+ bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS | BDRV_O_RDWR);
ret = bdrv_commit(bs);
switch(ret) {
case 0:
@@ -633,7 +603,7 @@ static int img_convert(int argc, char **argv)
total_sectors = 0;
for (bs_i = 0; bs_i < bs_n; bs_i++) {
- bs[bs_i] = bdrv_new_open(argv[optind + bs_i], fmt, 1);
+ bs[bs_i] = bdrv_new_open(argv[optind + bs_i], fmt, BRDV_O_FLAGS);
if (!bs[bs_i])
error("Could not open '%s'", argv[optind + bs_i]);
bdrv_get_geometry(bs[bs_i], &bs_sectors);
@@ -691,7 +661,7 @@ static int img_convert(int argc, char **argv)
}
}
- out_bs = bdrv_new_open(out_filename, out_fmt, 0);
+ out_bs = bdrv_new_open(out_filename, out_fmt, BRDV_O_FLAGS | BDRV_O_RDWR);
bs_i = 0;
bs_offset = 0;
@@ -889,7 +859,6 @@ static int img_info(int argc, char **argv)
{
int c;
const char *filename, *fmt;
- BlockDriver *drv;
BlockDriverState *bs;
char fmt_name[128], size_buf[128], dsize_buf[128];
uint64_t total_sectors;
@@ -916,19 +885,7 @@ static int img_info(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new("");
- if (!bs)
- error("Not enough memory");
- if (fmt) {
- drv = bdrv_find_format(fmt);
- if (!drv)
- error("Unknown file format '%s'", fmt);
- } else {
- drv = NULL;
- }
- if (bdrv_open(bs, filename, BRDV_O_FLAGS | BDRV_O_NO_BACKING, drv) < 0) {
- error("Could not open '%s'", filename);
- }
+ bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS | BDRV_O_NO_BACKING);
bdrv_get_format(bs, fmt_name, sizeof(fmt_name));
bdrv_get_geometry(bs, &total_sectors);
get_human_readable_size(size_buf, sizeof(size_buf), total_sectors * 512);
@@ -1028,13 +985,7 @@ static int img_snapshot(int argc, char **argv)
filename = argv[optind++];
/* Open the image */
- bs = bdrv_new("");
- if (!bs)
- error("Not enough memory");
-
- if (bdrv_open(bs, filename, bdrv_oflags, NULL) < 0) {
- error("Could not open '%s'", filename);
- }
+ bs = bdrv_new_open(filename, NULL, bdrv_oflags);
/* Perform the requested action */
switch(action) {
@@ -1080,7 +1031,7 @@ static int img_snapshot(int argc, char **argv)
static int img_rebase(int argc, char **argv)
{
BlockDriverState *bs, *bs_old_backing, *bs_new_backing;
- BlockDriver *drv, *old_backing_drv, *new_backing_drv;
+ BlockDriver *old_backing_drv, *new_backing_drv;
char *filename;
const char *fmt, *out_basefmt, *out_baseimg;
int c, flags, ret;
@@ -1124,22 +1075,8 @@ static int img_rebase(int argc, char **argv)
* Ignore the old backing file for unsafe rebase in case we want to correct
* the reference to a renamed or moved backing file.
*/
- bs = bdrv_new("");
- if (!bs)
- error("Not enough memory");
-
- drv = NULL;
- if (fmt) {
- drv = bdrv_find_format(fmt);
- if (drv == NULL) {
- error("Invalid format name: '%s'", fmt);
- }
- }
-
flags = BRDV_O_FLAGS | BDRV_O_RDWR | (unsafe ? BDRV_O_NO_BACKING : 0);
- if (bdrv_open(bs, filename, flags, drv) < 0) {
- error("Could not open '%s'", filename);
- }
+ bs = bdrv_new_open(filename, fmt, flags);
/* Find the right drivers for the backing files */
old_backing_drv = NULL;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 23/26] qemu-img: Fix BRDV_O_FLAGS typo
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (21 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 22/26] qemu-img: Eliminate bdrv_new_open() code duplication Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 24/26] linux-aio: Fix typo in read() EINTR check Kevin Wolf
` (3 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
It should be BDRV_O_FLAGS instead of BRDV_O_FLAGS.
