* [PULL 01/34] block/accounting: Use lock guard macros
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 02/34] block/curl: " Kevin Wolf
` (33 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Gan Qixin <ganqixin@huawei.com>
Replace manual lock()/unlock() calls with lock guard macros
(QEMU_LOCK_GUARD/WITH_QEMU_LOCK_GUARD) in block/accounting.
Signed-off-by: Gan Qixin <ganqixin@huawei.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20201203075055.127773-2-ganqixin@huawei.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/accounting.c | 32 +++++++++++++++-----------------
1 file changed, 15 insertions(+), 17 deletions(-)
diff --git a/block/accounting.c b/block/accounting.c
index 8d41c8a83a..2030851d79 100644
--- a/block/accounting.c
+++ b/block/accounting.c
@@ -199,29 +199,27 @@ static void block_account_one_io(BlockAcctStats *stats, BlockAcctCookie *cookie,
return;
}
- qemu_mutex_lock(&stats->lock);
-
- if (failed) {
- stats->failed_ops[cookie->type]++;
- } else {
- stats->nr_bytes[cookie->type] += cookie->bytes;
- stats->nr_ops[cookie->type]++;
- }
+ WITH_QEMU_LOCK_GUARD(&stats->lock) {
+ if (failed) {
+ stats->failed_ops[cookie->type]++;
+ } else {
+ stats->nr_bytes[cookie->type] += cookie->bytes;
+ stats->nr_ops[cookie->type]++;
+ }
- block_latency_histogram_account(&stats->latency_histogram[cookie->type],
- latency_ns);
+ block_latency_histogram_account(&stats->latency_histogram[cookie->type],
+ latency_ns);
- if (!failed || stats->account_failed) {
- stats->total_time_ns[cookie->type] += latency_ns;
- stats->last_access_time_ns = time_ns;
+ if (!failed || stats->account_failed) {
+ stats->total_time_ns[cookie->type] += latency_ns;
+ stats->last_access_time_ns = time_ns;
- QSLIST_FOREACH(s, &stats->intervals, entries) {
- timed_average_account(&s->latency[cookie->type], latency_ns);
+ QSLIST_FOREACH(s, &stats->intervals, entries) {
+ timed_average_account(&s->latency[cookie->type], latency_ns);
+ }
}
}
- qemu_mutex_unlock(&stats->lock);
-
cookie->type = BLOCK_ACCT_NONE;
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 02/34] block/curl: Use lock guard macros
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
2020-12-11 17:07 ` [PULL 01/34] block/accounting: Use lock guard macros Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 03/34] block/throttle-groups: " Kevin Wolf
` (32 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Gan Qixin <ganqixin@huawei.com>
Replace manual lock()/unlock() calls with lock guard macros
(QEMU_LOCK_GUARD/WITH_QEMU_LOCK_GUARD) in block/curl.
Signed-off-by: Gan Qixin <ganqixin@huawei.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20201203075055.127773-3-ganqixin@huawei.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/curl.c | 28 ++++++++++++++--------------
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/block/curl.c b/block/curl.c
index 4f907c47be..d24a4c5897 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -564,23 +564,23 @@ static void curl_detach_aio_context(BlockDriverState *bs)
BDRVCURLState *s = bs->opaque;
int i;
- qemu_mutex_lock(&s->mutex);
- for (i = 0; i < CURL_NUM_STATES; i++) {
- if (s->states[i].in_use) {
- curl_clean_state(&s->states[i]);
+ WITH_QEMU_LOCK_GUARD(&s->mutex) {
+ for (i = 0; i < CURL_NUM_STATES; i++) {
+ if (s->states[i].in_use) {
+ curl_clean_state(&s->states[i]);
+ }
+ if (s->states[i].curl) {
+ curl_easy_cleanup(s->states[i].curl);
+ s->states[i].curl = NULL;
+ }
+ g_free(s->states[i].orig_buf);
+ s->states[i].orig_buf = NULL;
}
- if (s->states[i].curl) {
- curl_easy_cleanup(s->states[i].curl);
- s->states[i].curl = NULL;
+ if (s->multi) {
+ curl_multi_cleanup(s->multi);
+ s->multi = NULL;
}
- g_free(s->states[i].orig_buf);
- s->states[i].orig_buf = NULL;
- }
- if (s->multi) {
- curl_multi_cleanup(s->multi);
- s->multi = NULL;
}
- qemu_mutex_unlock(&s->mutex);
timer_del(&s->timer);
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 03/34] block/throttle-groups: Use lock guard macros
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
2020-12-11 17:07 ` [PULL 01/34] block/accounting: Use lock guard macros Kevin Wolf
2020-12-11 17:07 ` [PULL 02/34] block/curl: " Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 04/34] block/iscsi: " Kevin Wolf
` (31 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Gan Qixin <ganqixin@huawei.com>
Replace manual lock()/unlock() calls with lock guard macros
(QEMU_LOCK_GUARD/WITH_QEMU_LOCK_GUARD) in block/throttle-groups.
Signed-off-by: Gan Qixin <ganqixin@huawei.com>
Message-Id: <20201203075055.127773-4-ganqixin@huawei.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/throttle-groups.c | 48 ++++++++++++++++++++---------------------
1 file changed, 23 insertions(+), 25 deletions(-)
diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index e2f2813c0f..abd16ed9db 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -546,7 +546,7 @@ void throttle_group_register_tgm(ThrottleGroupMember *tgm,
tgm->aio_context = ctx;
qatomic_set(&tgm->restart_pending, 0);
- qemu_mutex_lock(&tg->lock);
+ QEMU_LOCK_GUARD(&tg->lock);
/* If the ThrottleGroup is new set this ThrottleGroupMember as the token */
for (i = 0; i < 2; i++) {
if (!tg->tokens[i]) {
@@ -565,8 +565,6 @@ void throttle_group_register_tgm(ThrottleGroupMember *tgm,
qemu_co_mutex_init(&tgm->throttled_reqs_lock);
qemu_co_queue_init(&tgm->throttled_reqs[0]);
qemu_co_queue_init(&tgm->throttled_reqs[1]);
-
- qemu_mutex_unlock(&tg->lock);
}
/* Unregister a ThrottleGroupMember from its group, removing it from the list,
@@ -594,25 +592,25 @@ void throttle_group_unregister_tgm(ThrottleGroupMember *tgm)
/* Wait for throttle_group_restart_queue_entry() coroutines to finish */
AIO_WAIT_WHILE(tgm->aio_context, qatomic_read(&tgm->restart_pending) > 0);
- qemu_mutex_lock(&tg->lock);
- for (i = 0; i < 2; i++) {
- assert(tgm->pending_reqs[i] == 0);
- assert(qemu_co_queue_empty(&tgm->throttled_reqs[i]));
- assert(!timer_pending(tgm->throttle_timers.timers[i]));
- if (tg->tokens[i] == tgm) {
- token = throttle_group_next_tgm(tgm);
- /* Take care of the case where this is the last tgm in the group */
- if (token == tgm) {
- token = NULL;
+ WITH_QEMU_LOCK_GUARD(&tg->lock) {
+ for (i = 0; i < 2; i++) {
+ assert(tgm->pending_reqs[i] == 0);
+ assert(qemu_co_queue_empty(&tgm->throttled_reqs[i]));
+ assert(!timer_pending(tgm->throttle_timers.timers[i]));
+ if (tg->tokens[i] == tgm) {
+ token = throttle_group_next_tgm(tgm);
+ /* Take care of the case where this is the last tgm in the group */
+ if (token == tgm) {
+ token = NULL;
+ }
+ tg->tokens[i] = token;
}
- tg->tokens[i] = token;
}
- }
- /* remove the current tgm from the list */
- QLIST_REMOVE(tgm, round_robin);
- throttle_timers_destroy(&tgm->throttle_timers);
- qemu_mutex_unlock(&tg->lock);
+ /* remove the current tgm from the list */
+ QLIST_REMOVE(tgm, round_robin);
+ throttle_timers_destroy(&tgm->throttle_timers);
+ }
throttle_group_unref(&tg->ts);
tgm->throttle_state = NULL;
@@ -638,14 +636,14 @@ void throttle_group_detach_aio_context(ThrottleGroupMember *tgm)
assert(qemu_co_queue_empty(&tgm->throttled_reqs[1]));
/* Kick off next ThrottleGroupMember, if necessary */
- qemu_mutex_lock(&tg->lock);
- for (i = 0; i < 2; i++) {
- if (timer_pending(tt->timers[i])) {
- tg->any_timer_armed[i] = false;
- schedule_next_request(tgm, i);
+ WITH_QEMU_LOCK_GUARD(&tg->lock) {
+ for (i = 0; i < 2; i++) {
+ if (timer_pending(tt->timers[i])) {
+ tg->any_timer_armed[i] = false;
+ schedule_next_request(tgm, i);
+ }
}
}
- qemu_mutex_unlock(&tg->lock);
throttle_timers_detach_aio_context(tt);
tgm->aio_context = NULL;
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 04/34] block/iscsi: Use lock guard macros
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (2 preceding siblings ...)
2020-12-11 17:07 ` [PULL 03/34] block/throttle-groups: " Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 05/34] meson: Detect libfuse Kevin Wolf
` (30 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Gan Qixin <ganqixin@huawei.com>
Replace manual lock()/unlock() calls with lock guard macros
(QEMU_LOCK_GUARD/WITH_QEMU_LOCK_GUARD) in block/iscsi.
Signed-off-by: Gan Qixin <ganqixin@huawei.com>
Message-Id: <20201203075055.127773-5-ganqixin@huawei.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/iscsi.c | 50 ++++++++++++++++++++++++--------------------------
1 file changed, 24 insertions(+), 26 deletions(-)
diff --git a/block/iscsi.c b/block/iscsi.c
index e30a7e3606..7d4b3b56d5 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -322,25 +322,23 @@ iscsi_aio_cancel(BlockAIOCB *blockacb)
IscsiAIOCB *acb = (IscsiAIOCB *)blockacb;
IscsiLun *iscsilun = acb->iscsilun;
- qemu_mutex_lock(&iscsilun->mutex);
+ WITH_QEMU_LOCK_GUARD(&iscsilun->mutex) {
- /* If it was cancelled or completed already, our work is done here */
- if (acb->cancelled || acb->status != -EINPROGRESS) {
- qemu_mutex_unlock(&iscsilun->mutex);
- return;
- }
+ /* If it was cancelled or completed already, our work is done here */
+ if (acb->cancelled || acb->status != -EINPROGRESS) {
+ return;
+ }
- acb->cancelled = true;
+ acb->cancelled = true;
- qemu_aio_ref(acb); /* released in iscsi_abort_task_cb() */
+ qemu_aio_ref(acb); /* released in iscsi_abort_task_cb() */
- /* send a task mgmt call to the target to cancel the task on the target */
- if (iscsi_task_mgmt_abort_task_async(iscsilun->iscsi, acb->task,
- iscsi_abort_task_cb, acb) < 0) {
- qemu_aio_unref(acb); /* since iscsi_abort_task_cb() won't be called */
+ /* send a task mgmt call to the target to cancel the task on the target */
+ if (iscsi_task_mgmt_abort_task_async(iscsilun->iscsi, acb->task,
+ iscsi_abort_task_cb, acb) < 0) {
+ qemu_aio_unref(acb); /* since iscsi_abort_task_cb() won't be called */
+ }
}
-
- qemu_mutex_unlock(&iscsilun->mutex);
}
static const AIOCBInfo iscsi_aiocb_info = {
@@ -375,22 +373,22 @@ static void iscsi_timed_check_events(void *opaque)
{
IscsiLun *iscsilun = opaque;
- qemu_mutex_lock(&iscsilun->mutex);
+ WITH_QEMU_LOCK_GUARD(&iscsilun->mutex) {
+ /* check for timed out requests */
+ iscsi_service(iscsilun->iscsi, 0);
- /* check for timed out requests */
- iscsi_service(iscsilun->iscsi, 0);
+ if (iscsilun->request_timed_out) {
+ iscsilun->request_timed_out = false;
+ iscsi_reconnect(iscsilun->iscsi);
+ }
- if (iscsilun->request_timed_out) {
- iscsilun->request_timed_out = false;
- iscsi_reconnect(iscsilun->iscsi);
+ /*
+ * newer versions of libiscsi may return zero events. Ensure we are
+ * able to return to service once this situation changes.
+ */
+ iscsi_set_events(iscsilun);
}
- /* newer versions of libiscsi may return zero events. Ensure we are able
- * to return to service once this situation changes. */
- iscsi_set_events(iscsilun);
-
- qemu_mutex_unlock(&iscsilun->mutex);
-
timer_mod(iscsilun->event_timer,
qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + EVENT_INTERVAL);
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 05/34] meson: Detect libfuse
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (3 preceding siblings ...)
2020-12-11 17:07 ` [PULL 04/34] block/iscsi: " Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 06/34] fuse: Allow exporting BDSs via FUSE Kevin Wolf
` (29 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-2-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
meson_options.txt | 2 ++
configure | 7 +++++++
meson.build | 6 ++++++
3 files changed, 15 insertions(+)
diff --git a/meson_options.txt b/meson_options.txt
index f6f64785fe..8f9f2e3df6 100644
--- a/meson_options.txt
+++ b/meson_options.txt
@@ -66,6 +66,8 @@ option('virtiofsd', type: 'feature', value: 'auto',
description: 'build virtiofs daemon (virtiofsd)')
option('vhost_user_blk_server', type: 'feature', value: 'auto',
description: 'build vhost-user-blk server')
+option('fuse', type: 'feature', value: 'auto',
+ description: 'FUSE block device export')
option('capstone', type: 'combo', value: 'auto',
choices: ['disabled', 'enabled', 'auto', 'system', 'internal'],
diff --git a/configure b/configure
index 18c26e0389..b2f96c0da2 100755
--- a/configure
+++ b/configure
@@ -449,6 +449,7 @@ meson=""
ninja=""
skip_meson=no
gettext=""
+fuse="auto"
bogus_os="no"
malloc_trim="auto"
@@ -1525,6 +1526,10 @@ for opt do
;;
--disable-libdaxctl) libdaxctl=no
;;
+ --enable-fuse) fuse="enabled"
+ ;;
+ --disable-fuse) fuse="disabled"
+ ;;
*)
echo "ERROR: unknown option $opt"
echo "Try '$0 --help' for more information"
@@ -1850,6 +1855,7 @@ disabled with --disable-FEATURE, default is enabled if available:
xkbcommon xkbcommon support
rng-none dummy RNG, avoid using /dev/(u)random and getrandom()
libdaxctl libdaxctl support
+ fuse FUSE block device export
NOTE: The object files are built at the place where configure is launched
EOF
@@ -7014,6 +7020,7 @@ NINJA=$ninja $meson setup \
-Diconv=$iconv -Dcurses=$curses -Dlibudev=$libudev\
-Ddocs=$docs -Dsphinx_build=$sphinx_build -Dinstall_blobs=$blobs \
-Dvhost_user_blk_server=$vhost_user_blk_server \
+ -Dfuse=$fuse \
$cross_arg \
"$PWD" "$source_path"
diff --git a/meson.build b/meson.build
index 9ea05ab49f..6e8ef151d8 100644
--- a/meson.build
+++ b/meson.build
@@ -773,6 +773,10 @@ elif get_option('vhost_user_blk_server').disabled() or not have_system
have_vhost_user_blk_server = false
endif
+fuse = dependency('fuse3', required: get_option('fuse'),
+ version: '>=3.1', method: 'pkg-config',
+ static: enable_static)
+
#################
# config-host.h #
#################
@@ -807,6 +811,7 @@ config_host_data.set('CONFIG_KEYUTILS', keyutils.found())
config_host_data.set('CONFIG_GETTID', has_gettid)
config_host_data.set('CONFIG_MALLOC_TRIM', has_malloc_trim)
config_host_data.set('CONFIG_STATX', has_statx)
+config_host_data.set('CONFIG_FUSE', fuse.found())
config_host_data.set('QEMU_VERSION', '"@0@"'.format(meson.project_version()))
config_host_data.set('QEMU_VERSION_MAJOR', meson.project_version().split('.')[0])
config_host_data.set('QEMU_VERSION_MINOR', meson.project_version().split('.')[1])
@@ -2208,6 +2213,7 @@ endif
summary_info += {'thread sanitizer': config_host.has_key('CONFIG_TSAN')}
summary_info += {'rng-none': config_host.has_key('CONFIG_RNG_NONE')}
summary_info += {'Linux keyring': config_host.has_key('CONFIG_SECRET_KEYRING')}
+summary_info += {'FUSE exports': fuse.found()}
summary(summary_info, bool_yn: true)
if not supported_cpus.contains(cpu)
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 06/34] fuse: Allow exporting BDSs via FUSE
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (4 preceding siblings ...)
2020-12-11 17:07 ` [PULL 05/34] meson: Detect libfuse Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 07/34] fuse: Implement standard FUSE operations Kevin Wolf
` (28 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
block-export-add type=fuse allows mounting block graph nodes via FUSE on
some existing regular file. That file should then appears like a raw
disk image, and accesses to it result in accesses to the exported BDS.
Right now, we only implement the necessary block export functions to set
it up and shut it down. We do not implement any access functions, so
accessing the mount point only results in errors. This will be
addressed by a followup patch.
We keep a hash table of exported mount points, because we want to be
able to detect when users try to use a mount point twice. This is
because we invoke stat() to check whether the given mount point is a
regular file, but if that file is served by ourselves (because it is
already used as a mount point), then this stat() would have to be served
by ourselves, too, which is impossible to do while we (as the caller)
are waiting for it to settle. Therefore, keep track of mount point
paths to at least catch the most obvious instances of that problem.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-3-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qapi/block-export.json | 23 ++-
include/block/fuse.h | 30 ++++
block.c | 1 +
block/export/export.c | 4 +
block/export/fuse.c | 295 +++++++++++++++++++++++++++++++++++++++
MAINTAINERS | 6 +
block/export/meson.build | 2 +
7 files changed, 359 insertions(+), 2 deletions(-)
create mode 100644 include/block/fuse.h
create mode 100644 block/export/fuse.c
diff --git a/qapi/block-export.json b/qapi/block-export.json
index 4eeac7842d..430bc69f35 100644
--- a/qapi/block-export.json
+++ b/qapi/block-export.json
@@ -120,6 +120,21 @@
'*logical-block-size': 'size',
'*num-queues': 'uint16'} }
+##
+# @BlockExportOptionsFuse:
+#
+# Options for exporting a block graph node on some (file) mountpoint
+# as a raw image.
+#
+# @mountpoint: Path on which to export the block device via FUSE.
+# This must point to an existing regular file.
+#
+# Since: 6.0
+##
+{ 'struct': 'BlockExportOptionsFuse',
+ 'data': { 'mountpoint': 'str' },
+ 'if': 'defined(CONFIG_FUSE)' }
+
##
# @NbdServerAddOptions:
#
@@ -222,11 +237,13 @@
#
# @nbd: NBD export
# @vhost-user-blk: vhost-user-blk export (since 5.2)
+# @fuse: FUSE export (since: 6.0)
#
# Since: 4.2
##
{ 'enum': 'BlockExportType',
- 'data': [ 'nbd', 'vhost-user-blk' ] }
+ 'data': [ 'nbd', 'vhost-user-blk',
+ { 'name': 'fuse', 'if': 'defined(CONFIG_FUSE)' } ] }
##
# @BlockExportOptions:
@@ -267,7 +284,9 @@
'discriminator': 'type',
'data': {
'nbd': 'BlockExportOptionsNbd',
- 'vhost-user-blk': 'BlockExportOptionsVhostUserBlk'
+ 'vhost-user-blk': 'BlockExportOptionsVhostUserBlk',
+ 'fuse': { 'type': 'BlockExportOptionsFuse',
+ 'if': 'defined(CONFIG_FUSE)' }
} }
##
diff --git a/include/block/fuse.h b/include/block/fuse.h
new file mode 100644
index 0000000000..ffa91fe364
--- /dev/null
+++ b/include/block/fuse.h
@@ -0,0 +1,30 @@
+/*
+ * Present a block device as a raw image through FUSE
+ *
+ * Copyright (c) 2020 Max Reitz <mreitz@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; under version 2 or later of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef BLOCK_FUSE_H
+#define BLOCK_FUSE_H
+
+#ifdef CONFIG_FUSE
+
+#include "block/export.h"
+
+extern const BlockExportDriver blk_exp_fuse;
+
+#endif /* CONFIG_FUSE */
+
+#endif
diff --git a/block.c b/block.c
index f1cedac362..eb16fb48c6 100644
--- a/block.c
+++ b/block.c
@@ -26,6 +26,7 @@
#include "block/trace.h"
#include "block/block_int.h"
#include "block/blockjob.h"
+#include "block/fuse.h"
#include "block/nbd.h"
#include "block/qdict.h"
#include "qemu/error-report.h"
diff --git a/block/export/export.c b/block/export/export.c
index bad6f21b1c..b716c1522c 100644
--- a/block/export/export.c
+++ b/block/export/export.c
@@ -17,6 +17,7 @@
#include "sysemu/block-backend.h"
#include "sysemu/iothread.h"
#include "block/export.h"
+#include "block/fuse.h"
#include "block/nbd.h"
#include "qapi/error.h"
#include "qapi/qapi-commands-block-export.h"
@@ -31,6 +32,9 @@ static const BlockExportDriver *blk_exp_drivers[] = {
#ifdef CONFIG_VHOST_USER_BLK_SERVER
&blk_exp_vhost_user_blk,
#endif
+#ifdef CONFIG_FUSE
+ &blk_exp_fuse,
+#endif
};
/* Only accessed from the main thread */
diff --git a/block/export/fuse.c b/block/export/fuse.c
new file mode 100644
index 0000000000..0553bcd630
--- /dev/null
+++ b/block/export/fuse.c
@@ -0,0 +1,295 @@
+/*
+ * Present a block device as a raw image through FUSE
+ *
+ * Copyright (c) 2020 Max Reitz <mreitz@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; under version 2 or later of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#define FUSE_USE_VERSION 31
+
+#include "qemu/osdep.h"
+#include "block/aio.h"
+#include "block/block.h"
+#include "block/export.h"
+#include "block/fuse.h"
+#include "block/qapi.h"
+#include "qapi/error.h"
+#include "qapi/qapi-commands-block.h"
+#include "sysemu/block-backend.h"
+
+#include <fuse.h>
+#include <fuse_lowlevel.h>
+
+
+/* Prevent overly long bounce buffer allocations */
+#define FUSE_MAX_BOUNCE_BYTES (MIN(BDRV_REQUEST_MAX_BYTES, 64 * 1024 * 1024))
+
+
+typedef struct FuseExport {
+ BlockExport common;
+
+ struct fuse_session *fuse_session;
+ struct fuse_buf fuse_buf;
+ bool mounted, fd_handler_set_up;
+
+ char *mountpoint;
+ bool writable;
+} FuseExport;
+
+static GHashTable *exports;
+static const struct fuse_lowlevel_ops fuse_ops;
+
+static void fuse_export_shutdown(BlockExport *exp);
+static void fuse_export_delete(BlockExport *exp);
+
+static void init_exports_table(void);
+
+static int setup_fuse_export(FuseExport *exp, const char *mountpoint,
+ Error **errp);
+static void read_from_fuse_export(void *opaque);
+
+static bool is_regular_file(const char *path, Error **errp);
+
+
+static int fuse_export_create(BlockExport *blk_exp,
+ BlockExportOptions *blk_exp_args,
+ Error **errp)
+{
+ FuseExport *exp = container_of(blk_exp, FuseExport, common);
+ BlockExportOptionsFuse *args = &blk_exp_args->u.fuse;
+ int ret;
+
+ assert(blk_exp_args->type == BLOCK_EXPORT_TYPE_FUSE);
+
+ init_exports_table();
+
+ /*
+ * It is important to do this check before calling is_regular_file() --
+ * that function will do a stat(), which we would have to handle if we
+ * already exported something on @mountpoint. But we cannot, because
+ * we are currently caught up here.
