Commit graph

28543 commits

Author SHA1 Message Date
Wu, Josh
b137bd8c8d video: atmel_hlcdfb: enable dcache support
To support dcache, we need flush DMA descriptor buffer before enable lcd
DMA.

Also we need call lcd_set_flush_dcache(1) to make lcd driver flush the
lcd buffer if there is any change.

Cc: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
2014-06-14 18:07:02 +02:00
Wu, Josh
5ae0e38278 net: macb: enable dcache in macb
Add to code to flush the dcache after we writing in DMA buffer.
Also we need invalidate the dcache before we check the status in the
DMA buffer.

Tested in SAMA5D3x-EK with gmac0. Tftp download speed shows in below:
	Disable DCache: 1.1 MiB/s
	Enable DCache: 1.6 MiB/s
Increase speed with about 40%.

The code should have no impact with the boards which are not
enable_dcache().
Tested in AT91SAM9M10G45EK.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
2014-06-14 18:07:01 +02:00
Andreas Bießmann
ceef983bf9 macb: make checkpatch clean
This also renames the CONFIG_SYS_MACB_xx defines. They are used just local and
therefore don't need the CONFIG_SYS_ prefix.

Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
Reviewed-by: Josh Wu <josh.wu@atmel.com>
2014-06-14 18:07:00 +02:00
Andreas Bießmann
0e05543576 avr32: migrate cache functions
Unfortunately the avr32 cache implementation has another API than the one
described in common.h. Migrate the flush/invalidate dcache functions to the
common API to be usable in device drivers.

Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
CC: Josh Wu <josh.wu@atmel.com>
2014-06-14 18:06:58 +02:00
Akshay Saraswat
ed32522fe0 Exynos5420: DMC: Add software read leveling
Sometimes Read DQ and DQS are not in phase. Since, this
phase shift differs from board to board, we need to
calibrate it at DRAM init phase, that's read DQ calibration.
This patch adds SW Read DQ calibration routine to compensate
this skew.

Signed-off-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Akshay Saraswat <akshay.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
2014-06-13 17:05:14 +09:00
Doug Anderson
c9334fcda9 DMC: exynos5420: Gate CLKM to when reading PHY_CON13
when CLKM is running.  If we stop CLKM when sampling it the glitches
all go away, so we'll do that as per Samsung suggestion.

We also check the "is it locked" bits of PHY_CON13 and loop until they
show the the value sampled actually represents a locked value.  It
doesn't appear that the glitching and "is it locked" are related, but
it seems wise to wait until the PHY tells us the value is good before
we use it.  In practice we will not loop more than a couple times (and
usually won't loop at all).

Signed-off-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Akshay Saraswat <akshay.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
2014-06-13 17:05:13 +09:00
Akshay Saraswat
aacdd79095 Exynos5420: Remove code for enabling read leveling
This patch intends to remove all code which enables hardware read
leveling. All characterization environments may not cope up with
h/w read leveling enabled, so we must disable this.
Also, disabling h/w read leveling improves the MIF LVcc value
(LVcc value is the value at which DDR will fail to work properly).
Improving LVcc means we have enough voltage margin for MIF.
When h/w leveling is enabled, we have almost zero volatge margin.

Signed-off-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Akshay Saraswat <akshay.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
2014-06-13 17:05:13 +09:00
Akshay Saraswat
cfde7588d8 Exynos5: DMC: Modify the definition of ddr3_mem_ctrl_init
Passing fewer arguments is better and mem_iv_size is never
used. Let's keep only one argument and make it cleaner.

Signed-off-by: Hatim Ali <hatim.rv@samsung.com>
Signed-off-by: Akshay Saraswat <akshay.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
2014-06-13 17:05:13 +09:00
Minkyu Kang
7922a2d479 Revert "exynos: Enable PSHOLD in SPL"
This reverts commit eb0dd9986c.
2014-06-13 17:00:56 +09:00
Tom Rini
d8a97f934c Merge branch 'master' of git://git.denx.de/u-boot-mmc 2014-06-12 09:02:23 -04:00
Darwin Rambo
3d6a5a4dfc mmc: free allocated memory on initialization errors
Cleanup to balance malloc/free calls.

Signed-off-by: Darwin Rambo <drambo@broadcom.com>
Reviewed-by: Steve Rae <srae@broadcom.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
a5710920b7 cmd_mmc: make mmc dev always re-probe the HW
Currently, U-Boot behaves as follows:

- Begin with no SD card inserted in "mmc 1"
- Execute: mmc dev 1
- This fails, since there is no card
- User plugs in an SD card
- Execute: mmc dev 1
- This still fails, since the HW isn't reprobed.