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qemu-img.c | 20 ++++++++++----------
1 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
index b30effa..7203b8b 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -37,7 +37,7 @@ typedef struct img_cmd_t {
} img_cmd_t;
/* Default to cache=writeback as data integrity is not important for qemu-tcg. */
-#define BRDV_O_FLAGS BDRV_O_CACHE_WB
+#define BDRV_O_FLAGS BDRV_O_CACHE_WB
static void QEMU_NORETURN error(const char *fmt, ...)
{
@@ -345,7 +345,7 @@ static int img_create(int argc, char **argv)
}
}
- bs = bdrv_new_open(backing_file->value.s, fmt, BRDV_O_FLAGS);
+ bs = bdrv_new_open(backing_file->value.s, fmt, BDRV_O_FLAGS);
bdrv_get_geometry(bs, &size);
size *= 512;
bdrv_delete(bs);
@@ -400,7 +400,7 @@ static int img_check(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS);
+ bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS);
ret = bdrv_check(bs);
switch(ret) {
case 0:
@@ -446,7 +446,7 @@ static int img_commit(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS | BDRV_O_RDWR);
+ bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS | BDRV_O_RDWR);
ret = bdrv_commit(bs);
switch(ret) {
case 0:
@@ -603,7 +603,7 @@ static int img_convert(int argc, char **argv)
total_sectors = 0;
for (bs_i = 0; bs_i < bs_n; bs_i++) {
- bs[bs_i] = bdrv_new_open(argv[optind + bs_i], fmt, BRDV_O_FLAGS);
+ bs[bs_i] = bdrv_new_open(argv[optind + bs_i], fmt, BDRV_O_FLAGS);
if (!bs[bs_i])
error("Could not open '%s'", argv[optind + bs_i]);
bdrv_get_geometry(bs[bs_i], &bs_sectors);
@@ -661,7 +661,7 @@ static int img_convert(int argc, char **argv)
}
}
- out_bs = bdrv_new_open(out_filename, out_fmt, BRDV_O_FLAGS | BDRV_O_RDWR);
+ out_bs = bdrv_new_open(out_filename, out_fmt, BDRV_O_FLAGS | BDRV_O_RDWR);
bs_i = 0;
bs_offset = 0;
@@ -885,7 +885,7 @@ static int img_info(int argc, char **argv)
help();
filename = argv[optind++];
- bs = bdrv_new_open(filename, fmt, BRDV_O_FLAGS | BDRV_O_NO_BACKING);
+ bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS | BDRV_O_NO_BACKING);
bdrv_get_format(bs, fmt_name, sizeof(fmt_name));
bdrv_get_geometry(bs, &total_sectors);
get_human_readable_size(size_buf, sizeof(size_buf), total_sectors * 512);
@@ -1075,7 +1075,7 @@ static int img_rebase(int argc, char **argv)
* Ignore the old backing file for unsafe rebase in case we want to correct
* the reference to a renamed or moved backing file.
