mtd: spi-nor: add support for Macronix Octal flash
Follow patch <d9fe0810fca04a378b3d3fe14d33bdc22fbc1bfa> (Allow using Micron mt35xu512aba
in Octal DTR mode).
Enable Octal DTR mode with 20 dummy cycles to allow running at the
maximum supported frequency for adding Macronix flash in Octal DTR mode.
-https://www.mxic.com.tw/Lists/Datasheet/Attachments/7841/MX25LM51245G,%203V,%20512Mb,%20v1.1.pdf
Signed-off-by: JaimeLiao <jaimeliao.tw@gmail.com> Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tom Rini [Thu, 14 Jul 2022 22:43:51 +0000 (18:43 -0400)]
Merge branch '2022-07-14-migrate-wiki-to-sphinx'
- Merge the majority of the relevant wiki content to doc/process/ and
convert to Sphinx. Begin cleaning up and modernizing the content as
well to match current process. There is still more work to be done in
this regard.
Tom Rini [Tue, 12 Jul 2022 18:14:19 +0000 (14:14 -0400)]
doc: Add in the historical release statistics found on the wiki
The wiki had gitdm-generated release statistics starting with v1.3.0.
Re-generate this information as Sphinx. This aims to be as historically
accurate as possible and so some company renames were kept to their old
rather than current name until we had made the switch previously.
Tom Rini [Tue, 12 Jul 2022 21:34:15 +0000 (17:34 -0400)]
doc: process: Correct and expand slightly on the Merge Window concept
For quite a long time we've been using a 3 week, rather than 2 week,
merge window as it was only 2 weeks during the timeframe where we did 2
month rather than 3 month releases. This corrects the places that still
had 2 weeks and tries to make things a bit clearer overall.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Tue, 12 Jul 2022 21:34:14 +0000 (17:34 -0400)]
doc: Add doc/develop/release_cycle.rst
Migrate the RelaseCycle wiki page to Sphinx. In terms of visible
changes, we stop having a dynamic countdown to when the release is. And
we drop the year-based statistics, that were not being kept up to date.
For the moment, we only link to statistics for v2022.07 but will add
back the historical data in a subsequent patch.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Thu, 14 Jul 2022 12:07:46 +0000 (08:07 -0400)]
process.rst: Modernize the "Workflow of a Custodian" section
The "Workflow of a Custodian" section on the wiki had not been changed
in quite some time to reflect how the process has been functioning for
some time. First, update some links to point to modern and current
sources of information.
Second, and more overarching, reword much of the section. This expands
on the expectations of both custodians and developers when it comes to
rebasing patches. Rework the final points to be clearer that Custodians
are expected to do their best to test the changes and ask for help when
needed, as well as that pull requests are expected in a timely manner.
Cc: Claudius Heine <ch@denx.de> Cc: Martin Bonner <martingreybeard@gmail.com> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Thu, 14 Jul 2022 12:07:45 +0000 (08:07 -0400)]
process.rst: Perform minor cleanups
- Use gender-neutral language to refer to the user, consistently.
- Reword a few places so that they read more naturally.
- Make the long standing practice around "Twilight Time" more clear,
hopefully.
- Replace a reference to MAKEALL with a reference to CI testing as
that's the current requirement.
Cc: Claudius Heine <ch@denx.de> Cc: Martin Bonner <martingreybeard@gmail.com> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Thu, 14 Jul 2022 12:07:43 +0000 (08:07 -0400)]
doc: Migrate Process wiki page to Sphinx
Move the current Process wiki page to doc/develop/process.rst. The
changes here are for formatting or slight rewording so that it reads
well when linking to other Sphinx documents.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Thu, 14 Jul 2022 12:07:41 +0000 (08:07 -0400)]
doc: Migrate DesignPrinciples wiki page to Sphinx
Move the current DesignPrinciples wiki page to
doc/develop/designprinciples.rst. The changes here are for formatting
or slight rewording so that it reads well when linking to other Sphinx
documents.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Thu, 14 Jul 2022 12:07:40 +0000 (08:07 -0400)]
doc: Migrate CodingStyle wiki page to Sphinx
Move the current CodingStyle wiki page to doc/develop/codingstyle.rst.
