Merge remote-tracking branch 'remotes/dgilbert-gitlab/tags/pull-virti…

…ofs-20200123b' into staging

virtiofsd first pull v2

Import our virtiofsd.
This pulls in the daemon to drive a file system connected to the
existing qemu virtiofsd device.
It's derived from upstream libfuse with lots of changes (and a lot
trimmed out).
The daemon lives in the newly created qemu/tools/virtiofsd

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

v2
  drop the docs while we discuss where they should live
  and we need to redo the manpage in anything but texi

# gpg: Signature made Thu 23 Jan 2020 16:45:18 GMT
# gpg:                using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A  9FA9 0516 331E BC5B FDE7

* remotes/dgilbert-gitlab/tags/pull-virtiofs-20200123b: (108 commits)
  virtiofsd: add some options to the help message
  virtiofsd: stop all queue threads on exit in virtio_loop()
  virtiofsd/passthrough_ll: Pass errno to fuse_reply_err()
  virtiofsd: Convert lo_destroy to take the lo->mutex lock itself
  virtiofsd: add --thread-pool-size=NUM option
  virtiofsd: fix lo_destroy() resource leaks
  virtiofsd: prevent FUSE_INIT/FUSE_DESTROY races
  virtiofsd: process requests in a thread pool
  virtiofsd: use fuse_buf_writev to replace fuse_buf_write for better performance
  virtiofsd: add definition of fuse_buf_writev()
  virtiofsd: passthrough_ll: Use cache_readdir for directory open
  virtiofsd: Fix data corruption with O_APPEND write in writeback mode
  virtiofsd: Reset O_DIRECT flag during file open
  virtiofsd: convert more fprintf and perror to use fuse log infra
  virtiofsd: do not always set FUSE_FLOCK_LOCKS
  virtiofsd: introduce inode refcount to prevent use-after-free
  virtiofsd: passthrough_ll: fix refcounting on remove/rename
  libvhost-user: Fix some memtable remap cases
  virtiofsd: rename inode->refcount to inode->nlookup
  virtiofsd: prevent races with lo_dirp_put()
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

.gitignore

@@ -6,6 +6,7 @@
 /config-target.*
 /config.status
 /config-temp
+/tools/virtiofsd/50-qemu-virtiofsd.json
 /elf2dmp
 /trace-events-all
 /trace/generated-events.h

MAINTAINERS

@@ -1595,6 +1595,14 @@ T: git https://github.com/cohuck/qemu.git s390-next
 T: git https://github.com/borntraeger/qemu.git s390-next
 L: qemu-s390x@nongnu.org
 
+virtiofs
+M: Dr. David Alan Gilbert <dgilbert@redhat.com>
+M: Stefan Hajnoczi <stefanha@redhat.com>
+S: Supported
+F: tools/virtiofsd/*
+F: hw/virtio/vhost-user-fs*
+F: include/hw/virtio/vhost-user-fs.h
+
 virtio-input
 M: Gerd Hoffmann <kraxel@redhat.com>
 S: Maintained

Makefile

@@ -327,6 +327,11 @@ HELPERS-y += vhost-user-gpu$(EXESUF)
 vhost-user-json-y += contrib/vhost-user-gpu/50-qemu-gpu.json
 endif
 
+ifeq ($(CONFIG_LINUX)$(CONFIG_SECCOMP)$(CONFIG_LIBCAP_NG),yyy)
+HELPERS-y += virtiofsd$(EXESUF)
+vhost-user-json-y += tools/virtiofsd/50-qemu-virtiofsd.json
+endif
+
 # Sphinx does not allow building manuals into the same directory as
 # the source files, so if we're doing an in-tree QEMU build we must
 # build the manuals into a subdirectory (and then install them from
@@ -431,6 +436,7 @@ dummy := $(call unnest-vars,, \
                 elf2dmp-obj-y \
                 ivshmem-client-obj-y \
                 ivshmem-server-obj-y \
+                virtiofsd-obj-y \
                 rdmacm-mux-obj-y \
                 libvhost-user-obj-y \
                 vhost-user-scsi-obj-y \
@@ -670,6 +676,12 @@ rdmacm-mux$(EXESUF): LIBS += "-libumad"
 rdmacm-mux$(EXESUF): $(rdmacm-mux-obj-y) $(COMMON_LDADDS)
 	$(call LINK, $^)
 
