linux_dsm_epyc7002/arch/powerpc
Greg Kurz a9c55f22a0 powerpc/pseries: Don't enforce MSI affinity with kdump
commit f9619d5e5174867536b7e558683bc4408eab833f upstream.

Depending on the number of online CPUs in the original kernel, it is
likely for CPU  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 9ea69a55b3 ("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: 9ea69a55b3 ("powerpc/pseries: Pass MSI affinity to irq_create_mapping()")
Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210215094506.1196119-1-groug@kaod.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-03-17 17:06:10 +01:00
..
boot powerpc/boot: Fix build of dts/fsl 2020-12-30 11:54:01 +01:00
configs powerpc updates for 5.10 2020-10-16 12:21:15 -07:00
crypto
include powerpc/kexec_file: fix FDT size estimation for kdump kernel 2021-03-04 11:38:39 +01:00
kernel powerpc/32s: Add missing call to kuep_lock on syscall entry 2021-03-04 11:38:40 +01:00
kexec powerpc/kexec_file: fix FDT size estimation for kdump kernel 2021-03-04 11:38:39 +01:00
kvm KVM: PPC: Make the VMX instruction emulation routines static 2021-03-04 11:38:01 +01:00
lib powerpc/sstep: Fix incorrect return from analyze_instr() 2021-03-07 12:34:16 +01:00
math-emu
mm powerpc/mm: Fix verification of MMU_FTR_TYPE_44x 2020-12-30 11:54:16 +01:00
net bpf, powerpc: Fix misuse of fallthrough in bpf_jit_comp() 2020-09-29 16:39:11 +02:00
oprofile
perf powerpc/perf: Fix handling of privilege level checks in perf interrupt context 2021-03-17 17:06:10 +01:00
platforms powerpc/pseries: Don't enforce MSI affinity with kdump 2021-03-17 17:06:10 +01:00
purgatory
sysdev powerpc: sysdev: add missing iounmap() on error in mpic_msgr_probe() 2021-01-06 14:56:53 +01:00
tools
xmon powerpc/xmon: Change printk() to pr_cont() 2020-12-30 11:54:16 +01:00
Kbuild
Kconfig powerpc/47x: Disable 256k page size 2021-03-04 11:38:01 +01:00
Kconfig.debug
Makefile Kbuild fixes for v5.10 (2nd) 2020-12-06 10:31:39 -08:00
Makefile.postlink