+ * (Note that ideally we would want to resolve relative paths here,
+ * but bdrv_make_absolute_filename() might do the wrong thing for
+ * paths that contain colons, and realpath() would resolve symlinks,
+ * which we do not want: The mount point is not going to be the
+ * symlink's destination, but the link itself.)
+ * So this will not catch all potential clashes, but hopefully at
+ * least the most common one of specifying exactly the same path
+ * string twice.
+ */
+ if (g_hash_table_contains(exports, args->mountpoint)) {
+ error_setg(errp, "There already is a FUSE export on '%s'",
+ args->mountpoint);
+ ret = -EEXIST;
+ goto fail;
+ }
+
+ if (!is_regular_file(args->mountpoint, errp)) {
+ ret = -EINVAL;
+ goto fail;
+ }
+
+ exp->mountpoint = g_strdup(args->mountpoint);
+ exp->writable = blk_exp_args->writable;
+
+ ret = setup_fuse_export(exp, args->mountpoint, errp);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ return 0;
+
+fail:
+ fuse_export_delete(blk_exp);
+ return ret;
+}
+
+/**
+ * Allocates the global @exports hash table.
+ */
+static void init_exports_table(void)
+{
+ if (exports) {
+ return;
+ }
+
+ exports = g_hash_table_new_full(g_str_hash, g_str_equal, g_free, NULL);
+}
+
+/**
+ * Create exp->fuse_session and mount it.
+ */
+static int setup_fuse_export(FuseExport *exp, const char *mountpoint,
+ Error **errp)
+{
+ const char *fuse_argv[4];
+ char *mount_opts;
+ struct fuse_args fuse_args;
+ int ret;
+
+ /* Needs to match what fuse_init() sets. Only max_read must be supplied. */
+ mount_opts = g_strdup_printf("max_read=%zu", FUSE_MAX_BOUNCE_BYTES);
+
+ fuse_argv[0] = ""; /* Dummy program name */
+ fuse_argv[1] = "-o";
+ fuse_argv[2] = mount_opts;
+ fuse_argv[3] = NULL;
+ fuse_args = (struct fuse_args)FUSE_ARGS_INIT(3, (char **)fuse_argv);
+
+ exp->fuse_session = fuse_session_new(&fuse_args, &fuse_ops,
+ sizeof(fuse_ops), exp);
+ g_free(mount_opts);
+ if (!exp->fuse_session) {
+ error_setg(errp, "Failed to set up FUSE session");
+ ret = -EIO;
+ goto fail;
+ }
+
+ ret = fuse_session_mount(exp->fuse_session, mountpoint);
+ if (ret < 0) {
+ error_setg(errp, "Failed to mount FUSE session to export");
+ ret = -EIO;
+ goto fail;
+ }
+ exp->mounted = true;
+
+ g_hash_table_insert(exports, g_strdup(mountpoint), NULL);
+
+ aio_set_fd_handler(exp->common.ctx,
+ fuse_session_fd(exp->fuse_session), true,
+ read_from_fuse_export, NULL, NULL, exp);
+ exp->fd_handler_set_up = true;
+
+ return 0;
+
+fail:
+ fuse_export_shutdown(&exp->common);
+ return ret;
+}
+
+/**
+ * Callback to be invoked when the FUSE session FD can be read from.
+ * (This is basically the FUSE event loop.)
+ */
+static void read_from_fuse_export(void *opaque)
+{
+ FuseExport *exp = opaque;
+ int ret;
+
+ blk_exp_ref(&exp->common);
+
+ do {
+ ret = fuse_session_receive_buf(exp->fuse_session, &exp->fuse_buf);
+ } while (ret == -EINTR);
+ if (ret < 0) {
+ goto out;
+ }
+
+ fuse_session_process_buf(exp->fuse_session, &exp->fuse_buf);
+
+out:
+ blk_exp_unref(&exp->common);
+}
+
+static void fuse_export_shutdown(BlockExport *blk_exp)
+{
+ FuseExport *exp = container_of(blk_exp, FuseExport, common);
+
+ if (exp->fuse_session) {
+ fuse_session_exit(exp->fuse_session);
+
+ if (exp->fd_handler_set_up) {
+ aio_set_fd_handler(exp->common.ctx,
+ fuse_session_fd(exp->fuse_session), true,
+ NULL, NULL, NULL, NULL);
+ exp->fd_handler_set_up = false;
+ }
+ }
+
+ if (exp->mountpoint) {
+ /*
+ * Safe to drop now, because we will not handle any requests
+ * for this export anymore anyway.
+ */
+ g_hash_table_remove(exports, exp->mountpoint);
+ }
+}
+
+static void fuse_export_delete(BlockExport *blk_exp)
+{
+ FuseExport *exp = container_of(blk_exp, FuseExport, common);
+
+ if (exp->fuse_session) {
+ if (exp->mounted) {
+ fuse_session_unmount(exp->fuse_session);
+ }
+
+ fuse_session_destroy(exp->fuse_session);
+ }
+
+ free(exp->fuse_buf.mem);
+ g_free(exp->mountpoint);
+}
+
+/**
+ * Check whether @path points to a regular file. If not, put an
+ * appropriate message into *errp.
+ */
+static bool is_regular_file(const char *path, Error **errp)
+{
+ struct stat statbuf;
+ int ret;
+
+ ret = stat(path, &statbuf);
+ if (ret < 0) {
+ error_setg_errno(errp, errno, "Failed to stat '%s'", path);
+ return false;
+ }
+
+ if (!S_ISREG(statbuf.st_mode)) {
+ error_setg(errp, "'%s' is not a regular file", path);
+ return false;
+ }
+
+ return true;
+}
+
+/**
+ * A chance to set change some parameters supplied to FUSE_INIT.
+ */
+static void fuse_init(void *userdata, struct fuse_conn_info *conn)
+{
+ /*
+ * MIN_NON_ZERO() would not be wrong here, but what we set here
+ * must equal what has been passed to fuse_session_new().
+ * Therefore, as long as max_read must be passed as a mount option
+ * (which libfuse claims will be changed at some point), we have
+ * to set max_read to a fixed value here.
+ */
+ conn->max_read = FUSE_MAX_BOUNCE_BYTES;
+
+ conn->max_write = MIN_NON_ZERO(BDRV_REQUEST_MAX_BYTES, conn->max_write);
+}
+
+static const struct fuse_lowlevel_ops fuse_ops = {
+ .init = fuse_init,
+};
+
+const BlockExportDriver blk_exp_fuse = {
+ .type = BLOCK_EXPORT_TYPE_FUSE,
+ .instance_size = sizeof(FuseExport),
+ .create = fuse_export_create,
+ .delete = fuse_export_delete,
+ .request_shutdown = fuse_export_shutdown,
+};
diff --git a/MAINTAINERS b/MAINTAINERS
index aa39490a24..1950592d04 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3114,6 +3114,12 @@ F: include/qemu/vhost-user-server.h
F: tests/qtest/libqos/vhost-user-blk.c
F: util/vhost-user-server.c
+FUSE block device exports
+M: Max Reitz <mreitz@redhat.com>
+L: qemu-block@nongnu.org
+S: Supported
+F: block/export/fuse.c
+
Replication
M: Wen Congyang <wencongyang2@huawei.com>
M: Xie Changlong <xiechanglong.d@gmail.com>
diff --git a/block/export/meson.build b/block/export/meson.build
index 135b356775..0a08e384c7 100644
--- a/block/export/meson.build
+++ b/block/export/meson.build
@@ -3,3 +3,5 @@ blockdev_ss.add(files('export.c'))
if have_vhost_user_blk_server
blockdev_ss.add(files('vhost-user-blk-server.c'))
endif
+
+blockdev_ss.add(when: fuse, if_true: files('fuse.c'))
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 07/34] fuse: Implement standard FUSE operations
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (5 preceding siblings ...)
2020-12-11 17:07 ` [PULL 06/34] fuse: Allow exporting BDSs via FUSE Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 08/34] fuse: Allow growable exports Kevin Wolf
` (27 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
This makes the export actually useful instead of only producing errors
whenever it is accessed.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-4-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/export/fuse.c | 242 ++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 242 insertions(+)
diff --git a/block/export/fuse.c b/block/export/fuse.c
index 0553bcd630..d995829ab7 100644
--- a/block/export/fuse.c
+++ b/block/export/fuse.c
@@ -282,8 +282,250 @@ static void fuse_init(void *userdata, struct fuse_conn_info *conn)
conn->max_write = MIN_NON_ZERO(BDRV_REQUEST_MAX_BYTES, conn->max_write);
}
+/**
+ * Let clients look up files. Always return ENOENT because we only
+ * care about the mountpoint itself.
+ */
+static void fuse_lookup(fuse_req_t req, fuse_ino_t parent, const char *name)
+{
+ fuse_reply_err(req, ENOENT);
+}
+
+/**
+ * Let clients get file attributes (i.e., stat() the file).
+ */
+static void fuse_getattr(fuse_req_t req, fuse_ino_t inode,
+ struct fuse_file_info *fi)
+{
+ struct stat statbuf;
+ int64_t length, allocated_blocks;
+ time_t now = time(NULL);
+ FuseExport *exp = fuse_req_userdata(req);
+ mode_t mode;
+
+ length = blk_getlength(exp->common.blk);
+ if (length < 0) {
+ fuse_reply_err(req, -length);
+ return;
+ }
+
+ allocated_blocks = bdrv_get_allocated_file_size(blk_bs(exp->common.blk));
+ if (allocated_blocks <= 0) {
+ allocated_blocks = DIV_ROUND_UP(length, 512);
+ } else {
+ allocated_blocks = DIV_ROUND_UP(allocated_blocks, 512);
+ }
+
+ mode = S_IFREG | S_IRUSR;
+ if (exp->writable) {
+ mode |= S_IWUSR;
+ }
+
+ statbuf = (struct stat) {
+ .st_ino = inode,
+ .st_mode = mode,
+ .st_nlink = 1,
+ .st_uid = getuid(),
+ .st_gid = getgid(),
+ .st_size = length,
+ .st_blksize = blk_bs(exp->common.blk)->bl.request_alignment,
+ .st_blocks = allocated_blocks,
+ .st_atime = now,
+ .st_mtime = now,
+ .st_ctime = now,
+ };
+
+ fuse_reply_attr(req, &statbuf, 1.);
+}
+
+static int fuse_do_truncate(const FuseExport *exp, int64_t size,
+ bool req_zero_write, PreallocMode prealloc)
+{
+ uint64_t blk_perm, blk_shared_perm;
+ BdrvRequestFlags truncate_flags = 0;
+ int ret;
+
+ if (req_zero_write) {
+ truncate_flags |= BDRV_REQ_ZERO_WRITE;
+ }
+
+ blk_get_perm(exp->common.blk, &blk_perm, &blk_shared_perm);
+
+ ret = blk_set_perm(exp->common.blk, blk_perm | BLK_PERM_RESIZE,
+ blk_shared_perm, NULL);
+ if (ret < 0) {
+ return ret;
+ }
+
+ ret = blk_truncate(exp->common.blk, size, true, prealloc,
+ truncate_flags, NULL);
+
+ /* Must succeed, because we are only giving up the RESIZE permission */
+ blk_set_perm(exp->common.blk, blk_perm, blk_shared_perm, &error_abort);
+
+ return ret;
+}
+
+/**
+ * Let clients set file attributes. Only resizing is supported.
+ */
+static void fuse_setattr(fuse_req_t req, fuse_ino_t inode, struct stat *statbuf,
+ int to_set, struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+ int ret;
+
+ if (!exp->writable) {
+ fuse_reply_err(req, EACCES);
+ return;
+ }
+
+ if (to_set & ~FUSE_SET_ATTR_SIZE) {
+ fuse_reply_err(req, ENOTSUP);
+ return;
+ }
+
+ ret = fuse_do_truncate(exp, statbuf->st_size, true, PREALLOC_MODE_OFF);
+ if (ret < 0) {
+ fuse_reply_err(req, -ret);
+ return;
+ }
+
+ fuse_getattr(req, inode, fi);
+}
+
+/**
+ * Let clients open a file (i.e., the exported image).
+ */
+static void fuse_open(fuse_req_t req, fuse_ino_t inode,
+ struct fuse_file_info *fi)
+{
+ fuse_reply_open(req, fi);
+}
+
+/**
+ * Handle client reads from the exported image.
+ */
+static void fuse_read(fuse_req_t req, fuse_ino_t inode,
+ size_t size, off_t offset, struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+ int64_t length;
+ void *buf;
+ int ret;
+
+ /* Limited by max_read, should not happen */
+ if (size > FUSE_MAX_BOUNCE_BYTES) {
+ fuse_reply_err(req, EINVAL);
+ return;
+ }
+
+ /**
+ * Clients will expect short reads at EOF, so we have to limit
+ * offset+size to the image length.
+ */
+ length = blk_getlength(exp->common.blk);
+ if (length < 0) {
+ fuse_reply_err(req, -length);
+ return;
+ }
+
+ if (offset + size > length) {
+ size = length - offset;
+ }
+
+ buf = qemu_try_blockalign(blk_bs(exp->common.blk), size);
+ if (!buf) {
+ fuse_reply_err(req, ENOMEM);
+ return;
+ }
+
+ ret = blk_pread(exp->common.blk, offset, buf, size);
+ if (ret >= 0) {
+ fuse_reply_buf(req, buf, size);
+ } else {
+ fuse_reply_err(req, -ret);
+ }
+
+ qemu_vfree(buf);
+}
+
+/**
+ * Handle client writes to the exported image.
+ */
+static void fuse_write(fuse_req_t req, fuse_ino_t inode, const char *buf,
+ size_t size, off_t offset, struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+ int64_t length;
+ int ret;
+
+ /* Limited by max_write, should not happen */
+ if (size > BDRV_REQUEST_MAX_BYTES) {
+ fuse_reply_err(req, EINVAL);
+ return;
+ }
+
+ if (!exp->writable) {
+ fuse_reply_err(req, EACCES);
+ return;
+ }
+
+ /**
+ * Clients will expect short writes at EOF, so we have to limit
+ * offset+size to the image length.
+ */
+ length = blk_getlength(exp->common.blk);
+ if (length < 0) {
+ fuse_reply_err(req, -length);
+ return;
+ }
+
+ if (offset + size > length) {
+ size = length - offset;
+ }
+
+ ret = blk_pwrite(exp->common.blk, offset, buf, size, 0);
+ if (ret >= 0) {
+ fuse_reply_write(req, size);
+ } else {
+ fuse_reply_err(req, -ret);
+ }
+}
+
+/**
+ * Let clients fsync the exported image.
+ */
+static void fuse_fsync(fuse_req_t req, fuse_ino_t inode, int datasync,
+ struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+ int ret;
+
+ ret = blk_flush(exp->common.blk);
+ fuse_reply_err(req, ret < 0 ? -ret : 0);
+}
+
+/**
+ * Called before an FD to the exported image is closed. (libfuse
+ * notes this to be a way to return last-minute errors.)
+ */
+static void fuse_flush(fuse_req_t req, fuse_ino_t inode,
+ struct fuse_file_info *fi)
+{
+ fuse_fsync(req, inode, 1, fi);
+}
+
static const struct fuse_lowlevel_ops fuse_ops = {
.init = fuse_init,
+ .lookup = fuse_lookup,
+ .getattr = fuse_getattr,
+ .setattr = fuse_setattr,
+ .open = fuse_open,
+ .read = fuse_read,
+ .write = fuse_write,
+ .flush = fuse_flush,
+ .fsync = fuse_fsync,
};
const BlockExportDriver blk_exp_fuse = {
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 08/34] fuse: Allow growable exports
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (6 preceding siblings ...)
2020-12-11 17:07 ` [PULL 07/34] fuse: Implement standard FUSE operations Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 09/34] fuse: (Partially) implement fallocate() Kevin Wolf
` (26 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
These will behave more like normal files in that writes beyond the EOF
will automatically grow the export size.
As an optimization, keep the RESIZE permission for growable exports so
we do not have to take it for every post-EOF write. (This permission is
not released when the export is destroyed, because at that point the
BlockBackend is destroyed altogether anyway.)
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-5-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
qapi/block-export.json | 6 +++++-
block/export/fuse.c | 44 ++++++++++++++++++++++++++++++++++--------
2 files changed, 41 insertions(+), 9 deletions(-)
diff --git a/qapi/block-export.json b/qapi/block-export.json
index 430bc69f35..e819e70cac 100644
--- a/qapi/block-export.json
+++ b/qapi/block-export.json
@@ -129,10 +129,14 @@
# @mountpoint: Path on which to export the block device via FUSE.
# This must point to an existing regular file.
#
+# @growable: Whether writes beyond the EOF should grow the block node
+# accordingly. (default: false)
+#
# Since: 6.0
##
{ 'struct': 'BlockExportOptionsFuse',
- 'data': { 'mountpoint': 'str' },
+ 'data': { 'mountpoint': 'str',
+ '*growable': 'bool' },
'if': 'defined(CONFIG_FUSE)' }
##
diff --git a/block/export/fuse.c b/block/export/fuse.c
index d995829ab7..92d2f50bcc 100644
--- a/block/export/fuse.c
+++ b/block/export/fuse.c
@@ -45,6 +45,7 @@ typedef struct FuseExport {
char *mountpoint;
bool writable;
+ bool growable;
} FuseExport;
static GHashTable *exports;
@@ -72,6 +73,19 @@ static int fuse_export_create(BlockExport *blk_exp,
assert(blk_exp_args->type == BLOCK_EXPORT_TYPE_FUSE);
+ /* For growable exports, take the RESIZE permission */
+ if (args->growable) {
+ uint64_t blk_perm, blk_shared_perm;
+
+ blk_get_perm(exp->common.blk, &blk_perm, &blk_shared_perm);
+
+ ret = blk_set_perm(exp->common.blk, blk_perm | BLK_PERM_RESIZE,
+ blk_shared_perm, errp);
+ if (ret < 0) {
+ return ret;
+ }
+ }
+
init_exports_table();
/*
@@ -102,6 +116,7 @@ static int fuse_export_create(BlockExport *blk_exp,
exp->mountpoint = g_strdup(args->mountpoint);
exp->writable = blk_exp_args->writable;
+ exp->growable = args->growable;
ret = setup_fuse_export(exp, args->mountpoint, errp);
if (ret < 0) {
@@ -349,19 +364,24 @@ static int fuse_do_truncate(const FuseExport *exp, int64_t size,
truncate_flags |= BDRV_REQ_ZERO_WRITE;
}
- blk_get_perm(exp->common.blk, &blk_perm, &blk_shared_perm);
+ /* Growable exports have a permanent RESIZE permission */
+ if (!exp->growable) {
+ blk_get_perm(exp->common.blk, &blk_perm, &blk_shared_perm);
- ret = blk_set_perm(exp->common.blk, blk_perm | BLK_PERM_RESIZE,
- blk_shared_perm, NULL);
- if (ret < 0) {
- return ret;
+ ret = blk_set_perm(exp->common.blk, blk_perm | BLK_PERM_RESIZE,
+ blk_shared_perm, NULL);
+ if (ret < 0) {
+ return ret;
+ }
}
ret = blk_truncate(exp->common.blk, size, true, prealloc,
truncate_flags, NULL);
- /* Must succeed, because we are only giving up the RESIZE permission */
- blk_set_perm(exp->common.blk, blk_perm, blk_shared_perm, &error_abort);
+ if (!exp->growable) {
+ /* Must succeed, because we are only giving up the RESIZE permission */
+ blk_set_perm(exp->common.blk, blk_perm, blk_shared_perm, &error_abort);
+ }
return ret;
}
@@ -482,7 +502,15 @@ static void fuse_write(fuse_req_t req, fuse_ino_t inode, const char *buf,
}
if (offset + size > length) {
- size = length - offset;
+ if (exp->growable) {
+ ret = fuse_do_truncate(exp, offset + size, true, PREALLOC_MODE_OFF);
+ if (ret < 0) {
+ fuse_reply_err(req, -ret);
+ return;
+ }
+ } else {
+ size = length - offset;
+ }
}
ret = blk_pwrite(exp->common.blk, offset, buf, size, 0);
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 09/34] fuse: (Partially) implement fallocate()
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (7 preceding siblings ...)