*/
- flags = BRDV_O_FLAGS | BDRV_O_RDWR | (unsafe ? BDRV_O_NO_BACKING : 0);
+ flags = BDRV_O_FLAGS | BDRV_O_RDWR | (unsafe ? BDRV_O_NO_BACKING : 0);
bs = bdrv_new_open(filename, fmt, flags);
/* Find the right drivers for the backing files */
@@ -1106,7 +1106,7 @@ static int img_rebase(int argc, char **argv)
bs_old_backing = bdrv_new("old_backing");
bdrv_get_backing_filename(bs, backing_name, sizeof(backing_name));
- if (bdrv_open(bs_old_backing, backing_name, BRDV_O_FLAGS,
+ if (bdrv_open(bs_old_backing, backing_name, BDRV_O_FLAGS,
old_backing_drv))
{
error("Could not open old backing file '%s'", backing_name);
@@ -1114,7 +1114,7 @@ static int img_rebase(int argc, char **argv)
}
bs_new_backing = bdrv_new("new_backing");
- if (bdrv_open(bs_new_backing, out_baseimg, BRDV_O_FLAGS | BDRV_O_RDWR,
+ if (bdrv_open(bs_new_backing, out_baseimg, BDRV_O_FLAGS | BDRV_O_RDWR,
new_backing_drv))
{
error("Could not open new backing file '%s'", out_baseimg);
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 24/26] linux-aio: Fix typo in read() EINTR check
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (22 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 23/26] qemu-img: Fix BRDV_O_FLAGS typo Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 25/26] qcow2: Use QLIST_FOREACH_SAFE macro Kevin Wolf
` (2 subsequent siblings)
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
linux-aio.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/linux-aio.c b/linux-aio.c
index 5e892b0..68f4b3d 100644
--- a/linux-aio.c
+++ b/linux-aio.c
@@ -123,7 +123,7 @@ static void qemu_laio_completion_cb(void *opaque)
do {
ret = read(s->efd, &val, sizeof(val));
- } while (ret == 1 && errno == EINTR);
+ } while (ret == -1 && errno == EINTR);
if (ret == -1 && errno == EAGAIN)
break;
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 25/26] qcow2: Use QLIST_FOREACH_SAFE macro
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (23 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 24/26] linux-aio: Fix typo in read() EINTR check Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge() Kevin Wolf
2010-04-23 18:49 ` [Qemu-devel] Re: [PULL 00/26] Block patches Anthony Liguori
26 siblings, 0 replies; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/qcow2.c | 10 ++--------
1 files changed, 2 insertions(+), 8 deletions(-)
diff --git a/block/qcow2.c b/block/qcow2.c
index 30ded6a..f3e3cba 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -539,14 +539,8 @@ static void run_dependent_requests(QCowL2Meta *m)
QLIST_REMOVE(m, next_in_flight);
}
- /*
- * Restart all dependent requests.
- * Can't use QLIST_FOREACH here - the next link might not be the same
- * any more after the callback (request could depend on a different
- * request now)
- */
- for (req = m->dependent_requests.lh_first; req != NULL; req = next) {
- next = req->next_depend.le_next;
+ /* Restart all dependent requests */
+ QLIST_FOREACH_SAFE(req, &m->dependent_requests, next_depend, next) {
qcow_aio_write_cb(req, 0);
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge()
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (24 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 25/26] qcow2: Use QLIST_FOREACH_SAFE macro Kevin Wolf
@ 2010-04-23 15:30 ` Kevin Wolf
2010-04-23 20:22 ` Ryan Harper
2010-04-23 18:49 ` [Qemu-devel] Re: [PULL 00/26] Block patches Anthony Liguori
26 siblings, 1 reply; 30+ messages in thread
From: Kevin Wolf @ 2010-04-23 15:30 UTC (permalink / raw)
To: aliguori; +Cc: kwolf, qemu-devel
From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
A new iovec array is allocated when creating a merged write request.
This patch ensures that the iovec array is deleted in addition to its
qiov owner.
Reported-by: Leszek Urbanski <tygrys@moo.pl>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/block.c b/block.c
index 12cf434..7974215 100644
--- a/block.c
+++ b/block.c
@@ -1739,6 +1739,9 @@ static void multiwrite_user_cb(MultiwriteCB *mcb)
for (i = 0; i < mcb->num_callbacks; i++) {
mcb->callbacks[i].cb(mcb->callbacks[i].opaque, mcb->error);
+ if (mcb->callbacks[i].free_qiov) {
+ qemu_iovec_destroy(mcb->callbacks[i].free_qiov);
+ }
qemu_free(mcb->callbacks[i].free_qiov);
qemu_vfree(mcb->callbacks[i].free_buf);
}
--
1.6.6.1
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Qemu-devel] Re: [PULL 00/26] Block patches
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
` (25 preceding siblings ...)
