ot_usbdev: add initial host protocol and bus handling
#238
Conversation
pamaury
requested review from
AlexJones0,
jwnrt and
rivos-eblot
and removed request for
rivos-eblot
pamaury
force-pushed
the
usbdev2
branch
4 times, most recently
from
4db9cec to
4dc34ad
Compare
| ot_usbdev_server_write_packet(s, OT_USBDEV_SERVER_CMD_HELLO, | ||
| server->recv_pkt.id, &resp_hello, | ||
| sizeof(resp_hello)); |
There was a problem hiding this comment.
Can you not just send back the same hello packet now that you have checked its integrity, or am I missing something?
There was a problem hiding this comment.
You could but that feels like premature optimization, if the hello packet changed, it could become asymmetrical (e.g. different server and client versions).
pamaury
force-pushed
the
usbdev2
branch
3 times, most recently
from
5fcbdcf to
bdc6ba3
Compare
|
Once this passes CI, can this be merged or are there any big blockers left? The code will still be modified quite a lot to add support for data transfers but I am relunctant to base too much work on this PR until it's merged. |
There are a couple of remaining comments not closed (in the GitHub "hidden" conversation sections), and a couple of remaining details, mostly:
|
pamaury
force-pushed
the
usbdev2
branch
2 times, most recently
from
bcdf022 to
57d9ec5
Compare
|
I hope I caught everything this time. I may have missed some |
This commit creates a new chardev for the USBDEV block over which the driver can communicate with a virtual USB host. The protocol (still under development) is documented and implemented. At the moment, only bus events (vbus handling, connection, reset, stub suspend/resume) are implemented. Signed-off-by: Amaury Pouly <amaury.pouly@lowrisc.org>
4 participants
This PR creates a new chardev for the USBDEV block over which the driver can communicate with a virtual USB host. The protocol (still under development) is documented and implemented. At the moment, only bus events (vbus handling, connection, reset, stub suspend/resume) are implemented.
In order to test this implementation, a virtual USB host speaking the protocol must be used. I am currently testing this using a custom python script but I think it is too hacky to upstream at the moment.
See lowRISC/opentitan#28487 for the OpenTitan side