Rabin Vincent [Wed, 29 Oct 2014 22:21:37 +0000 (23:21 +0100)]
dlmalloc: ensure gd is set for early alloc
Attempting to run the sandbox leads to a segfault, because some dynamic
libraries (outside of u-boot) attempt to use malloc() to allocate memory
before u-boot's gd variable is initialized.
Check for gd not being NULL in the SYS_MALLOC_F_LEN handling, so that
malloc() doesn't crash when called at this point.
$ gdb -q --args ./u-boot
(gdb) r
Program received signal SIGSEGV, Segmentation fault.
0x0000000000412b9b in malloc (bytes=bytes@entry=37) at common/dlmalloc.c:2184
2184 if (!(gd->flags & GD_FLG_RELOC)) {
(gdb) p gd
$1 = (gd_t *) 0x0
(gdb) bt
#0 0x0000000000412b9b in malloc (bytes=bytes@entry=37) at common/dlmalloc.c:2184
#1 0x00007ffff75bf8e1 in set_binding_values (domainname=0x7ffff11f4f12 "libgpg-error", dirnamep=0x7fffffffe168, codesetp=0x0)
at bindtextdom.c:228
#2 0x00007ffff75bfb4c in set_binding_values (codesetp=0x0, dirnamep=0x7fffffffe168, domainname=<optimized out>) at bindtextdom.c:350
#3 __bindtextdomain (domainname=<optimized out>, dirname=0x7ffff11f4f00 "/usr/share/locale") at bindtextdom.c:348
#4 0x00007ffff11eca17 in ?? () from /lib/x86_64-linux-gnu/libgpg-error.so.0
#5 0x00007ffff7dea9fa in call_init (l=<optimized out>, argc=argc@entry=1, argv=argv@entry=0x7fffffffe208,
env=env@entry=0x7fffffffe218) at dl-init.c:78
#6 0x00007ffff7deaae3 in call_init (env=0x7fffffffe218, argv=0x7fffffffe208, argc=1, l=<optimized out>) at dl-init.c:36
#7 _dl_init (main_map=0x7ffff7ffe1a8, argc=1, argv=0x7fffffffe208, env=0x7fffffffe218) at dl-init.c:126
#8 0x00007ffff7ddd1ca in _dl_start_user () from /lib64/ld-linux-x86-64.so.2
Signed-off-by: Rabin Vincent <rabin@rab.in> Acked-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sat, 25 Oct 2014 11:38:31 +0000 (07:38 -0400)]
power_spi.c: Rewrite pmic_reg function
The pmic_spi_free function isn't ever used, and as the frameworks stand
today, cannot be, so remove it. Integrate the probe function into
pmic_reg as it's not really a "probe" today. Finally, add an err label
for the common failure cases.
Noam Camus [Wed, 22 Oct 2014 14:17:49 +0000 (17:17 +0300)]
image-fdt: boot_get_fdt() return value when no DTB exists
I believe that when no DTB is around we should return 1.
This why I fixed such scenarious to not return zero anymore.
Else kernel might get NULL pointer to DTB which doesn't exists.
Marek Vasut [Sat, 11 Oct 2014 16:42:58 +0000 (18:42 +0200)]
net: smc911x: Keep MAC programmed
Make sure to keep the MAC address programmed in the SMC911x ADDRH
and ADDRL registers. Linux can read those registers to determine
the MAC address on EEPROM-less configurations.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Tom Rini <trini@ti.com>
Masahiro Yamada [Thu, 6 Nov 2014 18:32:07 +0000 (03:32 +0900)]
generic-board: make compile-time noise for non-generic boards
Commit 0f605c1501f6 (Start the deprecation process for generic board)
added a run-time warning message. Let's be noisier for non-generic
boards to inform the dead line of the conversion.
This commit intentionally outputs a warning message to stdout.
We still have many unconverted boards. If we print the message
to stderr, MAKEALL and buildman will report tons of broken boards
and really important information will be buried.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Acked-by: Marek Vasut <marex@denx.de> Acked-by: Simon Glass <sjg@chromium.org> Cc: Tom Rini <trini@ti.com>
Igor Grinberg [Wed, 5 Nov 2014 12:25:35 +0000 (14:25 +0200)]
omap3: cm-t3517: add basic board support
CompuLab cm-t3517 is Computer on Module (CoM) based on AM3517 SoC.
Features: up to 256MB DDR2, up to 512MB NAND, USB hub, mUSB, WiFi, BT,
Analog audio codec, touch screen controller, LED.
Add basic support including:
LED, Serial console, NAND, MMC, GPIO, I2C, 256MB DRAM.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Igor Grinberg [Mon, 3 Nov 2014 09:32:23 +0000 (11:32 +0200)]
omap: hsmmc: assume cd gpio is active low
Switch the default CD GPIO polarity to active low.