+# relies on Linux-specific syscalls
+ifeq ($(CONFIG_LINUX)$(CONFIG_SECCOMP)$(CONFIG_LIBCAP_NG),yyy)
+virtiofsd$(EXESUF): $(virtiofsd-obj-y) libvhost-user.a $(COMMON_LDADDS)
+	$(call LINK, $^)
+endif
+
 vhost-user-gpu$(EXESUF): $(vhost-user-gpu-obj-y) $(libvhost-user-obj-y) libqemuutil.a libqemustub.a
 	$(call LINK, $^)
 

Makefile.objs

@@ -123,6 +123,7 @@ vhost-user-blk-obj-y = contrib/vhost-user-blk/
 rdmacm-mux-obj-y = contrib/rdmacm-mux/
 vhost-user-input-obj-y = contrib/vhost-user-input/
 vhost-user-gpu-obj-y = contrib/vhost-user-gpu/
+virtiofsd-obj-y = tools/virtiofsd/
 
 ######################################################################
 trace-events-subdirs =

configure

@@ -5197,6 +5197,19 @@ if compile_prog "" "" ; then
     strchrnul=yes
 fi
 
+#########################################
+# check if we have st_atim
+
+st_atim=no
+cat > $TMPC << EOF
+#include <sys/stat.h>
+#include <stddef.h>
+int main(void) { return offsetof(struct stat, st_atim); }
+EOF
+if compile_prog "" "" ; then
+    st_atim=yes
+fi
+
 ##########################################
 # check if trace backend exists
 
@@ -6895,6 +6908,9 @@ fi
 if test "$strchrnul" = "yes" ; then
   echo "HAVE_STRCHRNUL=y" >> $config_host_mak
 fi
+if test "$st_atim" = "yes" ; then
+  echo "HAVE_STRUCT_STAT_ST_ATIM=y" >> $config_host_mak
+fi
 if test "$byteswap_h" = "yes" ; then
   echo "CONFIG_BYTESWAP_H=y" >> $config_host_mak
 fi

contrib/libvhost-user/libvhost-user.c

@@ -392,26 +392,37 @@ vu_send_reply(VuDev *dev, int conn_fd, VhostUserMsg *vmsg)
     return vu_message_write(dev, conn_fd, vmsg);
 }
 
+/*
+ * Processes a reply on the slave channel.
+ * Entered with slave_mutex held and releases it before exit.
+ * Returns true on success.
+ */
 static bool
 vu_process_message_reply(VuDev *dev, const VhostUserMsg *vmsg)
 {
     VhostUserMsg msg_reply;
+    bool result = false;
 
     if ((vmsg->flags & VHOST_USER_NEED_REPLY_MASK) == 0) {
-        return true;
+        result = true;
+        goto out;
     }
 
     if (!vu_message_read(dev, dev->slave_fd, &msg_reply)) {
-        return false;
+        goto out;
     }
 
     if (msg_reply.request != vmsg->request) {
         DPRINT("Received unexpected msg type. Expected %d received %d",
                vmsg->request, msg_reply.request);
-        return false;
+        goto out;
     }
 
-    return msg_reply.payload.u64 == 0;
+    result = msg_reply.payload.u64 == 0;
+
+out:
+    pthread_mutex_unlock(&dev->slave_mutex);
+    return result;
 }
 
 /* Kick the log_call_fd if required. */
@@ -553,6 +564,21 @@ vu_reset_device_exec(VuDev *dev, VhostUserMsg *vmsg)
     return false;
 }
 
+static bool
+map_ring(VuDev *dev, VuVirtq *vq)
+{
+    vq->vring.desc = qva_to_va(dev, vq->vra.desc_user_addr);
+    vq->vring.used = qva_to_va(dev, vq->vra.used_user_addr);
+    vq->vring.avail = qva_to_va(dev, vq->vra.avail_user_addr);
+
+    DPRINT("Setting virtq addresses:\n");
+    DPRINT("    vring_desc  at %p\n", vq->vring.desc);
+    DPRINT("    vring_used  at %p\n", vq->vring.used);
+    DPRINT("    vring_avail at %p\n", vq->vring.avail);
+
+    return !(vq->vring.desc && vq->vring.used && vq->vring.avail);
+}
+
 static bool
 vu_set_mem_table_exec_postcopy(VuDev *dev, VhostUserMsg *vmsg)
 {
@@ -756,6 +782,14 @@ vu_set_mem_table_exec(VuDev *dev, VhostUserMsg *vmsg)
         close(vmsg->fds[i]);
     }
 
