rpms/qemu-kvm
7
0
Fork 0
Code Issues Pull Requests Releases Activity

* Mon Aug 15 2022 Miroslav Rezanina <mrezanin@redhat.com> - 7.0.0-11

Browse Source 
- kvm-QIOChannelSocket-Fix-zero-copy-flush-returning-code-.patch [bz#2107466]
- kvm-Add-dirty-sync-missed-zero-copy-migration-stat.patch [bz#2107466]
- kvm-migration-multifd-Report-to-user-when-zerocopy-not-w.patch [bz#2107466]
- kvm-migration-Avoid-false-positive-on-non-supported-scen.patch [bz#2107466]
- kvm-migration-add-remaining-params-has_-true-in-migratio.patch [bz#2107466]
- kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch [bz#2107466]
- kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch [bz#2112303]
- kvm-vdpa-Fix-bad-index-calculus-at-vhost_vdpa_get_vring_.patch [bz#2116876]
- kvm-vdpa-Fix-index-calculus-at-vhost_vdpa_svqs_start.patch [bz#2116876]
- kvm-vdpa-Fix-memory-listener-deletions-of-iova-tree.patch [bz#2116876]
- kvm-vdpa-Fix-file-descriptor-leak-on-get-features-error.patch [bz#2116876]
- Resolves: bz#2107466
  (zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together)
- Resolves: bz#2112303
  (virtio-blk: Can't boot fresh installation from used 512 cluster_size image under certain conditions)
- Resolves: bz#2116876
  (Fixes for vDPA control virtqueue support in Qemu)
This commit is contained in:
Miroslav Rezanina 2022-08-15 07:19:39 -04:00
parent 23d471091d
commit 85d5f0ed1b
12 changed files with 765 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
kvm-Add-dirty-sync-missed-zero-copy-migration-stat.patch Normal file
View File

@ -0,0 +1,87 @@
From ac346634c5731407baa9de709dbd4d5cc6f45301 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 11 Jul 2022 18:11:12 -0300
Subject: [PATCH 02/11] Add dirty-sync-missed-zero-copy migration stat
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [2/6] 115035fd0a4e4b9439c91fb0f5d1a2f9244ba369 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220711211112.18951-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit cf20c897338067ab4b70a4596fdccaf90c7e29a1)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 2 ++
monitor/hmp-cmds.c | 5 +++++
qapi/migration.json | 7 ++++++-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index 8fb3eae910..3a3a7a4a50 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1017,6 +1017,8 @@ static void populate_ram_info(MigrationInfo *info, MigrationState *s)
info->ram->normal_bytes = ram_counters.normal * page_size;
info->ram->mbps = s->mbps;
info->ram->dirty_sync_count = ram_counters.dirty_sync_count;
+ info->ram->dirty_sync_missed_zero_copy =
+ ram_counters.dirty_sync_missed_zero_copy;
info->ram->postcopy_requests = ram_counters.postcopy_requests;
info->ram->page_size = page_size;
info->ram->multifd_bytes = ram_counters.multifd_bytes;
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index 634968498b..9cec01de38 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -305,6 +305,11 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
monitor_printf(mon, "postcopy ram: %" PRIu64 " kbytes\n",
info->ram->postcopy_bytes >> 10);
}
+ if (info->ram->dirty_sync_missed_zero_copy) {
+ monitor_printf(mon,
+ "Zero-copy-send fallbacks happened: %" PRIu64 " times\n",
+ info->ram->dirty_sync_missed_zero_copy);
+ }
}
if (info->has_disk) {
diff --git a/qapi/migration.json b/qapi/migration.json
index 5105790cd0..9b38b3c21c 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -55,6 +55,10 @@
# @postcopy-bytes: The number of bytes sent during the post-copy phase
# (since 7.0).
#
+# @dirty-sync-missed-zero-copy: Number of times dirty RAM synchronization could
+# not avoid copying dirty pages. This is between
+# 0 and @dirty-sync-count * @multifd-channels.
+# (since 7.1)
# Since: 0.14
##
{ 'struct': 'MigrationStats',
@@ -65,7 +69,8 @@
'postcopy-requests' : 'int', 'page-size' : 'int',
'multifd-bytes' : 'uint64', 'pages-per-second' : 'uint64',
'precopy-bytes' : 'uint64', 'downtime-bytes' : 'uint64',
- 'postcopy-bytes' : 'uint64' } }
+ 'postcopy-bytes' : 'uint64',
+ 'dirty-sync-missed-zero-copy' : 'uint64' } }
##
# @XBZRLECacheStats:
--
2.31.1

