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

* Tue Jul 05 2022 Camilla Conte <cconte@redhat.com> - 7.0.0-8

Browse Source 
- kvm-tests-avocado-update-aarch64_virt-test-to-exercise-c.patch [bz#2060839]
- kvm-RHEL-only-tests-avocado-Switch-aarch64-tests-from-a5.patch [bz#2060839]
- kvm-RHEL-only-AArch64-Drop-unsupported-CPU-types.patch [bz#2060839]
- kvm-target-i386-deprecate-CPUs-older-than-x86_64-v2-ABI.patch [bz#2060839]
- kvm-target-s390x-deprecate-CPUs-older-than-z14.patch [bz#2060839]
- kvm-target-arm-deprecate-named-CPU-models.patch [bz#2060839]
- kvm-meson.build-Fix-docker-test-build-alpine-when-includ.patch [bz#1968509]
- kvm-QIOChannel-Add-flags-on-io_writev-and-introduce-io_f.patch [bz#1968509]
- kvm-QIOChannelSocket-Implement-io_writev-zero-copy-flag-.patch [bz#1968509]
- kvm-migration-Add-zero-copy-send-parameter-for-QMP-HMP-f.patch [bz#1968509]
- kvm-migration-Add-migrate_use_tls-helper.patch [bz#1968509]
- kvm-multifd-multifd_send_sync_main-now-returns-negative-.patch [bz#1968509]
- kvm-multifd-Send-header-packet-without-flags-if-zero-cop.patch [bz#1968509]
- kvm-multifd-Implement-zero-copy-write-in-multifd-migrati.patch [bz#1968509]
- kvm-QIOChannelSocket-Introduce-assert-and-reduce-ifdefs-.patch [bz#1968509]
- kvm-QIOChannelSocket-Fix-zero-copy-send-so-socket-flush-.patch [bz#1968509]
- kvm-migration-Change-zero_copy_send-from-migration-param.patch [bz#1968509]
- kvm-migration-Allow-migrate-recover-to-run-multiple-time.patch [bz#2096143]
- Resolves: bz#2060839
  (Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9)
- Resolves: bz#1968509
  (Use MSG_ZEROCOPY on QEMU Live Migration)
- Resolves: bz#2096143
  (The migration port is not released if use it again for recovering postcopy migration)
This commit is contained in:
Camilla Conte 2022-07-05 10:49:23 +00:00
parent 0ddb1bc433
commit da38d5c28e
19 changed files with 3234 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

420
kvm-QIOChannel-Add-flags-on-io_writev-and-introduce-io_f.patch Normal file
View File

@ -0,0 +1,420 @@
From cda3fcf14f2883fea633e25256f6c14a71271adf Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:31 -0300
Subject: [PATCH 08/18] QIOChannel: Add flags on io_writev and introduce
io_flush callback
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [2/11] 06acfb6b0cb2c25733c2eb198011f7623b5a7024 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Add flags to io_writev and introduce io_flush as optional callback to
QIOChannelClass, allowing the implementation of zero copy writes by
subclasses.
How to use them:
- Write data using qio_channel_writev*(...,QIO_CHANNEL_WRITE_FLAG_ZERO_COPY),
- Wait write completion with qio_channel_flush().
Notes:
As some zero copy write implementations work asynchronously, it's
recommended to keep the write buffer untouched until the return of
qio_channel_flush(), to avoid the risk of sending an updated buffer
instead of the buffer state during write.
As io_flush callback is optional, if a subclass does not implement it, then:
- io_flush will return 0 without changing anything.
Also, some functions like qio_channel_writev_full_all() were adapted to
receive a flag parameter. That allows shared code between zero copy and
non-zero copy writev, and also an easier implementation on new flags.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220513062836.965425-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit b88651cb4d4fa416fdbb6afaf5b26ec8c035eaad)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
chardev/char-io.c | 2 +-
hw/remote/mpqemu-link.c | 2 +-
include/io/channel.h | 38 +++++++++++++++++++++-
io/channel-buffer.c | 1 +
io/channel-command.c | 1 +
io/channel-file.c | 1 +
io/channel-socket.c | 2 ++
io/channel-tls.c | 1 +
io/channel-websock.c | 1 +
io/channel.c | 49 +++++++++++++++++++++++------
migration/rdma.c | 1 +
scsi/pr-manager-helper.c | 2 +-
tests/unit/test-io-channel-socket.c | 1 +
13 files changed, 88 insertions(+), 14 deletions(-)
diff --git a/chardev/char-io.c b/chardev/char-io.c
index 8ced184160..4451128cba 100644
--- a/chardev/char-io.c
+++ b/chardev/char-io.c
@@ -122,7 +122,7 @@ int io_channel_send_full(QIOChannel *ioc,
ret = qio_channel_writev_full(
ioc, &iov, 1,
- fds, nfds, NULL);
+ fds, nfds, 0, NULL);
if (ret == QIO_CHANNEL_ERR_BLOCK) {
if (offset) {
return offset;
diff --git a/hw/remote/mpqemu-link.c b/hw/remote/mpqemu-link.c
index 7e841820e5..e8f556bd27 100644
--- a/hw/remote/mpqemu-link.c
+++ b/hw/remote/mpqemu-link.c
@@ -69,7 +69,7 @@ bool mpqemu_msg_send(MPQemuMsg *msg, QIOChannel *ioc, Error **errp)
}
if (!qio_channel_writev_full_all(ioc, send, G_N_ELEMENTS(send),
- fds, nfds, errp)) {
+ fds, nfds, 0, errp)) {
ret = true;
} else {
trace_mpqemu_send_io_error(msg->cmd, msg->size, nfds);
diff --git a/include/io/channel.h b/include/io/channel.h
index 88988979f8..c680ee7480 100644
--- a/include/io/channel.h
+++ b/include/io/channel.h
@@ -32,12 +32,15 @@ OBJECT_DECLARE_TYPE(QIOChannel, QIOChannelClass,
#define QIO_CHANNEL_ERR_BLOCK -2
+#define QIO_CHANNEL_WRITE_FLAG_ZERO_COPY 0x1
+
typedef enum QIOChannelFeature QIOChannelFeature;
enum QIOChannelFeature {
QIO_CHANNEL_FEATURE_FD_PASS,
QIO_CHANNEL_FEATURE_SHUTDOWN,
QIO_CHANNEL_FEATURE_LISTEN,
+ QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY,
};
@@ -104,6 +107,7 @@ struct QIOChannelClass {
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp);
ssize_t (*io_readv)(QIOChannel *ioc,
const struct iovec *iov,
@@ -136,6 +140,8 @@ struct QIOChannelClass {
IOHandler *io_read,
IOHandler *io_write,
void *opaque);
+ int (*io_flush)(QIOChannel *ioc,
+ Error **errp);
};
/* General I/O handling functions */
@@ -228,6 +234,7 @@ ssize_t qio_channel_readv_full(QIOChannel *ioc,
* @niov: the length of the @iov array
* @fds: an array of file handles to send
* @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
* @errp: pointer to a NULL-initialized error object
*
* Write data to the IO channel, reading it from the
@@ -260,6 +267,7 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp);
/**
@@ -837,6 +845,7 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
* @niov: the length of the @iov array
* @fds: an array of file handles to send
* @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
* @errp: pointer to a NULL-initialized error object
*
*
@@ -846,6 +855,14 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
* to be written, yielding from the current coroutine
* if required.
*
+ * If QIO_CHANNEL_WRITE_FLAG_ZERO_COPY is passed in flags,
+ * instead of waiting for all requested data to be written,
+ * this function will wait until it's all queued for writing.
+ * In this case, if the buffer gets changed between queueing and
+ * sending, the updated buffer will be sent. If this is not a
+ * desired behavior, it's suggested to call qio_channel_flush()
+ * before reusing the buffer.
+ *
* Returns: 0 if all bytes were written, or -1 on error
*/
@@ -853,6 +870,25 @@ int qio_channel_writev_full_all(QIOChannel *ioc,
const struct iovec *iov,
size_t niov,
int *fds, size_t nfds,
- Error **errp);
+ int flags, Error **errp);
+
+/**
+ * qio_channel_flush:
+ * @ioc: the channel object
+ * @errp: pointer to a NULL-initialized error object
+ *
+ * Will block until every packet queued with
+ * qio_channel_writev_full() + QIO_CHANNEL_WRITE_FLAG_ZERO_COPY
+ * is sent, or return in case of any error.
+ *
+ * If not implemented, acts as a no-op, and returns 0.
+ *
+ * Returns -1 if any error is found,
+ * 1 if every send failed to use zero copy.
+ * 0 otherwise.
+ */
+
+int qio_channel_flush(QIOChannel *ioc,
+ Error **errp);
#endif /* QIO_CHANNEL_H */
diff --git a/io/channel-buffer.c b/io/channel-buffer.c
index baa4e2b089..bf52011be2 100644
--- a/io/channel-buffer.c
+++ b/io/channel-buffer.c
@@ -81,6 +81,7 @@ static ssize_t qio_channel_buffer_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelBuffer *bioc = QIO_CHANNEL_BUFFER(ioc);
diff --git a/io/channel-command.c b/io/channel-command.c
index 338da73ade..54560464ae 100644
--- a/io/channel-command.c
+++ b/io/channel-command.c
@@ -258,6 +258,7 @@ static ssize_t qio_channel_command_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelCommand *cioc = QIO_CHANNEL_COMMAND(ioc);
diff --git a/io/channel-file.c b/io/channel-file.c
index d7cf6d278f..ef6807a6be 100644
--- a/io/channel-file.c
+++ b/io/channel-file.c
@@ -114,6 +114,7 @@ static ssize_t qio_channel_file_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelFile *fioc = QIO_CHANNEL_FILE(ioc);
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 7a8d9f69c9..a1be2197ca 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -525,6 +525,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
@@ -620,6 +621,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
diff --git a/io/channel-tls.c b/io/channel-tls.c
index 2ae1b92fc0..4ce890a538 100644
--- a/io/channel-tls.c
+++ b/io/channel-tls.c
@@ -301,6 +301,7 @@ static ssize_t qio_channel_tls_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelTLS *tioc = QIO_CHANNEL_TLS(ioc);
diff --git a/io/channel-websock.c b/io/channel-websock.c
index 55145a6a8c..9619906ac3 100644
--- a/io/channel-websock.c
+++ b/io/channel-websock.c
@@ -1127,6 +1127,7 @@ static ssize_t qio_channel_websock_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelWebsock *wioc = QIO_CHANNEL_WEBSOCK(ioc);
diff --git a/io/channel.c b/io/channel.c
index e8b019dc36..0640941ac5 100644
--- a/io/channel.c
+++ b/io/channel.c
@@ -72,18 +72,32 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
- if ((fds || nfds) &&
- !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
+ if (fds || nfds) {
+ if (!qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
+ error_setg_errno(errp, EINVAL,
+ "Channel does not support file descriptor passing");
+ return -1;
+ }
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ error_setg_errno(errp, EINVAL,
+ "Zero Copy does not support file descriptor passing");
+ return -1;
+ }
+ }
+
+ if ((flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) &&
+ !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
error_setg_errno(errp, EINVAL,
- "Channel does not support file descriptor passing");
+ "Requested Zero Copy feature is not available");
return -1;
}
- return klass->io_writev(ioc, iov, niov, fds, nfds, errp);
+ return klass->io_writev(ioc, iov, niov, fds, nfds, flags, errp);
}
@@ -217,14 +231,14 @@ int qio_channel_writev_all(QIOChannel *ioc,
size_t niov,
Error **errp)
{
- return qio_channel_writev_full_all(ioc, iov, niov, NULL, 0, errp);
+ return qio_channel_writev_full_all(ioc, iov, niov, NULL, 0, 0, errp);
}
int qio_channel_writev_full_all(QIOChannel *ioc,
const struct iovec *iov,
size_t niov,
int *fds, size_t nfds,
- Error **errp)
+ int flags, Error **errp)
{
int ret = -1;
struct iovec *local_iov = g_new(struct iovec, niov);
@@ -237,8 +251,10 @@ int qio_channel_writev_full_all(QIOChannel *ioc,
while (nlocal_iov > 0) {
ssize_t len;
- len = qio_channel_writev_full(ioc, local_iov, nlocal_iov, fds, nfds,
- errp);
+
+ len = qio_channel_writev_full(ioc, local_iov, nlocal_iov, fds,
+ nfds, flags, errp);
+
if (len == QIO_CHANNEL_ERR_BLOCK) {
if (qemu_in_coroutine()) {
qio_channel_yield(ioc, G_IO_OUT);
@@ -277,7 +293,7 @@ ssize_t qio_channel_writev(QIOChannel *ioc,
size_t niov,
Error **errp)
{
- return qio_channel_writev_full(ioc, iov, niov, NULL, 0, errp);
+ return qio_channel_writev_full(ioc, iov, niov, NULL, 0, 0, errp);
}
@@ -297,7 +313,7 @@ ssize_t qio_channel_write(QIOChannel *ioc,
Error **errp)
{
struct iovec iov = { .iov_base = (char *)buf, .iov_len = buflen };
- return qio_channel_writev_full(ioc, &iov, 1, NULL, 0, errp);
+ return qio_channel_writev_full(ioc, &iov, 1, NULL, 0, 0, errp);
}
@@ -473,6 +489,19 @@ off_t qio_channel_io_seek(QIOChannel *ioc,
return klass->io_seek(ioc, offset, whence, errp);
}
+int qio_channel_flush(QIOChannel *ioc,
+ Error **errp)
+{
+ QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
+
+ if (!klass->io_flush ||
+ !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
+ return 0;
+ }
+
+ return klass->io_flush(ioc, errp);
+}
+
static void qio_channel_restart_read(void *opaque)
{
diff --git a/migration/rdma.c b/migration/rdma.c
index ef1e65ec36..672d1958a9 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -2840,6 +2840,7 @@ static ssize_t qio_channel_rdma_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelRDMA *rioc = QIO_CHANNEL_RDMA(ioc);
diff --git a/scsi/pr-manager-helper.c b/scsi/pr-manager-helper.c
index 451c7631b7..3be52a98d5 100644
--- a/scsi/pr-manager-helper.c
+++ b/scsi/pr-manager-helper.c
@@ -77,7 +77,7 @@ static int pr_manager_helper_write(PRManagerHelper *pr_mgr,
iov.iov_base = (void *)buf;
iov.iov_len = sz;
n_written = qio_channel_writev_full(QIO_CHANNEL(pr_mgr->ioc), &iov, 1,
- nfds ? &fd : NULL, nfds, errp);
+ nfds ? &fd : NULL, nfds, 0, errp);
if (n_written <= 0) {
assert(n_written != QIO_CHANNEL_ERR_BLOCK);
diff --git a/tests/unit/test-io-channel-socket.c b/tests/unit/test-io-channel-socket.c
index c49eec1f03..6713886d02 100644
--- a/tests/unit/test-io-channel-socket.c
+++ b/tests/unit/test-io-channel-socket.c
@@ -444,6 +444,7 @@ static void test_io_channel_unix_fd_pass(void)
G_N_ELEMENTS(iosend),
fdsend,
G_N_ELEMENTS(fdsend),
+ 0,
&error_abort);
qio_channel_readv_full(dst,
--
2.35.3

