Simon Glass [Sat, 15 Nov 2014 03:56:35 +0000 (20:56 -0700)]
x86: Allow an option ROM to be built into U-Boot
Some x86 machines require a binary blob containing 16-bit initialisation
code for their video hardware. Allow this to be built into the x86 ROM so
that it is accessible during boot.
Simon Glass [Sat, 15 Nov 2014 03:56:33 +0000 (20:56 -0700)]
pci: Add general support for execution of video ROMs
Some platforms don't have native code for dealing with their video
hardware. In some cases they use a binary blob to set it up and perform
required actions like setting the video mode. This approach is a hangover
from the old PC days where a ROM was provided and executed during startup.
Even now, these ROMs are supplied as a way to set up video. It avoids the
code for every video chip needing to be provided in the boot loader. But
it makes the video much less flexible - e.g. it is not possible to do
anything else while the video init is happening (including waiting hundreds
of milliseconds for display panels to start up).
In any case, to deal with this sad state of affairs, provide an API for
execution of x86 video ROMs, either natively or through emulation.
Simon Glass [Sat, 15 Nov 2014 03:56:32 +0000 (20:56 -0700)]
x86: Add support for running option ROMs natively
On x86 machines we can use an emulator to run option ROMS as with other
architectures. But with some additional effort (mostly due to the 16-bit
nature of option ROMs) we can run them natively. Add support for this.
Bin Meng [Thu, 20 Nov 2014 08:11:16 +0000 (16:11 +0800)]
x86: Refactor interrupt_init()
Rename interrupt_init() in arch/x86/lib/pcat_interrupts.c to
i8259_init() and create a new interrupt_init() in
arch/x86/cpu/interrupt.c to call i8259_init() followed by a
call to cpu_init_interrupts().
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org> Tested-by: Simon Glass <sjg@chromium.org>
Bin Meng [Thu, 20 Nov 2014 08:10:49 +0000 (16:10 +0800)]
x86: Call cpu_init_interrupts() from interrupt_init()
Currently cpu_init_interrupts() is called from cpu_init_r() to
setup the interrupt and exception of the cpu core, but at that
time the i8259 has not been initialized to mask all the irqs
and remap the master i8259 interrupt vector base, so the whole
system is at risk of being interrupted, and if interrupted,
wrong interrupt/exception message is shown.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org> Tested-by: Simon Glass <sjg@chromium.org>
Simon Glass [Sat, 15 Nov 2014 01:18:36 +0000 (18:18 -0700)]
x86: dts: Move PCI peripherals into a pci node
These peripherals should not be at the top level, since they exist inside
the PCI bus. We don't have a full device tree node for pci yet, but we
should at least put it at the right level.
Simon Glass [Sat, 15 Nov 2014 01:18:30 +0000 (18:18 -0700)]
pci: Add functions to read and write a BAR address
Some PCI functions cannot be auto-configured. Add a function to set up a
fixed BAR which can be used in these situations. Also add a function to read
the current address of a BAR.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sat, 15 Nov 2014 01:18:29 +0000 (18:18 -0700)]
x86: config: Enable plug-and-play for link PCI
Enable this option so that we can configure the available PCI devices. Also
make sure that PCI is available early after relocation as we use it for
several other subsystems.
Simon Glass [Sat, 15 Nov 2014 01:18:25 +0000 (18:18 -0700)]
x86: Factor out common values in the link script
Define the reset base in config.mk so that it does not need to be calculated
twice in the link script. Also tidy up the START_16 and RESET_VEC_LOC values
to fit with this new approach.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sat, 15 Nov 2014 01:18:24 +0000 (18:18 -0700)]
x86: Ensure that all relocation data is included in the image
Some toolchains put the relocation data into separate sections. Adjust the
linker script to catch this case. Without relocation data, U-Boot will not
boot.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sat, 15 Nov 2014 01:18:19 +0000 (18:18 -0700)]
Align embedded device tree correctly
Device trees must be aligned to a 4-byte boundary. This was dropped in the
Kbuild conversion. Bring it back, and use 16-byte alignment for good
measure.
Rabin Vincent [Fri, 21 Nov 2014 22:05:22 +0000 (23:05 +0100)]
Revert "hush: fix segfault on syntax error"
128059b92 ("hush: fix segfault on syntax error") attempted to fix a
segfault on syntax errors, but it broke Ctrl-C handling, and the
assumption that it made, that rcode could not be -1, is incorrect.
Revert this change.