56
kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch Normal file
View File

@ -0,0 +1,56 @@
From cb6dc39a5e5d2d981b4b1e983042b3fbb529d5d1 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Thu, 4 Aug 2022 04:10:43 -0300
Subject: [PATCH 06/11] QIOChannelSocket: Add support for MSG_ZEROCOPY + IPV6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [6/6] 2eb1aba8ebf267a6f67cfba2e489dc88619c7fd4 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
For using MSG_ZEROCOPY, there are two steps:
1 - io_writev() the packet, which enqueues the packet for sending, and
2 - io_flush(), which gets confirmation that all packets got correctly sent
Currently, if MSG_ZEROCOPY is used to send packets over IPV6, no error will
be reported in (1), but it will fail in the first time (2) happens.
This happens because (2) currently checks for cmsg_level & cmsg_type
associated with IPV4 only, before reporting any error.
Add checks for cmsg_level & cmsg_type associated with IPV6, and thus enable
support for MSG_ZEROCOPY + IPV6
Fixes: 2bc58ffc29 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 5258a7e2c0677d16e9e1d06845f60171adf0b290)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index eb7baa2184..efd5f60808 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -747,8 +747,8 @@ static int qio_channel_socket_flush(QIOChannel *ioc,
}
cm = CMSG_FIRSTHDR(&msg);
- if (cm->cmsg_level != SOL_IP &&
- cm->cmsg_type != IP_RECVERR) {
+ if (cm->cmsg_level != SOL_IP && cm->cmsg_type != IP_RECVERR &&
+ cm->cmsg_level != SOL_IPV6 && cm->cmsg_type != IPV6_RECVERR) {
error_setg_errno(errp, EPROTOTYPE,
"Wrong cmsg in errqueue");
return -1;
--
2.31.1

65
kvm-QIOChannelSocket-Fix-zero-copy-flush-returning-code-.patch Normal file
View File

@ -0,0 +1,65 @@
From 678981c6bb7c964e1591f6f8aba49e9602f64852 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 11 Jul 2022 18:11:11 -0300
Subject: [PATCH 01/11] QIOChannelSocket: Fix zero-copy flush returning code 1
when nothing sent
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [1/6] cebc887cb61de1572d8ae3232cde45e80c339404 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
If flush is called when no buffer was sent with MSG_ZEROCOPY, it currently
returns 1. This return code should be used only when Linux fails to use
MSG_ZEROCOPY on a lot of sendmsg().
Fix this by returning early from flush if no sendmsg(...,MSG_ZEROCOPY)
was attempted.
Fixes: 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220711211112.18951-2-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 927f93e099c4f9184e60a1bc61624ac2d04d0223)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 8ae8b212cf..eb7baa2184 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -717,12 +717,18 @@ static int qio_channel_socket_flush(QIOChannel *ioc,
struct cmsghdr *cm;
char control[CMSG_SPACE(sizeof(*serr))];
int received;
- int ret = 1;
+ int ret;
+
+ if (sioc->zero_copy_queued == sioc->zero_copy_sent) {
+ return 0;
+ }
msg.msg_control = control;
msg.msg_controllen = sizeof(control);
memset(control, 0, sizeof(control));
+ ret = 1;
+
while (sioc->zero_copy_sent < sioc->zero_copy_queued) {
received = recvmsg(sioc->fd, &msg, MSG_ERRQUEUE);
if (received < 0) {
--
2.31.1

93
kvm-migration-Avoid-false-positive-on-non-supported-scen.patch Normal file
View File

