Albert ARIBAUD [Fri, 5 Oct 2012 16:43:33 +0000 (18:43 +0200)]
arm: armv7: temporarily set -mno-unaligned-access
This patch aims at ensuring that the 2012.10 release works
out-of-the-box on as many targets as possible, by reinstating
commit 5347560f5427bcdd48a563b62180481606ac8044, which adds
option -mno-unaligned-access to armv7 builds.
This patch will be overriden immediately after release of 2012.10.
Dinh Nguyen [Thu, 4 Oct 2012 06:46:02 +0000 (06:46 +0000)]
ARM: Add Altera SOCFPGA Cyclone5
Add minimal support for Altera's SOCFPGA Cyclone 5 hardware.
Signed-off-by: Dinh Nguyen <dinguyen@altera.com> Signed-off-by: Chin Liang See <clsee@altera.com> Signed-off-by: Pavel Machek <pavel@denx.de> Reviewed-by: Marek Vasut <marex@denx.de> Acked-by: Tom Trini <trini@ti.com> Cc: Wolfgang Denx <wd@denx.de> Cc: Albert Aribaud <albert.u.boot@aribaud.net> Cc: Stefan Roese <sr@denx.de>
----
v8: Remove no_return attribute for reset_cpu
Without usb-serial cable plugged at this stage, some
garbage is seen in UART RX FIFO, which blocks autoboot
progress. The fix makes sure to empty the RX FIFO,
before we wait for user input to interrupt autoboot.
Michal Simek [Thu, 13 Sep 2012 20:23:35 +0000 (20:23 +0000)]
arm: Support new Xilinx Zynq platform
Add timer driver.
Signed-off-by: Michal Simek <monstr@monstr.eu> CC: Joe Hershberger <joe.hershberger@gmail.com> CC: Marek Vasut <marex@denx.de> Acked-by: Marek Vasut <marex@denx.de>
Michal Simek [Fri, 14 Sep 2012 00:55:24 +0000 (00:55 +0000)]
serial: Add Zynq serial driver
The driver is used on Xilinx Zynq platform.
Signed-off-by: Michal Simek <monstr@monstr.eu> CC: Joe Hershberger <joe.hershberger@gmail.com> CC: Marek Vasut <marex@denx.de> Acked-by: Marek Vasut <marex@denx.de>
Allen Martin [Wed, 18 Jul 2012 13:45:53 +0000 (13:45 +0000)]
arm: work around assembler bug
Disable sibling call optimization based on binutils version. This is
to work around a bug in the assember in binutils versions < 2.22.
Branches to weak symbols can be incorrectly optimized in thumb mode to
a short branch (b.n instruction) that won't reach when the symbol gets
preempted.
Rob Herring [Thu, 28 Jun 2012 03:54:11 +0000 (03:54 +0000)]
ARM: increase lmb stack space reservation to 4KB
The bootm initrd image copy to ram can collide with the stack in cases
where the print buffer size is large (i.e. 1K). The result is intermittent
initrd decompression errors depending on the initrd size MOD 4KB since
the initrd start address is 4KB aligned.
Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Stefan Roese [Wed, 30 May 2012 22:59:08 +0000 (22:59 +0000)]
ARM: Add X600 board support (SPEAr600 based)
This patch adds support for the X600 SPEAr600 based board. Its also
the first SPEAr600 board that uses the newly introduced SPEAr600
SPL support. Xloader is not necessary any more. By using the new
"u-boot.spr" make target, one image will generated containing both,
U-Boot SPL (with mkimage header as needed by the SPEAr BootROM, and
the main U-Boot with mkimage header.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Amit Virdi <amit.virdi@st.com> Cc: Vipin Kumar <vipin.kumar@st.com>
The kmnusa board uses a mv88e6352 switch that is connected to the main
eth interface of the kirkwood. Therefore the switch must be configured
so that the kirkwood's egiga eth inferface can be used.
Simon Baatz [Fri, 20 Jul 2012 09:59:14 +0000 (09:59 +0000)]
kirkwood: ib62x0: Invert SATA activity LEDs
The hardware design of the IB-NAS62x0 causes the SATA activity
LEDs to be on when idle by default. Reverse the polarity of the
activity LEDs in early init.