58
kvm-QIOChannelSocket-Fix-zero-copy-send-so-socket-flush-.patch Normal file
View File

@ -0,0 +1,58 @@
From e70f01749addd7d0b7aa7fa4fdedb664f98e6b9b Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 20 Jun 2022 02:39:43 -0300
Subject: [PATCH 16/18] QIOChannelSocket: Fix zero-copy send so socket flush
works
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [10/11] a2dfac987e24026b1a78e90b86234ca206b6401f (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Somewhere between v6 and v7 the of the zero-copy-send patchset a crucial
part of the flushing mechanism got missing: incrementing zero_copy_queued.
Without that, the flushing interface becomes a no-op, and there is no
guarantee the buffer is really sent.
This can go as bad as causing a corruption in RAM during migration.
Fixes: 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Reported-by: 徐闯 <xuchuangxclwt@bytedance.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 4f5a09714c983a3471fd12e3c7f3196e95c650c1)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 7490e5943d..8ae8b212cf 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -612,6 +612,11 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
"Unable to write to socket");
return -1;
}
+
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ sioc->zero_copy_queued++;
+ }
+
return ret;
}
#else /* WIN32 */
--
2.35.3

249
kvm-QIOChannelSocket-Implement-io_writev-zero-copy-flag-.patch Normal file
View File

@ -0,0 +1,249 @@
From 4aeba0365d30dabe2e70dc172683f0878a4a9621 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:32 -0300
Subject: [PATCH 09/18] QIOChannelSocket: Implement io_writev zero copy flag &
io_flush for CONFIG_LINUX
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [3/11] 9afeac1f5ac7675624660a0281726c09c8321180 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
For CONFIG_LINUX, implement the new zero copy flag and the optional callback
io_flush on QIOChannelSocket, but enables it only when MSG_ZEROCOPY
feature is available in the host kernel, which is checked on
qio_channel_socket_connect_sync()
qio_channel_socket_flush() was implemented by counting how many times
sendmsg(...,MSG_ZEROCOPY) was successfully called, and then reading the
socket's error queue, in order to find how many of them finished sending.
Flush will loop until those counters are the same, or until some error occurs.
Notes on using writev() with QIO_CHANNEL_WRITE_FLAG_ZERO_COPY:
1: Buffer
- As MSG_ZEROCOPY tells the kernel to use the same user buffer to avoid copying,
some caution is necessary to avoid overwriting any buffer before it's sent.
If something like this happen, a newer version of the buffer may be sent instead.
- If this is a problem, it's recommended to call qio_channel_flush() before freeing
or re-using the buffer.
2: Locked memory
- When using MSG_ZERCOCOPY, the buffer memory will be locked after queued, and
unlocked after it's sent.
- Depending on the size of each buffer, and how often it's sent, it may require
a larger amount of locked memory than usually available to non-root user.
- If the required amount of locked memory is not available, writev_zero_copy
will return an error, which can abort an operation like migration,
- Because of this, when an user code wants to add zero copy as a feature, it
requires a mechanism to disable it, so it can still be accessible to less
privileged users.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220513062836.965425-4-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 2bc58ffc2926a4efdd03edfb5909861fefc68c3d)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
include/io/channel-socket.h | 2 +
io/channel-socket.c | 116 ++++++++++++++++++++++++++++++++++--
2 files changed, 114 insertions(+), 4 deletions(-)
diff --git a/include/io/channel-socket.h b/include/io/channel-socket.h
index e747e63514..513c428fe4 100644
--- a/include/io/channel-socket.h
+++ b/include/io/channel-socket.h
@@ -47,6 +47,8 @@ struct QIOChannelSocket {
socklen_t localAddrLen;
struct sockaddr_storage remoteAddr;
socklen_t remoteAddrLen;
+ ssize_t zero_copy_queued;
+ ssize_t zero_copy_sent;
};
diff --git a/io/channel-socket.c b/io/channel-socket.c
index a1be2197ca..fbd2214d20 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -26,6 +26,14 @@
#include "io/channel-watch.h"
#include "trace.h"
#include "qapi/clone-visitor.h"
+#ifdef CONFIG_LINUX
+#include <linux/errqueue.h>
+#include <sys/socket.h>
+
+#if (defined(MSG_ZEROCOPY) && defined(SO_ZEROCOPY))
+#define QEMU_MSG_ZEROCOPY
+#endif
+#endif
#define SOCKET_MAX_FDS 16
@@ -55,6 +63,8 @@ qio_channel_socket_new(void)
sioc = QIO_CHANNEL_SOCKET(object_new(TYPE_QIO_CHANNEL_SOCKET));
sioc->fd = -1;
+ sioc->zero_copy_queued = 0;
+ sioc->zero_copy_sent = 0;
ioc = QIO_CHANNEL(sioc);
qio_channel_set_feature(ioc, QIO_CHANNEL_FEATURE_SHUTDOWN);
@@ -154,6 +164,16 @@ int qio_channel_socket_connect_sync(QIOChannelSocket *ioc,
return -1;
}
+#ifdef QEMU_MSG_ZEROCOPY
+ int ret, v = 1;
+ ret = setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &v, sizeof(v));
+ if (ret == 0) {
+ /* Zero copy available on host */
+ qio_channel_set_feature(QIO_CHANNEL(ioc),
+ QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY);
+ }
+#endif
+
return 0;
}
@@ -534,6 +554,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
char control[CMSG_SPACE(sizeof(int) * SOCKET_MAX_FDS)];
size_t fdsize = sizeof(int) * nfds;
struct cmsghdr *cmsg;
+ int sflags = 0;
memset(control, 0, CMSG_SPACE(sizeof(int) * SOCKET_MAX_FDS));
@@ -558,15 +579,31 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
memcpy(CMSG_DATA(cmsg), fds, fdsize);
}
+#ifdef QEMU_MSG_ZEROCOPY
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ sflags = MSG_ZEROCOPY;
+ }
+#endif
+
retry:
- ret = sendmsg(sioc->fd, &msg, 0);
+ ret = sendmsg(sioc->fd, &msg, sflags);
if (ret <= 0) {
- if (errno == EAGAIN) {
+ switch (errno) {
+ case EAGAIN:
return QIO_CHANNEL_ERR_BLOCK;
- }
- if (errno == EINTR) {
+ case EINTR:
goto retry;
+#ifdef QEMU_MSG_ZEROCOPY
+ case ENOBUFS:
+ if (sflags & MSG_ZEROCOPY) {
+ error_setg_errno(errp, errno,
+ "Process can't lock enough memory for using MSG_ZEROCOPY");
+ return -1;
+ }
+ break;
+#endif
}
+
error_setg_errno(errp, errno,
"Unable to write to socket");
return -1;
@@ -660,6 +697,74 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
}
#endif /* WIN32 */
+
+#ifdef QEMU_MSG_ZEROCOPY
+static int qio_channel_socket_flush(QIOChannel *ioc,
+ Error **errp)
+{
+ QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
+ struct msghdr msg = {};
+ struct sock_extended_err *serr;
+ struct cmsghdr *cm;
+ char control[CMSG_SPACE(sizeof(*serr))];
+ int received;
+ int ret = 1;
+
+ msg.msg_control = control;
+ msg.msg_controllen = sizeof(control);
+ memset(control, 0, sizeof(control));
+
+ while (sioc->zero_copy_sent < sioc->zero_copy_queued) {
+ received = recvmsg(sioc->fd, &msg, MSG_ERRQUEUE);
+ if (received < 0) {
+ switch (errno) {
+ case EAGAIN:
+ /* Nothing on errqueue, wait until something is available */
+ qio_channel_wait(ioc, G_IO_ERR);
+ continue;
+ case EINTR:
+ continue;
+ default:
+ error_setg_errno(errp, errno,
+ "Unable to read errqueue");
+ return -1;
+ }
+ }
+
+ cm = CMSG_FIRSTHDR(&msg);
+ if (cm->cmsg_level != SOL_IP &&
+ cm->cmsg_type != IP_RECVERR) {
+ error_setg_errno(errp, EPROTOTYPE,
+ "Wrong cmsg in errqueue");
+ return -1;
+ }
+
+ serr = (void *) CMSG_DATA(cm);
+ if (serr->ee_errno != SO_EE_ORIGIN_NONE) {
+ error_setg_errno(errp, serr->ee_errno,
+ "Error on socket");
+ return -1;
+ }
+ if (serr->ee_origin != SO_EE_ORIGIN_ZEROCOPY) {
+ error_setg_errno(errp, serr->ee_origin,
+ "Error not from zero copy");
+ return -1;
+ }
+
+ /* No errors, count successfully finished sendmsg()*/
+ sioc->zero_copy_sent += serr->ee_data - serr->ee_info + 1;
+
+ /* If any sendmsg() succeeded using zero copy, return 0 at the end */
+ if (serr->ee_code != SO_EE_CODE_ZEROCOPY_COPIED) {
+ ret = 0;
+ }
+ }
+
+ return ret;
+}
+
+#endif /* QEMU_MSG_ZEROCOPY */
+
static int
qio_channel_socket_set_blocking(QIOChannel *ioc,
bool enabled,
@@ -790,6 +895,9 @@ static void qio_channel_socket_class_init(ObjectClass *klass,
ioc_klass->io_set_delay = qio_channel_socket_set_delay;
ioc_klass->io_create_watch = qio_channel_socket_create_watch;
ioc_klass->io_set_aio_fd_handler = qio_channel_socket_set_aio_fd_handler;
+#ifdef QEMU_MSG_ZEROCOPY
+ ioc_klass->io_flush = qio_channel_socket_flush;
+#endif
}
static const TypeInfo qio_channel_socket_info = {
--
2.35.3

82
kvm-QIOChannelSocket-Introduce-assert-and-reduce-ifdefs-.patch Normal file
View File

@ -0,0 +1,82 @@
From 60bf942a58db12c821f2a6a49e2e0b04b99bec30 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 20 Jun 2022 02:39:42 -0300
Subject: [PATCH 15/18] QIOChannelSocket: Introduce assert and reduce ifdefs to
improve readability
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [9/11] eaa02d68301852ccc98bdacc7387d8d03be1cb05 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
During implementation of MSG_ZEROCOPY feature, a lot of #ifdefs were
introduced, particularly at qio_channel_socket_writev().
Rewrite some of those changes so it's easier to read.
Also, introduce an assert to help detect incorrect zero-copy usage is when
it's disabled on build.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Fixed up thinko'd g_assert_unreachable->g_assert_not_reached
(cherry picked from commit 803ca43e4c7fcf32f9f68c118301ccd0c83ece3f)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index fbd2214d20..7490e5943d 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -579,11 +579,17 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
memcpy(CMSG_DATA(cmsg), fds, fdsize);
}
-#ifdef QEMU_MSG_ZEROCOPY
if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+#ifdef QEMU_MSG_ZEROCOPY
sflags = MSG_ZEROCOPY;
- }
+#else
+ /*
+ * We expect QIOChannel class entry point to have
+ * blocked this code path already
+ */
+ g_assert_not_reached();
#endif
+ }
retry:
ret = sendmsg(sioc->fd, &msg, sflags);
@@ -593,15 +599,13 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
return QIO_CHANNEL_ERR_BLOCK;
case EINTR:
goto retry;
-#ifdef QEMU_MSG_ZEROCOPY
case ENOBUFS:
- if (sflags & MSG_ZEROCOPY) {
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
error_setg_errno(errp, errno,
"Process can't lock enough memory for using MSG_ZEROCOPY");
return -1;
}
break;
-#endif
}
error_setg_errno(errp, errno,
--
2.35.3