@ -0,0 +1,93 @@
From 0753565af588dfa78b3529e359b1590e15fcbdb3 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Tue, 19 Jul 2022 09:23:45 -0300
Subject: [PATCH 04/11] migration: Avoid false-positive on non-supported
scenarios for zero-copy-send
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [4/6] f5c7ed6710d92668acb81d0118a71fab0b4e3d43 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Migration with zero-copy-send currently has it's limitations, as it can't
be used with TLS nor any kind of compression. In such scenarios, it should
output errors during parameter / capability setting.
But currently there are some ways of setting this not-supported scenarios
without printing the error message:
!) For 'compression' capability, it works by enabling it together with
zero-copy-send. This happens because the validity test for zero-copy uses
the helper unction migrate_use_compression(), which check for compression
presence in s->enabled_capabilities[MIGRATION_CAPABILITY_COMPRESS].
The point here is: the validity test happens before the capability gets
enabled. If all of them get enabled together, this test will not return
error.
In order to fix that, replace migrate_use_compression() by directly testing
the cap_list parameter migrate_caps_check().
2) For features enabled by parameters such as TLS & 'multifd_compression',
there was also a possibility of setting non-supported scenarios: setting
zero-copy-send first, then setting the unsupported parameter.
In order to fix that, also add a check for parameters conflicting with
zero-copy-send on migrate_params_check().
3) XBZRLE is also a compression capability, so it makes sense to also add
it to the list of capabilities which are not supported with zero-copy-send.
Fixes: 1abaec9a1b2c ("migration: Change zero_copy_send from migration parameter to migration capability")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220719122345.253713-1-leobras@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 90eb69e4f1a16b388d0483543bf6bfc69a9966e4)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index 3a3a7a4a50..343629d59c 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1265,7 +1265,9 @@ static bool migrate_caps_check(bool *cap_list,
#ifdef CONFIG_LINUX
if (cap_list[MIGRATION_CAPABILITY_ZERO_COPY_SEND] &&
(!cap_list[MIGRATION_CAPABILITY_MULTIFD] ||
- migrate_use_compression() ||
+ cap_list[MIGRATION_CAPABILITY_COMPRESS] ||
+ cap_list[MIGRATION_CAPABILITY_XBZRLE] ||
+ migrate_multifd_compression() ||
migrate_use_tls())) {
error_setg(errp,
"Zero copy only available for non-compressed non-TLS multifd migration");
@@ -1502,6 +1504,17 @@ static bool migrate_params_check(MigrationParameters *params, Error **errp)
error_prepend(errp, "Invalid mapping given for block-bitmap-mapping: ");
return false;
}
+
+#ifdef CONFIG_LINUX
+ if (migrate_use_zero_copy_send() &&
+ ((params->has_multifd_compression && params->multifd_compression) ||
+ (params->has_tls_creds && params->tls_creds && *params->tls_creds))) {
+ error_setg(errp,
+ "Zero copy only available for non-compressed non-TLS multifd migration");
+ return false;
+ }
+#endif
+
return true;
}
--
2.31.1

62
kvm-migration-add-remaining-params-has_-true-in-migratio.patch Normal file
View File

@ -0,0 +1,62 @@
From 9698c0e8dd9b4f5dbc237a3f98ac46297dac85fb Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 25 Jul 2022 22:02:35 -0300
Subject: [PATCH 05/11] migration: add remaining params->has_* = true in
migration_instance_init()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [5/6] 50bbad254e2356b3ae16f6e00a3db8fd0b22dde9 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Some of params->has_* = true are missing in migration_instance_init, this
causes migrate_params_check() to skip some tests, allowing some
unsupported scenarios.
Fix this by adding all missing params->has_* = true in
migration_instance_init().
Fixes: 69ef1f36b0 ("migration: define 'tls-creds' and 'tls-hostname' migration parameters")
Fixes: 1d58872a91 ("migration: do not wait for free thread")
Fixes: d2f1d29b95 ("migration: add support for a "tls-authz" migration parameter")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220726010235.342927-1-leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit df67aa3e61e2c83459da7d815962d9706f1528fc)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/migration/migration.c b/migration/migration.c
index 343629d59c..5e78028df4 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -4332,6 +4332,7 @@ static void migration_instance_init(Object *obj)
/* Set has_* up only for parameter checks */
params->has_compress_level = true;
params->has_compress_threads = true;
+ params->has_compress_wait_thread = true;
params->has_decompress_threads = true;
params->has_throttle_trigger_threshold = true;
params->has_cpu_throttle_initial = true;
@@ -4352,6 +4353,9 @@ static void migration_instance_init(Object *obj)
params->has_announce_max = true;
params->has_announce_rounds = true;
params->has_announce_step = true;
+ params->has_tls_creds = true;
+ params->has_tls_hostname = true;
+ params->has_tls_authz = true;
qemu_sem_init(&ms->postcopy_pause_sem, 0);
qemu_sem_init(&ms->postcopy_pause_rp_sem, 0);
--
2.31.1

