Devices in CONFIG_SYS_POST_I2C_IGNORES list may be absent
and the rule is not to report I2C POST failure for devices
in this list. Currently this doesn't work since probing for
these devices isn't done and thus they are not marked as
successfully probed. Ignore optional devices when checking
for devices that didn't respond.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
These boards have long reached EOL, and there has been no indication
of any active users of such hardware for years. Get rid of the dead
weight.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Wolfgang Grandegger <wg@denx.de>
Remove the parts depending either on disabled CONFIG_SERIAL_MULTI
or ifdefs around CONFIG_SERIAL_MULTI parts since CONFIG_SERIAL_MULTI
is now enabled by default.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: Tom Rini <trini@ti.com>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Stefan Roese <sr@denx.de>
Since commit 96f5c4b the needed functions (get_ticks() and get_tbclk() )
are defined for kirkwood CPUs as well. This warning is then not relevant
anymore.
Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
cc: Holger Brunck <holger.brunck@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Albert Aribaud <albert.u.boot@aribaud.net>
These calls should not be made directly any more, since bootstage
will call the show_boot_...() functions as needed.
Signed-off-by: Simon Glass <sjg@chromium.org>
Rather than the caller negating our progress numbers to indicate an
error has occurred, which seems hacky, add a function to indicate this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
This reverts commit 3e16abe0e4.
The logic of this patch is broken - testing for CONFIG_SYS_POST_FPU in
the Makefile cannot work, as this is only a bit that may (or may not)
be set in the CONFIG_POST variable.
The patch cases build errors on a number of boards, so we revert it.
Signed-off-by: Wolfgang Denk <wd@denx.de>
The code and comment disagreed: the comment claimed that r6...r31
were copied, and consequently the arrays for "src" and "dst" were
declared with 26 entries, but the actual code ("lmw r5,0(r3)" and
"stmw r5,0(r4)") copied _27_ words (r5 through r31), which resulted
in false "POST cpu Error at multi test" messages.
Fix the comment and the array sizes.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Stefan Roese <sr@denx.de>
Cc: Kumar Gala <galak@kernel.crashing.org>
Cc: Kim Phillips <kim.phillips@freescale.com>
Cc: Andy Fleming <afleming@gmail.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Anatolij Gustschin <agust@denx.de>
Clean up and document the code:
- get rid of unneeded code block
- add comment which code is generated
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Stefan Roese <sr@denx.de>
Cc: Kumar Gala <galak@kernel.crashing.org>
Cc: Kim Phillips <kim.phillips@freescale.com>
Cc: Andy Fleming <afleming@gmail.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Stefan Roese <sr@denx.de>
Cc: Kumar Gala <galak@kernel.crashing.org>
Cc: Kim Phillips <kim.phillips@freescale.com>
Cc: Andy Fleming <afleming@gmail.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
It appears that with recent versions of GCC the explicit
"-mhard-float" command line option takes precedence over the
``asm(".gnu_attribute 4, 2");'' in the source file, so this no longer
helps to avoid the warnings we get when linking code that uses FP
instructions with other code that was built using soft-float.
We can remove the ".gnu_attribute" (which appears to carry no other
information, at least so far) from the object files, but we also have
to make sure we don't pull in the __gcc_qsub() and __gcc_qmul()
functions from the standard libgcc, as these would again "infect" our
linking. We copy this code from:
gcc-4.2.2/gcc/config/rs6000/darwin-ldouble.c
This old version was chosen because it was still available under a
compatible license (GCC v2+). The file was stripped down to the
needed parts, and reformatted so it passes checkpatch with only one
warning (do not add new typedefs).