237
kvm-RHEL-only-AArch64-Drop-unsupported-CPU-types.patch Normal file
View File

@ -0,0 +1,237 @@
From 055edf068196622a3e1868c9e4c991d410272a6d Mon Sep 17 00:00:00 2001
From: Andrew Jones <drjones@redhat.com>
Date: Wed, 15 Jun 2022 15:28:27 +0200
Subject: [PATCH 03/18] RHEL-only: AArch64: Drop unsupported CPU types
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [3/6] 21f54c86dc87e5e75a64459b5a385686bc09640c (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2066824
Upstream Status: RHEL only
We only need to support AArch64 cpu types and we only need three
types:
1) A base type to use with TCG, i.e. a cpu type with only base
features. 'cortex-a57' serves this role and is currently used
by libguestfs.
2) The 'max' type, which is for both KVM and TCG and is good for
tests that just specify 'max' but run under both. 'max' with
TCG also provides the VM with all the CPU features TCG
supports, which is good for VMs that need features not
provided by the basic cortex-a57.
3) The host type which is used with KVM.
Signed-off-by: Andrew Jones <drjones@redhat.com>
---
hw/arm/virt.c | 4 ++++
target/arm/cpu64.c | 6 ++++++
target/arm/cpu_tcg.c | 12 ++----------
tests/qtest/arm-cpu-features.c | 6 ++++++
4 files changed, 18 insertions(+), 10 deletions(-)
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 95d012d6eb..74119976d3 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -239,12 +239,16 @@ static const int a15irqmap[] = {
};
static const char *valid_cpus[] = {
+#if 0 /* Disabled for Red Hat Enterprise Linux */
ARM_CPU_TYPE_NAME("cortex-a7"),
ARM_CPU_TYPE_NAME("cortex-a15"),
ARM_CPU_TYPE_NAME("cortex-a53"),
+#endif /* disabled for RHEL */
ARM_CPU_TYPE_NAME("cortex-a57"),
+#if 0 /* Disabled for Red Hat Enterprise Linux */
ARM_CPU_TYPE_NAME("cortex-a72"),
ARM_CPU_TYPE_NAME("a64fx"),
+#endif /* disabled for RHEL */
ARM_CPU_TYPE_NAME("host"),
ARM_CPU_TYPE_NAME("max"),
};
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index eb44c05822..e80b831073 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -146,6 +146,7 @@ static void aarch64_a57_initfn(Object *obj)
define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void aarch64_a53_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -249,6 +250,7 @@ static void aarch64_a72_initfn(Object *obj)
cpu->gic_vprebits = 5;
define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo);
}
+#endif /* disabled for RHEL */
void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp)
{
@@ -923,6 +925,7 @@ static void aarch64_max_initfn(Object *obj)
qdev_property_add_static(DEVICE(obj), &arm_cpu_lpa2_property);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void aarch64_a64fx_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -969,12 +972,15 @@ static void aarch64_a64fx_initfn(Object *obj)
/* TODO: Add A64FX specific HPC extension registers */
}
+#endif /* disabled for RHEL */
static const ARMCPUInfo aarch64_cpus[] = {
{ .name = "cortex-a57", .initfn = aarch64_a57_initfn },
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "cortex-a53", .initfn = aarch64_a53_initfn },
{ .name = "cortex-a72", .initfn = aarch64_a72_initfn },
{ .name = "a64fx", .initfn = aarch64_a64fx_initfn },
+#endif /* disabled for RHEL */
{ .name = "max", .initfn = aarch64_max_initfn },
#if defined(CONFIG_KVM) || defined(CONFIG_HVF)
{ .name = "host", .initfn = aarch64_host_initfn },
diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
index 3826fa5122..74727fc92c 100644
--- a/target/arm/cpu_tcg.c
+++ b/target/arm/cpu_tcg.c
@@ -19,10 +19,10 @@
#include "hw/boards.h"
#endif
+#if 0 /* Disabled for Red Hat Enterprise Linux */
/* CPU models. These are not needed for the AArch64 linux-user build. */
#if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
-#if 0 /* Disabled for Red Hat Enterprise Linux */
#if !defined(CONFIG_USER_ONLY) && defined(CONFIG_TCG)
static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
{
@@ -376,7 +376,6 @@ static void cortex_a9_initfn(Object *obj)
cpu->ccsidr[1] = 0x200fe019; /* 16k L1 icache. */
define_arm_cp_regs(cpu, cortexa9_cp_reginfo);
}
-#endif /* disabled for RHEL */
#ifndef CONFIG_USER_ONLY
static uint64_t a15_l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -402,7 +401,6 @@ static const ARMCPRegInfo cortexa15_cp_reginfo[] = {
REGINFO_SENTINEL
};
-#if 0 /* Disabled for Red Hat Enterprise Linux */
static void cortex_a7_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -448,7 +446,6 @@ static void cortex_a7_initfn(Object *obj)
cpu->ccsidr[2] = 0x711fe07a; /* 4096K L2 unified cache */
define_arm_cp_regs(cpu, cortexa15_cp_reginfo); /* Same as A15 */
}
-#endif /* disabled for RHEL */
static void cortex_a15_initfn(Object *obj)
{
@@ -492,7 +489,6 @@ static void cortex_a15_initfn(Object *obj)
define_arm_cp_regs(cpu, cortexa15_cp_reginfo);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
static void cortex_m0_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -933,7 +929,6 @@ static void arm_v7m_class_init(ObjectClass *oc, void *data)
cc->gdb_core_xml_file = "arm-m-profile.xml";
}
-#endif /* disabled for RHEL */
#ifndef TARGET_AARCH64
/*
@@ -1013,7 +1008,6 @@ static void arm_max_initfn(Object *obj)
#endif /* !TARGET_AARCH64 */
static const ARMCPUInfo arm_tcg_cpus[] = {
-#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "arm926", .initfn = arm926_initfn },
{ .name = "arm946", .initfn = arm946_initfn },
{ .name = "arm1026", .initfn = arm1026_initfn },
@@ -1029,9 +1023,7 @@ static const ARMCPUInfo arm_tcg_cpus[] = {
{ .name = "cortex-a7", .initfn = cortex_a7_initfn },
{ .name = "cortex-a8", .initfn = cortex_a8_initfn },
{ .name = "cortex-a9", .initfn = cortex_a9_initfn },
-#endif /* disabled for RHEL */
{ .name = "cortex-a15", .initfn = cortex_a15_initfn },
-#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "cortex-m0", .initfn = cortex_m0_initfn,
.class_init = arm_v7m_class_init },
{ .name = "cortex-m3", .initfn = cortex_m3_initfn,
@@ -1062,7 +1054,6 @@ static const ARMCPUInfo arm_tcg_cpus[] = {
{ .name = "pxa270-b1", .initfn = pxa270b1_initfn },
{ .name = "pxa270-c0", .initfn = pxa270c0_initfn },
{ .name = "pxa270-c5", .initfn = pxa270c5_initfn },
-#endif /* disabled for RHEL */
#ifndef TARGET_AARCH64
{ .name = "max", .initfn = arm_max_initfn },
#endif
@@ -1090,3 +1081,4 @@ static void arm_tcg_cpu_register_types(void)
type_init(arm_tcg_cpu_register_types)
#endif /* !CONFIG_USER_ONLY || !TARGET_AARCH64 */
+#endif /* disabled for RHEL */
diff --git a/tests/qtest/arm-cpu-features.c b/tests/qtest/arm-cpu-features.c
index f76652143a..fe2a0a070d 100644
--- a/tests/qtest/arm-cpu-features.c
+++ b/tests/qtest/arm-cpu-features.c
@@ -440,8 +440,10 @@ static void test_query_cpu_model_expansion(const void *data)
assert_error(qts, "host", "The CPU type 'host' requires KVM", NULL);
/* Test expected feature presence/absence for some cpu types */
+#if 0 /* Disabled for Red Hat Enterprise Linux */
assert_has_feature_enabled(qts, "cortex-a15", "pmu");
assert_has_not_feature(qts, "cortex-a15", "aarch64");
+#endif /* disabled for RHEL */
/* Enabling and disabling pmu should always work. */
assert_has_feature_enabled(qts, "max", "pmu");
@@ -458,6 +460,7 @@ static void test_query_cpu_model_expansion(const void *data)
assert_has_feature_enabled(qts, "cortex-a57", "pmu");
assert_has_feature_enabled(qts, "cortex-a57", "aarch64");
+#if 0 /* Disabled for Red Hat Enterprise Linux */
assert_has_feature_enabled(qts, "a64fx", "pmu");
assert_has_feature_enabled(qts, "a64fx", "aarch64");
/*
@@ -470,6 +473,7 @@ static void test_query_cpu_model_expansion(const void *data)
"{ 'sve384': true }");
assert_error(qts, "a64fx", "cannot enable sve640",
"{ 'sve640': true }");
+#endif /* disabled for RHEL */
sve_tests_default(qts, "max");
pauth_tests_default(qts, "max");
@@ -505,9 +509,11 @@ static void test_query_cpu_model_expansion_kvm(const void *data)
QDict *resp;
char *error;
+#if 0 /* Disabled for Red Hat Enterprise Linux */
assert_error(qts, "cortex-a15",
"We cannot guarantee the CPU type 'cortex-a15' works "
"with KVM on this host", NULL);
+#endif /* disabled for RHEL */
assert_has_feature_enabled(qts, "host", "aarch64");
--
2.35.3

95
kvm-RHEL-only-tests-avocado-Switch-aarch64-tests-from-a5.patch Normal file
View File

@ -0,0 +1,95 @@
From d710394f68eb0b6116dd8ac76f619c192e0d5972 Mon Sep 17 00:00:00 2001
From: Andrew Jones <drjones@redhat.com>
Date: Wed, 15 Jun 2022 15:28:27 +0200
Subject: [PATCH 02/18] RHEL-only: tests/avocado: Switch aarch64 tests from a53
to a57
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [2/6] e85ef69b42c411a6997e4da10ba05176368769b3 (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2066824
Upstream Status: RHEL only
We plan to remove the cortex-a53 from the supported cpu types. Switch
all avocado tests that use it to the cortex-a57, which will work the
same and we intend to keep. We don't want to try and upstream this
change since the better upstream change would be to switch from the
a53 to 'max', but the upstream tests also need to use later guest
kernels to use 'max' (see qemu upstream commit 0942820408dc
("hw/arm/virt: Disable LPA2 for -machine virt-6.2")
Signed-off-by: Andrew Jones <drjones@redhat.com>
---
tests/avocado/replay_kernel.py | 2 +-
tests/avocado/reverse_debugging.py | 2 +-
tests/avocado/tcg_plugins.py | 6 +++---
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/avocado/replay_kernel.py b/tests/avocado/replay_kernel.py
index 0b2b0dc692..3a7b5f0748 100644
--- a/tests/avocado/replay_kernel.py
+++ b/tests/avocado/replay_kernel.py
@@ -147,7 +147,7 @@ def test_aarch64_virt(self):
"""
:avocado: tags=arch:aarch64
:avocado: tags=machine:virt
- :avocado: tags=cpu:cortex-a53
+ :avocado: tags=cpu:cortex-a57
"""
kernel_url = ('https://archives.fedoraproject.org/pub/archive/fedora'
'/linux/releases/29/Everything/aarch64/os/images/pxeboot'
diff --git a/tests/avocado/reverse_debugging.py b/tests/avocado/reverse_debugging.py
index d2921e70c3..66d185ed42 100644
--- a/tests/avocado/reverse_debugging.py
+++ b/tests/avocado/reverse_debugging.py
@@ -198,7 +198,7 @@ def test_aarch64_virt(self):
"""
:avocado: tags=arch:aarch64
:avocado: tags=machine:virt
- :avocado: tags=cpu:cortex-a53
+ :avocado: tags=cpu:cortex-a57
"""
kernel_url = ('https://archives.fedoraproject.org/pub/archive/fedora'
'/linux/releases/29/Everything/aarch64/os/images/pxeboot'
diff --git a/tests/avocado/tcg_plugins.py b/tests/avocado/tcg_plugins.py
index 642d2e49e3..93b3afd823 100644
--- a/tests/avocado/tcg_plugins.py
+++ b/tests/avocado/tcg_plugins.py
@@ -68,7 +68,7 @@ def test_aarch64_virt_insn(self):
:avocado: tags=accel:tcg
:avocado: tags=arch:aarch64
:avocado: tags=machine:virt
- :avocado: tags=cpu:cortex-a53
+ :avocado: tags=cpu:cortex-a57
"""
kernel_path = self._grab_aarch64_kernel()
kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE +
@@ -94,7 +94,7 @@ def test_aarch64_virt_insn_icount(self):
:avocado: tags=accel:tcg
:avocado: tags=arch:aarch64
:avocado: tags=machine:virt
- :avocado: tags=cpu:cortex-a53
+ :avocado: tags=cpu:cortex-a57
"""
kernel_path = self._grab_aarch64_kernel()
kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE +
@@ -120,7 +120,7 @@ def test_aarch64_virt_mem_icount(self):
:avocado: tags=accel:tcg
:avocado: tags=arch:aarch64
:avocado: tags=machine:virt
- :avocado: tags=cpu:cortex-a53
+ :avocado: tags=cpu:cortex-a57
"""
kernel_path = self._grab_aarch64_kernel()
kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE +
--
2.35.3