83
kvm-migration-multifd-Report-to-user-when-zerocopy-not-w.patch Normal file
View File

@ -0,0 +1,83 @@
From 78bbe28d5f5691330239041448cccfb339eed779 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 11 Jul 2022 18:11:13 -0300
Subject: [PATCH 03/11] migration/multifd: Report to user when zerocopy not
working
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 111: zero-copy-send fixes & improvements
RH-Commit: [3/6] 4f9165325b3cb8ff16d8b3b7649ff780fae0e2ad (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 2107466
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Some errors, like the lack of Scatter-Gather support by the network
interface(NETIF_F_SG) may cause sendmsg(...,MSG_ZEROCOPY) to fail on using
zero-copy, which causes it to fall back to the default copying mechanism.
After each full dirty-bitmap scan there should be a zero-copy flush
happening, which checks for errors each of the previous calls to
sendmsg(...,MSG_ZEROCOPY). If all of them failed to use zero-copy, then
increment dirty_sync_missed_zero_copy migration stat to let the user know
about it.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220711211112.18951-4-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit d59c40cc483729f2e67c80e58df769ad19976fe9)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/multifd.c | 2 ++
migration/ram.c | 5 +++++
migration/ram.h | 2 ++
3 files changed, 9 insertions(+)
diff --git a/migration/multifd.c b/migration/multifd.c
index 0b5b41c53f..96e5f0a058 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -626,6 +626,8 @@ int multifd_send_sync_main(QEMUFile *f)
if (ret < 0) {
error_report_err(err);
return -1;
+ } else if (ret == 1) {
+ dirty_sync_missed_zero_copy();
}
}
}
diff --git a/migration/ram.c b/migration/ram.c
index ee40e4a718..c437ff1b1f 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -406,6 +406,11 @@ static void ram_transferred_add(uint64_t bytes)
ram_counters.transferred += bytes;
}
+void dirty_sync_missed_zero_copy(void)
+{
+ ram_counters.dirty_sync_missed_zero_copy++;
+}
+
/* used by the search for pages to send */
struct PageSearchStatus {
/* Current block being searched */
diff --git a/migration/ram.h b/migration/ram.h
index 2c6dc3675d..34adf5cb92 100644
--- a/migration/ram.h
+++ b/migration/ram.h
@@ -86,4 +86,6 @@ void ram_write_tracking_prepare(void);
int ram_write_tracking_start(void);
void ram_write_tracking_stop(void);
+void dirty_sync_missed_zero_copy(void);
+
#endif
--
2.31.1

63
kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch Normal file
View File

@ -0,0 +1,63 @@
From 03996a8a826c9186e4a16e1b4757f1ef5947a503 Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Fri, 5 Aug 2022 11:42:14 +0200
Subject: [PATCH 07/11] pc-bios/s390-ccw: Fix booting with logical block size <
physical block size
RH-Author: Thomas Huth <thuth@redhat.com>
RH-MergeRequest: 113: pc-bios/s390-ccw: Fix booting with logical block size < physical block size
RH-Commit: [1/1] a45ff477bc7d7011ea6c4d42a1aade213d1e4690 (thuth/qemu-kvm-cs9)
RH-Bugzilla: 2112303
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Claudio Imbrenda <None>
For accessing single blocks during boot, it's the logical block size that
matters. (Physical block sizes are rather interesting e.g. for creating
file systems with the correct alignment for speed reasons etc.).
So the s390-ccw bios has to use the logical block size for calculating
sector numbers during the boot phase, the "physical_block_exp" shift
value must not be taken into account. This change fixes the boot process
when the guest hast been installed on a disk where the logical block size
differs from the physical one, e.g. if the guest has been installed
like this:
qemu-system-s390x -nographic -accel kvm -m 2G \
-drive if=none,id=d1,file=fedora.iso,format=raw,media=cdrom \
-device virtio-scsi -device scsi-cd,drive=d1 \
-drive if=none,id=d2,file=test.qcow2,format=qcow2
-device virtio-blk,drive=d2,physical_block_size=4096,logical_block_size=512
Linux correctly uses the logical block size of 512 for the installation,
but the s390-ccw bios tries to boot from a disk with 4096 block size so
far, as long as this patch has not been applied yet (well, it used to work
by accident in the past due to the virtio_assume_scsi() hack that used to
enforce 512 byte sectors on all virtio-block disks, but that hack has been
well removed in commit 5447de2619050a0a4d to fix other scenarios).
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2112303
Message-Id: <20220805094214.285223-1-thuth@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 393296de19650e1400ca265914cfdeb313725363)
---
pc-bios/s390-ccw/virtio-blkdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pc-bios/s390-ccw/virtio-blkdev.c b/pc-bios/s390-ccw/virtio-blkdev.c
index 8271c47296..794f99b42c 100644
--- a/pc-bios/s390-ccw/virtio-blkdev.c
+++ b/pc-bios/s390-ccw/virtio-blkdev.c
@@ -173,7 +173,7 @@ int virtio_get_block_size(void)
switch (vdev->senseid.cu_model) {
case VIRTIO_ID_BLOCK:
- return vdev->config.blk.blk_size << vdev->config.blk.physical_block_exp;
+ return vdev->config.blk.blk_size;
case VIRTIO_ID_SCSI:
return vdev->scsi_block_size;
}
--
2.31.1