The changes here are for formatting or slight rewording so that it reads
well when linking to other Sphinx documents.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Weijie Gao [Fri, 20 May 2022 03:24:10 +0000 (11:24 +0800)]
tools: mtk_image: add support for MT7621 NAND images
The BootROM of MT7621 requires a image header for SPL to record its size
and load address when booting from NAND.
To create such an image, one can use the following command line:
mkimage -T mtk_image -a 0x80200000 -e 0x80200000 -n "mt7621=1"
-d u-boot-spl-ddr.bin u-boot-spl-ddr.img
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:24:04 +0000 (11:24 +0800)]
spl: nand: support loading legacy image with payload compressed
Add support to load legacy image with payload compressed. This redirects
the boot flow for all legacy images. If the payload is not compressed, the
actual behavior will remain unchanged.
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:23:58 +0000 (11:23 +0800)]
spl: spl_legacy: fix the use of SPL_COPY_PAYLOAD_ONLY
If the payload is compressed, SPL_COPY_PAYLOAD_ONLY should always be set
since the payload will not be directly read to its load address. The
payload will first be read to a temporary buffer, and then be decompressed
to its load address, without image header.
If the payload is not compressed, and SPL_COPY_PAYLOAD_ONLY is set, image
header should be skipped on loading. Otherwise image header should also be
read to its load address.
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:23:47 +0000 (11:23 +0800)]
nand: raw: add support for MediaTek MT7621 SoC
This patch adds NAND flash controller driver for MediaTek MT7621 SoC.
The NAND flash controller of MT7621 supports only SLC NAND flashes.
It supports 4~12 bits correction with maximum 4KB page size.
Weijie Gao [Fri, 20 May 2022 03:23:37 +0000 (11:23 +0800)]
net: mediatek: use regmap api to modify ethsys registers
The address returned by regmap_get_range() is not remapped. Directly r/w
to this address is ok for ARM platforms since it's idential to the virtual
address.
But for MIPS platform only virtual address should be used for access.
To solve this issue, the regmap api regmap_read/regmap_write should be used
since they will remap address before accessing.
Reviewed-by: Ramon Fried <rfried.dev@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:23:31 +0000 (11:23 +0800)]
net: mediatek: remap iobase address
The iobase address from dts node is actually physical address. It's
identical to the virtual address in ARM platform. This is ok because this
driver was used only by ARM platforms (mt7622/mt7623 ...).
But now this driver will be used by mt7621 which is a MIPS SoC. For MIPS
platform the physical address space is mapped to KSEG0 and KSEG1 and this
makes the virtual address apparently not idential to its physical address.
To solve this issue, this patch replaces dev_read_addr with dev_remap_addr
to get the remapped iobase address.
Reviewed-by: Ramon Fried <rfried.dev@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:22:36 +0000 (11:22 +0800)]
clk: mtmips: add clock driver for MediaTek MT7621 SoC
This patch adds a clock driver for MediaTek MT7621 SoC.
This driver provides clock gate control as well as getting clock frequency
for CPU/SYS/XTAL and some peripherals.
Reviewed-by: Sean Anderson <seanga2@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:22:31 +0000 (11:22 +0800)]
doc: mediatek: add documentation for mt7621 reference boards
The MT7621 requires external binary blob being executed during u-boot's
boot-up flow. It's necessary to provide a guide here for users to correctly
build the u-boot.
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:22:26 +0000 (11:22 +0800)]
mips: mtmips: add two reference boards for mt7621
The mt7621_rfb board supports integrated giga PHYs plus one external
giga PHYs. It also has up to 512MiB DDR3, 16MB SPI-NOR, 3 mini PCI-e x1
slots, SDXC and USB.
The mt7621_nand_rfb board is almost the same as mt7621_rfb board, but it
uses NAND flash and SDXC is not available.