2010-04-23 15:30 ` [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge() Kevin Wolf
@ 2010-04-23 18:49 ` Anthony Liguori
26 siblings, 0 replies; 30+ messages in thread
From: Anthony Liguori @ 2010-04-23 18:49 UTC (permalink / raw)
To: Kevin Wolf; +Cc: qemu-devel
On 04/23/2010 10:30 AM, Kevin Wolf wrote:
> Hi Anthony,
>
> this is the first part of the block patches that accumulated in my block branch
> during your absence. I consider these patches ready to be merged into master.
> I've not included another 13 patches that might still need discussion/review or
> depend on such patches. I'm going to send another pull request for them some
> time next week.
>
> Kevin
>
> The following changes since commit 6c557ab975fc8e5edb4167a241266c7c4657054a:
> Serge Ziryukin (1):
> audio/sdlaudio: remove unused variable
>
> are available in the git repository at:
>
> git://repo.or.cz/qemu/kevin.git for-anthony
>
Pulled. Thanks for gathering these up!
Regards,
Anthony Liguori
> Bruce Rogers (1):
> Remove un-needed code
>
> Christoph Hellwig (3):
> block: get rid of the BDRV_O_FILE flag
> block: split raw_getlength
> cleanup block driver option handling in vl.c
>
> Kevin Wolf (14):
> qemu-config: qemu_read_config_file() reads the normal config file
> qemu-config: Make qemu_config_parse more generic
> blkdebug: Basic request passthrough
> blkdebug: Inject errors
> Make qemu-config available for tools
> blkdebug: Add events and rules
> qcow2: Trigger blkdebug events
> qcow2: Fix creation of large images
> Replace calls of old bdrv_open
> qcow2: Return 0/-errno in write_l2_entries
> qcow2: Fix error return code in qcow2_alloc_cluster_link_l2
> qcow2: Return 0/-errno in write_l1_entry
> qcow2: Return 0/-errno in l2_allocate
> block.h: bdrv_create2 doesn't exist any more
>
> Stefan Hajnoczi (8):
> block: Do not export bdrv_first
> block: Convert bdrv_first to QTAILQ
> block: Convert first_drv to QLIST
> qemu-img: Eliminate bdrv_new_open() code duplication
> qemu-img: Fix BRDV_O_FLAGS typo
> linux-aio: Fix typo in read() EINTR check
> qcow2: Use QLIST_FOREACH_SAFE macro
> block: Free iovec arrays allocated by multiwrite_merge()
>
> Makefile.objs | 6 +-
> block-migration.c | 63 ++++---
> block.c | 119 +++++++------
> block.h | 66 ++++++-
> block/blkdebug.c | 475 ++++++++++++++++++++++++++++++++++++++++++++++++
> block/qcow2-cluster.c | 78 +++++---
> block/qcow2-refcount.c | 18 ++
> block/qcow2.c | 63 +++++--
> block/raw-posix.c | 65 +++++---
> block/vmdk.c | 2 +-
> block/vvfat.c | 5 +-
> block_int.h | 9 +-
> hw/qdev-properties.c | 19 ++-
> hw/qdev.h | 1 -
> hw/virtio-blk.c | 1 -
> hw/xen_disk.c | 2 +-
> linux-aio.c | 2 +-
> monitor.c | 2 +-
> qemu-config.c | 48 +++---
> qemu-config.h | 4 +-
> qemu-img.c | 93 ++--------
> qemu-io.c | 28 ++--
> qemu-nbd.c | 2 +-
> vl.c | 82 +++------
> 24 files changed, 902 insertions(+), 351 deletions(-)
> create mode 100644 block/blkdebug.c
>
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge()
2010-04-23 15:30 ` [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge() Kevin Wolf
@ 2010-04-23 20:22 ` Ryan Harper
2010-04-25 1:42 ` Aurelien Jarno
0 siblings, 1 reply; 30+ messages in thread
From: Ryan Harper @ 2010-04-23 20:22 UTC (permalink / raw)
To: Kevin Wolf; +Cc: aliguori, qemu-devel
* Kevin Wolf <kwolf@redhat.com> [2010-04-23 11:29]:
> From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
>
> A new iovec array is allocated when creating a merged write request.
> This patch ensures that the iovec array is deleted in addition to its
> qiov owner.
>
Submit for [STABLE] ?