87
kvm-meson.build-Fix-docker-test-build-alpine-when-includ.patch Normal file
View File

@ -0,0 +1,87 @@
From 7c489b54b0bb33445113fbf16e88feb23be68013 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:30 -0300
Subject: [PATCH 07/18] meson.build: Fix docker-test-build@alpine when
including linux/errqueue.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [1/11] f058eb846fcf611d527a1dd3b0cc399cdc17e3ee (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
A build error happens in alpine CI when linux/errqueue.h is included
in io/channel-socket.c, due to redefining of 'struct __kernel_timespec':
===
ninja: job failed: [...]
In file included from /usr/include/linux/errqueue.h:6,
from ../io/channel-socket.c:29:
/usr/include/linux/time_types.h:7:8: error: redefinition of 'struct __kernel_timespec'
7 | struct __kernel_timespec {
| ^~~~~~~~~~~~~~~~~
In file included from /usr/include/liburing.h:19,
from /builds/user/qemu/include/block/aio.h:18,
from /builds/user/qemu/include/io/channel.h:26,
from /builds/user/qemu/include/io/channel-socket.h:24,
from ../io/channel-socket.c:24:
/usr/include/liburing/compat.h:9:8: note: originally defined here
9 | struct __kernel_timespec {
| ^~~~~~~~~~~~~~~~~
ninja: subcommand failed
===
As above error message suggests, 'struct __kernel_timespec' was already
defined by liburing/compat.h.
Fix alpine CI by adding test to disable liburing in configure step if a
redefinition happens between linux/errqueue.h and liburing/compat.h.
[dgilbert: This has been fixed in Alpine issue 13813 and liburing]
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220513062836.965425-2-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 354081d43de44ebd3497fe08f7f0121a5517d528)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
meson.build | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/meson.build b/meson.build
index 5a7c10e639..13e3323380 100644
--- a/meson.build
+++ b/meson.build
@@ -471,12 +471,23 @@ if not get_option('linux_aio').auto() or have_block
required: get_option('linux_aio'),
kwargs: static_kwargs)
endif
+
+linux_io_uring_test = '''
+ #include <liburing.h>
+ #include <linux/errqueue.h>
+
+ int main(void) { return 0; }'''
+
linux_io_uring = not_found
if not get_option('linux_io_uring').auto() or have_block
linux_io_uring = dependency('liburing', version: '>=0.3',
required: get_option('linux_io_uring'),
method: 'pkg-config', kwargs: static_kwargs)
+ if not cc.links(linux_io_uring_test)
+ linux_io_uring = not_found
+ endif
endif
+
libnfs = not_found
if not get_option('libnfs').auto() or have_block
libnfs = dependency('libnfs', version: '>=1.9.3',
--
2.35.3

106
kvm-migration-Add-migrate_use_tls-helper.patch Normal file
View File

@ -0,0 +1,106 @@
From 828f6c106eedcb7a48e551ffda15af56ff92a899 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:34 -0300
Subject: [PATCH 11/18] migration: Add migrate_use_tls() helper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [5/11] 06e945297c3b9c0ce5864885aafcdba1e5746bc2 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
A lot of places check parameters.tls_creds in order to evaluate if TLS is
in use, and sometimes call migrate_get_current() just for that test.
Add new helper function migrate_use_tls() in order to simplify testing
for TLS usage.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220513062836.965425-6-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit d2fafb6a6814a8998607d0baf691265032996a0f)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/channel.c | 3 +--
migration/migration.c | 9 +++++++++
migration/migration.h | 1 +
migration/multifd.c | 5 +----
4 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/migration/channel.c b/migration/channel.c
index c4fc000a1a..086b5c0d8b 100644
--- a/migration/channel.c
+++ b/migration/channel.c
@@ -38,8 +38,7 @@ void migration_channel_process_incoming(QIOChannel *ioc)
trace_migration_set_incoming_channel(
ioc, object_get_typename(OBJECT(ioc)));
- if (s->parameters.tls_creds &&
- *s->parameters.tls_creds &&
+ if (migrate_use_tls() &&
!object_dynamic_cast(OBJECT(ioc),
TYPE_QIO_CHANNEL_TLS)) {
migration_tls_channel_process_incoming(s, ioc, &local_err);
diff --git a/migration/migration.c b/migration/migration.c
index 0a6b3b9f4d..d91efb66fe 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -2582,6 +2582,15 @@ bool migrate_use_zero_copy_send(void)
}
#endif
+int migrate_use_tls(void)
+{
+ MigrationState *s;
+
+ s = migrate_get_current();
+
+ return s->parameters.tls_creds && *s->parameters.tls_creds;
+}
+
int migrate_use_xbzrle(void)
{
MigrationState *s;
diff --git a/migration/migration.h b/migration/migration.h
index 5bcb7628ef..c2cabb8a14 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -381,6 +381,7 @@ bool migrate_use_zero_copy_send(void);
#else
#define migrate_use_zero_copy_send() (false)
#endif
+int migrate_use_tls(void);
int migrate_use_xbzrle(void);
uint64_t migrate_xbzrle_cache_size(void);
bool migrate_colo_enabled(void);
diff --git a/migration/multifd.c b/migration/multifd.c
index 76b57a7177..43998ad117 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -784,14 +784,11 @@ static bool multifd_channel_connect(MultiFDSendParams *p,
QIOChannel *ioc,
Error *error)
{
- MigrationState *s = migrate_get_current();
-
trace_multifd_set_outgoing_channel(
ioc, object_get_typename(OBJECT(ioc)), p->tls_hostname, error);
if (!error) {
- if (s->parameters.tls_creds &&
- *s->parameters.tls_creds &&
+ if (migrate_use_tls() &&
!object_dynamic_cast(OBJECT(ioc),
TYPE_QIO_CHANNEL_TLS)) {
multifd_tls_channel_connect(p, ioc, &error);
--
2.35.3

250
kvm-migration-Add-zero-copy-send-parameter-for-QMP-HMP-f.patch Normal file
View File

@ -0,0 +1,250 @@
From d6500340dc3c1152b5efe04ef3daa50c17a55e30 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:33 -0300
Subject: [PATCH 10/18] migration: Add zero-copy-send parameter for QMP/HMP for
Linux
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [4/11] 514d98d595992c53ff98de750035e080ded8972e (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Add property that allows zero-copy migration of memory pages
on the sending side, and also includes a helper function
migrate_use_zero_copy_send() to check if it's enabled.
No code is introduced to actually do the migration, but it allow
future implementations to enable/disable this feature.
On non-Linux builds this parameter is compiled-out.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220513062836.965425-5-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit abb6295b3ace5d17c3a65936913fc346616dbf14)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 32 ++++++++++++++++++++++++++++++++
migration/migration.h | 5 +++++
migration/socket.c | 11 +++++++++--
monitor/hmp-cmds.c | 6 ++++++
qapi/migration.json | 24 ++++++++++++++++++++++++
5 files changed, 76 insertions(+), 2 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 695f0f2900..0a6b3b9f4d 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -899,6 +899,10 @@ MigrationParameters *qmp_query_migrate_parameters(Error **errp)
params->multifd_zlib_level = s->parameters.multifd_zlib_level;
params->has_multifd_zstd_level = true;
params->multifd_zstd_level = s->parameters.multifd_zstd_level;
+#ifdef CONFIG_LINUX
+ params->has_zero_copy_send = true;
+ params->zero_copy_send = s->parameters.zero_copy_send;
+#endif
params->has_xbzrle_cache_size = true;
params->xbzrle_cache_size = s->parameters.xbzrle_cache_size;
params->has_max_postcopy_bandwidth = true;
@@ -1555,6 +1559,11 @@ static void migrate_params_test_apply(MigrateSetParameters *params,
if (params->has_multifd_compression) {
dest->multifd_compression = params->multifd_compression;
}
+#ifdef CONFIG_LINUX
+ if (params->has_zero_copy_send) {
+ dest->zero_copy_send = params->zero_copy_send;
+ }
+#endif
if (params->has_xbzrle_cache_size) {
dest->xbzrle_cache_size = params->xbzrle_cache_size;
}
@@ -1667,6 +1676,11 @@ static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
if (params->has_multifd_compression) {
s->parameters.multifd_compression = params->multifd_compression;
}
+#ifdef CONFIG_LINUX
+ if (params->has_zero_copy_send) {
+ s->parameters.zero_copy_send = params->zero_copy_send;
+ }
+#endif
if (params->has_xbzrle_cache_size) {
s->parameters.xbzrle_cache_size = params->xbzrle_cache_size;
xbzrle_cache_resize(params->xbzrle_cache_size, errp);
@@ -2557,6 +2571,17 @@ int migrate_multifd_zstd_level(void)
return s->parameters.multifd_zstd_level;
}
+#ifdef CONFIG_LINUX
+bool migrate_use_zero_copy_send(void)
+{
+ MigrationState *s;
+
+ s = migrate_get_current();
+
+ return s->parameters.zero_copy_send;
+}
+#endif
+
int migrate_use_xbzrle(void)
{
MigrationState *s;
@@ -4200,6 +4225,10 @@ static Property migration_properties[] = {
DEFINE_PROP_UINT8("multifd-zstd-level", MigrationState,
parameters.multifd_zstd_level,
DEFAULT_MIGRATE_MULTIFD_ZSTD_LEVEL),
+#ifdef CONFIG_LINUX
+ DEFINE_PROP_BOOL("zero_copy_send", MigrationState,
+ parameters.zero_copy_send, false),
+#endif
DEFINE_PROP_SIZE("xbzrle-cache-size", MigrationState,
parameters.xbzrle_cache_size,
DEFAULT_MIGRATE_XBZRLE_CACHE_SIZE),
@@ -4297,6 +4326,9 @@ static void migration_instance_init(Object *obj)
params->has_multifd_compression = true;
params->has_multifd_zlib_level = true;
params->has_multifd_zstd_level = true;
+#ifdef CONFIG_LINUX
+ params->has_zero_copy_send = true;
+#endif
params->has_xbzrle_cache_size = true;
params->has_max_postcopy_bandwidth = true;
params->has_max_cpu_throttle = true;
diff --git a/migration/migration.h b/migration/migration.h
index 2de861df01..5bcb7628ef 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -376,6 +376,11 @@ MultiFDCompression migrate_multifd_compression(void);
int migrate_multifd_zlib_level(void);
int migrate_multifd_zstd_level(void);
+#ifdef CONFIG_LINUX
+bool migrate_use_zero_copy_send(void);
+#else
+#define migrate_use_zero_copy_send() (false)
+#endif
int migrate_use_xbzrle(void);
uint64_t migrate_xbzrle_cache_size(void);
bool migrate_colo_enabled(void);
diff --git a/migration/socket.c b/migration/socket.c
index 05705a32d8..3754d8f72c 100644
--- a/migration/socket.c
+++ b/migration/socket.c
@@ -74,9 +74,16 @@ static void socket_outgoing_migration(QIOTask *task,
if (qio_task_propagate_error(task, &err)) {
trace_migration_socket_outgoing_error(error_get_pretty(err));
- } else {
- trace_migration_socket_outgoing_connected(data->hostname);
+ goto out;
}
+
+ trace_migration_socket_outgoing_connected(data->hostname);
+
+ if (migrate_use_zero_copy_send()) {
+ error_setg(&err, "Zero copy send not available in migration");
+ }
+
+out:
migration_channel_connect(data->s, sioc, data->hostname, err);
object_unref(OBJECT(sioc));
}
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index 634968498b..55b48d3733 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -1309,6 +1309,12 @@ void hmp_migrate_set_parameter(Monitor *mon, const QDict *qdict)
p->has_multifd_zstd_level = true;
visit_type_uint8(v, param, &p->multifd_zstd_level, &err);
break;
+#ifdef CONFIG_LINUX
+ case MIGRATION_PARAMETER_ZERO_COPY_SEND:
+ p->has_zero_copy_send = true;
+ visit_type_bool(v, param, &p->zero_copy_send, &err);
+ break;
+#endif
case MIGRATION_PARAMETER_XBZRLE_CACHE_SIZE:
p->has_xbzrle_cache_size = true;
if (!visit_type_size(v, param, &cache_size, &err)) {
diff --git a/qapi/migration.json b/qapi/migration.json
index 27d7b28158..4d833ecdd6 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -741,6 +741,13 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+# When true, enables a zero-copy mechanism for sending
+# memory pages, if host supports it.
+# Requires that QEMU be permitted to use locked memory
+# for guest RAM pages.
+# Defaults to false. (Since 7.1)
+#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
# aliases may for example be the corresponding names on the
@@ -780,6 +787,7 @@
'xbzrle-cache-size', 'max-postcopy-bandwidth',
'max-cpu-throttle', 'multifd-compression',
'multifd-zlib-level' ,'multifd-zstd-level',
+ { 'name': 'zero-copy-send', 'if' : 'CONFIG_LINUX'},
'block-bitmap-mapping' ] }
##
@@ -906,6 +914,13 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+# When true, enables a zero-copy mechanism for sending
+# memory pages, if host supports it.
+# Requires that QEMU be permitted to use locked memory
+# for guest RAM pages.
+# Defaults to false. (Since 7.1)
+#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
# aliases may for example be the corresponding names on the
@@ -960,6 +975,7 @@
'*multifd-compression': 'MultiFDCompression',
'*multifd-zlib-level': 'uint8',
'*multifd-zstd-level': 'uint8',
+ '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
'*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
##
@@ -1106,6 +1122,13 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+# When true, enables a zero-copy mechanism for sending
+# memory pages, if host supports it.
+# Requires that QEMU be permitted to use locked memory
+# for guest RAM pages.
+# Defaults to false. (Since 7.1)
+#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
# aliases may for example be the corresponding names on the
@@ -1158,6 +1181,7 @@
'*multifd-compression': 'MultiFDCompression',
'*multifd-zlib-level': 'uint8',
'*multifd-zstd-level': 'uint8',
+ '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
'*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
##
--
2.35.3