2020-12-11 17:07 ` [PULL 08/34] fuse: Allow growable exports Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 10/34] fuse: Implement hole detection through lseek Kevin Wolf
` (25 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
This allows allocating areas after the (old) EOF as part of a growing
resize, writing zeroes, and discarding.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-6-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/export/fuse.c | 84 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 84 insertions(+)
diff --git a/block/export/fuse.c b/block/export/fuse.c
index 92d2f50bcc..0b9d226b2f 100644
--- a/block/export/fuse.c
+++ b/block/export/fuse.c
@@ -521,6 +521,89 @@ static void fuse_write(fuse_req_t req, fuse_ino_t inode, const char *buf,
}
}
+/**
+ * Let clients perform various fallocate() operations.
+ */
+static void fuse_fallocate(fuse_req_t req, fuse_ino_t inode, int mode,
+ off_t offset, off_t length,
+ struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+ int64_t blk_len;
+ int ret;
+
+ if (!exp->writable) {
+ fuse_reply_err(req, EACCES);
+ return;
+ }
+
+ blk_len = blk_getlength(exp->common.blk);
+ if (blk_len < 0) {
+ fuse_reply_err(req, -blk_len);
+ return;
+ }
+
+ if (mode & FALLOC_FL_KEEP_SIZE) {
+ length = MIN(length, blk_len - offset);
+ }
+
+ if (mode & FALLOC_FL_PUNCH_HOLE) {
+ if (!(mode & FALLOC_FL_KEEP_SIZE)) {
+ fuse_reply_err(req, EINVAL);
+ return;
+ }
+
+ do {
+ int size = MIN(length, BDRV_REQUEST_MAX_BYTES);
+
+ ret = blk_pdiscard(exp->common.blk, offset, size);
+ offset += size;
+ length -= size;
+ } while (ret == 0 && length > 0);
+ } else if (mode & FALLOC_FL_ZERO_RANGE) {
+ if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + length > blk_len) {
+ /* No need for zeroes, we are going to write them ourselves */
+ ret = fuse_do_truncate(exp, offset + length, false,
+ PREALLOC_MODE_OFF);
+ if (ret < 0) {
+ fuse_reply_err(req, -ret);
+ return;
+ }
+ }
+
+ do {
+ int size = MIN(length, BDRV_REQUEST_MAX_BYTES);
+
+ ret = blk_pwrite_zeroes(exp->common.blk,
+ offset, size, 0);
+ offset += size;
+ length -= size;
+ } while (ret == 0 && length > 0);
+ } else if (!mode) {
+ /* We can only fallocate at the EOF with a truncate */
+ if (offset < blk_len) {
+ fuse_reply_err(req, EOPNOTSUPP);
+ return;
+ }
+
+ if (offset > blk_len) {
+ /* No preallocation needed here */
+ ret = fuse_do_truncate(exp, offset, true, PREALLOC_MODE_OFF);
+ if (ret < 0) {
+ fuse_reply_err(req, -ret);
+ return;
+ }
+ }
+
+ ret = fuse_do_truncate(exp, offset + length, true,
+ PREALLOC_MODE_FALLOC);
+ } else {
+ ret = -EOPNOTSUPP;
+ }
+
+ fuse_reply_err(req, ret < 0 ? -ret : 0);
+}
+
/**
* Let clients fsync the exported image.
*/
@@ -552,6 +635,7 @@ static const struct fuse_lowlevel_ops fuse_ops = {
.open = fuse_open,
.read = fuse_read,
.write = fuse_write,
+ .fallocate = fuse_fallocate,
.flush = fuse_flush,
.fsync = fuse_fsync,
};
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 10/34] fuse: Implement hole detection through lseek
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (8 preceding siblings ...)
2020-12-11 17:07 ` [PULL 09/34] fuse: (Partially) implement fallocate() Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 11/34] iotests: Do not needlessly filter _make_test_img Kevin Wolf
` (24 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
This is a relatively new feature in libfuse (available since 3.8.0,
which was released in November 2019), so we have to add a dedicated
check whether it is available before making use of it.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-7-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
meson_options.txt | 2 ++
configure | 8 ++++-
block/export/fuse.c | 77 +++++++++++++++++++++++++++++++++++++++++++++
meson.build | 20 ++++++++++++
4 files changed, 106 insertions(+), 1 deletion(-)
diff --git a/meson_options.txt b/meson_options.txt
index 8f9f2e3df6..74ac853548 100644
--- a/meson_options.txt
+++ b/meson_options.txt
@@ -68,6 +68,8 @@ option('vhost_user_blk_server', type: 'feature', value: 'auto',
description: 'build vhost-user-blk server')
option('fuse', type: 'feature', value: 'auto',
description: 'FUSE block device export')
+option('fuse_lseek', type : 'feature', value : 'auto',
+ description: 'SEEK_HOLE/SEEK_DATA support for FUSE exports')
option('capstone', type: 'combo', value: 'auto',
choices: ['disabled', 'enabled', 'auto', 'system', 'internal'],
diff --git a/configure b/configure
index b2f96c0da2..3f823ed163 100755
--- a/configure
+++ b/configure
@@ -450,6 +450,7 @@ ninja=""
skip_meson=no
gettext=""
fuse="auto"
+fuse_lseek="auto"
bogus_os="no"
malloc_trim="auto"
@@ -1530,6 +1531,10 @@ for opt do
;;
--disable-fuse) fuse="disabled"
;;
+ --enable-fuse-lseek) fuse_lseek="enabled"
+ ;;
+ --disable-fuse-lseek) fuse_lseek="disabled"
+ ;;
*)
echo "ERROR: unknown option $opt"
echo "Try '$0 --help' for more information"
@@ -1856,6 +1861,7 @@ disabled with --disable-FEATURE, default is enabled if available:
rng-none dummy RNG, avoid using /dev/(u)random and getrandom()
libdaxctl libdaxctl support
fuse FUSE block device export
+ fuse-lseek SEEK_HOLE/SEEK_DATA support for FUSE exports
NOTE: The object files are built at the place where configure is launched
EOF
@@ -7020,7 +7026,7 @@ NINJA=$ninja $meson setup \
-Diconv=$iconv -Dcurses=$curses -Dlibudev=$libudev\
-Ddocs=$docs -Dsphinx_build=$sphinx_build -Dinstall_blobs=$blobs \
-Dvhost_user_blk_server=$vhost_user_blk_server \
- -Dfuse=$fuse \
+ -Dfuse=$fuse -Dfuse_lseek=$fuse_lseek \
$cross_arg \
"$PWD" "$source_path"
diff --git a/block/export/fuse.c b/block/export/fuse.c
index 0b9d226b2f..38f74c94da 100644
--- a/block/export/fuse.c
+++ b/block/export/fuse.c
@@ -627,6 +627,80 @@ static void fuse_flush(fuse_req_t req, fuse_ino_t inode,
fuse_fsync(req, inode, 1, fi);
}
+#ifdef CONFIG_FUSE_LSEEK
+/**
+ * Let clients inquire allocation status.
+ */
+static void fuse_lseek(fuse_req_t req, fuse_ino_t inode, off_t offset,
+ int whence, struct fuse_file_info *fi)
+{
+ FuseExport *exp = fuse_req_userdata(req);
+
+ if (whence != SEEK_HOLE && whence != SEEK_DATA) {
+ fuse_reply_err(req, EINVAL);
+ return;
+ }
+
+ while (true) {
+ int64_t pnum;
+ int ret;
+
+ ret = bdrv_block_status_above(blk_bs(exp->common.blk), NULL,
+ offset, INT64_MAX, &pnum, NULL, NULL);
+ if (ret < 0) {
+ fuse_reply_err(req, -ret);
+ return;
+ }
+
+ if (!pnum && (ret & BDRV_BLOCK_EOF)) {
+ int64_t blk_len;
+
+ /*
+ * If blk_getlength() rounds (e.g. by sectors), then the
+ * export length will be rounded, too. However,
+ * bdrv_block_status_above() may return EOF at unaligned
+ * offsets. We must not let this become visible and thus
+ * always simulate a hole between @offset (the real EOF)
+ * and @blk_len (the client-visible EOF).
+ */
+
+ blk_len = blk_getlength(exp->common.blk);
+ if (blk_len < 0) {
+ fuse_reply_err(req, -blk_len);
+ return;
+ }
+
+ if (offset > blk_len || whence == SEEK_DATA) {
+ fuse_reply_err(req, ENXIO);
+ } else {
+ fuse_reply_lseek(req, offset);
+ }
+ return;
+ }
+
+ if (ret & BDRV_BLOCK_DATA) {
+ if (whence == SEEK_DATA) {
+ fuse_reply_lseek(req, offset);
+ return;
+ }
+ } else {
+ if (whence == SEEK_HOLE) {
+ fuse_reply_lseek(req, offset);
+ return;
+ }
+ }
+
+ /* Safety check against infinite loops */
+ if (!pnum) {
+ fuse_reply_err(req, ENXIO);
+ return;
+ }
+
+ offset += pnum;
+ }
+}
+#endif
+
static const struct fuse_lowlevel_ops fuse_ops = {
.init = fuse_init,
.lookup = fuse_lookup,
@@ -638,6 +712,9 @@ static const struct fuse_lowlevel_ops fuse_ops = {
.fallocate = fuse_fallocate,
.flush = fuse_flush,
.fsync = fuse_fsync,
+#ifdef CONFIG_FUSE_LSEEK
+ .lseek = fuse_lseek,
+#endif
};
const BlockExportDriver blk_exp_fuse = {
diff --git a/meson.build b/meson.build
index 6e8ef151d8..f344b25955 100644
--- a/meson.build
+++ b/meson.build
@@ -773,10 +773,28 @@ elif get_option('vhost_user_blk_server').disabled() or not have_system
have_vhost_user_blk_server = false
endif
+if get_option('fuse').disabled() and get_option('fuse_lseek').enabled()
+ error('Cannot enable fuse-lseek while fuse is disabled')
+endif
+
fuse = dependency('fuse3', required: get_option('fuse'),
version: '>=3.1', method: 'pkg-config',
static: enable_static)
+fuse_lseek = not_found
+if not get_option('fuse_lseek').disabled()
+ if fuse.version().version_compare('>=3.8')
+ # Dummy dependency
+ fuse_lseek = declare_dependency()
+ elif get_option('fuse_lseek').enabled()
+ if fuse.found()
+ error('fuse-lseek requires libfuse >=3.8, found ' + fuse.version())
+ else
+ error('fuse-lseek requires libfuse, which was not found')
+ endif
+ endif
+endif
+
#################
# config-host.h #
#################
@@ -812,6 +830,7 @@ config_host_data.set('CONFIG_GETTID', has_gettid)
config_host_data.set('CONFIG_MALLOC_TRIM', has_malloc_trim)
config_host_data.set('CONFIG_STATX', has_statx)
config_host_data.set('CONFIG_FUSE', fuse.found())
+config_host_data.set('CONFIG_FUSE_LSEEK', fuse_lseek.found())
config_host_data.set('QEMU_VERSION', '"@0@"'.format(meson.project_version()))
config_host_data.set('QEMU_VERSION_MAJOR', meson.project_version().split('.')[0])
config_host_data.set('QEMU_VERSION_MINOR', meson.project_version().split('.')[1])
@@ -2214,6 +2233,7 @@ summary_info += {'thread sanitizer': config_host.has_key('CONFIG_TSAN')}
summary_info += {'rng-none': config_host.has_key('CONFIG_RNG_NONE')}
summary_info += {'Linux keyring': config_host.has_key('CONFIG_SECRET_KEYRING')}
summary_info += {'FUSE exports': fuse.found()}
+summary_info += {'FUSE lseek': fuse_lseek.found()}
summary(summary_info, bool_yn: true)
if not supported_cpus.contains(cpu)
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 11/34] iotests: Do not needlessly filter _make_test_img
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (9 preceding siblings ...)
2020-12-11 17:07 ` [PULL 10/34] fuse: Implement hole detection through lseek Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 12/34] iotests: Do not pipe _make_test_img Kevin Wolf
` (23 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
In most cases, _make_test_img does not need a _filter_imgfmt on top. It
does that by itself.
(The exception is when IMGFMT has been overwritten but TEST_IMG has not.
In such cases, we do need a _filter_imgfmt on top to filter the test's
original IMGFMT from TEST_IMG.)
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-8-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/161 | 12 ++++++------
tests/qemu-iotests/175 | 6 +++---
tests/qemu-iotests/249 | 6 +++---
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/tests/qemu-iotests/161 b/tests/qemu-iotests/161
index e270976d87..bbf7dbbc5c 100755
--- a/tests/qemu-iotests/161
+++ b/tests/qemu-iotests/161
@@ -48,9 +48,9 @@ _supported_os Linux
IMG_SIZE=1M
# Create the images
-TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE | _filter_imgfmt
-TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT | _filter_imgfmt
-_make_test_img -b "$TEST_IMG.int" -F $IMGFMT -F $IMGFMT | _filter_imgfmt
+TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
+TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT
+_make_test_img -b "$TEST_IMG.int" -F $IMGFMT -F $IMGFMT
# First test: reopen $TEST.IMG changing the detect-zeroes option on
# its backing file ($TEST_IMG.int).
@@ -105,9 +105,9 @@ echo
echo "*** Commit and then change an option on the backing file"
echo
# Create the images again
-TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE | _filter_imgfmt
-TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT| _filter_imgfmt
-_make_test_img -b "$TEST_IMG.int" -F $IMGFMT | _filter_imgfmt
+TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
+TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT
+_make_test_img -b "$TEST_IMG.int" -F $IMGFMT
_launch_qemu -drive if=none,file="${TEST_IMG}"
_send_qemu_cmd $QEMU_HANDLE \
diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175
index 00a626aa63..c3c2aed653 100755
--- a/tests/qemu-iotests/175
+++ b/tests/qemu-iotests/175
@@ -89,20 +89,20 @@ min_blocks=$(stat -c '%b' "$TEST_DIR/empty")
echo
echo "== creating image with default preallocation =="
-_make_test_img -o extent_size_hint=0 $size | _filter_imgfmt
+_make_test_img -o extent_size_hint=0 $size
stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size
for mode in off full falloc; do
echo
echo "== creating image with preallocation $mode =="
- _make_test_img -o preallocation=$mode,extent_size_hint=0 $size | _filter_imgfmt
+ _make_test_img -o preallocation=$mode,extent_size_hint=0 $size
stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size
done
for new_size in 4096 1048576; do
echo
echo "== resize empty image with block_resize =="
- _make_test_img -o extent_size_hint=0 0 | _filter_imgfmt
+ _make_test_img -o extent_size_hint=0 0
_block_resize $TEST_IMG $new_size >/dev/null
stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $new_size
done
diff --git a/tests/qemu-iotests/249 b/tests/qemu-iotests/249
index 68f13ed328..a9aa9303eb 100755
--- a/tests/qemu-iotests/249
+++ b/tests/qemu-iotests/249
@@ -48,9 +48,9 @@ _supported_os Linux
IMG_SIZE=1M
# Create the images: base <- int <- active
-TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE | _filter_imgfmt
-TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT | _filter_imgfmt
-_make_test_img -b "$TEST_IMG.int" -F $IMGFMT | _filter_imgfmt
+TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
+TEST_IMG="$TEST_IMG.int" _make_test_img -b "$TEST_IMG.base" -F $IMGFMT
+_make_test_img -b "$TEST_IMG.int" -F $IMGFMT
# Launch QEMU with these two drives:
# none0: base (read-only)
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 12/34] iotests: Do not pipe _make_test_img
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (10 preceding siblings ...)
2020-12-11 17:07 ` [PULL 11/34] iotests: Do not needlessly filter _make_test_img Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 13/34] iotests: Use convert -n in some cases Kevin Wolf
` (22 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Executing _make_test_img as part of a pipe will undo all variable
changes it has done. As such, this could not work with FUSE (because
we want to remember all of our exports and their qemu instances).
Replace the pipe by a temporary file in 071 and 174 (the two tests that
can run on FUSE).
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-9-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/071 | 19 +++++++++++++++----
tests/qemu-iotests/174 | 10 +++++++++-
2 files changed, 24 insertions(+), 5 deletions(-)
diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
index 88faebcc1d..18fe9054b0 100755
--- a/tests/qemu-iotests/071
+++ b/tests/qemu-iotests/071
@@ -61,8 +61,17 @@ echo
echo "=== Testing blkverify through filename ==="
echo
-TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
- _filter_imgfmt
+# _make_test_img may set variables that we need to retain. Everything
+# in a pipe is executed in a subshell, so doing so would throw away
+# all changes. Therefore, we have to store the output in some temp
+# file and filter that.
+scratch_out="$TEST_DIR/img-create.out"
+
+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE \
+ >"$scratch_out"
+_filter_imgfmt <"$scratch_out"
+rm -f "$scratch_out"
+
_make_test_img $IMG_SIZE
$QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base $TEST_IMG" \
-c 'read 0 512' -c 'write -P 42 0x38000 512' -c 'read -P 42 0x38000 512' | _filter_qemu_io
@@ -76,8 +85,10 @@ echo
echo "=== Testing blkverify through file blockref ==="
echo
-TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
- _filter_imgfmt
+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE \
+ >"$scratch_out"
+_filter_imgfmt <"$scratch_out"
+
_make_test_img $IMG_SIZE
$QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base,file.test.driver=$IMGFMT,file.test.file.filename=$TEST_IMG" \
-c 'read 0 512' -c 'write -P 42 0x38000 512' -c 'read -P 42 0x38000 512' | _filter_qemu_io
diff --git a/tests/qemu-iotests/174 b/tests/qemu-iotests/174
index e2f14a38c6..1b0dd2e8b7 100755
--- a/tests/qemu-iotests/174
+++ b/tests/qemu-iotests/174
@@ -40,7 +40,15 @@ _unsupported_fmt raw
size=256K
-IMGFMT=raw IMGKEYSECRET= _make_test_img --no-opts $size | _filter_imgfmt
+
+# _make_test_img may set variables that we need to retain. Everything
+# in a pipe is executed in a subshell, so doing so would throw away
+# all changes. Therefore, we have to store the output in some temp
+# file and filter that.
+scratch_out="$TEST_DIR/img-create.out"
+IMGFMT=raw IMGKEYSECRET= _make_test_img --no-opts $size >"$scratch_out"
+_filter_imgfmt <"$scratch_out"
+rm -f "$scratch_out"
echo
echo "== reading wrong format should fail =="
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 13/34] iotests: Use convert -n in some cases
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (11 preceding siblings ...)
2020-12-11 17:07 ` [PULL 12/34] iotests: Do not pipe _make_test_img Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 14/34] iotests/046: Avoid renaming images Kevin Wolf
` (21 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
qemu-img convert (without -n) can often be replaced by a combination of
_make_test_img + qemu-img convert -n. Doing so allows converting to
protocols that do not allow direct file creation, such as FUSE exports.
The only problem is that for formats other than qcow2 and qed (qcow1 at
least), this may lead to high disk usage for some reason, so we cannot
do it everywhere.
But we can do it in 028 and 089, so let us do that so they can run on
FUSE exports. Also, in 028 this allows us to remove a 9-line comment
that used to explain why we cannot safely filter drive-backup's image
creation output.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-10-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/028 | 14 ++++----------
tests/qemu-iotests/028.out | 3 +++
tests/qemu-iotests/089 | 3 ++-
tests/qemu-iotests/089.out | 1 +
4 files changed, 10 insertions(+), 11 deletions(-)
diff --git a/tests/qemu-iotests/028 b/tests/qemu-iotests/028
index 6dd3ae09a3..864dc4a4e2 100755
--- a/tests/qemu-iotests/028
+++ b/tests/qemu-iotests/028
@@ -116,16 +116,10 @@ else
QEMU_COMM_TIMEOUT=1
fi
-# Silence output since it contains the disk image path and QEMU's readline
-# character echoing makes it very hard to filter the output. Plus, there
-# is no telling how many times the command will repeat before succeeding.
-# (Note that creating the image results in a "Formatting..." message over
-# stdout, which is the same channel the monitor uses. We cannot reliably
-# wait for it because the monitor output may interact with it in such a
-# way that _timed_wait_for cannot read it. However, once the block job is
-# done, we know that the "Formatting..." message must have appeared
-# already, so the output is still deterministic.)