50
kvm-vdpa-Fix-bad-index-calculus-at-vhost_vdpa_get_vring_.patch Normal file
View File

@ -0,0 +1,50 @@
From e19adb058502e24580dbc4f6f944cd951ca288ed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Thu, 12 May 2022 19:57:44 +0200
Subject: [PATCH 08/11] vdpa: Fix bad index calculus at
vhost_vdpa_get_vring_base
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Eugenio Pérez <eperezma@redhat.com>
RH-MergeRequest: 114: vdpa: Fix memory listener deletions of iova tree
RH-Commit: [1/4] 754fb8960684fa7a91bddb18c8df58c3b947ee75 (eperezmartin/qemu-kvm)
RH-Bugzilla: 2116876
RH-Acked-by: Jason Wang <jasowang@redhat.com>
RH-Acked-by: Cindy Lu <lulu@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
Fixes: 6d0b222666 ("vdpa: Adapt vhost_vdpa_get_vring_base to SVQ")
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20220512175747.142058-4-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 639036477ef890958415967e753ca2cbb348c16c)
---
hw/virtio/vhost-vdpa.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
index e3e5bce4bb..a7dfac530f 100644
--- a/hw/virtio/vhost-vdpa.c
+++ b/hw/virtio/vhost-vdpa.c
@@ -1193,11 +1193,11 @@ static int vhost_vdpa_get_vring_base(struct vhost_dev *dev,
struct vhost_vring_state *ring)
{
struct vhost_vdpa *v = dev->opaque;
+ int vdpa_idx = ring->index - dev->vq_index;
int ret;
if (v->shadow_vqs_enabled) {
- VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs,
- ring->index);
+ VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, vdpa_idx);
/*
* Setting base as last used idx, so destination will see as available
--
2.31.1

58
kvm-vdpa-Fix-file-descriptor-leak-on-get-features-error.patch Normal file
View File

@ -0,0 +1,58 @@
From 71857062b7aea29fc418e107244cf4083cd78cd7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Tue, 2 Aug 2022 13:24:46 +0200
Subject: [PATCH 11/11] vdpa: Fix file descriptor leak on get features error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Eugenio Pérez <eperezma@redhat.com>
RH-MergeRequest: 114: vdpa: Fix memory listener deletions of iova tree
RH-Commit: [4/4] bdfe6ed4539ecf68dc8bc4519755f9d5c096447d (eperezmartin/qemu-kvm)
RH-Bugzilla: 2116876
RH-Acked-by: Jason Wang <jasowang@redhat.com>
RH-Acked-by: Cindy Lu <lulu@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
File descriptor vdpa_device_fd is not free in the case of returning
error from vhost_vdpa_get_features. Fixing it by making all errors go to
the same error path.
Resolves: Coverity CID 1490785
Fixes: 8170ab3f43 ("vdpa: Extract get features part from vhost_vdpa_get_max_queue_pairs")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20220802112447.249436-2-eperezma@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
(cherry picked from commit aed5da45daf734ddc543c0791e877dac75e16f61)
---
net/vhost-vdpa.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 50672bcd66..411e71e6c2 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -566,7 +566,7 @@ int net_init_vhost_vdpa(const Netdev *netdev, const char *name,
g_autofree NetClientState **ncs = NULL;
g_autoptr(VhostIOVATree) iova_tree = NULL;
NetClientState *nc;
- int queue_pairs, r, i, has_cvq = 0;
+ int queue_pairs, r, i = 0, has_cvq = 0;
assert(netdev->type == NET_CLIENT_DRIVER_VHOST_VDPA);
opts = &netdev->u.vhost_vdpa;
@@ -582,7 +582,7 @@ int net_init_vhost_vdpa(const Netdev *netdev, const char *name,
r = vhost_vdpa_get_features(vdpa_device_fd, &features, errp);
if (unlikely(r < 0)) {
- return r;
+ goto err;
}
queue_pairs = vhost_vdpa_get_max_queue_pairs(vdpa_device_fd, features,
--
2.31.1

