IB/ipath: Fix SDMA error recovery in absence of link status change
What's fixed:
in ipath_cancel_sends()
We need to unconditionally set ABORTING. So, swap the tests
so the set_bit() isn't shadowed by the &&.
If we've disarmed the piobufs, then we need to unconditionally
set DISARMED. So, move it out from the overly protective if
at the bottom.
in sdma_abort_task()
Abort_task was written knowing that the SDMA engine would always
be reset (and restarted) on error. A recent change broke that
fundamental assumption by taking the restart portion and making
it conditional on a link status change. But, SDMA can go boom
without a link status change in some conditions.
Signed-off-by: John Gregor <john.gregor@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
diff --git a/drivers/infiniband/hw/ipath/ipath_sdma.c b/drivers/infiniband/hw/ipath/ipath_sdma.c
index 0d07682..3697449 100644
--- a/drivers/infiniband/hw/ipath/ipath_sdma.c
+++ b/drivers/infiniband/hw/ipath/ipath_sdma.c
@@ -308,13 +308,15 @@
spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags);
/*
- * Don't restart sdma here. Wait until link is up to ACTIVE.
- * VL15 MADs used to bring the link up use PIO, and multiple
- * link transitions otherwise cause the sdma engine to be
+ * Don't restart sdma here (with the exception
+ * below). Wait until link is up to ACTIVE. VL15 MADs
+ * used to bring the link up use PIO, and multiple link
+ * transitions otherwise cause the sdma engine to be
* stopped and started multiple times.
- * The disable is done here, including the shadow, so the
- * state is kept consistent.
- * See ipath_restart_sdma() for the actual starting of sdma.
+ * The disable is done here, including the shadow,
+ * so the state is kept consistent.
+ * See ipath_restart_sdma() for the actual starting
+ * of sdma.
*/
spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
dd->ipath_sendctrl &= ~INFINIPATH_S_SDMAENABLE;
@@ -326,6 +328,13 @@
/* make sure I see next message */
dd->ipath_sdma_abort_jiffies = 0;
+ /*
+ * Not everything that takes SDMA offline is a link
+ * status change. If the link was up, restart SDMA.
+ */
+ if (dd->ipath_flags & IPATH_LINKACTIVE)
+ ipath_restart_sdma(dd);
+
goto done;
}
@@ -427,7 +436,12 @@
goto done;
}
- dd->ipath_sdma_status = 0;
+ /*
+ * Set initial status as if we had been up, then gone down.
+ * This lets initial start on transition to ACTIVE be the
+ * same as restart after link flap.
+ */
+ dd->ipath_sdma_status = IPATH_SDMA_ABORT_ABORTED;
dd->ipath_sdma_abort_jiffies = 0;
dd->ipath_sdma_generation = 0;
dd->ipath_sdma_descq_tail = 0;
@@ -618,6 +632,9 @@
ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch);
spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
+ /* notify upper layers */
+ ipath_ib_piobufavail(dd->verbs_dev);
+
bail:
return;
}