Re: [kvm-unit-tests PATCH v2 1/8] s390x: lib: css: disabling a subchannel

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

 





On 3/29/21 10:00 AM, Thomas Huth wrote:
On 25/03/2021 10.39, Pierre Morel wrote:
Some tests require to disable a subchannel.
Let's implement the css_disable() function.

Signed-off-by: Pierre Morel <pmorel@xxxxxxxxxxxxx>
Reviewed-by: Cornelia Huck <cohuck@xxxxxxxxxx>
---
  lib/s390x/css.h     |  1 +
  lib/s390x/css_lib.c | 67 +++++++++++++++++++++++++++++++++++++++++++++
  2 files changed, 68 insertions(+)

diff --git a/lib/s390x/css.h b/lib/s390x/css.h
index 7e3d261..b0de3a3 100644
--- a/lib/s390x/css.h
+++ b/lib/s390x/css.h
@@ -284,6 +284,7 @@ int css_enumerate(void);
  #define IO_SCH_ISC      3
  int css_enable(int schid, int isc);
  bool css_enabled(int schid);
+int css_disable(int schid);
  /* Library functions */
  int start_ccw1_chain(unsigned int sid, struct ccw1 *ccw);
diff --git a/lib/s390x/css_lib.c b/lib/s390x/css_lib.c
index efc7057..f5c4f37 100644
--- a/lib/s390x/css_lib.c
+++ b/lib/s390x/css_lib.c
@@ -186,6 +186,73 @@ bool css_enabled(int schid)
      }
      return true;
  }
+
+/*
+ * css_disable: disable the subchannel
+ * @schid: Subchannel Identifier
+ * Return value:
+ *   On success: 0
+ *   On error the CC of the faulty instruction
+ *      or -1 if the retry count is exceeded.
+ */
+int css_disable(int schid)
+{
+    struct pmcw *pmcw = &schib.pmcw;
+    int retry_count = 0;
+    int cc;
+
+    /* Read the SCHIB for this subchannel */
+    cc = stsch(schid, &schib);
+    if (cc) {
+        report_info("stsch: sch %08x failed with cc=%d", schid, cc);
+        return cc;
+    }
+
+    if (!(pmcw->flags & PMCW_ENABLE)) {
+        report_info("stsch: sch %08x already disabled", schid);
+        return 0;
+    }
+
+retry:

I have to saythat I really dislike writing loops with gotos if it can be avoided easily. What about:

for (retry_count = 0; retry_count < MAX_ENABLE_RETRIES; retry_count++) ?

(and maybe rename that variable to "retries" to keep it short?)

hum, you already said that.
Sorry, I forgot and duplicated css_enable()

done.


...


+    /* Read the SCHIB again to verify the enablement */

"verify the disablement" ?

:) yes


+    cc = stsch(schid, &schib);
+    if (cc) {
+        report_info("stsch: updating sch %08x failed with cc=%d",
+                schid, cc);
+        return cc;
+    }
+
+    if (!(pmcw->flags & PMCW_ENABLE)) {
+        if (retry_count)
+            report_info("stsch: sch %08x successfully disabled after %d retries",
+                    schid, retry_count);
+        return 0;
+    }
+
+    if (retry_count++ < MAX_ENABLE_RETRIES) {
+        /* the hardware was not ready, give it some time */
+        mdelay(10);
+        goto retry;
+    }
+
+    report_info("msch: modifying sch %08x failed after %d retries. pmcw flags: %04x",
+            schid, retry_count, pmcw->flags);
+    return -1;
+}
  /*
   * css_enable: enable the subchannel with the specified ISC
   * @schid: Subchannel Identifier


  Thomas


Thanks,
Pierre

--
Pierre Morel
IBM Lab Boeblingen



[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux