iommu/vt-d: Skip TE disabling on quirky gfx dedicated iommu
authorLu Baolu <baolu.lu@linux.intel.com>
Thu, 23 Jul 2020 01:34:37 +0000 (09:34 +0800)
committerJoerg Roedel <jroedel@suse.de>
Fri, 24 Jul 2020 12:33:39 +0000 (14:33 +0200)
commitb1012ca8dc4f9b1a1fe8e2cb1590dd6d43ea3849
tree22c29fd43a23c036aa3b73f9b298bd717d375c2d
parent02f3effddfd04f3f08a24d23a82d1c1c6d89b777
iommu/vt-d: Skip TE disabling on quirky gfx dedicated iommu

The VT-d spec requires (10.4.4 Global Command Register, TE field) that:

Hardware implementations supporting DMA draining must drain any in-flight
DMA read/write requests queued within the Root-Complex before completing
the translation enable command and reflecting the status of the command
through the TES field in the Global Status register.

Unfortunately, some integrated graphic devices fail to do so after some
kind of power state transition. As the result, the system might stuck in
iommu_disable_translation(), waiting for the completion of TE transition.

This provides a quirk list for those devices and skips TE disabling if
the qurik hits.

Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=208363
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=206571
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Tested-by: Koba Ko <koba.ko@canonical.com>
Tested-by: Jun Miao <jun.miao@windriver.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200723013437.2268-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
drivers/iommu/intel/dmar.c
drivers/iommu/intel/iommu.c
include/linux/dmar.h
include/linux/intel-iommu.h