Reported-by: Stephen Warren <swarren@wwwdotorg.org> Reported-by: Przemyslaw Marczak <p.marczak@samsung.com> Signed-off-by: Rabin Vincent <rabin@rab.in>
Nikita Kiryanov [Fri, 21 Nov 2014 10:47:26 +0000 (12:47 +0200)]
arm: imx: stop sata on boot
Ideally, the Linux kernel should get the hardware in its most
untouched state. For the most part, U-Boot does not reset the various
subsystems it touches before boot, and usually Linux deals with it, but
on some boards (cm_fx6) the Linux kernel fails to detect the ssd
correctly if sata is used by U-Boot.
Power off sata on OS boot so that Linux will have a clean state to work
with.
Nikita Kiryanov [Fri, 21 Nov 2014 10:47:24 +0000 (12:47 +0200)]
cmd_sata: implement sata stop command
Implement sata stop command.
This introduces the __sata_stop() weak function, which mirrors
the weak __sata_initialize() function, giving users the option of
undoing the custom steps performed in overrides of sata_initialize().
Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il> Cc: Marek Vasut <marex@denx.de> Cc: Tom Rini <trini@ti.com>
Nikita Kiryanov [Fri, 21 Nov 2014 10:47:23 +0000 (12:47 +0200)]
sata: implement reset_sata for dwc_ahsata
Add reset_sata() to the sata driver interface and implement it
for dwc_ahsata. This function cleans up after sata_init(), and
therefore accepts a device number like sata_init() does.
A dummy implementation is provided for the rest of the drivers.
Fabio Estevam [Thu, 20 Nov 2014 15:17:56 +0000 (13:17 -0200)]
gw_ventana: Use the generic spl_sd.cfg
gw_ventana can boot from SPI or NAND and both of these interfaces boot from
the same 0x400 offset.
This means that we could simplify the code and replace the custom gw_ventana.cfg
with the generic spl_sd.cfg, as it provides the same boot offset of 0x400.
Cc: Tim Harvey <tharvey@gateworks.com> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Hans de Goede [Fri, 14 Nov 2014 08:34:31 +0000 (09:34 +0100)]
ARM: bootm: Allow booting in secure mode on hyp capable systems
Older Linux kernels will not properly boot in hyp mode, add support for a
bootm_boot_mode environment variable, which can be set to "sec" or "nonsec"
to force booting in secure or non-secure mode when build with non-sec support.
The default behavior can be selected through CONFIG_ARMV7_BOOT_SEC_DEFAULT,
when this is set booting in secure mode is the default. The default setting
for this Kconfig option is N, preserving the current behavior of booting in
non-secure mode by default when non-secure mode is supported.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Marc Zyngier <marc.zyngier@arm.com> Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Hans de Goede [Fri, 14 Nov 2014 08:34:30 +0000 (09:34 +0100)]
ARM: Add arch/arm/cpu/armv7/Kconfig with non-secure and virt options
Add arch/arm/cpu/armv7/Kconfig with non-secure and virt options, this is a
preparation patch for adding an env variable to choose between secure /
non-secure boot on non-secure boot capable systems, specifically this
prepares for adding CONFIG_ARMV7_BOOT_SEC_DEFAULT as a proper Kconfig option.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Tom Rini [Fri, 14 Nov 2014 08:34:29 +0000 (09:34 +0100)]
Kconfig: Add EXPERT option
For similar reasons to why the Linux Kernel has an EXPERT option, we too
want an option to allow for tweaking of some options that while normally
should remain hidden, may need to be changed in some cases.
Signed-off-by: Tom Rini <trini@ti.com> Acked-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Suriyan Ramasami [Mon, 17 Nov 2014 22:39:38 +0000 (14:39 -0800)]
fs: API changes enabling extra parameter to return size of type loff_t
The sandbox/ext4/fat/generic fs commands do not gracefully deal with files
greater than 2GB. Negative values are returned in such cases.
To handle this, the fs functions have been modified to take an additional
parameter of type "* loff_t" which is then populated. The return value
of the fs functions are used only for error conditions.
Signed-off-by: Suriyan Ramasami <suriyan.r@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
[trini: Update board/gdsys/p1022/controlcenterd-id.c,
drivers/fpga/zynqpl.c for changes] Signed-off-by: Tom Rini <trini@ti.com>
Suriyan Ramasami [Mon, 17 Nov 2014 22:39:34 +0000 (14:39 -0800)]
sandbox: Use md5sum and fatwrite to enable testing of fs commands
Enable md5sum to obtain the MD5 of the read and written files to check
their contents for validity.