Reviewed-by: Stefan Roese <sr@denx.de> Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:21:51 +0000 (11:21 +0800)]
mips: add support for noncached_alloc()
This patch adds support for noncached_alloc() which was only supported by
ARM platform.
Unlike the ARM platform, MMU is not used in u-boot for MIPS. Instead, KSEG
is provided to access uncached memory. So most code of this patch is copied
from cache.c of ARM platform, with only two differences:
1. MMU is untouched in noncached_set_region()
2. Address returned by noncached_alloc() is converted using KSEG1ADDR()
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Weijie Gao [Fri, 20 May 2022 03:21:45 +0000 (11:21 +0800)]
mips: add __image_copy_len for SPL linker script
This patch adds __image_copy_len needed by TPL of MT7621 SoC.
The __image_copy_len represents the binary blob size of both SPL/TPL
binaries. To achieve this, __text_start/end are added for calculation.
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Tom Rini [Tue, 12 Jul 2022 21:34:13 +0000 (17:34 -0400)]
doc: Add statistics page for v2022.07
Our statistics pages have always been generated by gitdm. After
patching gitdm to generate an acceptable Sphinx output for tables,
include that and some other basic formatting here.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Sean Anderson [Sat, 25 Jun 2022 17:12:20 +0000 (13:12 -0400)]
doc: mkimage: Further document -o and -R
Despite the original description of these options, they are not always
image names, or even files. Some image types use these options to convey
configuration directly. Re-document these options as configuration options.
Additionally, add a new section documenting the format of the configuration
for each image type which uses it. In general, if configuration is used
directly (without a separate file) I have added documentation for it. If
the configuration points to a separate file, I have referenced that file's
documentation. Where there is no such documentation, I have added it.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:19 +0000 (13:12 -0400)]
mkimage: Add long options
The mkimage command has had many options added over the years.
Unfortunately, we are starting to run out of short options. Recent options
don't have any obvious relation to their meaning (e.g. -o/-g). Fortunately,
long options exist. Add long options for each current short option.
For the curious, the remaining short options are HIkLmMPQSuUwWXyYzZ.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:18 +0000 (13:12 -0400)]
doc: mkimage: Remove AUTHORS section
Per man-pages(7), "use of an AUTHORS section is strongly discouraged."
Remove it, and instead add some copyright notices and an SPDX. The default
license for U-Boot is GPL2, so that's what I put. The copyright dates are
based on the commit dates.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:15 +0000 (13:12 -0400)]
doc: mkimage: Edit options for style and consistency
This makes a variety of changes for the options to make them
typographically consistent, clarify their meaning, and fix grammatical (or
other) errors. Many of the changes here are stylistic, though there are a
few fixes. The main changes I made across the board were:
- All options are bolded and parameters italicised
- All single quotes are properly matched (instead of using apostrophes)
- Minor background info has been added to clarify many underdocumented
options
- Default values for options are documented
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:13 +0000 (13:12 -0400)]
doc: mkimage: Rearrange/remove some options
This moves some options which work in any mode to the general options
section. -p is moved to after -E/-B since those options are related. This
also adds documentation for -h and -V.
The -F, -l, and -G options are documented twice. Remove the second
documentation in each case. The synopsis for -l also suggests an implied
second uimage-file-name parameter. E.g.
mkimage [-l uimage-file-name] uimage-file-name
This is misleading, so remove it. Wrap a few lines to 80 characters as
well.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:12 +0000 (13:12 -0400)]
doc: mkimage: Use subsection macro
The options are divided up into several subsections. Use the appropriate
macro. While we're at it, rename the headings to better reflect the
contents of their sections.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:11 +0000 (13:12 -0400)]
doc: mkimage: Regularize option documentation
Square brackets are commonly used to denote optional parts of a command.
However, all option arguments are mandatory. Remove these brackets. This
also removes some unnecessary quotation marks, and uses hyphens to connect
words in option arguments. This is intended to just clean up the
formatting, leaving content corrections to later patches.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:10 +0000 (13:12 -0400)]
doc: mkimage: Reformat examples
This puts each example in a new paragraph and uses a hanging indent for
continued lines to increase clarity. We use tabs instead of .in or .RS for
the indent because it renders properly in both man and mandoc (which is
what many common HTML man pages use). The only nit is that the tab stops in
man default to something like 2", so reduce that to 1". We also escape
every "minus" as recommended by man-pages(7).
