qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support
@ 2015-09-24 16:22 marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 01/21] vhost-user: unit test for new messages marcandre.lureau
                   ` (22 more replies)
  0 siblings, 23 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Hi,

The following series implement shareable log for vhost-user to support
memory tracking during live migration. On qemu-side, the solution is
fairly straightfoward since vhost already supports the dirty log, only
vhost-user couldn't access the log memory until then.

The series includes "protocol feature negotiation" series proposed
earlier by Michael S. Tsirkinm and "vhost user: Add live migration"
series from Thibaut Collet .

v4->v5:
- rebase on top of last Michael S. Tsirkin PULL request
- block live migration if !PROTOCOL_F_LOG_SHMFD
- wait for a reply after SET_LOG_BASE
- split vhost_set_log_base from the rest of vhost_call refactoring
- use a seperate global vhost_log_shm

v3->v4:
- add the proto negotiation & the migration series
- replace the varargs vhost_call() approach for callbacks
- only share-allocate when the backend needs it

v2->v3:
- changed some patch summary
- added migration tests
- added a patch to replace error message with a trace

The development branch I used is:
https://github.com/elmarco/qemu branch "vhost-user"

Marc-André Lureau (18):
  configure: probe for memfd
  util: add linux-only memfd fallback
  util: add memfd helpers
  vhost: alloc shareable log
  vhost: document log resizing
  vhost: add vhost_set_log_base op
  vhost-user: send log shm fd along with log_base
  vhost-user: add a migration blocker
  vhost: use a function for each call
  vhost: only use shared log if in use by backend
  vhost-user: document migration log
  net: add trace_vhost_user_event
  vhost-user-test: move wait_for_fds() out
  vhost-user-test: remove useless static check
  vhost-user-test: wrap server in TestServer struct
  vhost-user-test: learn to tweak various qemu arguments
  vhost-user-test: add live-migration test
  vhost-user-test: check ownership during migration

Michael S. Tsirkin (1):
  vhost-user: unit test for new messages

Thibaut Collet (2):
  vhost user: add support of live migration
  vhost user: add rarp sending after live migration for legacy guest

 configure                         |  19 ++
 docs/specs/vhost-user.txt         |  63 ++++-
 hw/net/vhost_net.c                |  35 ++-
 hw/scsi/vhost-scsi.c              |   7 +-
 hw/virtio/vhost-backend.c         | 121 +++++++-
 hw/virtio/vhost-user.c            | 571 ++++++++++++++++++++++++--------------
 hw/virtio/vhost.c                 | 101 ++++---
 include/hw/virtio/vhost-backend.h |  77 ++++-
 include/hw/virtio/vhost.h         |  15 +-
 include/net/vhost_net.h           |   1 +
 include/qemu/memfd.h              |  24 ++
 net/vhost-user.c                  |  34 ++-
 tests/vhost-user-test.c           | 390 ++++++++++++++++++++++----
 trace-events                      |   3 +
 util/Makefile.objs                |   2 +-
 util/memfd.c                      | 126 +++++++++
 16 files changed, 1260 insertions(+), 329 deletions(-)
 create mode 100644 include/qemu/memfd.h
 create mode 100644 util/memfd.c

-- 
2.4.3

^ permalink raw reply	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 01/21] vhost-user: unit test for new messages
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 02/21] configure: probe for memfd marcandre.lureau
                   ` (21 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel; +Cc: thibaut.collet, jasowang, mst, haifeng.lin, pbonzini

From: "Michael S. Tsirkin" <mst@redhat.com>

Data is empty for now, but do make sure master
sets the new feature bit flag.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 tests/vhost-user-test.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index e301db7..77b7b68 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -53,6 +53,8 @@
 
 #define VHOST_MEMORY_MAX_NREGIONS    8
 
+#define VHOST_USER_F_PROTOCOL_FEATURES 30
+
 typedef enum VhostUserRequest {
     VHOST_USER_NONE = 0,
     VHOST_USER_GET_FEATURES = 1,
@@ -69,6 +71,8 @@ typedef enum VhostUserRequest {
     VHOST_USER_SET_VRING_KICK = 12,
     VHOST_USER_SET_VRING_CALL = 13,
     VHOST_USER_SET_VRING_ERR = 14,
+    VHOST_USER_GET_PROTOCOL_FEATURES = 15,
+    VHOST_USER_SET_PROTOCOL_FEATURES = 16,
     VHOST_USER_MAX
 } VhostUserRequest;
 
@@ -293,11 +297,26 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
         /* send back features to qemu */
         msg.flags |= VHOST_USER_REPLY_MASK;
         msg.size = sizeof(m.u64);
+        msg.u64 = 0x1ULL << VHOST_USER_F_PROTOCOL_FEATURES;
+        p = (uint8_t *) &msg;
+        qemu_chr_fe_write_all(chr, p, VHOST_USER_HDR_SIZE + msg.size);
+        break;
+
+    case VHOST_USER_SET_FEATURES:
+	g_assert_cmpint(msg.u64 & (0x1ULL << VHOST_USER_F_PROTOCOL_FEATURES),
+			!=, 0ULL);
+        break;
+
+    case VHOST_USER_GET_PROTOCOL_FEATURES:
+        /* send back features to qemu */
+        msg.flags |= VHOST_USER_REPLY_MASK;
+        msg.size = sizeof(m.u64);
         msg.u64 = 0;
         p = (uint8_t *) &msg;
         qemu_chr_fe_write_all(chr, p, VHOST_USER_HDR_SIZE + msg.size);
         break;
 
+
     case VHOST_USER_GET_VRING_BASE:
         /* send back vring base to qemu */
         msg.flags |= VHOST_USER_REPLY_MASK;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 02/21] configure: probe for memfd
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 01/21] vhost-user: unit test for new messages marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 03/21] util: add linux-only memfd fallback marcandre.lureau
                   ` (20 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Check if memfd_create() is part of system libc.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 configure | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/configure b/configure
index 52f5b79..a7cdfa4 100755
--- a/configure
+++ b/configure
@@ -3455,6 +3455,22 @@ if compile_prog "" "" ; then
   eventfd=yes
 fi
 
+# check if memfd is supported
+memfd=no
+cat > $TMPC << EOF
+#include <sys/memfd.h>
+
+int main(void)
+{
+    return memfd_create("foo", MFD_ALLOW_SEALING);
+}
+EOF
+if compile_prog "" "" ; then
+  memfd=yes
+fi
+
+
+
 # check for fallocate
 fallocate=no
 cat > $TMPC << EOF
@@ -4840,6 +4856,9 @@ fi
 if test "$eventfd" = "yes" ; then
   echo "CONFIG_EVENTFD=y" >> $config_host_mak
 fi
+if test "$memfd" = "yes" ; then
+  echo "CONFIG_MEMFD=y" >> $config_host_mak
+fi
 if test "$fallocate" = "yes" ; then
   echo "CONFIG_FALLOCATE=y" >> $config_host_mak
 fi
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 03/21] util: add linux-only memfd fallback
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 01/21] vhost-user: unit test for new messages marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 02/21] configure: probe for memfd marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 04/21] util: add memfd helpers marcandre.lureau
                   ` (19 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Implement memfd_create() fallback if not available in system libc.
memfd_create() is still not included in glibc today, atlhough it's been
available since Linux 3.17 in Oct 2014.

memfd has numerous advantages over traditional shm/mmap for ipc memory
sharing with fd handler, which we are going to make use of for
vhost-user logging memory in following patches.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 include/qemu/memfd.h | 20 +++++++++++++++++++
 util/Makefile.objs   |  2 +-
 util/memfd.c         | 56 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+), 1 deletion(-)
 create mode 100644 include/qemu/memfd.h
 create mode 100644 util/memfd.c

diff --git a/include/qemu/memfd.h b/include/qemu/memfd.h
new file mode 100644
index 0000000..8b1fe6a
--- /dev/null
+++ b/include/qemu/memfd.h
@@ -0,0 +1,20 @@
+#ifndef QEMU_MEMFD_H
+#define QEMU_MEMFD_H
+
+#include "config-host.h"
+
+#ifndef F_LINUX_SPECIFIC_BASE
+#define F_LINUX_SPECIFIC_BASE 1024
+#endif
+
+#ifndef F_ADD_SEALS
+#define F_ADD_SEALS (F_LINUX_SPECIFIC_BASE + 9)
+#define F_GET_SEALS (F_LINUX_SPECIFIC_BASE + 10)
+
+#define F_SEAL_SEAL     0x0001  /* prevent further seals from being set */
+#define F_SEAL_SHRINK   0x0002  /* prevent file from shrinking */
+#define F_SEAL_GROW     0x0004  /* prevent file from growing */
+#define F_SEAL_WRITE    0x0008  /* prevent writes */
+#endif
+
+#endif /* QEMU_MEMFD_H */
diff --git a/util/Makefile.objs b/util/Makefile.objs
index 114d657..84c5485 100644
--- a/util/Makefile.objs
+++ b/util/Makefile.objs
@@ -1,6 +1,6 @@
 util-obj-y = osdep.o cutils.o unicode.o qemu-timer-common.o
 util-obj-$(CONFIG_WIN32) += oslib-win32.o qemu-thread-win32.o event_notifier-win32.o
-util-obj-$(CONFIG_POSIX) += oslib-posix.o qemu-thread-posix.o event_notifier-posix.o qemu-openpty.o
+util-obj-$(CONFIG_POSIX) += oslib-posix.o qemu-thread-posix.o event_notifier-posix.o qemu-openpty.o memfd.o
 util-obj-y += envlist.o path.o module.o
 util-obj-$(call lnot,$(CONFIG_INT128)) += host-utils.o
 util-obj-y += bitmap.o bitops.o hbitmap.o
diff --git a/util/memfd.c b/util/memfd.c
new file mode 100644
index 0000000..a98d57e
--- /dev/null
+++ b/util/memfd.c
@@ -0,0 +1,56 @@
+/*
+ * memfd.c
+ *
+ * Copyright (c) 2015 Red Hat, Inc.
+ *
+ * QEMU library functions on POSIX which are shared between QEMU and
+ * the QEMU tools.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "config-host.h"
+
+#include "qemu/memfd.h"
+
+#ifdef CONFIG_MEMFD
+#include <sys/memfd.h>
+#elif defined CONFIG_LINUX
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <asm/unistd.h>
+
+#ifndef MFD_CLOEXEC
+#define MFD_CLOEXEC 0x0001U
+#endif
+
+#ifndef MFD_ALLOW_SEALING
+#define MFD_ALLOW_SEALING 0x0002U
+#endif
+
+static inline int memfd_create(const char *name, unsigned int flags)
+{
+    return syscall(__NR_memfd_create, name, flags);
+}
+#else /* !LINUX */
+static inline int memfd_create(const char *name, unsigned int flags)
+{
+    return -1;
+}
+#endif
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 04/21] util: add memfd helpers
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (2 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 03/21] util: add linux-only memfd fallback marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 05/21] vhost: alloc shareable log marcandre.lureau
                   ` (18 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Add qemu_memfd_alloc/free() helpers.

The function helps to allocate and seal a memfd, and implements an
open/unlink/mmap fallback for system that do not support memfd.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 include/qemu/memfd.h |  4 +++
 util/memfd.c         | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 76 insertions(+), 2 deletions(-)

diff --git a/include/qemu/memfd.h b/include/qemu/memfd.h
index 8b1fe6a..950fb88 100644
--- a/include/qemu/memfd.h
+++ b/include/qemu/memfd.h
@@ -17,4 +17,8 @@
 #define F_SEAL_WRITE    0x0008  /* prevent writes */
 #endif
 
+void *qemu_memfd_alloc(const char *name, size_t size, unsigned int seals,
+                       int *fd);
+void qemu_memfd_free(void *ptr, size_t size, int fd);
+
 #endif /* QEMU_MEMFD_H */
diff --git a/util/memfd.c b/util/memfd.c
index a98d57e..8b2b785 100644
--- a/util/memfd.c
+++ b/util/memfd.c
@@ -27,6 +27,14 @@
 
 #include "config-host.h"
 
+#include <glib.h>
+#include <glib/gprintf.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <fcntl.h>
+#include <sys/mman.h>
+
 #include "qemu/memfd.h"
 
 #ifdef CONFIG_MEMFD
@@ -44,13 +52,75 @@
 #define MFD_ALLOW_SEALING 0x0002U
 #endif
 
-static inline int memfd_create(const char *name, unsigned int flags)
+static int memfd_create(const char *name, unsigned int flags)
 {
     return syscall(__NR_memfd_create, name, flags);
 }
 #else /* !LINUX */
-static inline int memfd_create(const char *name, unsigned int flags)
+static int memfd_create(const char *name, unsigned int flags)
 {
     return -1;
 }
 #endif
+
+void *qemu_memfd_alloc(const char *name, size_t size, unsigned int seals,
+                       int *fd)
+{
+    void *ptr;
+    int mfd;
+
+    mfd = memfd_create(name, MFD_ALLOW_SEALING|MFD_CLOEXEC);
+    if (mfd != -1) {
+        if (ftruncate(mfd, size) == -1) {
+            perror("ftruncate");
+            close(mfd);
+            return NULL;
+        }
+
+        if (fcntl(mfd, F_ADD_SEALS, seals) == -1) {
+            perror("fcntl");
+            close(mfd);
+            return NULL;
+        }
+    } else {
+        const char *tmpdir = getenv("TMPDIR");
+        gchar *fname;
+
+        tmpdir = tmpdir ? tmpdir : "/tmp";
+
+        fname = g_strdup_printf("%s/memfd-XXXXXX", tmpdir);
+        mfd = mkstemp(fname);
+        unlink(fname);
+        g_free(fname);
+
+        if (mfd == -1) {
+            perror("mkstemp");
+            return NULL;
+        }
+
+        if (ftruncate(mfd, size) == -1) {
+            perror("ftruncate");
+            close(mfd);
+            return NULL;
+        }
+    }
+
+    ptr = mmap(0, size, PROT_READ|PROT_WRITE, MAP_SHARED, mfd, 0);
+    if (ptr == MAP_FAILED) {
+        perror("mmap");
+        close(mfd);
+        return NULL;
+    }
+
+    *fd = mfd;
+    return ptr;
+}
+
+void qemu_memfd_free(void *ptr, size_t size, int fd)
+{
+    if (ptr) {
+        munmap(ptr, size);
+    }
+
+    close(fd);
+}
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 05/21] vhost: alloc shareable log
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (3 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 04/21] util: add memfd helpers marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 06/21] vhost: document log resizing marcandre.lureau
                   ` (17 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

If the backend is of type VHOST_BACKEND_TYPE_USER, allocate
shareable memory. Next patch will only allocate when the backend
has the required feature.

vhost_log_get() now uses 2 globals "vhost_log" and "vhost_log_shm", that
way there is a common non-shareable log and a common shareable one.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost.c         | 49 ++++++++++++++++++++++++++++++++++++++---------
 include/hw/virtio/vhost.h |  3 ++-
 2 files changed, 42 insertions(+), 10 deletions(-)

diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index c0ed5b2..bf88618 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -18,6 +18,7 @@
 #include "qemu/atomic.h"
 #include "qemu/range.h"
 #include "qemu/error-report.h"
+#include "qemu/memfd.h"
 #include <linux/vhost.h>
 #include "exec/address-spaces.h"
 #include "hw/virtio/virtio-bus.h"
@@ -25,6 +26,7 @@
 #include "migration/migration.h"
 
 static struct vhost_log *vhost_log;
+static struct vhost_log *vhost_log_shm;
 
 static void vhost_dev_sync_region(struct vhost_dev *dev,
                                   MemoryRegionSection *section,
@@ -286,25 +288,45 @@ static uint64_t vhost_get_log_size(struct vhost_dev *dev)
     }
     return log_size;
 }
-static struct vhost_log *vhost_log_alloc(uint64_t size)
+
+static struct vhost_log *vhost_log_alloc(uint64_t size, bool share)
 {
-    struct vhost_log *log = g_malloc0(sizeof *log + size * sizeof(*(log->log)));
+    struct vhost_log *log;
+    uint64_t logsize = size * sizeof(*(log->log));
+    int fd = -1;
+
+    log = g_new0(struct vhost_log, 1);
+    if (share) {
+        log->log = qemu_memfd_alloc("vhost-log", logsize,
+                                    F_SEAL_GROW|F_SEAL_SHRINK|F_SEAL_SEAL, &fd);
+        memset(log->log, 0, logsize);
+    } else {
+        log->log = g_malloc0(logsize);
+    }
 
     log->size = size;
     log->refcnt = 1;
+    log->fd = fd;
 
     return log;
 }
 
-static struct vhost_log *vhost_log_get(uint64_t size)
+static struct vhost_log *vhost_log_get(uint64_t size, bool share)
 {
-    if (!vhost_log || vhost_log->size != size) {
-        vhost_log = vhost_log_alloc(size);
+    struct vhost_log *log = share ? vhost_log_shm : vhost_log;
+
+    if (!log || log->size != size) {
+        log = vhost_log_alloc(size, share);
+        if (share) {
+            vhost_log_shm = log;
+        } else {
+            vhost_log = log;
+        }
     } else {
-        ++vhost_log->refcnt;
+        ++log->refcnt;
     }
 
-    return vhost_log;
+    return log;
 }
 
 static void vhost_log_put(struct vhost_dev *dev, bool sync)
@@ -321,16 +343,24 @@ static void vhost_log_put(struct vhost_dev *dev, bool sync)
         if (dev->log_size && sync) {
             vhost_log_sync_range(dev, 0, dev->log_size * VHOST_LOG_CHUNK - 1);
         }
+
         if (vhost_log == log) {
+            g_free(log->log);
             vhost_log = NULL;
+        } else if (vhost_log_shm == log) {
+            qemu_memfd_free(log->log, log->size * sizeof(*(log->log)),
+                            log->fd);
+            vhost_log_shm = NULL;
         }
+
         g_free(log);
     }
 }
 
 static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
 {
-    struct vhost_log *log = vhost_log_get(size);
+    bool share = dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
+    struct vhost_log *log = vhost_log_get(size, share);
     uint64_t log_base = (uintptr_t)log->log;
     int r;
 
@@ -1134,9 +1164,10 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
 
     if (hdev->log_enabled) {
         uint64_t log_base;
+        bool share = hdev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
 
         hdev->log_size = vhost_get_log_size(hdev);
-        hdev->log = vhost_log_get(hdev->log_size);
+        hdev->log = vhost_log_get(hdev->log_size, share);
         log_base = (uintptr_t)hdev->log->log;
         r = hdev->vhost_ops->vhost_call(hdev, VHOST_SET_LOG_BASE,
                                         hdev->log_size ? &log_base : NULL);
diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h
index c3758f3..7e7dc45 100644
--- a/include/hw/virtio/vhost.h
+++ b/include/hw/virtio/vhost.h
@@ -31,7 +31,8 @@ typedef unsigned long vhost_log_chunk_t;
 struct vhost_log {
     unsigned long long size;
     int refcnt;
-    vhost_log_chunk_t log[0];
+    int fd;
+    vhost_log_chunk_t *log;
 };
 
 struct vhost_memory;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 06/21] vhost: document log resizing
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (4 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 05/21] vhost: alloc shareable log marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 07/21] vhost: add vhost_set_log_base op marcandre.lureau
                   ` (16 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index bf88618..7bc3862 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -364,6 +364,8 @@ static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
     uint64_t log_base = (uintptr_t)log->log;
     int r;
 
+    /* inform backend of log switching, this must be done before
+       releasing the current log, to ensure no logging is lost */
     r = dev->vhost_ops->vhost_call(dev, VHOST_SET_LOG_BASE, &log_base);
     assert(r >= 0);
     vhost_log_put(dev, true);
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 07/21] vhost: add vhost_set_log_base op
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (5 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 06/21] vhost: document log resizing marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 08/21] vhost-user: send log shm fd along with log_base marcandre.lureau
                   ` (15 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Split VHOST_SET_LOG_BASE call in a seperate function callback, so that
type safety works and more arguments can be added in the next patches.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost-backend.c         |  8 ++++++++
 hw/virtio/vhost-user.c            | 17 ++++++++++++++++-
 hw/virtio/vhost.c                 |  6 +++---
 include/hw/virtio/vhost-backend.h |  5 +++++
 4 files changed, 32 insertions(+), 4 deletions(-)

diff --git a/hw/virtio/vhost-backend.c b/hw/virtio/vhost-backend.c
index 72d1392..88d33fd 100644
--- a/hw/virtio/vhost-backend.c
+++ b/hw/virtio/vhost-backend.c
@@ -11,6 +11,7 @@
 #include "hw/virtio/vhost.h"
 #include "hw/virtio/vhost-backend.h"
 #include "qemu/error-report.h"
+#include "linux/vhost.h"
 
 #include <sys/ioctl.h>
 
@@ -49,12 +50,19 @@ static int vhost_kernel_get_vq_index(struct vhost_dev *dev, int idx)
     return idx - dev->vq_index;
 }
 
+static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base)
+{
+    return vhost_kernel_call(dev, VHOST_SET_LOG_BASE, &base);
+}
+
 static const VhostOps kernel_ops = {
         .backend_type = VHOST_BACKEND_TYPE_KERNEL,
         .vhost_call = vhost_kernel_call,
         .vhost_backend_init = vhost_kernel_init,
         .vhost_backend_cleanup = vhost_kernel_cleanup,
         .vhost_backend_get_vq_index = vhost_kernel_get_vq_index,
+
+        .vhost_set_log_base = vhost_set_log_base,
 };
 
 int vhost_set_backend_type(struct vhost_dev *dev, VhostBackendType backend_type)
diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index b11c0d2..fd4fcbf 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -242,7 +242,6 @@ static int vhost_user_call(struct vhost_dev *dev, unsigned long int request,
         break;
 
     case VHOST_USER_SET_FEATURES:
-    case VHOST_USER_SET_LOG_BASE:
     case VHOST_USER_SET_PROTOCOL_FEATURES:
         msg.u64 = *((__u64 *) arg);
         msg.size = sizeof(m.u64);
@@ -367,6 +366,20 @@ static int vhost_user_call(struct vhost_dev *dev, unsigned long int request,
     return 0;
 }
 
+static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base)
+{
+    VhostUserMsg msg = {
+        .request = VHOST_USER_SET_LOG_BASE,
+        .flags = VHOST_USER_VERSION,
+        .u64 = base,
+        .size = sizeof(m.u64),
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    return 0;
+}
+
 static int vhost_user_init(struct vhost_dev *dev, void *opaque)
 {
     unsigned long long features;
@@ -447,4 +460,6 @@ const VhostOps user_ops = {
         .vhost_backend_cleanup = vhost_user_cleanup,
         .vhost_backend_get_vq_index = vhost_user_get_vq_index,
         .vhost_backend_set_vring_enable = vhost_user_set_vring_enable,
+
+        .vhost_set_log_base = vhost_set_log_base,
 };
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 7bc3862..7f01cd6 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -366,7 +366,7 @@ static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
 