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Kumar Gala <galak@kernel.crashing.org>
Cc: Stefan Roese <sr@denx.de>
Cc: Andy Fleming <afleming@gmail.com>
Cc: Kim Phillips <kim.phillips@freescale.com>
Tested-by: Stefan Roese <sr@denx.de>
Tested-by: Anatolij Gustschin <agust@denx.de>
commit f31a911fe (arm, post: add missing post_time_ms for arm)
enables get_ticks and get_tbclk for all arm based boards,
but kirkwood has currently no implementation for this. So
undefine this for kirkwood boards.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
cc: Heiko Schocher <hs@denx.de>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Valentin Longchamp <valentin.longchamp@keymile.com>
Signed-off-by: Christian Riesch <christian.riesch@omicron.at>
Cc: Tom Rini <trini@ti.com>
Cc: Heiko Schocher <hs@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Fix:
gdc.c: In function 'gdc_test_reg_one':
gdc.c:66:6: warning: variable 'ret' set but not used
[-Wunused-but-set-variable]
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Stefan Roese <sr@denx.de>
Acked-by: Stefan Roese <sr@denx.de>
Fix:
20001122-1.c: In function 'fpu_post_test_math1':
20001122-1.c:37:22: warning: variable 'p' set but not used [-Wunused-but-set-variable]
Signed-off-by: Stefan Roese <sr@denx.de>
Fix:
post.c: In function 'post_log':
post.c:425:7: warning: variable 'i' set but not used
[-Wunused-but-set-variable]
Signed-off-by: Wolfgang Denk <wd@denx.de>
This test is similar to the actual POST memory test but quicker and
far less complete. It checks the address and data lines and then only
tests some regularly placed sub regions of the RAM.
This can be useful when we want to test the RAM but we do not have enough
time to run the full memory test.
The POST memory test code was rearranged in order to avoid code duplication
between the two tests but the memory test functionnality remains the same.
Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Ackey-by: Mike Frysinger <vapier@gentoo.org>
This is needed for a further patch adding a new memory test.
Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Some boards have the environment variables defined in a slow EEPROM. post_run
accesses these environment variables to define which tests have to be run (in
post_get_flags). This is very slow before the code relocation on some boards
with a slow I2C EEPROM for environement variables.
This patch adds a config option to skip the fetching of the test flags in the
environment variables. The test flags assigned to the tests then only are the
ones statically defined for the test in post/tests.c.
Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
The current post_log_word in global data is currently split into 2x
16 bits: half for the test start, half for the test success.
Since we alredy have more than 16 POST tests defined and more could
be defined, this may result in an overflow and the post_output_backlog
would not work for the tests defined further of these 16 positions.
An additional field is added to global data so that we can now support up
to 32 (depending of architecture) tests. The post_log_word is only used
to record the start of the test and the new field post_log_res for the
test success (or failure). The post_output_backlog is for this change
also adapted.
Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
change bd->bi_memsize to gd->ram_size, as this is defined
on all archs, so this post test can used on none powerpc
archs too.
Signed-off-by: Heiko Schocher <hs@denx.de>
cc: Wolfgang Denk <hs@denx.de>
cc: Mike Frysinger <vapier@gentoo.org>
The post.c code is missing braces around the pass case, and as a
result, the diagnostic function will post both fail and pass for
a failed test. The reason for this bug is probably the incorrect
indentation used, so when reading the code it seems like there
are proper braces.
Indent the code to the correct depth and put proper braces around
the "else" branch of the "if" statement.
Signed-off-by: James Kosin <jkosin@intcomgrp.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This adds a simple flash test to automatically verify erasing,
writing, and reading of sectors. The code is based on existing
Blackfin tests but generalized for everyone to use.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Now that we have the generic GPIO layer, we can easily provide a common
implementation for the post_hotkeys_pressed() function based on it.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
If we built POST on PPC's that didn't enable CONFIG_SYS_POST_FPU we'd
get the following warning with newer toolchains:
powerpc-linux-gnu-ld: Warning: lib_powerpc/fpu/libpostpowerpcfpu.o
uses hard float, libpost.o uses soft float
We actually worked around this sometime ago with the following commit:
commit ce82ff0538
Author: Yuri Tikhonov <yur@emcraft.com>
Date: Sat Dec 20 14:54:21 2008 +0300
FPU POST: fix warnings when building with 2.18 binutils
However, this only took into effect if CONFIG_SYS_POST_FPU was enabled.
We can simply move the GNU_FPOST_ATTR out of the CONFIG_SYS_POST_FPU
ifdef block to address the issue.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This patch changes the PPC4xx ethernet POST loop test count from
currently 192 (256 - 64) to a default of 10. While doing this the max
frame size is increased. Each loop run uses a different frame size,
starting with a max of 1514 bytes, down to 64. The default loop
count of 10 can be overriden using CONFIG_SYS_POST_ETH_LOOPS in the
board config header.
The TEST_NUM loop has been removed as it was never used.