With this change, U-Boot behaves as follows:

- Begin with no SD card inserted in "mmc 1"
- Execute: mmc dev 1
- This fails, since there is no card
- User plugs in an SD card
- Execute: mmc dev 1
- The newly present SD card is detected

I know that "mmc rescan" will force the HW to be reprobed, but I feel it
makes more sense if "mmc dev" always reprobes the HW after selecting the
current MMC device. This allows scripts to just execute "mmc dev", and
not have to also execute "mmc rescan" to check for media presense.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
941944e445 cmd_mmc: Use init_mmc_device() from do_mmc_rescan()
The body of init_mmc_device() is now identical to that of do_mmc_rescan()
except for the error codes returned. Modify do_mmc_rescan() to simply
call init_mmc_device() and convert the error codes, to avoid code
duplication.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
1ae24a5041 cmd_mmc: add force_init parameter to init_mmc_device()
This allows callers to inject mmc->has_init = 0 between finding the
MMC device, and calling mmc_init(), which forces mmc_init() to rescan
the HW. Future changes will use this feature.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
ecdd57e274 disk: default to HW partition 0 if not specified
Currently, get_device()/get_dev_hwpart() for MMC devices does not select
an explicit HW partition unless the user explicitly requests one, i.e. by
requesting device "mmc 0.0" rather than just "mmc 0". I think it makes
more sense if the default is to select HW partition 0 (main data area)
if the user didn't request a specific partition. Otherwise, the following
happens, which feels wrong:

Select HW partition 1 (boot0):
mmc dev 0 1

Attempts to access SW partition 1 on HW partition 1 (boot0), rather than
SW partition 1 on HW partition 0 (main data area):
ls mmc 0:1 /

With this patch, the second command above re-selects the main data area.

Many device types don't support HW partitions at all, so if HW partition
0 is selected (either explicitly or as the default) and there's no
select_hwpart function, we simply skip attempting to select a HW
partition.

Some MMC devices (i.e. SD cards) don't support HW partitions. However,
this patch still works, since mmc_start_init() sets the current
partition number to 0, and mmc_select_hwpart() succeeds if the requested
partition is already selected.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
60dc58f735 cmd_mmc: default to HW partition 0 if not specified
Currently, "mmc dev 0" does not change the selected HW partition. I
think it makes more sense if "mmc dev 0" is an alias for "mmc dev 0 0",
i.e. that HW partition 0 (main data area) is always selected by default
if the user didn't request a specific partition. Otherwise, the following
happens, which feels wrong:

Select HW partition 1 (boot0):
mmc dev 0 1

Doesn't change the HW partition, so it's still 1 (boot0):
mmc dev 0

With this patch, the second command above re-selects the main data area.

Note that some MMC devices (i.e. SD cards) don't support HW partitions.
However, this patch still works, since mmc_start_init() sets the current
partition number to 0, and mmc_select_hwpart() succeeds if the requested
partition is already selected.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 15:21:12 +03:00
Stephen Warren
d4622df342 mmc: return meaningful error codes from mmc_select_hwpart
Rather than just returning -1 everywhere, try to return something
meaningful from mmc_select_hwpart(). Note that most other MMC functions
don't do this, including functions called from mmc_select_hwpart(), so
I'm not sure how effective this will be. Still, it's one less place with
hard-coded -1.

Suggested-by: Pantelis Antoniou <panto@antoniou-consulting.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
2014-06-12 14:15:59 +03:00
Steve Rae
6736ec15c5 i2c: kona: Resolve Kona I2C driver issue
- "i2c mw" command hangs (with some compilers)

Signed-off-by: Steve Rae <srae@broadcom.com>
2014-06-12 11:42:50 +02:00
Jeroen Hofstee
e153b13c8e common/xyzModem.c: move empty statements to newline
To prevent a warning for clang the loop without a body
is made more clear by moving it to a line of its own.
This prevents a clang warning.

cc: sbabic@denx.de
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
46f46fd48e jffs2:jffs2_1pass.c: remove double braces
Clang interpretes an if condition like  "if ((a = b) == NULL)
as it tries to assign a value in a statement. Hence if you do
"if ((something)) it warns you that you might be confused.
Hence drop the double braces for plane if statements.

Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
2716077cda board:keymile: remove unnecessary double braces
Clang interpretes an if condition like  "if ((a = b) == NULL)
as it tries to assign a value in a statement. Hence if you do
"if ((something)) it warns you that you might be confused.
Hence drop the double braces for plane if statements.

cc: Holger Brunck <holger.brunck@keymile.com>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
930e4254ec common/cli_hush.c: remove unnecessary double braces
Clang interpretes an if condition like  "if ((a = b) == NULL)
as it tries to assign a value in a statement. Hence if you do
"if ((something)) it warns you that you might be confused.
Hence drop the double braces for plane if statements.

Simon Glass <sjg@chromium.org>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
867abdac5e LzmaTools: don't self assign values
It seems the code tries to trick the compiler the argument
is actually used. However compilers became too smart to
fool them so easily an now warn. Gcc and clang don't seem
to emit a warning when the argument is unused. If so it
should be decorated with unused / (void).

Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
c84f9e0d42 imximage_hynix.cfg: fix unterminated comment
cc: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
1b34e880e0 cosmetic: board: pm9263 rewrite old style stuct init
this prevent some warnings when compiling with clang

cc: Stelian Pop <stelian@popies.net>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:06 -04:00
Jeroen Hofstee
7ffdc831f9 tps6586x.h: fix inclusion guard
cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
Acked-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:27:06 -04:00
Vasili Galka
7d89982b7a Remove ${objtree}/include/asm/proc/ link
mkconfig links ${objtree}/include/asm/proc/ to
${srctree}/arch/${arch}/include/asm/proc-armv/. This seems to be a
remnant from the past. Ever since its introduction in 2003 it is used
only in ARM build and always links to same place, so let's simplify
the code, remove it and reference directly where needed.

Successful MAKEALL for ARM and PowerPC verified on Linux.

Signed-off-by: Vasili Galka <vvv444@gmail.com>
2014-06-11 16:27:05 -04:00
Vasili Galka
4b9ca09399 cosmetic: Whitespace fix
Signed-off-by: Vasili Galka <vvv444@gmail.com>
2014-06-11 16:27:05 -04:00
Jeroen Hofstee
c346e466e0 cosmetic: atmel: replace old style struct init
This prevents some warnings when building with clang.
cc:: andreas.devel@googlemail.com
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:05 -04:00
Jeroen Hofstee
46a5707d9c ext4: correctly zero filename
Since ALLOC_CACHE_ALIGN_BUFFER declares a char* for filename
sizeof(filename) is not the size of the buffer. Use the already
known length instead.

cc: Uma Shankar <uma.shankar@samsung.com>
cc: Manjunatha C Achar <a.manjunatha@samsung.com>
cc: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
Acked-by: Marek Vasut <marex@denx.de>
2014-06-11 16:27:05 -04:00
Jeroen Hofstee
8b9cc866c1 common: hash: zero end the string instead of the pointer
if algo->digest_size is zero nothing is set in the str_output
buffer. An attempt is made to zero end the buffer, but the
pointer to the buffer is set to zero instead. I am unaware if
it causes any actual problems, but solves the following warning:

common/hash.c:217:13: warning: expression which evaluates to zero treated as
a null pointer constant of type 'char *' [-Wnon-literal-null-conversion]
                str_ptr = '\0';
                          ^~~~

cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
c71630838d kbuild: move spl/Makefile to scripts/Makefile.spl
All files under spl/ and tpl/ are generated during the build process
except spl/Makefile.

We can simplify clean-rule and git-ignore by moving spl/Makefile
to somewhere else.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
2014-06-11 16:27:05 -04:00
Simon Glass
ddc94378db m68k: Fix warnings with gcc 4.6
Most of the warnings seem to be related to using 'int' for size_t. Change
this and fix up the remaining warnings and problems. For bootm, the warning
was masked by others, and there is an actual bug in the code.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:27:05 -04:00
Chris Packham
34e4a2ec0a docs: driver-model: Fix spelling
Signed-off-by: Chris Packham <judge.packham@gmail.com>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
7050f0de7e .gitignore: move *.exe pattern to the top gitignore for Cygwin
GCC on Cygwin generates executables with .exe extension,
for example:
 scripts/basic/fixdep.exe
 scripts/docproc.exe

To ignore them, *.exe pattern should be moved
from tools/.gitignore to ./.gitignore

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
96b09a97f5 kbuild: remove unnecessary adjustment for Cygwin
"SFX = .exe" was originally added for Cygwin environment.

It is true that GCC on Cygwin spits executables with .exe extention.

For example,

  gcc -o foo foo.c

will generate "foo.exe", not "foo".

