On Tue, 24 Jan 2017 11:02:10 +0200, Felipe Balbi <felipe.balbi@xxxxxxxxxxxxxxx> wrote: > yeah, order matters. That's documented for f_fs. The order in the EP0 write call yes. The mangles-HS-wMaxPacketSize-if-no-LS/FS-given is not (or I could not find it). > Well, we can't easily change the way it works because it's an ABI to > userspace. I would have very much preferred for us to pass descriptors > using ioctl(). That way we could have: > > ioctl(fd, FFS_IOCTL_SS_EP_DESC, ep1_ss_desc); > ioctl(fd, FFS_IOCTL_SS_EP_COMP_DESC, ep1_ss_comp_desc); > > and so on. If you mean the EP0 writes, indeed it's userspace ABI and I agree a more structured/fine-grained access method would be nicer (current error reporting is terrible... too many way to trigger the same errno). But I was actually thinking of the epautoconf API. Is it also exposed to userspace ? Actually, thinking a bit more about it I think epautoconf should only alter wMaxPacketSize when caller is probing the max supported value, and in such case provide the minimum between speed-dependent limit and controller limit. And I would remove the hardcoded 64B limit on non-SS bulk descriptors. Regards, -- Vincent Pelletier
Attachment:
pgpcAFNX8G5yE.pgp
Description: Signature digitale OpenPGP