-silent=y _send_qemu_cmd $h "drive_backup disk ${TEST_IMG}.copy" "(qemu)"
+TEST_IMG="$TEST_IMG.copy" _make_test_img $image_size
+_send_qemu_cmd $h "drive_backup -n disk ${TEST_IMG}.copy" "(qemu)" \
+ | _filter_imgfmt
+
silent=y qemu_cmd_repeat=20 _send_qemu_cmd $h "info block-jobs" "No active jobs"
_send_qemu_cmd $h "info block-jobs" "No active jobs"
_send_qemu_cmd $h 'quit' ""
diff --git a/tests/qemu-iotests/028.out b/tests/qemu-iotests/028.out
index 5a68de5c46..e580488216 100644
--- a/tests/qemu-iotests/028.out
+++ b/tests/qemu-iotests/028.out
@@ -468,6 +468,9 @@ No errors were found on the image.
block-backup
+Formatting 'TEST_DIR/t.IMGFMT.copy', fmt=IMGFMT size=4294968832
+QEMU X.Y.Z monitor - type 'help' for more information
+(qemu) drive_backup -n disk TEST_DIR/t.IMGFMT.copy
(qemu) info block-jobs
No active jobs
=== IO: pattern 195
diff --git a/tests/qemu-iotests/089 b/tests/qemu-iotests/089
index 66c5415abe..03a2ccf1e8 100755
--- a/tests/qemu-iotests/089
+++ b/tests/qemu-iotests/089
@@ -62,7 +62,8 @@ TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
$QEMU_IO -c 'write -P 42 0 512' -c 'write -P 23 512 512' \
-c 'write -P 66 1024 512' "$TEST_IMG.base" | _filter_qemu_io
-$QEMU_IMG convert -f raw -O $IMGFMT "$TEST_IMG.base" "$TEST_IMG"
+_make_test_img $IMG_SIZE
+$QEMU_IMG convert -f raw -O $IMGFMT -n "$TEST_IMG.base" "$TEST_IMG"
$QEMU_IO_PROG --cache $CACHEMODE --aio $AIOMODE \
-c 'read -P 42 0 512' -c 'read -P 23 512 512' \
diff --git a/tests/qemu-iotests/089.out b/tests/qemu-iotests/089.out
index 15682c2886..c53fc4823a 100644
--- a/tests/qemu-iotests/089.out
+++ b/tests/qemu-iotests/089.out
@@ -9,6 +9,7 @@ wrote 512/512 bytes at offset 512
512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
wrote 512/512 bytes at offset 1024
512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
read 512/512 bytes at offset 0
512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
read 512/512 bytes at offset 512
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 14/34] iotests/046: Avoid renaming images
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (12 preceding siblings ...)
2020-12-11 17:07 ` [PULL 13/34] iotests: Use convert -n in some cases Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 15/34] iotests: Derive image names from $TEST_IMG Kevin Wolf
` (20 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
This generally does not work on non-file protocols. It is better to
create the image with the final name from the start, and most tests do
this already. Let 046 follow suit.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-11-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/046 | 5 +++--
tests/qemu-iotests/046.out | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/tests/qemu-iotests/046 b/tests/qemu-iotests/046
index 88b3363c19..40a9f30087 100755
--- a/tests/qemu-iotests/046
+++ b/tests/qemu-iotests/046
@@ -47,6 +47,8 @@ size=128M
echo
echo "== creating backing file for COW tests =="
+TEST_IMG_SAVE=$TEST_IMG
+TEST_IMG="$TEST_IMG.base"
_make_test_img $size
backing_io()
@@ -67,8 +69,7 @@ backing_io()
backing_io 0 32 write | $QEMU_IO "$TEST_IMG" | _filter_qemu_io
-mv "$TEST_IMG" "$TEST_IMG.base"
-
+TEST_IMG=$TEST_IMG_SAVE
_make_test_img -b "$TEST_IMG.base" -F $IMGFMT 6G
echo
diff --git a/tests/qemu-iotests/046.out b/tests/qemu-iotests/046.out
index b022bcddd5..66ad987ab3 100644
--- a/tests/qemu-iotests/046.out
+++ b/tests/qemu-iotests/046.out
@@ -1,7 +1,7 @@
QA output created by 046
== creating backing file for COW tests ==
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=134217728
wrote 65536/65536 bytes at offset 0
64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
wrote 65536/65536 bytes at offset 65536
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 15/34] iotests: Derive image names from $TEST_IMG
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (13 preceding siblings ...)
2020-12-11 17:07 ` [PULL 14/34] iotests/046: Avoid renaming images Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 16/34] iotests/091: Use _cleanup_qemu instad of "wait" Kevin Wolf
` (19 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Avoid creating images with custom filenames in $TEST_DIR, because
non-file protocols may want to keep $TEST_IMG (and all other test
images) in some other directory.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-12-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/200 | 3 +--
tests/qemu-iotests/200.out | 4 ++--
tests/qemu-iotests/229 | 3 +--
tests/qemu-iotests/229.out | 6 +++---
4 files changed, 7 insertions(+), 9 deletions(-)
diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
index 59f7854b9f..a7aabbd032 100755
--- a/tests/qemu-iotests/200
+++ b/tests/qemu-iotests/200
@@ -44,8 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
_supported_fmt qcow2 qed
_supported_proto file
-BACKING_IMG="${TEST_DIR}/backing.img"
-TEST_IMG="${TEST_DIR}/test.img"
+BACKING_IMG="$TEST_IMG.base"
TEST_IMG="$BACKING_IMG" _make_test_img 512M
_make_test_img -F $IMGFMT -b "$BACKING_IMG" 512M
diff --git a/tests/qemu-iotests/200.out b/tests/qemu-iotests/200.out
index a6776070e4..5883f16ac3 100644
--- a/tests/qemu-iotests/200.out
+++ b/tests/qemu-iotests/200.out
@@ -1,6 +1,6 @@
QA output created by 200
-Formatting 'TEST_DIR/backing.img', fmt=IMGFMT size=536870912
-Formatting 'TEST_DIR/test.img', fmt=IMGFMT size=536870912 backing_file=TEST_DIR/backing.img backing_fmt=IMGFMT
+Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=536870912
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=536870912 backing_file=TEST_DIR/t.IMGFMT.base backing_fmt=IMGFMT
wrote 314572800/314572800 bytes at offset 512
300 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
diff --git a/tests/qemu-iotests/229 b/tests/qemu-iotests/229
index 89a5359f32..5f759fa587 100755
--- a/tests/qemu-iotests/229
+++ b/tests/qemu-iotests/229
@@ -51,8 +51,7 @@ _supported_os Linux
_unsupported_imgopts data_file
-DEST_IMG="$TEST_DIR/d.$IMGFMT"
-TEST_IMG="$TEST_DIR/b.$IMGFMT"
+DEST_IMG="$TEST_IMG.dest"
BLKDEBUG_CONF="$TEST_DIR/blkdebug.conf"
_make_test_img 2M
diff --git a/tests/qemu-iotests/229.out b/tests/qemu-iotests/229.out
index 4de6dfaa28..7eed393013 100644
--- a/tests/qemu-iotests/229.out
+++ b/tests/qemu-iotests/229.out
@@ -1,6 +1,6 @@
QA output created by 229
-Formatting 'TEST_DIR/b.IMGFMT', fmt=IMGFMT size=2097152
-Formatting 'TEST_DIR/d.IMGFMT', fmt=IMGFMT size=2097152
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
+Formatting 'TEST_DIR/t.IMGFMT.dest', fmt=IMGFMT size=2097152
wrote 2097152/2097152 bytes at offset 0
2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
{'execute': 'qmp_capabilities'}
@@ -8,7 +8,7 @@ wrote 2097152/2097152 bytes at offset 0
=== Starting drive-mirror, causing error & stop ===
-{'execute': 'drive-mirror', 'arguments': {'device': 'testdisk', 'format': 'IMGFMT', 'target': 'blkdebug:TEST_DIR/blkdebug.conf:TEST_DIR/d.IMGFMT', 'sync': 'full', 'mode': 'existing', 'on-source-error': 'stop', 'on-target-error': 'stop' }}
+{'execute': 'drive-mirror', 'arguments': {'device': 'testdisk', 'format': 'IMGFMT', 'target': 'blkdebug:TEST_DIR/blkdebug.conf:TEST_DIR/t.IMGFMT.dest', 'sync': 'full', 'mode': 'existing', 'on-source-error': 'stop', 'on-target-error': 'stop' }}
{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "testdisk"}}
{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "testdisk"}}
{"return": {}}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 16/34] iotests/091: Use _cleanup_qemu instad of "wait"
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (14 preceding siblings ...)
2020-12-11 17:07 ` [PULL 15/34] iotests: Derive image names from $TEST_IMG Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 17/34] iotests: Restrict some Python tests to file Kevin Wolf
` (18 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
If the test environment has some other child processes running (like a
storage daemon that provides a FUSE export), then "wait" will never
finish. Use wait=yes _cleanup_qemu instead.
(We need to discard the output so there is no change to the reference
output.)
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-13-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/091 | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/091 b/tests/qemu-iotests/091
index 68fbfd777b..8a4ce5b7e2 100755
--- a/tests/qemu-iotests/091
+++ b/tests/qemu-iotests/091
@@ -96,7 +96,8 @@ _send_qemu_cmd $h2 'qemu-io disk flush' "(qemu)"
_send_qemu_cmd $h2 'quit' ""
_send_qemu_cmd $h1 'quit' ""
-wait
+wait=yes _cleanup_qemu >/dev/null
+
echo "Check image pattern"
${QEMU_IO} -c "read -P 0x22 0 4M" "${TEST_IMG}" | _filter_testdir | _filter_qemu_io
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 17/34] iotests: Restrict some Python tests to file
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (15 preceding siblings ...)
2020-12-11 17:07 ` [PULL 16/34] iotests/091: Use _cleanup_qemu instad of "wait" Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 18/34] iotests: Let _make_test_img guess $TEST_IMG_FILE Kevin Wolf
` (17 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Most Python tests are restricted to the file protocol (without
explicitly saying so), but these are the ones that would break
./check -fuse -qcow2.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-14-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/206 | 3 ++-
tests/qemu-iotests/242 | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/tests/qemu-iotests/206 b/tests/qemu-iotests/206
index 11bc51f256..0a3ee5ef00 100755
--- a/tests/qemu-iotests/206
+++ b/tests/qemu-iotests/206
@@ -23,7 +23,8 @@
import iotests
from iotests import imgfmt
-iotests.script_initialize(supported_fmts=['qcow2'])
+iotests.script_initialize(supported_fmts=['qcow2'],
+ supported_protocols=['file'])
iotests.verify_working_luks()
with iotests.FilePath('t.qcow2') as disk_path, \
diff --git a/tests/qemu-iotests/242 b/tests/qemu-iotests/242
index 64f1bd95e4..a16de3085f 100755
--- a/tests/qemu-iotests/242
+++ b/tests/qemu-iotests/242
@@ -24,7 +24,8 @@ import struct
from iotests import qemu_img_create, qemu_io, qemu_img_pipe, \
file_path, img_info_log, log, filter_qemu_io
-iotests.script_initialize(supported_fmts=['qcow2'])
+iotests.script_initialize(supported_fmts=['qcow2'],
+ supported_protocols=['file'])
disk = file_path('disk')
chunk = 256 * 1024
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 18/34] iotests: Let _make_test_img guess $TEST_IMG_FILE
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (16 preceding siblings ...)
2020-12-11 17:07 ` [PULL 17/34] iotests: Restrict some Python tests to file Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 19/34] iotests/287: Clean up subshell test image Kevin Wolf
` (16 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
When most iotests want to create a test image that is named differently
from the default $TEST_IMG, they do something like this:
TEST_IMG="$TEST_IMG.base" _make_test_img $options
This works fine with the "file" protocol, but not so much for anything
else: _make_test_img tries to create an image under $TEST_IMG_FILE
first, and only under $TEST_IMG if the former is not set; and on
everything but "file", $TEST_IMG_FILE is set.
There are two ways we can fix this: First, we could make all tests
adjust not only TEST_IMG, but also TEST_IMG_FILE if that is present
(e.g. with something like _set_test_img_suffix $suffix that would affect
not only TEST_IMG but also TEST_IMG_FILE, if necessary). This is a
pretty clean solution, and this is maybe what we should have done from
the start.
But it would also require changes to most existing bash tests. So the
alternative is this: Let _make_test_img see whether $TEST_IMG_FILE still
points to the original value. If so, it is possible that the caller has
adjusted $TEST_IMG but not $TEST_IMG_FILE. In such a case, we can (for
most protocols) derive the corresponding $TEST_IMG_FILE value from
$TEST_IMG value and thus work around what technically is the caller
misbehaving.
This second solution is less clean, but it is robust against people
keeping their old habit of adjusting TEST_IMG only, and requires much
less changes. So this patch implements it.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-15-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/common.rc | 40 +++++++++++++++++++++++++++++++++---
1 file changed, 37 insertions(+), 3 deletions(-)
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index 494490a272..23f46da2db 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -268,6 +268,7 @@ else
TEST_IMG=$IMGPROTO:$TEST_DIR/t.$IMGFMT
fi
fi
+ORIG_TEST_IMG_FILE=$TEST_IMG_FILE
ORIG_TEST_IMG="$TEST_IMG"
if [ -z "$TEST_DIR" ]; then
@@ -330,6 +331,30 @@ _get_data_file()
| sed -e "s#\\\$TEST_IMG#$1#"
}
+# Translate a $TEST_IMG to its corresponding $TEST_IMG_FILE for
+# different protocols
+_test_img_to_test_img_file()
+{
+ case "$IMGPROTO" in
+ file)
+ echo "$1"
+ ;;
+
+ nfs)
+ echo "$1" | sed -e "s#nfs://127.0.0.1##"
+ ;;
+
+ ssh)
+ echo "$1" | \
+ sed -e "s#ssh://\\($USER@\\)\\?127.0.0.1\\(:[0-9]\\+\\)\\?##"
+ ;;
+
+ *)
+ return 1
+ ;;
+ esac
+}
+
_make_test_img()
{
# extra qemu-img options can be added by tests
@@ -343,10 +368,19 @@ _make_test_img()
local opts_param=false
local misc_params=()
- if [ -n "$TEST_IMG_FILE" ]; then
- img_name=$TEST_IMG_FILE
- else
+ if [ -z "$TEST_IMG_FILE" ]; then
img_name=$TEST_IMG
+ elif [ "$IMGOPTSSYNTAX" != "true" -a \
+ "$TEST_IMG_FILE" = "$ORIG_TEST_IMG_FILE" ]; then
+ # Handle cases of tests only updating TEST_IMG, but not TEST_IMG_FILE
+ img_name=$(_test_img_to_test_img_file "$TEST_IMG")
+ if [ "$?" != 0 ]; then
+ img_name=$TEST_IMG_FILE
+ fi
+ else
+ # $TEST_IMG_FILE is not the default value, so it definitely has been
+ # modified by the test
+ img_name=$TEST_IMG_FILE
fi
if [ -n "$IMGOPTS" ]; then
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 19/34] iotests/287: Clean up subshell test image
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (17 preceding siblings ...)
2020-12-11 17:07 ` [PULL 18/34] iotests: Let _make_test_img guess $TEST_IMG_FILE Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 20/34] storage-daemon: Call bdrv_close_all() on exit Kevin Wolf
` (15 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
287 creates an image in a subshell (thanks to the pipe) to see whether
that is possible with compression_type=zstd. If _make_test_img were to
modify any global state, this global state would then be lost before we
could cleanup the image.
When using FUSE as the test protocol, this global state is important, so
clean up the image before the state is lost.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-16-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/287 | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/qemu-iotests/287 b/tests/qemu-iotests/287
index f98a4cadc1..036cc09e82 100755
--- a/tests/qemu-iotests/287
+++ b/tests/qemu-iotests/287
@@ -51,8 +51,8 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
CLUSTER_SIZE=65536
# Check if we can run this test.
-if IMGOPTS='compression_type=zstd' _make_test_img 64M |
- grep "Invalid parameter 'zstd'"; then
+output=$(_make_test_img -o 'compression_type=zstd' 64M; _cleanup_test_img)
+if echo "$output" | grep -q "Invalid parameter 'zstd'"; then
_notrun "ZSTD is disabled"
fi
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 20/34] storage-daemon: Call bdrv_close_all() on exit
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (18 preceding siblings ...)
2020-12-11 17:07 ` [PULL 19/34] iotests/287: Clean up subshell test image Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:07 ` [PULL 21/34] iotests: Give access to the qemu-storage-daemon Kevin Wolf
` (14 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Otherwise, exports and block devices are not properly shut down and
closed, unless the users explicitly issues blockdev-del and
block-export-del commands for each of them.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-17-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
storage-daemon/qemu-storage-daemon.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/storage-daemon/qemu-storage-daemon.c b/storage-daemon/qemu-storage-daemon.c
index 7c914b0dc1..e0c87edbdd 100644
--- a/storage-daemon/qemu-storage-daemon.c
+++ b/storage-daemon/qemu-storage-daemon.c
@@ -314,6 +314,9 @@ int main(int argc, char *argv[])
main_loop_wait(false);
}
+ bdrv_drain_all_begin();
+ bdrv_close_all();
+
monitor_cleanup();
qemu_chr_cleanup();
user_creatable_cleanup();
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 21/34] iotests: Give access to the qemu-storage-daemon
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (19 preceding siblings ...)
2020-12-11 17:07 ` [PULL 20/34] storage-daemon: Call bdrv_close_all() on exit Kevin Wolf
@ 2020-12-11 17:07 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 22/34] iotests: Allow testing FUSE exports Kevin Wolf
` (13 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:07 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-18-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/check | 11 +++++++++++
tests/qemu-iotests/common.rc | 17 +++++++++++++++++
2 files changed, 28 insertions(+)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
index 678b6e4910..3c1fa4435a 100755
--- a/tests/qemu-iotests/check
+++ b/tests/qemu-iotests/check
@@ -644,6 +644,17 @@ if [ -z $QEMU_NBD_PROG ]; then
fi
export QEMU_NBD_PROG="$(type -p "$QEMU_NBD_PROG")"
+if [ -z "$QSD_PROG" ]; then
+ if [ -x "$build_iotests/qemu-storage-daemon" ]; then
+ export QSD_PROG="$build_iotests/qemu-storage-daemon"
+ elif [ -x "$build_root/storage-daemon/qemu-storage-daemon" ]; then
+ export QSD_PROG="$build_root/storage-daemon/qemu-storage-daemon"
+ else
+ _init_error "qemu-storage-daemon not found"
+ fi
+fi
+export QSD_PROG="$(type -p "$QSD_PROG")"
+
if [ -x "$build_iotests/socket_scm_helper" ]
then
export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper"
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index 23f46da2db..20589e59a5 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -124,6 +124,7 @@ fi
: ${VALGRIND_QEMU_IMG=$VALGRIND_QEMU}
: ${VALGRIND_QEMU_IO=$VALGRIND_QEMU}
: ${VALGRIND_QEMU_NBD=$VALGRIND_QEMU}
+: ${VALGRIND_QSD=$VALGRIND_QEMU}
# The Valgrind own parameters may be set with
# its environment variable VALGRIND_OPTS, e.g.
@@ -211,6 +212,21 @@ _qemu_nbd_wrapper()
return $RETVAL
}
+_qemu_storage_daemon_wrapper()
+{
+ local VALGRIND_LOGFILE="${TEST_DIR}"/$$.valgrind
+ (
+ if [ -n "${QSD_NEED_PID}" ]; then
+ echo $BASHPID > "${QEMU_TEST_DIR}/qemu-storage-daemon.pid"
+ fi
+ VALGRIND_QEMU="${VALGRIND_QSD}" _qemu_proc_exec "${VALGRIND_LOGFILE}" \
+ "$QSD_PROG" $QSD_OPTIONS "$@"
+ )
+ RETVAL=$?
+ _qemu_proc_valgrind_log "${VALGRIND_LOGFILE}" $RETVAL
+ return $RETVAL
+}
+
# Valgrind bug #409141 https://bugs.kde.org/show_bug.cgi?id=409141
# Until valgrind 3.16+ is ubiquitous, we must work around a hang in
# valgrind when issuing sigkill. Disable valgrind for this invocation.
@@ -223,6 +239,7 @@ export QEMU=_qemu_wrapper
export QEMU_IMG=_qemu_img_wrapper
export QEMU_IO=_qemu_io_wrapper
export QEMU_NBD=_qemu_nbd_wrapper
+export QSD=_qemu_storage_daemon_wrapper
if [ "$IMGOPTSSYNTAX" = "true" ]; then
DRIVER="driver=$IMGFMT"
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 22/34] iotests: Allow testing FUSE exports
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (20 preceding siblings ...)