98
kvm-migration-Allow-migrate-recover-to-run-multiple-time.patch Normal file
View File

@ -0,0 +1,98 @@
From fd6f516a94e635bc42e58448f314db575814a834 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Thu, 31 Mar 2022 11:08:45 -0400
Subject: [PATCH 18/18] migration: Allow migrate-recover to run multiple times
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 104: migration: Allow migrate-recover to run multiple times
RH-Commit: [1/1] afd726e54c069ae800e2d01f34e768d6bac7dcb9 (peterx/qemu-kvm)
RH-Bugzilla: 2096143
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Hanna Reitz <hreitz@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Previously migration didn't have an easy way to cleanup the listening
transport, migrate recovery only allows to execute once. That's done with a
trick flag in postcopy_recover_triggered.
Now the facility is already there.
Drop postcopy_recover_triggered and instead allows a new migrate-recover to
release the previous listener transport.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220331150857.74406-8-peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 08401c0426bc1a5ce4609afd1cda5dd39abbf9fa)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 13 ++-----------
migration/migration.h | 1 -
migration/savevm.c | 3 ---
3 files changed, 2 insertions(+), 15 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 2a141bfaf3..8fb3eae910 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -2166,11 +2166,8 @@ void qmp_migrate_recover(const char *uri, Error **errp)
return;
}
- if (qatomic_cmpxchg(&mis->postcopy_recover_triggered,
- false, true) == true) {
- error_setg(errp, "Migrate recovery is triggered already");
- return;
- }
+ /* If there's an existing transport, release it */
+ migration_incoming_transport_cleanup(mis);
/*
* Note that this call will never start a real migration; it will
@@ -2178,12 +2175,6 @@ void qmp_migrate_recover(const char *uri, Error **errp)
* to continue using that newly established channel.
*/
qemu_start_incoming_migration(uri, errp);
-
- /* Safe to dereference with the assert above */
- if (*errp) {
- /* Reset the flag so user could still retry */
- qatomic_set(&mis->postcopy_recover_triggered, false);
- }
}
void qmp_migrate_pause(Error **errp)
diff --git a/migration/migration.h b/migration/migration.h
index c2cabb8a14..fbc8690ec8 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -139,7 +139,6 @@ struct MigrationIncomingState {
struct PostcopyBlocktimeContext *blocktime_ctx;
/* notify PAUSED postcopy incoming migrations to try to continue */
- bool postcopy_recover_triggered;
QemuSemaphore postcopy_pause_sem_dst;
QemuSemaphore postcopy_pause_sem_fault;
diff --git a/migration/savevm.c b/migration/savevm.c
index 02ed94c180..d9076897b8 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -2589,9 +2589,6 @@ static bool postcopy_pause_incoming(MigrationIncomingState *mis)
assert(migrate_postcopy_ram());
- /* Clear the triggered bit to allow one recovery */
- mis->postcopy_recover_triggered = false;
-
/*
* Unregister yank with either from/to src would work, since ioc behind it
* is the same
--
2.35.3

289
kvm-migration-Change-zero_copy_send-from-migration-param.patch Normal file
View File

@ -0,0 +1,289 @@
From 7e2a037f3f349c21201152cecce32d8c8ff0bea0 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 20 Jun 2022 02:39:45 -0300
Subject: [PATCH 17/18] migration: Change zero_copy_send from migration
parameter to migration capability
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [11/11] e4a955607947896a49398ac8400241a0adac51a1 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
When originally implemented, zero_copy_send was designed as a Migration
paramenter.
But taking into account how is that supposed to work, and how
the difference between a capability and a parameter, it only makes sense
that zero-copy-send would work better as a capability.
Taking into account how recently the change got merged, it was decided
that it's still time to make it right, and convert zero_copy_send into
a Migration capability.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: always define the capability, even on non-Linux but error if
set; avoids build problems with the capability
(cherry picked from commit 1abaec9a1b2c23f7aa94709a422128d9e42c3e0b)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 58 +++++++++++++++++++------------------------
monitor/hmp-cmds.c | 6 -----
qapi/migration.json | 33 +++++++-----------------
3 files changed, 34 insertions(+), 63 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 102236fba0..2a141bfaf3 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -163,7 +163,8 @@ INITIALIZE_MIGRATE_CAPS_SET(check_caps_background_snapshot,
MIGRATION_CAPABILITY_COMPRESS,
MIGRATION_CAPABILITY_XBZRLE,
MIGRATION_CAPABILITY_X_COLO,
- MIGRATION_CAPABILITY_VALIDATE_UUID);
+ MIGRATION_CAPABILITY_VALIDATE_UUID,
+ MIGRATION_CAPABILITY_ZERO_COPY_SEND);
/* When we add fault tolerance, we could have several
migrations at once. For now we don't need to add
@@ -899,10 +900,6 @@ MigrationParameters *qmp_query_migrate_parameters(Error **errp)
params->multifd_zlib_level = s->parameters.multifd_zlib_level;
params->has_multifd_zstd_level = true;
params->multifd_zstd_level = s->parameters.multifd_zstd_level;
-#ifdef CONFIG_LINUX
- params->has_zero_copy_send = true;
- params->zero_copy_send = s->parameters.zero_copy_send;
-#endif
params->has_xbzrle_cache_size = true;
params->xbzrle_cache_size = s->parameters.xbzrle_cache_size;
params->has_max_postcopy_bandwidth = true;
@@ -1263,6 +1260,24 @@ 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() ||
+ migrate_use_tls())) {
+ error_setg(errp,
+ "Zero copy only available for non-compressed non-TLS multifd migration");
+ return false;
+ }
+#else
+ if (cap_list[MIGRATION_CAPABILITY_ZERO_COPY_SEND]) {
+ error_setg(errp,
+ "Zero copy currently only available on Linux");
+ return false;
+ }
+#endif
+
+
/* incoming side only */
if (runstate_check(RUN_STATE_INMIGRATE) &&
!migrate_multifd_is_allowed() &&
@@ -1485,16 +1500,6 @@ 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 (params->zero_copy_send &&
- (!migrate_use_multifd() ||
- params->multifd_compression != MULTIFD_COMPRESSION_NONE ||
- (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;
}
@@ -1568,11 +1573,6 @@ static void migrate_params_test_apply(MigrateSetParameters *params,
if (params->has_multifd_compression) {
dest->multifd_compression = params->multifd_compression;
}
-#ifdef CONFIG_LINUX
- if (params->has_zero_copy_send) {
- dest->zero_copy_send = params->zero_copy_send;
- }
-#endif
if (params->has_xbzrle_cache_size) {
dest->xbzrle_cache_size = params->xbzrle_cache_size;
}
@@ -1685,11 +1685,6 @@ static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
if (params->has_multifd_compression) {
s->parameters.multifd_compression = params->multifd_compression;
}
-#ifdef CONFIG_LINUX
- if (params->has_zero_copy_send) {
- s->parameters.zero_copy_send = params->zero_copy_send;
- }
-#endif
if (params->has_xbzrle_cache_size) {
s->parameters.xbzrle_cache_size = params->xbzrle_cache_size;
xbzrle_cache_resize(params->xbzrle_cache_size, errp);
@@ -2587,7 +2582,7 @@ bool migrate_use_zero_copy_send(void)
s = migrate_get_current();
- return s->parameters.zero_copy_send;
+ return s->enabled_capabilities[MIGRATION_CAPABILITY_ZERO_COPY_SEND];
}
#endif
@@ -4243,10 +4238,6 @@ static Property migration_properties[] = {
DEFINE_PROP_UINT8("multifd-zstd-level", MigrationState,
parameters.multifd_zstd_level,
DEFAULT_MIGRATE_MULTIFD_ZSTD_LEVEL),
-#ifdef CONFIG_LINUX
- DEFINE_PROP_BOOL("zero_copy_send", MigrationState,
- parameters.zero_copy_send, false),
-#endif
DEFINE_PROP_SIZE("xbzrle-cache-size", MigrationState,
parameters.xbzrle_cache_size,
DEFAULT_MIGRATE_XBZRLE_CACHE_SIZE),
@@ -4284,6 +4275,10 @@ static Property migration_properties[] = {
DEFINE_PROP_MIG_CAP("x-multifd", MIGRATION_CAPABILITY_MULTIFD),
DEFINE_PROP_MIG_CAP("x-background-snapshot",
MIGRATION_CAPABILITY_BACKGROUND_SNAPSHOT),
+#ifdef CONFIG_LINUX
+ DEFINE_PROP_MIG_CAP("x-zero-copy-send",
+ MIGRATION_CAPABILITY_ZERO_COPY_SEND),
+#endif
DEFINE_PROP_END_OF_LIST(),
};
@@ -4344,9 +4339,6 @@ static void migration_instance_init(Object *obj)
params->has_multifd_compression = true;
params->has_multifd_zlib_level = true;
params->has_multifd_zstd_level = true;
-#ifdef CONFIG_LINUX
- params->has_zero_copy_send = true;
-#endif
params->has_xbzrle_cache_size = true;
params->has_max_postcopy_bandwidth = true;
params->has_max_cpu_throttle = true;
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index 55b48d3733..634968498b 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -1309,12 +1309,6 @@ void hmp_migrate_set_parameter(Monitor *mon, const QDict *qdict)
p->has_multifd_zstd_level = true;
visit_type_uint8(v, param, &p->multifd_zstd_level, &err);
break;
-#ifdef CONFIG_LINUX
- case MIGRATION_PARAMETER_ZERO_COPY_SEND:
- p->has_zero_copy_send = true;
- visit_type_bool(v, param, &p->zero_copy_send, &err);
- break;
-#endif
case MIGRATION_PARAMETER_XBZRLE_CACHE_SIZE:
p->has_xbzrle_cache_size = true;
if (!visit_type_size(v, param, &cache_size, &err)) {
diff --git a/qapi/migration.json b/qapi/migration.json
index 4d833ecdd6..5105790cd0 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -463,6 +463,13 @@
# procedure starts. The VM RAM is saved with running VM.
# (since 6.0)
#
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+# When true, enables a zero-copy mechanism for sending
+# memory pages, if host supports it.
+# Requires that QEMU be permitted to use locked memory
+# for guest RAM pages.
+# (since 7.1)
+#
# Features:
# @unstable: Members @x-colo and @x-ignore-shared are experimental.
#
@@ -476,7 +483,8 @@
'block', 'return-path', 'pause-before-switchover', 'multifd',
'dirty-bitmaps', 'postcopy-blocktime', 'late-block-activate',
{ 'name': 'x-ignore-shared', 'features': [ 'unstable' ] },
- 'validate-uuid', 'background-snapshot'] }
+ 'validate-uuid', 'background-snapshot',
+ 'zero-copy-send'] }
##
# @MigrationCapabilityStatus:
@@ -741,12 +749,6 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
-# @zero-copy-send: Controls behavior on sending memory pages on migration.
-# When true, enables a zero-copy mechanism for sending
-# memory pages, if host supports it.
-# Requires that QEMU be permitted to use locked memory
-# for guest RAM pages.
-# Defaults to false. (Since 7.1)
#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
@@ -787,7 +789,6 @@
'xbzrle-cache-size', 'max-postcopy-bandwidth',
'max-cpu-throttle', 'multifd-compression',
'multifd-zlib-level' ,'multifd-zstd-level',
- { 'name': 'zero-copy-send', 'if' : 'CONFIG_LINUX'},
'block-bitmap-mapping' ] }
##
@@ -914,13 +915,6 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
-# @zero-copy-send: Controls behavior on sending memory pages on migration.
-# When true, enables a zero-copy mechanism for sending
-# memory pages, if host supports it.
-# Requires that QEMU be permitted to use locked memory
-# for guest RAM pages.
-# Defaults to false. (Since 7.1)
-#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
# aliases may for example be the corresponding names on the
@@ -975,7 +969,6 @@
'*multifd-compression': 'MultiFDCompression',
'*multifd-zlib-level': 'uint8',
'*multifd-zstd-level': 'uint8',
- '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
'*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
##
@@ -1122,13 +1115,6 @@
# will consume more CPU.
# Defaults to 1. (Since 5.0)
#
-# @zero-copy-send: Controls behavior on sending memory pages on migration.
-# When true, enables a zero-copy mechanism for sending
-# memory pages, if host supports it.
-# Requires that QEMU be permitted to use locked memory
-# for guest RAM pages.
-# Defaults to false. (Since 7.1)
-#
# @block-bitmap-mapping: Maps block nodes and bitmaps on them to
# aliases for the purpose of dirty bitmap migration. Such
# aliases may for example be the corresponding names on the
@@ -1181,7 +1167,6 @@
'*multifd-compression': 'MultiFDCompression',
'*multifd-zlib-level': 'uint8',
'*multifd-zstd-level': 'uint8',
- '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
'*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
##
--
2.35.3

