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

Re: [libvirt] [PATCH 6/7] node_device: introduce new capability FC_RPORT



John Ferlan <jferlan redhat com> [2017-05-25, 03:26PM -0400]:


On 05/22/2017 02:38 AM, Bjoern Walk wrote:
Similar to scsi_host and fc_host, there is a relation between a
scsi_target and its transport specific fc_remote_port. Let's expose this
relation and relevant information behind it.

An example for a virsh nodedev-dumpxml:

    virsh # nodedev-dumpxml scsi_target0_0_0
    <device>
      <name>scsi_target0_0_0</name>
      <path>/sys/devices/[...]/host0/rport-0:0-0/target0:0:0</path>
      <parent>scsi_host0</parent>
      <capability type='scsi_target'>
        <target>target0:0:0</target>
        <capability type='fc_remote_port'>
          <rport>rport-0:0-0</rport>
          <wwpn>0x9d73bc45f0e21a86</wwpn>
        </capability>
      </capability>
    </device>

Reviewed-by: Boris Fiuczynski <fiuczy linux vnet ibm com>
Signed-off-by: Bjoern Walk <bwalk linux vnet ibm com>
---
 docs/schemas/nodedev.rng                     | 20 ++++++++
 src/conf/node_device_conf.c                  | 75 +++++++++++++++++++++++++++-
 src/conf/node_device_conf.h                  |  7 +++
 src/node_device/node_device_driver.c         |  5 +-
 src/node_device/node_device_linux_sysfs.c    | 56 +++++++++++++++++++++
 src/node_device/node_device_linux_sysfs.h    |  2 +
 src/node_device/node_device_udev.c           |  4 +-
 tests/nodedevschemadata/scsi_target1_0_0.xml | 12 +++++
 tests/nodedevxml2xmltest.c                   |  1 +
 9 files changed, 178 insertions(+), 4 deletions(-)
 create mode 100644 tests/nodedevschemadata/scsi_target1_0_0.xml

diff --git a/docs/schemas/nodedev.rng b/docs/schemas/nodedev.rng
index 87bfb0c28..5bcf31787 100644
--- a/docs/schemas/nodedev.rng
+++ b/docs/schemas/nodedev.rng
@@ -458,6 +458,20 @@
     </optional>
   </define>

+  <define name='capsfcrport'>
+    <attribute name='type'>
+      <value>fc_remote_port</value>
+    </attribute>
+
+    <element name='rport'>
+      <text/>
+    </element>
+
+    <element name='wwpn'>
+      <ref name='wwn'/>

NB: 'wwn' is a string...

+    </element>
+  </define>
+
   <define name='capscsitarget'>
     <attribute name='type'>
       <value>scsi_target</value>
@@ -466,6 +480,12 @@
     <element name='target'>
       <text/>
     </element>
+
+    <optional>
+      <element name='capability'>
+        <ref name='capsfcrport'/>
+      </element>
+    </optional>
   </define>

   <define name='capscsi'>
diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c
index 28d4907f5..ed003d37b 100644
--- a/src/conf/node_device_conf.c
+++ b/src/conf/node_device_conf.c
@@ -563,6 +563,16 @@ virNodeDeviceDefFormat(const virNodeDeviceDef *def)
         case VIR_NODE_DEV_CAP_SCSI_TARGET:
             virBufferEscapeString(&buf, "<target>%s</target>\n",
                                   data->scsi_target.name);
+            if (data->scsi_target.flags & VIR_NODE_DEV_CAP_FLAG_FC_RPORT) {
+                virBufferAddLit(&buf, "<capability type='fc_remote_port'>\n");
+                virBufferAdjustIndent(&buf, 2);
+                virBufferAsprintf(&buf, "<rport>%s</rport>\n",
+                                  data->scsi_target.rport);
+                virBufferAsprintf(&buf, "<wwpn>0x%llx</wwpn>\n",
+                                  data->scsi_target.wwpn);

If you don't convert to a number, then you don't need to perform this
conversion as it's just a %s print.