Signed-off-by: Simon Baatz <gmbnomis@gmail.com> Cc: Luka Perkov <uboot@lukaperkov.net> Cc: Prafulla Wadaskar <prafulla@marvell.com> Acked-by: Luka Perkov <uboot@lukaperkov.net>
Size of the SDRAM chips might differ between any two (otherwise
identical) instances of the same board.
So add a function kw_sdram_size_adjust() which reads out the current
ram size for a given bank, and adjusts the Kirkwood's SDRAM window size
register accordingly.
Some boards might be equipped with different SDRAM configurations.
When that is the case, CPU CS Window Size Register (CS[0]n Size)
should be set to the biggest value through board.cfg file; then its
value can be fixed at runtime according to the detected SDRAM size.
Signed-off-by: Luka Perkov <uboot@lukaperkov.net> Tested-by: Wojciech Dubowik <wojciech.dubowik@neratec.com> Tested-by: Tim Fletcher <tim@night-shade.org.uk>
Simon Guinot [Thu, 6 Sep 2012 10:51:42 +0000 (10:51 +0000)]
ARM: add support for Network Space v2 Lite and Mini
This patch adds support for the LaCie boards Network Space v2 (Lite and
Mini). This two boards are derived from the Network Space v2 and a lot
of hardware caracteristics are shared.
- CPU: Marvell 88F6192 800Mhz
- SDRAM memory: 128MB DDR2 200Mhz
- 1 SATA port: internal
- Gigabit ethernet: PHY Marvell 88E1318
- Flash memory: SPI NOR 512KB (Macronix MX25L4005A)
- i2c EEPROM: 512 bytes (24C04 type)
- 2 USB2 ports (Lite only): host and host/device
- 1 push button
- 1 SATA LED (bi-color, blue and red)
Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Rmobile common timer driver diverts the same driver as SH architecture.
When it builds at the same place with source, it is no problem, but when
it builds out of source, it cannot build.
This patch revises this problem.
Reported-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
arm: rmobile: kzm9g: Add CONFIG_GLOBAL_TIMER to board config file
kzm9g board use global timer. But by commit 813ffda31, timer function of
rmobile was changed that global timer might be used, when CONFIG_GLOBAL_TIMER
was defined.
This add CONFIG_GLOBAL_TIMER to board config file.
arm: rmobile: kzm9g: separate cpu_rev to integer and fraction
According to SoC document, revision info is separated to integer part and
fracton part.
So I separete rmobile_get_cpu_rev() to rmobile_get_cpu_rev_integer() and
rmobile_get_cpu_rev_fraction().
The value of CONFIG_BAUDRATE is treated as string and put as initial value of
environment variable. If it begin with '(', it is wrongly parsed to 0 in number.
So I removed '(' and ')'.
Latest rmobile code was tested by using old gcc (gcc-4.4).
When we use gcc-4.6 (or later), the build is made, but does not work.
This solves a problem not to work by add -march=armv5 to compiple option
when we built in gcc-4.6 (or later).
I tested by linaro's compiler version 2012.04-20120426.
arm: rmobile: kzm9g: Modify bus controller setting for CS4
Problem:
Linux kernel hangs up when it write a file to NFS mounted directory.
Solution:
Modify bus controller setting for CS4, which connected smsc9221 ethernet
controller.
Detail:
Modify CS4BCR bit[29:28] (IWW[1:0]) from 00 to 01.
Modify CS4BCR bit[20:19] (IWRRD[1:0]) from 00 to 01.
Modify CS4BCR bit[17:16] (IWRRS[1:0]) from 00 to 01.
Modify CS4WCR bit[27:26] (WSW[1:0]) from 10 to 11
Modify CS4WCR bit[25:24] (WHW[1:0]) from 01 to 10
Modify CS4WCR bit[18:16] (WW[2:0]) from 101 to 111
Modify CS4WCR bit[13:11] (SW[2:0]) from 010 to 011
Modify CS4WCR bit[10:7] (WR[3:0]) from 1000 to 1011
am335x-evm config: decrease bootdelay to 1s and mount rootfs RO
A fast boot is important to the beaglebone, so save 2 seconds here by
decreasing bootdelay. This is still plenty time to break into the prompt,
I do that at least once a day.