The main reason for this change is to reduce the boot time on boards
using this POST test, like the lwmon5 board. This change reduces the
boot time by about 600ms on the lwmon5 board.
Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
This patch fixes a problem in the Denali (440EPx) SDRAM ECC POST test.
When cache is enabled in the SDRAM area, the values written to SDRAM
need to be flushed from cache to SDRAM using the dcfb instruction.
Without this patch the POST ECC test failed. Now its working again on
platforms with cache enabled in SDRAM.
Signed-off-by: Stefan Roese <sr@denx.de>
Before this commit, weak symbols were not overridden by non-weak symbols
found in archive libraries when linking with recent versions of
binutils. As stated in the System V ABI, "the link editor does not
extract archive members to resolve undefined weak symbols".
This commit changes all Makefiles to use partial linking (ld -r) instead
of creating library archives, which forces all symbols to participate in
linking, allowing non-weak symbols to override weak symbols as intended.
This approach is also used by Linux, from which the gmake function
cmd_link_o_target (defined in config.mk and used in all Makefiles) is
inspired.
The name of each former library archive is preserved except for
extensions which change from ".a" to ".o". This commit updates
references accordingly where needed, in particular in some linker
scripts.
This commit reveals board configurations that exclude some features but
include source files that depend these disabled features in the build,
resulting in undefined symbols. Known such cases include:
- disabling CMD_NET but not CMD_NFS;
- enabling CONFIG_OF_LIBFDT but not CONFIG_QE.
Signed-off-by: Sebastien Carlier <sebastien.carlier@gmail.com>
If CONFIG_SYS_POST_I2C_ADDRS is not defined and CONFIG_SYS_POST_I2C
is activated, i2c_probe() is not called in the for statement,
because missing curly bracket.
Signed-off-by: Heiko Schocher <hs@denx.de>
By now, the majority of architectures have working relocation
support, so the few remaining architectures have become exceptions.
To make this more obvious, we make working relocation now the default
case, and flag the remaining cases with CONFIG_NEEDS_MANUAL_RELOC.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Tested-by: Heiko Schocher <hs@denx.de>
Tested-by: Reinhard Meyer <u-boot@emk-elektronik.de>
Commit 7e263ce "post/i2c: Clean up detection logic" added a "const"
qualifier to the declaration of i2c_addr_list[], missing the fact that
the list gets modified later in the code, which results in build
errors like these:
i2c.c: In function 'i2c_post_test':
i2c.c:88: error: assignment of read-only location
Remove the incorrect "const".
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Peter Tyser <ptyser@xes-inc.com>
Cc: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
Add the ability to not report an I2C POST error for a set of given I2C
addresses on bootup. This is useful for cases when a device may or may
not be present, and neither case is considered an error. For example:
- Some form factors such as XMC and Compact PCI Express have an I2C
EEPROM whose address changes based on geographical address. Eg
installed in one slot its EEPROM address is, 0x50, in another its
0x51, etc. This allows multiple devices to have their EEPROMs present
on the same I2C bus. Thus the I2C devices present for an XMC or
CPCIe card depend on if and where other cards are installed in the
same system.
- Some cards have optional I2C devices. Eg one hardware build
configuration has different I2C devices than another and software
can't determine if the optional device should be present or not.
- Some cards have optional daughtercards with I2C devices on them.
- I2C EEPROMs address range depends on their size. Its possible to
support differently size EEPROMs by only probing the EEPROM's base
address and ignoring the other addresses that are impacted by its
size.
A new CONFIG_SYS_POST_I2C_IGNORES define has been added which specifies
a list of I2C addresses for the I2C POST to ignore.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
According to the I2C specification device address 0 is the "general call
address", ie a broadcast address. The I2C specification states that the
format of a general call uses at least 2 bytes, which U-Boot's probing
routine does not adhere to.
Not probing device address 0 will prevent possible issues with devices
that accept general calls. Additionally, this change shouldn't reduce
POST coverage since each I2C device should still be accessed via its
own, unique address.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The logic previously used in the I2C post was a bit convoluted.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
- Clean up ifdeffery
- Update coding style
No functional change should have occurred.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Add CONFIG_POST_UART to implement a board specific UART POST test.
This is done since lwmon5 needs to set POST_ALWAYS to run this
test on each reboot. And we don't want to change the default
behavious of this this.
Signed-off-by: Stefan Roese <sr@denx.de>
