-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 1 Nov 2010 07:06:44 -0500 "Olin Lathrop" wrote: [snip] > You seem to be over thinking this. Is there any reason to believe > that setting STALL doesn't have the overall desired effect? If you > get a packet with the wrong toggle, then something else is already > wrong. Does it really matter which the SIE sends back for that? > Presumably the next packet will have correct toggle and stall will be > sent back. Well, in the interests of bullet-proof development, I thought that if DTSEN takes precedence over BSTALL, you could set DTSEN=3D1, DTS=3D0, BSTALL=3D1 on an OUT BD to handle the status stage of a control transfer with an inbound data stage. Since the status stage should always consist of a zero-length OUT transaction with a DATA1 packet in it, this would mean a legitimate status stage would mismatch DTS and get ACKed, whereas a host refusing to follow the rules and sending a DATA0 instead would match DTS and get STALLed. I realize that the USB specification mostly says that a device can do anything it wants if the host screws up, so I could just accept either packet type, but I guess coming from a full-size-computer environment I consider anything on the far end of a communications link to be "untrusted", and tend to carefully verify whatever I can. It's not really a big deal, I just read the relevant pages of the datasheet, noticed the undocumented situation, and thought of a possible use for it. Chris -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (GNU/Linux) Comment: GnuPT 2.7.2 iEYEARECAAYFAkzPqGgACgkQXUF6hOTGP7dzuwCffWFA7qeKj1JBSadZmZuJ6/jp HN8AnjVuiL+5wjQ4YEiu20BcbMiTxINd =3DElGs -----END PGP SIGNATURE----- --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .