ACPI: APEI: fix synchronous external aborts in user-mode
authorXiaofei Tan <tanxiaofei@huawei.com>
Fri, 11 Jun 2021 12:37:07 +0000 (20:37 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 14 Jul 2021 14:56:12 +0000 (16:56 +0200)
commite0afab5181d0801b270233af18f2db88d75824f7
tree736437bc4e91ecc249522d6097ed36058e6c787d
parentf626452df805bb4e69501bd9eac6a6c9866e4e2f
ACPI: APEI: fix synchronous external aborts in user-mode

[ Upstream commit ccb5ecdc2ddeaff744ee075b54cdff8a689e8fa7 ]

Before commit 8fcc4ae6faf8 ("arm64: acpi: Make apei_claim_sea()
synchronise with APEI's irq work"), do_sea() would unconditionally
signal the affected task from the arch code. Since that change,
the GHES driver sends the signals.

This exposes a problem as errors the GHES driver doesn't understand
or doesn't handle effectively are silently ignored. It will cause
the errors get taken again, and circulate endlessly. User-space task
get stuck in this loop.

Existing firmware on Kunpeng9xx systems reports cache errors with the
'ARM Processor Error' CPER records.

Do memory failure handling for ARM Processor Error Section just like
for Memory Error Section.

Fixes: 8fcc4ae6faf8 ("arm64: acpi: Make apei_claim_sea() synchronise with APEI's irq work")
Signed-off-by: Xiaofei Tan <tanxiaofei@huawei.com>
Reviewed-by: James Morse <james.morse@arm.com>
[ rjw: Subject edit ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/acpi/apei/ghes.c