2020-12-11 17:07 ` [PULL 21/34] iotests: Give access to the qemu-storage-daemon Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 23/34] iotests: Enable fuse for many tests Kevin Wolf
` (12 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
This pretends FUSE exports are a kind of protocol. As such, they are
always tested under the format node. This is probably the best way to
test them, actually, because this will generate more I/O load and more
varied patterns.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201027190600.192171-19-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/check | 6 ++
tests/qemu-iotests/common.filter | 5 +-
tests/qemu-iotests/common.rc | 124 +++++++++++++++++++++++++++++++
3 files changed, 134 insertions(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
index 3c1fa4435a..952762d5ed 100755
--- a/tests/qemu-iotests/check
+++ b/tests/qemu-iotests/check
@@ -270,6 +270,7 @@ image protocol options
-rbd test rbd
-sheepdog test sheepdog
-nbd test nbd
+ -fuse test fuse
-ssh test ssh
-nfs test nfs
@@ -382,6 +383,11 @@ testlist options
xpand=false
;;
+ -fuse)
+ IMGPROTO=fuse
+ xpand=false
+ ;;
+
-ssh)
IMGPROTO=ssh
xpand=false
diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
index 838ed15793..172ea5752e 100644
--- a/tests/qemu-iotests/common.filter
+++ b/tests/qemu-iotests/common.filter
@@ -44,7 +44,8 @@ _filter_qom_path()
_filter_testdir()
{
$SED -e "s#$TEST_DIR/#TEST_DIR/#g" \
- -e "s#$SOCK_DIR/#SOCK_DIR/#g"
+ -e "s#$SOCK_DIR/#SOCK_DIR/#g" \
+ -e "s#SOCK_DIR/fuse-#TEST_DIR/#g"
}
# replace occurrences of the actual IMGFMT value with IMGFMT
@@ -127,6 +128,7 @@ _filter_img_create_filenames()
-e "s#$IMGPROTO:$TEST_DIR#TEST_DIR#g" \
-e "s#$TEST_DIR#TEST_DIR#g" \
-e "s#$SOCK_DIR#SOCK_DIR#g" \
+ -e 's#SOCK_DIR/fuse-#TEST_DIR/#g' \
-e "s#$IMGFMT#IMGFMT#g" \
-e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
-e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g'
@@ -227,6 +229,7 @@ _filter_img_info()
-e "s#$IMGFMT#IMGFMT#g" \
-e 's#nbd://127.0.0.1:[0-9]\\+$#TEST_DIR/t.IMGFMT#g' \
-e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
+ -e 's#SOCK_DIR/fuse-#TEST_DIR/#g' \
-e "/encrypted: yes/d" \
-e "/cluster_size: [0-9]\\+/d" \
-e "/table_size: [0-9]\\+/d" \
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index 20589e59a5..29354654cc 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -257,6 +257,9 @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then
TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix"
TEST_IMG="$TEST_IMG,file.path=$SOCK_DIR/nbd"
+ elif [ "$IMGPROTO" = "fuse" ]; then
+ TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+ TEST_IMG="$DRIVER,file.filename=$SOCK_DIR/fuse-t.$IMGFMT"
elif [ "$IMGPROTO" = "ssh" ]; then
TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
@@ -273,6 +276,9 @@ else
elif [ "$IMGPROTO" = "nbd" ]; then
TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"
+ elif [ "$IMGPROTO" = "fuse" ]; then
+ TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+ TEST_IMG="$SOCK_DIR/fuse-t.$IMGFMT"
elif [ "$IMGPROTO" = "ssh" ]; then
TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
REMOTE_TEST_DIR="ssh://\\($USER@\\)\\?127.0.0.1\\(:[0-9]\\+\\)\\?$TEST_DIR"
@@ -288,6 +294,9 @@ fi
ORIG_TEST_IMG_FILE=$TEST_IMG_FILE
ORIG_TEST_IMG="$TEST_IMG"
+FUSE_PIDS=()
+FUSE_EXPORTS=()
+
if [ -z "$TEST_DIR" ]; then
TEST_DIR=$PWD/scratch
fi
@@ -357,6 +366,10 @@ _test_img_to_test_img_file()
echo "$1"
;;
+ fuse)
+ echo "$1" | sed -e "s#$SOCK_DIR/fuse-#$TEST_DIR/#"
+ ;;
+
nfs)
echo "$1" | sed -e "s#nfs://127.0.0.1##"
;;
@@ -385,6 +398,11 @@ _make_test_img()
local opts_param=false
local misc_params=()
+ if [[ $IMGPROTO == fuse && $TEST_IMG == $SOCK_DIR/fuse-* ]]; then
+ # The caller may be trying to overwrite an existing image
+ _rm_test_img "$TEST_IMG"
+ fi
+
if [ -z "$TEST_IMG_FILE" ]; then
img_name=$TEST_IMG
elif [ "$IMGOPTSSYNTAX" != "true" -a \
@@ -469,11 +487,105 @@ _make_test_img()
eval "$QEMU_NBD -v -t -k '$SOCK_DIR/nbd' -f $IMGFMT -e 42 -x '' $TEST_IMG_FILE >/dev/null &"
sleep 1 # FIXME: qemu-nbd needs to be listening before we continue
fi
+
+ if [ $IMGPROTO = "fuse" -a -f "$img_name" ]; then
+ local export_mp
+ local pid
+ local pidfile
+ local timeout
+
+ export_mp=$(echo "$img_name" | sed -e "s#$TEST_DIR/#$SOCK_DIR/fuse-#")
+ if ! echo "$export_mp" | grep -q "^$SOCK_DIR"; then
+ echo 'Cannot use FUSE exports with images outside of TEST_DIR' >&2
+ return 1
+ fi
+
+ touch "$export_mp"
+ rm -f "$SOCK_DIR/fuse-output"
+
+ # Usually, users would export formatted nodes. But we present fuse as a
+ # protocol-level driver here, so we have to leave the format to the
+ # client.
+ QSD_NEED_PID=y $QSD \
+ --blockdev file,node-name=export-node,filename=$img_name,discard=unmap \
+ --export fuse,id=fuse-export,node-name=export-node,mountpoint="$export_mp",writable=on,growable=on \
+ &
+
+ pidfile="$QEMU_TEST_DIR/qemu-storage-daemon.pid"
+
+ # Wait for the PID file
+ while [ ! -f "$pidfile" ]; do
+ sleep 0.5
+ done
+
+ pid=$(cat "$pidfile")
+ rm -f "$pidfile"
+
+ FUSE_PIDS+=($pid)
+ FUSE_EXPORTS+=("$export_mp")
+ fi
}
_rm_test_img()
{
local img=$1
+
+ if [[ $IMGPROTO == fuse && $img == $SOCK_DIR/fuse-* ]]; then
+ # Drop a FUSE export
+ local df_output
+ local i
+ local image_file
+ local index=''
+ local timeout
+
+ for i in "${!FUSE_EXPORTS[@]}"; do
+ if [ "${FUSE_EXPORTS[i]}" = "$img" ]; then
+ index=$i
+ break
+ fi
+ done
+
+ if [ -z "$index" ]; then
+ # Probably gone already
+ return 0
+ fi
+
+ kill "${FUSE_PIDS[index]}"
+
+ # Wait until the mount is gone
+ timeout=10 # *0.5 s
+ while true; do
+ # Will show the mount point; if the mount is still there,
+ # it will be $img.
+ df_output=$(df "$img" 2>/dev/null)
+
+ # But df may also show an error ("Transpoint endpoint not
+ # connected"), so retry in such cases
+ if [ -n "$df_output" ]; then
+ if ! echo "$df_output" | grep -q "$img"; then
+ break
+ fi
+ fi
+
+ sleep 0.5
+
+ timeout=$((timeout - 1))
+ if [ "$timeout" = 0 ]; then
+ echo 'Failed to take down FUSE export' >&2
+ return 1
+ fi
+ done
+
+ rm -f "$img"
+
+ unset "FUSE_PIDS[$index]"
+ unset "FUSE_EXPORTS[$index]"
+
+ image_file=$(echo "$img" | sed -e "s#$SOCK_DIR/fuse-#$TEST_DIR/#")
+ _rm_test_img "$image_file"
+ return
+ fi
+
if [ "$IMGFMT" = "vmdk" ]; then
# Remove all the extents for vmdk
"$QEMU_IMG" info "$img" 2>/dev/null | grep 'filename:' | cut -f 2 -d: \
@@ -496,6 +608,17 @@ _cleanup_test_img()
rm -f "$TEST_IMG_FILE"
;;
+ fuse)
+ local mp
+
+ for mp in "${FUSE_EXPORTS[@]}"; do
+ _rm_test_img "$mp"
+ done
+
+ FUSE_PIDS=()
+ FUSE_EXPORTS=()
+ ;;
+
file)
_rm_test_img "$TEST_DIR/t.$IMGFMT"
_rm_test_img "$TEST_DIR/t.$IMGFMT.orig"
@@ -562,6 +685,7 @@ _img_info()
sed -e "s#$REMOTE_TEST_DIR#TEST_DIR#g" \
-e "s#$IMGPROTO:$TEST_DIR#TEST_DIR#g" \
-e "s#$TEST_DIR#TEST_DIR#g" \
+ -e "s#$SOCK_DIR/fuse-#TEST_DIR/#g" \
-e "s#$IMGFMT#IMGFMT#g" \
-e "/^disk size:/ D" \
-e "/actual-size/ D" | \
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 23/34] iotests: Enable fuse for many tests
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (21 preceding siblings ...)
2020-12-11 17:08 ` [PULL 22/34] iotests: Allow testing FUSE exports Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 24/34] iotests/308: Add test for FUSE exports Kevin Wolf
` (11 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
Many tests (that do not support generic protocols) can run just fine
with FUSE-exported images, so allow them to. Note that this is no
attempt at being definitely complete. There are some tests that might
be modified to run on FUSE, but this patch still skips them. This patch
only tries to pick the rather low-hanging fruits.
Note that 221 and 250 only pass when .lseek is correctly implemented,
which is only possible with a libfuse that is 3.8 or newer.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-20-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/025 | 2 +-
tests/qemu-iotests/026 | 2 +-
tests/qemu-iotests/028 | 2 +-
tests/qemu-iotests/031 | 2 +-
tests/qemu-iotests/034 | 2 +-
tests/qemu-iotests/036 | 2 +-
tests/qemu-iotests/037 | 2 +-
tests/qemu-iotests/038 | 2 +-
tests/qemu-iotests/039 | 2 +-
tests/qemu-iotests/046 | 2 +-
tests/qemu-iotests/050 | 2 +-
tests/qemu-iotests/054 | 2 +-
tests/qemu-iotests/060 | 2 +-
tests/qemu-iotests/071 | 2 +-
tests/qemu-iotests/079 | 2 +-
tests/qemu-iotests/080 | 2 +-
tests/qemu-iotests/089 | 2 +-
tests/qemu-iotests/090 | 2 +-
tests/qemu-iotests/091 | 2 +-
tests/qemu-iotests/095 | 2 +-
tests/qemu-iotests/097 | 2 +-
tests/qemu-iotests/098 | 2 +-
tests/qemu-iotests/102 | 2 +-
tests/qemu-iotests/103 | 2 +-
tests/qemu-iotests/106 | 2 +-
tests/qemu-iotests/107 | 2 +-
tests/qemu-iotests/108 | 2 +-
tests/qemu-iotests/111 | 2 +-
tests/qemu-iotests/112 | 2 +-
tests/qemu-iotests/115 | 2 +-
tests/qemu-iotests/117 | 2 +-
tests/qemu-iotests/120 | 2 +-
tests/qemu-iotests/121 | 2 +-
tests/qemu-iotests/127 | 2 +-
tests/qemu-iotests/133 | 2 +-
tests/qemu-iotests/137 | 2 +-
tests/qemu-iotests/138 | 2 +-
tests/qemu-iotests/140 | 2 +-
tests/qemu-iotests/154 | 2 +-
tests/qemu-iotests/161 | 2 +-
tests/qemu-iotests/171 | 2 +-
tests/qemu-iotests/175 | 2 +-
tests/qemu-iotests/176 | 2 +-
tests/qemu-iotests/177 | 2 +-
tests/qemu-iotests/179 | 2 +-
tests/qemu-iotests/183 | 2 +-
tests/qemu-iotests/186 | 2 +-
tests/qemu-iotests/187 | 2 +-
tests/qemu-iotests/191 | 2 +-
tests/qemu-iotests/195 | 2 +-
tests/qemu-iotests/200 | 2 +-
tests/qemu-iotests/204 | 2 +-
tests/qemu-iotests/214 | 2 +-
tests/qemu-iotests/217 | 2 +-
tests/qemu-iotests/220 | 2 +-
tests/qemu-iotests/221 | 2 +-
tests/qemu-iotests/229 | 2 +-
tests/qemu-iotests/247 | 2 +-
tests/qemu-iotests/249 | 2 +-
tests/qemu-iotests/250 | 2 +-
tests/qemu-iotests/252 | 2 +-
tests/qemu-iotests/265 | 2 +-
tests/qemu-iotests/268 | 2 +-
tests/qemu-iotests/272 | 2 +-
tests/qemu-iotests/273 | 2 +-
tests/qemu-iotests/279 | 2 +-
tests/qemu-iotests/286 | 2 +-
tests/qemu-iotests/287 | 2 +-
tests/qemu-iotests/289 | 2 +-
tests/qemu-iotests/290 | 2 +-
tests/qemu-iotests/291 | 2 +-
tests/qemu-iotests/292 | 2 +-
tests/qemu-iotests/293 | 2 +-
tests/qemu-iotests/294 | 2 +-
tests/qemu-iotests/305 | 2 +-
75 files changed, 75 insertions(+), 75 deletions(-)
diff --git a/tests/qemu-iotests/025 b/tests/qemu-iotests/025
index e05d833452..1569d912f4 100755
--- a/tests/qemu-iotests/025
+++ b/tests/qemu-iotests/025
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.pattern
_supported_fmt raw qcow2 qed luks
-_supported_proto file sheepdog rbd nfs
+_supported_proto file sheepdog rbd nfs fuse
echo "=== Creating image"
echo
diff --git a/tests/qemu-iotests/026 b/tests/qemu-iotests/026
index b9713eb591..9ecc5880b1 100755
--- a/tests/qemu-iotests/026
+++ b/tests/qemu-iotests/026
@@ -41,7 +41,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Currently only qcow2 supports rebasing
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_default_cache_mode writethrough
_supported_cache_modes writethrough none
# The refcount table tests expect a certain minimum width for refcount entries
diff --git a/tests/qemu-iotests/028 b/tests/qemu-iotests/028
index 864dc4a4e2..57d34aae99 100755
--- a/tests/qemu-iotests/028
+++ b/tests/qemu-iotests/028
@@ -46,7 +46,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Any format supporting backing files except vmdk and qcow which do not support
# smaller backing files.
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# Choose a size that is not necessarily a cluster size multiple for image
diff --git a/tests/qemu-iotests/031 b/tests/qemu-iotests/031
index 646ecd593f..2bcbc5886e 100755
--- a/tests/qemu-iotests/031
+++ b/tests/qemu-iotests/031
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# We want to test compat=0.10, which does not support external data
# files or refcount widths other than 16
_unsupported_imgopts data_file 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
diff --git a/tests/qemu-iotests/034 b/tests/qemu-iotests/034
index ac2d687c71..08f7aea6d5 100755
--- a/tests/qemu-iotests/034
+++ b/tests/qemu-iotests/034
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow qcow2 vmdk qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_unsupported_imgopts "subformat=monolithicFlat" \
"subformat=twoGbMaxExtentFlat" \
diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
index cf522de7a1..6b82638080 100755
--- a/tests/qemu-iotests/036
+++ b/tests/qemu-iotests/036
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# Only qcow2v3 and later supports feature bits;
# qcow2.py does not support external data files;
# this test requires a cluster size large enough for the feature table
diff --git a/tests/qemu-iotests/037 b/tests/qemu-iotests/037
index e1187ac24a..bb893c43dc 100755
--- a/tests/qemu-iotests/037
+++ b/tests/qemu-iotests/037
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow qcow2 vmdk qed
-_supported_proto file
+_supported_proto file fuse
_unsupported_imgopts "subformat=monolithicFlat" \
"subformat=twoGbMaxExtentFlat" \
"subformat=twoGbMaxExtentSparse" \
diff --git a/tests/qemu-iotests/038 b/tests/qemu-iotests/038
index a253231f5b..30f1f73c25 100755
--- a/tests/qemu-iotests/038
+++ b/tests/qemu-iotests/038
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
CLUSTER_SIZE=2M
diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039
index 42f6503138..ad3867c3fc 100755
--- a/tests/qemu-iotests/039
+++ b/tests/qemu-iotests/039
@@ -40,7 +40,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_default_cache_mode writethrough
_supported_cache_modes writethrough
diff --git a/tests/qemu-iotests/046 b/tests/qemu-iotests/046
index 40a9f30087..ed6fae3529 100755
--- a/tests/qemu-iotests/046
+++ b/tests/qemu-iotests/046
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# data_file does not support compressed clusters
_unsupported_imgopts data_file
diff --git a/tests/qemu-iotests/050 b/tests/qemu-iotests/050
index 4b0a390c43..741bdb610e 100755
--- a/tests/qemu-iotests/050
+++ b/tests/qemu-iotests/050
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
echo
echo "== Creating images =="
diff --git a/tests/qemu-iotests/054 b/tests/qemu-iotests/054
index a8905b60d0..40922db2b1 100755
--- a/tests/qemu-iotests/054
+++ b/tests/qemu-iotests/054
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
echo
echo "creating too large image (1 EB)"
diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
index 94c0d5accc..4b81d1aa51 100755
--- a/tests/qemu-iotests/060
+++ b/tests/qemu-iotests/060
@@ -46,7 +46,7 @@ _filter_io_error()
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# These tests only work for compat=1.1 images without an external
# data file with refcount_bits=16
diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
index 18fe9054b0..49faae6684 100755
--- a/tests/qemu-iotests/071
+++ b/tests/qemu-iotests/071
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_require_drivers blkdebug blkverify
# blkdebug can only inject errors on bs->file, not on the data_file,
# so thie test does not work with external data files
diff --git a/tests/qemu-iotests/079 b/tests/qemu-iotests/079
index 3642b51feb..0f0d94a2ac 100755
--- a/tests/qemu-iotests/079
+++ b/tests/qemu-iotests/079
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file nfs
+_supported_proto file nfs fuse
# Some containers (e.g. non-x86 on Travis) do not allow large files
_require_large_file 4G
diff --git a/tests/qemu-iotests/080 b/tests/qemu-iotests/080
index 7588c63b6c..bda8617c38 100755
--- a/tests/qemu-iotests/080
+++ b/tests/qemu-iotests/080
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# - Internal snapshots are (currently) impossible with refcount_bits=1,
# and generally impossible with external data files
diff --git a/tests/qemu-iotests/089 b/tests/qemu-iotests/089
index 03a2ccf1e8..f0929b64c0 100755
--- a/tests/qemu-iotests/089
+++ b/tests/qemu-iotests/089
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# Because anything other than 16 would change the output of qemu_io -c info
_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
diff --git a/tests/qemu-iotests/090 b/tests/qemu-iotests/090
index 1246e4f910..87e872ebf4 100755
--- a/tests/qemu-iotests/090
+++ b/tests/qemu-iotests/090
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file nfs
+_supported_proto file nfs fuse
# External data files do not support compressed clusters
_unsupported_imgopts data_file
diff --git a/tests/qemu-iotests/091 b/tests/qemu-iotests/091
index 8a4ce5b7e2..8dee168bf6 100755
--- a/tests/qemu-iotests/091
+++ b/tests/qemu-iotests/091
@@ -44,7 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_supported_cache_modes writethrough none writeback
_default_cache_mode none writeback
diff --git a/tests/qemu-iotests/095 b/tests/qemu-iotests/095
index 77a5f0f3f5..7604ae6966 100755
--- a/tests/qemu-iotests/095
+++ b/tests/qemu-iotests/095
@@ -44,7 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
size_smaller=5M
size_larger=100M
diff --git a/tests/qemu-iotests/097 b/tests/qemu-iotests/097
index d910a8b107..1837d4e8e0 100755
--- a/tests/qemu-iotests/097
+++ b/tests/qemu-iotests/097
@@ -41,7 +41,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Any format supporting backing files and bdrv_make_empty
_supported_fmt qcow qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
diff --git a/tests/qemu-iotests/098 b/tests/qemu-iotests/098
index f2ccdd7909..a35ce7205e 100755
--- a/tests/qemu-iotests/098
+++ b/tests/qemu-iotests/098
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.pattern
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# The code path we want to test here only works for compat=1.1 images;
# blkdebug can only inject errors on bs->file, so external data files
# do not work with this test
diff --git a/tests/qemu-iotests/102 b/tests/qemu-iotests/102
index b898df436f..2cc3efd1ed 100755
--- a/tests/qemu-iotests/102
+++ b/tests/qemu-iotests/102
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
IMG_SIZE=64K
diff --git a/tests/qemu-iotests/103 b/tests/qemu-iotests/103
index 8c1ebe0443..220481db4c 100755
--- a/tests/qemu-iotests/103
+++ b/tests/qemu-iotests/103
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file nfs
+_supported_proto file nfs fuse
# Internal snapshots are (currently) impossible with refcount_bits=1,
# and generally impossible with external data files
_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
diff --git a/tests/qemu-iotests/106 b/tests/qemu-iotests/106
index a20659d443..20ad7bd5a2 100755
--- a/tests/qemu-iotests/106
+++ b/tests/qemu-iotests/106
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt raw
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# in kB
diff --git a/tests/qemu-iotests/107 b/tests/qemu-iotests/107
index 268ba27688..d24829ccf9 100755
--- a/tests/qemu-iotests/107
+++ b/tests/qemu-iotests/107
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file nfs
+_supported_proto file nfs fuse
IMG_SIZE=64K
diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
index 5f7076fba4..ba67748bdf 100755
--- a/tests/qemu-iotests/108
+++ b/tests/qemu-iotests/108
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# This test directly modifies a refblock so it relies on refcount_bits being 16;
# and the low-level modification it performs are not tuned for external data
diff --git a/tests/qemu-iotests/111 b/tests/qemu-iotests/111
index 3b43d1bd83..bd839a39f4 100755
--- a/tests/qemu-iotests/111
+++ b/tests/qemu-iotests/111
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qed qcow qcow2 vmdk
-_supported_proto file
+_supported_proto file fuse
_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
_make_test_img -b "$TEST_IMG.inexistent"
diff --git a/tests/qemu-iotests/112 b/tests/qemu-iotests/112
index 20ff5c224a..6e413f5651 100755
--- a/tests/qemu-iotests/112
+++ b/tests/qemu-iotests/112
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# This test will set refcount_bits on its own which would conflict with the
# manual setting; compat will be overridden as well;
# and external data files do not work well with our refcount testing
diff --git a/tests/qemu-iotests/115 b/tests/qemu-iotests/115
index d254b18342..7f53987d1b 100755
--- a/tests/qemu-iotests/115
+++ b/tests/qemu-iotests/115
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# This test relies on refcounts being 64 bits wide (which does not work with
# compat=0.10)
_unsupported_imgopts 'refcount_bits=\([^6]\|.\([^4]\|$\)\)' 'compat=0.10'
diff --git a/tests/qemu-iotests/117 b/tests/qemu-iotests/117
index f37b34f8b1..9039555ac4 100755
--- a/tests/qemu-iotests/117
+++ b/tests/qemu-iotests/117
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_make_test_img 64k
diff --git a/tests/qemu-iotests/120 b/tests/qemu-iotests/120
index 2931a7550f..45c55c1c01 100755
--- a/tests/qemu-iotests/120
+++ b/tests/qemu-iotests/120
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt generic
-_supported_proto file
+_supported_proto file fuse
_unsupported_fmt luks
_require_drivers raw
diff --git a/tests/qemu-iotests/121 b/tests/qemu-iotests/121
index 90ea0db737..8357ce089a 100755
--- a/tests/qemu-iotests/121
+++ b/tests/qemu-iotests/121
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# Refcount structures are used much differently with external data
# files
diff --git a/tests/qemu-iotests/127 b/tests/qemu-iotests/127
index e2ac2f5536..77fdfd0205 100755
--- a/tests/qemu-iotests/127
+++ b/tests/qemu-iotests/127
@@ -41,7 +41,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_require_devices virtio-scsi scsi-hd
diff --git a/tests/qemu-iotests/133 b/tests/qemu-iotests/133
index 4070fd9457..bc82d8ebd7 100755
--- a/tests/qemu-iotests/133
+++ b/tests/qemu-iotests/133
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
TEST_IMG="$TEST_IMG.base" _make_test_img 64M
_make_test_img -b "$TEST_IMG.base" -F $IMGFMT
diff --git a/tests/qemu-iotests/137 b/tests/qemu-iotests/137
index 7ae86892f7..de555a91c9 100755
--- a/tests/qemu-iotests/137
+++ b/tests/qemu-iotests/137
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# We are going to use lazy-refcounts
_unsupported_imgopts 'compat=0.10'
diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
index 1d5b0bed6d..e87a64eb89 100755
--- a/tests/qemu-iotests/138
+++ b/tests/qemu-iotests/138
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This tests qcow2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# With an external data file, data clusters are not refcounted
# (so qemu-img check would not do much);
diff --git a/tests/qemu-iotests/140 b/tests/qemu-iotests/140
index 309b177e77..ff6b904fa0 100755
--- a/tests/qemu-iotests/140
+++ b/tests/qemu-iotests/140
@@ -44,7 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt generic
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_make_test_img 64k
diff --git a/tests/qemu-iotests/154 b/tests/qemu-iotests/154
index 7f1c0d9bd9..34a1c051b6 100755
--- a/tests/qemu-iotests/154
+++ b/tests/qemu-iotests/154
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
CLUSTER_SIZE=4k
diff --git a/tests/qemu-iotests/161 b/tests/qemu-iotests/161
index bbf7dbbc5c..4fb7d0cbf0 100755
--- a/tests/qemu-iotests/161
+++ b/tests/qemu-iotests/161
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Any format implementing BlockDriver.bdrv_change_backing_file
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
IMG_SIZE=1M
diff --git a/tests/qemu-iotests/171 b/tests/qemu-iotests/171
index 341064a1c6..f3582edb10 100755
--- a/tests/qemu-iotests/171
+++ b/tests/qemu-iotests/171
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt raw
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175
index c3c2aed653..21a77a2bf5 100755
--- a/tests/qemu-iotests/175
+++ b/tests/qemu-iotests/175
@@ -71,7 +71,7 @@ EOF
. ./common.filter
_supported_fmt raw
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_default_cache_mode none
diff --git a/tests/qemu-iotests/176 b/tests/qemu-iotests/176
index 2565ff12ee..5ce3b27069 100755
--- a/tests/qemu-iotests/176
+++ b/tests/qemu-iotests/176
@@ -45,7 +45,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This test is specific to qcow2
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# Persistent dirty bitmaps require compat=1.1;
# Internal snapshots forbid using an external data file
diff --git a/tests/qemu-iotests/177 b/tests/qemu-iotests/177
index 5d4a77a6ab..595bfd4236 100755
--- a/tests/qemu-iotests/177
+++ b/tests/qemu-iotests/177
@@ -40,7 +40,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# tests specific to compat=1.1.