182
kvm-multifd-Implement-zero-copy-write-in-multifd-migrati.patch Normal file
View File

@ -0,0 +1,182 @@
From c1a2866d158ac67179fa0d17f1710302eb9a3866 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:37 -0300
Subject: [PATCH 14/18] multifd: Implement zero copy write in multifd migration
(multifd-zero-copy)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [8/11] b93009cc94b2cc4b464b4f68ebfb37b870dd6f7d (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Implement zero copy send on nocomp_send_write(), by making use of QIOChannel
writev + flags & flush interface.
Change multifd_send_sync_main() so flush_zero_copy() can be called
after each iteration in order to make sure all dirty pages are sent before
a new iteration is started. It will also flush at the beginning and at the
end of migration.
Also make it return -1 if flush_zero_copy() fails, in order to cancel
the migration process, and avoid resuming the guest in the target host
without receiving all current RAM.
This will work fine on RAM migration because the RAM pages are not usually freed,
and there is no problem on changing the pages content between writev_zero_copy() and
the actual sending of the buffer, because this change will dirty the page and
cause it to be re-sent on a next iteration anyway.
A lot of locked memory may be needed in order to use multifd migration
with zero-copy enabled, so disabling the feature should be necessary for
low-privileged users trying to perform multifd migrations.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220513062836.965425-9-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 5b1d9bab2da4fca3a3caee97c430e5709cb32b7b)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 11 ++++++++++-
migration/multifd.c | 37 +++++++++++++++++++++++++++++++++++--
migration/multifd.h | 2 ++
migration/socket.c | 5 +++--
4 files changed, 50 insertions(+), 5 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index d91efb66fe..102236fba0 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1485,7 +1485,16 @@ 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 (params->zero_copy_send &&
+ (!migrate_use_multifd() ||
+ params->multifd_compression != MULTIFD_COMPRESSION_NONE ||
+ (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;
}
diff --git a/migration/multifd.c b/migration/multifd.c
index 8fca6c970e..0b5b41c53f 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -571,6 +571,7 @@ void multifd_save_cleanup(void)
int multifd_send_sync_main(QEMUFile *f)
{
int i;
+ bool flush_zero_copy;
if (!migrate_use_multifd()) {
return 0;
@@ -581,6 +582,20 @@ int multifd_send_sync_main(QEMUFile *f)
return -1;
}
}
+
+ /*
+ * When using zero-copy, it's necessary to flush the pages before any of
+ * the pages can be sent again, so we'll make sure the new version of the
+ * pages will always arrive _later_ than the old pages.
+ *
+ * Currently we achieve this by flushing the zero-page requested writes
+ * per ram iteration, but in the future we could potentially optimize it
+ * to be less frequent, e.g. only after we finished one whole scanning of
+ * all the dirty bitmaps.
+ */
+
+ flush_zero_copy = migrate_use_zero_copy_send();
+
for (i = 0; i < migrate_multifd_channels(); i++) {
MultiFDSendParams *p = &multifd_send_state->params[i];
@@ -602,6 +617,17 @@ int multifd_send_sync_main(QEMUFile *f)
ram_counters.transferred += p->packet_len;
qemu_mutex_unlock(&p->mutex);
qemu_sem_post(&p->sem);
+
+ if (flush_zero_copy && p->c) {
+ int ret;
+ Error *err = NULL;
+
+ ret = qio_channel_flush(p->c, &err);
+ if (ret < 0) {
+ error_report_err(err);
+ return -1;
+ }
+ }
}
for (i = 0; i < migrate_multifd_channels(); i++) {
MultiFDSendParams *p = &multifd_send_state->params[i];
@@ -686,8 +712,8 @@ static void *multifd_send_thread(void *opaque)
p->iov[0].iov_base = p->packet;
}
- ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
- &local_err);
+ ret = qio_channel_writev_full_all(p->c, p->iov, p->iovs_num, NULL,
+ 0, p->write_flags, &local_err);
if (ret != 0) {
break;
}
@@ -928,6 +954,13 @@ int multifd_save_setup(Error **errp)
/* We need one extra place for the packet header */
p->iov = g_new0(struct iovec, page_count + 1);
p->normal = g_new0(ram_addr_t, page_count);
+
+ if (migrate_use_zero_copy_send()) {
+ p->write_flags = QIO_CHANNEL_WRITE_FLAG_ZERO_COPY;
+ } else {
+ p->write_flags = 0;
+ }
+
socket_send_channel_create(multifd_new_send_channel_async, p);
}
diff --git a/migration/multifd.h b/migration/multifd.h
index cd495195ce..7ec688fb4f 100644
--- a/migration/multifd.h
+++ b/migration/multifd.h
@@ -96,6 +96,8 @@ typedef struct {
uint32_t packet_len;
/* pointer to the packet */
MultiFDPacket_t *packet;
+ /* multifd flags for sending ram */
+ int write_flags;
/* multifd flags for each packet */
uint32_t flags;
/* size of the next packet that contains pages */
diff --git a/migration/socket.c b/migration/socket.c
index 3754d8f72c..4fd5e85f50 100644
--- a/migration/socket.c
+++ b/migration/socket.c
@@ -79,8 +79,9 @@ static void socket_outgoing_migration(QIOTask *task,
trace_migration_socket_outgoing_connected(data->hostname);
- if (migrate_use_zero_copy_send()) {
- error_setg(&err, "Zero copy send not available in migration");
+ if (migrate_use_zero_copy_send() &&
+ !qio_channel_has_feature(sioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
+ error_setg(&err, "Zero copy send feature not detected in host kernel");
}
out:
--
2.35.3

102
kvm-multifd-Send-header-packet-without-flags-if-zero-cop.patch Normal file
View File

@ -0,0 +1,102 @@
From 63255c13492f42a3236d96e706e5f8e70bb4e219 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:36 -0300
Subject: [PATCH 13/18] multifd: Send header packet without flags if
zero-copy-send is enabled
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [7/11] 137eea685e387d3d6aff187ec3fcac05bc16b6e3 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Since d48c3a0445 ("multifd: Use a single writev on the send side"),
sending the header packet and the memory pages happens in the same
writev, which can potentially make the migration faster.
Using channel-socket as example, this works well with the default copying
mechanism of sendmsg(), but with zero-copy-send=true, it will cause
the migration to often break.
This happens because the header packet buffer gets reused quite often,
and there is a high chance that by the time the MSG_ZEROCOPY mechanism get
to send the buffer, it has already changed, sending the wrong data and
causing the migration to abort.
It means that, as it is, the buffer for the header packet is not suitable
for sending with MSG_ZEROCOPY.
In order to enable zero copy for multifd, send the header packet on an
individual write(), without any flags, and the remanining pages with a
writev(), as it was happening before. This only changes how a migration
with zero-copy-send=true works, not changing any current behavior for
migrations with zero-copy-send=false.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220513062836.965425-8-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit b7dbdd8e76cd03453c234dbb9578d20969859d74)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/multifd.c | 22 +++++++++++++++++++---
1 file changed, 19 insertions(+), 3 deletions(-)
diff --git a/migration/multifd.c b/migration/multifd.c
index cdb57439a7..8fca6c970e 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -619,6 +619,7 @@ static void *multifd_send_thread(void *opaque)
MultiFDSendParams *p = opaque;
Error *local_err = NULL;
int ret = 0;
+ bool use_zero_copy_send = migrate_use_zero_copy_send();
trace_multifd_send_thread_start(p->id);
rcu_register_thread();
@@ -641,9 +642,14 @@ static void *multifd_send_thread(void *opaque)
if (p->pending_job) {
uint64_t packet_num = p->packet_num;
uint32_t flags = p->flags;
- p->iovs_num = 1;
p->normal_num = 0;
+ if (use_zero_copy_send) {
+ p->iovs_num = 0;
+ } else {
+ p->iovs_num = 1;
+ }
+
for (int i = 0; i < p->pages->num; i++) {
p->normal[p->normal_num] = p->pages->offset[i];
p->normal_num++;
@@ -667,8 +673,18 @@ static void *multifd_send_thread(void *opaque)
trace_multifd_send(p->id, packet_num, p->normal_num, flags,
p->next_packet_size);
- p->iov[0].iov_len = p->packet_len;
- p->iov[0].iov_base = p->packet;
+ if (use_zero_copy_send) {
+ /* Send header first, without zerocopy */
+ ret = qio_channel_write_all(p->c, (void *)p->packet,
+ p->packet_len, &local_err);
+ if (ret != 0) {
+ break;
+ }
+ } else {
+ /* Send header using the same writev call */
+ p->iov[0].iov_len = p->packet_len;
+ p->iov[0].iov_base = p->packet;
+ }
ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
&local_err);
--
2.35.3

163
kvm-multifd-multifd_send_sync_main-now-returns-negative-.patch Normal file
View File

@ -0,0 +1,163 @@
From 4ca5375a936bc87829c6e2b4620f56c73a5efc70 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:35 -0300
Subject: [PATCH 12/18] multifd: multifd_send_sync_main now returns negative on
error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [6/11] c8ebdee4327d463c74f4b2eeb42d3c964f314c94 (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Even though multifd_send_sync_main() currently emits error_reports, it's
callers don't really check it before continuing.
Change multifd_send_sync_main() to return -1 on error and 0 on success.
Also change all it's callers to make use of this change and possibly fail
earlier.
(This change is important to next patch on multifd zero copy
implementation, to make it sure an error in zero-copy flush does not go
unnoticed.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220513062836.965425-7-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 33d70973a3a6e8c6b62bcbc64d9e488961981007)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/multifd.c | 10 ++++++----
migration/multifd.h | 2 +-
migration/ram.c | 29 ++++++++++++++++++++++-------
3 files changed, 29 insertions(+), 12 deletions(-)
diff --git a/migration/multifd.c b/migration/multifd.c
index 43998ad117..cdb57439a7 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -568,17 +568,17 @@ void multifd_save_cleanup(void)
multifd_send_state = NULL;
}
-void multifd_send_sync_main(QEMUFile *f)
+int multifd_send_sync_main(QEMUFile *f)
{
int i;
if (!migrate_use_multifd()) {
- return;
+ return 0;
}
if (multifd_send_state->pages->num) {
if (multifd_send_pages(f) < 0) {
error_report("%s: multifd_send_pages fail", __func__);
- return;
+ return -1;
}
}
for (i = 0; i < migrate_multifd_channels(); i++) {
@@ -591,7 +591,7 @@ void multifd_send_sync_main(QEMUFile *f)
if (p->quit) {
error_report("%s: channel %d has already quit", __func__, i);
qemu_mutex_unlock(&p->mutex);
- return;
+ return -1;
}
p->packet_num = multifd_send_state->packet_num++;
@@ -610,6 +610,8 @@ void multifd_send_sync_main(QEMUFile *f)
qemu_sem_wait(&p->sem_sync);
}
trace_multifd_send_sync_main(multifd_send_state->packet_num);
+
+ return 0;
}
static void *multifd_send_thread(void *opaque)
diff --git a/migration/multifd.h b/migration/multifd.h
index 4dda900a0b..cd495195ce 100644
--- a/migration/multifd.h
+++ b/migration/multifd.h
@@ -22,7 +22,7 @@ int multifd_load_cleanup(Error **errp);
bool multifd_recv_all_channels_created(void);
bool multifd_recv_new_channel(QIOChannel *ioc, Error **errp);
void multifd_recv_sync_main(void);
-void multifd_send_sync_main(QEMUFile *f);
+int multifd_send_sync_main(QEMUFile *f);
int multifd_queue_page(QEMUFile *f, RAMBlock *block, ram_addr_t offset);
/* Multifd Compression flags */
diff --git a/migration/ram.c b/migration/ram.c
index 0ef4bd63eb..fb6db54642 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -2903,6 +2903,7 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
{
RAMState **rsp = opaque;
RAMBlock *block;
+ int ret;
if (compress_threads_save_setup()) {
return -1;
@@ -2937,7 +2938,11 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
ram_control_before_iterate(f, RAM_CONTROL_SETUP);
ram_control_after_iterate(f, RAM_CONTROL_SETUP);
- multifd_send_sync_main(f);
+ ret = multifd_send_sync_main(f);
+ if (ret < 0) {
+ return ret;
+ }
+
qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
qemu_fflush(f);
@@ -3046,7 +3051,11 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
out:
if (ret >= 0
&& migration_is_setup_or_active(migrate_get_current()->state)) {
- multifd_send_sync_main(rs->f);
+ ret = multifd_send_sync_main(rs->f);
+ if (ret < 0) {
+ return ret;
+ }
+
qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
qemu_fflush(f);
ram_transferred_add(8);
@@ -3106,13 +3115,19 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
ram_control_after_iterate(f, RAM_CONTROL_FINISH);
}
- if (ret >= 0) {
- multifd_send_sync_main(rs->f);
- qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
- qemu_fflush(f);
+ if (ret < 0) {
+ return ret;
}
- return ret;
+ ret = multifd_send_sync_main(rs->f);
+ if (ret < 0) {
+ return ret;
+ }
+
+ qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
+ qemu_fflush(f);
+
+ return 0;
}
static void ram_save_pending(QEMUFile *f, void *opaque, uint64_t max_size,
--
2.35.3