Use map_sysmem() to map buffer in a sandbox environment.
Signed-off-by: Suriyan Ramasami <suriyan.r@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
This fixes following warning when compiled with MIPS64
common/board_f.c: In function 'display_text_info':
common/board_f.c:150:2: warning: format '%X' expects argument i
of type 'unsigned int', but argument 2 has type 'long unsigned int' [-Wformat=]
debug("U-Boot code: %08X -> %08lX BSS: -> %08lX\n",
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Masahiro Yamada [Thu, 13 Nov 2014 10:29:09 +0000 (19:29 +0900)]
spl: fix descending condition to drivers/mtd/nand/
SPL should not reference CONFIG_CMD_NAND to decide whether or not
it should build drivers/mtd/nand. CONFIG_CMD_NAND should be only
used to select the NAND utility command on the command parser.
Masahiro Yamada [Thu, 13 Nov 2014 10:29:07 +0000 (19:29 +0900)]
hush: add CONFIG_HUSH_PARSER to Kconfig
The README file states that the macros beginning with "CONFIG_SYS_"
depend on the hardware etc. and should not be meddled with if you do
not what you're doing.
We have already screwed up with this policy; we have given the prefix
"CONFIG_SYS_" to many user-selectable configurations.
Here, "CONFIG_SYS_HUSH_PARSER" is one of them. Users can enable it
if they want to use a more powerful command line parser, or disable it
if they only need a simple one.
This commit attempts to rename CONFIG_SYS_HUSH_PARSER to
CONFIG_HUSH_PARSER and move it to Kconfig.
Every board maintainer is expected to enable CONFIG_HUSH_PARSER
(= add "CONFIG_HUSH_PARSER=y" to his defconfig file) and remove
"#define CONFIG_SYS_HUSH_PARSER" from his header file.
Since commit 0defddc851ed (config: Add a default CONFIG_SYS_PROMPT),
each board header does not need to define CONFIG_SYS_PROMPT
as long as it uses the default prompt "=> ".
Masahiro Yamada [Thu, 13 Nov 2014 03:28:41 +0000 (12:28 +0900)]
x86: use CONFIG_SYS_COREBOOT to descend into coreboot/ directory
The references of CONFIG_SYS_COREBOOT in arch/x86/cpu/coreboot/Makefile
are redundant because the build system descends into the directory
only when CONFIG_SYS_COREBOOT is defined.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Cc: Simon Glass <sjg@chromium.org> Acked-by: Simon Glass <sjg@chromium.org>
Masahiro Yamada [Thu, 13 Nov 2014 03:28:40 +0000 (12:28 +0900)]
kbuild: Descend into SOC directory from CPU directory
Some CPUs of some architectures have SOC directories.
At present, the build system directly descends into SOC directories
from the top Makefile, but it should generally descend into each
directory from its parent directory.
Thierry Reding [Thu, 13 Nov 2014 01:26:50 +0000 (18:26 -0700)]
Add pr_fmt() macro
This macro can be overridden in source files (before including common.h)
and can be used to specify a prefix for debug and error messages. An
example of how to use this is shown below:
#define pr_fmt(fmt) "foo: " fmt
#include <common.h>
...
debug("bar");
The resulting message will read:
foo: bar
Acked-by: Simon Glass <sjg@chromium.org> Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Thierry Reding [Thu, 13 Nov 2014 01:26:49 +0000 (18:26 -0700)]
pci: Honour pci_skip_dev()
When enumerating devices, honour the pci_skip_dev() function. This can
be used by PCI controller drivers to restrict which devices will be
probed.
This is required by the NVIDIA Tegra PCIe controller driver, which will
fail with a data abort exception if an access is attempted to a device
number larger than 0 outside of bus 0. pci_skip_dev() is therefore
implemented to prevent any such accesses.
Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Thierry Reding [Thu, 13 Nov 2014 01:26:48 +0000 (18:26 -0700)]
pci: Abort early if bus does not exist
When listing the devices on a PCI bus, the current code will blindly try
to access all devices. Internally this causes pci_bus_to_hose() to be
repeatedly called and output an error message every time. Prevent this
by calling pci_bus_to_hose() once and abort early if no bus was found.
Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Thierry Reding [Thu, 13 Nov 2014 01:26:47 +0000 (18:26 -0700)]
vsprintf: Add modifier for phys_addr_t
Provide a new modifier to vsprintf() to print phys_addr_t variables to
avoid having to cast or #ifdef when printing them out. The %pa modifier
is used for this purpose, so phys_addr_t variables need to be passed by
reference, like so:
phys_addr_t start = 0;
printf("start: %pa\n", &start);
Depending on the size of phys_addr_t this will print out the address
with 8 or 16 hexadecimal digits following a 0x prefix.
Signed-off-by: Thierry Reding <treding@nvidia.com> Tested-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Some filesystems have a UUID stored in its superblock. To
allow using root=UUID=... for the kernel command line we
need a way to read-out the filesystem UUID.
changes rfc -> v1:
- make the environment variable an option parameter. If not
given, the UUID is printed out. If given, it is stored in the env
variable.
- corrected typos
- return error codes
changes v1 -> v2:
- fix return code of do_fs_uuid(..)
- document do_fs_uuid(..)
- implement fs_uuid_unsuported(..) be more consistent with the
way other optional functionality works
changes v2 -> v3:
- change ext4fs_uuid(..) to make use of #if .. #else .. #endif
construct to get rid of unreachable code
Hit any key to stop autoboot: 0
=> fsuuid
fsuuid - Look up a filesystem UUID
Usage:
fsuuid <interface> <dev>:<part>
- print filesystem UUID
fsuuid <interface> <dev>:<part> <varname>
- set environment variable to filesystem UUID
Stefan Roese [Tue, 11 Nov 2014 18:03:55 +0000 (19:03 +0100)]
spl: Change debug to printf for "Unsupported boot-device"
We had the problem on an AM33xx platform, that SPL detected an
unsupported boot-device. But since this message is a debug message
it took a bit of time to really know, where the hangup in SPL
resulted from. So let's change this debug message to a printf
and also print the detected boot-device that is not supported.
This makes debugging of such cases much easier.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Wolfgang Denk <wd@denx.de> Cc: Tom Rini <trini@ti.com> Acked-by: Heiko Schocher <hs@denx.de>
Masahiro Yamada [Thu, 6 Nov 2014 18:03:31 +0000 (03:03 +0900)]
linux/kernel.h: sync min, max, min3, max3 macros with Linux
U-Boot has never cared about the type when we get max/min of two
values, but Linux Kernel does. This commit gets min, max, min3, max3
macros synced with the kernel introducing type checks.
Many of references of those macros must be fixed to suppress warnings.
We have two options:
- Use min, max, min3, max3 only when the arguments have the same type
(or add casts to the arguments)
- Use min_t/max_t instead with the appropriate type for the first
argument
Masahiro Yamada [Mon, 17 Nov 2014 08:19:41 +0000 (17:19 +0900)]
dm: core: remove unnecessary return condition in uclass lookup
These conditions never happen.
- There is no real uclass with UCLASS_INVALID id.
- uclass never becomes NULL because ll_entry_start() always returns
a valid pointer.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Acked-by: Simon Glass <sjg@chromium.org>
Masahiro Yamada [Mon, 17 Nov 2014 08:19:39 +0000 (17:19 +0900)]
dm: core: remove meaningless if conditional
If the variable "ret" is equal to "-ENOENT", it is trapped at [1] and
never reaches [2]. At [3], the condition "ret != -ENOENT" is always
true.
if (ret == -ENOENT) { <------------------ [1]
continue;
} else if (ret == -ENODEV) {
dm_dbg("Device '%s' has no compatible string\n", name);
break;
} else if (ret) { <------------------ [2]
dm_warn("Device tree error at offset %d\n", offset);
if (!result || ret != -ENOENT) <------------------ [3]
result = ret;
break;
}
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Acked-by: Simon Glass <sjg@chromium.org>
Simon Glass [Tue, 11 Nov 2014 17:46:23 +0000 (10:46 -0700)]
dm: spi: Use device_bind_driver() instead of our own function
The SPI function does the same thing, so we may as well just use the new
generic function. The 'cs' parameter was not actually used, so can be
dropped.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com> Acked-by: Heiko Schocher <hs@denx.de>
Simon Glass [Tue, 11 Nov 2014 17:46:20 +0000 (10:46 -0700)]
dm: fdt: Correct handling of aliases with embedded digits
Since we scan from left to right looking for the first digit, "i2c0" returns
2 instead of 0 for the alias number. Adjust the code to scan from right to
left instead.
Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Heiko Schocher <hs@denx.de> Reviewed-by: Tom Rini <trini@ti.com>
projects / u-boot.git / log