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
CLUSTER_SIZE=1M
size=128M
diff --git a/tests/qemu-iotests/179 b/tests/qemu-iotests/179
index 11a20cb1bf..7ada04c641 100755
--- a/tests/qemu-iotests/179
+++ b/tests/qemu-iotests/179
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# v2 images can't mark clusters as zero
diff --git a/tests/qemu-iotests/183 b/tests/qemu-iotests/183
index acdbefa310..d889a3b19c 100755
--- a/tests/qemu-iotests/183
+++ b/tests/qemu-iotests/183
@@ -44,7 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
_supported_os Linux FreeBSD NetBSD
_supported_fmt qcow2 raw qed quorum
-_supported_proto file
+_supported_proto file fuse
size=64M
_make_test_img $size
diff --git a/tests/qemu-iotests/186 b/tests/qemu-iotests/186
index 3ea0442d44..0db25b0e68 100755
--- a/tests/qemu-iotests/186
+++ b/tests/qemu-iotests/186
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_require_drivers null-co
if [ "$QEMU_DEFAULT_MACHINE" != "pc" ]; then
diff --git a/tests/qemu-iotests/187 b/tests/qemu-iotests/187
index c6e1dc57a0..f262d83e3a 100755
--- a/tests/qemu-iotests/187
+++ b/tests/qemu-iotests/187
@@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
size=64M
_make_test_img $size
diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
index d17462e1e4..95a891350d 100755
--- a/tests/qemu-iotests/191
+++ b/tests/qemu-iotests/191
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# An external data file would change the query-named-block-nodes output
_unsupported_imgopts data_file
diff --git a/tests/qemu-iotests/195 b/tests/qemu-iotests/195
index 2351d55fe1..967af5b7b5 100755
--- a/tests/qemu-iotests/195
+++ b/tests/qemu-iotests/195
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
do_run_qemu()
{
diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
index a7aabbd032..046539154f 100755
--- a/tests/qemu-iotests/200
+++ b/tests/qemu-iotests/200
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
BACKING_IMG="$TEST_IMG.base"
diff --git a/tests/qemu-iotests/204 b/tests/qemu-iotests/204
index 6770fa9b96..536bb8b534 100755
--- a/tests/qemu-iotests/204
+++ b/tests/qemu-iotests/204
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# This test assumes that discard leaves zero clusters; see test 177 for
# other tests that also work in older images
_unsupported_imgopts 'compat=0.10'
diff --git a/tests/qemu-iotests/214 b/tests/qemu-iotests/214
index af677d90b8..75ae7a14b5 100755
--- a/tests/qemu-iotests/214
+++ b/tests/qemu-iotests/214
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# Repairing the corrupted image requires qemu-img check to store a
# refcount up to 3, which requires at least two refcount bits.
diff --git a/tests/qemu-iotests/217 b/tests/qemu-iotests/217
index d89116ccad..7385342498 100755
--- a/tests/qemu-iotests/217
+++ b/tests/qemu-iotests/217
@@ -36,7 +36,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This test is specific to qcow2
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# This test needs clusters with at least a refcount of 2 so that
# OFLAG_COPIED is not set. refcount_bits=1 is therefore unsupported.
diff --git a/tests/qemu-iotests/220 b/tests/qemu-iotests/220
index a9259b7127..9ba3b3fdcb 100755
--- a/tests/qemu-iotests/220
+++ b/tests/qemu-iotests/220
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.pattern
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# To use a different refcount width but 16 bits we need compat=1.1,
# and external data files do not support compressed clusters.
diff --git a/tests/qemu-iotests/221 b/tests/qemu-iotests/221
index 0e9096fec7..ca62b3baa1 100755
--- a/tests/qemu-iotests/221
+++ b/tests/qemu-iotests/221
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt raw
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_default_cache_mode writeback
diff --git a/tests/qemu-iotests/229 b/tests/qemu-iotests/229
index 5f759fa587..273ac2472d 100755
--- a/tests/qemu-iotests/229
+++ b/tests/qemu-iotests/229
@@ -44,7 +44,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Needs backing file and backing format support
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# blkdebug can only inject errors on bs->file, so external data files
# do not work with this test
diff --git a/tests/qemu-iotests/247 b/tests/qemu-iotests/247
index 87e37b39e2..6cf2679750 100755
--- a/tests/qemu-iotests/247
+++ b/tests/qemu-iotests/247
@@ -41,7 +41,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Requires backing files and .bdrv_change_backing_file support
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
size=128M
diff --git a/tests/qemu-iotests/249 b/tests/qemu-iotests/249
index a9aa9303eb..29453b8c90 100755
--- a/tests/qemu-iotests/249
+++ b/tests/qemu-iotests/249
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Any format implementing BlockDriver.bdrv_change_backing_file
_supported_fmt qcow2 qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
IMG_SIZE=1M
diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
index 9bb6b94d74..3df275c76b 100755
--- a/tests/qemu-iotests/250
+++ b/tests/qemu-iotests/250
@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# This test does not make much sense with external data files
_unsupported_imgopts data_file
diff --git a/tests/qemu-iotests/252 b/tests/qemu-iotests/252
index 6662f4c9de..1d74afff99 100755
--- a/tests/qemu-iotests/252
+++ b/tests/qemu-iotests/252
@@ -42,7 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# zero cluster support
_supported_fmt qcow2
_unsupported_imgopts 'compat=0.10'
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
CLUSTER_SIZE=65536
diff --git a/tests/qemu-iotests/265 b/tests/qemu-iotests/265
index 00f2ec769e..0e800fb524 100755
--- a/tests/qemu-iotests/265
+++ b/tests/qemu-iotests/265
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# qcow2-specific test
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
echo '--- Writing to the image ---'
diff --git a/tests/qemu-iotests/268 b/tests/qemu-iotests/268
index 78c3f4db3a..ddf4312284 100755
--- a/tests/qemu-iotests/268
+++ b/tests/qemu-iotests/268
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
echo
echo "== Required alignment larger than cluster size =="
diff --git a/tests/qemu-iotests/272 b/tests/qemu-iotests/272
index c2f782d47b..de475bf6f0 100755
--- a/tests/qemu-iotests/272
+++ b/tests/qemu-iotests/272
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This is a qcow2 regression test
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# External data files do not support compression;
# We need an exact cluster size (2M) and refcount width (2) so we can
diff --git a/tests/qemu-iotests/273 b/tests/qemu-iotests/273
index 41ffbf39c4..79b4ab4b05 100755
--- a/tests/qemu-iotests/273
+++ b/tests/qemu-iotests/273
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# This is a qcow2 regression test
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# External data files would add nodes to the block graph, so it would
# not match the reference output
diff --git a/tests/qemu-iotests/279 b/tests/qemu-iotests/279
index 5515d4ed01..2a6315cf17 100755
--- a/tests/qemu-iotests/279
+++ b/tests/qemu-iotests/279
@@ -36,7 +36,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
# Backing files are required...
_supported_fmt qcow qcow2 vmdk qed
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_unsupported_imgopts "subformat=monolithicFlat" \
"subformat=twoGbMaxExtentFlat" \
diff --git a/tests/qemu-iotests/286 b/tests/qemu-iotests/286
index f14445ba4a..f64e0eccea 100755
--- a/tests/qemu-iotests/286
+++ b/tests/qemu-iotests/286
@@ -35,7 +35,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.qemu
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# Internal snapshots are (currently) impossible with refcount_bits=1,
# and generally impossible with external data files
_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
diff --git a/tests/qemu-iotests/287 b/tests/qemu-iotests/287
index 036cc09e82..3bb383fd4b 100755
--- a/tests/qemu-iotests/287
+++ b/tests/qemu-iotests/287
@@ -32,7 +32,7 @@ status=1 # failure is the default!
# This tests qocw2-specific low-level functionality
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_unsupported_imgopts 'compat=0.10' data_file
diff --git a/tests/qemu-iotests/289 b/tests/qemu-iotests/289
index 1c11d4030e..fe69bde1eb 100755
--- a/tests/qemu-iotests/289
+++ b/tests/qemu-iotests/289
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.pattern
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
# This is a v3-exclusive test;
# As for data_file, error paths often very much depend on whether
# there is an external data file or not; so we create one exactly when
diff --git a/tests/qemu-iotests/290 b/tests/qemu-iotests/290
index 01ee14dcfb..35c38d4f80 100755
--- a/tests/qemu-iotests/290
+++ b/tests/qemu-iotests/290
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_unsupported_imgopts 'compat=0.10' refcount_bits data_file
diff --git a/tests/qemu-iotests/291 b/tests/qemu-iotests/291
index ecef9eec62..bc70d5e389 100755
--- a/tests/qemu-iotests/291
+++ b/tests/qemu-iotests/291
@@ -36,7 +36,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.nbd
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_require_command QEMU_NBD
# compat=0.10 does not support bitmaps
diff --git a/tests/qemu-iotests/292 b/tests/qemu-iotests/292
index 83ab19231d..3ae2772e3b 100755
--- a/tests/qemu-iotests/292
+++ b/tests/qemu-iotests/292
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
# We need qemu-img map to show the file where the data is allocated,
# but with an external data file, it will show that instead of the
diff --git a/tests/qemu-iotests/293 b/tests/qemu-iotests/293
index f86fe3b413..3363bf07f0 100755
--- a/tests/qemu-iotests/293
+++ b/tests/qemu-iotests/293
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2 luks
-_supported_proto file #TODO
+_supported_proto file fuse #TODO
_require_working_luks
QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT
diff --git a/tests/qemu-iotests/294 b/tests/qemu-iotests/294
index 9c95ed8c9a..87da35db49 100755
--- a/tests/qemu-iotests/294
+++ b/tests/qemu-iotests/294
@@ -34,7 +34,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt luks
-_supported_proto file #TODO
+_supported_proto file fuse #TODO
QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT
diff --git a/tests/qemu-iotests/305 b/tests/qemu-iotests/305
index 768818af4a..5a415eb2a4 100755
--- a/tests/qemu-iotests/305
+++ b/tests/qemu-iotests/305
@@ -38,7 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
. ./common.filter
_supported_fmt qcow2
-_supported_proto file
+_supported_proto file fuse
_supported_os Linux
_unsupported_imgopts cluster_size refcount_bits extended_l2 compat=0.10 data_file
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 24/34] iotests/308: Add test for FUSE exports
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (22 preceding siblings ...)
2020-12-11 17:08 ` [PULL 23/34] iotests: Enable fuse for many tests Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 25/34] file-posix: check the use_lock before setting the file lock Kevin Wolf
` (10 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
We have good coverage of the normal I/O paths now, but what remains is a
test that tests some more special cases: Exporting an image on itself
(thus turning a formatted image into a raw one), some error cases, and
non-writable and non-growable exports.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201027190600.192171-21-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/308 | 339 +++++++++++++++++++++++++++++++++++++
tests/qemu-iotests/308.out | 97 +++++++++++
tests/qemu-iotests/group | 1 +
3 files changed, 437 insertions(+)
create mode 100755 tests/qemu-iotests/308
create mode 100644 tests/qemu-iotests/308.out
diff --git a/tests/qemu-iotests/308 b/tests/qemu-iotests/308
new file mode 100755
index 0000000000..b30f4400f6
--- /dev/null
+++ b/tests/qemu-iotests/308
@@ -0,0 +1,339 @@
+#!/usr/bin/env bash
+#
+# Test FUSE exports (in ways that are not captured by the generic
+# tests)
+#
+# Copyright (C) 2020 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+seq=$(basename "$0")
+echo "QA output created by $seq"
+
+status=1 # failure is the default!
+
+_cleanup()
+{
+ _cleanup_qemu
+ _cleanup_test_img
+ rmdir "$EXT_MP" 2>/dev/null
+ rm -f "$EXT_MP"
+ rm -f "$COPIED_IMG"
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+. ./common.qemu
+
+# Generic format, but needs a plain filename
+_supported_fmt generic
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
+ _unsupported_fmt $IMGFMT
+fi
+# We need the image to have exactly the specified size, and VPC does
+# not allow that by default
+_unsupported_fmt vpc
+
+_supported_proto file # We create the FUSE export manually
+_supported_os Linux # We need /dev/urandom
+
+# $1: Export ID
+# $2: Options (beyond the node-name and ID)
+# $3: Expected return value (defaults to 'return')
+# $4: Node to export (defaults to 'node-format')
+fuse_export_add()
+{
+ _send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'block-export-add',
+ 'arguments': {
+ 'type': 'fuse',
+ 'id': '$1',
+ 'node-name': '${4:-node-format}',
+ $2
+ } }" \
+ "${3:-return}" \
+ | _filter_imgfmt
+}
+
+# $1: Export ID
+fuse_export_del()
+{
+ _send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'block-export-del',
+ 'arguments': {
+ 'id': '$1'
+ } }" \
+ 'return'
+
+ _send_qemu_cmd $QEMU_HANDLE \
+ '' \
+ 'BLOCK_EXPORT_DELETED'
+}
+
+# Return the length of the protocol file
+# $1: Protocol node export mount point
+# $2: Original file (to compare)
+get_proto_len()
+{
+ len1=$(stat -c '%s' "$1")
+ len2=$(stat -c '%s' "$2")
+
+ if [ "$len1" != "$len2" ]; then
+ echo 'ERROR: Length of export and original differ:' >&2
+ echo "$len1 != $len2" >&2
+ else
+ echo '(OK: Lengths of export and original are the same)' >&2
+ fi
+
+ echo "$len1"
+}
+
+COPIED_IMG="$TEST_IMG.copy"
+EXT_MP="$TEST_IMG.fuse"
+
+echo '=== Set up ==='
+
+# Create image with random data
+_make_test_img 64M
+$QEMU_IO -c 'write -s /dev/urandom 0 64M' "$TEST_IMG" | _filter_qemu_io
+
+_launch_qemu
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'qmp_capabilities'}" \
+ 'return'
+
+# Separate blockdev-add calls for format and protocol so we can remove
+# the format layer later on
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'blockdev-add',
+ 'arguments': {
+ 'driver': 'file',
+ 'node-name': 'node-protocol',
+ 'filename': '$TEST_IMG'
+ } }" \
+ 'return'
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'blockdev-add',
+ 'arguments': {
+ 'driver': '$IMGFMT',
+ 'node-name': 'node-format',
+ 'file': 'node-protocol'
+ } }" \
+ 'return'
+
+echo
+echo '=== Mountpoint not present ==='
+
+rmdir "$EXT_MP" 2>/dev/null
+rm -f "$EXT_MP"
+output=$(fuse_export_add 'export-err' "'mountpoint': '$EXT_MP'" error)
+
+if echo "$output" | grep -q "Invalid parameter 'fuse'"; then
+ _notrun 'No FUSE support'
+fi
+
+echo "$output"
+
+echo
+echo '=== Mountpoint is a directory ==='
+
+mkdir "$EXT_MP"
+fuse_export_add 'export-err' "'mountpoint': '$EXT_MP'" error
+rmdir "$EXT_MP"
+
+echo
+echo '=== Mountpoint is a regular file ==='
+
+touch "$EXT_MP"
+fuse_export_add 'export-mp' "'mountpoint': '$EXT_MP'"
+
+# Check that the export presents the same data as the original image
+$QEMU_IMG compare -f raw -F $IMGFMT -U "$EXT_MP" "$TEST_IMG"
+
+echo
+echo '=== Mount over existing file ==='
+
+# This is the coolest feature of FUSE exports: You can transparently
+# make images in any format appear as raw images
+fuse_export_add 'export-img' "'mountpoint': '$TEST_IMG'"
+
+# Accesses both exports at the same time, so we get a concurrency test
+$QEMU_IMG compare -f raw -F raw -U "$EXT_MP" "$TEST_IMG"
+
+# Just to be sure, we later want to compare the data offline. Also,
+# this allows us to see that cp works without complaining.
+# (This is not a given, because cp will expect a short read at EOF.
+# Internally, qemu does not allow short reads, so we have to check
+# whether the FUSE export driver lets them work.)
+cp "$TEST_IMG" "$COPIED_IMG"
+
+# $TEST_IMG will be in mode 0400 because it is read-only; we are going
+# to write to the copy, so make it writable
+chmod 0600 "$COPIED_IMG"
+
+echo
+echo '=== Double export ==='
+
+# We have already seen that exporting a node twice works fine, but you
+# cannot export anything twice on the same mount point. The reason is
+# that qemu has to stat the given mount point, and this would have to
+# be answered by the same qemu instance if it already has an export
+# there. However, it cannot answer the stat because it is itself
+# caught up in that same stat.
+fuse_export_add 'export-err' "'mountpoint': '$EXT_MP'" error
+
+echo
+echo '=== Remove export ==='
+
+# Double-check that $EXT_MP appears as a non-empty file (the raw image)
+$QEMU_IMG info -f raw "$EXT_MP" | grep 'virtual size'
+
+fuse_export_del 'export-mp'
+
+# See that the file appears empty again
+$QEMU_IMG info -f raw "$EXT_MP" | grep 'virtual size'
+
+echo
+echo '=== Writable export ==='
+
+fuse_export_add 'export-mp' "'mountpoint': '$EXT_MP', 'writable': true"
+
+# Check that writing to the read-only export fails
+$QEMU_IO -f raw -c 'write -P 42 1M 64k' "$TEST_IMG" | _filter_qemu_io
+
+# But here it should work
+$QEMU_IO -f raw -c 'write -P 42 1M 64k' "$EXT_MP" | _filter_qemu_io
+
+# (Adjust the copy, too)
+$QEMU_IO -f raw -c 'write -P 42 1M 64k' "$COPIED_IMG" | _filter_qemu_io
+
+echo
+echo '=== Resizing exports ==='
+
+# Here, we need to export the protocol node -- the format layer may
+# not be growable, simply because the format does not support it.
