From: CJ Chen <cjchen@igel.co.jp>

The riscv_iommu_trap_ops MemoryRegionOps specifies that
unaligned accesses are not valid for this device but that
it does implement them. This doesn't make much sense, and
we want to add an assertion that registered MRs don't specify
this invalid combination of settings.

Drop .impl.unaligned = true, with no behaviour change.

Signed-off-by: CJ Chen <cjchen@igel.co.jp>
Acked-by: Tomoyuki Hirose <hrstmyk811m@gmail.com>
Reported-by: Tomoyuki Hirose <hrstmyk811m@gmail.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[PMM: reworded commit message]
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/r/20260428093339.2087081-2-peter.maydell@linaro.org
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 hw/riscv/riscv-iommu.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
index 7ba3240552..2b40ab2ce0 100644
--- a/hw/riscv/riscv-iommu.c
+++ b/hw/riscv/riscv-iommu.c
@@ -2434,7 +2434,6 @@ static const MemoryRegionOps riscv_iommu_trap_ops = {
     .impl = {
         .min_access_size = 4,
         .max_access_size = 8,
-        .unaligned = true,
     },
     .valid = {
         .min_access_size = 4,
-- 
2.53.0