     /* inform backend of log switching, this must be done before
        releasing the current log, to ensure no logging is lost */
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_LOG_BASE, &log_base);
+    r = dev->vhost_ops->vhost_set_log_base(dev, log_base);
     assert(r >= 0);
     vhost_log_put(dev, true);
     dev->log = log;
@@ -1171,8 +1171,8 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
         hdev->log_size = vhost_get_log_size(hdev);
         hdev->log = vhost_log_get(hdev->log_size, share);
         log_base = (uintptr_t)hdev->log->log;
-        r = hdev->vhost_ops->vhost_call(hdev, VHOST_SET_LOG_BASE,
-                                        hdev->log_size ? &log_base : NULL);
+        r = hdev->vhost_ops->vhost_set_log_base(hdev,
+                                                hdev->log_size ? log_base : 0);
         if (r < 0) {
             r = -errno;
             goto fail_log;
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index 3a0f6e2..c6b688a 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -27,13 +27,18 @@ typedef int (*vhost_backend_cleanup)(struct vhost_dev *dev);
 typedef int (*vhost_backend_get_vq_index)(struct vhost_dev *dev, int idx);
 typedef int (*vhost_backend_set_vring_enable)(struct vhost_dev *dev, int enable);
 
+typedef int (*vhost_set_log_base_op)(struct vhost_dev *dev, uint64_t base);
+
 typedef struct VhostOps {
     VhostBackendType backend_type;
     vhost_call vhost_call;
+
     vhost_backend_init vhost_backend_init;
     vhost_backend_cleanup vhost_backend_cleanup;
     vhost_backend_get_vq_index vhost_backend_get_vq_index;
     vhost_backend_set_vring_enable vhost_backend_set_vring_enable;
+
+    vhost_set_log_base_op vhost_set_log_base;
 } VhostOps;
 
 extern const VhostOps user_ops;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 08/21] vhost-user: send log shm fd along with log_base
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (6 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 07/21] vhost: add vhost_set_log_base op marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 09/21] vhost-user: add a migration blocker marcandre.lureau
                   ` (14 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Send the shm for the dirty pages logging if the backend supports
VHOST_USER_PROTOCOL_F_LOG_SHMFD. Wait for a reply to make sure
the old log is no longer used.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost-backend.c         |  3 ++-
 hw/virtio/vhost-user.c            | 32 ++++++++++++++++++++++++++++----
 hw/virtio/vhost.c                 |  5 +++--
 include/hw/virtio/vhost-backend.h |  4 +++-
 4 files changed, 36 insertions(+), 8 deletions(-)

diff --git a/hw/virtio/vhost-backend.c b/hw/virtio/vhost-backend.c
index 88d33fd..87ab028 100644
--- a/hw/virtio/vhost-backend.c
+++ b/hw/virtio/vhost-backend.c
@@ -50,7 +50,8 @@ static int vhost_kernel_get_vq_index(struct vhost_dev *dev, int idx)
     return idx - dev->vq_index;
 }
 
-static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base)
+static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base,
+                              struct vhost_log *log)
 {
     return vhost_kernel_call(dev, VHOST_SET_LOG_BASE, &base);
 }
diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index fd4fcbf..6420efd 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -25,9 +25,10 @@
 
 #define VHOST_MEMORY_MAX_NREGIONS    8
 #define VHOST_USER_F_PROTOCOL_FEATURES 30
-#define VHOST_USER_PROTOCOL_FEATURE_MASK 0x1ULL
 
-#define VHOST_USER_PROTOCOL_F_MQ    0
+#define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
+#define VHOST_USER_PROTOCOL_F_MQ         0
+#define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
 
 typedef enum VhostUserRequest {
     VHOST_USER_NONE = 0,
@@ -366,8 +367,13 @@ static int vhost_user_call(struct vhost_dev *dev, unsigned long int request,
     return 0;
 }
 
-static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base)
+static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base,
+                              struct vhost_log *log)
 {
+    int fds[VHOST_MEMORY_MAX_NREGIONS];
+    size_t fd_num = 0;
+    bool shmfd = virtio_has_feature(dev->protocol_features,
+                                    VHOST_USER_PROTOCOL_F_LOG_SHMFD);
     VhostUserMsg msg = {
         .request = VHOST_USER_SET_LOG_BASE,
         .flags = VHOST_USER_VERSION,
@@ -375,7 +381,25 @@ static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base)
         .size = sizeof(m.u64),
     };
 
-    vhost_user_write(dev, &msg, NULL, 0);
+    if (shmfd && log->fd != -1) {
+        fds[fd_num++] = log->fd;
+    }
+
+    vhost_user_write(dev, &msg, fds, fd_num);
+
+    if (shmfd) {
+        msg.size = 0;
+        if (vhost_user_read(dev, &msg) < 0) {
+            return 0;
+        }
+
+        if (msg.request != VHOST_USER_SET_LOG_BASE) {
+            error_report("Received unexpected msg type. "
+                         "Expected %d received %d",
+                         VHOST_USER_SET_LOG_BASE, msg.request);
+            return -1;
+        }
+    }
 
     return 0;
 }
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 7f01cd6..9b73f83 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -366,7 +366,7 @@ static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
 
     /* inform backend of log switching, this must be done before
        releasing the current log, to ensure no logging is lost */
-    r = dev->vhost_ops->vhost_set_log_base(dev, log_base);
+    r = dev->vhost_ops->vhost_set_log_base(dev, log_base, log);
     assert(r >= 0);
     vhost_log_put(dev, true);
     dev->log = log;
@@ -1172,7 +1172,8 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
         hdev->log = vhost_log_get(hdev->log_size, share);
         log_base = (uintptr_t)hdev->log->log;
         r = hdev->vhost_ops->vhost_set_log_base(hdev,
-                                                hdev->log_size ? log_base : 0);
+                                                hdev->log_size ? log_base : 0,
+                                                hdev->log);
         if (r < 0) {
             r = -errno;
             goto fail_log;
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index c6b688a..485279b 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -19,6 +19,7 @@ typedef enum VhostBackendType {
 } VhostBackendType;
 
 struct vhost_dev;
+struct vhost_log;
 
 typedef int (*vhost_call)(struct vhost_dev *dev, unsigned long int request,
              void *arg);
@@ -27,7 +28,8 @@ typedef int (*vhost_backend_cleanup)(struct vhost_dev *dev);
 typedef int (*vhost_backend_get_vq_index)(struct vhost_dev *dev, int idx);
 typedef int (*vhost_backend_set_vring_enable)(struct vhost_dev *dev, int enable);
 
-typedef int (*vhost_set_log_base_op)(struct vhost_dev *dev, uint64_t base);
+typedef int (*vhost_set_log_base_op)(struct vhost_dev *dev, uint64_t base,
+                                     struct vhost_log *log);
 
 typedef struct VhostOps {
     VhostBackendType backend_type;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 09/21] vhost-user: add a migration blocker
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (7 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 08/21] vhost-user: send log shm fd along with log_base marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 10/21] vhost: use a function for each call marcandre.lureau
                   ` (13 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

If VHOST_USER_PROTOCOL_F_LOG_SHMFD is not announced, block vhost-user
migration.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost-user.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index 6420efd..ff87e87 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -15,6 +15,7 @@
 #include "qemu/error-report.h"
 #include "qemu/sockets.h"
 #include "exec/ram_addr.h"
+#include "migration/migration.h"
 
 #include <fcntl.h>
 #include <unistd.h>
@@ -442,6 +443,14 @@ static int vhost_user_init(struct vhost_dev *dev, void *opaque)
         }
     }
 
+    if (!virtio_has_feature(dev->protocol_features,
+                            VHOST_USER_PROTOCOL_F_LOG_SHMFD)) {
+        migrate_add_blocker(dev->migration_blocker);
+        error_setg(&dev->migration_blocker,
+                   "Migration disabled: vhost-user backend lacks "
+                   "VHOST_USER_PROTOCOL_F_LOG_SHMFD feature.");
+    }
+
     return 0;
 }
 
@@ -466,6 +475,10 @@ static int vhost_user_cleanup(struct vhost_dev *dev)
     assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
 
     dev->opaque = 0;
+    if (dev->migration_blocker) {
+        migrate_del_blocker(dev->migration_blocker);
+        error_free(dev->migration_blocker);
+    }
 
     return 0;
 }
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 10/21] vhost: use a function for each call
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (8 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 09/21] vhost-user: add a migration blocker marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend marcandre.lureau
                   ` (12 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Replace the generic vhost_call() by specific functions for each
function call to help with type safety and changing arguments.

While doing this, I found that "unsigned long long" and "uint64_t" were
used interchangeably and causing compilation warnings, using uint64_t
instead, as the vhost & protocol specifies.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/net/vhost_net.c                |  16 +-
 hw/scsi/vhost-scsi.c              |   7 +-
 hw/virtio/vhost-backend.c         | 124 ++++++++-
 hw/virtio/vhost-user.c            | 518 ++++++++++++++++++++++----------------
 hw/virtio/vhost.c                 |  36 +--
 include/hw/virtio/vhost-backend.h |  63 ++++-
 include/hw/virtio/vhost.h         |  12 +-
 7 files changed, 501 insertions(+), 275 deletions(-)

diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
index 2bce891..1ab4133 100644
--- a/hw/net/vhost_net.c
+++ b/hw/net/vhost_net.c
@@ -252,8 +252,7 @@ static int vhost_net_start_one(struct vhost_net *net,
         file.fd = net->backend;
         for (file.index = 0; file.index < net->dev.nvqs; ++file.index) {
             const VhostOps *vhost_ops = net->dev.vhost_ops;
-            r = vhost_ops->vhost_call(&net->dev, VHOST_NET_SET_BACKEND,
-                                      &file);
+            r = vhost_ops->vhost_net_set_backend(&net->dev, &file);
             if (r < 0) {
                 r = -errno;
                 goto fail;
@@ -266,8 +265,7 @@ fail:
     if (net->nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP) {
         while (file.index-- > 0) {
             const VhostOps *vhost_ops = net->dev.vhost_ops;
-            int r = vhost_ops->vhost_call(&net->dev, VHOST_NET_SET_BACKEND,
-                                          &file);
+            int r = vhost_ops->vhost_net_set_backend(&net->dev, &file);
             assert(r >= 0);
         }
     }
@@ -289,15 +287,13 @@ static void vhost_net_stop_one(struct vhost_net *net,
     if (net->nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP) {
         for (file.index = 0; file.index < net->dev.nvqs; ++file.index) {
             const VhostOps *vhost_ops = net->dev.vhost_ops;
-            int r = vhost_ops->vhost_call(&net->dev, VHOST_NET_SET_BACKEND,
-                                          &file);
+            int r = vhost_ops->vhost_net_set_backend(&net->dev, &file);
             assert(r >= 0);
         }
     } else if (net->nc->info->type == NET_CLIENT_OPTIONS_KIND_VHOST_USER) {
         for (file.index = 0; file.index < net->dev.nvqs; ++file.index) {
             const VhostOps *vhost_ops = net->dev.vhost_ops;
-            int r = vhost_ops->vhost_call(&net->dev, VHOST_RESET_DEVICE,
-                                          NULL);
+            int r = vhost_ops->vhost_reset_device(&net->dev);
             assert(r >= 0);
         }
     }
@@ -428,8 +424,8 @@ int vhost_set_vring_enable(NetClientState *nc, int enable)
     VHostNetState *net = get_vhost_net(nc);
     const VhostOps *vhost_ops = net->dev.vhost_ops;
 
-    if (vhost_ops->vhost_backend_set_vring_enable) {
-        return vhost_ops->vhost_backend_set_vring_enable(&net->dev, enable);
+    if (vhost_ops->vhost_set_vring_enable) {
+        return vhost_ops->vhost_set_vring_enable(&net->dev, enable);
     }
 
     return 0;
diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
index bac9ddb..a0034ab 100644
--- a/hw/scsi/vhost-scsi.c
+++ b/hw/scsi/vhost-scsi.c
@@ -45,7 +45,7 @@ static int vhost_scsi_set_endpoint(VHostSCSI *s)
 
     memset(&backend, 0, sizeof(backend));
     pstrcpy(backend.vhost_wwpn, sizeof(backend.vhost_wwpn), vs->conf.wwpn);
-    ret = vhost_ops->vhost_call(&s->dev, VHOST_SCSI_SET_ENDPOINT, &backend);
+    ret = vhost_ops->vhost_scsi_set_endpoint(&s->dev, &backend);
     if (ret < 0) {
         return -errno;
     }
@@ -60,7 +60,7 @@ static void vhost_scsi_clear_endpoint(VHostSCSI *s)
 
     memset(&backend, 0, sizeof(backend));
     pstrcpy(backend.vhost_wwpn, sizeof(backend.vhost_wwpn), vs->conf.wwpn);
-    vhost_ops->vhost_call(&s->dev, VHOST_SCSI_CLEAR_ENDPOINT, &backend);
+    vhost_ops->vhost_scsi_clear_endpoint(&s->dev, &backend);
 }
 
 static int vhost_scsi_start(VHostSCSI *s)
@@ -76,8 +76,7 @@ static int vhost_scsi_start(VHostSCSI *s)
         return -ENOSYS;
     }
 
-    ret = vhost_ops->vhost_call(&s->dev,
-                                VHOST_SCSI_GET_ABI_VERSION, &abi_version);
+    ret = vhost_ops->vhost_scsi_get_abi_version(&s->dev, &abi_version);
     if (ret < 0) {
         return -errno;
     }
diff --git a/hw/virtio/vhost-backend.c b/hw/virtio/vhost-backend.c
index 87ab028..3820af4 100644
--- a/hw/virtio/vhost-backend.c
+++ b/hw/virtio/vhost-backend.c
@@ -43,27 +43,135 @@ static int vhost_kernel_cleanup(struct vhost_dev *dev)
     return close(fd);
 }
 
-static int vhost_kernel_get_vq_index(struct vhost_dev *dev, int idx)
+static int vhost_kernel_net_set_backend(struct vhost_dev *dev,
+                                        struct vhost_vring_file *file)
 {
-    assert(idx >= dev->vq_index && idx < dev->vq_index + dev->nvqs);
+    return vhost_kernel_call(dev, VHOST_NET_SET_BACKEND, file);
+}
 
-    return idx - dev->vq_index;
+static int vhost_kernel_scsi_set_endpoint(struct vhost_dev *dev,
+                                          struct vhost_scsi_target *target)
+{
+    return vhost_kernel_call(dev, VHOST_SCSI_SET_ENDPOINT, target);
+}
+
+static int vhost_kernel_scsi_clear_endpoint(struct vhost_dev *dev,
+                                            struct vhost_scsi_target *target)
+{
+    return vhost_kernel_call(dev, VHOST_SCSI_CLEAR_ENDPOINT, target);
+}
+
+static int vhost_kernel_scsi_get_abi_version(struct vhost_dev *dev, int *version)
+{
+    return vhost_kernel_call(dev, VHOST_SCSI_GET_ABI_VERSION, version);
 }
 
-static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base,
-                              struct vhost_log *log)
+static int vhost_kernel_set_log_base(struct vhost_dev *dev, uint64_t base,
+                                     struct vhost_log *log)
 {
     return vhost_kernel_call(dev, VHOST_SET_LOG_BASE, &base);
 }
 
+static int vhost_kernel_set_mem_table(struct vhost_dev *dev,
+                                      struct vhost_memory *mem)
+{
+    return vhost_kernel_call(dev, VHOST_SET_MEM_TABLE, mem);
+}
+
+static int vhost_kernel_set_vring_addr(struct vhost_dev *dev,
+                                       struct vhost_vring_addr *addr)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_ADDR, addr);
+}
+
+static int vhost_kernel_set_vring_endian(struct vhost_dev *dev,
+                                         struct vhost_vring_state *ring)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_ENDIAN, ring);
+}
+
+static int vhost_kernel_set_vring_num(struct vhost_dev *dev,
+                                      struct vhost_vring_state *ring)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_NUM, ring);
+}
+
+static int vhost_kernel_set_vring_base(struct vhost_dev *dev,
+                                       struct vhost_vring_state *ring)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_BASE, ring);
+}
+
+static int vhost_kernel_get_vring_base(struct vhost_dev *dev,
+                                       struct vhost_vring_state *ring)
+{
+    return vhost_kernel_call(dev, VHOST_GET_VRING_BASE, ring);
+}
+
+static int vhost_kernel_set_vring_kick(struct vhost_dev *dev,
+                                       struct vhost_vring_file *file)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_KICK, file);
+}
+
+static int vhost_kernel_set_vring_call(struct vhost_dev *dev,
+                                       struct vhost_vring_file *file)
+{
+    return vhost_kernel_call(dev, VHOST_SET_VRING_CALL, file);
+}
+
+static int vhost_kernel_set_features(struct vhost_dev *dev,
+                                     uint64_t features)
+{
+    return vhost_kernel_call(dev, VHOST_SET_FEATURES, &features);
+}
+
+static int vhost_kernel_get_features(struct vhost_dev *dev,
+                                     uint64_t *features)
+{
+    return vhost_kernel_call(dev, VHOST_GET_FEATURES, features);
+}
+
+static int vhost_kernel_set_owner(struct vhost_dev *dev)
+{
+    return vhost_kernel_call(dev, VHOST_SET_OWNER, NULL);
+}
+
+static int vhost_kernel_reset_device(struct vhost_dev *dev)
+{
+    return vhost_kernel_call(dev, VHOST_RESET_DEVICE, NULL);
+}
+
+static int vhost_kernel_get_vq_index(struct vhost_dev *dev, int idx)
+{
+    assert(idx >= dev->vq_index && idx < dev->vq_index + dev->nvqs);
+
+    return idx - dev->vq_index;
+}
+
 static const VhostOps kernel_ops = {
         .backend_type = VHOST_BACKEND_TYPE_KERNEL,
-        .vhost_call = vhost_kernel_call,
         .vhost_backend_init = vhost_kernel_init,
         .vhost_backend_cleanup = vhost_kernel_cleanup,
-        .vhost_backend_get_vq_index = vhost_kernel_get_vq_index,
 
-        .vhost_set_log_base = vhost_set_log_base,
+        .vhost_net_set_backend = vhost_kernel_net_set_backend,
+        .vhost_scsi_set_endpoint = vhost_kernel_scsi_set_endpoint,
+        .vhost_scsi_clear_endpoint = vhost_kernel_scsi_clear_endpoint,
+        .vhost_scsi_get_abi_version = vhost_kernel_scsi_get_abi_version,
+        .vhost_set_log_base = vhost_kernel_set_log_base,
+        .vhost_set_mem_table = vhost_kernel_set_mem_table,
+        .vhost_set_vring_addr = vhost_kernel_set_vring_addr,
+        .vhost_set_vring_endian = vhost_kernel_set_vring_endian,
+        .vhost_set_vring_num = vhost_kernel_set_vring_num,
+        .vhost_set_vring_base = vhost_kernel_set_vring_base,
+        .vhost_get_vring_base = vhost_kernel_get_vring_base,
+        .vhost_set_vring_kick = vhost_kernel_set_vring_kick,
+        .vhost_set_vring_call = vhost_kernel_set_vring_call,
+        .vhost_set_features = vhost_kernel_set_features,
+        .vhost_get_features = vhost_kernel_get_features,
+        .vhost_set_owner = vhost_kernel_set_owner,
+        .vhost_reset_device = vhost_kernel_reset_device,
+        .vhost_get_vq_index = vhost_kernel_get_vq_index,
 };
 
 int vhost_set_backend_type(struct vhost_dev *dev, VhostBackendType backend_type)
diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index ff87e87..41e0364 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -99,37 +99,6 @@ static bool ioeventfd_enabled(void)
     return kvm_enabled() && kvm_eventfds_enabled();
 }
 
-static unsigned long int ioctl_to_vhost_user_request[VHOST_USER_MAX] = {
-    -1,                     /* VHOST_USER_NONE */
-    VHOST_GET_FEATURES,     /* VHOST_USER_GET_FEATURES */
-    VHOST_SET_FEATURES,     /* VHOST_USER_SET_FEATURES */
-    VHOST_SET_OWNER,        /* VHOST_USER_SET_OWNER */
-    VHOST_RESET_DEVICE,      /* VHOST_USER_RESET_DEVICE */
-    VHOST_SET_MEM_TABLE,    /* VHOST_USER_SET_MEM_TABLE */
-    VHOST_SET_LOG_BASE,     /* VHOST_USER_SET_LOG_BASE */
-    VHOST_SET_LOG_FD,       /* VHOST_USER_SET_LOG_FD */
-    VHOST_SET_VRING_NUM,    /* VHOST_USER_SET_VRING_NUM */
-    VHOST_SET_VRING_ADDR,   /* VHOST_USER_SET_VRING_ADDR */
-    VHOST_SET_VRING_BASE,   /* VHOST_USER_SET_VRING_BASE */
-    VHOST_GET_VRING_BASE,   /* VHOST_USER_GET_VRING_BASE */
-    VHOST_SET_VRING_KICK,   /* VHOST_USER_SET_VRING_KICK */
-    VHOST_SET_VRING_CALL,   /* VHOST_USER_SET_VRING_CALL */
-    VHOST_SET_VRING_ERR     /* VHOST_USER_SET_VRING_ERR */
-};
-
-static VhostUserRequest vhost_user_request_translate(unsigned long int request)
-{
-    VhostUserRequest idx;
-
-    for (idx = 0; idx < VHOST_USER_MAX; idx++) {
-        if (ioctl_to_vhost_user_request[idx] == request) {
-            break;
-        }
-    }
-
-    return (idx == VHOST_USER_MAX) ? VHOST_USER_NONE : idx;
-}
-
 static int vhost_user_read(struct vhost_dev *dev, VhostUserMsg *msg)
 {
     CharDriverState *chr = dev->opaque;
@@ -176,12 +145,35 @@ fail:
     return -1;
 }
 
+static bool vhost_user_one_time_request(VhostUserRequest request)
+{
+    switch (request) {
+    case VHOST_USER_SET_OWNER:
+    case VHOST_USER_RESET_DEVICE:
+    case VHOST_USER_SET_MEM_TABLE:
+    case VHOST_USER_GET_QUEUE_NUM:
+        return true;
+    default:
+        return false;
+    }
+}
+
+/* most non-init callers ignore the error */
 static int vhost_user_write(struct vhost_dev *dev, VhostUserMsg *msg,
                             int *fds, int fd_num)
 {
     CharDriverState *chr = dev->opaque;
     int size = VHOST_USER_HDR_SIZE + msg->size;
 
+    /*
+     * For non-vring specific requests, like VHOST_USER_SET_MEM_TABLE,
+     * we just need send it once in the first time. For later such
+     * request, we just ignore it.
+     */
+    if (vhost_user_one_time_request(msg->request) && dev->vq_index != 0) {
+        return 0;
+    }
+
     if (fd_num) {
         qemu_chr_fe_set_msgfds(chr, fds, fd_num);
     }
@@ -190,231 +182,317 @@ static int vhost_user_write(struct vhost_dev *dev, VhostUserMsg *msg,
             0 : -1;
 }
 