45
kvm-vdpa-Fix-index-calculus-at-vhost_vdpa_svqs_start.patch Normal file
View File

@ -0,0 +1,45 @@
From 6335431b70dd55c1d52152d726fa462db2e10eb8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Thu, 12 May 2022 19:57:45 +0200
Subject: [PATCH 09/11] vdpa: Fix index calculus at vhost_vdpa_svqs_start
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Eugenio Pérez <eperezma@redhat.com>
RH-MergeRequest: 114: vdpa: Fix memory listener deletions of iova tree
RH-Commit: [2/4] 9ce732e6bba426f8e00020ee6ad77f972f3e75b5 (eperezmartin/qemu-kvm)
RH-Bugzilla: 2116876
RH-Acked-by: Jason Wang <jasowang@redhat.com>
RH-Acked-by: Cindy Lu <lulu@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
With the introduction of MQ the index of the vq needs to be calculated
with the device model vq_index.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20220512175747.142058-5-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 1c82fdfef8a227518ffecae9d419bcada995c202)
---
hw/virtio/vhost-vdpa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
index a7dfac530f..f877b354fa 100644
--- a/hw/virtio/vhost-vdpa.c
+++ b/hw/virtio/vhost-vdpa.c
@@ -1032,7 +1032,7 @@ static bool vhost_vdpa_svqs_start(struct vhost_dev *dev)
VirtQueue *vq = virtio_get_queue(dev->vdev, dev->vq_index + i);
VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
struct vhost_vring_addr addr = {
- .index = i,
+ .index = dev->vq_index + i,
};
int r;
bool ok = vhost_vdpa_svq_setup(dev, svq, i, &err);
--
2.31.1

61
kvm-vdpa-Fix-memory-listener-deletions-of-iova-tree.patch Normal file
View File

@ -0,0 +1,61 @@
From b212edc97a471c75f8b8b44ee2a3a2cf82ef14d9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Fri, 22 Jul 2022 10:26:30 +0200
Subject: [PATCH 10/11] vdpa: Fix memory listener deletions of iova tree
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Eugenio Pérez <eperezma@redhat.com>
RH-MergeRequest: 114: vdpa: Fix memory listener deletions of iova tree
RH-Commit: [3/4] ad71f098b3fa8654962ac7872b5393c37c9825f2 (eperezmartin/qemu-kvm)
RH-Bugzilla: 2116876
RH-Acked-by: Jason Wang <jasowang@redhat.com>
RH-Acked-by: Cindy Lu <lulu@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
vhost_vdpa_listener_region_del is always deleting the first iova entry
of the tree, since it's using the needle iova instead of the result's
one.
This was detected using a vga virtual device in the VM using vdpa SVQ.
It makes some extra memory adding and deleting, so the wrong one was
mapped / unmapped. This was undetected before since all the memory was
mappend and unmapped totally without that device, but other conditions
could trigger it too:
* mem_region was with .iova = 0, .translated_addr = (correct GPA).
* iova_tree_find_iova returned right result, but does not update
mem_region.
* iova_tree_remove always removed region with .iova = 0. Right iova were
sent to the device.
* Next map will fill the first region with .iova = 0, causing a mapping
with the same iova and device complains, if the next action is a map.
* Next unmap will cause to try to unmap again iova = 0, causing the
device to complain that no region was mapped at iova = 0.
Fixes: 34e3c94edaef ("vdpa: Add custom IOTLB translations to SVQ")
Reported-by: Lei Yang <leiyang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit 75a8ce64f6e37513698857fb4284170da163ed06)
---
hw/virtio/vhost-vdpa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
index f877b354fa..03dc6014b0 100644
--- a/hw/virtio/vhost-vdpa.c
+++ b/hw/virtio/vhost-vdpa.c
@@ -288,7 +288,7 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
result = vhost_iova_tree_find_iova(v->iova_tree, &mem_region);
iova = result->iova;
- vhost_iova_tree_remove(v->iova_tree, &mem_region);
+ vhost_iova_tree_remove(v->iova_tree, result);
}
vhost_vdpa_iotlb_batch_begin_once(v);
ret = vhost_vdpa_dma_unmap(v, iova, int128_get64(llsize));
--
2.31.1

