04 Jan, 2012
1 commit
-
The DB8500 ED (Early Drop) and V1 are only available inside of
ST-Ericsson or partners, we have actively replaced and scrapped
these prototypes. All Nova products on the open market (such as
the Snowball board) are based on V2 and later ASIC variants.
So let us focus on supporting the silicon that will be used and
delete this to get a clear overview.Cc: Daniel Lezcano
Acked-by: Samuel Ortiz
Signed-off-by: Linus Walleij
Signed-off-by: Arnd Bergmann
24 Oct, 2011
13 commits
-
panic() is too heavy for this, indeed the PRCMU is critical for
the system but not to the point that we should stop everything,
if we can still get a prompt or so.Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This refactors the mfd/dbx500-prcmu drivers to use a switch()
statement rather than nested if/else-construction.Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This renames the PRCMU clock force initialization function
to have a less generic name.Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This rectifies the device name of the MCDE voltage domain
regulator consumer and adds a number of other consumers to
the voltage domains.Cc: Liam Girdwood
Cc: Mark Brown
Signed-off-by: Bengt Jonsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This implements the reset code retrieval function so we can
ipso facto get to know how the system was reset.Signed-off-by: Sebastian Rasmussen
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This is a tweak for the case where the modem goes to sleep while
emitting the AC_WAKE_ACK anyway. Also print the modem errors as
critical, since they jeopardize the entire platform when they
occur.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This implements the watchdog accessor functions for the DB8500
PRCMU, making it possible to implement the watchdog driver.Signed-off-by: Jonas Aberg
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This implements the accessor function for hardware accelerator
power state settings.Signed-off-by: Bengt Jonsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
This extends the DB8500 PRCMU driver with accessor calls for the
PRCMU PLL and SGA clocks.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
Now that we have a shared API between the DB8500 and DB5500
PRCMU's, switch to using this neutral API instead. We delete the
parts of db8500-prcmu.h that is now PRCMU-neutral, and calls will
be diverted to respective driver. Common registers are in
dbx500-prcmu-regs.h and common accessors and defines in
This way we get a a lot more
abstraction and code reuse.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
Some clocks may be force enabled when we probe the
driver, but they need to be turned off by default so
we have a known state. We call this the register
initialization function if we need more stuff in
there in the future.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
We have a few more mailboxes and fixed messages in the
DB8500 PRCMU, update to match the latest specification.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz -
Instead of carrying around the __PRCMU_BASE in every read or
write to the PRCMU registers, move it out to the register
definition file and define registers along with their base
offset so that the code gets easier to read.Signed-off-by: Mattias Nilsson
Signed-off-by: Linus Walleij
Signed-off-by: Samuel Ortiz
28 May, 2011
1 commit
-
With the addition of a device platform mfd_cell pointer, MFD drivers
can go back to passing platform data back to their sub drivers.
This allows for an mfd_cell->mfd_data removal and thus keep the
sub drivers MFD agnostic. This is mostly needed for non MFD aware
sub drivers.Signed-off-by: Mattias Wallin
Signed-off-by: Samuel Ortiz
25 May, 2011
3 commits
-
The DB8500 has ePOD:s (electronic power domains) which are possible
to switch on/off to deactivate silicon blocks on the DB8500 SoC
by cutting their power without retention. We model these as simple
regulators with one bit on/off settings.Acked-by: Liam Girdwood
Acked-by: Mark Brown
Signed-off-by: Bengt Jonsson
Signed-off-by: Sundar Iyer
Signed-off-by: Jonas Aberg
Signed-off-by: Virupax Sadashivpetimath
Signed-off-by: Martin Persson
Signed-off-by: Linus Walleij -
This updates the DB8500 PRCMU driver to the latest version
available internally. Nominally we would update the dependent
CPUfreq driver at the same time but since that is being moved
around in this patch set we postpone that by simply deactivating
it for the time being.This is a snapshot of the current PRCMU firmware API as it looks
right now. The PRCMU firmware is still subject to change. This
also updates the CPUfreq driver to a newer version that will
utilize the new API.Acked-by: Samuel Ortiz
Signed-off-by: Mattias Nilsson
Signed-off-by: Martin Persson
Signed-off-by: Per Fransson
Signed-off-by: Jonas Aaberg
Signed-off-by: Sebastien Rault
Signed-off-by: Bengt Jonsson
Signed-off-by: Rickard Andersson
Signed-off-by: Linus Walleij -
We have decided that this function arbiter fits better in the MFD
subsystem. Since we need to concatenate the split header files we move
it basically like this:mv mach-ux500/prcmu-db8500.c drivers/mfd/db8500-prcmu.c
mv mach-ux500/include/mach/prcmu-defs.h include/linux/mfd/db8500-prcmu.h
mv mach-ux500/include/mach/prcmu-regs.h drivers/mfd/db8500-prcmu-regs.h
mach-ux500/include/mach/prcmu.h >> include/linux/mfd/db8500-prcmu.h
rm arch/arm/mach-ux500/include/mach/prcmu.hThen we update different #include statements and Makefile orders etc
to make the PRCMU driver compile, link and boot in the new place
without really changing any code.Acked-by: Samuel Ortiz
Signed-off-by: Linus Walleij