Mount the rootfs RO by default, this is needed to make fsck succeed
without resorting to an initramfs.
Joel A Fernandes [Tue, 18 Sep 2012 04:30:51 +0000 (04:30 +0000)]
am33xx: Fix fetching of mmc1 bootmode from bootrom for AM33XX
U-boot should not ignore getting the bootmode passed on from the bootrom.
With this, U-boot SPL knows it was loaded from MMC1 and use this info to
read full U-boot from MMC1 as well.
Cc: pprakash@ti.com Cc: trini@ti.com Signed-off-by: Joel A Fernandes <joelagnel@ti.com> Signed-off-by: Tom Rini <trini@ti.com>
This patch adds support for networking in SPL. Some devices are
capable of loading SPL via network so it makes sense to load the
main U-Boot binary via network too. This patch tries to use
existing network code as much as possible. Unfortunately, it depends
on environment which in turn depends on other code so SPL size
is increased significantly. No effort was done to decouple network
code and environment so far.
Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com> Acked-by: Joe Hershberger <joe.hershberger@ni.com> Signed-off-by: Tom Rini <trini@ti.com>
Vendor Class Identifier option is common to BOOTP and DHCP and
can be useful without PXE. So send VCI in both BOOTP and DHCP
requests if CONFIG_BOOTP_VCI_STRING is defined.
Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com> Signed-off-by: Tom Rini <trini@ti.com>
Commit cfda5aeab89d73779e26f0d34cf10f64caa67431
(cmd_fat: use common get_device_and_partition function)
introduced a warning:
cmd_fat.c: In function 'do_fat_fswrite':
cmd_fat.c:178:8: warning: unused variable 'ep' [-Wunused-variable]
Fix it.
Signed-off-by: Anatolij Gustschin <agust@denx.de> Reported-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Tested-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Lei Wen [Fri, 28 Sep 2012 04:26:45 +0000 (04:26 +0000)]
lib: zlib: remove the limitation for cannot using 0 as start
We often need the requirement that compressing those memory range start
from 0, but the default deflate code in zlib prevent us to do this.
Considering the special case of uboot, that it could access all memory
range, it is reasonable to be able to also take the address space from 0
into compression.
Pavel Herrmann [Thu, 27 Sep 2012 23:18:03 +0000 (23:18 +0000)]
remove unnecessary code in ata_piix
We set sata_curr_device to 0 right after returning from init_sata(), so there's
no point in setting it to the last scanned driver at this point.
Note: there are more duplicities with cmd_sata, but those might be required,
as the code seems to reset the entire controller on every scan, ignoring the
requested port number.
Signed-off-by: Pavel Herrmann <morpheus.ibis@gmail.com>
Stephen Warren [Fri, 28 Sep 2012 05:34:09 +0000 (05:34 +0000)]
disk: allow - or empty string to fall back to $bootdevice
Commit 10a37fd "disk: get_device_and_partition() "auto" partition"
prevented the use of "-" on the command-line to request fallback to the
$bootdevice environment variable instead. This patch allows that, or an
empty string "" to be used.
Tom Rini [Mon, 27 Aug 2012 21:58:28 +0000 (14:58 -0700)]
SPL: Rework how we inform about un-headered images
First, remove the puts from the case where we don't have an mkimage
header as this is somewhat common and intentional for no-arg target
images. Second, rework the final switch statement in board_init_r to,
in the case of !CONFIG_SPL_OS_BOOT be only about doing debug prints
about if we know what the magic is or not (the CONFIG_SPL_OS_BOOT case
is unchanged). Then we call jump_to_image_no_args(). This gives us the
same behavior as before but with slightly smaller code.
Stefan Roese [Tue, 28 Aug 2012 08:50:59 +0000 (10:50 +0200)]
SPL: Enable use of custom defined U-Boot entry point
By setting CONFIG_SYS_UBOOT_START boards can now use a different entry
point for their U-Boot image. So the U-Boot entry point is not fixed
to CONFIG_SYS_TEXT_BASE any more.
Signed-off-by: Stefan Roese <sr@denx.de> Signed-off-by: Tom Rini <trini@ti.com>
projects / u-boot.git / log