> Reported-by: Leszek Urbanski <tygrys@moo.pl>
> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> ---
> block.c | 3 +++
> 1 files changed, 3 insertions(+), 0 deletions(-)
>
> diff --git a/block.c b/block.c
> index 12cf434..7974215 100644
> --- a/block.c
> +++ b/block.c
> @@ -1739,6 +1739,9 @@ static void multiwrite_user_cb(MultiwriteCB *mcb)
>
> for (i = 0; i < mcb->num_callbacks; i++) {
> mcb->callbacks[i].cb(mcb->callbacks[i].opaque, mcb->error);
> + if (mcb->callbacks[i].free_qiov) {
> + qemu_iovec_destroy(mcb->callbacks[i].free_qiov);
> + }
> qemu_free(mcb->callbacks[i].free_qiov);
> qemu_vfree(mcb->callbacks[i].free_buf);
> }
> --
> 1.6.6.1
>
>
--
Ryan Harper
Software Engineer; Linux Technology Center
IBM Corp., Austin, Tx
ryanh@us.ibm.com
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge()
2010-04-23 20:22 ` Ryan Harper
@ 2010-04-25 1:42 ` Aurelien Jarno
0 siblings, 0 replies; 30+ messages in thread
From: Aurelien Jarno @ 2010-04-25 1:42 UTC (permalink / raw)
To: Ryan Harper; +Cc: Kevin Wolf, aliguori, qemu-devel
On Fri, Apr 23, 2010 at 03:22:14PM -0500, Ryan Harper wrote:
> * Kevin Wolf <kwolf@redhat.com> [2010-04-23 11:29]:
> > From: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
> >
> > A new iovec array is allocated when creating a merged write request.
> > This patch ensures that the iovec array is deleted in addition to its
> > qiov owner.
> >
>
> Submit for [STABLE] ?
>
I have just cherry-picked it into stable
--
Aurelien Jarno GPG: 1024D/F1BCDB73
aurelien@aurel32.net http://www.aurel32.net
^ permalink raw reply [flat|nested] 30+ messages in thread
end of thread, other threads:[~2010-04-25 1:42 UTC | newest]
Thread overview: 30+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-04-23 15:30 [Qemu-devel] [PULL 00/26] Block patches Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 01/26] qemu-config: qemu_read_config_file() reads the normal config file Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 02/26] qemu-config: Make qemu_config_parse more generic Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 03/26] blkdebug: Basic request passthrough Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 04/26] blkdebug: Inject errors Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 05/26] Make qemu-config available for tools Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 06/26] blkdebug: Add events and rules Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 07/26] qcow2: Trigger blkdebug events Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 08/26] qcow2: Fix creation of large images Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 09/26] Replace calls of old bdrv_open Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 10/26] block: get rid of the BDRV_O_FILE flag Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 11/26] block: split raw_getlength Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 12/26] qcow2: Return 0/-errno in write_l2_entries Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 13/26] qcow2: Fix error return code in qcow2_alloc_cluster_link_l2 Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 14/26] qcow2: Return 0/-errno in write_l1_entry Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 15/26] qcow2: Return 0/-errno in l2_allocate Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 16/26] cleanup block driver option handling in vl.c Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 17/26] block: Do not export bdrv_first Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 18/26] block: Convert bdrv_first to QTAILQ Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 19/26] Remove un-needed code Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 20/26] block.h: bdrv_create2 doesn't exist any more Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 21/26] block: Convert first_drv to QLIST Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 22/26] qemu-img: Eliminate bdrv_new_open() code duplication Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 23/26] qemu-img: Fix BRDV_O_FLAGS typo Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 24/26] linux-aio: Fix typo in read() EINTR check Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 25/26] qcow2: Use QLIST_FOREACH_SAFE macro Kevin Wolf
2010-04-23 15:30 ` [Qemu-devel] [PATCH 26/26] block: Free iovec arrays allocated by multiwrite_merge() Kevin Wolf
2010-04-23 20:22 ` Ryan Harper
2010-04-25 1:42 ` Aurelien Jarno
2010-04-23 18:49 ` [Qemu-devel] Re: [PULL 00/26] Block patches Anthony Liguori
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.