-static bool vhost_user_one_time_request(VhostUserRequest request)
+static int vhost_user_set_log_base(struct vhost_dev *dev, uint64_t base,
+                                   struct vhost_log *log)
 {
-    switch (request) {
-    case VHOST_USER_SET_OWNER:
-    case VHOST_USER_RESET_DEVICE:
-    case VHOST_USER_SET_MEM_TABLE:
-    case VHOST_USER_GET_QUEUE_NUM:
-        return true;
-    default:
-        return false;
+    int fds[VHOST_MEMORY_MAX_NREGIONS];
+    size_t fd_num = 0;
+    bool shmfd = virtio_has_feature(dev->protocol_features,
+                                    VHOST_USER_PROTOCOL_F_LOG_SHMFD);
+    VhostUserMsg msg = {
+        .request = VHOST_USER_SET_LOG_BASE,
+        .flags = VHOST_USER_VERSION,
+        .u64 = base,
+        .size = sizeof(m.u64),
+    };
+
+    if (shmfd && log->fd != -1) {
+        fds[fd_num++] = log->fd;
     }
+
+    vhost_user_write(dev, &msg, fds, fd_num);
+
+    if (shmfd) {
+        msg.size = 0;
+        if (vhost_user_read(dev, &msg) < 0) {
+            return 0;
+        }
+
+        if (msg.request != VHOST_USER_SET_LOG_BASE) {
+            error_report("Received unexpected msg type. "
+                         "Expected %d received %d",
+                         VHOST_USER_SET_LOG_BASE, msg.request);
+            return -1;
+        }
+    }
+
+    return 0;
 }
 
-static int vhost_user_call(struct vhost_dev *dev, unsigned long int request,
-        void *arg)
+static int vhost_user_set_mem_table(struct vhost_dev *dev,
+                                    struct vhost_memory *mem)
 {
-    VhostUserMsg msg;
-    VhostUserRequest msg_request;
-    struct vhost_vring_file *file = 0;
-    int need_reply = 0;
     int fds[VHOST_MEMORY_MAX_NREGIONS];
     int i, fd;
     size_t fd_num = 0;
+    VhostUserMsg msg = {
+        .request = VHOST_USER_SET_MEM_TABLE,
+        .flags = VHOST_USER_VERSION,
+    };
 
-    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
-
-    /* only translate vhost ioctl requests */
-    if (request > VHOST_USER_MAX) {
-        msg_request = vhost_user_request_translate(request);
-    } else {
-        msg_request = request;
+    for (i = 0; i < dev->mem->nregions; ++i) {
+        struct vhost_memory_region *reg = dev->mem->regions + i;
+        ram_addr_t ram_addr;
+
+        assert((uintptr_t)reg->userspace_addr == reg->userspace_addr);
+        qemu_ram_addr_from_host((void *)(uintptr_t)reg->userspace_addr,
+                                &ram_addr);
+        fd = qemu_get_ram_fd(ram_addr);
+        if (fd > 0) {
+            msg.memory.regions[fd_num].userspace_addr = reg->userspace_addr;
+            msg.memory.regions[fd_num].memory_size  = reg->memory_size;
+            msg.memory.regions[fd_num].guest_phys_addr = reg->guest_phys_addr;
+            msg.memory.regions[fd_num].mmap_offset = reg->userspace_addr -
+                (uintptr_t) qemu_get_ram_block_host_ptr(ram_addr);
+            assert(fd_num < VHOST_MEMORY_MAX_NREGIONS);
+            fds[fd_num++] = fd;
+        }
     }
 
-    /*
-     * For non-vring specific requests, like VHOST_USER_SET_MEM_TABLE,
-     * we just need send it once in the first time. For later such
-     * request, we just ignore it.
-     */
-    if (vhost_user_one_time_request(msg_request) && dev->vq_index != 0) {
-        return 0;
+    msg.memory.nregions = fd_num;
+
+    if (!fd_num) {
+        error_report("Failed initializing vhost-user memory map, "
+                     "consider using -object memory-backend-file share=on");
+        return -1;
     }
 
-    msg.request = msg_request;
-    msg.flags = VHOST_USER_VERSION;
-    msg.size = 0;
+    msg.size = sizeof(m.memory.nregions);
+    msg.size += sizeof(m.memory.padding);
+    msg.size += fd_num * sizeof(VhostUserMemoryRegion);
 
-    switch (msg_request) {
-    case VHOST_USER_GET_FEATURES:
-    case VHOST_USER_GET_PROTOCOL_FEATURES:
-    case VHOST_USER_GET_QUEUE_NUM:
-        need_reply = 1;
-        break;
+    vhost_user_write(dev, &msg, fds, fd_num);
 
-    case VHOST_USER_SET_FEATURES:
-    case VHOST_USER_SET_PROTOCOL_FEATURES:
-        msg.u64 = *((__u64 *) arg);
-        msg.size = sizeof(m.u64);
-        break;
+    return 0;
+}
 
-    case VHOST_USER_SET_OWNER:
-    case VHOST_USER_RESET_DEVICE:
-        break;
+static int vhost_user_set_vring_addr(struct vhost_dev *dev,
+                                     struct vhost_vring_addr *addr)
+{
+    VhostUserMsg msg = {
+        .request = VHOST_USER_SET_VRING_ADDR,
+        .flags = VHOST_USER_VERSION,
+        .addr = *addr,
+        .size = sizeof(*addr),
+    };
 
-    case VHOST_USER_SET_MEM_TABLE:
-        for (i = 0; i < dev->mem->nregions; ++i) {
-            struct vhost_memory_region *reg = dev->mem->regions + i;
-            ram_addr_t ram_addr;
-
-            assert((uintptr_t)reg->userspace_addr == reg->userspace_addr);
-            qemu_ram_addr_from_host((void *)(uintptr_t)reg->userspace_addr, &ram_addr);
-            fd = qemu_get_ram_fd(ram_addr);
-            if (fd > 0) {
-                msg.memory.regions[fd_num].userspace_addr = reg->userspace_addr;
-                msg.memory.regions[fd_num].memory_size  = reg->memory_size;
-                msg.memory.regions[fd_num].guest_phys_addr = reg->guest_phys_addr;
-                msg.memory.regions[fd_num].mmap_offset = reg->userspace_addr -
-                    (uintptr_t) qemu_get_ram_block_host_ptr(ram_addr);
-                assert(fd_num < VHOST_MEMORY_MAX_NREGIONS);
-                fds[fd_num++] = fd;
-            }
-        }
+    vhost_user_write(dev, &msg, NULL, 0);
 
-        msg.memory.nregions = fd_num;
+    return 0;
+}
 
-        if (!fd_num) {
-            error_report("Failed initializing vhost-user memory map, "
-                    "consider using -object memory-backend-file share=on");
-            return -1;
-        }
+static int vhost_user_set_vring_endian(struct vhost_dev *dev,
+                                       struct vhost_vring_state *ring)
+{
+    error_report("vhost-user trying to send unhandled ioctl");
+    return -1;
+}
 
-        msg.size = sizeof(m.memory.nregions);
-        msg.size += sizeof(m.memory.padding);
-        msg.size += fd_num * sizeof(VhostUserMemoryRegion);
-
-        break;
-
-    case VHOST_USER_SET_LOG_FD:
-        fds[fd_num++] = *((int *) arg);
-        break;
-
-    case VHOST_USER_SET_VRING_NUM:
-    case VHOST_USER_SET_VRING_BASE:
-    case VHOST_USER_SET_VRING_ENABLE:
-        memcpy(&msg.state, arg, sizeof(struct vhost_vring_state));
-        msg.size = sizeof(m.state);
-        break;
-
-    case VHOST_USER_GET_VRING_BASE:
-        memcpy(&msg.state, arg, sizeof(struct vhost_vring_state));
-        msg.size = sizeof(m.state);
-        need_reply = 1;
-        break;
-
-    case VHOST_USER_SET_VRING_ADDR:
-        memcpy(&msg.addr, arg, sizeof(struct vhost_vring_addr));
-        msg.size = sizeof(m.addr);
-        break;
-
-    case VHOST_USER_SET_VRING_KICK:
-    case VHOST_USER_SET_VRING_CALL:
-    case VHOST_USER_SET_VRING_ERR:
-        file = arg;
-        msg.u64 = file->index & VHOST_USER_VRING_IDX_MASK;
-        msg.size = sizeof(m.u64);
-        if (ioeventfd_enabled() && file->fd > 0) {
-            fds[fd_num++] = file->fd;
-        } else {
-            msg.u64 |= VHOST_USER_VRING_NOFD_MASK;
-        }
-        break;
-    default:
-        error_report("vhost-user trying to send unhandled ioctl");
+static int vhost_set_vring(struct vhost_dev *dev,
+                           unsigned long int request,
+                           struct vhost_vring_state *ring)
+{
+    VhostUserMsg msg = {
+        .request = request,
+        .flags = VHOST_USER_VERSION,
+        .state = *ring,
+        .size = sizeof(*ring),
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    return 0;
+}
+
+static int vhost_user_set_vring_num(struct vhost_dev *dev,
+                                    struct vhost_vring_state *ring)
+{
+    return vhost_set_vring(dev, VHOST_SET_VRING_NUM, ring);
+}
+
+static int vhost_user_set_vring_base(struct vhost_dev *dev,
+                                     struct vhost_vring_state *ring)
+{
+    return vhost_set_vring(dev, VHOST_SET_VRING_BASE, ring);
+}
+
+static int vhost_user_set_vring_enable(struct vhost_dev *dev, int enable)
+{
+    struct vhost_vring_state state = {
+        .index = dev->vq_index,
+        .num   = enable,
+    };
+
+    if (!(dev->protocol_features & (1ULL << VHOST_USER_PROTOCOL_F_MQ))) {
         return -1;
-        break;
     }
 
-    if (vhost_user_write(dev, &msg, fds, fd_num) < 0) {
+    return vhost_set_vring(dev, VHOST_USER_SET_VRING_ENABLE, &state);
+}
+
+
+static int vhost_user_get_vring_base(struct vhost_dev *dev,
+                                     struct vhost_vring_state *ring)
+{
+    VhostUserMsg msg = {
+        .request = VHOST_USER_GET_VRING_BASE,
+        .flags = VHOST_USER_VERSION,
+        .state = *ring,
+        .size = sizeof(*ring),
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    if (vhost_user_read(dev, &msg) < 0) {
         return 0;
     }
 
-    if (need_reply) {
-        if (vhost_user_read(dev, &msg) < 0) {
-            return 0;
-        }
-
-        if (msg_request != msg.request) {
-            error_report("Received unexpected msg type."
-                    " Expected %d received %d", msg_request, msg.request);
-            return -1;
-        }
+    if (msg.request != VHOST_USER_GET_VRING_BASE) {
+        error_report("Received unexpected msg type. Expected %d received %d",
+                     VHOST_USER_GET_VRING_BASE, msg.request);
+        return -1;
+    }
 
-        switch (msg_request) {
-        case VHOST_USER_GET_FEATURES:
-        case VHOST_USER_GET_PROTOCOL_FEATURES:
-        case VHOST_USER_GET_QUEUE_NUM:
-            if (msg.size != sizeof(m.u64)) {
-                error_report("Received bad msg size.");
-                return -1;
-            }
-            *((__u64 *) arg) = msg.u64;
-            break;
-        case VHOST_USER_GET_VRING_BASE:
-            if (msg.size != sizeof(m.state)) {
-                error_report("Received bad msg size.");
-                return -1;
-            }
-            memcpy(arg, &msg.state, sizeof(struct vhost_vring_state));
-            break;
-        default:
-            error_report("Received unexpected msg type.");
-            return -1;
-            break;
-        }
+    if (msg.size != sizeof(m.state)) {
+        error_report("Received bad msg size.");
+        return -1;
     }
 
+    *ring = msg.state;
+
     return 0;
 }
 
-static int vhost_set_log_base(struct vhost_dev *dev, uint64_t base,
-                              struct vhost_log *log)
+static int vhost_set_vring_file(struct vhost_dev *dev,
+                                VhostUserRequest request,
+                                struct vhost_vring_file *file)
 {
     int fds[VHOST_MEMORY_MAX_NREGIONS];
     size_t fd_num = 0;
-    bool shmfd = virtio_has_feature(dev->protocol_features,
-                                    VHOST_USER_PROTOCOL_F_LOG_SHMFD);
     VhostUserMsg msg = {
-        .request = VHOST_USER_SET_LOG_BASE,
+        .request = request,
         .flags = VHOST_USER_VERSION,
-        .u64 = base,
+        .u64 = file->index & VHOST_USER_VRING_IDX_MASK,
         .size = sizeof(m.u64),
     };
 
-    if (shmfd && log->fd != -1) {
-        fds[fd_num++] = log->fd;
+    if (ioeventfd_enabled() && file->fd > 0) {
+        fds[fd_num++] = file->fd;
+    } else {
+        msg.u64 |= VHOST_USER_VRING_NOFD_MASK;
     }
 
     vhost_user_write(dev, &msg, fds, fd_num);
 
-    if (shmfd) {
-        msg.size = 0;
-        if (vhost_user_read(dev, &msg) < 0) {
-            return 0;
-        }
+    return 0;
+}
 
-        if (msg.request != VHOST_USER_SET_LOG_BASE) {
-            error_report("Received unexpected msg type. "
-                         "Expected %d received %d",
-                         VHOST_USER_SET_LOG_BASE, msg.request);
-            return -1;
-        }
+static int vhost_user_set_vring_kick(struct vhost_dev *dev,
+                                     struct vhost_vring_file *file)
+{
+    return vhost_set_vring_file(dev, VHOST_USER_SET_VRING_KICK, file);
+}
+
+static int vhost_user_set_vring_call(struct vhost_dev *dev,
+                                     struct vhost_vring_file *file)
+{
+    return vhost_set_vring_file(dev, VHOST_USER_SET_VRING_CALL, file);
+}
+
+static int vhost_user_set_u64(struct vhost_dev *dev, int request, uint64_t u64)
+{
+    VhostUserMsg msg = {
+        .request = request,
+        .flags = VHOST_USER_VERSION,
+        .u64 = u64,
+        .size = sizeof(m.u64),
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    return 0;
+}
+
+static int vhost_user_set_features(struct vhost_dev *dev,
+                                   uint64_t features)
+{
+    return vhost_user_set_u64(dev, VHOST_USER_SET_FEATURES, features);
+}
+
+static int vhost_user_set_protocol_features(struct vhost_dev *dev,
+                                            uint64_t features)
+{
+    return vhost_user_set_u64(dev, VHOST_USER_SET_PROTOCOL_FEATURES, features);
+}
+
+static int vhost_user_get_u64(struct vhost_dev *dev, int request, uint64_t *u64)
+{
+    VhostUserMsg msg = {
+        .request = request,
+        .flags = VHOST_USER_VERSION,
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    if (vhost_user_read(dev, &msg) < 0) {
+        return 0;
+    }
+
+    if (msg.request != request) {
+        error_report("Received unexpected msg type. Expected %d received %d",
+                     request, msg.request);
+        return -1;
+    }
+
+    if (msg.size != sizeof(m.u64)) {
+        error_report("Received bad msg size.");
+        return -1;
     }
 
+    *u64 = msg.u64;
+
+    return 0;
+}
+
+static int vhost_user_get_features(struct vhost_dev *dev, uint64_t *features)
+{
+    return vhost_user_get_u64(dev, VHOST_USER_GET_FEATURES, features);
+}
+
+static int vhost_user_set_owner(struct vhost_dev *dev)
+{
+    VhostUserMsg msg = {
+        .request = VHOST_USER_SET_OWNER,
+        .flags = VHOST_USER_VERSION,
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
+    return 0;
+}
+
+static int vhost_user_reset_device(struct vhost_dev *dev)
+{
+    VhostUserMsg msg = {
+        .request = VHOST_USER_RESET_DEVICE,
+        .flags = VHOST_USER_VERSION,
+    };
+
+    vhost_user_write(dev, &msg, NULL, 0);
+
     return 0;
 }
 
 static int vhost_user_init(struct vhost_dev *dev, void *opaque)
 {
-    unsigned long long features;
+    uint64_t features;
     int err;
 
     assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
 
     dev->opaque = opaque;
 
-    err = vhost_user_call(dev, VHOST_USER_GET_FEATURES, &features);
+    err = vhost_user_get_features(dev, &features);
     if (err < 0) {
         return err;
     }
@@ -422,21 +500,22 @@ static int vhost_user_init(struct vhost_dev *dev, void *opaque)
     if (virtio_has_feature(features, VHOST_USER_F_PROTOCOL_FEATURES)) {
         dev->backend_features |= 1ULL << VHOST_USER_F_PROTOCOL_FEATURES;
 
-        err = vhost_user_call(dev, VHOST_USER_GET_PROTOCOL_FEATURES, &features);
+        err = vhost_user_get_u64(dev, VHOST_USER_GET_PROTOCOL_FEATURES,
+                                 &features);
         if (err < 0) {
             return err;
         }
 
         dev->protocol_features = features & VHOST_USER_PROTOCOL_FEATURE_MASK;
-        err = vhost_user_call(dev, VHOST_USER_SET_PROTOCOL_FEATURES,
-                              &dev->protocol_features);
+        err = vhost_user_set_protocol_features(dev, dev->protocol_features);
         if (err < 0) {
             return err;
         }
 
         /* query the max queues we support if backend supports Multiple Queue */
         if (dev->protocol_features & (1ULL << VHOST_USER_PROTOCOL_F_MQ)) {
-            err = vhost_user_call(dev, VHOST_USER_GET_QUEUE_NUM, &dev->max_queues);
+            err = vhost_user_get_u64(dev, VHOST_USER_GET_QUEUE_NUM,
+                                     &dev->max_queues);
             if (err < 0) {
                 return err;
             }
@@ -454,22 +533,6 @@ static int vhost_user_init(struct vhost_dev *dev, void *opaque)
     return 0;
 }
 
-static int vhost_user_set_vring_enable(struct vhost_dev *dev, int enable)
-{
-    struct vhost_vring_state state = {
-        .index = dev->vq_index,
-        .num   = enable,
-    };
-
-    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
-
-    if (!(dev->protocol_features & (1ULL << VHOST_USER_PROTOCOL_F_MQ))) {
-        return -1;
-    }
-
-    return vhost_user_call(dev, VHOST_USER_SET_VRING_ENABLE, &state);
-}
-
 static int vhost_user_cleanup(struct vhost_dev *dev)
 {
     assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
@@ -492,11 +555,22 @@ static int vhost_user_get_vq_index(struct vhost_dev *dev, int idx)
 
 const VhostOps user_ops = {
         .backend_type = VHOST_BACKEND_TYPE_USER,
-        .vhost_call = vhost_user_call,
         .vhost_backend_init = vhost_user_init,
         .vhost_backend_cleanup = vhost_user_cleanup,
-        .vhost_backend_get_vq_index = vhost_user_get_vq_index,
-        .vhost_backend_set_vring_enable = vhost_user_set_vring_enable,
 
-        .vhost_set_log_base = vhost_set_log_base,
+        .vhost_set_log_base = vhost_user_set_log_base,
+        .vhost_set_mem_table = vhost_user_set_mem_table,
+        .vhost_set_vring_addr = vhost_user_set_vring_addr,
+        .vhost_set_vring_endian = vhost_user_set_vring_endian,
+        .vhost_set_vring_num = vhost_user_set_vring_num,
+        .vhost_set_vring_base = vhost_user_set_vring_base,
+        .vhost_get_vring_base = vhost_user_get_vring_base,
+        .vhost_set_vring_kick = vhost_user_set_vring_kick,
+        .vhost_set_vring_call = vhost_user_set_vring_call,
+        .vhost_set_features = vhost_user_set_features,
+        .vhost_get_features = vhost_user_get_features,
+        .vhost_set_owner = vhost_user_set_owner,
+        .vhost_reset_device = vhost_user_reset_device,
+        .vhost_get_vq_index = vhost_user_get_vq_index,
+        .vhost_set_vring_enable = vhost_user_set_vring_enable,
 };
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 9b73f83..d9f2131 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -532,7 +532,7 @@ static void vhost_commit(MemoryListener *listener)
     }
 
     if (!dev->log_enabled) {
-        r = dev->vhost_ops->vhost_call(dev, VHOST_SET_MEM_TABLE, dev->mem);
+        r = dev->vhost_ops->vhost_set_mem_table(dev, dev->mem);
         assert(r >= 0);
         dev->memory_changed = false;
         return;
@@ -545,7 +545,7 @@ static void vhost_commit(MemoryListener *listener)
     if (dev->log_size < log_size) {
         vhost_dev_log_resize(dev, log_size + VHOST_LOG_BUFFER);
     }
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_MEM_TABLE, dev->mem);
+    r = dev->vhost_ops->vhost_set_mem_table(dev, dev->mem);
     assert(r >= 0);
     /* To log less, can only decrease log size after table update. */
     if (dev->log_size > log_size + VHOST_LOG_BUFFER) {
@@ -613,7 +613,7 @@ static int vhost_virtqueue_set_addr(struct vhost_dev *dev,
         .log_guest_addr = vq->used_phys,
         .flags = enable_log ? (1 << VHOST_VRING_F_LOG) : 0,
     };
-    int r = dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_ADDR, &addr);
+    int r = dev->vhost_ops->vhost_set_vring_addr(dev, &addr);
     if (r < 0) {
         return -errno;
     }
@@ -627,7 +627,7 @@ static int vhost_dev_set_features(struct vhost_dev *dev, bool enable_log)
     if (enable_log) {
         features |= 0x1ULL << VHOST_F_LOG_ALL;
     }
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_FEATURES, &features);
+    r = dev->vhost_ops->vhost_set_features(dev, features);
     return r < 0 ? -errno : 0;
 }
 
@@ -732,7 +732,7 @@ static int vhost_virtqueue_set_vring_endian_legacy(struct vhost_dev *dev,
         .num = is_big_endian
     };
 
-    if (!dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_ENDIAN, &s)) {
+    if (!dev->vhost_ops->vhost_set_vring_endian(dev, &s)) {
         return 0;
     }
 
@@ -751,7 +751,7 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
 {
     hwaddr s, l, a;
     int r;
-    int vhost_vq_index = dev->vhost_ops->vhost_backend_get_vq_index(dev, idx);
+    int vhost_vq_index = dev->vhost_ops->vhost_get_vq_index(dev, idx);
     struct vhost_vring_file file = {
         .index = vhost_vq_index
     };
@@ -762,13 +762,13 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
 
 
     vq->num = state.num = virtio_queue_get_num(vdev, idx);
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_NUM, &state);
+    r = dev->vhost_ops->vhost_set_vring_num(dev, &state);
     if (r) {
         return -errno;
     }
 
     state.num = virtio_queue_get_last_avail_idx(vdev, idx);
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_BASE, &state);
+    r = dev->vhost_ops->vhost_set_vring_base(dev, &state);
     if (r) {
         return -errno;
     }
@@ -820,7 +820,7 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
     }
 
     file.fd = event_notifier_get_fd(virtio_queue_get_host_notifier(vvq));
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_KICK, &file);
+    r = dev->vhost_ops->vhost_set_vring_kick(dev, &file);
     if (r) {
         r = -errno;
         goto fail_kick;
@@ -853,13 +853,13 @@ static void vhost_virtqueue_stop(struct vhost_dev *dev,
                                     struct vhost_virtqueue *vq,
                                     unsigned idx)
 {
-    int vhost_vq_index = dev->vhost_ops->vhost_backend_get_vq_index(dev, idx);
+    int vhost_vq_index = dev->vhost_ops->vhost_get_vq_index(dev, idx);
     struct vhost_vring_state state = {
         .index = vhost_vq_index,
     };
     int r;
 