129
kvm-target-arm-deprecate-named-CPU-models.patch Normal file
View File

@ -0,0 +1,129 @@
From 1f8528b71d96c01dd6106f11681f4a4e2776ef5f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 21 Mar 2022 12:05:42 +0000
Subject: [PATCH 06/18] target/arm: deprecate named CPU models
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [6/6] afddeb9e898206fd04499f01c48caf7dc1a8b8ef (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
KVM requires use of the 'host' CPU model, so named CPU models are only
needed for TCG. Since we don't consider TCG to be supported we can
deprecate all the named CPU models. TCG users can rely on 'max' model.
Note: this has the effect of deprecating the default built-in CPU
model 'cortex-a57'. Applications using QEMU are expected to make an
explicit choice about which CPU model they want, since no builtin
default can suit all purposes.
https://bugzilla.redhat.com/show_bug.cgi?id=2060839
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
target/arm/cpu-qom.h | 1 +
target/arm/cpu.c | 5 +++++
target/arm/cpu.h | 2 ++
target/arm/cpu64.c | 8 +++++++-
target/arm/helper.c | 2 ++
5 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/target/arm/cpu-qom.h b/target/arm/cpu-qom.h
index 64c44cef2d..82e97249bc 100644
--- a/target/arm/cpu-qom.h
+++ b/target/arm/cpu-qom.h
@@ -35,6 +35,7 @@ typedef struct ARMCPUInfo {
const char *name;
void (*initfn)(Object *obj);
void (*class_init)(ObjectClass *oc, void *data);
+ const char *deprecation_note;
} ARMCPUInfo;
void arm_cpu_register(const ARMCPUInfo *info);
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index 5d4ca7a227..c74b0fb462 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -2105,8 +2105,13 @@ static void arm_cpu_instance_init(Object *obj)
static void cpu_register_class_init(ObjectClass *oc, void *data)
{
ARMCPUClass *acc = ARM_CPU_CLASS(oc);
+ CPUClass *cc = CPU_CLASS(oc);
acc->info = data;
+
+ if (acc->info->deprecation_note) {
+ cc->deprecation_note = acc->info->deprecation_note;
+ }
}
void arm_cpu_register(const ARMCPUInfo *info)
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 23879de5fa..c0c9f680e5 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -33,6 +33,8 @@
#define KVM_HAVE_MCE_INJECTION 1
#endif
+#define RHEL_CPU_DEPRECATION "use 'host' / 'max'"
+
#define EXCP_UDEF 1 /* undefined instruction */
#define EXCP_SWI 2 /* software interrupt */
#define EXCP_PREFETCH_ABORT 3
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index e80b831073..c8f152891c 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -975,7 +975,8 @@ static void aarch64_a64fx_initfn(Object *obj)
#endif /* disabled for RHEL */
static const ARMCPUInfo aarch64_cpus[] = {
- { .name = "cortex-a57", .initfn = aarch64_a57_initfn },
+ { .name = "cortex-a57", .initfn = aarch64_a57_initfn,
+ .deprecation_note = RHEL_CPU_DEPRECATION },
#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "cortex-a53", .initfn = aarch64_a53_initfn },
{ .name = "cortex-a72", .initfn = aarch64_a72_initfn },
@@ -1052,8 +1053,13 @@ static void aarch64_cpu_instance_init(Object *obj)
static void cpu_register_class_init(ObjectClass *oc, void *data)
{
ARMCPUClass *acc = ARM_CPU_CLASS(oc);
+ CPUClass *cc = CPU_CLASS(oc);
acc->info = data;
+
+ if (acc->info->deprecation_note) {
+ cc->deprecation_note = acc->info->deprecation_note;
+ }
}
void aarch64_cpu_register(const ARMCPUInfo *info)
diff --git a/target/arm/helper.c b/target/arm/helper.c
index 7d14650615..3d34f63e49 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -8560,6 +8560,7 @@ void arm_cpu_list(void)
static void arm_cpu_add_definition(gpointer data, gpointer user_data)
{
ObjectClass *oc = data;
+ CPUClass *cc = CPU_CLASS(oc);
CpuDefinitionInfoList **cpu_list = user_data;
CpuDefinitionInfo *info;
const char *typename;
@@ -8569,6 +8570,7 @@ static void arm_cpu_add_definition(gpointer data, gpointer user_data)
info->name = g_strndup(typename,
strlen(typename) - strlen("-" TYPE_ARM_CPU));
info->q_typename = g_strdup(typename);
+ info->deprecated = !!cc->deprecation_note;
QAPI_LIST_PREPEND(*cpu_list, info);
}
--
2.35.3

273
kvm-target-i386-deprecate-CPUs-older-than-x86_64-v2-ABI.patch Normal file
View File

@ -0,0 +1,273 @@
From 577b04770e47aed0f88acb4a415ed04ddbe087f1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Thu, 17 Mar 2022 17:59:22 +0000
Subject: [PATCH 04/18] target/i386: deprecate CPUs older than x86_64-v2 ABI
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [4/6] 71f6043f11b31ffa841a2e14d24972e571c18a9e (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RHEL-9 is compiled with the x86_64-v2 ABI. We use this as a baseline to
select which CPUs we want to support, such that there is at least one
supported guest CPU that can be launched for every physical machine
capable of running RHEL-9 KVM.
Supported CPUs:
* QEMU models
base (QEMU internal)
host (host passthrough)
max (host passthrough for KVM,
all emulated features for TCG)
* Intel models
Icelake-Server
Icelake-Server-noTSX
Cascadelake-Server (2019)
Cascadelake-Server-noTSX (2019)
Skylake-Server (2016)
Skylake-Server-IBRS (2016)
Skylake-Server-noTSX-IBRS (2016)
Skylake-Client (2015)
Skylake-Client-IBRS (2015)
Skylake-Client-noTSX-IBRS (2015)
Broadwell (2014)
Broadwell-IBRS (2014)
Broadwell-noTSX (2014)
Broadwell-noTSX-IBRS (2014)
Haswell (2013)
Haswell-IBRS (2013)
Haswell-noTSX (2013)
Haswell-noTSX-IBRS (2013)
IvyBridge (2012)
IvyBridge-IBRS (2012)
SandyBridge (2011)
SandyBridge-IBRS (2011)
Westmere (2010)
Westmere-IBRS (2010)
Nehalem (2008)
Nehalem-IBRS (2008)
Cooperlake (2020)
Snowridge (2019)
KnightsMill (2017)
Denverton (2016)
* AMD models
EPYC-Milan (2021)
EPYC-Rome (2019)
EPYC (2017)
EPYC-IBPB (2017)
Opteron_G5 (2012)
Opteron_G4 (2011)
* Other
Dhyana (2018)
(I've omitted the many -vNNN versions for brevity)
Deprecated CPUs:
486
athlon
Conroe
core2duo
coreduo
Icelake-Client (already deprecated upstream)
Icelake-Client-noTSX (already deprecated upstream)
kvm32
kvm64
n270
Opteron_G1
Opteron_G2
Opteron_G3
Penryn
pentium2
pentium3
pentium
phenom
qemu32
qemu64
The deprecated CPU models are subject to removal in a future
major version of RHEL.
Note: this has the effect of deprecating the default built-in CPU
model 'qemu64'. Applications using QEMU are expected to make an
explicit choice about which CPU model they want, since no builtin
default can suit all purposes.
https://bugzilla.redhat.com/show_bug.cgi?id=2060839
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
target/i386/cpu.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index cb6b5467d0..87cb641b5f 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -1780,9 +1780,13 @@ static const CPUCaches epyc_milan_cache_info = {
* PT in VMX operation
*/
+#define RHEL_CPU_DEPRECATION \
+ "use at least 'Nehalem' / 'Opteron_G4', or 'host' / 'max'"
+
static const X86CPUDefinition builtin_x86_defs[] = {
{
.name = "qemu64",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 0xd,
.vendor = CPUID_VENDOR_AMD,
.family = 15,
@@ -1803,6 +1807,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "phenom",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 5,
.vendor = CPUID_VENDOR_AMD,
.family = 16,
@@ -1835,6 +1840,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "core2duo",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 10,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -1877,6 +1883,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "kvm64",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 0xd,
.vendor = CPUID_VENDOR_INTEL,
.family = 15,
@@ -1918,6 +1925,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "qemu32",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 4,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -1932,6 +1940,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "kvm32",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 5,
.vendor = CPUID_VENDOR_INTEL,
.family = 15,
@@ -1962,6 +1971,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "coreduo",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 10,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -1995,6 +2005,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "486",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 1,
.vendor = CPUID_VENDOR_INTEL,
.family = 4,
@@ -2007,6 +2018,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "pentium",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 1,
.vendor = CPUID_VENDOR_INTEL,
.family = 5,
@@ -2019,6 +2031,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "pentium2",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 2,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -2031,6 +2044,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "pentium3",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 3,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -2043,6 +2057,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "athlon",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 2,
.vendor = CPUID_VENDOR_AMD,
.family = 6,
@@ -2058,6 +2073,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "n270",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 10,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -2083,6 +2099,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "Conroe",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 10,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -2123,6 +2140,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "Penryn",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 10,
.vendor = CPUID_VENDOR_INTEL,
.family = 6,
@@ -3832,6 +3850,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "Opteron_G1",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 5,
.vendor = CPUID_VENDOR_AMD,
.family = 15,
@@ -3852,6 +3871,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "Opteron_G2",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 5,
.vendor = CPUID_VENDOR_AMD,
.family = 15,
@@ -3874,6 +3894,7 @@ static const X86CPUDefinition builtin_x86_defs[] = {
},
{
.name = "Opteron_G3",
+ .deprecation_note = RHEL_CPU_DEPRECATION,
.level = 5,
.vendor = CPUID_VENDOR_AMD,
.family = 16,
--
2.35.3

194
kvm-target-s390x-deprecate-CPUs-older-than-z14.patch Normal file
View File

@ -0,0 +1,194 @@
From 8459c305914e2a7a19dcd1662d54a89def7acfa6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Thu, 17 Mar 2022 17:59:22 +0000
Subject: [PATCH 05/18] target/s390x: deprecate CPUs older than z14
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [5/6] 2da9e06cf452287673f94f880a7eb8b2b37b7278 (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RHEL-9 is compiled with the z14 ABI. We use this as a baseline to
select which CPUs we want to support, such that there is at least one
supported guest CPU that can be launched for every physical
machine capable of running RHEL-9 KVM.
Supported CPUs:
gen15a-base
gen15a
gen15b-base
gen15b
gen16a-base
gen16a
gen16b-base
gen16b
max
qemu
z14.2-base
z14.2
z14-base
z14
z14ZR1-base
z14ZR1
Deprecated CPUs:
z10BC.2-base
z10BC.2
z10BC-base
z10BC
z10EC.2-base
z10EC.2
z10EC.3-base
z10EC.3
z10EC-base
z10EC
z114-base
z114
z13.2-base
z13.2
z13-base
z13s-base
z13s
z13
z196.2-base
z196.2
z196-base
z196
z800-base
z800
z890.2-base
z890.2
z890.3-base
z890.3
z890-base
z890
z900.2-base
z900.2
z900.3-base
z900.3
z900-base
z900
z990.2-base
z990.2
z990.3-base
z990.3
z990.4-base
z990.4
z990.5-base
z990.5
z990-base
z990
z9BC.2-base
z9BC.2
z9BC-base
z9BC
z9EC.2-base
z9EC.2
z9EC.3-base
z9EC.3
z9EC-base
z9EC
zBC12-base
zBC12
zEC12.2-base
zEC12.2
zEC12-base
zEC12
https://bugzilla.redhat.com/show_bug.cgi?id=2060839
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
target/s390x/cpu_models.c | 11 +++++++++++
target/s390x/cpu_models.h | 2 ++
target/s390x/cpu_models_sysemu.c | 2 ++
3 files changed, 15 insertions(+)
diff --git a/target/s390x/cpu_models.c b/target/s390x/cpu_models.c
index 6d71428056..9b9fc41676 100644
--- a/target/s390x/cpu_models.c
+++ b/target/s390x/cpu_models.c
@@ -45,6 +45,9 @@
* of a following release have been a superset of the previous release. With
* generation 15 one base feature and one optional feature have been deprecated.
*/
+
+#define RHEL_CPU_DEPRECATION "use at least 'z14', or 'host' / 'qemu' / 'max'"
+
static S390CPUDef s390_cpu_defs[] = {
CPUDEF_INIT(0x2064, 7, 1, 38, 0x00000000U, "z900", "IBM zSeries 900 GA1"),
CPUDEF_INIT(0x2064, 7, 2, 38, 0x00000000U, "z900.2", "IBM zSeries 900 GA2"),
@@ -852,22 +855,30 @@ static void s390_host_cpu_model_class_init(ObjectClass *oc, void *data)
static void s390_base_cpu_model_class_init(ObjectClass *oc, void *data)
{
S390CPUClass *xcc = S390_CPU_CLASS(oc);
+ CPUClass *cc = CPU_CLASS(oc);
/* all base models are migration safe */
xcc->cpu_def = (const S390CPUDef *) data;
xcc->is_migration_safe = true;
xcc->is_static = true;
xcc->desc = xcc->cpu_def->desc;
+ if (xcc->cpu_def->gen < 14) {
+ cc->deprecation_note = RHEL_CPU_DEPRECATION;
+ }
}
static void s390_cpu_model_class_init(ObjectClass *oc, void *data)
{
S390CPUClass *xcc = S390_CPU_CLASS(oc);
+ CPUClass *cc = CPU_CLASS(oc);
/* model that can change between QEMU versions */
xcc->cpu_def = (const S390CPUDef *) data;
xcc->is_migration_safe = true;
xcc->desc = xcc->cpu_def->desc;
+ if (xcc->cpu_def->gen < 14) {
+ cc->deprecation_note = RHEL_CPU_DEPRECATION;
+ }
}
static void s390_qemu_cpu_model_class_init(ObjectClass *oc, void *data)
diff --git a/target/s390x/cpu_models.h b/target/s390x/cpu_models.h
index 74d1f87e4f..372160bcd7 100644
--- a/target/s390x/cpu_models.h
+++ b/target/s390x/cpu_models.h
@@ -38,6 +38,8 @@ struct S390CPUDef {
S390FeatBitmap full_feat;
/* used to init full_feat from generated data */
S390FeatInit full_init;
+ /* if deprecated, provides a suggestion */
+ const char *deprecation_note;
};
/* CPU model based on a CPU definition */
diff --git a/target/s390x/cpu_models_sysemu.c b/target/s390x/cpu_models_sysemu.c
index 6a04ccab1b..f3b7c304ec 100644
--- a/target/s390x/cpu_models_sysemu.c
+++ b/target/s390x/cpu_models_sysemu.c
@@ -61,6 +61,7 @@ static void create_cpu_model_list(ObjectClass *klass, void *opaque)
CpuDefinitionInfo *info;
char *name = g_strdup(object_class_get_name(klass));
S390CPUClass *scc = S390_CPU_CLASS(klass);
+ CPUClass *cc = CPU_CLASS(klass);
/* strip off the -s390x-cpu */
g_strrstr(name, "-" TYPE_S390_CPU)[0] = 0;
@@ -70,6 +71,7 @@ static void create_cpu_model_list(ObjectClass *klass, void *opaque)
info->migration_safe = scc->is_migration_safe;
info->q_static = scc->is_static;
info->q_typename = g_strdup(object_class_get_name(klass));
+ info->deprecated = !!cc->deprecation_note;
/* check for unavailable features */
if (cpu_list_data->model) {
Object *obj;
--
2.35.3

