ARM: mvebu: fix HW I/O coherency related deadlocks
authorThomas Petazzoni <thomas.petazzoni@free-electrons.com>
Thu, 16 Jun 2016 13:42:25 +0000 (15:42 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 27 Jul 2016 16:47:39 +0000 (09:47 -0700)
commite9a2ce2a68381f0d638cadf1e4070f73a7df43c0
tree0b4128dae3b5abba0b7f631cd5d03d87c502ac09
parente4ad472e64c43a34a17c2636978d017b7dd8569c
ARM: mvebu: fix HW I/O coherency related deadlocks

commit c5379ba8fccd99d5f99632c789f0393d84a57805 upstream.

Until now, our understanding for HW I/O coherency to work on the
Cortex-A9 based Marvell SoC was that only the PCIe regions should be
mapped strongly-ordered. However, we were still encountering some
deadlocks, especially when testing the CESA crypto engine. After
checking with the HW designers, it was concluded that all the MMIO
registers should be mapped as strongly ordered for the HW I/O coherency
mechanism to work properly.

This fixes some easy to reproduce deadlocks with the CESA crypto engine
driver (dmcrypt on a sufficiently large disk partition).

Tested-by: Terry Stockert <stockert@inkblotadmirer.me>
Tested-by: Romain Perier <romain.perier@free-electrons.com>
Cc: Terry Stockert <stockert@inkblotadmirer.me>
Cc: Romain Perier <romain.perier@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/arm/mach-mvebu/coherency.c