-    r = dev->vhost_ops->vhost_call(dev, VHOST_GET_VRING_BASE, &state);
+    r = dev->vhost_ops->vhost_get_vring_base(dev, &state);
     if (r < 0) {
         fprintf(stderr, "vhost VQ %d ring restore failed: %d\n", idx, r);
         fflush(stderr);
@@ -906,7 +906,7 @@ static void vhost_eventfd_del(MemoryListener *listener,
 static int vhost_virtqueue_init(struct vhost_dev *dev,
                                 struct vhost_virtqueue *vq, int n)
 {
-    int vhost_vq_index = dev->vhost_ops->vhost_backend_get_vq_index(dev, n);
+    int vhost_vq_index = dev->vhost_ops->vhost_get_vq_index(dev, n);
     struct vhost_vring_file file = {
         .index = vhost_vq_index,
     };
@@ -916,7 +916,7 @@ static int vhost_virtqueue_init(struct vhost_dev *dev,
     }
 
     file.fd = event_notifier_get_fd(&vq->masked_notifier);
-    r = dev->vhost_ops->vhost_call(dev, VHOST_SET_VRING_CALL, &file);
+    r = dev->vhost_ops->vhost_set_vring_call(dev, &file);
     if (r) {
         r = -errno;
         goto fail_call;
@@ -948,12 +948,12 @@ int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
         return -errno;
     }
 
-    r = hdev->vhost_ops->vhost_call(hdev, VHOST_SET_OWNER, NULL);
+    r = hdev->vhost_ops->vhost_set_owner(hdev);
     if (r < 0) {
         goto fail;
     }
 
-    r = hdev->vhost_ops->vhost_call(hdev, VHOST_GET_FEATURES, &features);
+    r = hdev->vhost_ops->vhost_get_features(hdev, &features);
     if (r < 0) {
         goto fail;
     }
@@ -1106,8 +1106,8 @@ void vhost_virtqueue_mask(struct vhost_dev *hdev, VirtIODevice *vdev, int n,
         file.fd = event_notifier_get_fd(virtio_queue_get_guest_notifier(vvq));
     }
 
-    file.index = hdev->vhost_ops->vhost_backend_get_vq_index(hdev, n);
-    r = hdev->vhost_ops->vhost_call(hdev, VHOST_SET_VRING_CALL, &file);
+    file.index = hdev->vhost_ops->vhost_get_vq_index(hdev, n);
+    r = hdev->vhost_ops->vhost_set_vring_call(hdev, &file);
     assert(r >= 0);
 }
 
@@ -1149,7 +1149,7 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
     if (r < 0) {
         goto fail_features;
     }
-    r = hdev->vhost_ops->vhost_call(hdev, VHOST_SET_MEM_TABLE, hdev->mem);
+    r = hdev->vhost_ops->vhost_set_mem_table(hdev, hdev->mem);
     if (r < 0) {
         r = -errno;
         goto fail_mem;
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index 485279b..868a78a 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -20,27 +20,76 @@ typedef enum VhostBackendType {
 
 struct vhost_dev;
 struct vhost_log;
+struct vhost_memory;
+struct vhost_vring_file;
+struct vhost_vring_state;
+struct vhost_vring_addr;
+struct vhost_scsi_target;
 
-typedef int (*vhost_call)(struct vhost_dev *dev, unsigned long int request,
-             void *arg);
 typedef int (*vhost_backend_init)(struct vhost_dev *dev, void *opaque);
 typedef int (*vhost_backend_cleanup)(struct vhost_dev *dev);
-typedef int (*vhost_backend_get_vq_index)(struct vhost_dev *dev, int idx);
-typedef int (*vhost_backend_set_vring_enable)(struct vhost_dev *dev, int enable);
 
+typedef int (*vhost_net_set_backend_op)(struct vhost_dev *dev,
+                                struct vhost_vring_file *file);
+typedef int (*vhost_scsi_set_endpoint_op)(struct vhost_dev *dev,
+                                  struct vhost_scsi_target *target);
+typedef int (*vhost_scsi_clear_endpoint_op)(struct vhost_dev *dev,
+                                    struct vhost_scsi_target *target);
+typedef int (*vhost_scsi_get_abi_version_op)(struct vhost_dev *dev,
+                                             int *version);
 typedef int (*vhost_set_log_base_op)(struct vhost_dev *dev, uint64_t base,
                                      struct vhost_log *log);
+typedef int (*vhost_set_mem_table_op)(struct vhost_dev *dev,
+                                      struct vhost_memory *mem);
+typedef int (*vhost_set_vring_addr_op)(struct vhost_dev *dev,
+                                       struct vhost_vring_addr *addr);
+typedef int (*vhost_set_vring_endian_op)(struct vhost_dev *dev,
+                                         struct vhost_vring_state *ring);
+typedef int (*vhost_set_vring_num_op)(struct vhost_dev *dev,
+                                      struct vhost_vring_state *ring);
+typedef int (*vhost_set_vring_base_op)(struct vhost_dev *dev,
+                                       struct vhost_vring_state *ring);
+typedef int (*vhost_get_vring_base_op)(struct vhost_dev *dev,
+                                       struct vhost_vring_state *ring);
+typedef int (*vhost_set_vring_kick_op)(struct vhost_dev *dev,
+                                       struct vhost_vring_file *file);
+typedef int (*vhost_set_vring_call_op)(struct vhost_dev *dev,
+                                       struct vhost_vring_file *file);
+typedef int (*vhost_set_features_op)(struct vhost_dev *dev,
+                                     uint64_t features);
+typedef int (*vhost_get_features_op)(struct vhost_dev *dev,
+                                     uint64_t *features);
+typedef int (*vhost_set_owner_op)(struct vhost_dev *dev);
+typedef int (*vhost_reset_device_op)(struct vhost_dev *dev);
+typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
+typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
+                                         int enable);
 
 typedef struct VhostOps {
     VhostBackendType backend_type;
-    vhost_call vhost_call;
 
     vhost_backend_init vhost_backend_init;
     vhost_backend_cleanup vhost_backend_cleanup;
-    vhost_backend_get_vq_index vhost_backend_get_vq_index;
-    vhost_backend_set_vring_enable vhost_backend_set_vring_enable;
 
+    vhost_net_set_backend_op vhost_net_set_backend;
+    vhost_scsi_set_endpoint_op vhost_scsi_set_endpoint;
+    vhost_scsi_clear_endpoint_op vhost_scsi_clear_endpoint;
+    vhost_scsi_get_abi_version_op vhost_scsi_get_abi_version;
     vhost_set_log_base_op vhost_set_log_base;
+    vhost_set_mem_table_op vhost_set_mem_table;
+    vhost_set_vring_addr_op vhost_set_vring_addr;
+    vhost_set_vring_endian_op vhost_set_vring_endian;
+    vhost_set_vring_num_op vhost_set_vring_num;
+    vhost_set_vring_base_op vhost_set_vring_base;
+    vhost_get_vring_base_op vhost_get_vring_base;
+    vhost_set_vring_kick_op vhost_set_vring_kick;
+    vhost_set_vring_call_op vhost_set_vring_call;
+    vhost_set_features_op vhost_set_features;
+    vhost_get_features_op vhost_get_features;
+    vhost_set_owner_op vhost_set_owner;
+    vhost_reset_device_op vhost_reset_device;
+    vhost_get_vq_index_op vhost_get_vq_index;
+    vhost_set_vring_enable_op vhost_set_vring_enable;
 } VhostOps;
 
 extern const VhostOps user_ops;
diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h
index 7e7dc45..c8c42cc 100644
--- a/include/hw/virtio/vhost.h
+++ b/include/hw/virtio/vhost.h
@@ -45,14 +45,14 @@ struct vhost_dev {
     int nvqs;
     /* the first virtqueue which would be used by this vhost dev */
     int vq_index;
-    unsigned long long features;
-    unsigned long long acked_features;
-    unsigned long long backend_features;
-    unsigned long long protocol_features;
-    unsigned long long max_queues;
+    uint64_t features;
+    uint64_t acked_features;
+    uint64_t backend_features;
+    uint64_t protocol_features;
+    uint64_t max_queues;
     bool started;
     bool log_enabled;
-    unsigned long long log_size;
+    uint64_t log_size;
     Error *migration_blocker;
     bool memory_changed;
     hwaddr mem_changed_start_addr;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (9 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 10/21] vhost: use a function for each call marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-29 14:59   ` Michael S. Tsirkin
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log marcandre.lureau
                   ` (11 subsequent siblings)
  22 siblings, 1 reply; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Do not allocate a shared log if the backend doesn't support it.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 hw/virtio/vhost-user.c            |  9 +++++++++
 hw/virtio/vhost.c                 | 15 ++++++++++-----
 include/hw/virtio/vhost-backend.h |  4 ++++
 3 files changed, 23 insertions(+), 5 deletions(-)

diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index 41e0364..455caba 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -553,6 +553,14 @@ static int vhost_user_get_vq_index(struct vhost_dev *dev, int idx)
     return idx;
 }
 
+static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
+{
+    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
+
+    return virtio_has_feature(dev->protocol_features,
+                              VHOST_USER_PROTOCOL_F_LOG_SHMFD);
+}
+
 const VhostOps user_ops = {
         .backend_type = VHOST_BACKEND_TYPE_USER,
         .vhost_backend_init = vhost_user_init,
@@ -573,4 +581,5 @@ const VhostOps user_ops = {
         .vhost_reset_device = vhost_user_reset_device,
         .vhost_get_vq_index = vhost_user_get_vq_index,
         .vhost_set_vring_enable = vhost_user_set_vring_enable,
+        .vhost_requires_shm_log = vhost_user_requires_shm_log,
 };
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index d9f2131..ec27c7b 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -357,10 +357,15 @@ static void vhost_log_put(struct vhost_dev *dev, bool sync)
     }
 }
 
-static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
+static bool vhost_dev_log_is_shared(struct vhost_dev *dev)
 {
-    bool share = dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
-    struct vhost_log *log = vhost_log_get(size, share);
+    return dev->vhost_ops->vhost_requires_shm_log &&
+           dev->vhost_ops->vhost_requires_shm_log(dev);
+}
+
+static inline void vhost_dev_log_resize(struct vhost_dev *dev, uint64_t size)
+{
+    struct vhost_log *log = vhost_log_get(size, vhost_dev_log_is_shared(dev));
     uint64_t log_base = (uintptr_t)log->log;
     int r;
 
@@ -1166,10 +1171,10 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
 
     if (hdev->log_enabled) {
         uint64_t log_base;
-        bool share = hdev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
 
         hdev->log_size = vhost_get_log_size(hdev);
-        hdev->log = vhost_log_get(hdev->log_size, share);
+        hdev->log = vhost_log_get(hdev->log_size,
+                                  vhost_dev_log_is_shared(hdev));
         log_base = (uintptr_t)hdev->log->log;
         r = hdev->vhost_ops->vhost_set_log_base(hdev,
                                                 hdev->log_size ? log_base : 0,
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index 868a78a..e07118c 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -11,6 +11,8 @@
 #ifndef VHOST_BACKEND_H_
 #define VHOST_BACKEND_H_
 
+#include <stdbool.h>
+
 typedef enum VhostBackendType {
     VHOST_BACKEND_TYPE_NONE = 0,
     VHOST_BACKEND_TYPE_KERNEL = 1,
@@ -64,6 +66,7 @@ typedef int (*vhost_reset_device_op)(struct vhost_dev *dev);
 typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
 typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
                                          int enable);
+typedef bool (*vhost_requires_shm_log_op)(struct vhost_dev *dev);
 
 typedef struct VhostOps {
     VhostBackendType backend_type;
@@ -90,6 +93,7 @@ typedef struct VhostOps {
     vhost_reset_device_op vhost_reset_device;
     vhost_get_vq_index_op vhost_get_vq_index;
     vhost_set_vring_enable_op vhost_set_vring_enable;
+    vhost_requires_shm_log_op vhost_requires_shm_log;
 } VhostOps;
 
 extern const VhostOps user_ops;
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (10 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:38   ` Eric Blake
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 13/21] net: add trace_vhost_user_event marcandre.lureau
                   ` (10 subsequent siblings)
  22 siblings, 1 reply; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 docs/specs/vhost-user.txt | 48 +++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 46 insertions(+), 2 deletions(-)

diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
index 4eadad1..e0292a0 100644
--- a/docs/specs/vhost-user.txt
+++ b/docs/specs/vhost-user.txt
@@ -115,11 +115,13 @@ the ones that do:
  * VHOST_GET_FEATURES
  * VHOST_GET_PROTOCOL_FEATURES
  * VHOST_GET_VRING_BASE
+ * VHOST_SET_LOG_BASE (if VHOST_USER_PROTOCOL_F_LOG_SHMFD)
 
 There are several messages that the master sends with file descriptors passed
 in the ancillary data:
 
  * VHOST_SET_MEM_TABLE
+ * VHOST_SET_LOG_BASE (if VHOST_USER_PROTOCOL_F_LOG_SHMFD)
  * VHOST_SET_LOG_FD
  * VHOST_SET_VRING_KICK
  * VHOST_SET_VRING_CALL
@@ -140,8 +142,7 @@ Multiple queue support
 
 Multiple queue is treated as a protocol extension, hence the slave has to
 implement protocol features first. The multiple queues feature is supported
-only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set:
-#define VHOST_USER_PROTOCOL_F_MQ    0
+only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set.
 
 The max number of queues the slave supports can be queried with message
 VHOST_USER_GET_PROTOCOL_FEATURES. Master should stop when the number of
@@ -152,6 +153,48 @@ queue in the sent message to identify a specified queue. One queue pair
 is enabled initially. More queues are enabled dynamically, by sending
 message VHOST_USER_SET_VRING_ENABLE.
 
+Migration
+---------
+
+During live migration, the master may need to track the modifications
+the slave makes to the memory mapped regions. The client should mark
+the dirty pages in a log. Once it complies to this logging, it may
+declare the VHOST_F_LOG_ALL vhost feature.
+
+All the modifications to memory pointed by vring "descriptor" should
+be marked. Modifications to "used" vring should be marked if
+VHOST_VRING_F_LOG is part of ring's features.
+
+Dirty pages are of size:
+#define VHOST_LOG_PAGE 0x1000
+
+The log memory fd is provided in the ancillary data of
+VHOST_USER_SET_LOG_BASE message when the slave has
+VHOST_USER_PROTOCOL_F_LOG_SHMFD protocol feature.
+
+The size of the log may be computed by using all the known guest
+addresses. The log covers from address 0 to the maximum of guest
+regions. In pseudo-code, to mark page at "addr" as dirty:
+
+page = addr / VHOST_LOG_PAGE
+log[page / 8] |= 1 << page % 8
+
+Use atomic operations, as the log may be concurrently manipulated.
+
+VHOST_USER_SET_LOG_FD is an optional message with an eventfd in
+ancillary data, it may be used to inform the master that the log has
+been modified.
+
+Once the source has finished migration, VHOST_USER_RESET_OWNER message
+will be sent by the source. No further update must be done before the
+destination takes over with new regions & rings.
+
+Protocol features
+-----------------
+
+#define VHOST_USER_PROTOCOL_F_MQ             0
+#define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
+
 Message types
 -------------
 
@@ -236,6 +279,7 @@ Message types
       Id: 6
       Equivalent ioctl: VHOST_SET_LOG_BASE
       Master payload: u64
+      Slave payload: N/A
 
       Sets the logging base address.
 
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 13/21] net: add trace_vhost_user_event
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (11 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 14/21] vhost user: add support of live migration marcandre.lureau
                   ` (9 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Replace error_report() and use tracing instead. It's not an error to get
a connection or a disconnection, so silence this and trace it instead.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 net/vhost-user.c | 4 ++--
 trace-events     | 3 +++
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/vhost-user.c b/net/vhost-user.c
index 8f354eb..9b38431 100644
--- a/net/vhost-user.c
+++ b/net/vhost-user.c
@@ -15,6 +15,7 @@
 #include "qemu/config-file.h"
 #include "qemu/error-report.h"
 #include "qmp-commands.h"
+#include "trace.h"
 
 typedef struct VhostUserState {
     NetClientState nc;
@@ -148,18 +149,17 @@ static void net_vhost_user_event(void *opaque, int event)
                                           NET_CLIENT_OPTIONS_KIND_NIC,
                                           MAX_QUEUE_NUM);
     s = DO_UPCAST(VhostUserState, nc, ncs[0]);
+    trace_vhost_user_event(s->chr->label, event);
     switch (event) {
     case CHR_EVENT_OPENED:
         if (vhost_user_start(queues, ncs) < 0) {
             exit(1);
         }
         qmp_set_link(name, true, &err);
-        error_report("chardev \"%s\" went up", s->chr->label);
         break;
     case CHR_EVENT_CLOSED:
         qmp_set_link(name, true, &err);
         vhost_user_stop(queues, ncs);
-        error_report("chardev \"%s\" went down", s->chr->label);
         break;
     }
 
diff --git a/trace-events b/trace-events
index 25c53e0..0434705 100644
--- a/trace-events
+++ b/trace-events
@@ -1694,3 +1694,6 @@ qcrypto_tls_creds_x509_load_cert_list(void *creds, const char *file) "TLS creds
 
 # crypto/tlssession.c
 qcrypto_tls_session_new(void *session, void *creds, const char *hostname, const char *aclname, int endpoint) "TLS session new session=%p creds=%p hostname=%s aclname=%s endpoint=%d"
+
+# net/vhost-user.c
+vhost_user_event(const char *chr, int event) "chr: %s got event: %d"
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 14/21] vhost user: add support of live migration
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (12 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 13/21] net: add trace_vhost_user_event marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest marcandre.lureau
                   ` (8 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel; +Cc: thibaut.collet, jasowang, mst, haifeng.lin, pbonzini

From: Thibaut Collet <thibaut.collet@6wind.com>

Some vhost user backends are able to support live migration.
To provide this service the following features must be added:
1. Add the VIRTIO_NET_F_GUEST_ANNOUNCE capability to vhost-net when netdev
   backend is vhost-user.
2. Provide a nop receive callback to vhost-user.
   This callback is called by:
    *  qemu_announce_self after a migration to send fake RARP to avoid network
       outage for peers talking to the migrated guest.
         - For guest with GUEST_ANNOUNCE capabilities, guest already sends GARP
           when the bit VIRTIO_NET_S_ANNOUNCE is set.
           => These packets must be discarded.
         - For guest without GUEST_ANNOUNCE capabilities, migration termination
           is notified when the guest sends packets.
           => These packets can be discarded.
    * virtio_net_tx_bh with a dummy boot to send fake bootp/dhcp request.
      BIOS guest manages virtio driver to send 4 bootp/dhcp request in case of
      dummy boot.
      => These packets must be discarded.

Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
---
 hw/net/vhost_net.c |  2 ++
 net/vhost-user.c   | 10 ++++++++++
 2 files changed, 12 insertions(+)

diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
index 1ab4133..840f443 100644
--- a/hw/net/vhost_net.c
+++ b/hw/net/vhost_net.c
@@ -85,6 +85,8 @@ static const int user_feature_bits[] = {
     VIRTIO_NET_F_CTRL_MAC_ADDR,
     VIRTIO_NET_F_CTRL_GUEST_OFFLOADS,
 
+    VIRTIO_NET_F_GUEST_ANNOUNCE,
+
     VIRTIO_NET_F_MQ,
 
     VHOST_INVALID_FEATURE_BIT
diff --git a/net/vhost-user.c b/net/vhost-user.c
index 9b38431..87917a5 100644
--- a/net/vhost-user.c
+++ b/net/vhost-user.c
@@ -103,6 +103,15 @@ err:
     return -1;
 }
 
+static ssize_t vhost_user_receive(NetClientState *nc, const uint8_t *buf,
+                                  size_t size)
+{
+    /* Discard the request that is received and managed by backend
+     * by an other way.
+     */
+    return size;
+}
+
 static void vhost_user_cleanup(NetClientState *nc)
 {
     VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
@@ -132,6 +141,7 @@ static bool vhost_user_has_ufo(NetClientState *nc)
 static NetClientInfo net_vhost_user_info = {
         .type = NET_CLIENT_OPTIONS_KIND_VHOST_USER,
         .size = sizeof(VhostUserState),
+        .receive = vhost_user_receive,
         .cleanup = vhost_user_cleanup,
         .has_vnet_hdr = vhost_user_has_vnet_hdr,
         .has_ufo = vhost_user_has_ufo,
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (13 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 14/21] vhost user: add support of live migration marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 17:24   ` Thibaut Collet
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 16/21] vhost-user-test: move wait_for_fds() out marcandre.lureau
                   ` (7 subsequent siblings)
  22 siblings, 1 reply; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Thibaut Collet <thibaut.collet@6wind.com>

A new vhost user message is added to allow QEMU to ask to vhost user backend to
broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
capability.

This new message is sent only if the backend supports the new
VHOST_USER_PROTOCOL_F_RARP protocol feature.
The payload of this new message is the MAC address of the guest (not known by
the backend). The MAC address is copied in the first 6 bytes of a u64 to avoid
to create a new payload message type.

This new message has no equivalent ioctl so a new callback is added in the
userOps structure to send the request.

Upon reception of this new message the vhost user backend must generate and
broadcast a fake RARP request to notify the migration is terminated.

Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
[Rebased and fixed checkpatch errors - Marc-André]
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 docs/specs/vhost-user.txt         | 15 +++++++++++++++
 hw/net/vhost_net.c                | 17 +++++++++++++++++
 hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
 include/hw/virtio/vhost-backend.h |  3 +++
 include/net/vhost_net.h           |  1 +
 net/vhost-user.c                  | 24 ++++++++++++++++++++++--
 6 files changed, 88 insertions(+), 2 deletions(-)

diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
index e0292a0..e0d71e2 100644
--- a/docs/specs/vhost-user.txt
+++ b/docs/specs/vhost-user.txt
@@ -194,6 +194,7 @@ Protocol features
 
 #define VHOST_USER_PROTOCOL_F_MQ             0
 #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
+#define VHOST_USER_PROTOCOL_F_RARP           2
 
 Message types
 -------------
@@ -381,3 +382,17 @@ Message types
       Master payload: vring state description
 
       Signal slave to enable or disable corresponding vring.
+
+ * VHOST_USER_SEND_RARP
+
+      Id: 19
+      Equivalent ioctl: N/A
+      Master payload: u64
+
+      Ask vhost user backend to broadcast a fake RARP to notify the migration
+      is terminated for guest that does not support GUEST_ANNOUNCE.
+      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present in
+      VHOST_USER_GET_FEATURES and protocol feature bit VHOST_USER_PROTOCOL_F_RARP
+      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
+      The first 6 bytes of the payload contain the mac address of the guest to
+      allow the vhost user backend to construct and broadcast the fake RARP.
diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
index 840f443..da66b64 100644
--- a/hw/net/vhost_net.c
+++ b/hw/net/vhost_net.c
@@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
     g_free(net);
 }
 
+int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
+{
+    const VhostOps *vhost_ops = net->dev.vhost_ops;
+    int r = -1;
+
+    if (vhost_ops->vhost_migration_done) {
+        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
+    }
+
+    return r;
+}
+
 bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
 {
     return vhost_virtqueue_pending(&net->dev, idx);
@@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net, VirtIODevice *dev,
 {
 }
 
+int vhost_net_notify_migration_done(struct vhost_net *net)
+{
+    return -1;
+}
+
 VHostNetState *get_vhost_net(NetClientState *nc)
 {
     return 0;
diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index 455caba..b7f3699 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -10,6 +10,7 @@
 
 #include "hw/virtio/vhost.h"
 #include "hw/virtio/vhost-backend.h"
+#include "hw/virtio/virtio-net.h"
 #include "sysemu/char.h"
 #include "sysemu/kvm.h"
 #include "qemu/error-report.h"
@@ -30,6 +31,7 @@
 #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
 #define VHOST_USER_PROTOCOL_F_MQ         0
 #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
+#define VHOST_USER_PROTOCOL_F_RARP       2
 
 typedef enum VhostUserRequest {
     VHOST_USER_NONE = 0,
@@ -51,6 +53,7 @@ typedef enum VhostUserRequest {
     VHOST_USER_SET_PROTOCOL_FEATURES = 16,
     VHOST_USER_GET_QUEUE_NUM = 17,
     VHOST_USER_SET_VRING_ENABLE = 18,
+    VHOST_USER_SEND_RARP = 19,
     VHOST_USER_MAX
 } VhostUserRequest;
 
@@ -561,6 +564,32 @@ static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
                               VHOST_USER_PROTOCOL_F_LOG_SHMFD);
 }
 
+static int vhost_user_migration_done(struct vhost_dev *dev, char* mac_addr)
+{
+    VhostUserMsg msg = { 0 };
+    int err;
+
+    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
+
+    /* If guest supports GUEST_ANNOUNCE do nothing */
+    if (virtio_has_feature(dev->acked_features, VIRTIO_NET_F_GUEST_ANNOUNCE)) {
+        return 0;
+    }
+
+    /* if backend supports VHOST_USER_PROTOCOL_F_RARP ask it to send the RARP */
+    if (virtio_has_feature(dev->protocol_features,
+                           VHOST_USER_PROTOCOL_F_RARP)) {
+        msg.request = VHOST_USER_SEND_RARP;
+        msg.flags = VHOST_USER_VERSION;
+        memcpy((char *)&msg.u64, mac_addr, 6);
+        msg.size = sizeof(m.u64);
+
+        err = vhost_user_write(dev, &msg, NULL, 0);
+        return err;
+    }
+    return -1;
+}
+
 const VhostOps user_ops = {
         .backend_type = VHOST_BACKEND_TYPE_USER,
         .vhost_backend_init = vhost_user_init,
@@ -582,4 +611,5 @@ const VhostOps user_ops = {
         .vhost_get_vq_index = vhost_user_get_vq_index,
         .vhost_set_vring_enable = vhost_user_set_vring_enable,
         .vhost_requires_shm_log = vhost_user_requires_shm_log,
+        .vhost_migration_done = vhost_user_migration_done,
 };
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index e07118c..5918c01 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -67,6 +67,8 @@ typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
 typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
                                          int enable);
 typedef bool (*vhost_requires_shm_log_op)(struct vhost_dev *dev);
+typedef int (*vhost_migration_done_op)(struct vhost_dev *dev,
+                                       char *mac_addr);
 
 typedef struct VhostOps {
     VhostBackendType backend_type;
@@ -94,6 +96,7 @@ typedef struct VhostOps {
     vhost_get_vq_index_op vhost_get_vq_index;
     vhost_set_vring_enable_op vhost_set_vring_enable;
     vhost_requires_shm_log_op vhost_requires_shm_log;
+    vhost_migration_done_op vhost_migration_done;
 } VhostOps;
 
 extern const VhostOps user_ops;
diff --git a/include/net/vhost_net.h b/include/net/vhost_net.h
index 0188c4d..3389b41 100644
--- a/include/net/vhost_net.h
+++ b/include/net/vhost_net.h
@@ -27,6 +27,7 @@ void vhost_net_ack_features(VHostNetState *net, uint64_t features);
 bool vhost_net_virtqueue_pending(VHostNetState *net, int n);
 void vhost_net_virtqueue_mask(VHostNetState *net, VirtIODevice *dev,
                               int idx, bool mask);
+int vhost_net_notify_migration_done(VHostNetState *net, char* mac_addr);
 VHostNetState *get_vhost_net(NetClientState *nc);
 
 int vhost_set_vring_enable(NetClientState * nc, int enable);
diff --git a/net/vhost-user.c b/net/vhost-user.c
index 87917a5..cfe11b8 100644
--- a/net/vhost-user.c
+++ b/net/vhost-user.c
@@ -106,9 +106,29 @@ err:
 static ssize_t vhost_user_receive(NetClientState *nc, const uint8_t *buf,
                                   size_t size)
 {
-    /* Discard the request that is received and managed by backend
-     * by an other way.
+    /* In case of RARP (message size is 60) notify backup to send a fake RARP.
+       This fake RARP will be sent by backend only for guest
+       without GUEST_ANNOUNCE capability.
      */
+    if (size == 60) {
+        VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
+        int r;
+        static int display_rarp_failure = 1;
+        char mac_addr[6];
+
+        /* extract guest mac address from the RARP message */
+        memcpy(mac_addr, &buf[6], 6);
+
+        r = vhost_net_notify_migration_done(s->vhost_net, mac_addr);
+
+        if ((r != 0) && (display_rarp_failure)) {
+            fprintf(stderr,
+                    "Vhost user backend fails to broadcast fake RARP\n");
+            fflush(stderr);
+            display_rarp_failure = 0;
+        }
+    }
+
     return size;
 }
 
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 16/21] vhost-user-test: move wait_for_fds() out
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (14 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 17/21] vhost-user-test: remove useless static check marcandre.lureau
                   ` (6 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

This function is a precondition for most vhost-user tests.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index 77b7b68..88714ff 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -205,15 +205,11 @@ static GThread *_thread_new(const gchar *name, GThreadFunc func, gpointer data)
     return thread;
 }
 
-static void read_guest_mem(void)
+static void wait_for_fds(void)
 {
-    uint32_t *guest_mem;
     gint64 end_time;
-    int i, j;
-    size_t size;
 
     g_mutex_lock(data_mutex);
-
     end_time = _get_time() + 5 * G_TIME_SPAN_SECOND;
     while (!fds_num) {
         if (!_cond_wait_until(data_cond, data_mutex, end_time)) {
@@ -227,6 +223,17 @@ static void read_guest_mem(void)
     g_assert_cmpint(fds_num, >, 0);
     g_assert_cmpint(fds_num, ==, memory.nregions);
 
+    g_mutex_unlock(data_mutex);
+}
+
+static void read_guest_mem(void)
+{
+    uint32_t *guest_mem;
+    int i, j;
+    size_t size;
+
+    wait_for_fds();
+
     /* iterate all regions */
     for (i = 0; i < fds_num; i++) {
 
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 17/21] vhost-user-test: remove useless static check
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (15 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 16/21] vhost-user-test: move wait_for_fds() out marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 18/21] vhost-user-test: wrap server in TestServer struct marcandre.lureau
                   ` (5 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index 88714ff..6662ca9 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -262,7 +262,6 @@ static void read_guest_mem(void)
         munmap(guest_mem, memory.regions[i].memory_size);
     }
 
-    g_assert_cmpint(1, ==, 1);
     g_mutex_unlock(data_mutex);
 }
 
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 18/21] vhost-user-test: wrap server in TestServer struct
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (16 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 17/21] vhost-user-test: remove useless static check marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 19/21] vhost-user-test: learn to tweak various qemu arguments marcandre.lureau
                   ` (4 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

In the coming patches, a test will use several servers
simultaneously. Wrap the server in a struct, out of the global scope.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 146 ++++++++++++++++++++++++++++++------------------
 1 file changed, 92 insertions(+), 54 deletions(-)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index 6662ca9..8daef01 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -115,10 +115,18 @@ static VhostUserMsg m __attribute__ ((unused));
 #define VHOST_USER_VERSION    (0x1)
 /*****************************************************************************/
 
-int fds_num = 0, fds[VHOST_MEMORY_MAX_NREGIONS];
-static VhostUserMemory memory;
-static GMutex *data_mutex;
-static GCond *data_cond;
+typedef struct TestServer {
+    gchar *socket_path;
+    gchar *chr_name;
+    CharDriverState *chr;
+    int fds_num;
+    int fds[VHOST_MEMORY_MAX_NREGIONS];
+    VhostUserMemory memory;
+    GMutex *data_mutex;
+    GCond *data_cond;
+} TestServer;
+
+static const char *hugefs;
 
 static gint64 _get_time(void)
 {
@@ -205,64 +213,63 @@ static GThread *_thread_new(const gchar *name, GThreadFunc func, gpointer data)
     return thread;
 }
 
-static void wait_for_fds(void)
+static void wait_for_fds(TestServer *s)
 {
     gint64 end_time;
 
-    g_mutex_lock(data_mutex);
+    g_mutex_lock(s->data_mutex);
     end_time = _get_time() + 5 * G_TIME_SPAN_SECOND;
-    while (!fds_num) {
-        if (!_cond_wait_until(data_cond, data_mutex, end_time)) {
+    while (!s->fds_num) {
+        if (!_cond_wait_until(s->data_cond, s->data_mutex, end_time)) {
             /* timeout has passed */
-            g_assert(fds_num);
+            g_assert(s->fds_num);
             break;
         }
     }
 
     /* check for sanity */
-    g_assert_cmpint(fds_num, >, 0);
-    g_assert_cmpint(fds_num, ==, memory.nregions);
+    g_assert_cmpint(s->fds_num, >, 0);
+    g_assert_cmpint(s->fds_num, ==, s->memory.nregions);
 
-    g_mutex_unlock(data_mutex);
+    g_mutex_unlock(s->data_mutex);
 }
 
-static void read_guest_mem(void)
+static void read_guest_mem(TestServer *s)
 {
     uint32_t *guest_mem;
     int i, j;
     size_t size;
 
-    wait_for_fds();
+    wait_for_fds(s);
 
     /* iterate all regions */
-    for (i = 0; i < fds_num; i++) {
+    for (i = 0; i < s->fds_num; i++) {
 
         /* We'll check only the region statring at 0x0*/
-        if (memory.regions[i].guest_phys_addr != 0x0) {
+        if (s->memory.regions[i].guest_phys_addr != 0x0) {
             continue;
         }
 
-        g_assert_cmpint(memory.regions[i].memory_size, >, 1024);
+        g_assert_cmpint(s->memory.regions[i].memory_size, >, 1024);
 
-        size =  memory.regions[i].memory_size + memory.regions[i].mmap_offset;
+        size = s->memory.regions[i].memory_size +
+            s->memory.regions[i].mmap_offset;
 
         guest_mem = mmap(0, size, PROT_READ | PROT_WRITE,
-                         MAP_SHARED, fds[i], 0);
+                         MAP_SHARED, s->fds[i], 0);
 
         g_assert(guest_mem != MAP_FAILED);
-        guest_mem += (memory.regions[i].mmap_offset / sizeof(*guest_mem));
+        guest_mem += (s->memory.regions[i].mmap_offset / sizeof(*guest_mem));
 
         for (j = 0; j < 256; j++) {
-            uint32_t a = readl(memory.regions[i].guest_phys_addr + j*4);
+            uint32_t a = readl(s->memory.regions[i].guest_phys_addr + j*4);
             uint32_t b = guest_mem[j];
 
             g_assert_cmpint(a, ==, b);
         }
 
-        munmap(guest_mem, memory.regions[i].memory_size);
+        munmap(guest_mem, s->memory.regions[i].memory_size);
     }
-
-    g_mutex_unlock(data_mutex);
 }
 
 static void *thread_function(void *data)
@@ -280,7 +287,8 @@ static int chr_can_read(void *opaque)
 
 static void chr_read(void *opaque, const uint8_t *buf, int size)
 {
-    CharDriverState *chr = opaque;
+    TestServer *s = opaque;
+    CharDriverState *chr = s->chr;
     VhostUserMsg msg;
     uint8_t *p = (uint8_t *) &msg;
     int fd;
@@ -290,12 +298,12 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
         return;
     }
 
-    g_mutex_lock(data_mutex);
+    g_mutex_lock(s->data_mutex);
     memcpy(p, buf, VHOST_USER_HDR_SIZE);
 
     if (msg.size) {
         p += VHOST_USER_HDR_SIZE;
-        qemu_chr_fe_read_all(chr, p, msg.size);
+        g_assert_cmpint(qemu_chr_fe_read_all(chr, p, msg.size), ==, msg.size);
     }
 
     switch (msg.request) {
@@ -334,11 +342,11 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
 
     case VHOST_USER_SET_MEM_TABLE:
         /* received the mem table */
-        memcpy(&memory, &msg.memory, sizeof(msg.memory));
-        fds_num = qemu_chr_fe_get_msgfds(chr, fds, sizeof(fds) / sizeof(int));
+        memcpy(&s->memory, &msg.memory, sizeof(msg.memory));
+        s->fds_num = qemu_chr_fe_get_msgfds(chr, s->fds, G_N_ELEMENTS(s->fds));
 
         /* signal the test that it can continue */
-        g_cond_signal(data_cond);
+        g_cond_signal(s->data_cond);
         break;
 
     case VHOST_USER_SET_VRING_KICK:
@@ -355,7 +363,53 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
     default:
         break;
     }
-    g_mutex_unlock(data_mutex);
+
+    g_mutex_unlock(s->data_mutex);
+}
+
+static TestServer *test_server_new(const gchar *name)
+{
+    TestServer *server = g_new0(TestServer, 1);
+    gchar *chr_path;
+
+    server->socket_path = g_strdup_printf("/tmp/vhost-%s-%d.sock",
+                                          name, getpid());
+
+    chr_path = g_strdup_printf("unix:%s,server,nowait", server->socket_path);
+    server->chr_name = g_strdup_printf("chr-%s", name);
+    server->chr = qemu_chr_new(server->chr_name, chr_path, NULL);
+    g_free(chr_path);
+
+    qemu_chr_add_handlers(server->chr, chr_can_read, chr_read, NULL, server);
+
+    /* run the main loop thread so the chardev may operate */
+    server->data_mutex = _mutex_new();
+    server->data_cond = _cond_new();
+
+    return server;
+}
+
+#define GET_QEMU_CMD(s)                                         \
+    g_strdup_printf(QEMU_CMD, (hugefs), (s)->socket_path)
+
+
+static void test_server_free(TestServer *server)
+{
+    int i;
+
+    qemu_chr_delete(server->chr);
+
+    for (i = 0; i < server->fds_num; i++) {
+        close(server->fds[i]);
+    }
+
+    unlink(server->socket_path);
+    g_free(server->socket_path);
+
+    _cond_free(server->data_cond);
+    _mutex_free(server->data_mutex);
+
+    g_free(server);
 }
 
 static const char *init_hugepagefs(void)
@@ -394,41 +448,28 @@ static const char *init_hugepagefs(void)
 int main(int argc, char **argv)
 {
     QTestState *s = NULL;
-    CharDriverState *chr = NULL;
-    const char *hugefs = 0;
-    char *socket_path = 0;
-    char *qemu_cmd = 0;
-    char *chr_path = 0;
+    TestServer *server = NULL;
+    char *qemu_cmd;
     int ret;
 
     g_test_init(&argc, &argv, NULL);
 
     module_call_init(MODULE_INIT_QOM);
+    qemu_add_opts(&qemu_chardev_opts);
 
     hugefs = init_hugepagefs();
     if (!hugefs) {
         return 0;
     }
 
-    socket_path = g_strdup_printf("/tmp/vhost-%d.sock", getpid());
-
-    /* create char dev and add read handlers */
-    qemu_add_opts(&qemu_chardev_opts);
-    chr_path = g_strdup_printf("unix:%s,server,nowait", socket_path);
-    chr = qemu_chr_new("chr0", chr_path, NULL);
-    g_free(chr_path);
-    qemu_chr_add_handlers(chr, chr_can_read, chr_read, NULL, chr);
-
-    /* run the main loop thread so the chardev may operate */
-    data_mutex = _mutex_new();
-    data_cond = _cond_new();
     _thread_new(NULL, thread_function, NULL);
 
-    qemu_cmd = g_strdup_printf(QEMU_CMD, hugefs, socket_path);
+    server = test_server_new("test");
+    qemu_cmd = GET_QEMU_CMD(server);
     s = qtest_start(qemu_cmd);
     g_free(qemu_cmd);
 
-    qtest_add_func("/vhost-user/read-guest-mem", read_guest_mem);
+    qtest_add_data_func("/vhost-user/read-guest-mem", server, read_guest_mem);
 
     ret = g_test_run();
 
@@ -437,10 +478,7 @@ int main(int argc, char **argv)
     }
 
     /* cleanup */
-    unlink(socket_path);
-    g_free(socket_path);
-    _cond_free(data_cond);
-    _mutex_free(data_mutex);
+    test_server_free(server);
 
     return ret;
 }
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 19/21] vhost-user-test: learn to tweak various qemu arguments
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (17 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 18/21] vhost-user-test: wrap server in TestServer struct marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 20/21] vhost-user-test: add live-migration test marcandre.lureau
                   ` (3 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Add a new macro to make the qemu command line with other
values of memory size, and specific chardev id.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index 8daef01..b2ac454 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -37,10 +37,10 @@
 #endif
 
 #define QEMU_CMD_ACCEL  " -machine accel=tcg"
-#define QEMU_CMD_MEM    " -m 512 -object memory-backend-file,id=mem,size=512M,"\
+#define QEMU_CMD_MEM    " -m %d -object memory-backend-file,id=mem,size=%dM,"\
                         "mem-path=%s,share=on -numa node,memdev=mem"
-#define QEMU_CMD_CHR    " -chardev socket,id=chr0,path=%s"
-#define QEMU_CMD_NETDEV " -netdev vhost-user,id=net0,chardev=chr0,vhostforce"
+#define QEMU_CMD_CHR    " -chardev socket,id=%s,path=%s"
+#define QEMU_CMD_NETDEV " -netdev vhost-user,id=net0,chardev=%s,vhostforce"
 #define QEMU_CMD_NET    " -device virtio-net-pci,netdev=net0 "
 #define QEMU_CMD_ROM    " -option-rom ../pc-bios/pxe-virtio.rom"
 
@@ -389,9 +389,13 @@ static TestServer *test_server_new(const gchar *name)
     return server;
 }
 
-#define GET_QEMU_CMD(s)                                         \
-    g_strdup_printf(QEMU_CMD, (hugefs), (s)->socket_path)
+#define GET_QEMU_CMD(s)                                 \
+    g_strdup_printf(QEMU_CMD, 512, 512, (hugefs),       \
+        (s)->chr_name, (s)->socket_path, (s)->chr_name)
 
+#define GET_QEMU_CMDE(s, mem, extra, ...)                               \
+    g_strdup_printf(QEMU_CMD extra, (mem), (mem), (hugefs),             \
+        (s)->chr_name, (s)->socket_path, (s)->chr_name, ##__VA_ARGS__)
 
 static void test_server_free(TestServer *server)
 {
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 20/21] vhost-user-test: add live-migration test
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (18 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 19/21] vhost-user-test: learn to tweak various qemu arguments marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 21/21] vhost-user-test: check ownership during migration marcandre.lureau
                   ` (2 subsequent siblings)
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

This test checks that the log fd is given to the migration source, and
mark dirty pages during migration.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 171 +++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 168 insertions(+), 3 deletions(-)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index b2ac454..7b5f130 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -13,6 +13,7 @@
 
 #include "libqtest.h"
 #include "qemu/option.h"
+#include "qemu/range.h"
 #include "sysemu/char.h"
 #include "sysemu/sysemu.h"
 
@@ -54,6 +55,9 @@
 #define VHOST_MEMORY_MAX_NREGIONS    8
 
 #define VHOST_USER_F_PROTOCOL_FEATURES 30
+#define VHOST_USER_PROTOCOL_F_LOG_SHMFD 1
+
+#define VHOST_LOG_PAGE 0x1000
 
 typedef enum VhostUserRequest {
     VHOST_USER_NONE = 0,
@@ -124,6 +128,7 @@ typedef struct TestServer {
     VhostUserMemory memory;
     GMutex *data_mutex;
     GCond *data_cond;
+    int log_fd;
 } TestServer;
 
 static const char *hugefs;
@@ -311,7 +316,8 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
         /* send back features to qemu */
         msg.flags |= VHOST_USER_REPLY_MASK;
         msg.size = sizeof(m.u64);
-        msg.u64 = 0x1ULL << VHOST_USER_F_PROTOCOL_FEATURES;
+        msg.u64 = 0x1ULL << VHOST_F_LOG_ALL |
+            0x1ULL << VHOST_USER_F_PROTOCOL_FEATURES;
         p = (uint8_t *) &msg;
         qemu_chr_fe_write_all(chr, p, VHOST_USER_HDR_SIZE + msg.size);
         break;
@@ -325,12 +331,11 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
         /* send back features to qemu */
         msg.flags |= VHOST_USER_REPLY_MASK;
         msg.size = sizeof(m.u64);
-        msg.u64 = 0;
+        msg.u64 = 1 << VHOST_USER_PROTOCOL_F_LOG_SHMFD;
         p = (uint8_t *) &msg;
         qemu_chr_fe_write_all(chr, p, VHOST_USER_HDR_SIZE + msg.size);
         break;
 
-
     case VHOST_USER_GET_VRING_BASE:
         /* send back vring base to qemu */
         msg.flags |= VHOST_USER_REPLY_MASK;
@@ -360,6 +365,21 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
          */
         qemu_set_nonblock(fd);
         break;
+
+    case VHOST_USER_SET_LOG_BASE:
+        if (s->log_fd != -1) {
+            close(s->log_fd);
+            s->log_fd = -1;
+        }
+        qemu_chr_fe_get_msgfds(chr, &s->log_fd, 1);
+        msg.flags |= VHOST_USER_REPLY_MASK;
+        msg.size = 0;
+        p = (uint8_t *) &msg;
+        qemu_chr_fe_write_all(chr, p, VHOST_USER_HDR_SIZE);
+
+        g_cond_signal(s->data_cond);
+        break;
+
     default:
         break;
     }
@@ -386,6 +406,8 @@ static TestServer *test_server_new(const gchar *name)
     server->data_mutex = _mutex_new();
     server->data_cond = _cond_new();
 
