USB and suspend. Will it still charge? / 100mA on dead ports?

[Andy Green]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:

| Normal PC doesn't "power down USB port" I think you'll find.  The port
| has to be powered to detect next device insertion.  You can confirm with
| one of these USB reading lamps for example that have no smarts inside.
|
|> what about being in 500ma mode? ie enough power to charge? windows i
know will
|> not go above 100ma unless u have a driver for the device... or i'm pretty
|> sure... ?

Yeah this is all the undefined land of the power spec that doesn't
exist.  At least it makes content for the power spec for future products :-)

Typically in PCs the only enforcement of current limiting is a hardware
one at something above 500mA per port, there is no hardware "100mA mode"
normally that I saw.  It's an accounting issue for the global power
supply in the PC about 100mA or 500mA, managed by the PC USB driver
considering all ports, not really an issue for one port.

In the case it was negotiated and happy for the device to have 500mA, if
it did logically downgrade that to 100mA because the device went to
suspend (dunno it is true), it only makes a problem if the user then
plugs in a high current device in another port.  But seems you can be
right: that may happen and we don't address it.  If we do address it, it
means charging while off or in suspend will give very bad performance
from a 100mA limit.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkgO6c8ACgkQOjLpvpq7dMrTUACgkk723CBvXPU8/dX+Ar5naRHJ
17YAn1AIGit2qFmUwVw+gv39lfCfHIXj
=gYY7
-----END PGP SIGNATURE-----