+    for (i = 0; i < dev->max_queues; i++) {
+        if (dev->vq[i].vring.desc) {
+            if (map_ring(dev, &dev->vq[i])) {
+                vu_panic(dev, "remaping queue %d during setmemtable", i);
+            }
+        }
+    }
+
     return false;
 }
 
@@ -842,18 +876,12 @@ vu_set_vring_addr_exec(VuDev *dev, VhostUserMsg *vmsg)
     DPRINT("    avail_user_addr:  0x%016" PRIx64 "\n", vra->avail_user_addr);
     DPRINT("    log_guest_addr:   0x%016" PRIx64 "\n", vra->log_guest_addr);
 
+    vq->vra = *vra;
     vq->vring.flags = vra->flags;
-    vq->vring.desc = qva_to_va(dev, vra->desc_user_addr);
-    vq->vring.used = qva_to_va(dev, vra->used_user_addr);
-    vq->vring.avail = qva_to_va(dev, vra->avail_user_addr);
     vq->vring.log_guest_addr = vra->log_guest_addr;
 
-    DPRINT("Setting virtq addresses:\n");
-    DPRINT("    vring_desc  at %p\n", vq->vring.desc);
-    DPRINT("    vring_used  at %p\n", vq->vring.used);
-    DPRINT("    vring_avail at %p\n", vq->vring.avail);
 
-    if (!(vq->vring.desc && vq->vring.used && vq->vring.avail)) {
+    if (map_ring(dev, vq)) {
         vu_panic(dev, "Invalid vring_addr message");
         return false;
     }
@@ -1105,10 +1133,13 @@ bool vu_set_queue_host_notifier(VuDev *dev, VuVirtq *vq, int fd,
         return false;
     }
 
+    pthread_mutex_lock(&dev->slave_mutex);
     if (!vu_message_write(dev, dev->slave_fd, &vmsg)) {
+        pthread_mutex_unlock(&dev->slave_mutex);
         return false;
     }
 
+    /* Also unlocks the slave_mutex */
     return vu_process_message_reply(dev, &vmsg);
 }
 
@@ -1628,6 +1659,7 @@ vu_deinit(VuDev *dev)
         close(dev->slave_fd);
         dev->slave_fd = -1;
     }
+    pthread_mutex_destroy(&dev->slave_mutex);
 
     if (dev->sock != -1) {
         close(dev->sock);
@@ -1663,6 +1695,7 @@ vu_init(VuDev *dev,
     dev->remove_watch = remove_watch;
     dev->iface = iface;
     dev->log_call_fd = -1;
+    pthread_mutex_init(&dev->slave_mutex, NULL);
     dev->slave_fd = -1;
     dev->max_queues = max_queues;
 

contrib/libvhost-user/libvhost-user.h

@@ -19,6 +19,7 @@
 #include <stddef.h>
 #include <sys/poll.h>
 #include <linux/vhost.h>
+#include <pthread.h>
 #include "standard-headers/linux/virtio_ring.h"
 
 /* Based on qemu/hw/virtio/vhost-user.c */
@@ -326,6 +327,9 @@ typedef struct VuVirtq {
     int err_fd;
     unsigned int enable;
     bool started;
+
+    /* Guest addresses of our ring */
+    struct vhost_vring_addr vra;
 } VuVirtq;
 
 enum VuWatchCondtion {
@@ -355,6 +359,8 @@ struct VuDev {
     VuVirtq *vq;
     VuDevInflightInfo inflight_info;
     int log_call_fd;
+    /* Must be held while using slave_fd */
+    pthread_mutex_t slave_mutex;
     int slave_fd;
     uint64_t log_size;
     uint8_t *log_table;

docs/interop/vhost-user.json

@@ -31,6 +31,7 @@
 # @rproc-serial: virtio remoteproc serial link
 # @scsi: virtio scsi
 # @vsock: virtio vsock transport
+# @fs: virtio fs (since 4.2)
 #
 # Since: 4.0
 ##
@@ -50,7 +51,8 @@
       'rpmsg',
       'rproc-serial',
       'scsi',
-      'vsock'
+      'vsock',
+      'fs'
   ]
 }
 

hw/virtio/vhost-user.c

@@ -1061,7 +1061,7 @@ static void slave_read(void *opaque)
                                                           fd[0]);
         break;
     default:
-        error_report("Received unexpected msg type.");
+        error_report("Received unexpected msg type: %d.", hdr.request);
         ret = -EINVAL;
     }