Some common code uses more of the io.h funcs than we currently provide, so
pull in all of the ones from the linux kernel.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

These need to be rethought, but until that happens, isolate the hack so
that we can extend the common code without breaking things.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Give the CF/IDE code its own file to keep things cleanly separated.  While
we're here, clean up the code to use common functions.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Some SPI devices like to see high idle values rather than low.
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The single initcode function was growing unwieldy, so split it up the
distinct steps into their own function.  This should making digesting the
result much easier on people.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

We need the definition of bd_t in this header, so pull in asm/u-boot.h.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

While the initdram() function makes sense on some arches, it doesn't for
Blackfin systems as it's always implemented the same way.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The length of the sections is fixed at link time, so let the linker do the
calculation rather than doing it ourselves at runtime.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NValentin Yakovenkov <yakovenkov@niistt.ru>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Very little additional code overhead, and only works when the user sets an
env var ahead of time, so default to on makes sense.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The tinyboards like to run at a little lower voltage than the default, and
they prefer to boot over the network.  For the latter, extend the common
code a little to make this easier.

Also fix the cm-bf527 env sector size while we're in here to reflect the
flash that is actually in use.
Signed-off-by: NHarald Krapfenbauer <Harald.Krapfenbauer@bluetechnix.at>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Use the common config option for extracting the environment for embedding
into LDR files and clarify the LDR-specific option.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

For MUSB devices that do not support multipoint (hubs), we have to emulate
a root hub so that we can support core operations like resetting ports.
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Adds required definitions for EHCI support in sequoia configuration file.
But still keeps the OHCI as default driver.
Signed-off-by: NChris Zhang <chris@seamicro.com>

Currently ppc440epx uses OHCI for USB full-speed support. This change adds
support for EHCI.
Signed-off-by: NChris Zhang <chris@seamicro.com>

In USB ehci driver, the port reset is not terminated. EHCI spec says "A host
 controller must terminate the reset and stabilize the state of the port within
 2 milliseconds". Without termination, a port stays at reset state. This is
 observed on ppc4xx(sequoia) boards.
Signed-off-by: NChris Zhang <chris@seamicro.com>

This patch fixes this warning during compile:

omap3.c: In function 'musb_platform_init':
omap3.c:126: warning: label 'end' defined but not used

Problem reported by: Dirk Behme[dirk.behme@googlemail.com]
Signed-off-by: NSanjeev Premi <premi@ti.com>

Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NRemy Bohmer <linux@bohmer.net>

The RX Control/Status register has bits that we want to preserve, so don't
just write out a single bit.  Preserve the others bits in the process.

The original code posted to the u-boot list had this behavior, but looks
like it was lost somewhere along the way to merging.
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NRemy Bohmer <linux@bohmer.net>

The multipoint handling under MUSB is optional, and some parts (like the
Blackfin processor) do not implement support for it.
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NRemy Bohmer <linux@bohmer.net>

The dynamic FIFO handling under MUSB is optional, and some parts (like
the Blackfin processor) do not implement support for it.

Due to this, the FIFO reading/writing steps need special handling, so
mark the common versions weak so drivers can override.
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NRemy Bohmer <linux@bohmer.net>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NRemy Bohmer <linux@bohmer.net>

Adding USB configuration. Default is set for USB MSC host.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NSwaminathan S <swami.iyer@ti.com>

Tested USB host functionality on DA830 EVM.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NSwaminathan S <swami.iyer@ti.com>

Added DA8xx GPIO base addresses in gpio_defs.h and pointers
to different BANKs which can be used to program GPIOs.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NSwaminathan S <swami.iyer@ti.com>

USB0_DRVVBUS pinmux configuration is required for USB functinality
in uboot.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NSwaminathan S <swami.iyer@ti.com>

Add support to configure bus parking mode and master in bus arbitration
configuration (ACR). Add this for the kmeter1 port:

Configure bus arbiter with recommended values from Freescale
to improve bus latency/throughput for application with
intensive QuiccEngine activity.
Signed-off-by: NHeiko Schocher <hs@denx.de>
Signed-off-by: NKim Phillips <kim.phillips@freescale.com>