project/bcm63xx/u-boot.git
12 years agomx35pdk: Remove CONFIG_SYS_I2C_SLAVE definition
Fabio Estevam [Sat, 2 Jun 2012 21:53:26 +0000 (21:53 +0000)]
mx35pdk: Remove CONFIG_SYS_I2C_SLAVE definition

According to include/i2c.h:

"/*
 * Many boards/controllers/drivers don't support an I2C slave interface so
 * provide a default slave address for them for use in common code.  A real
 * value for CONFIG_SYS_I2C_SLAVE should be defined for any board which does
 * support a slave interface.
 */
#ifndef CONFIG_SYS_I2C_SLAVE
#define CONFIG_SYS_I2C_SLAVE 0xfe
#endif
"

As the mxc_i2c driver does not support slave mode, there is no need
to define CONFIG_SYS_I2C_SLAVE in i.MX board file.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agoimx31_phycore: Remove CONFIG_SYS_I2C_SLAVE definition
Fabio Estevam [Sat, 2 Jun 2012 21:53:25 +0000 (21:53 +0000)]
imx31_phycore: Remove CONFIG_SYS_I2C_SLAVE definition

According to include/i2c.h:

"/*
 * Many boards/controllers/drivers don't support an I2C slave interface so
 * provide a default slave address for them for use in common code.  A real
 * value for CONFIG_SYS_I2C_SLAVE should be defined for any board which does
 * support a slave interface.
 */
#ifndef CONFIG_SYS_I2C_SLAVE
#define CONFIG_SYS_I2C_SLAVE 0xfe
#endif
"

As the mxc_i2c driver does not support slave mode, there is no need
to define CONFIG_SYS_I2C_SLAVE in i.MX board file.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53ard: Remove unused CONFIG_MII_GASKET
Fabio Estevam [Thu, 7 Jun 2012 08:05:41 +0000 (08:05 +0000)]
mx53ard: Remove unused CONFIG_MII_GASKET

CONFIG_MII_GASKET is not defined anywhere, so remove it.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx6: Avoid writing to read-only bits in imximage.cfg
Vikram Narayanan [Tue, 12 Jun 2012 04:50:33 +0000 (04:50 +0000)]
mx6: Avoid writing to read-only bits in imximage.cfg

If in case this is valid according to the latest datasheet, ignore this patch.
Acked-by: Marek Vasut <marex@denx.de>
12 years agom28evk: use same notation to alloc the 128kB stack
Otavio Salvador [Sat, 12 May 2012 13:40:11 +0000 (13:40 +0000)]
m28evk: use same notation to alloc the 128kB stack

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Wolfgang Denk <wd@denx.de>
12 years agom28evk: drop duplicated definition of CONFIG_OF_LIBFDT
Otavio Salvador [Tue, 15 May 2012 10:59:30 +0000 (10:59 +0000)]
m28evk: drop duplicated definition of CONFIG_OF_LIBFDT

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Cc: Marek Vasut <marex@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Acked-by: Marek Vasut <marex@denx.de>
12 years agomx28evk: fix board config include guardian macro name
Otavio Salvador [Sat, 12 May 2012 13:40:16 +0000 (13:40 +0000)]
mx28evk: fix board config include guardian macro name

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx28evk: ensure command definition is in alphabetic order
Otavio Salvador [Sat, 12 May 2012 13:40:13 +0000 (13:40 +0000)]
mx28evk: ensure command definition is in alphabetic order

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agom28evk: fix board config include guardian macro name
Otavio Salvador [Sat, 12 May 2012 13:40:15 +0000 (13:40 +0000)]
m28evk: fix board config include guardian macro name

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Marek Vasut <marex@denx.de>
12 years agoMX28: Fix a typo in mx28_reg_8 macro
Otavio Salvador [Fri, 11 May 2012 05:37:59 +0000 (05:37 +0000)]
MX28: Fix a typo in mx28_reg_8 macro

The macro mistakenly referred to 32bit struct instead of 8bit one.

