Hi Michał (again) > On Jun 18, 2019, at 12:59 PM, Gix, Brian <brian.gix@xxxxxxxxx> wrote: > > Hi Michał, > >> On Jun 18, 2019, at 4:27 AM, Michał Lowas-Rzechonek <michal.lowas-rzechonek@xxxxxxxxxxx> wrote: >> >> This patch enables us to implement more I/O layers, particularly non-HCI >> ones. >> >> As of Bluetooth 5.1, standard HCI commands don't allow precise control >> over mesh-related advertising, making this I/O layer very inefficient. >> Therefore, it is desirable to have an option to use a non-HCI transport >> to talk to radio transceiver, at least until Bluetooth extends HCI so >> that standard BLE Controllers achieve good performance. > > The intent if the architecture was to allow for finer control with future controllers, and thus a mesh-io.c which doesn’t assume anything about underlying controllers, and then an *initial* mesh-io-generic.c which is what gets used by 4.x controllers... as more controllers are supported with finer tuning available, then mesh-io-<bt5+>.c can be added to the abstracted list. We are also talking about how to share controllers between bluetoothd and meshd... one idea is using the MGMT interface to send and receive specific advertisements... that would necessitate a mesh-io-mgmt.c. However, it will also require support in the kernel. > >> >> Another use case would be a non-local radio: the mesh stack can run on a >> secure device, but due to physical constraints it might need to use >> radio transceiver located at a distance, connected e.g. via LAN. >> >> Michał Lowas-Rzechonek (1): >> mesh: Move HCI handling to mesh-io-generic >> >> mesh/main.c | 4 +- >> mesh/mesh-io-api.h | 3 +- >> mesh/mesh-io-generic.c | 203 ++++++++++++++++++++++++++++++++++++----- >> mesh/mesh-io.c | 17 ++-- >> mesh/mesh-io.h | 2 +- >> mesh/mesh.c | 189 +++----------------------------------- >> mesh/mesh.h | 4 +- >> 7 files changed, 209 insertions(+), 213 deletions(-) >> >> -- >> 2.19.1 >>
