-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Somebody in the thread at some point said: | Andy Green wrote: |> This is yet another case for an always-on MPU to solve it in future designs. | | This one is actually a little bit trickier: even if we had that | MPU, it could do nothing to change what the PMU is doing, since the | PMU doesn't even get out of reset. From section 8.17.11 of the PMU | manual: | | | The I2C-bus module is only enabled in the Active state when the | | reset for the host controller is released
You are right, but it doesn't stop us. We need to add a P-Channel MOSFET on at least AUTO controlled by MPU to be able to remove the rail from its load. ~ - MPU disconnects high power loads ~ - MPU brings up PMU fully ~ - MPU waits for nRSTHC from PMU is high, writes low "always active" ~ current limits to PMU by I2C ~ - connect loads in turn according to power sequencing demands ~ - MPU polls until all power rails to be OK ~ - MPU brings CPU out of reset using MPU GPIO - -Andy -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkh3SbcACgkQOjLpvpq7dMqJVACdHbwHvl+dowlBfZZbzmU62AmF aGQAnA+a6lyjfioCZN3wwl/S/25hXtkP =eepz -----END PGP SIGNATURE-----