43
qemu-kvm.spec
View File

@ -151,7 +151,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}:%{version} \
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 7.0.0
Release: 10%{?rcrel}%{?dist}%{?cc_suffix}
Release: 11%{?rcrel}%{?dist}%{?cc_suffix}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
# Epoch 15 used for RHEL 8
# Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@ -422,6 +422,28 @@ Patch133: kvm-kvm-don-t-use-perror-without-useful-errno.patch
Patch134: kvm-multifd-Copy-pages-before-compressing-them-with-zlib.patch
# For bz#2099934 - Guest reboot on destination host after postcopy migration completed
Patch135: kvm-Revert-migration-Simplify-unqueue_page.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch136: kvm-QIOChannelSocket-Fix-zero-copy-flush-returning-code-.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch137: kvm-Add-dirty-sync-missed-zero-copy-migration-stat.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch138: kvm-migration-multifd-Report-to-user-when-zerocopy-not-w.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch139: kvm-migration-Avoid-false-positive-on-non-supported-scen.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch140: kvm-migration-add-remaining-params-has_-true-in-migratio.patch
# For bz#2107466 - zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together
Patch141: kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch
# For bz#2112303 - virtio-blk: Can't boot fresh installation from used 512 cluster_size image under certain conditions
Patch142: kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch
# For bz#2116876 - Fixes for vDPA control virtqueue support in Qemu
Patch143: kvm-vdpa-Fix-bad-index-calculus-at-vhost_vdpa_get_vring_.patch
# For bz#2116876 - Fixes for vDPA control virtqueue support in Qemu
Patch144: kvm-vdpa-Fix-index-calculus-at-vhost_vdpa_svqs_start.patch
# For bz#2116876 - Fixes for vDPA control virtqueue support in Qemu
Patch145: kvm-vdpa-Fix-memory-listener-deletions-of-iova-tree.patch
# For bz#2116876 - Fixes for vDPA control virtqueue support in Qemu
Patch146: kvm-vdpa-Fix-file-descriptor-leak-on-get-features-error.patch
# Source-git patches
@ -1457,6 +1479,25 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%endif
%changelog
* Mon Aug 15 2022 Miroslav Rezanina <mrezanin@redhat.com> - 7.0.0-11
- kvm-QIOChannelSocket-Fix-zero-copy-flush-returning-code-.patch [bz#2107466]
- kvm-Add-dirty-sync-missed-zero-copy-migration-stat.patch [bz#2107466]
- kvm-migration-multifd-Report-to-user-when-zerocopy-not-w.patch [bz#2107466]
- kvm-migration-Avoid-false-positive-on-non-supported-scen.patch [bz#2107466]
- kvm-migration-add-remaining-params-has_-true-in-migratio.patch [bz#2107466]
- kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch [bz#2107466]
- kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch [bz#2112303]
- kvm-vdpa-Fix-bad-index-calculus-at-vhost_vdpa_get_vring_.patch [bz#2116876]
- kvm-vdpa-Fix-index-calculus-at-vhost_vdpa_svqs_start.patch [bz#2116876]
- kvm-vdpa-Fix-memory-listener-deletions-of-iova-tree.patch [bz#2116876]
- kvm-vdpa-Fix-file-descriptor-leak-on-get-features-error.patch [bz#2116876]
- Resolves: bz#2107466
(zerocopy capability can be enabled when set migrate capabilities with multifd and compress/xbzrle together)
- Resolves: bz#2112303
(virtio-blk: Can't boot fresh installation from used 512 cluster_size image under certain conditions)
- Resolves: bz#2116876
(Fixes for vDPA control virtqueue support in Qemu)
* Mon Aug 08 2022 Miroslav Rezanina <mrezanin@redhat.com> - 7.0.0-10
- kvm-vhost-Track-descriptor-chain-in-private-at-SVQ.patch [bz#1939363]
- kvm-vhost-Fix-device-s-used-descriptor-dequeue.patch [bz#1939363]