+                virBufferAdjustIndent(&buf, -2);
+                virBufferAddLit(&buf, "</capability>\n");
+            }
             break;
         case VIR_NODE_DEV_CAP_SCSI:
             virNodeDeviceCapSCSIDefFormat(&buf, data);
@@ -932,8 +942,10 @@ virNodeDevCapSCSITargetParseXML(xmlXPathContextPtr ctxt,
                                 xmlNodePtr node,
                                 virNodeDevCapSCSITargetPtr scsi_target)
 {
-    xmlNodePtr orignode;
-    int ret = -1;
+    xmlNodePtr orignode, *nodes = NULL;
+    int ret = -1, n = 0;
+    size_t i;
+    char *type = NULL, *wwpn = NULL;

     orignode = ctxt->node;
     ctxt->node = node;
@@ -946,10 +958,68 @@ virNodeDevCapSCSITargetParseXML(xmlXPathContextPtr ctxt,
         goto out;
     }

+    if ((n = virXPathNodeSet("./capability", ctxt, &nodes)) < 0)
+        goto out;
+
+    for (i = 0; i < n; ++i) {
+        type = virXMLPropString(nodes[i], "type");
+
+        if (!type) {
+            virReportError(VIR_ERR_INTERNAL_ERROR,
+                           _("missing type for SCSI target capability for '%s'"),
+                           def->name);
+            goto out;
+        }
+
+        if (STREQ(type, "fc_remote_port")) {
+            xmlNodePtr orignode2;
+
+            scsi_target->flags |= VIR_NODE_DEV_CAP_FLAG_FC_RPORT;
+
+            orignode2 = ctxt->node;
+            ctxt->node = nodes[i];

Like the virNodeDevCapSCSIHostParseXML I assume this only parses the
first 'fc_remote_port'... Trying to think whether a VIR_FREE would be
necessary here (and perhaps from SCSIHost code just in case there were
more than one <rport> or <wwpn> provided...


More then one fc_remote_port per scsi_target would not make any sense in
my opinion, at least if I understood the entity relation in the kernel
correctly. The RNG only (hopefully) allows for one fc_remote_port
capability as well.

+
+            if (virNodeDevCapsDefParseString("string(./rport[1])",
+                                             ctxt,
+                                             &scsi_target->rport) < 0) {
+                virReportError(VIR_ERR_INTERNAL_ERROR,
+                               _("missing rport name for '%s'"), def->name);
+                goto out;
+            }
+
+            if (virNodeDevCapsDefParseString("string(./wwpn[1])",
+                                             ctxt, &wwpn) < 0) {

If you don't do the conversion, this could just be a straight parse into
scsi_target->wwpn

+                virReportError(VIR_ERR_INTERNAL_ERROR,
+                               _("missing wwpn identifier for '%s'"),
+                               def->name);
+                goto out;
+            }
+
+            if (virStrToLong_ullp(wwpn, NULL, 16, &scsi_target->wwpn) < 0) {
+                virReportError(VIR_ERR_INTERNAL_ERROR,
+                               _("invalid wwpn identifier '%s' for '%s'"),
+                               wwpn, def->name);

Why even both with this conversion?

Yeah, actually you are right. I don't ever use the numerical value of
the WWPN. Somehow in the back of my head my SCSI guru yells at me: "WWN
should be stored in memory as 64-bit integer values!". But true,
without the conversion the code will be easier.


+                goto out;
+            }
+
+            ctxt->node = orignode2;
+        } else {
+            virReportError(VIR_ERR_INTERNAL_ERROR,
+                           _("unknown SCSI target capability type '%s' for '%s'"),
+                           type, def->name);
+            goto out;
+        }
+
+        VIR_FREE(type);
+        VIR_FREE(wwpn);
+    }
+
     ret = 0;

  out:
     ctxt->node = orignode;
+    VIR_FREE(type);
+    VIR_FREE(wwpn);

If you don't do the conversion, then this VIR_FREE is unnecessary.