+
+# Remove all exports and the format node first so permissions will not
+# get in the way
+fuse_export_del 'export-mp'
+fuse_export_del 'export-img'
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'blockdev-del',
+ 'arguments': {
+ 'node-name': 'node-format'
+ } }" \
+ 'return'
+
+# Now export the protocol node
+fuse_export_add \
+ 'export-mp' \
+ "'mountpoint': '$EXT_MP', 'writable': true" \
+ 'return' \
+ 'node-protocol'
+
+echo
+echo '--- Try growing non-growable export ---'
+
+# Get the current size so we can write beyond the EOF
+orig_len=$(get_proto_len "$EXT_MP" "$TEST_IMG")
+orig_disk_usage=$(stat -c '%b' "$TEST_IMG")
+
+# Should fail (exports are non-growable by default)
+# (Note that qemu-io can never write beyond the EOF, so we have to use
+# dd here)
+dd if=/dev/zero of="$EXT_MP" bs=1 count=64k seek=$orig_len 2>&1 \
+ | _filter_testdir | _filter_imgfmt
+
+echo
+echo '--- Resize export ---'
+
+# But we can truncate it explicitly; even with fallocate
+fallocate -o "$orig_len" -l 64k "$EXT_MP"
+
+new_len=$(get_proto_len "$EXT_MP" "$TEST_IMG")
+if [ "$new_len" != "$((orig_len + 65536))" ]; then
+ echo 'ERROR: Unexpected post-truncate image size:'
+ echo "$new_len != $((orig_len + 65536))"
+else
+ echo 'OK: Post-truncate image size is as expected'
+fi
+
+new_disk_usage=$(stat -c '%b' "$TEST_IMG")
+if [ "$new_disk_usage" -gt "$orig_disk_usage" ]; then
+ echo 'OK: Disk usage grew with fallocate'
+else
+ echo 'ERROR: Disk usage did not grow despite fallocate:'
+ echo "$orig_disk_usage => $new_disk_usage"
+fi
+
+echo
+echo '--- Try growing growable export ---'
+
+# Now export as growable
+fuse_export_del 'export-mp'
+fuse_export_add \
+ 'export-mp' \
+ "'mountpoint': '$EXT_MP', 'writable': true, 'growable': true" \
+ 'return' \
+ 'node-protocol'
+
+# Now we should be able to write beyond the EOF
+dd if=/dev/zero of="$EXT_MP" bs=1 count=64k seek=$new_len 2>&1 \
+ | _filter_testdir | _filter_imgfmt
+
+new_len=$(get_proto_len "$EXT_MP" "$TEST_IMG")
+if [ "$new_len" != "$((orig_len + 131072))" ]; then
+ echo 'ERROR: Unexpected post-grow image size:'
+ echo "$new_len != $((orig_len + 131072))"
+else
+ echo 'OK: Post-grow image size is as expected'
+fi
+
+echo
+echo '--- Shrink export ---'
+
+# Now go back to the original size
+truncate -s "$orig_len" "$EXT_MP"
+
+new_len=$(get_proto_len "$EXT_MP" "$TEST_IMG")
+if [ "$new_len" != "$orig_len" ]; then
+ echo 'ERROR: Unexpected post-truncate image size:'
+ echo "$new_len != $orig_len"
+else
+ echo 'OK: Post-truncate image size is as expected'
+fi
+
+echo
+echo '=== Tear down ==='
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'quit'}" \
+ 'return'
+
+wait=yes _cleanup_qemu
+
+echo
+echo '=== Compare copy with original ==='
+
+$QEMU_IMG compare -f raw -F $IMGFMT "$COPIED_IMG" "$TEST_IMG"
+
+# success, all done
+echo "*** done"
+rm -f $seq.full
+status=0
diff --git a/tests/qemu-iotests/308.out b/tests/qemu-iotests/308.out
new file mode 100644
index 0000000000..b93aceed2e
--- /dev/null
+++ b/tests/qemu-iotests/308.out
@@ -0,0 +1,97 @@
+QA output created by 308
+=== Set up ===
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
+wrote 67108864/67108864 bytes at offset 0
+64 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+{'execute': 'qmp_capabilities'}
+{"return": {}}
+{'execute': 'blockdev-add', 'arguments': { 'driver': 'file', 'node-name': 'node-protocol', 'filename': 'TEST_DIR/t.IMGFMT' } }
+{"return": {}}
+{'execute': 'blockdev-add', 'arguments': { 'driver': 'IMGFMT', 'node-name': 'node-format', 'file': 'node-protocol' } }
+{"return": {}}
+
+=== Mountpoint not present ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-err', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse' } }
+{"error": {"class": "GenericError", "desc": "Failed to stat 'TEST_DIR/t.IMGFMT.fuse': No such file or directory"}}
+
+=== Mountpoint is a directory ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-err', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse' } }
+{"error": {"class": "GenericError", "desc": "'TEST_DIR/t.IMGFMT.fuse' is not a regular file"}}
+
+=== Mountpoint is a regular file ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-mp', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse' } }
+{"return": {}}
+Images are identical.
+
+=== Mount over existing file ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-img', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT' } }
+{"return": {}}
+Images are identical.
+
+=== Double export ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-err', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse' } }
+{"error": {"class": "GenericError", "desc": "There already is a FUSE export on 'TEST_DIR/t.IMGFMT.fuse'"}}
+
+=== Remove export ===
+virtual size: 64 MiB (67108864 bytes)
+{'execute': 'block-export-del', 'arguments': { 'id': 'export-mp' } }
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "export-mp"}}
+virtual size: 0 B (0 bytes)
+
+=== Writable export ===
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-mp', 'node-name': 'node-format', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse', 'writable': true } }
+{"return": {}}
+write failed: Permission denied
+wrote 65536/65536 bytes at offset 1048576
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 65536/65536 bytes at offset 1048576
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Resizing exports ===
+{'execute': 'block-export-del', 'arguments': { 'id': 'export-mp' } }
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "export-mp"}}
+{'execute': 'block-export-del', 'arguments': { 'id': 'export-img' } }
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "export-img"}}
+{'execute': 'blockdev-del', 'arguments': { 'node-name': 'node-format' } }
+{"return": {}}
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-mp', 'node-name': 'node-protocol', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse', 'writable': true } }
+{"return": {}}
+
+--- Try growing non-growable export ---
+(OK: Lengths of export and original are the same)
+dd: error writing 'TEST_DIR/t.IMGFMT.fuse': Input/output error
+1+0 records in
+0+0 records out
+
+--- Resize export ---
+(OK: Lengths of export and original are the same)
+OK: Post-truncate image size is as expected
+OK: Disk usage grew with fallocate
+
+--- Try growing growable export ---
+{'execute': 'block-export-del', 'arguments': { 'id': 'export-mp' } }
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "export-mp"}}
+{'execute': 'block-export-add', 'arguments': { 'type': 'fuse', 'id': 'export-mp', 'node-name': 'node-protocol', 'mountpoint': 'TEST_DIR/t.IMGFMT.fuse', 'writable': true, 'growable': true } }
+{"return": {}}
+65536+0 records in
+65536+0 records out
+(OK: Lengths of export and original are the same)
+OK: Post-grow image size is as expected
+
+--- Shrink export ---
+(OK: Lengths of export and original are the same)
+OK: Post-truncate image size is as expected
+
+=== Tear down ===
+{'execute': 'quit'}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "SHUTDOWN", "data": {"guest": false, "reason": "host-qmp-quit"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "export-mp"}}
+
+=== Compare copy with original ===
+Images are identical.
+*** done
diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
index 2960dff728..9a8394b4cd 100644
--- a/tests/qemu-iotests/group
+++ b/tests/qemu-iotests/group
@@ -315,4 +315,5 @@
304 rw quick
305 rw quick
307 rw quick export
+308 rw
309 rw auto quick
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 25/34] file-posix: check the use_lock before setting the file lock
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (23 preceding siblings ...)
2020-12-11 17:08 ` [PULL 24/34] iotests/308: Add test for FUSE exports Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 26/34] iotests/221: Discard image before qemu-img map Kevin Wolf
` (9 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Li Feng <fengli@smartx.com>
The scenario is that when accessing a volume on an NFS filesystem
without supporting the file lock, Qemu will complain "Failed to lock
byte 100", even when setting the file.locking = off.
We should do file lock related operations only when the file.locking is
enabled, otherwise, the syscall of 'fcntl' will return non-zero.
Signed-off-by: Li Feng <fengli@smartx.com>
Message-Id: <1607341446-85506-1-git-send-email-fengli@smartx.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/file-posix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index d5fd1dbcd2..806764f7e3 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3104,7 +3104,7 @@ static int raw_check_perm(BlockDriverState *bs, uint64_t perm, uint64_t shared,
}
/* Copy locks to the new fd */
- if (s->perm_change_fd) {
+ if (s->perm_change_fd && s->use_lock) {
ret = raw_apply_lock_bytes(NULL, s->perm_change_fd, perm, ~shared,
false, errp);
if (ret < 0) {
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 26/34] iotests/221: Discard image before qemu-img map
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (24 preceding siblings ...)
2020-12-11 17:08 ` [PULL 25/34] file-posix: check the use_lock before setting the file lock Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 27/34] can-host: Fix crash when 'canbus' property is not set Kevin Wolf
` (8 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Max Reitz <mreitz@redhat.com>
See the new comment for why this should be done.
I do not have a reproducer on master, but when using FUSE block exports,
this test breaks depending on the underlying filesystem (for me, it
works on tmpfs, but fails on xfs, because the block allocated by
file-posix has 16 kB there instead of 4 kB).
Suggested-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20201207152245.66987-1-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
tests/qemu-iotests/221 | 7 +++++++
tests/qemu-iotests/221.out | 14 ++++++--------
2 files changed, 13 insertions(+), 8 deletions(-)
diff --git a/tests/qemu-iotests/221 b/tests/qemu-iotests/221
index ca62b3baa1..7e6086b205 100755
--- a/tests/qemu-iotests/221
+++ b/tests/qemu-iotests/221
@@ -46,6 +46,13 @@ echo "=== Check mapping of unaligned raw image ==="
echo
_make_test_img 65537 # qemu-img create rounds size up
+
+# file-posix allocates the first block of any images when it is created;
+# the size of this block depends on the host page size and the file
+# system block size, none of which are constant. Discard the whole
+# image so we will not see this allocation in qemu-img map's output.
+$QEMU_IO -c 'discard 0 65537' "$TEST_IMG" | _filter_qemu_io
+
$QEMU_IMG map --output=json "$TEST_IMG" | _filter_qemu_img_map
truncate --size=65537 "$TEST_IMG" # so we resize it and check again
diff --git a/tests/qemu-iotests/221.out b/tests/qemu-iotests/221.out
index dca024a0c3..93846c7dab 100644
--- a/tests/qemu-iotests/221.out
+++ b/tests/qemu-iotests/221.out
@@ -3,18 +3,16 @@ QA output created by 221
=== Check mapping of unaligned raw image ===
Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=65537
-[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
-{ "start": 4096, "length": 61952, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
-[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
-{ "start": 4096, "length": 61952, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
+discard 65537/65537 bytes at offset 0
+64.001 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+[{ "start": 0, "length": 66048, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
+[{ "start": 0, "length": 66048, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
wrote 1/1 bytes at offset 65536
1 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
-{ "start": 4096, "length": 61440, "depth": 0, "zero": true, "data": false, "offset": OFFSET},
+[{ "start": 0, "length": 65536, "depth": 0, "zero": true, "data": false, "offset": OFFSET},
{ "start": 65536, "length": 1, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
{ "start": 65537, "length": 511, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
-[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
-{ "start": 4096, "length": 61440, "depth": 0, "zero": true, "data": false, "offset": OFFSET},
+[{ "start": 0, "length": 65536, "depth": 0, "zero": true, "data": false, "offset": OFFSET},
{ "start": 65536, "length": 1, "depth": 0, "zero": false, "data": true, "offset": OFFSET},
{ "start": 65537, "length": 511, "depth": 0, "zero": true, "data": false, "offset": OFFSET}]
*** done
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 27/34] can-host: Fix crash when 'canbus' property is not set
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (25 preceding siblings ...)
2020-12-11 17:08 ` [PULL 26/34] iotests/221: Discard image before qemu-img map Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 28/34] block/file-posix: fix workaround in raw_do_pwrite_zeroes() Kevin Wolf
` (7 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
Providing the 'if' property, but not 'canbus' segfaults like this:
#0 0x0000555555b0f14d in can_bus_insert_client (bus=0x0, client=0x555556aa9af0) at ../net/can/can_core.c:88
#1 0x00005555559c3803 in can_host_connect (ch=0x555556aa9ac0, errp=0x7fffffffd568) at ../net/can/can_host.c:62
#2 0x00005555559c386a in can_host_complete (uc=0x555556aa9ac0, errp=0x7fffffffd568) at ../net/can/can_host.c:72
#3 0x0000555555d52de9 in user_creatable_complete (uc=0x555556aa9ac0, errp=0x7fffffffd5c8) at ../qom/object_interfaces.c:23
Add the missing NULL check.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201130105615.21799-5-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
net/can/can_host.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/net/can/can_host.c b/net/can/can_host.c
index be4547d913..ba7f163d0a 100644
--- a/net/can/can_host.c
+++ b/net/can/can_host.c
@@ -53,6 +53,11 @@ static void can_host_connect(CanHostState *ch, Error **errp)
CanHostClass *chc = CAN_HOST_GET_CLASS(ch);
Error *local_err = NULL;
+ if (ch->bus == NULL) {
+ error_setg(errp, "'canbus' property not set");
+ return;
+ }
+
chc->connect(ch, &local_err);
if (local_err) {
error_propagate(errp, local_err);
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 28/34] block/file-posix: fix workaround in raw_do_pwrite_zeroes()
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (26 preceding siblings ...)
2020-12-11 17:08 ` [PULL 27/34] can-host: Fix crash when 'canbus' property is not set Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 29/34] block/io: bdrv_refresh_limits(): use ERRP_GUARD Kevin Wolf
` (6 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
We should not set overlap_bytes:
1. Don't worry: it is calculated by bdrv_mark_request_serialising() and
will be equal to or greater than bytes anyway.
2. If the request was already aligned up to some greater alignment,
than we may break things: we reduce overlap_bytes, and further
bdrv_mark_request_serialising() may not help, as it will not restore
old bigger alignment.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20201203222713.13507-2-vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/file-posix.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 806764f7e3..9bee3d88d0 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2952,7 +2952,6 @@ raw_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes,
end = INT64_MAX & -(uint64_t)bs->bl.request_alignment;
req->bytes = end - req->offset;
- req->overlap_bytes = req->bytes;
bdrv_mark_request_serialising(req, bs->bl.request_alignment);
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 29/34] block/io: bdrv_refresh_limits(): use ERRP_GUARD
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (27 preceding siblings ...)
2020-12-11 17:08 ` [PULL 28/34] block/file-posix: fix workaround in raw_do_pwrite_zeroes() Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 30/34] block/io: bdrv_check_byte_request(): drop bdrv_is_inserted() Kevin Wolf
` (5 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
This simplifies following commit.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20201203222713.13507-3-vsementsov@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/io.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/block/io.c b/block/io.c
index ec5e152bb7..3e91074c9f 100644
--- a/block/io.c
+++ b/block/io.c
@@ -135,10 +135,10 @@ static void bdrv_merge_limits(BlockLimits *dst, const BlockLimits *src)
void bdrv_refresh_limits(BlockDriverState *bs, Error **errp)
{
+ ERRP_GUARD();
BlockDriver *drv = bs->drv;
BdrvChild *c;
bool have_limits;
- Error *local_err = NULL;
memset(&bs->bl, 0, sizeof(bs->bl));
@@ -156,9 +156,8 @@ void bdrv_refresh_limits(BlockDriverState *bs, Error **errp)
QLIST_FOREACH(c, &bs->children, next) {
if (c->role & (BDRV_CHILD_DATA | BDRV_CHILD_FILTERED | BDRV_CHILD_COW))
{
- bdrv_refresh_limits(c->bs, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ bdrv_refresh_limits(c->bs, errp);
+ if (*errp) {
return;
}
bdrv_merge_limits(&bs->bl, &c->bs->bl);
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 30/34] block/io: bdrv_check_byte_request(): drop bdrv_is_inserted()
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (28 preceding siblings ...)
2020-12-11 17:08 ` [PULL 29/34] block/io: bdrv_refresh_limits(): use ERRP_GUARD Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 31/34] block: introduce BDRV_MAX_LENGTH Kevin Wolf
` (4 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Move bdrv_is_inserted() calls into callers.
We are going to make bdrv_check_byte_request() a clean thing.
bdrv_is_inserted() is not about checking the request, it's about
checking the bs. So, it should be separate.
With this patch we probably change error path for some failure
scenarios. But depending on the fact that querying too big request on
empty cdrom (or corrupted qcow2 node with no drv) will result in EIO
and not ENOMEDIUM would be very strange. More over, we are going to
move to 64bit requests, so larger requests will be allowed anyway.
More over, keeping in mind that cdrom is the only driver that has
.bdrv_is_inserted() handler it's strange that we should care so much
about it in generic block layer, intuitively we should just do read and
write, and cdrom driver should return correct errors if it is not
inserted. But it's a work for another series.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20201203222713.13507-4-vsementsov@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/io.c | 25 ++++++++++++-------------
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/block/io.c b/block/io.c
index 3e91074c9f..ef75a5abb4 100644
--- a/block/io.c
+++ b/block/io.c
@@ -884,17 +884,12 @@ static bool coroutine_fn bdrv_wait_serialising_requests(BdrvTrackedRequest *self
return waited;
}
-static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
- size_t size)
+static int bdrv_check_byte_request(int64_t offset, size_t size)
{
if (size > BDRV_REQUEST_MAX_BYTES) {
return -EIO;
}
- if (!bdrv_is_inserted(bs)) {
- return -ENOMEDIUM;
- }
-
if (offset < 0) {
return -EIO;
}
@@ -1642,7 +1637,11 @@ int coroutine_fn bdrv_co_preadv_part(BdrvChild *child,
trace_bdrv_co_preadv(bs, offset, bytes, flags);
- ret = bdrv_check_byte_request(bs, offset, bytes);
+ if (!bdrv_is_inserted(bs)) {
+ return -ENOMEDIUM;
+ }
+
+ ret = bdrv_check_byte_request(offset, bytes);
if (ret < 0) {
return ret;
}
@@ -2054,11 +2053,11 @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *child,
trace_bdrv_co_pwritev(child->bs, offset, bytes, flags);
- if (!bs->drv) {
+ if (!bdrv_is_inserted(bs)) {
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(bs, offset, bytes);
+ ret = bdrv_check_byte_request(offset, bytes);
if (ret < 0) {
return ret;
}
@@ -3045,10 +3044,10 @@ static int coroutine_fn bdrv_co_copy_range_internal(
assert(!(read_flags & BDRV_REQ_NO_FALLBACK));
assert(!(write_flags & BDRV_REQ_NO_FALLBACK));
- if (!dst || !dst->bs) {
+ if (!dst || !dst->bs || !bdrv_is_inserted(dst->bs)) {
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(dst->bs, dst_offset, bytes);
+ ret = bdrv_check_byte_request(dst_offset, bytes);
if (ret) {
return ret;
}
@@ -3056,10 +3055,10 @@ static int coroutine_fn bdrv_co_copy_range_internal(
return bdrv_co_pwrite_zeroes(dst, dst_offset, bytes, write_flags);
}
- if (!src || !src->bs) {
+ if (!src || !src->bs || !bdrv_is_inserted(src->bs)) {
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(src->bs, src_offset, bytes);
+ ret = bdrv_check_byte_request(src_offset, bytes);
if (ret) {
return ret;
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 31/34] block: introduce BDRV_MAX_LENGTH
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (29 preceding siblings ...)
2020-12-11 17:08 ` [PULL 30/34] block/io: bdrv_check_byte_request(): drop bdrv_is_inserted() Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 32/34] block: Simplify qmp_block_resize() error paths Kevin Wolf
` (3 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
We are going to modify block layer to work with 64bit requests. And
first step is moving to int64_t type for both offset and bytes
arguments in all block request related functions.
It's mostly safe (when widening signed or unsigned int to int64_t), but
switching from uint64_t is questionable.
So, let's first establish the set of requests we want to work with.
First signed int64_t should be enough, as off_t is signed anyway. Then,
obviously offset + bytes should not overflow.
And most interesting: (offset + bytes) being aligned up should not
overflow as well. Aligned to what alignment? First thing that comes in
mind is bs->bl.request_alignment, as we align up request to this
alignment. But there is another thing: look at
bdrv_mark_request_serialising(). It aligns request up to some given
alignment. And this parameter may be bdrv_get_cluster_size(), which is
often a lot greater than bs->bl.request_alignment.
Note also, that bdrv_mark_request_serialising() uses signed int64_t for
calculations. So, actually, we already depend on some restrictions.
Happily, bdrv_get_cluster_size() returns int and
bs->bl.request_alignment has 32bit unsigned type, but defined to be a
power of 2 less than INT_MAX. So, we may establish, that INT_MAX is
absolute maximum for any kind of alignment that may occur with the
request.
Note, that bdrv_get_cluster_size() is not documented to return power
of 2, still bdrv_mark_request_serialising() behaves like it is.
Also, backup uses bdi.cluster_size and is not prepared to it not being
power of 2.
So, let's establish that Qemu supports only power-of-2 clusters and
alignments.
So, alignment can't be greater than 2^30.