+    server->log_fd = -1;
+
     return server;
 }
 
@@ -407,15 +429,157 @@ static void test_server_free(TestServer *server)
         close(server->fds[i]);
     }
 
+    if (server->log_fd != -1) {
+        close(server->log_fd);
+    }
+
     unlink(server->socket_path);
     g_free(server->socket_path);
 
     _cond_free(server->data_cond);
     _mutex_free(server->data_mutex);
 
+    g_free(server->chr_name);
     g_free(server);
 }
 
+static void wait_for_log_fd(TestServer *s)
+{
+    gint64 end_time;
+
+    g_mutex_lock(s->data_mutex);
+    end_time = _get_time() + 5 * G_TIME_SPAN_SECOND;
+    while (s->log_fd == -1) {
+        if (!_cond_wait_until(s->data_cond, s->data_mutex, end_time)) {
+            /* timeout has passed */
+            g_assert(s->log_fd != -1);
+            break;
+        }
+    }
+
+    g_mutex_unlock(s->data_mutex);
+}
+
+static void write_guest_mem(TestServer *s, uint32 seed)
+{
+    uint32_t *guest_mem;
+    int i, j;
+    size_t size;
+
+    wait_for_fds(s);
+
+    /* iterate all regions */
+    for (i = 0; i < s->fds_num; i++) {
+
+        /* We'll write only the region statring at 0x0 */
+        if (s->memory.regions[i].guest_phys_addr != 0x0) {
+            continue;
+        }
+
+        g_assert_cmpint(s->memory.regions[i].memory_size, >, 1024);
+
+        size = s->memory.regions[i].memory_size +
+            s->memory.regions[i].mmap_offset;
+
+        guest_mem = mmap(0, size, PROT_READ | PROT_WRITE,
+                         MAP_SHARED, s->fds[i], 0);
+
+        g_assert(guest_mem != MAP_FAILED);
+        guest_mem += (s->memory.regions[i].mmap_offset / sizeof(*guest_mem));
+
+        for (j = 0; j < 256; j++) {
+            guest_mem[j] = seed + j;
+        }
+
+        munmap(guest_mem, s->memory.regions[i].memory_size);
+        break;
+    }
+}
+
+static guint64 get_log_size(TestServer *s)
+{
+    guint64 log_size = 0;
+    int i;
+
+    for (i = 0; i < s->memory.nregions; ++i) {
+        VhostUserMemoryRegion *reg = &s->memory.regions[i];
+        guint64 last = range_get_last(reg->guest_phys_addr,
+                                       reg->memory_size);
+        log_size = MAX(log_size, last / (8 * VHOST_LOG_PAGE) + 1);
+    }
+
+    return log_size;
+}
+
+static void test_migrate(void)
+{
+    TestServer *s = test_server_new("src");
+    TestServer *dest = test_server_new("dest");
+    const char *uri = "tcp:127.0.0.1:1234";
+    QTestState *global = global_qtest, *from, *to;
+    gchar *cmd;
+    QDict *rsp;
+    guint8 *log;
+    guint64 size;
+
+    cmd = GET_QEMU_CMDE(s, 2, "");
+    from = qtest_start(cmd);
+    g_free(cmd);
+
+    wait_for_fds(s);
+    size = get_log_size(s);
+    g_assert_cmpint(size, ==, (2 * 1024 * 1024) / (VHOST_LOG_PAGE * 8));
+
+    cmd = GET_QEMU_CMDE(dest, 2, " -incoming %s", uri);
+    to = qtest_init(cmd);
+    g_free(cmd);
+
+    /* slow down migration to have time to fiddle with log */
+    /* TODO: qtest could learn to break on some places */
+    rsp = qmp("{ 'execute': 'migrate_set_speed',"
+              "'arguments': { 'value': 10 } }");
+    g_assert(qdict_haskey(rsp, "return"));
+    QDECREF(rsp);
+
+    cmd = g_strdup_printf("{ 'execute': 'migrate',"
+                          "'arguments': { 'uri': '%s' } }",
+                          uri);
+    rsp = qmp(cmd);
+    g_free(cmd);
+    g_assert(qdict_haskey(rsp, "return"));
+    QDECREF(rsp);
+
+    wait_for_log_fd(s);
+
+    log = mmap(0, size, PROT_READ | PROT_WRITE, MAP_SHARED, s->log_fd, 0);
+    g_assert(log != MAP_FAILED);
+
+    /* modify first page */
+    write_guest_mem(s, 0x42);
+    log[0] = 1;
+    munmap(log, size);
+
+    /* speed things up */
+    rsp = qmp("{ 'execute': 'migrate_set_speed',"
+              "'arguments': { 'value': 0 } }");
+    g_assert(qdict_haskey(rsp, "return"));
+    QDECREF(rsp);
+
+    qmp_eventwait("STOP");
+
+    global_qtest = to;
+    qmp_eventwait("RESUME");
+
+    read_guest_mem(dest);
+
+    qtest_quit(to);
+    test_server_free(dest);
+    qtest_quit(from);
+    test_server_free(s);
+
+    global_qtest = global;
+}
+
 static const char *init_hugepagefs(void)
 {
     const char *path;
@@ -474,6 +638,7 @@ int main(int argc, char **argv)
     g_free(qemu_cmd);
 
     qtest_add_data_func("/vhost-user/read-guest-mem", server, read_guest_mem);
+    qtest_add_func("/vhost-user/migrate", test_migrate);
 
     ret = g_test_run();
 
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* [Qemu-devel] [PATCH v5 21/21] vhost-user-test: check ownership during migration
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (19 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 20/21] vhost-user-test: add live-migration test marcandre.lureau
@ 2015-09-24 16:22 ` marcandre.lureau
  2015-09-27 13:19 ` [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support Michael S. Tsirkin
  2015-09-29 15:07 ` Michael S. Tsirkin
  22 siblings, 0 replies; 35+ messages in thread
From: marcandre.lureau @ 2015-09-24 16:22 UTC (permalink / raw)
  To: qemu-devel
  Cc: haifeng.lin, mst, thibaut.collet, jasowang, pbonzini,
	Marc-André Lureau

From: Marc-André Lureau <marcandre.lureau@redhat.com>

Check that backend source and destination do not have simultaneous
ownership during migration.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/vhost-user-test.c | 38 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
index 7b5f130..99a5cfe 100644
--- a/tests/vhost-user-test.c
+++ b/tests/vhost-user-test.c
@@ -380,6 +380,10 @@ static void chr_read(void *opaque, const uint8_t *buf, int size)
         g_cond_signal(s->data_cond);
         break;
 
+    case VHOST_USER_RESET_OWNER:
+        s->fds_num = 0;
+        break;
+
     default:
         break;
     }
@@ -511,12 +515,37 @@ static guint64 get_log_size(TestServer *s)
     return log_size;
 }
 
+typedef struct TestMigrateSource {
+    GSource source;
+    TestServer *src;
+    TestServer *dest;
+} TestMigrateSource;
+
+static gboolean
+test_migrate_source_check(GSource *source)
+{
+    TestMigrateSource *t = (TestMigrateSource *)source;
+    gboolean overlap = t->src->fds_num > 0 && t->dest->fds_num > 0;
+
+    g_assert(!overlap);
+
+    return FALSE;
+}
+
+GSourceFuncs test_migrate_source_funcs = {
+    NULL,
+    test_migrate_source_check,
+    NULL,
+    NULL
+};
+
 static void test_migrate(void)
 {
     TestServer *s = test_server_new("src");
     TestServer *dest = test_server_new("dest");
     const char *uri = "tcp:127.0.0.1:1234";
     QTestState *global = global_qtest, *from, *to;
+    GSource *source;
     gchar *cmd;
     QDict *rsp;
     guint8 *log;
@@ -534,6 +563,12 @@ static void test_migrate(void)
     to = qtest_init(cmd);
     g_free(cmd);
 
+    source = g_source_new(&test_migrate_source_funcs,
+                          sizeof(TestMigrateSource));
+    ((TestMigrateSource *)source)->src = s;
+    ((TestMigrateSource *)source)->dest = dest;
+    g_source_attach(source, NULL);
+
     /* slow down migration to have time to fiddle with log */
     /* TODO: qtest could learn to break on some places */
     rsp = qmp("{ 'execute': 'migrate_set_speed',"
@@ -572,6 +607,9 @@ static void test_migrate(void)
 
     read_guest_mem(dest);
 
+    g_source_destroy(source);
+    g_source_unref(source);
+
     qtest_quit(to);
     test_server_free(dest);
     qtest_quit(from);
-- 
2.4.3

^ permalink raw reply related	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log marcandre.lureau
@ 2015-09-24 16:38   ` Eric Blake
  2015-09-24 16:41     ` Marc-André Lureau
  0 siblings, 1 reply; 35+ messages in thread
From: Eric Blake @ 2015-09-24 16:38 UTC (permalink / raw)
  To: marcandre.lureau, qemu-devel
  Cc: thibaut.collet, jasowang, pbonzini, haifeng.lin, mst

[-- Attachment #1: Type: text/plain, Size: 1121 bytes --]

On 09/24/2015 10:22 AM, marcandre.lureau@redhat.com wrote:
> From: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> ---
>  docs/specs/vhost-user.txt | 48 +++++++++++++++++++++++++++++++++++++++++++++--
>  1 file changed, 46 insertions(+), 2 deletions(-)

> @@ -140,8 +142,7 @@ Multiple queue support
>  
>  Multiple queue is treated as a protocol extension, hence the slave has to
>  implement protocol features first. The multiple queues feature is supported
> -only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set:
> -#define VHOST_USER_PROTOCOL_F_MQ    0
> +only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set.

Should this hunk be in an earlier patch,

> +
> +Protocol features
> +-----------------
> +
> +#define VHOST_USER_PROTOCOL_F_MQ             0
> +#define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1

along with this section (of course, with just bit 0 in that patch)?

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log
  2015-09-24 16:38   ` Eric Blake
@ 2015-09-24 16:41     ` Marc-André Lureau
  2015-09-24 16:47       ` Eric Blake
  0 siblings, 1 reply; 35+ messages in thread
From: Marc-André Lureau @ 2015-09-24 16:41 UTC (permalink / raw)
  To: Eric Blake
  Cc: haifeng lin, mst, thibaut collet, jasowang, qemu-devel, pbonzini,
	marcandre lureau


Hi

----- Original Message -----
> On 09/24/2015 10:22 AM, marcandre.lureau@redhat.com wrote:
> > From: Marc-André Lureau <marcandre.lureau@redhat.com>
> > 
> > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> > ---
> >  docs/specs/vhost-user.txt | 48
> >  +++++++++++++++++++++++++++++++++++++++++++++--
> >  1 file changed, 46 insertions(+), 2 deletions(-)
> 
> > @@ -140,8 +142,7 @@ Multiple queue support
> >  
> >  Multiple queue is treated as a protocol extension, hence the slave has to
> >  implement protocol features first. The multiple queues feature is
> >  supported
> > -only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set:
> > -#define VHOST_USER_PROTOCOL_F_MQ    0
> > +only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set.
> 
> Should this hunk be in an earlier patch,

The series is on top of last Michael PULL request. I just moved the define
in a common section, since now they are 2 and later 3. Does that make sense?

> 
> > +
> > +Protocol features
> > +-----------------
> > +
> > +#define VHOST_USER_PROTOCOL_F_MQ             0
> > +#define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
> 
> along with this section (of course, with just bit 0 in that patch)?
> 
> --
> Eric Blake   eblake redhat com    +1-919-301-3266
> Libvirt virtualization library http://libvirt.org
> 
> 

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log
  2015-09-24 16:41     ` Marc-André Lureau
@ 2015-09-24 16:47       ` Eric Blake
  0 siblings, 0 replies; 35+ messages in thread
From: Eric Blake @ 2015-09-24 16:47 UTC (permalink / raw)
  To: Marc-André Lureau
  Cc: haifeng lin, mst, thibaut collet, jasowang, qemu-devel, pbonzini,
	marcandre lureau

[-- Attachment #1: Type: text/plain, Size: 710 bytes --]

On 09/24/2015 10:41 AM, Marc-André Lureau wrote:

>>> -only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set:
>>> -#define VHOST_USER_PROTOCOL_F_MQ    0
>>> +only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set.
>>
>> Should this hunk be in an earlier patch,
> 
> The series is on top of last Michael PULL request. I just moved the define
> in a common section, since now they are 2 and later 3. Does that make sense?

Sure. My bad for not checking whether it was this series that introduced
bit 0 vs. it being pre-existing from an earlier series.


-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest marcandre.lureau
@ 2015-09-24 17:24   ` Thibaut Collet
  2015-09-24 21:53     ` Marc-André Lureau
  0 siblings, 1 reply; 35+ messages in thread
From: Thibaut Collet @ 2015-09-24 17:24 UTC (permalink / raw)
  To: marcandre.lureau
  Cc: Paolo Bonzini, Jason Wang, Michael S. Tsirkin, qemu-devel,
	Linhaifeng

On Thu, Sep 24, 2015 at 6:22 PM,  <marcandre.lureau@redhat.com> wrote:
> From: Thibaut Collet <thibaut.collet@6wind.com>
>
> A new vhost user message is added to allow QEMU to ask to vhost user backend to
> broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
> capability.
>
> This new message is sent only if the backend supports the new
> VHOST_USER_PROTOCOL_F_RARP protocol feature.
> The payload of this new message is the MAC address of the guest (not known by
> the backend). The MAC address is copied in the first 6 bytes of a u64 to avoid
> to create a new payload message type.
>
> This new message has no equivalent ioctl so a new callback is added in the
> userOps structure to send the request.
>
> Upon reception of this new message the vhost user backend must generate and
> broadcast a fake RARP request to notify the migration is terminated.
>
> Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
> [Rebased and fixed checkpatch errors - Marc-André]
> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> ---
>  docs/specs/vhost-user.txt         | 15 +++++++++++++++
>  hw/net/vhost_net.c                | 17 +++++++++++++++++
>  hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
>  include/hw/virtio/vhost-backend.h |  3 +++
>  include/net/vhost_net.h           |  1 +
>  net/vhost-user.c                  | 24 ++++++++++++++++++++++--
>  6 files changed, 88 insertions(+), 2 deletions(-)
>
> diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
> index e0292a0..e0d71e2 100644
> --- a/docs/specs/vhost-user.txt
> +++ b/docs/specs/vhost-user.txt
> @@ -194,6 +194,7 @@ Protocol features
>
>  #define VHOST_USER_PROTOCOL_F_MQ             0
>  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
> +#define VHOST_USER_PROTOCOL_F_RARP           2
>
>  Message types
>  -------------
> @@ -381,3 +382,17 @@ Message types
>        Master payload: vring state description
>
>        Signal slave to enable or disable corresponding vring.
> +
> + * VHOST_USER_SEND_RARP
> +
> +      Id: 19
> +      Equivalent ioctl: N/A
> +      Master payload: u64
> +
> +      Ask vhost user backend to broadcast a fake RARP to notify the migration
> +      is terminated for guest that does not support GUEST_ANNOUNCE.
> +      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present in
> +      VHOST_USER_GET_FEATURES and protocol feature bit VHOST_USER_PROTOCOL_F_RARP
> +      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
> +      The first 6 bytes of the payload contain the mac address of the guest to
> +      allow the vhost user backend to construct and broadcast the fake RARP.
> diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
> index 840f443..da66b64 100644
> --- a/hw/net/vhost_net.c
> +++ b/hw/net/vhost_net.c
> @@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
>      g_free(net);
>  }
>
> +int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
> +{
> +    const VhostOps *vhost_ops = net->dev.vhost_ops;
> +    int r = -1;
> +
> +    if (vhost_ops->vhost_migration_done) {
> +        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
> +    }
> +
> +    return r;
> +}
> +
>  bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
>  {
>      return vhost_virtqueue_pending(&net->dev, idx);
> @@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net, VirtIODevice *dev,
>  {
>  }
>
> +int vhost_net_notify_migration_done(struct vhost_net *net)
> +{
> +    return -1;
> +}
> +
>  VHostNetState *get_vhost_net(NetClientState *nc)
>  {
>      return 0;
> diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> index 455caba..b7f3699 100644
> --- a/hw/virtio/vhost-user.c
> +++ b/hw/virtio/vhost-user.c
> @@ -10,6 +10,7 @@
>
>  #include "hw/virtio/vhost.h"
>  #include "hw/virtio/vhost-backend.h"
> +#include "hw/virtio/virtio-net.h"
>  #include "sysemu/char.h"
>  #include "sysemu/kvm.h"
>  #include "qemu/error-report.h"
> @@ -30,6 +31,7 @@
>  #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
>  #define VHOST_USER_PROTOCOL_F_MQ         0
>  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
> +#define VHOST_USER_PROTOCOL_F_RARP       2

The VHOST_USER_PROTOCOL_FEATURE_MASK  must be changed and set to 0x7ULL

>
>  typedef enum VhostUserRequest {
>      VHOST_USER_NONE = 0,
> @@ -51,6 +53,7 @@ typedef enum VhostUserRequest {
>      VHOST_USER_SET_PROTOCOL_FEATURES = 16,
>      VHOST_USER_GET_QUEUE_NUM = 17,
>      VHOST_USER_SET_VRING_ENABLE = 18,
> +    VHOST_USER_SEND_RARP = 19,
>      VHOST_USER_MAX
>  } VhostUserRequest;
>
> @@ -561,6 +564,32 @@ static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
>                                VHOST_USER_PROTOCOL_F_LOG_SHMFD);
>  }
>
> +static int vhost_user_migration_done(struct vhost_dev *dev, char* mac_addr)
> +{
> +    VhostUserMsg msg = { 0 };
> +    int err;
> +
> +    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
> +
> +    /* If guest supports GUEST_ANNOUNCE do nothing */
> +    if (virtio_has_feature(dev->acked_features, VIRTIO_NET_F_GUEST_ANNOUNCE)) {
> +        return 0;
> +    }
> +
> +    /* if backend supports VHOST_USER_PROTOCOL_F_RARP ask it to send the RARP */
> +    if (virtio_has_feature(dev->protocol_features,
> +                           VHOST_USER_PROTOCOL_F_RARP)) {
> +        msg.request = VHOST_USER_SEND_RARP;
> +        msg.flags = VHOST_USER_VERSION;
> +        memcpy((char *)&msg.u64, mac_addr, 6);
> +        msg.size = sizeof(m.u64);
> +
> +        err = vhost_user_write(dev, &msg, NULL, 0);
> +        return err;
> +    }
> +    return -1;
> +}
> +
>  const VhostOps user_ops = {
>          .backend_type = VHOST_BACKEND_TYPE_USER,
>          .vhost_backend_init = vhost_user_init,
> @@ -582,4 +611,5 @@ const VhostOps user_ops = {
>          .vhost_get_vq_index = vhost_user_get_vq_index,
>          .vhost_set_vring_enable = vhost_user_set_vring_enable,
>          .vhost_requires_shm_log = vhost_user_requires_shm_log,
> +        .vhost_migration_done = vhost_user_migration_done,
>  };
> diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
> index e07118c..5918c01 100644
> --- a/include/hw/virtio/vhost-backend.h
> +++ b/include/hw/virtio/vhost-backend.h
> @@ -67,6 +67,8 @@ typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
>  typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
>                                           int enable);
>  typedef bool (*vhost_requires_shm_log_op)(struct vhost_dev *dev);
> +typedef int (*vhost_migration_done_op)(struct vhost_dev *dev,
> +                                       char *mac_addr);
>
>  typedef struct VhostOps {
>      VhostBackendType backend_type;
> @@ -94,6 +96,7 @@ typedef struct VhostOps {
>      vhost_get_vq_index_op vhost_get_vq_index;
>      vhost_set_vring_enable_op vhost_set_vring_enable;
>      vhost_requires_shm_log_op vhost_requires_shm_log;
> +    vhost_migration_done_op vhost_migration_done;
>  } VhostOps;
>
>  extern const VhostOps user_ops;
> diff --git a/include/net/vhost_net.h b/include/net/vhost_net.h
> index 0188c4d..3389b41 100644
> --- a/include/net/vhost_net.h
> +++ b/include/net/vhost_net.h
> @@ -27,6 +27,7 @@ void vhost_net_ack_features(VHostNetState *net, uint64_t features);
>  bool vhost_net_virtqueue_pending(VHostNetState *net, int n);
>  void vhost_net_virtqueue_mask(VHostNetState *net, VirtIODevice *dev,
>                                int idx, bool mask);
> +int vhost_net_notify_migration_done(VHostNetState *net, char* mac_addr);
>  VHostNetState *get_vhost_net(NetClientState *nc);
>
>  int vhost_set_vring_enable(NetClientState * nc, int enable);
> diff --git a/net/vhost-user.c b/net/vhost-user.c
> index 87917a5..cfe11b8 100644
> --- a/net/vhost-user.c
> +++ b/net/vhost-user.c
> @@ -106,9 +106,29 @@ err:
>  static ssize_t vhost_user_receive(NetClientState *nc, const uint8_t *buf,
>                                    size_t size)
>  {
> -    /* Discard the request that is received and managed by backend
> -     * by an other way.
> +    /* In case of RARP (message size is 60) notify backup to send a fake RARP.
> +       This fake RARP will be sent by backend only for guest
> +       without GUEST_ANNOUNCE capability.
>       */
> +    if (size == 60) {
> +        VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
> +        int r;
> +        static int display_rarp_failure = 1;
> +        char mac_addr[6];
> +
> +        /* extract guest mac address from the RARP message */
> +        memcpy(mac_addr, &buf[6], 6);
> +
> +        r = vhost_net_notify_migration_done(s->vhost_net, mac_addr);
> +
> +        if ((r != 0) && (display_rarp_failure)) {
> +            fprintf(stderr,
> +                    "Vhost user backend fails to broadcast fake RARP\n");
> +            fflush(stderr);
> +            display_rarp_failure = 0;
> +        }
> +    }
> +
>      return size;
>  }
>
> --
> 2.4.3
>

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-24 17:24   ` Thibaut Collet
@ 2015-09-24 21:53     ` Marc-André Lureau
  2015-09-27 13:12       ` Michael S. Tsirkin
  0 siblings, 1 reply; 35+ messages in thread
From: Marc-André Lureau @ 2015-09-24 21:53 UTC (permalink / raw)
  To: Thibaut Collet
  Cc: Linhaifeng, Michael S. Tsirkin, Jason Wang, qemu-devel,
	Paolo Bonzini, marcandre lureau