157
kvm-tests-avocado-update-aarch64_virt-test-to-exercise-c.patch Normal file
View File

@ -0,0 +1,157 @@
From f52aa60217634c96fef59ce76b803a94610bf5c8 Mon Sep 17 00:00:00 2001
From: Andrew Jones <drjones@redhat.com>
Date: Wed, 15 Jun 2022 15:28:27 +0200
Subject: [PATCH 01/18] tests/avocado: update aarch64_virt test to exercise
-cpu max
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Daniel P. Berrangé <berrange@redhat.com>
RH-MergeRequest: 94: i386, aarch64, s390x: deprecate many named CPU models
RH-Commit: [1/6] df6839e567180a4c32afd98852f68b2279e00f7c (berrange/centos-src-qemu)
RH-Bugzilla: 2060839
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2066824
commit 11593544df6f8febb3ce87015c22b429bf43c4c7
Author: Alex Bennée <alex.bennee@linaro.org>
Date: Tue Apr 19 10:09:56 2022 +0100
tests/avocado: update aarch64_virt test to exercise -cpu max
The Fedora 29 kernel is quite old and importantly fails when running
in LPA2 scenarios. As it's not really exercising much of the CPU space
replace it with a custom 5.16.12 kernel with all the architecture
options turned on. There is a minimal buildroot initramfs included in
the kernel which has a few tools for stress testing the memory
subsystem. The userspace also targets the Neoverse N1 processor so
would fail with a v8.0 cpu like cortex-a53.
While we are at it move the test into its own file so it can have an
assigned maintainer.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20220419091020.3008144-2-alex.bennee@linaro.org>
Signed-off-by: Andrew Jones <drjones@redhat.com>
---
MAINTAINERS | 1 +
tests/avocado/boot_linux_console.py | 25 -------------
tests/avocado/machine_aarch64_virt.py | 51 +++++++++++++++++++++++++++
3 files changed, 52 insertions(+), 25 deletions(-)
create mode 100644 tests/avocado/machine_aarch64_virt.py
diff --git a/MAINTAINERS b/MAINTAINERS
index 2fe20a49ab..bfe8806f60 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -942,6 +942,7 @@ S: Maintained
F: hw/arm/virt*
F: include/hw/arm/virt.h
F: docs/system/arm/virt.rst
+F: tests/avocado/machine_aarch64_virt.py
Xilinx Zynq
M: Edgar E. Iglesias <edgar.iglesias@gmail.com>
diff --git a/tests/avocado/boot_linux_console.py b/tests/avocado/boot_linux_console.py
index b40a3abc81..45a2ceda22 100644
--- a/tests/avocado/boot_linux_console.py
+++ b/tests/avocado/boot_linux_console.py
@@ -325,31 +325,6 @@ def test_mips_malta32el_nanomips_64k_dbg(self):
kernel_hash = '18d1c68f2e23429e266ca39ba5349ccd0aeb7180'
self.do_test_mips_malta32el_nanomips(kernel_url, kernel_hash)
- def test_aarch64_virt(self):
- """
- :avocado: tags=arch:aarch64
- :avocado: tags=machine:virt
- :avocado: tags=accel:tcg
- :avocado: tags=cpu:cortex-a53
- """
- kernel_url = ('https://archives.fedoraproject.org/pub/archive/fedora'
- '/linux/releases/29/Everything/aarch64/os/images/pxeboot'
- '/vmlinuz')
- kernel_hash = '8c73e469fc6ea06a58dc83a628fc695b693b8493'
- kernel_path = self.fetch_asset(kernel_url, asset_hash=kernel_hash)
-
- self.vm.set_console()
- kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE +
- 'console=ttyAMA0')
- self.require_accelerator("tcg")
- self.vm.add_args('-cpu', 'cortex-a53',
- '-accel', 'tcg',
- '-kernel', kernel_path,
- '-append', kernel_command_line)
- self.vm.launch()
- console_pattern = 'Kernel command line: %s' % kernel_command_line
- self.wait_for_console_pattern(console_pattern)
-
def test_aarch64_xlnx_versal_virt(self):
"""
:avocado: tags=arch:aarch64
diff --git a/tests/avocado/machine_aarch64_virt.py b/tests/avocado/machine_aarch64_virt.py
new file mode 100644
index 0000000000..21848cba70
--- /dev/null
+++ b/tests/avocado/machine_aarch64_virt.py
@@ -0,0 +1,51 @@
+# Functional test that boots a Linux kernel and checks the console
+#
+# Copyright (c) 2022 Linaro Ltd.
+#
+# Author:
+# Alex Bennée <alex.bennee@linaro.org>
+#
+# SPDX-License-Identifier: GPL-2.0-or-later
+
+import time
+
+from avocado_qemu import QemuSystemTest
+from avocado_qemu import wait_for_console_pattern
+from avocado_qemu import exec_command
+
+class Aarch64VirtMachine(QemuSystemTest):
+ KERNEL_COMMON_COMMAND_LINE = 'printk.time=0 '
+
+ def wait_for_console_pattern(self, success_message, vm=None):
+ wait_for_console_pattern(self, success_message,
+ failure_message='Kernel panic - not syncing',
+ vm=vm)
+
+ def test_aarch64_virt(self):
+ """
+ :avocado: tags=arch:aarch64
+ :avocado: tags=machine:virt
+ :avocado: tags=accel:tcg
+ :avocado: tags=cpu:max
+ """
+ kernel_url = ('https://fileserver.linaro.org/s/'
+ 'z6B2ARM7DQT3HWN/download')
+
+ kernel_hash = 'ed11daab50c151dde0e1e9c9cb8b2d9bd3215347'
+ kernel_path = self.fetch_asset(kernel_url, asset_hash=kernel_hash)
+
+ self.vm.set_console()
+ kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE +
+ 'console=ttyAMA0')
+ self.require_accelerator("tcg")
+ self.vm.add_args('-cpu', 'max,pauth-impdef=on',
+ '-accel', 'tcg',
+ '-kernel', kernel_path,
+ '-append', kernel_command_line)
+ self.vm.launch()
+ self.wait_for_console_pattern('Welcome to Buildroot')
+ time.sleep(0.1)
+ exec_command(self, 'root')
+ time.sleep(0.1)
+ exec_command(self, 'cat /proc/self/maps')
+ time.sleep(0.1)
--
2.35.3

64
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: 7%{?rcrel}%{?dist}%{?cc_suffix}
Release: 8%{?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)
@ -288,6 +288,42 @@ Patch66: kvm-Enable-virtio-iommu-pci-on-x86_64.patch
Patch67: kvm-linux-aio-fix-unbalanced-plugged-counter-in-laio_io_.patch
# For bz#2092788 - Stalled IO Operations in VM
Patch68: kvm-linux-aio-explain-why-max-batch-is-checked-in-laio_i.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch69: kvm-tests-avocado-update-aarch64_virt-test-to-exercise-c.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch70: kvm-RHEL-only-tests-avocado-Switch-aarch64-tests-from-a5.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch71: kvm-RHEL-only-AArch64-Drop-unsupported-CPU-types.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch72: kvm-target-i386-deprecate-CPUs-older-than-x86_64-v2-ABI.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch73: kvm-target-s390x-deprecate-CPUs-older-than-z14.patch
# For bz#2060839 - Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9
Patch74: kvm-target-arm-deprecate-named-CPU-models.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch75: kvm-meson.build-Fix-docker-test-build-alpine-when-includ.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch76: kvm-QIOChannel-Add-flags-on-io_writev-and-introduce-io_f.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch77: kvm-QIOChannelSocket-Implement-io_writev-zero-copy-flag-.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch78: kvm-migration-Add-zero-copy-send-parameter-for-QMP-HMP-f.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch79: kvm-migration-Add-migrate_use_tls-helper.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch80: kvm-multifd-multifd_send_sync_main-now-returns-negative-.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch81: kvm-multifd-Send-header-packet-without-flags-if-zero-cop.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch82: kvm-multifd-Implement-zero-copy-write-in-multifd-migrati.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch83: kvm-QIOChannelSocket-Introduce-assert-and-reduce-ifdefs-.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch84: kvm-QIOChannelSocket-Fix-zero-copy-send-so-socket-flush-.patch
# For bz#1968509 - Use MSG_ZEROCOPY on QEMU Live Migration
Patch85: kvm-migration-Change-zero_copy_send-from-migration-param.patch
# For bz#2096143 - The migration port is not released if use it again for recovering postcopy migration
Patch86: kvm-migration-Allow-migrate-recover-to-run-multiple-time.patch
# Source-git patches
@ -1323,6 +1359,32 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%endif
%changelog
* Tue Jul 05 2022 Camilla Conte <cconte@redhat.com> - 7.0.0-8
- kvm-tests-avocado-update-aarch64_virt-test-to-exercise-c.patch [bz#2060839]
- kvm-RHEL-only-tests-avocado-Switch-aarch64-tests-from-a5.patch [bz#2060839]
- kvm-RHEL-only-AArch64-Drop-unsupported-CPU-types.patch [bz#2060839]
- kvm-target-i386-deprecate-CPUs-older-than-x86_64-v2-ABI.patch [bz#2060839]
- kvm-target-s390x-deprecate-CPUs-older-than-z14.patch [bz#2060839]
- kvm-target-arm-deprecate-named-CPU-models.patch [bz#2060839]
- kvm-meson.build-Fix-docker-test-build-alpine-when-includ.patch [bz#1968509]
- kvm-QIOChannel-Add-flags-on-io_writev-and-introduce-io_f.patch [bz#1968509]
- kvm-QIOChannelSocket-Implement-io_writev-zero-copy-flag-.patch [bz#1968509]
- kvm-migration-Add-zero-copy-send-parameter-for-QMP-HMP-f.patch [bz#1968509]
- kvm-migration-Add-migrate_use_tls-helper.patch [bz#1968509]
- kvm-multifd-multifd_send_sync_main-now-returns-negative-.patch [bz#1968509]
- kvm-multifd-Send-header-packet-without-flags-if-zero-cop.patch [bz#1968509]
- kvm-multifd-Implement-zero-copy-write-in-multifd-migrati.patch [bz#1968509]
- kvm-QIOChannelSocket-Introduce-assert-and-reduce-ifdefs-.patch [bz#1968509]
- kvm-QIOChannelSocket-Fix-zero-copy-send-so-socket-flush-.patch [bz#1968509]
- kvm-migration-Change-zero_copy_send-from-migration-param.patch [bz#1968509]
- kvm-migration-Allow-migrate-recover-to-run-multiple-time.patch [bz#2096143]
- Resolves: bz#2060839
(Consider deprecating CPU models like "kvm64" / "qemu64" on RHEL 9)
- Resolves: bz#1968509
(Use MSG_ZEROCOPY on QEMU Live Migration)
- Resolves: bz#2096143
(The migration port is not released if use it again for recovering postcopy migration)
* Mon Jun 27 2022 Miroslav Rezanina <mrezanin@redhat.com> - 7.0.0-7
- kvm-coroutine-ucontext-use-QEMU_DEFINE_STATIC_CO_TLS.patch [bz#1952483]
- kvm-coroutine-use-QEMU_DEFINE_STATIC_CO_TLS.patch [bz#1952483]