[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH V9 1/2] libxl: Add support for Virtio disk configuration




On 10.06.22 20:18, Oleksandr wrote:

Hello Anthony


On 10.06.22 18:12, Anthony PERARD wrote:

Hello Anthony

On Wed, Jun 01, 2022 at 08:57:40PM +0300, Oleksandr Tyshchenko wrote:
diff --git a/tools/libs/light/libxl_disk.c b/tools/libs/light/libxl_disk.c
index a5ca778..e90bc25 100644
--- a/tools/libs/light/libxl_disk.c
+++ b/tools/libs/light/libxl_disk.c
@@ -163,6 +163,25 @@ static int libxl__device_disk_setdefault(libxl__gc *gc, uint32_t domid,       rc = libxl__resolve_domid(gc, disk->backend_domname, &disk->backend_domid);
      if (rc < 0) return rc;
  +    if (disk->specification == LIBXL_DISK_SPECIFICATION_UNKNOWN)
+        disk->specification = LIBXL_DISK_SPECIFICATION_XEN;
+
+    if (disk->specification == LIBXL_DISK_SPECIFICATION_XEN &&
+        disk->transport != LIBXL_DISK_TRANSPORT_UNKNOWN) {
+        LOGD(ERROR, domid, "Transport is only supported for specification virtio");
+        return ERROR_FAIL;
Could you return ERROR_INVAL instead?

yes



+    }
+
+    /* Force transport mmio for specification virtio for now */
+    if (disk->specification == LIBXL_DISK_SPECIFICATION_VIRTIO) {
+        if (!(disk->transport == LIBXL_DISK_TRANSPORT_UNKNOWN ||
+              disk->transport == LIBXL_DISK_TRANSPORT_MMIO)) {
+            LOGD(ERROR, domid, "Unsupported transport for specification virtio");
+            return ERROR_FAIL;
Same here.

yes



+        }
+        disk->transport = LIBXL_DISK_TRANSPORT_MMIO;
+    }
+
      /* Force Qdisk backend for CDROM devices of guests with a device model. */
      if (disk->is_cdrom != 0 &&
          libxl__domain_type(gc, domid) == LIBXL_DOMAIN_TYPE_HVM) {
@@ -575,6 +660,41 @@ cleanup:
      return rc;
  }
  +static int libxl__device_disk_get_path(libxl__gc *gc, uint32_t domid,
+                                       char **path)
+{
+    const char *dir;
+    int rc;
+
+    /*
+     * As we don't know exactly what device kind to be used here, guess it
+     * by checking the presence of the corresponding path in Xenstore.
+     * First, try to read path for vbd device (default) and if not exists +     * read path for virtio_disk device. This will work as long as both Xen PV
+     * and Virtio disk devices are not assigned to the same guest.
+     */
+    *path = GCSPRINTF("%s/device/%s",
+                      libxl__xs_libxl_path(gc, domid),
+ libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VBD));
+
+    rc = libxl__xs_read_checked(gc, XBT_NULL, *path, &dir);
+    if (rc)
+        return rc;
+
+    if (dir)
+        return 0;
+
+    *path = GCSPRINTF("%s/device/%s",
+                      libxl__xs_libxl_path(gc, domid),
+ libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VIRTIO_DISK));
+
+    rc = libxl__xs_read_checked(gc, XBT_NULL, *path, &dir);
+    if (rc)
+        return rc;
+
+    return 0;
I still don't like this implementation of get_path() which return a
different answer depending on the environment which can change from one
call to the next. I think get_path() was introduced when the path for
the kind of device didn't correspond to the common path which other kind
of devices uses. And when get_path() is implemented, it always returns
the same answer (see libxl_pci.c for the only implementation).

I don't really know how to deal with a type of device that have two
different frontend kind at the moment. (But maybe there's something in
libxl_usb.c which could be useful as a potential example, but one of the
kind doesn't use xenstore so is probably easier to deal with.). So I
guess we are stuck with an implementation of get_path() which may return
a different answer depending on thing outside of libxl's full control.

So, could you at least make it much harder to have libxl's view of a
guest in a weird state? I mean:
- always check both xenstore paths
   -> if both exist, return an error
   -> if only one exist, return that one.
   -> default to the "vbd" kind, otherwise.

I think I got your idea, will try to do this way

I hope that this (it seems working) function follows your suggestion. If no objections, I will use it for V10.


static int libxl__device_disk_get_path(libxl__gc *gc, uint32_t domid,
                                       char **path)
{
    const char *xen_dir, *virtio_dir;
    char *xen_path, *virtio_path;
    int rc;

    xen_path = GCSPRINTF("%s/device/%s",
                         libxl__xs_libxl_path(gc, domid),
libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VBD));

    rc = libxl__xs_read_checked(gc, XBT_NULL, xen_path, &xen_dir);
    if (rc)
        return rc;

    virtio_path = GCSPRINTF("%s/device/%s",
                            libxl__xs_libxl_path(gc, domid),
libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VIRTIO_DISK));

    rc = libxl__xs_read_checked(gc, XBT_NULL, virtio_path, &virtio_dir);
    if (rc)
        return rc;

    if (xen_dir && virtio_dir) {
        LOGD(ERROR, domid, "Invalid configuration, both xen and virtio paths are present");
        return ERROR_INVAL;
    } else if (virtio_dir)
        *path = virtio_path;
    else
        *path = xen_path;

    return 0;
}





That would be better than the current implementation which returns the
"virtio" path by default.

agree,

making virtio by default wasn't the original intention, it was entirely involuntarily))



Thanks,

--
Regards,

Oleksandr Tyshchenko




 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.