The current hsmmc driver assumption that the CD GPIO is active high, but
in the real hardware, usually the opposite holds.
The usual SD card socket has a mechanical switch which is grounded as
soon as a card is inserted.
Of course there might be some board logic which inverts the signal, but
as far as current users are concerned, there is no such logic.
Current U-Boot users either not using the CD functionality, or have a
different way (e.g. external to SoC GPIO controller) for checking the
card presence.
This patch also brings the polarity assumption in line with the Linux
kernel and adds appropriate comments.
This patch also might spare issues once the TWL GPIO driver will be
converted to the DM.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il> Cc: Pantelis Antoniou <panto@antoniou-consulting.com> Cc: Dmitry Lifshitz <lifshitz@compulab.co.il> Reviewed-by: Tom Rini <trini@ti.com>
Igor Grinberg [Mon, 3 Nov 2014 09:32:22 +0000 (11:32 +0200)]
omap3: cm-t35: move get_board_serial() fallback
The fallback is used for cases when CONFIG_SERIAL_TAG defined, but the
eeprom is not used.
The fallback is useful for more than one CompuLab board, so move it to a
common location.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il> Reviewed-by: Tom Rini <trini@ti.com>
Igor Grinberg [Mon, 3 Nov 2014 09:32:20 +0000 (11:32 +0200)]
omap3: cm-t35: extract the splash code from board
The splash screen loading code can be reused by other compulab boards.
For now extract it to a common location for further reuse.
This also switches the splash code dependency from CONFIG_LCD to
CONFIG_SPLASH_SCREEN as it should normally be.
In addition this patch fixes the accidental dependency of the
get_board_mem_timings() function on CONFIG_LCD, by just moving the
splash code and leaving the above function intact.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il> Reviewed-by: Tom Rini <trini@ti.com>
Igor Grinberg [Wed, 5 Nov 2014 11:29:54 +0000 (13:29 +0200)]
am335x: make get_board_rev() function weak
Current get_board_rev() function returns a hard coded value which is
obviously incorrect for the majority of boards.
Allow boards to provide a correct implementation by making this
function weak.
In addition open code the trivial and useless BOARD_REV_ID define and
adjust the comment.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il> Cc: Tom Rini <trini@ti.com>
keystone2: config: align names of images with MCSDK
The Multicore Software Development Kit (MCSDK) provides foundational
software for TI KeyStone II device platforms. It's supposed to be used
with uboot, and it's convenient to have the same names for images, so
correct environment image names according to the last MCSDK3.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@ti.com> Reviewed-by: Tom Rini <trini@ti.com>
keystone2: config: restructure handling of default env settings
Currently to customize env for various ks2 boards, individual
variables such as NAME_FS etc are defined and included in the
common config.h to define CONFIG_EXTRA_ENV_SETTINGS. This
doesn't scale well if a variable is not applicable on a
specific board. Using this scheme, we have to define variables
with empty value and it's ugly. Instead, to allow board specific
customization of default env variable, define a common
CONFIG_EXTRA_ENV_KS2_SETTINGS for all common variables and define
board specific variables in individual board specific config.h
using CONFIG_EXTRA_ENV_KS2_BOARD_SETTINGS. Use the common and board
specific variables to define CONFIG_EXTRA_ENV_SETTINGS. This way
more variables can be added in future for individual boards
without affecting the other config.h files.
Signed-off-by: Murali Karicheri <m-karicheri2@ti.com> Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@ti.com> Reviewed-by: Tom Rini <trini@ti.com>
Hao Zhang [Thu, 30 Oct 2014 16:59:43 +0000 (18:59 +0200)]
net: phy: marvell: add errata w/a for 88E151* chips
As per Marvell Release Notes - Alaska 88E1510/88E1518/88E1512/88E1514
Rev A0, Errata Section 3.1 Marvell PHY has an errata which requires
that certain registers get written in order to restart
autonegotiation.
Signed-off-by: Hao Zhang <hzhang@ti.com> Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@ti.com> Reviewed-by: Stefan Roese <sr@denx.de>
Khoronzhuk, Ivan [Wed, 29 Oct 2014 11:09:32 +0000 (13:09 +0200)]
net: keystone_serdes: add keystone K2L SoC support
Keystone2 Lamar SoC uses the same keystone SerDes driver.
All Keystone2 EVM boards currently use SerDes driver, so move
CONFIG_TI_KEYSTONE_SERDES to common configuration file.
Acked-by: Murali Karicheri <m-karicheri2@ti.com> Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@ti.com>
Nikita Kiryanov [Wed, 29 Oct 2014 15:56:22 +0000 (17:56 +0200)]
arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo
The 1GB DRAM configuration on mx6 solo uses 2 chip selects, but
the code tests 1GB DRAM configuration as if it is all present on one
chip select, and thus cannot see the full range of available memory.