Hi

----- Original Message -----
> On Thu, Sep 24, 2015 at 6:22 PM,  <marcandre.lureau@redhat.com> wrote:
> > From: Thibaut Collet <thibaut.collet@6wind.com>
> >
> > A new vhost user message is added to allow QEMU to ask to vhost user
> > backend to
> > broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
> > capability.
> >
> > This new message is sent only if the backend supports the new
> > VHOST_USER_PROTOCOL_F_RARP protocol feature.
> > The payload of this new message is the MAC address of the guest (not known
> > by
> > the backend). The MAC address is copied in the first 6 bytes of a u64 to
> > avoid
> > to create a new payload message type.
> >
> > This new message has no equivalent ioctl so a new callback is added in the
> > userOps structure to send the request.
> >
> > Upon reception of this new message the vhost user backend must generate and
> > broadcast a fake RARP request to notify the migration is terminated.
> >
> > Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
> > [Rebased and fixed checkpatch errors - Marc-André]
> > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> > ---
> >  docs/specs/vhost-user.txt         | 15 +++++++++++++++
> >  hw/net/vhost_net.c                | 17 +++++++++++++++++
> >  hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
> >  include/hw/virtio/vhost-backend.h |  3 +++
> >  include/net/vhost_net.h           |  1 +
> >  net/vhost-user.c                  | 24 ++++++++++++++++++++++--
> >  6 files changed, 88 insertions(+), 2 deletions(-)
> >
> > diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
> > index e0292a0..e0d71e2 100644
> > --- a/docs/specs/vhost-user.txt
> > +++ b/docs/specs/vhost-user.txt
> > @@ -194,6 +194,7 @@ Protocol features
> >
> >  #define VHOST_USER_PROTOCOL_F_MQ             0
> >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
> > +#define VHOST_USER_PROTOCOL_F_RARP           2
> >
> >  Message types
> >  -------------
> > @@ -381,3 +382,17 @@ Message types
> >        Master payload: vring state description
> >
> >        Signal slave to enable or disable corresponding vring.
> > +
> > + * VHOST_USER_SEND_RARP
> > +
> > +      Id: 19
> > +      Equivalent ioctl: N/A
> > +      Master payload: u64
> > +
> > +      Ask vhost user backend to broadcast a fake RARP to notify the
> > migration
> > +      is terminated for guest that does not support GUEST_ANNOUNCE.
> > +      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present
> > in
> > +      VHOST_USER_GET_FEATURES and protocol feature bit
> > VHOST_USER_PROTOCOL_F_RARP
> > +      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
> > +      The first 6 bytes of the payload contain the mac address of the
> > guest to
> > +      allow the vhost user backend to construct and broadcast the fake
> > RARP.
> > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
> > index 840f443..da66b64 100644
> > --- a/hw/net/vhost_net.c
> > +++ b/hw/net/vhost_net.c
> > @@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
> >      g_free(net);
> >  }
> >
> > +int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
> > +{
> > +    const VhostOps *vhost_ops = net->dev.vhost_ops;
> > +    int r = -1;
> > +
> > +    if (vhost_ops->vhost_migration_done) {
> > +        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
> > +    }
> > +
> > +    return r;
> > +}
> > +
> >  bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
> >  {
> >      return vhost_virtqueue_pending(&net->dev, idx);
> > @@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net,
> > VirtIODevice *dev,
> >  {
> >  }
> >
> > +int vhost_net_notify_migration_done(struct vhost_net *net)
> > +{
> > +    return -1;
> > +}
> > +
> >  VHostNetState *get_vhost_net(NetClientState *nc)
> >  {
> >      return 0;
> > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> > index 455caba..b7f3699 100644
> > --- a/hw/virtio/vhost-user.c
> > +++ b/hw/virtio/vhost-user.c
> > @@ -10,6 +10,7 @@
> >
> >  #include "hw/virtio/vhost.h"
> >  #include "hw/virtio/vhost-backend.h"
> > +#include "hw/virtio/virtio-net.h"
> >  #include "sysemu/char.h"
> >  #include "sysemu/kvm.h"
> >  #include "qemu/error-report.h"
> > @@ -30,6 +31,7 @@
> >  #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
> >  #define VHOST_USER_PROTOCOL_F_MQ         0
> >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
> > +#define VHOST_USER_PROTOCOL_F_RARP       2
> 
> The VHOST_USER_PROTOCOL_FEATURE_MASK  must be changed and set to 0x7ULL
> 

Good catch (too many rebases, having a test would help to prevent this kind of mistake)

thanks

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-24 21:53     ` Marc-André Lureau
@ 2015-09-27 13:12       ` Michael S. Tsirkin
  2015-09-27 15:13         ` Marc-André Lureau
  0 siblings, 1 reply; 35+ messages in thread
From: Michael S. Tsirkin @ 2015-09-27 13:12 UTC (permalink / raw)
  To: Marc-André Lureau
  Cc: Linhaifeng, Thibaut Collet, Jason Wang, qemu-devel,
	marcandre lureau, Paolo Bonzini

On Thu, Sep 24, 2015 at 05:53:05PM -0400, Marc-André Lureau wrote:
> Hi
> 
> ----- Original Message -----
> > On Thu, Sep 24, 2015 at 6:22 PM,  <marcandre.lureau@redhat.com> wrote:
> > > From: Thibaut Collet <thibaut.collet@6wind.com>
> > >
> > > A new vhost user message is added to allow QEMU to ask to vhost user
> > > backend to
> > > broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
> > > capability.
> > >
> > > This new message is sent only if the backend supports the new
> > > VHOST_USER_PROTOCOL_F_RARP protocol feature.
> > > The payload of this new message is the MAC address of the guest (not known
> > > by
> > > the backend). The MAC address is copied in the first 6 bytes of a u64 to
> > > avoid
> > > to create a new payload message type.
> > >
> > > This new message has no equivalent ioctl so a new callback is added in the
> > > userOps structure to send the request.
> > >
> > > Upon reception of this new message the vhost user backend must generate and
> > > broadcast a fake RARP request to notify the migration is terminated.
> > >
> > > Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
> > > [Rebased and fixed checkpatch errors - Marc-André]
> > > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> > > ---
> > >  docs/specs/vhost-user.txt         | 15 +++++++++++++++
> > >  hw/net/vhost_net.c                | 17 +++++++++++++++++
> > >  hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
> > >  include/hw/virtio/vhost-backend.h |  3 +++
> > >  include/net/vhost_net.h           |  1 +
> > >  net/vhost-user.c                  | 24 ++++++++++++++++++++++--
> > >  6 files changed, 88 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
> > > index e0292a0..e0d71e2 100644
> > > --- a/docs/specs/vhost-user.txt
> > > +++ b/docs/specs/vhost-user.txt
> > > @@ -194,6 +194,7 @@ Protocol features
> > >
> > >  #define VHOST_USER_PROTOCOL_F_MQ             0
> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
> > > +#define VHOST_USER_PROTOCOL_F_RARP           2
> > >
> > >  Message types
> > >  -------------
> > > @@ -381,3 +382,17 @@ Message types
> > >        Master payload: vring state description
> > >
> > >        Signal slave to enable or disable corresponding vring.
> > > +
> > > + * VHOST_USER_SEND_RARP
> > > +
> > > +      Id: 19
> > > +      Equivalent ioctl: N/A
> > > +      Master payload: u64
> > > +
> > > +      Ask vhost user backend to broadcast a fake RARP to notify the
> > > migration
> > > +      is terminated for guest that does not support GUEST_ANNOUNCE.
> > > +      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present
> > > in
> > > +      VHOST_USER_GET_FEATURES and protocol feature bit
> > > VHOST_USER_PROTOCOL_F_RARP
> > > +      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
> > > +      The first 6 bytes of the payload contain the mac address of the
> > > guest to
> > > +      allow the vhost user backend to construct and broadcast the fake
> > > RARP.
> > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
> > > index 840f443..da66b64 100644
> > > --- a/hw/net/vhost_net.c
> > > +++ b/hw/net/vhost_net.c
> > > @@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
> > >      g_free(net);
> > >  }
> > >
> > > +int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
> > > +{
> > > +    const VhostOps *vhost_ops = net->dev.vhost_ops;
> > > +    int r = -1;
> > > +
> > > +    if (vhost_ops->vhost_migration_done) {
> > > +        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
> > > +    }
> > > +
> > > +    return r;
> > > +}
> > > +
> > >  bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
> > >  {
> > >      return vhost_virtqueue_pending(&net->dev, idx);
> > > @@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net,
> > > VirtIODevice *dev,
> > >  {
> > >  }
> > >
> > > +int vhost_net_notify_migration_done(struct vhost_net *net)
> > > +{
> > > +    return -1;
> > > +}
> > > +
> > >  VHostNetState *get_vhost_net(NetClientState *nc)
> > >  {
> > >      return 0;
> > > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> > > index 455caba..b7f3699 100644
> > > --- a/hw/virtio/vhost-user.c
> > > +++ b/hw/virtio/vhost-user.c
> > > @@ -10,6 +10,7 @@
> > >
> > >  #include "hw/virtio/vhost.h"
> > >  #include "hw/virtio/vhost-backend.h"
> > > +#include "hw/virtio/virtio-net.h"
> > >  #include "sysemu/char.h"
> > >  #include "sysemu/kvm.h"
> > >  #include "qemu/error-report.h"
> > > @@ -30,6 +31,7 @@
> > >  #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
> > >  #define VHOST_USER_PROTOCOL_F_MQ         0
> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
> > > +#define VHOST_USER_PROTOCOL_F_RARP       2
> > 
> > The VHOST_USER_PROTOCOL_FEATURE_MASK  must be changed and set to 0x7ULL
> > 
Better, change VHOST_USER_PROTOCOL_FEATURE_MASK to be
calculated based on other macros.

> 
> Good catch (too many rebases, having a test would help to prevent this kind of mistake)
> 
> thanks

So there will be v6 with a fix?

-- 
MST

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (20 preceding siblings ...)
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 21/21] vhost-user-test: check ownership during migration marcandre.lureau
@ 2015-09-27 13:19 ` Michael S. Tsirkin
  2015-09-29 15:07 ` Michael S. Tsirkin
  22 siblings, 0 replies; 35+ messages in thread
From: Michael S. Tsirkin @ 2015-09-27 13:19 UTC (permalink / raw)
  To: marcandre.lureau
  Cc: haifeng.lin, thibaut.collet, jasowang, qemu-devel, pbonzini

On Thu, Sep 24, 2015 at 06:22:00PM +0200, marcandre.lureau@redhat.com wrote:
> From: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
> Hi,
> 
> The following series implement shareable log for vhost-user to support
> memory tracking during live migration. On qemu-side, the solution is
> fairly straightfoward since vhost already supports the dirty log, only
> vhost-user couldn't access the log memory until then.
> 
> The series includes "protocol feature negotiation" series proposed
> earlier by Michael S. Tsirkinm and "vhost user: Add live migration"
> series from Thibaut Collet .

I applied patches 1-14 in my tree for now.

This way you don't need to repost them.

If you do change some of these you can post a fixup patch
(with subject fixup! <original subject>)
or tell me to drop some of the old patches.

> v4->v5:
> - rebase on top of last Michael S. Tsirkin PULL request
> - block live migration if !PROTOCOL_F_LOG_SHMFD
> - wait for a reply after SET_LOG_BASE
> - split vhost_set_log_base from the rest of vhost_call refactoring
> - use a seperate global vhost_log_shm
> 
> v3->v4:
> - add the proto negotiation & the migration series
> - replace the varargs vhost_call() approach for callbacks
> - only share-allocate when the backend needs it
> 
> v2->v3:
> - changed some patch summary
> - added migration tests
> - added a patch to replace error message with a trace
> 
> The development branch I used is:
> https://github.com/elmarco/qemu branch "vhost-user"
> 
> Marc-André Lureau (18):
>   configure: probe for memfd
>   util: add linux-only memfd fallback
>   util: add memfd helpers
>   vhost: alloc shareable log
>   vhost: document log resizing
>   vhost: add vhost_set_log_base op
>   vhost-user: send log shm fd along with log_base
>   vhost-user: add a migration blocker
>   vhost: use a function for each call
>   vhost: only use shared log if in use by backend
>   vhost-user: document migration log
>   net: add trace_vhost_user_event
>   vhost-user-test: move wait_for_fds() out
>   vhost-user-test: remove useless static check
>   vhost-user-test: wrap server in TestServer struct
>   vhost-user-test: learn to tweak various qemu arguments
>   vhost-user-test: add live-migration test
>   vhost-user-test: check ownership during migration
> 
> Michael S. Tsirkin (1):
>   vhost-user: unit test for new messages
> 
> Thibaut Collet (2):
>   vhost user: add support of live migration
>   vhost user: add rarp sending after live migration for legacy guest
> 
>  configure                         |  19 ++
>  docs/specs/vhost-user.txt         |  63 ++++-
>  hw/net/vhost_net.c                |  35 ++-
>  hw/scsi/vhost-scsi.c              |   7 +-
>  hw/virtio/vhost-backend.c         | 121 +++++++-
>  hw/virtio/vhost-user.c            | 571 ++++++++++++++++++++++++--------------
>  hw/virtio/vhost.c                 | 101 ++++---
>  include/hw/virtio/vhost-backend.h |  77 ++++-
>  include/hw/virtio/vhost.h         |  15 +-
>  include/net/vhost_net.h           |   1 +
>  include/qemu/memfd.h              |  24 ++
>  net/vhost-user.c                  |  34 ++-
>  tests/vhost-user-test.c           | 390 ++++++++++++++++++++++----
>  trace-events                      |   3 +
>  util/Makefile.objs                |   2 +-
>  util/memfd.c                      | 126 +++++++++
>  16 files changed, 1260 insertions(+), 329 deletions(-)
>  create mode 100644 include/qemu/memfd.h
>  create mode 100644 util/memfd.c
> 
> -- 
> 2.4.3

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-27 13:12       ` Michael S. Tsirkin
@ 2015-09-27 15:13         ` Marc-André Lureau
  2015-09-27 15:49           ` Thibaut Collet
  0 siblings, 1 reply; 35+ messages in thread
From: Marc-André Lureau @ 2015-09-27 15:13 UTC (permalink / raw)
  To: Michael S. Tsirkin
  Cc: Thibaut Collet, Jason Wang, Paolo Bonzini, Linhaifeng, qemu-devel

Hi

On Sun, Sep 27, 2015 at 3:12 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Thu, Sep 24, 2015 at 05:53:05PM -0400, Marc-André Lureau wrote:
>> Hi
>>
>> ----- Original Message -----
>> > On Thu, Sep 24, 2015 at 6:22 PM,  <marcandre.lureau@redhat.com> wrote:
>> > > From: Thibaut Collet <thibaut.collet@6wind.com>
>> > >
>> > > A new vhost user message is added to allow QEMU to ask to vhost user
>> > > backend to
>> > > broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
>> > > capability.
>> > >
>> > > This new message is sent only if the backend supports the new
>> > > VHOST_USER_PROTOCOL_F_RARP protocol feature.
>> > > The payload of this new message is the MAC address of the guest (not known
>> > > by
>> > > the backend). The MAC address is copied in the first 6 bytes of a u64 to
>> > > avoid
>> > > to create a new payload message type.
>> > >
>> > > This new message has no equivalent ioctl so a new callback is added in the
>> > > userOps structure to send the request.
>> > >
>> > > Upon reception of this new message the vhost user backend must generate and
>> > > broadcast a fake RARP request to notify the migration is terminated.
>> > >
>> > > Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
>> > > [Rebased and fixed checkpatch errors - Marc-André]
>> > > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>> > > ---
>> > >  docs/specs/vhost-user.txt         | 15 +++++++++++++++
>> > >  hw/net/vhost_net.c                | 17 +++++++++++++++++
>> > >  hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
>> > >  include/hw/virtio/vhost-backend.h |  3 +++
>> > >  include/net/vhost_net.h           |  1 +
>> > >  net/vhost-user.c                  | 24 ++++++++++++++++++++++--
>> > >  6 files changed, 88 insertions(+), 2 deletions(-)
>> > >
>> > > diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
>> > > index e0292a0..e0d71e2 100644
>> > > --- a/docs/specs/vhost-user.txt
>> > > +++ b/docs/specs/vhost-user.txt
>> > > @@ -194,6 +194,7 @@ Protocol features
>> > >
>> > >  #define VHOST_USER_PROTOCOL_F_MQ             0
>> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
>> > > +#define VHOST_USER_PROTOCOL_F_RARP           2
>> > >
>> > >  Message types
>> > >  -------------
>> > > @@ -381,3 +382,17 @@ Message types
>> > >        Master payload: vring state description
>> > >
>> > >        Signal slave to enable or disable corresponding vring.
>> > > +
>> > > + * VHOST_USER_SEND_RARP
>> > > +
>> > > +      Id: 19
>> > > +      Equivalent ioctl: N/A
>> > > +      Master payload: u64
>> > > +
>> > > +      Ask vhost user backend to broadcast a fake RARP to notify the
>> > > migration
>> > > +      is terminated for guest that does not support GUEST_ANNOUNCE.
>> > > +      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present
>> > > in
>> > > +      VHOST_USER_GET_FEATURES and protocol feature bit
>> > > VHOST_USER_PROTOCOL_F_RARP
>> > > +      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
>> > > +      The first 6 bytes of the payload contain the mac address of the
>> > > guest to
>> > > +      allow the vhost user backend to construct and broadcast the fake
>> > > RARP.
>> > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
>> > > index 840f443..da66b64 100644
>> > > --- a/hw/net/vhost_net.c
>> > > +++ b/hw/net/vhost_net.c
>> > > @@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
>> > >      g_free(net);
>> > >  }
>> > >
>> > > +int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
>> > > +{
>> > > +    const VhostOps *vhost_ops = net->dev.vhost_ops;
>> > > +    int r = -1;
>> > > +
>> > > +    if (vhost_ops->vhost_migration_done) {
>> > > +        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
>> > > +    }
>> > > +
>> > > +    return r;
>> > > +}
>> > > +
>> > >  bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
>> > >  {
>> > >      return vhost_virtqueue_pending(&net->dev, idx);
>> > > @@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net,
>> > > VirtIODevice *dev,
>> > >  {
>> > >  }
>> > >
>> > > +int vhost_net_notify_migration_done(struct vhost_net *net)
>> > > +{
>> > > +    return -1;
>> > > +}
>> > > +
>> > >  VHostNetState *get_vhost_net(NetClientState *nc)
>> > >  {
>> > >      return 0;
>> > > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
>> > > index 455caba..b7f3699 100644
>> > > --- a/hw/virtio/vhost-user.c
>> > > +++ b/hw/virtio/vhost-user.c
>> > > @@ -10,6 +10,7 @@
>> > >
>> > >  #include "hw/virtio/vhost.h"
>> > >  #include "hw/virtio/vhost-backend.h"
>> > > +#include "hw/virtio/virtio-net.h"
>> > >  #include "sysemu/char.h"
>> > >  #include "sysemu/kvm.h"
>> > >  #include "qemu/error-report.h"
>> > > @@ -30,6 +31,7 @@
>> > >  #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
>> > >  #define VHOST_USER_PROTOCOL_F_MQ         0
>> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
>> > > +#define VHOST_USER_PROTOCOL_F_RARP       2
>> >
>> > The VHOST_USER_PROTOCOL_FEATURE_MASK  must be changed and set to 0x7ULL
>> >
> Better, change VHOST_USER_PROTOCOL_FEATURE_MASK to be
> calculated based on other macros.

yeah, I actually thought about something like this a few days ago:
https://github.com/elmarco/qemu/commit/b4e4ec2e4c78f6b12e34822d8a3f3a101d065d80

Is that what you have in mind or rather make a bit mask of all the
features? Maybe the bitmask of all the features is simpler.

>
>>
>> Good catch (too many rebases, having a test would help to prevent this kind of mistake)
>>
>> thanks
>
> So there will be v6 with a fix?

If there is no further changes in the series, I guess you could fix it
when cherry-picking. But if I have to send a v6, I'll include the fix.

thanks



-- 
Marc-André Lureau

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-27 15:13         ` Marc-André Lureau
@ 2015-09-27 15:49           ` Thibaut Collet
  2015-09-27 18:04             ` Marc-André Lureau
  0 siblings, 1 reply; 35+ messages in thread
From: Thibaut Collet @ 2015-09-27 15:49 UTC (permalink / raw)
  To: Marc-André Lureau
  Cc: qemu-devel, Paolo Bonzini, Jason Wang, Linhaifeng,
	Michael S. Tsirkin

