Skip to content

L
linux-davinci
Switch branch/tag
  1. 18 Sep, 2006 2 commits
  3. 14 Sep, 2006 3 commits
    • David Brownell's avatar
      MUSB: More TUSB OTG support · e3071039
      David Brownell authored 
      Add more host/OTG functionality.

New Features:
   -	Cable based role switching (using the ID pin) mostly works.  This gives
	much of the end-user value for USB OTG; there's also dynamic role
	switching (using special HNP protocol with another OTG device), which
	surely won't work yet.
   -	Root can write 0/N to /sys/module/musb_hdrc/parameters/musb_otg to
   	disable OTG functionality, forcing B-Default mode (with optional SRP
	support, but without 8mA current limits) when it's set.  Change this
	before the gadget driver initializes (e.g. before "modprobe").

Bugfixes:
   -	Fix many OTG mode startup glitches, notably for the root hub.
   -	Add workaround for "issue 4".

Cleanups:
   -	Remove more mode-specific #ifdeffery
   -	ReMoveSomeCamelCasing
   -	Quieter startup
   -	Quash another Kconfig rebellion
   -	Add missing bit declaration
   -	Tersify TUSB irq messages

Open Issues:
   -	Nothing passes remote wakeup down to root hub (unchanged)
   -	Strange VBUS errors (unchanged; workaround: use hub modified for
	N770, providing 5V supply till TUSB board does so reliably);
   -	Plug in/out as peripheral many times; fine.  As host (with modified
	hub), ditto.  Now peripheral again ... fails.
Signed-off-by: default avatarDavid Brownell <dbrownell@users.sourceforge.net>
      e3071039
    • David Brownell's avatar
      MUSB: TUSB OTG enumeration support · 89c57744
      David Brownell authored 
      This gives TUSB host side enumeration support for low speed devices (like
USB mice), through a Mini-A connector that's not removed ... basically, a
non-OTG configuration, with the ID pin always grounded.

Basically it punts the "turn VBUS power on/off" to board-specific logic,
and implements it for TUSB using software switching (rather than having
the controller do it).

There are various issues, notably

    (a)	remote wakeup not getting passed to the root hub then down
	to the device (mouse) that issued the wakeup;

    (b)	strange "vbus error" reports on device connection if nothing
    	is hooked up at driver initialization; and

    (c) full or high speed devices see spurious disconnect events
	right after they've been reset (hw bug?);

Includes various small cleanups too, notably starting to obey the OTG
state machine and use the OTG timer.
Signed-off-by: default avatarDavid Brownell <dbrownell@users.sourceforge.net>
      89c57744
    • Imre Deak's avatar
      ARM: OMAP2: fb: clock refcounting for dynamic sleep mode · 829702a4
      Imre Deak authored 
      Hold reference counts on related clocks only for the duration when they
are needed, so that dynamic sleep logic can enter the most ideal power
state.
Signed-off-by: default avatarImre Deak <imre.deak@solidboot.com>
Signed-off-by: default avatarJuha Yrjola <juha.yrjola@solidboot.com>
      829702a4
  5. 13 Sep, 2006 9 commits
  7. 12 Sep, 2006 12 commits
  9. 11 Sep, 2006 14 commits