Refactor the check to detect 1GB DRAM correctly.
Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il> Cc: Igor Grinberg <grinberg@compulab.co.il> Cc: Stefano Babic <sbabic@denx.de> Acked-by: Igor Grinberg <grinberg@compulab.co.il>
Simon Glass [Fri, 31 Oct 2014 02:25:49 +0000 (20:25 -0600)]
dm: sunxi: Modify the GPIO driver to support driver model
This adds driver model support to the sunxi GPIO driver, using the device
tree to trigger binding of the driver. The driver will still operate
without driver model too.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Fri, 31 Oct 2014 02:25:48 +0000 (20:25 -0600)]
dm: sunxi: Make sure that GPIOs are requested
The scsi_init() function uses a GPIO so should request it. There is no
way to return an error here, and the request may be made multiple times,
so just ignore errors for now.
Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Fri, 31 Oct 2014 02:25:47 +0000 (20:25 -0600)]
dm: sunxi: Add pinmux functions which take a bank parameter
With driver model we will have access to a bank pointer, so we want to
use it rather than converting back to a number, and then back to a
bank pointer. Add functions to provide this feature.
Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Fri, 31 Oct 2014 02:25:46 +0000 (20:25 -0600)]
dm: sunxi: Add a new config for an FDT-based pcDuino3
For now we won't want to mess with the existing configurations. Create a
new one which will enable device tree and driver model. Note that this
brings the device tree binary into u-boot-sunxi-with-spl.bin.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Ian Campbell [Fri, 24 Oct 2014 20:20:49 +0000 (21:20 +0100)]
sunxi: kconfig: Introduce CONFIG_TARGET_<BOARD>
This was done automatically with the following bits of scripting.
The Kconfig choice content was generated with this script snippet:
for i in $(git grep -l CONFIG_ARCH_SUNXI configs/*) ; do
TARGET=$(sed -n -e 's/CONFIG_SYS_EXTRA_OPTIONS="\([^,"]\+\).*/\1/p' $i);
MACH=$(sed -n -e 's/.*CONFIG_\(MACH_SUN.I\)=./\1/p' $i)
echo "config TARGET_$TARGET"
echo " bool \"$TARGET\""
echo " depends on $MACH"
echo
done
defconfigs were updated with a sed script (t):
# Extract board from first entry of CONFIG_SYS_EXTRA_OPTIONS,
/^CONFIG_SYS_EXTRA_OPTIONS/ {
s/^\(CONFIG_SYS_EXTRA_OPTIONS="\)\([^,"]\+\),\?\(.*\)/\1\3\nCONFIG_TARGET_\2=y/;
# Print and delete first line (CONFIG_SYS_EXTRA_OPTIONS), leaving
# CONFIG_TARGET_<BOARD> in pattern space
P;D;
};
# Move CONFIG_TARGET_<BOARD> to hold space
/^CONFIG_TARGET/{h;n}
# Print CONFIG_TARGET_<BOARD> after CONFIG_MACH_<SOC> in either SPL or
# normal mode.
/^CONFIG_MACH/{p;g;p;n};
/^\+S:CONFIG_MACH/{p;g;s/^CONFIG_TARGET/+S:&/;p;n};
# Print any remaining lines normally
p;
Run as:
sed -i -n -f t $(git grep -l CONFIG_ARCH_SUNXI configs/*)
and then manually removing the one instance of CONFIG_SYS_EXTRA_OPTIONS="" from
Colombus_defconfig
board/sunxi/Makefile was updated with:
sed -e 's/^\(obj-\$(CONFIG_\)\(.*\)\().*+= dram_.*\)/\1TARGET_\2\3/g' board/sunxi/Makefile
and manually retabbing a few lines to line up again.
The board descriptions could certainly be improved.
Signed-off-by: Ian Campbell <ijc@hellion.org.uk> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Ian Campbell [Fri, 24 Oct 2014 20:20:47 +0000 (21:20 +0100)]
sunxi: Use CONFIG_MACH_SUN?I from Kconfig instead of CONFIG_SUN?I
Mostly automatic with:
sed -i -e 's/CONFIG_\(SUN[45678]I\)/CONFIG_MACH_\1/g' $(git grep -l CONFIG_SUN[45678]I)
followed by removing the relevant #defines from include/configs/sun?i.h by
hand.
Signed-off-by: Ian Campbell <ijc@hellion.org.uk> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Ian Campbell [Fri, 24 Oct 2014 20:20:46 +0000 (21:20 +0100)]
sunxi: Kconfig: Make SPL_FEL a toplevel Kconfig option
It's unfortunate that this needs to be present in both .config and spl/.config
since it makes it slightly hard to enable FEL mode for a regular defconfig. It
can be done with:
echo CONFIG_SPL_FEL=y >> .config
echo CONFIG_SPL_FEL=y >> spl/.config
Ideally only one of those would be needed.