On Sun, Sep 27, 2015 at 5:13 PM, Marc-André Lureau
<marcandre.lureau@gmail.com> wrote:
> Hi
>
> On Sun, Sep 27, 2015 at 3:12 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>> On Thu, Sep 24, 2015 at 05:53:05PM -0400, Marc-André Lureau wrote:
>>> Hi
>>>
>>> ----- Original Message -----
>>> > On Thu, Sep 24, 2015 at 6:22 PM,  <marcandre.lureau@redhat.com> wrote:
>>> > > From: Thibaut Collet <thibaut.collet@6wind.com>
>>> > >
>>> > > A new vhost user message is added to allow QEMU to ask to vhost user
>>> > > backend to
>>> > > broadcast a fake RARP after live migration for guest without GUEST_ANNOUNCE
>>> > > capability.
>>> > >
>>> > > This new message is sent only if the backend supports the new
>>> > > VHOST_USER_PROTOCOL_F_RARP protocol feature.
>>> > > The payload of this new message is the MAC address of the guest (not known
>>> > > by
>>> > > the backend). The MAC address is copied in the first 6 bytes of a u64 to
>>> > > avoid
>>> > > to create a new payload message type.
>>> > >
>>> > > This new message has no equivalent ioctl so a new callback is added in the
>>> > > userOps structure to send the request.
>>> > >
>>> > > Upon reception of this new message the vhost user backend must generate and
>>> > > broadcast a fake RARP request to notify the migration is terminated.
>>> > >
>>> > > Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
>>> > > [Rebased and fixed checkpatch errors - Marc-André]
>>> > > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>>> > > ---
>>> > >  docs/specs/vhost-user.txt         | 15 +++++++++++++++
>>> > >  hw/net/vhost_net.c                | 17 +++++++++++++++++
>>> > >  hw/virtio/vhost-user.c            | 30 ++++++++++++++++++++++++++++++
>>> > >  include/hw/virtio/vhost-backend.h |  3 +++
>>> > >  include/net/vhost_net.h           |  1 +
>>> > >  net/vhost-user.c                  | 24 ++++++++++++++++++++++--
>>> > >  6 files changed, 88 insertions(+), 2 deletions(-)
>>> > >
>>> > > diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
>>> > > index e0292a0..e0d71e2 100644
>>> > > --- a/docs/specs/vhost-user.txt
>>> > > +++ b/docs/specs/vhost-user.txt
>>> > > @@ -194,6 +194,7 @@ Protocol features
>>> > >
>>> > >  #define VHOST_USER_PROTOCOL_F_MQ             0
>>> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD      1
>>> > > +#define VHOST_USER_PROTOCOL_F_RARP           2
>>> > >
>>> > >  Message types
>>> > >  -------------
>>> > > @@ -381,3 +382,17 @@ Message types
>>> > >        Master payload: vring state description
>>> > >
>>> > >        Signal slave to enable or disable corresponding vring.
>>> > > +
>>> > > + * VHOST_USER_SEND_RARP
>>> > > +
>>> > > +      Id: 19
>>> > > +      Equivalent ioctl: N/A
>>> > > +      Master payload: u64
>>> > > +
>>> > > +      Ask vhost user backend to broadcast a fake RARP to notify the
>>> > > migration
>>> > > +      is terminated for guest that does not support GUEST_ANNOUNCE.
>>> > > +      Only legal if feature bit VHOST_USER_F_PROTOCOL_FEATURES is present
>>> > > in
>>> > > +      VHOST_USER_GET_FEATURES and protocol feature bit
>>> > > VHOST_USER_PROTOCOL_F_RARP
>>> > > +      is present in VHOST_USER_GET_PROTOCOL_FEATURES.
>>> > > +      The first 6 bytes of the payload contain the mac address of the
>>> > > guest to
>>> > > +      allow the vhost user backend to construct and broadcast the fake
>>> > > RARP.
>>> > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
>>> > > index 840f443..da66b64 100644
>>> > > --- a/hw/net/vhost_net.c
>>> > > +++ b/hw/net/vhost_net.c
>>> > > @@ -388,6 +388,18 @@ void vhost_net_cleanup(struct vhost_net *net)
>>> > >      g_free(net);
>>> > >  }
>>> > >
>>> > > +int vhost_net_notify_migration_done(struct vhost_net *net, char* mac_addr)
>>> > > +{
>>> > > +    const VhostOps *vhost_ops = net->dev.vhost_ops;
>>> > > +    int r = -1;
>>> > > +
>>> > > +    if (vhost_ops->vhost_migration_done) {
>>> > > +        r = vhost_ops->vhost_migration_done(&net->dev, mac_addr);
>>> > > +    }
>>> > > +
>>> > > +    return r;
>>> > > +}
>>> > > +
>>> > >  bool vhost_net_virtqueue_pending(VHostNetState *net, int idx)
>>> > >  {
>>> > >      return vhost_virtqueue_pending(&net->dev, idx);
>>> > > @@ -479,6 +491,11 @@ void vhost_net_virtqueue_mask(VHostNetState *net,
>>> > > VirtIODevice *dev,
>>> > >  {
>>> > >  }
>>> > >
>>> > > +int vhost_net_notify_migration_done(struct vhost_net *net)
>>> > > +{
>>> > > +    return -1;
>>> > > +}
>>> > > +
>>> > >  VHostNetState *get_vhost_net(NetClientState *nc)
>>> > >  {
>>> > >      return 0;
>>> > > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
>>> > > index 455caba..b7f3699 100644
>>> > > --- a/hw/virtio/vhost-user.c
>>> > > +++ b/hw/virtio/vhost-user.c
>>> > > @@ -10,6 +10,7 @@
>>> > >
>>> > >  #include "hw/virtio/vhost.h"
>>> > >  #include "hw/virtio/vhost-backend.h"
>>> > > +#include "hw/virtio/virtio-net.h"
>>> > >  #include "sysemu/char.h"
>>> > >  #include "sysemu/kvm.h"
>>> > >  #include "qemu/error-report.h"
>>> > > @@ -30,6 +31,7 @@
>>> > >  #define VHOST_USER_PROTOCOL_FEATURE_MASK 0x3ULL
>>> > >  #define VHOST_USER_PROTOCOL_F_MQ         0
>>> > >  #define VHOST_USER_PROTOCOL_F_LOG_SHMFD  1
>>> > > +#define VHOST_USER_PROTOCOL_F_RARP       2
>>> >
>>> > The VHOST_USER_PROTOCOL_FEATURE_MASK  must be changed and set to 0x7ULL
>>> >
>> Better, change VHOST_USER_PROTOCOL_FEATURE_MASK to be
>> calculated based on other macros.
>
> yeah, I actually thought about something like this a few days ago:
> https://github.com/elmarco/qemu/commit/b4e4ec2e4c78f6b12e34822d8a3f3a101d065d80
>

I agree an automatic computation of the FEATURE_MASK from the
different protocol feature will avoid issues wiht rebase or merge
operation.

Maybe a solution like that is clearer ?

-#define VHOST_USER_PROTOCOL_FEATURE_MASK 0x7ULL
-#define VHOST_USER_PROTOCOL_F_MQ 0
-#define VHOST_USER_PROTOCOL_F_LOG_SHMFD 1
-#define VHOST_USER_PROTOCOL_F_RARP 2
+typedef enum VhostUserProtocolFeature {
+VHOST_USER_PROTOCOL_F_MQ = 0,
+VHOST_USER_PROTOCOL_F_LOG_SHMFD = 1,
+VHOST_USER_PROTOCOL_F_RARP = 2,
+VHOST_USER_PROTOCOL_F_MAX
+} VhostUserProtocolFeature ;

+#define VHOST_USER_PROTOCOL_FEATURE_MASK ((1 << VHOST_USER_PROTOCOL_F_MAX) - 1)


> Is that what you have in mind or rather make a bit mask of all the
> features? Maybe the bitmask of all the features is simpler.
>
>>
>>>
>>> Good catch (too many rebases, having a test would help to prevent this kind of mistake)
>>>
>>> thanks
>>
>> So there will be v6 with a fix?
>
> If there is no further changes in the series, I guess you could fix it
> when cherry-picking. But if I have to send a v6, I'll include the fix.
>
> thanks
>
>
>
> --
> Marc-André Lureau

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest
  2015-09-27 15:49           ` Thibaut Collet
@ 2015-09-27 18:04             ` Marc-André Lureau
  0 siblings, 0 replies; 35+ messages in thread
From: Marc-André Lureau @ 2015-09-27 18:04 UTC (permalink / raw)
  To: Thibaut Collet
  Cc: qemu-devel, Paolo Bonzini, Jason Wang, Linhaifeng,
	Michael S. Tsirkin

hi

On Sun, Sep 27, 2015 at 5:49 PM, Thibaut Collet
<thibaut.collet@6wind.com> wrote:
> Maybe a solution like that is clearer ?
>
> -#define VHOST_USER_PROTOCOL_FEATURE_MASK 0x7ULL
> -#define VHOST_USER_PROTOCOL_F_MQ 0
> -#define VHOST_USER_PROTOCOL_F_LOG_SHMFD 1
> -#define VHOST_USER_PROTOCOL_F_RARP 2
> +typedef enum VhostUserProtocolFeature {
> +VHOST_USER_PROTOCOL_F_MQ = 0,
> +VHOST_USER_PROTOCOL_F_LOG_SHMFD = 1,
> +VHOST_USER_PROTOCOL_F_RARP = 2,
> +VHOST_USER_PROTOCOL_F_MAX
> +} VhostUserProtocolFeature ;
>
> +#define VHOST_USER_PROTOCOL_FEATURE_MASK ((1 << VHOST_USER_PROTOCOL_F_MAX) - 1)


I agree with that, I don't see a good reason to keep the defines.

-- 
Marc-André Lureau

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend
  2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend marcandre.lureau
@ 2015-09-29 14:59   ` Michael S. Tsirkin
  2015-09-29 15:26     ` Marc-André Lureau
  0 siblings, 1 reply; 35+ messages in thread
From: Michael S. Tsirkin @ 2015-09-29 14:59 UTC (permalink / raw)
  To: marcandre.lureau
  Cc: haifeng.lin, thibaut.collet, jasowang, qemu-devel, pbonzini

On Thu, Sep 24, 2015 at 06:22:11PM +0200, marcandre.lureau@redhat.com wrote:
> From: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
> Do not allocate a shared log if the backend doesn't support it.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

So squash this in the previous patch introducing the shared log?
This split makes review harder, not easier.

> ---
>  hw/virtio/vhost-user.c            |  9 +++++++++
>  hw/virtio/vhost.c                 | 15 ++++++++++-----
>  include/hw/virtio/vhost-backend.h |  4 ++++
>  3 files changed, 23 insertions(+), 5 deletions(-)
> 
> diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> index 41e0364..455caba 100644
> --- a/hw/virtio/vhost-user.c
> +++ b/hw/virtio/vhost-user.c
> @@ -553,6 +553,14 @@ static int vhost_user_get_vq_index(struct vhost_dev *dev, int idx)
>      return idx;
>  }
>  
> +static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
> +{
> +    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
> +
> +    return virtio_has_feature(dev->protocol_features,
> +                              VHOST_USER_PROTOCOL_F_LOG_SHMFD);
> +}
> +
>  const VhostOps user_ops = {
>          .backend_type = VHOST_BACKEND_TYPE_USER,
>          .vhost_backend_init = vhost_user_init,
> @@ -573,4 +581,5 @@ const VhostOps user_ops = {
>          .vhost_reset_device = vhost_user_reset_device,
>          .vhost_get_vq_index = vhost_user_get_vq_index,
>          .vhost_set_vring_enable = vhost_user_set_vring_enable,
> +        .vhost_requires_shm_log = vhost_user_requires_shm_log,
>  };
> diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
> index d9f2131..ec27c7b 100644
> --- a/hw/virtio/vhost.c
> +++ b/hw/virtio/vhost.c
> @@ -357,10 +357,15 @@ static void vhost_log_put(struct vhost_dev *dev, bool sync)
>      }
>  }
>  
> -static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t size)
> +static bool vhost_dev_log_is_shared(struct vhost_dev *dev)
>  {
> -    bool share = dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
> -    struct vhost_log *log = vhost_log_get(size, share);
> +    return dev->vhost_ops->vhost_requires_shm_log &&
> +           dev->vhost_ops->vhost_requires_shm_log(dev);
> +}
> +
> +static inline void vhost_dev_log_resize(struct vhost_dev *dev, uint64_t size)
> +{
> +    struct vhost_log *log = vhost_log_get(size, vhost_dev_log_is_shared(dev));
>      uint64_t log_base = (uintptr_t)log->log;
>      int r;
>  
> @@ -1166,10 +1171,10 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev)
>  
>      if (hdev->log_enabled) {
>          uint64_t log_base;
> -        bool share = hdev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
>  
>          hdev->log_size = vhost_get_log_size(hdev);
> -        hdev->log = vhost_log_get(hdev->log_size, share);
> +        hdev->log = vhost_log_get(hdev->log_size,
> +                                  vhost_dev_log_is_shared(hdev));
>          log_base = (uintptr_t)hdev->log->log;
>          r = hdev->vhost_ops->vhost_set_log_base(hdev,
>                                                  hdev->log_size ? log_base : 0,
> diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
> index 868a78a..e07118c 100644
> --- a/include/hw/virtio/vhost-backend.h
> +++ b/include/hw/virtio/vhost-backend.h
> @@ -11,6 +11,8 @@
>  #ifndef VHOST_BACKEND_H_
>  #define VHOST_BACKEND_H_
>  
> +#include <stdbool.h>
> +
>  typedef enum VhostBackendType {
>      VHOST_BACKEND_TYPE_NONE = 0,
>      VHOST_BACKEND_TYPE_KERNEL = 1,
> @@ -64,6 +66,7 @@ typedef int (*vhost_reset_device_op)(struct vhost_dev *dev);
>  typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
>  typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
>                                           int enable);
> +typedef bool (*vhost_requires_shm_log_op)(struct vhost_dev *dev);
>  
>  typedef struct VhostOps {
>      VhostBackendType backend_type;
> @@ -90,6 +93,7 @@ typedef struct VhostOps {
>      vhost_reset_device_op vhost_reset_device;
>      vhost_get_vq_index_op vhost_get_vq_index;
>      vhost_set_vring_enable_op vhost_set_vring_enable;
> +    vhost_requires_shm_log_op vhost_requires_shm_log;
>  } VhostOps;
>  
>  extern const VhostOps user_ops;
> -- 
> 2.4.3

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support
  2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
                   ` (21 preceding siblings ...)
  2015-09-27 13:19 ` [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support Michael S. Tsirkin
@ 2015-09-29 15:07 ` Michael S. Tsirkin
  22 siblings, 0 replies; 35+ messages in thread
From: Michael S. Tsirkin @ 2015-09-29 15:07 UTC (permalink / raw)
  To: marcandre.lureau
  Cc: haifeng.lin, thibaut.collet, jasowang, qemu-devel, pbonzini

On Thu, Sep 24, 2015 at 06:22:00PM +0200, marcandre.lureau@redhat.com wrote:
> From: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
> Hi,
> 
> The following series implement shareable log for vhost-user to support
> memory tracking during live migration. On qemu-side, the solution is
> fairly straightfoward since vhost already supports the dirty log, only
> vhost-user couldn't access the log memory until then.
> 
> The series includes "protocol feature negotiation" series proposed
> earlier by Michael S. Tsirkinm and "vhost user: Add live migration"
> series from Thibaut Collet .

So I dropped everything except the 1st patch from my tree
since there were too many failures.

> v4->v5:
> - rebase on top of last Michael S. Tsirkin PULL request
> - block live migration if !PROTOCOL_F_LOG_SHMFD
> - wait for a reply after SET_LOG_BASE
> - split vhost_set_log_base from the rest of vhost_call refactoring
> - use a seperate global vhost_log_shm
> 
> v3->v4:
> - add the proto negotiation & the migration series
> - replace the varargs vhost_call() approach for callbacks
> - only share-allocate when the backend needs it
> 
> v2->v3:
> - changed some patch summary
> - added migration tests
> - added a patch to replace error message with a trace
> 
> The development branch I used is:
> https://github.com/elmarco/qemu branch "vhost-user"
> 
> Marc-André Lureau (18):
>   configure: probe for memfd
>   util: add linux-only memfd fallback
>   util: add memfd helpers
>   vhost: alloc shareable log
>   vhost: document log resizing
>   vhost: add vhost_set_log_base op
>   vhost-user: send log shm fd along with log_base
>   vhost-user: add a migration blocker
>   vhost: use a function for each call
>   vhost: only use shared log if in use by backend
>   vhost-user: document migration log
>   net: add trace_vhost_user_event
>   vhost-user-test: move wait_for_fds() out
>   vhost-user-test: remove useless static check
>   vhost-user-test: wrap server in TestServer struct
>   vhost-user-test: learn to tweak various qemu arguments
>   vhost-user-test: add live-migration test
>   vhost-user-test: check ownership during migration
> 
> Michael S. Tsirkin (1):
>   vhost-user: unit test for new messages
> 
> Thibaut Collet (2):
>   vhost user: add support of live migration
>   vhost user: add rarp sending after live migration for legacy guest
> 
>  configure                         |  19 ++
>  docs/specs/vhost-user.txt         |  63 ++++-
>  hw/net/vhost_net.c                |  35 ++-
>  hw/scsi/vhost-scsi.c              |   7 +-
>  hw/virtio/vhost-backend.c         | 121 +++++++-
>  hw/virtio/vhost-user.c            | 571 ++++++++++++++++++++++++--------------
>  hw/virtio/vhost.c                 | 101 ++++---
>  include/hw/virtio/vhost-backend.h |  77 ++++-
>  include/hw/virtio/vhost.h         |  15 +-
>  include/net/vhost_net.h           |   1 +
>  include/qemu/memfd.h              |  24 ++
>  net/vhost-user.c                  |  34 ++-
>  tests/vhost-user-test.c           | 390 ++++++++++++++++++++++----
>  trace-events                      |   3 +
>  util/Makefile.objs                |   2 +-
>  util/memfd.c                      | 126 +++++++++
>  16 files changed, 1260 insertions(+), 329 deletions(-)
>  create mode 100644 include/qemu/memfd.h
>  create mode 100644 util/memfd.c
> 
> -- 
> 2.4.3

^ permalink raw reply	[flat|nested] 35+ messages in thread

* Re: [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend
  2015-09-29 14:59   ` Michael S. Tsirkin
@ 2015-09-29 15:26     ` Marc-André Lureau
  0 siblings, 0 replies; 35+ messages in thread
From: Marc-André Lureau @ 2015-09-29 15:26 UTC (permalink / raw)
  To: Michael S. Tsirkin
  Cc: haifeng lin, thibaut collet, jasowang, qemu-devel, pbonzini,
	marcandre lureau

Hi

----- Original Message -----
> On Thu, Sep 24, 2015 at 06:22:11PM +0200, marcandre.lureau@redhat.com wrote:
> > From: Marc-André Lureau <marcandre.lureau@redhat.com>
> > 
> > Do not allocate a shared log if the backend doesn't support it.
> > 
> > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
> So squash this in the previous patch introducing the shared log?
> This split makes review harder, not easier.

Ok, I'll try to reorder things

> 
> > ---
> >  hw/virtio/vhost-user.c            |  9 +++++++++
> >  hw/virtio/vhost.c                 | 15 ++++++++++-----
> >  include/hw/virtio/vhost-backend.h |  4 ++++
> >  3 files changed, 23 insertions(+), 5 deletions(-)
> > 
> > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> > index 41e0364..455caba 100644
> > --- a/hw/virtio/vhost-user.c
> > +++ b/hw/virtio/vhost-user.c
> > @@ -553,6 +553,14 @@ static int vhost_user_get_vq_index(struct vhost_dev
> > *dev, int idx)
> >      return idx;
> >  }
> >  
> > +static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
> > +{
> > +    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);
> > +
> > +    return virtio_has_feature(dev->protocol_features,
> > +                              VHOST_USER_PROTOCOL_F_LOG_SHMFD);
> > +}
> > +
> >  const VhostOps user_ops = {
> >          .backend_type = VHOST_BACKEND_TYPE_USER,
> >          .vhost_backend_init = vhost_user_init,
> > @@ -573,4 +581,5 @@ const VhostOps user_ops = {
> >          .vhost_reset_device = vhost_user_reset_device,
> >          .vhost_get_vq_index = vhost_user_get_vq_index,
> >          .vhost_set_vring_enable = vhost_user_set_vring_enable,
> > +        .vhost_requires_shm_log = vhost_user_requires_shm_log,
> >  };
> > diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
> > index d9f2131..ec27c7b 100644
> > --- a/hw/virtio/vhost.c
> > +++ b/hw/virtio/vhost.c
> > @@ -357,10 +357,15 @@ static void vhost_log_put(struct vhost_dev *dev, bool
> > sync)
> >      }
> >  }
> >  
> > -static inline void vhost_dev_log_resize(struct vhost_dev* dev, uint64_t
> > size)
> > +static bool vhost_dev_log_is_shared(struct vhost_dev *dev)
> >  {
> > -    bool share = dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER;
> > -    struct vhost_log *log = vhost_log_get(size, share);
> > +    return dev->vhost_ops->vhost_requires_shm_log &&
> > +           dev->vhost_ops->vhost_requires_shm_log(dev);
> > +}
> > +
> > +static inline void vhost_dev_log_resize(struct vhost_dev *dev, uint64_t
> > size)
> > +{
> > +    struct vhost_log *log = vhost_log_get(size,
> > vhost_dev_log_is_shared(dev));
> >      uint64_t log_base = (uintptr_t)log->log;
> >      int r;
> >  
> > @@ -1166,10 +1171,10 @@ int vhost_dev_start(struct vhost_dev *hdev,
> > VirtIODevice *vdev)
> >  
> >      if (hdev->log_enabled) {
> >          uint64_t log_base;
> > -        bool share = hdev->vhost_ops->backend_type ==
> > VHOST_BACKEND_TYPE_USER;
> >  
> >          hdev->log_size = vhost_get_log_size(hdev);
> > -        hdev->log = vhost_log_get(hdev->log_size, share);
> > +        hdev->log = vhost_log_get(hdev->log_size,
> > +                                  vhost_dev_log_is_shared(hdev));
> >          log_base = (uintptr_t)hdev->log->log;
> >          r = hdev->vhost_ops->vhost_set_log_base(hdev,
> >                                                  hdev->log_size ? log_base
> >                                                  : 0,
> > diff --git a/include/hw/virtio/vhost-backend.h
> > b/include/hw/virtio/vhost-backend.h
> > index 868a78a..e07118c 100644
> > --- a/include/hw/virtio/vhost-backend.h
> > +++ b/include/hw/virtio/vhost-backend.h
> > @@ -11,6 +11,8 @@
> >  #ifndef VHOST_BACKEND_H_
> >  #define VHOST_BACKEND_H_
> >  
> > +#include <stdbool.h>
> > +
> >  typedef enum VhostBackendType {
> >      VHOST_BACKEND_TYPE_NONE = 0,
> >      VHOST_BACKEND_TYPE_KERNEL = 1,
> > @@ -64,6 +66,7 @@ typedef int (*vhost_reset_device_op)(struct vhost_dev
> > *dev);
> >  typedef int (*vhost_get_vq_index_op)(struct vhost_dev *dev, int idx);
> >  typedef int (*vhost_set_vring_enable_op)(struct vhost_dev *dev,
> >                                           int enable);
> > +typedef bool (*vhost_requires_shm_log_op)(struct vhost_dev *dev);
> >  
> >  typedef struct VhostOps {
> >      VhostBackendType backend_type;
> > @@ -90,6 +93,7 @@ typedef struct VhostOps {
> >      vhost_reset_device_op vhost_reset_device;
> >      vhost_get_vq_index_op vhost_get_vq_index;
> >      vhost_set_vring_enable_op vhost_set_vring_enable;
> > +    vhost_requires_shm_log_op vhost_requires_shm_log;
> >  } VhostOps;
> >  
> >  extern const VhostOps user_ops;
> > --
> > 2.4.3
> 

^ permalink raw reply	[flat|nested] 35+ messages in thread

end of thread, other threads:[~2015-09-29 15:26 UTC | newest]

Thread overview: 35+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-09-24 16:22 [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 01/21] vhost-user: unit test for new messages marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 02/21] configure: probe for memfd marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 03/21] util: add linux-only memfd fallback marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 04/21] util: add memfd helpers marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 05/21] vhost: alloc shareable log marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 06/21] vhost: document log resizing marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 07/21] vhost: add vhost_set_log_base op marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 08/21] vhost-user: send log shm fd along with log_base marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 09/21] vhost-user: add a migration blocker marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 10/21] vhost: use a function for each call marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 11/21] vhost: only use shared log if in use by backend marcandre.lureau
2015-09-29 14:59   ` Michael S. Tsirkin
2015-09-29 15:26     ` Marc-André Lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 12/21] vhost-user: document migration log marcandre.lureau
2015-09-24 16:38   ` Eric Blake
2015-09-24 16:41     ` Marc-André Lureau
2015-09-24 16:47       ` Eric Blake
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 13/21] net: add trace_vhost_user_event marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 14/21] vhost user: add support of live migration marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 15/21] vhost user: add rarp sending after live migration for legacy guest marcandre.lureau
2015-09-24 17:24   ` Thibaut Collet
2015-09-24 21:53     ` Marc-André Lureau
2015-09-27 13:12       ` Michael S. Tsirkin
2015-09-27 15:13         ` Marc-André Lureau
2015-09-27 15:49           ` Thibaut Collet
2015-09-27 18:04             ` Marc-André Lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 16/21] vhost-user-test: move wait_for_fds() out marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 17/21] vhost-user-test: remove useless static check marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 18/21] vhost-user-test: wrap server in TestServer struct marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 19/21] vhost-user-test: learn to tweak various qemu arguments marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 20/21] vhost-user-test: add live-migration test marcandre.lureau
2015-09-24 16:22 ` [Qemu-devel] [PATCH v5 21/21] vhost-user-test: check ownership during migration marcandre.lureau
2015-09-27 13:19 ` [Qemu-devel] [PATCH v5 00/21] vhost-user: add migration support Michael S. Tsirkin
2015-09-29 15:07 ` Michael S. Tsirkin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).