By default, the DPAA2_USE_PHYS_IOVA setting with the make build-system
is "y", which is overridden to "n" in the config file specifically for
the arm64-dpaa2-linuxapp-gcc builds. The lack of this setting in meson
builds causes issues for 32-bit build, as the alternative code paths
have compiler warnings e.g. for i686 builds. Therefore we should align
the meson and make settings, setting the value to "true" by default and
overriding it to "false" for dpaa2-specific builds.
Fixes:
6ec78c2463ac ("build: add meson support for dpaaX platforms")
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
['RTE_CACHE_LINE_SIZE', 64],
['RTE_MAX_NUMA_NODES', 1],
['RTE_MAX_LCORE', 16],
- ['RTE_LIBRTE_DPAA2_USE_PHYS_IOVA', 'n']]
+ ['RTE_LIBRTE_DPAA2_USE_PHYS_IOVA', false]]
## Arm implementer ID (ARM DDI 0487C.a, Section G7.2.106, Page G7-5321)
impl_generic = ['Generic armv8', flags_generic, machine_args_generic]
# values which have defaults which may be overridden
dpdk_conf.set('RTE_MAX_VFIO_GROUPS', 64)
dpdk_conf.set('RTE_DRIVER_MEMPOOL_BUCKET_SIZE_KB', 64)
+dpdk_conf.set('RTE_LIBRTE_DPAA2_USE_PHYS_IOVA', true)
compile_time_cpuflags = []
if host_machine.cpu_family().startswith('x86')