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:09 +0000 (13:12 -0400)]
doc: mkimage: Use empty request instead of blank lines
Blank lines do not have well-defined semantics in fill mode (the default).
Instead, use empty requests (.) where vertical space is necessary for
readability. There are a few places where we use a paragraph instead.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Sean Anderson [Sat, 25 Jun 2022 17:12:08 +0000 (13:12 -0400)]
doc: mkimage: Use standard style for synopsis
The synopsis section is a bit messy. As an example, "uimage file name" is
printed in italics, bold, and roman (depending on the line). This cleans
things up and converts the synopsis section to use standard style. The
.SY/.YS macros set up appropriate formatting for command synopsis sections
(such as disabling hyphenation and setting a hanging indent). All parts of
the synopsis now use the following style:
- Bold for parts of the command which should be typed in by the user (such
as the program name and flags)
- Italic for parts which should be replaced (such as uimage-file-name)
- Roman for parts which should not be typed at all (such as brackets)
Multi-word variables now use hyphens to connect their words instead of
spaces. This makes it clearer that all the words are part of the same
variable. Additionally, "option ..." is used to denote where other options
may be specified, as this appears to be standard style.
In addition to the above style changes, this also makes some changes to
content. The use of the term "legacy" has been removed, since this simply
refers to any non-FIT image type. Additionally, wording like "uimage file
name" has been replaced with "image-file-name" to better reflect that
images may or may not be uImages. Lastly, the "auto" value for -f is
documented in the synopsis.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Wed, 13 Jul 2022 12:09:20 +0000 (08:09 -0400)]
Merge tag 'u-boot-stm32-20220712' of https://source.denx.de/u-boot/custodians/u-boot-stm
- Alignment with Linux kernel device tree v5.19 for stm32mp15 and stm32mp13
- Add OP-TEE nodes for stm32mp13x, alligned with upstreamed OP-TEE
- Introduce of_to_plat ops in stm32_sdmmc2 driver
- Activate more features in stm32mp13 defconfig and support of STM32MP13x Rev.Y
- Drop fastboot and stm32prog trigger gpios on STM32MP15x DHCOM board
Kory Maincent [Wed, 22 Jun 2022 08:59:31 +0000 (10:59 +0200)]
usb: kbd: allow probing even if usbkbd not in stdin
For now the driver does not probe if usbkbd was not present in stdin.
This presents two issues, we can not probe the driver before setting stdin
and we can not use this driver in other manner than stdin console.
This patch fixes this by adding an else statement. It simply probes the
driver without console management in the case "usbkbd" is not in stdin.
Rui Miguel Silva [Wed, 29 Jun 2022 10:06:16 +0000 (11:06 +0100)]
corstone1000: enable isp1763 usb controller and mmc
MPS3 board have a ISP1763 usb controller, enable it to be used
for mass storage access for example. Enable the usb command
also and for the FVP support for mass storage enable the mmc
command.
Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Rui Miguel Silva [Wed, 29 Jun 2022 10:06:15 +0000 (11:06 +0100)]
usb: add isp1760 family driver
ISP1760/61/63 are a family of usb controllers, here the main
goal is to support the ISP1763 hcd part found in the MPS3 FPGA
board form Arm. This is based on the kernel driver and ported
to u-boot.
Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Where it is visible that there is no gpio.h in platform headers:
$ ls arch/arm/mach-keystone/include/mach/
clock_defs.h clock-k2e.h clock-k2hk.h ddr3.h hardware-k2e.h
hardware-k2hk.h i2c_defs.h mon.h mux-k2g.h xhci-keystone.h
clock.h clock-k2g.h clock-k2l.h hardware.h hardware-k2g.h
hardware-k2l.h mmc_host_def.h msmc.h psc_defs.h
Signed-off-by: Michal Simek <michal.simek@amd.com> Reviewed-by: Tom Rini <trini@konsulko.com>
Johann Neuhauser [Thu, 30 Jun 2022 14:19:48 +0000 (16:19 +0200)]
arm: dts: stm32mp1: Drop fastboot and stm32prog trigger gpios on DHCOM
PA13 and PA14 are used for USB power control and can't be used
to enforce fastboot or stm32prog mode by pressing a button.
