[PATCH v2] powerpc/pseries: Don't enforce MSI affinity with kdump

From: Greg Kurz
Date: Mon Feb 15 2021 - 04:46:12 EST


Depending on the number of online CPUs in the original kernel, it is
likely for CPU #0 to be offline in a kdump kernel. The associated IRQs
in the affinity mappings provided by irq_create_affinity_masks() are
thus not started by irq_startup(), as per-design with managed IRQs.

This can be a problem with multi-queue block devices driven by blk-mq :
such a non-started IRQ is very likely paired with the single queue
enforced by blk-mq during kdump (see blk_mq_alloc_tag_set()). This
causes the device to remain silent and likely hangs the guest at
some point.

This is a regression caused by commit 9ea69a55b3b9 ("powerpc/pseries:
Pass MSI affinity to irq_create_mapping()"). Note that this only happens
with the XIVE interrupt controller because XICS has a workaround to bypass
affinity, which is activated during kdump with the "noirqdistrib" kernel
parameter.

The issue comes from a combination of factors:
- discrepancy between the number of queues detected by the multi-queue
block driver, that was used to create the MSI vectors, and the single
queue mode enforced later on by blk-mq because of kdump (i.e. keeping
all queues fixes the issue)
- CPU#0 offline (i.e. kdump always succeed with CPU#0)

Given that I couldn't reproduce on x86, which seems to always have CPU#0
online even during kdump, I'm not sure where this should be fixed. Hence
going for another approach : fine-grained affinity is for performance
and we don't really care about that during kdump. Simply revert to the
previous working behavior of ignoring affinity masks in this case only.

Fixes: 9ea69a55b3b9 ("powerpc/pseries: Pass MSI affinity to irq_create_mapping()")
Cc: lvivier@xxxxxxxxxx
Cc: stable@xxxxxxxxxxxxxxx
Reviewed-by: Laurent Vivier <lvivier@xxxxxxxxxx>
Reviewed-by: Cédric Le Goater <clg@xxxxxxxx>
Signed-off-by: Greg Kurz <groug@xxxxxxxx>
---

v2: - added missing #include <linux/crash_dump.h>

arch/powerpc/platforms/pseries/msi.c | 25 +++++++++++++++++++++++--
1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c
index b3ac2455faad..637300330507 100644
--- a/arch/powerpc/platforms/pseries/msi.c
+++ b/arch/powerpc/platforms/pseries/msi.c
@@ -4,6 +4,7 @@
* Copyright 2006-2007 Michael Ellerman, IBM Corp.
*/

+#include <linux/crash_dump.h>
#include <linux/device.h>
#include <linux/irq.h>
#include <linux/msi.h>
@@ -458,8 +459,28 @@ static int rtas_setup_msi_irqs(struct pci_dev *pdev, int nvec_in, int type)
return hwirq;
}

- virq = irq_create_mapping_affinity(NULL, hwirq,
- entry->affinity);
+ /*
+ * Depending on the number of online CPUs in the original
+ * kernel, it is likely for CPU #0 to be offline in a kdump
+ * kernel. The associated IRQs in the affinity mappings
+ * provided by irq_create_affinity_masks() are thus not
+ * started by irq_startup(), as per-design for managed IRQs.
+ * This can be a problem with multi-queue block devices driven
+ * by blk-mq : such a non-started IRQ is very likely paired
+ * with the single queue enforced by blk-mq during kdump (see
+ * blk_mq_alloc_tag_set()). This causes the device to remain
+ * silent and likely hangs the guest at some point.
+ *
+ * We don't really care for fine-grained affinity when doing
+ * kdump actually : simply ignore the pre-computed affinity
+ * masks in this case and let the default mask with all CPUs
+ * be used when creating the IRQ mappings.
+ */
+ if (is_kdump_kernel())
+ virq = irq_create_mapping(NULL, hwirq);
+ else
+ virq = irq_create_mapping_affinity(NULL, hwirq,
+ entry->affinity);

if (!virq) {
pr_debug("rtas_msi: Failed mapping hwirq %d\n", hwirq);
--
2.26.2