On 03.12.18 10:50, Cornelia Huck wrote:
On Fri, 30 Nov 2018 15:32:15 +0100
Michael Mueller <mimu@xxxxxxxxxxxxx> wrote:
By initializing the GIB, it will be used by the kvm host.
Signed-off-by: Michael Mueller <mimu@xxxxxxxxxxxxx>
---
Documentation/kmsg/s390/kvm | 12 ++++++++++++
arch/s390/kvm/kvm-s390.c | 10 +++++++++-
2 files changed, 21 insertions(+), 1 deletion(-)
diff --git a/Documentation/kmsg/s390/kvm b/Documentation/kmsg/s390/kvm
index 76ffe2012254..6908476a54f1 100644
--- a/Documentation/kmsg/s390/kvm
+++ b/Documentation/kmsg/s390/kvm
@@ -10,3 +10,15 @@
* If this problem persists after a reload of the kvm kernel module, gather
* Linux debug data and contact your support organization.
*/
+
+/*?
+ * Text: "Failed to initialize GIB rc=%d\n"
+ * Severity: Error
+ * Parameter:
+ * @1: return code of the GIB initialization call
+ * Description:
+ * The initialization if the GIB (Guest Information Block) was not successful.
+ * User action:
+ * If this problem persists after a reload of the kvm kernel module, gather
+ * Linux debug data and contact your support organization.
+ */
This does not exist upstream. Wrong branch?
No, not the wrong branch, wrong file. These are the very first kmsgs for
kvm and
it turns out that we defined
#define KMSG_COMPONENT "kvm-s390"
as the location.
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 2d10e175862c..5f068f3942da 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -428,16 +428,24 @@ int kvm_arch_init(void *opaque)
goto out_debug_unreg;
}
+ rc = kvm_s390_gib_init(GAL_ISC);
+ if (rc) {
+ pr_err("Failed to initialize GIB rc=%d\n", rc);
+ goto out_debug_unreg;
+ }
+
kvm_s390_cpu_feat_init();
/* Register floating interrupt controller interface. */
rc = kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
if (rc) {
pr_err("Failed to register FLIC rc=%d\n", rc);
- goto out_debug_unreg;
+ goto out_gib_destroy;
}
return 0;
+out_gib_destroy:
+ kvm_s390_gib_destroy();
out_debug_unreg:
debug_unregister(kvm_s390_dbf);
return rc;