Defining CONFIG_FASTBOOT/CONFIG_CMD_STM32PROG without this patch applied
results in fastboot/stm32prog always starting, because PA13/PA14 are always
low during boot. So drop the wrong trigger gpios definitions.
Signed-off-by: Johann Neuhauser <jneuhauser@dh-electronics.com> Reviewed-by: Marek Vasut <marex@denx.de> Reviewed-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
- ARM: dts: stm32: Add alternate pinmux for ethernet0 pins
- ARM: dts: stm32: Add alternate pinmux for mco2 pins
- ARM: dts: stm32: fix pinctrl node name warnings (MPU soc)
- ARM: dts: stm32: stm32mp15-pinctrl: add spi1-1 pinmux group
- dt-bindings: clock: add IDs for SCMI clocks on stm32mp15
- dt-bindings: reset: add IDs for SCMI reset domains on stm32mp15
- dt-bindings: clock: stm32mp15: rename CK_SCMI define
- dt-bindings: reset: stm32mp15: rename RST_SCMI define
- dt-bindings: reset: add MCU HOLD BOOT ID for SCMI reset domains
on stm32mp15
- dt-bindings: clk: cleanup comments
- ARM: dts: align SPI NOR node name with dtschema
- ARM: dts: stm32: enable optee firmware and SCMI support on STM32MP15
- ARM: dts: stm32: Add SCMI version of STM32 boards (DK1/DK2/ED1/EV1)
- ARM: dts: stm32: move SCMI related nodes in a dedicated file for
stm32mp15
+ patch from stm32-dt-for-v5.19-fixes-2
- ARM: dts: stm32: move SCMI related nodes in a dedicated file for
stm32mp15
- ARM: dts: stm32: fix pwr regulators references to use scmi
- ARM: dts: stm32: use the correct clock source for CEC on stm32mp151
- ARM: dts: stm32: DSI should use LSE SCMI clock on DK1/ED1 STM32 board
- ARM: dts: stm32: delete fixed clock node on STM32MP15-SCMI
- ARM: dts: stm32: add missing usbh clock and fix clk order on stm32mp15
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com> Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Patrick Delaunay [Mon, 11 Jul 2022 17:45:50 +0000 (19:45 +0200)]
configs: stm32m15: support STM32MP_BOARD_EXTRA_ENV for st boards
Correctly handle STM32MP_BOARD_EXTRA_ENV define in stm32mp15_st_common.h;
the STM32MP_BOARD_EXTRA_ENV is added in CONFIG_EXTRA_ENV_SETTINGS
definition, as it is done "stm32mp15_st_common.h"
Without this patch, the content of STM32MP_BOARD_EXTRA_ENV is not used in
the default environment for STMicroelectronics boards.
Fixes: a7d908a54337 ("configs: stm32mp1: set the console variable for extlinux.conf") Reported-by: Gatien CHEVALLIER <gatien.chevallier@foss.st.com> Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Tom Rini [Mon, 27 Jun 2022 17:35:51 +0000 (13:35 -0400)]
powerpc: Use the poison value of 0xdeadbeef directly in DDR init
On p1_p2_rdb_pc platforms, we set ddr_data_init to the "poison" value of
0xdeadbeef rather than a real calculated / derived value. Do this
directly and comment rather than via CONFIG.
Tom Rini [Mon, 27 Jun 2022 17:35:48 +0000 (13:35 -0400)]
omap3: emif4: More clearly hard-code cs0 size
We have a single platform that is both in the OMAP3 family of parts, but
has an EMIF4 memory controller. Currently we hard-code the size of
chip select 0. Make this more clear by putting the value in the
function rather than a CONFIG option.
projects / uboot.git / log