Finally to be safe with calculations, to not calculate different
maximums for different nodes (depending on cluster size and
request_alignment), let's simply set QEMU_ALIGN_DOWN(INT64_MAX, 2^30)
as absolute maximum bytes length for Qemu. Actually, it's not much less
than INT64_MAX.
OK, then, let's apply it to block/io.
Let's consider all block/io entry points of offset/bytes:
4 bytes/offset interface functions: bdrv_co_preadv_part(),
bdrv_co_pwritev_part(), bdrv_co_copy_range_internal() and
bdrv_co_pdiscard() and we check them all with bdrv_check_request().
We also have one entry point with only offset: bdrv_co_truncate().
Check the offset.
And one public structure: BdrvTrackedRequest. Happily, it has only
three external users:
file-posix.c: adopted by this patch
write-threshold.c: only read fields
test-write-threshold.c: sets obviously small constant values
Better is to make the structure private and add corresponding
interfaces.. Still it's not obvious what kind of interface is needed
for file-posix.c. Let's keep it public but add corresponding
assertions.
After this patch we'll convert functions in block/io.c to int64_t bytes
and offset parameters. We can assume that offset/bytes pair always
satisfy new restrictions, and make
corresponding assertions where needed. If we reach some offset/bytes
point in block/io.c missing bdrv_check_request() it is considered a
bug. As well, if block/io.c modifies a offset/bytes request, expanding
it more then aligning up to request_alignment, it's a bug too.
For all io requests except for discard we keep for now old restriction
of 32bit request length.
iotest 206 output error message changed, as now test disk size is
larger than new limit. Add one more test case with new maximum disk
size to cover too-big-L1 case.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20201203222713.13507-5-vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
include/block/block.h | 10 +++++++
include/block/block_int.h | 8 ++++++
block.c | 17 +++++++++++-
block/file-posix.c | 6 ++---
block/io.c | 51 +++++++++++++++++++++++++++++-------
tests/test-write-threshold.c | 4 +++
tests/qemu-iotests/206 | 2 +-
tests/qemu-iotests/206.out | 6 +++++
8 files changed, 90 insertions(+), 14 deletions(-)
diff --git a/include/block/block.h b/include/block/block.h
index c9d7c58765..5b81e33e94 100644
--- a/include/block/block.h
+++ b/include/block/block.h
@@ -142,6 +142,16 @@ typedef struct HDGeometry {
INT_MAX >> BDRV_SECTOR_BITS)
#define BDRV_REQUEST_MAX_BYTES (BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS)
+/*
+ * We want allow aligning requests and disk length up to any 32bit alignment
+ * and don't afraid of overflow.
+ * To achieve it, and in the same time use some pretty number as maximum disk
+ * size, let's define maximum "length" (a limit for any offset/bytes request and
+ * for disk size) to be the greatest power of 2 less than INT64_MAX.
+ */
+#define BDRV_MAX_ALIGNMENT (1L << 30)
+#define BDRV_MAX_LENGTH (QEMU_ALIGN_DOWN(INT64_MAX, BDRV_MAX_ALIGNMENT))
+
/*
* Allocation status flags for bdrv_block_status() and friends.
*
diff --git a/include/block/block_int.h b/include/block/block_int.h
index 95d9333be1..1eeafc118c 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -70,6 +70,12 @@ enum BdrvTrackedRequestType {
BDRV_TRACKED_TRUNCATE,
};
+/*
+ * That is not quite good that BdrvTrackedRequest structure is public,
+ * as block/io.c is very careful about incoming offset/bytes being
+ * correct. Be sure to assert bdrv_check_request() succeeded after any
+ * modification of BdrvTrackedRequest object out of block/io.c
+ */
typedef struct BdrvTrackedRequest {
BlockDriverState *bs;
int64_t offset;
@@ -87,6 +93,8 @@ typedef struct BdrvTrackedRequest {
struct BdrvTrackedRequest *waiting_for;
} BdrvTrackedRequest;
+int bdrv_check_request(int64_t offset, int64_t bytes);
+
struct BlockDriver {
const char *format_name;
int instance_size;
diff --git a/block.c b/block.c
index eb16fb48c6..8f177504d4 100644
--- a/block.c
+++ b/block.c
@@ -962,6 +962,11 @@ int refresh_total_sectors(BlockDriverState *bs, int64_t hint)
}
bs->total_sectors = hint;
+
+ if (bs->total_sectors * BDRV_SECTOR_SIZE > BDRV_MAX_LENGTH) {
+ return -EFBIG;
+ }
+
return 0;
}
@@ -5535,6 +5540,7 @@ void bdrv_get_backing_filename(BlockDriverState *bs,
int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
{
+ int ret;
BlockDriver *drv = bs->drv;
/* if bs->drv == NULL, bs is closed, so there's nothing to do here */
if (!drv) {
@@ -5548,7 +5554,16 @@ int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
return -ENOTSUP;
}
memset(bdi, 0, sizeof(*bdi));
- return drv->bdrv_get_info(bs, bdi);
+ ret = drv->bdrv_get_info(bs, bdi);
+ if (ret < 0) {
+ return ret;
+ }
+
+ if (bdi->cluster_size > BDRV_MAX_ALIGNMENT) {
+ return -EINVAL;
+ }
+
+ return 0;
}
ImageInfoSpecific *bdrv_get_specific_info(BlockDriverState *bs,
diff --git a/block/file-posix.c b/block/file-posix.c
index 9bee3d88d0..83e2cc5530 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2926,7 +2926,6 @@ raw_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes,
#ifdef CONFIG_FALLOCATE
if (offset + bytes > bs->total_sectors * BDRV_SECTOR_SIZE) {
BdrvTrackedRequest *req;
- uint64_t end;
/*
* This is a workaround for a bug in the Linux XFS driver,
@@ -2950,8 +2949,9 @@ raw_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes,
assert(req->offset <= offset);
assert(req->offset + req->bytes >= offset + bytes);
- end = INT64_MAX & -(uint64_t)bs->bl.request_alignment;
- req->bytes = end - req->offset;
+ req->bytes = BDRV_MAX_LENGTH - req->offset;
+
+ assert(bdrv_check_request(req->offset, req->bytes) == 0);
bdrv_mark_request_serialising(req, bs->bl.request_alignment);
}
diff --git a/block/io.c b/block/io.c
index ef75a5abb4..6343d85476 100644
--- a/block/io.c
+++ b/block/io.c
@@ -176,6 +176,13 @@ void bdrv_refresh_limits(BlockDriverState *bs, Error **errp)
/* Then let the driver override it */
if (drv->bdrv_refresh_limits) {
drv->bdrv_refresh_limits(bs, errp);
+ if (*errp) {
+ return;
+ }
+ }
+
+ if (bs->bl.request_alignment > BDRV_MAX_ALIGNMENT) {
+ error_setg(errp, "Driver requires too large request alignment");
}
}
@@ -884,13 +891,31 @@ static bool coroutine_fn bdrv_wait_serialising_requests(BdrvTrackedRequest *self
return waited;
}
-static int bdrv_check_byte_request(int64_t offset, size_t size)
+int bdrv_check_request(int64_t offset, int64_t bytes)
{
- if (size > BDRV_REQUEST_MAX_BYTES) {
+ if (offset < 0 || bytes < 0) {
return -EIO;
}
- if (offset < 0) {
+ if (bytes > BDRV_MAX_LENGTH) {
+ return -EIO;
+ }
+
+ if (offset > BDRV_MAX_LENGTH - bytes) {
+ return -EIO;
+ }
+
+ return 0;
+}
+
+static int bdrv_check_request32(int64_t offset, int64_t bytes)
+{
+ int ret = bdrv_check_request(offset, bytes);
+ if (ret < 0) {
+ return ret;
+ }
+
+ if (bytes > BDRV_REQUEST_MAX_BYTES) {
return -EIO;
}
@@ -1641,7 +1666,7 @@ int coroutine_fn bdrv_co_preadv_part(BdrvChild *child,
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(offset, bytes);
+ ret = bdrv_check_request32(offset, bytes);
if (ret < 0) {
return ret;
}
@@ -2057,7 +2082,7 @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *child,
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(offset, bytes);
+ ret = bdrv_check_request32(offset, bytes);
if (ret < 0) {
return ret;
}
@@ -2787,8 +2812,9 @@ int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset,
return -EPERM;
}
- if (offset < 0 || bytes < 0 || bytes > INT64_MAX - offset) {
- return -EIO;
+ ret = bdrv_check_request(offset, bytes);
+ if (ret < 0) {
+ return ret;
}
/* Do nothing if disabled. */
@@ -3047,7 +3073,7 @@ static int coroutine_fn bdrv_co_copy_range_internal(
if (!dst || !dst->bs || !bdrv_is_inserted(dst->bs)) {
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(dst_offset, bytes);
+ ret = bdrv_check_request32(dst_offset, bytes);
if (ret) {
return ret;
}
@@ -3058,7 +3084,7 @@ static int coroutine_fn bdrv_co_copy_range_internal(
if (!src || !src->bs || !bdrv_is_inserted(src->bs)) {
return -ENOMEDIUM;
}
- ret = bdrv_check_byte_request(src_offset, bytes);
+ ret = bdrv_check_request32(src_offset, bytes);
if (ret) {
return ret;
}
@@ -3188,6 +3214,13 @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,
return -EINVAL;
}
+ ret = bdrv_check_request(offset, 0);
+ if (ret < 0) {
+ error_setg(errp, "Required too big image size, it must be not greater "
+ "than %" PRId64, BDRV_MAX_LENGTH);
+ return ret;
+ }
+
old_size = bdrv_getlength(bs);
if (old_size < 0) {
error_setg_errno(errp, -old_size, "Failed to get old image size");
diff --git a/tests/test-write-threshold.c b/tests/test-write-threshold.c
index 97ca12f710..4cf032652d 100644
--- a/tests/test-write-threshold.c
+++ b/tests/test-write-threshold.c
@@ -64,6 +64,8 @@ static void test_threshold_not_trigger(void)
req.offset = 1024;
req.bytes = 1024;
+ assert(bdrv_check_request(req.offset, req.bytes) == 0);
+
bdrv_write_threshold_set(&bs, threshold);
amount = bdrv_write_threshold_exceeded(&bs, &req);
g_assert_cmpuint(amount, ==, 0);
@@ -82,6 +84,8 @@ static void test_threshold_trigger(void)
req.offset = (4 * 1024 * 1024) - 1024;
req.bytes = 2 * 1024;
+ assert(bdrv_check_request(req.offset, req.bytes) == 0);
+
bdrv_write_threshold_set(&bs, threshold);
amount = bdrv_write_threshold_exceeded(&bs, &req);
g_assert_cmpuint(amount, >=, 1024);
diff --git a/tests/qemu-iotests/206 b/tests/qemu-iotests/206
index 0a3ee5ef00..d12d7cb566 100755
--- a/tests/qemu-iotests/206
+++ b/tests/qemu-iotests/206
@@ -202,7 +202,7 @@ with iotests.FilePath('t.qcow2') as disk_path, \
vm.launch()
for size in [ 1234, 18446744073709551104, 9223372036854775808,
- 9223372036854775296 ]:
+ 9223372036854775296, 9223372035781033984 ]:
vm.blockdev_create({ 'driver': imgfmt,
'file': 'node0',
'size': size })
diff --git a/tests/qemu-iotests/206.out b/tests/qemu-iotests/206.out
index a100849fcb..e8a36de00b 100644
--- a/tests/qemu-iotests/206.out
+++ b/tests/qemu-iotests/206.out
@@ -180,6 +180,12 @@ Job failed: Could not resize image: Image size cannot be negative
{"execute": "blockdev-create", "arguments": {"job-id": "job0", "options": {"driver": "qcow2", "file": "node0", "size": 9223372036854775296}}}
{"return": {}}
+Job failed: Could not resize image: Required too big image size, it must be not greater than 9223372035781033984
+{"execute": "job-dismiss", "arguments": {"id": "job0"}}
+{"return": {}}
+
+{"execute": "blockdev-create", "arguments": {"job-id": "job0", "options": {"driver": "qcow2", "file": "node0", "size": 9223372035781033984}}}
+{"return": {}}
Job failed: Could not resize image: Failed to grow the L1 table: File too large
{"execute": "job-dismiss", "arguments": {"id": "job0"}}
{"return": {}}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 32/34] block: Simplify qmp_block_resize() error paths
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (30 preceding siblings ...)
2020-12-11 17:08 ` [PULL 31/34] block: introduce BDRV_MAX_LENGTH Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 33/34] block: Fix locking in qmp_block_resize() Kevin Wolf
` (2 subsequent siblings)
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
The only thing that happens after the 'out:' label is blk_unref(blk).
However, blk = NULL in all of the error cases, so instead of jumping to
'out:', we can just return directly.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201203172311.68232-2-kwolf@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
blockdev.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/blockdev.c b/blockdev.c
index 6c7be7c522..660c735c81 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2454,7 +2454,7 @@ void coroutine_fn qmp_block_resize(bool has_device, const char *device,
int64_t size, Error **errp)
{
Error *local_err = NULL;
- BlockBackend *blk = NULL;
+ BlockBackend *blk;
BlockDriverState *bs;
AioContext *old_ctx;
@@ -2468,17 +2468,17 @@ void coroutine_fn qmp_block_resize(bool has_device, const char *device,
if (size < 0) {
error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "size", "a >0 size");
- goto out;
+ return;
}
if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_RESIZE, NULL)) {
error_setg(errp, QERR_DEVICE_IN_USE, device);
- goto out;
+ return;
}
blk = blk_new_with_bs(bs, BLK_PERM_RESIZE, BLK_PERM_ALL, errp);
if (!blk) {
- goto out;
+ return;
}
bdrv_drained_begin(bs);
@@ -2487,7 +2487,6 @@ void coroutine_fn qmp_block_resize(bool has_device, const char *device,
bdrv_co_leave(bs, old_ctx);
bdrv_drained_end(bs);
-out:
bdrv_co_lock(bs);
blk_unref(blk);
bdrv_co_unlock(bs);
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 33/34] block: Fix locking in qmp_block_resize()
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (31 preceding siblings ...)
2020-12-11 17:08 ` [PULL 32/34] block: Simplify qmp_block_resize() error paths Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-11 17:08 ` [PULL 34/34] block: Fix deadlock in bdrv_co_yield_to_drain() Kevin Wolf
2020-12-12 16:06 ` [PULL 00/34] Block layer patches Peter Maydell
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
The drain functions assume that we hold the AioContext lock of the
drained block node. Make sure to actually take the lock.
Cc: qemu-stable@nongnu.org
Fixes: eb94b81a94bce112e6b206df846c1551aaf6cab6
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201203172311.68232-3-kwolf@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
blockdev.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/blockdev.c b/blockdev.c
index 660c735c81..412354b4b6 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2481,13 +2481,16 @@ void coroutine_fn qmp_block_resize(bool has_device, const char *device,
return;
}
+ bdrv_co_lock(bs);
bdrv_drained_begin(bs);
+ bdrv_co_unlock(bs);
+
old_ctx = bdrv_co_enter(bs);
blk_truncate(blk, size, false, PREALLOC_MODE_OFF, 0, errp);
bdrv_co_leave(bs, old_ctx);
- bdrv_drained_end(bs);
bdrv_co_lock(bs);
+ bdrv_drained_end(bs);
blk_unref(blk);
bdrv_co_unlock(bs);
}
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PULL 34/34] block: Fix deadlock in bdrv_co_yield_to_drain()
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (32 preceding siblings ...)
2020-12-11 17:08 ` [PULL 33/34] block: Fix locking in qmp_block_resize() Kevin Wolf
@ 2020-12-11 17:08 ` Kevin Wolf
2020-12-12 16:06 ` [PULL 00/34] Block layer patches Peter Maydell
34 siblings, 0 replies; 36+ messages in thread
From: Kevin Wolf @ 2020-12-11 17:08 UTC (permalink / raw)
To: qemu-block; +Cc: kwolf, peter.maydell, qemu-devel
If bdrv_co_yield_to_drain() is called for draining a block node that
runs in a different AioContext, it keeps that AioContext locked while it
yields and schedules a BH in the AioContext to do the actual drain.
As long as executing the BH is the very next thing that the event loop
of the node's AioContext does, this actually happens to work, but when
it tries to execute something else that wants to take the AioContext
lock, it will deadlock. (In the bug report, this other thing is a
virtio-scsi device running virtio_scsi_data_plane_handle_cmd().)
Instead, always drop the AioContext lock across the yield and reacquire
it only when the coroutine is reentered. The BH needs to unconditionally
take the lock for itself now.
This fixes the 'block_resize' QMP command on a block node that runs in
an iothread.
Cc: qemu-stable@nongnu.org
Fixes: eb94b81a94bce112e6b206df846c1551aaf6cab6
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1903511
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20201203172311.68232-4-kwolf@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block/io.c | 41 ++++++++++++++++++++++++-----------------
1 file changed, 24 insertions(+), 17 deletions(-)
diff --git a/block/io.c b/block/io.c
index 6343d85476..24205f5168 100644
--- a/block/io.c
+++ b/block/io.c
@@ -312,17 +312,7 @@ static void bdrv_co_drain_bh_cb(void *opaque)
if (bs) {
AioContext *ctx = bdrv_get_aio_context(bs);
- AioContext *co_ctx = qemu_coroutine_get_aio_context(co);
-
- /*
- * When the coroutine yielded, the lock for its home context was
- * released, so we need to re-acquire it here. If it explicitly
- * acquired a different context, the lock is still held and we don't
- * want to lock it a second time (or AIO_WAIT_WHILE() would hang).
- */
- if (ctx == co_ctx) {
- aio_context_acquire(ctx);
- }
+ aio_context_acquire(ctx);
bdrv_dec_in_flight(bs);
if (data->begin) {
assert(!data->drained_end_counter);
@@ -334,9 +324,7 @@ static void bdrv_co_drain_bh_cb(void *opaque)
data->ignore_bds_parents,
data->drained_end_counter);
}
- if (ctx == co_ctx) {
- aio_context_release(ctx);
- }
+ aio_context_release(ctx);
} else {
assert(data->begin);
bdrv_drain_all_begin();
@@ -354,13 +342,16 @@ static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs,
int *drained_end_counter)
{
BdrvCoDrainData data;
+ Coroutine *self = qemu_coroutine_self();
+ AioContext *ctx = bdrv_get_aio_context(bs);
+ AioContext *co_ctx = qemu_coroutine_get_aio_context(self);
/* Calling bdrv_drain() from a BH ensures the current coroutine yields and
* other coroutines run if they were queued by aio_co_enter(). */
assert(qemu_in_coroutine());
data = (BdrvCoDrainData) {
- .co = qemu_coroutine_self(),
+ .co = self,
.bs = bs,
.done = false,
.begin = begin,
@@ -374,13 +365,29 @@ static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs,
if (bs) {
bdrv_inc_in_flight(bs);
}
- replay_bh_schedule_oneshot_event(bdrv_get_aio_context(bs),
- bdrv_co_drain_bh_cb, &data);
+
+ /*
+ * Temporarily drop the lock across yield or we would get deadlocks.
+ * bdrv_co_drain_bh_cb() reaquires the lock as needed.
+ *
+ * When we yield below, the lock for the current context will be
+ * released, so if this is actually the lock that protects bs, don't drop
+ * it a second time.
+ */
+ if (ctx != co_ctx) {
+ aio_context_release(ctx);
+ }
+ replay_bh_schedule_oneshot_event(ctx, bdrv_co_drain_bh_cb, &data);
qemu_coroutine_yield();
/* If we are resumed from some other event (such as an aio completion or a
* timer callback), it is a bug in the caller that should be fixed. */
assert(data.done);
+
+ /* Reaquire the AioContext of bs if we dropped it */
+ if (ctx != co_ctx) {
+ aio_context_acquire(ctx);
+ }
}
void bdrv_do_drained_begin_quiesce(BlockDriverState *bs,
--
2.29.2
^ permalink raw reply related [flat|nested] 36+ messages in thread* Re: [PULL 00/34] Block layer patches
2020-12-11 17:07 [PULL 00/34] Block layer patches Kevin Wolf
` (33 preceding siblings ...)
2020-12-11 17:08 ` [PULL 34/34] block: Fix deadlock in bdrv_co_yield_to_drain() Kevin Wolf
@ 2020-12-12 16:06 ` Peter Maydell
34 siblings, 0 replies; 36+ messages in thread
From: Peter Maydell @ 2020-12-12 16:06 UTC (permalink / raw)
To: Kevin Wolf; +Cc: QEMU Developers, Qemu-block
On Fri, 11 Dec 2020 at 17:08, Kevin Wolf <kwolf@redhat.com> wrote:
>
> The following changes since commit b785d25e91718a660546a6550f64b3c543af7754:
>
> Merge remote-tracking branch 'remotes/bonzini-gitlab/tags/for-upstream' into staging (2020-12-11 13:50:35 +0000)
>
> are available in the Git repository at:
>
> git://repo.or.cz/qemu/kevin.git tags/for-upstream
>
> for you to fetch changes up to 960d5fb3e8ee09bc5f1a5c84f66dce42a6cef920:
>
> block: Fix deadlock in bdrv_co_yield_to_drain() (2020-12-11 17:52:40 +0100)
>
> ----------------------------------------------------------------
> Block layer patches:
>
> - Support for FUSE exports
> - Fix deadlock in bdrv_co_yield_to_drain()
> - Use lock guard macros
> - Some preparational patches for 64 bit block layer
> - file-posix: Fix request extension to INT64_MAX in raw_do_pwrite_zeroes()
>
Applied, thanks.
Please update the changelog at https://wiki.qemu.org/ChangeLog/6.0
for any user-visible changes.
-- PMM
^ permalink raw reply [flat|nested] 36+ messages in thread