>From patch 3/3:
The Socionext Synquacer SoC's implementation of GICv3 has a so-called
'pre-ITS', which maps 32-bit writes targeted at a separate window of
size '4 << device_id_bits' onto writes to GITS_TRANSLATER with device
ID taken from bits [device_id_bits + 1:2] of the window offset.
Writes that target GITS_TRANSLATER directly are reported as originating
from device ID #0.
So add a workaround for this. Given that this breaks isolation, clear
the IRQ_DOMAIN_FLAG_MSI_REMAP flag as well.
v4: - add patch to allow the quirk hook to signal whether it was applied or
not; this is necessary for hardware whose IIDR is not sufficient to
identify it
- use DT sub-node rather than property to describe the quirk
- remove pre_its_size field, and use get_msi_base() function pointer
instead
v3: - add patch to pull device ID space discovery forward, so we can quirk
it as well (as we already do for Cavium)
- use existing quirks framework as much as possible
- get rid of ITS_WORKAROUND_xxx flag: it is no longer needed after the
refactoring
v2: - use a 32-bit host address/size rather than a PCI address, to factor
out the involvement of an SMMU (which the platform does have, but it
is unclear atm if it can be exposed to the OS)
- add msi_domain_flags member to move the quirk flag checks out of the
common code path
Ard Biesheuvel (3):
drivers/irqchip: gicv3: probe device ID space before quirks handling
drivers/irqchip: gic: make quirks matching conditional on init return
value
drivers/irqchip: gicv3: add workaround for Synquacer pre-ITS
Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt | 4 +
arch/arm64/Kconfig | 8 ++
drivers/irqchip/irq-gic-common.c | 5 +-
drivers/irqchip/irq-gic-common.h | 2 +-
drivers/irqchip/irq-gic-v3-its.c | 113 ++++++++++++++++----
5 files changed, 110 insertions(+), 22 deletions(-)
--
2.11.0
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