Signed-off-by: Ian Campbell <ijc@hellion.org.uk> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Ian Campbell [Fri, 24 Oct 2014 20:20:45 +0000 (21:20 +0100)]
sunxi: kconfig: Rename TARGET_SUN[45678]I to MACH_SUN[45678]I.
Now we have CONFIG_ARCH_SUNXI as the toplevel, CONFIG_MACH_SUN[45678]I as the
per-SoC option and leave CONFIG_TARGET_BLAH free for individual boards in the
future.
Chen-Yu Tsai [Fri, 24 Oct 2014 08:07:25 +0000 (16:07 +0800)]
ARM: sunxi: Fix Ippo-q8h-v5 defconfig filename
The defconfig for Ippo-q8h-v5 was incorrectly committed as
"Ippo_q8h". This patch renames it correctly, as well as
correct a typo in my name in the maintainers listing.
Signed-off-by: Chen-Yu Tsai <wens@csie.org> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Dirk Eibach [Mon, 3 Nov 2014 09:56:31 +0000 (10:56 +0100)]
ppc: Fix ppc4xx CONFIG_SYS_GENERIC_BOARD
I realized that with v2014.10 u-boot is not starting up on
ppc4xx boards with CONFIG_SYS_GENERIC_BOARD set.
I bisected it down to this: d54d7eb support blackfin board initialization in generic board_f
With d29437a ppc: Make ppc4xx ready for CONFIG_SYS_GENERIC_BOARD
I set gd to a defined value for ppc4xx in cpu_init_f().
I did not realize that the gd struct has also to be memset()
to zero at this point.
But at least commit d54d7eb assumes it is:
@@ -516,11 +528,13 @@ static int reserve_malloc(void)
/* (permanently) allocate a Board Info struct */
static int reserve_board(void)
{
- gd->start_addr_sp -= sizeof(bd_t);
- gd->bd = (bd_t *)map_sysmem(gd->start_addr_sp, sizeof(bd_t));
- memset(gd->bd, '\0', sizeof(bd_t));
- debug("Reserving %zu Bytes for Board Info at: %08lx\n",
- sizeof(bd_t), gd->start_addr_sp);
+ if (!gd->bd) {
+ gd->start_addr_sp -= sizeof(bd_t);
+ gd->bd = (bd_t *)map_sysmem(gd->start_addr_sp, sizeof(bd_t));
+ memset(gd->bd, '\0', sizeof(bd_t));
+ debug("Reserving %zu Bytes for Board Info at: %08lx\n",
+ sizeof(bd_t), gd->start_addr_sp);
+ }
return 0;
}
#endif
This might also be an issue on other architectures, so maintainers should
check.
Signed-off-by: Dirk Eibach <dirk.eibach@gdsys.cc> Signed-off-by: Stefan Roese <sr@denx.de>
Masahiro Yamada [Thu, 30 Oct 2014 06:30:58 +0000 (15:30 +0900)]
sparc: move CONFIG_SYS_TEXT_BASE to Kconfig
Defining CONFIG_SYS_TEXT_BASE in config.mk is very old style.
Create CONFIG_SYS_TEXT_BASE option in Kconfig, but let it
depend on CONFIG_SPARC because we do not want to disturb
the other architectures that still define CONFIG_SYS_TEXT_BASE
in their header files.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Cc: Daniel Hellstrom <daniel@gaisler.com>
Hans de Goede [Wed, 24 Sep 2014 12:06:11 +0000 (14:06 +0200)]
usb: kbd: Add (optional) support for using an interrupt queue for polling
Waiting an interrupt packet to complete in usb_kbd_poll_for_event, causes
a 40 ms latency for each call to usb_kbd_testc, which is undesirable.
Using control messages leads to lower (but still not 0) latency, but some
devices do not work well with control messages (e.g. my kvm behaves funny
with them).
This commit adds support for using the int_queue mechanism which at least
the ehci-hcd driver supports. This allows polling with 0 latency, while
using interrupt packets.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Hans de Goede [Sat, 20 Sep 2014 15:03:52 +0000 (17:03 +0200)]
usb: Remove unnecessary portnr lookup from usb_new_device
If the device has a parent, it is instantiated from usb_hub_port_connect_change
and the portnr is right there in dev->portnr, so there is no need for this
whole dance to look it up.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
net: sh-eth: Remove definition of RX_DESC_SIZE and TX_DESC_SIZE
RX_DESC_SIZE and TX_DESC_SIZE mean the size of the data descriptor for sh-eth.
We can acquire this in sizeof. It is not necessary to define these in define.
projects / u-boot / log