But GNU make is also nicely adjusted for Cygwin.

For example,

  foo: foo.c
          gcc -o $@ $<

will compare the timestamp between "foo.exe" and "foo.c".

You do not have to tweak Makefiles like this:

  foo$(SFX): foo.c
          gcc -o $@ $<

And "make clean" works as well without adjustment for Cygwin because
the command "rm foo" on Cygwin will delete both "foo" and "foo.exe".

In conclusion, makefiles do not need special care for Cygwin.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
899a8cbbc2 .gitignore: drop *.dts.tmp pattern
This pattern was added by commit cc4f427b to ignore the intermidiate
file for generating DTB.

When Kbuild was introduced, dts/Makefile was totally re-written.
This ignore pattern is already useless.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
ad80c4a322 kbuild, tools: generate wrapper C sources automatically by Makefile
There are many source files shared between U-boot image and tools.
Instead of adding a lot of dummy wrapper files that just include
the corresponding file in lib/ or common/ directory,
Makefile should automatically generate them.

The original inspiration for this came from
scripts/Makefile.asm-generic of Linux Kernel.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:27:05 -04:00
Darwin Dingel
4a36be9bdb disk: part_dos.c: Add a PBR check when MBR checking fails
Bug: SDCard with a messed up partition but still has a FAT signature
intact is readable in Linux but unreadable in uboot with 'fatls'.

Fix: When partition info checking fails, there is no checking for a
FAT signature (DOS_PBR) which will fail 'fatls'. FAT signature checking
is done when no valid partition is found in partition table. If FAT
signature is found, the disk will be read as PBR and continue
processing.

Signed-off-by: Darwin Dingel <darwin.dingel@alliedtelesis.co.nz>
2014-06-11 16:27:05 -04:00
Masahiro Yamada
15c939f970 kbuild: export HOSTCXX and HOSTCXXFLAGS
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
2014-06-11 16:25:47 -04:00
Thomas Betker
52715f8931 Use run_command_repeatable()
Replace run_command() by run_command_repeatable() in places which
depend on the return code to indicate repeatability.

Signed-off-by: Thomas Betker <thomas.betker@rohde-schwarz.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:47 -04:00
Thomas Betker
1d43bfd2d5 Add run_command_repeatable()
run_command() returns 0 on success and 1 on error. However, there are some
invocations which expect 0 or 1 for success (not repeatable or repeatable)
and -1 for error; add run_command_repeatable() for this purpose.

Signed-off-by: Thomas Betker <thomas.betker@rohde-schwarz.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:47 -04:00
Thomas Betker
73671dad49 Check run_command() return code properly
run_command() returns 0 for success, 1 for failure. Fix places which
assume that failure is indicated by a negative return code.

Signed-off-by: Thomas Betker <thomas.betker@rohde-schwarz.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Tested-by: Stefan Roese <sr@denx.de>
2014-06-11 16:25:47 -04:00
Simon Glass
32e2c42a83 am33xx/omap: Add a new board to enable verified boot
Enable verified boot functionality for a new am335x_boneblack_vboot target.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:47 -04:00
Simon Glass
dd42a4abf6 am33xx/omap: Enable FIT support
Enable booting a FIT containing a kernel/device tree.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:47 -04:00
Simon Glass
5cc16cbf25 am33xx/omap: Enable CONFIG_OF_CONTROL
Add support for device tree control and add device tree files for the
beaglebone black initially.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:46 -04:00
Simon Glass
0e1612a7d1 arm: ti: Increase malloc size to 16MB for armv7 boards
The current size of 1MB is not enough use to use DFU. Increase it for
ARMv7 boards, all of which should have 32MB or more SDRAM.

With this change it is possible to do 'dfu mmc 0' on a Beaglebone Black.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:46 -04:00
Simon Glass
a946811569 mkimage: Automatically make space in FDT when full
When adding hashes or signatures, the target FDT may be full. Detect this
and automatically try again after making 1KB of space.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:46 -04:00
Simon Glass
ef0af64b1c Improve error handling in fit_common
Make the error handling common, and make sure the file is always closed
on error. Rename the parameter to be more description and add comments.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:46 -04:00
Simon Glass
4f427a421f fdt: Update functions which write to an FDT to return -ENOSPC
When writing values into an FDT it is possible that there will be
insufficient space. If the caller gets a useful error then it can
potentially deal with the situation.

Adjust these functions to return -ENOSPC when the FDT is full.

Signed-off-by: Simon Glass <sjg@chromium.org>
2014-06-11 16:25:46 -04:00