Coverity complained that nodes was leaked, so I added a VIR_FREE(nodes);
here.

     return ret;
 }

@@ -2132,6 +2202,7 @@ virNodeDevCapsDefFree(virNodeDevCapsDefPtr caps)
         break;
     case VIR_NODE_DEV_CAP_SCSI_TARGET:
         VIR_FREE(data->scsi_target.name);
+        VIR_FREE(data->scsi_target.rport);

Without the conversion there's a VIR_FREE here for wwpn

         break;
     case VIR_NODE_DEV_CAP_SCSI:
         VIR_FREE(data->scsi.type);
diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h
index 285841edc..f5267efda 100644
--- a/src/conf/node_device_conf.h
+++ b/src/conf/node_device_conf.h
@@ -93,6 +93,10 @@ typedef enum {
 } virNodeDevSCSIHostCapFlags;

 typedef enum {
+    VIR_NODE_DEV_CAP_FLAG_FC_RPORT			= (1 << 0),
+} virNodeDevSCSITargetCapsFlags;
+
+typedef enum {
     VIR_NODE_DEV_CAP_FLAG_PCI_PHYSICAL_FUNCTION     = (1 << 0),
     VIR_NODE_DEV_CAP_FLAG_PCI_VIRTUAL_FUNCTION      = (1 << 1),
     VIR_NODE_DEV_CAP_FLAG_PCIE                      = (1 << 2),
@@ -227,6 +231,9 @@ typedef struct _virNodeDevCapSCSITarget virNodeDevCapSCSITarget;
 typedef virNodeDevCapSCSITarget *virNodeDevCapSCSITargetPtr;
 struct _virNodeDevCapSCSITarget {
     char *name;
+    unsigned int flags; /* enum virNodeDevSCSITargetCapsFlags */
+    char *rport;
+    unsigned long long wwpn;

and this would be a char *wwpn;

 };

 typedef struct _virNodeDevCapSCSI virNodeDevCapSCSI;
diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c
index 286af26bc..acb8b89af 100644
--- a/src/node_device/node_device_driver.c
+++ b/src/node_device/node_device_driver.c
@@ -56,6 +56,10 @@ static int update_caps(virNodeDeviceObjPtr dev)
         case VIR_NODE_DEV_CAP_SCSI_HOST:
             nodeDeviceSysfsGetSCSIHostCaps(&cap->data.scsi_host);
             break;
+        case VIR_NODE_DEV_CAP_SCSI_TARGET:
+            nodeDeviceSysfsGetSCSITargetCaps(dev->def->sysfs_path,
+                                             &cap->data.scsi_target);
+            break;
         case VIR_NODE_DEV_CAP_NET:
             if (virNetDevGetLinkInfo(cap->data.net.ifname, &cap->data.net.lnk) < 0)
                 return -1;
@@ -76,7 +80,6 @@ static int update_caps(virNodeDeviceObjPtr dev)
         case VIR_NODE_DEV_CAP_SYSTEM:
         case VIR_NODE_DEV_CAP_USB_DEV:
         case VIR_NODE_DEV_CAP_USB_INTERFACE:
-        case VIR_NODE_DEV_CAP_SCSI_TARGET:
         case VIR_NODE_DEV_CAP_SCSI:
         case VIR_NODE_DEV_CAP_STORAGE:
         case VIR_NODE_DEV_CAP_FC_HOST:
diff --git a/src/node_device/node_device_linux_sysfs.c b/src/node_device/node_device_linux_sysfs.c
index 1b7aa9435..90452fb29 100644
--- a/src/node_device/node_device_linux_sysfs.c
+++ b/src/node_device/node_device_linux_sysfs.c
@@ -26,11 +26,13 @@
 #include <sys/stat.h>
 #include <stdlib.h>

+#include "dirname.h"
 #include "node_device_driver.h"
 #include "node_device_hal.h"
 #include "node_device_linux_sysfs.h"
 #include "virerror.h"
 #include "viralloc.h"
+#include "virfcp.h"
 #include "virlog.h"
 #include "virfile.h"
 #include "virscsihost.h"
@@ -124,6 +126,54 @@ nodeDeviceSysfsGetSCSIHostCaps(virNodeDevCapSCSIHostPtr scsi_host)
 }


+int
+nodeDeviceSysfsGetSCSITargetCaps(const char *sysfsPath,
+                                 virNodeDevCapSCSITargetPtr scsi_target)
+{
+    int ret = -1;
+    char *dir = NULL, *rport = NULL, *wwpn = NULL;
+
+    VIR_DEBUG("Checking if '%s' is an FC remote port", scsi_target->name);
+
+    /* /sys/devices/[...]/host0/rport-0:0-0/target0:0:0 -> rport-0:0-0 */
+    if (!(dir = mdir_name(sysfsPath)))
+        return -1;
+
+    if (VIR_STRDUP(rport, last_component(dir)) < 0)
+        goto cleanup;
+
+    if (!virFCIsCapableRport(rport))
+        goto cleanup;
+
+    VIR_FREE(scsi_target->rport);
+    scsi_target->rport = rport;

  VIR_STEAL_PTR(scsi_target->rport, rport);


Yeah, that's better.

+
+    if (virFCReadRportValue(scsi_target->rport, "port_name", &wwpn) < 0) {
+        VIR_WARN("Failed to read port_name for '%s'", scsi_target->rport);
+        goto cleanup;
+    }
+
+    if (virStrToLong_ullp(wwpn, NULL, 16, &scsi_target->wwpn) < 0) {
+        VIR_WARN("Failed to parse value of port_name '%s'", wwpn);
+        goto cleanup;
+    }
+
+    scsi_target->flags |= VIR_NODE_DEV_CAP_FLAG_FC_RPORT;
+    ret = 0;
+
+ cleanup:
+    if (ret < 0) {
+        VIR_FREE(scsi_target->rport);
+        scsi_target->wwpn = 0;
+        scsi_target->flags &= ~VIR_NODE_DEV_CAP_FLAG_FC_RPORT;
+    }

Coverity complained that rport could be leaked, so I added the
VIR_STEAL_PTR above and a VIR_FREE(rport); here.

+    VIR_FREE(dir);
+    VIR_FREE(wwpn);
+
+    return ret;
+}
+
+
 static int
 nodeDeviceSysfsGetPCISRIOVCaps(const char *sysfsPath,
                                virNodeDevCapPCIDevPtr pci_dev)
@@ -230,6 +280,12 @@ nodeDeviceSysfsGetSCSIHostCaps(virNodeDevCapSCSIHostPtr scsi_host ATTRIBUTE_UNUS
     return -1;
 }


Reviewed-by: John Ferlan <jferlan redhat com>

Thanks.


I can make the suggested changes if you'd like or you can send a v2 or a
squash diff...  Let me know

John

--
libvir-list mailing list
libvir-list redhat com
https://www.redhat.com/mailman/listinfo/libvir-list


--
IBM Systems
Linux on z Systems & Virtualization Development
------------------------------------------------------------------------
IBM Deutschland
Schönaicher Str. 220
71032 Böblingen
Phone: +49 7031 16 1819
E-Mail: bwalk de ibm com
------------------------------------------------------------------------
IBM Deutschland Research & Development GmbH
Vorsitzende des Aufsichtsrats: Martina Koederitz
Geschäftsführung: Dirk Wittkopp
Sitz der Gesellschaft: Böblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294

Attachment: signature.asc
Description: PGP signature


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