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Marek Vasut <marex@denx.de>
Acked-by: Marek Vasut <marex@denx.de>
12 years agomx6qsabrelite: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:40:03 +0000 (07:40 +0000)]
mx6qsabrelite: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Jason Liu <r64343@freescale.com>
12 years agomx6qarm2: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:40:02 +0000 (07:40 +0000)]
mx6qarm2: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Jason Liu <r64343@freescale.com>
12 years agoima3-mx53: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:40:01 +0000 (07:40 +0000)]
ima3-mx53: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agovision2: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:40:00 +0000 (07:40 +0000)]
vision2: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agoefikamx: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:59 +0000 (07:39 +0000)]
efikamx: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx51evk: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:58 +0000 (07:39 +0000)]
mx51evk: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53smd: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:57 +0000 (07:39 +0000)]
mx53smd: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53ard: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:56 +0000 (07:39 +0000)]
mx53ard: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53evk: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:55 +0000 (07:39 +0000)]
mx53evk: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53loco: Remove unused SOBJS
Fabio Estevam [Tue, 29 May 2012 07:39:54 +0000 (07:39 +0000)]
mx53loco: Remove unused SOBJS

There is no .S file in this directory, so just remove SOBJS.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Marek Vasut <marex@denx.de>
12 years agomx53: Fix mask for SATA reference clock
Fabio Estevam [Fri, 11 May 2012 14:39:11 +0000 (14:39 +0000)]
mx53: Fix mask for SATA reference clock

SATA_ALT_REF_CLK field corresponds to bits 1 and 2 of offset 0x180c.

Fix the mask for these bits.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53loco: Fix revision of Dialog boards
Fabio Estevam [Tue, 29 May 2012 05:54:39 +0000 (05:54 +0000)]
mx53loco: Fix revision of Dialog boards

Original code was assuming that the fuse revision version for all mx53loco boards
based on Dialog PMIC was the same, which is not the case.

Force the revision of all Dialog-based boards to 0.

This fixes a kernel crash when PMIC is accessed in the 2.6.35 kernel
for Dialog rev E boards.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agoi.mx: i.mx6x: NO_MUX_I/NO_PAD_I not set correctly
Jason Liu [Wed, 11 Apr 2012 01:18:02 +0000 (01:18 +0000)]
i.mx: i.mx6x: NO_MUX_I/NO_PAD_I not set correctly

If one PAD does not have mux or pad config register, we need
set the NO_MUX_I/NO_PAD_I to 0, the old value is not correct

Signed-off-by: Jason Liu <jason.hui@linaro.org>
CC: Stefano Babic <sbabic@denx.de>
Acked-by: Stefano Babic <sbabic@denx.de>
12 years agoEXYNOS: SMDK5250: Enable the pinmux setup
Rajeshwari Shinde [Wed, 6 Jun 2012 19:54:30 +0000 (19:54 +0000)]
EXYNOS: SMDK5250: Enable the pinmux setup

Use the pinmux configuration function for SMDK5250.

Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Chander Kashyap <chander.kashyap@linaro.org>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
12 years agoEXYNOS5: PINMUX: Added default pinumx settings
Rajeshwari Shinde [Wed, 6 Jun 2012 19:54:29 +0000 (19:54 +0000)]
EXYNOS5: PINMUX: Added default pinumx settings

This patch performs the pinmux configuration in a common file.
As of now only EXYNOS5 pinmux for SDMMC, UART and Ethernet is
supported.

Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
Signed-off-by: Che-Liang Chiou <clchiou@chromium.org>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Chander Kashyap <chander.kashyap@linaro.org>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
12 years agomisc:pmic:trats: Correct procedure of enabling/disabling USB regulators
Łukasz Majewski [Wed, 25 Apr 2012 23:30:18 +0000 (23:30 +0000)]
misc:pmic:trats: Correct procedure of enabling/disabling USB regulators

In the MAX8997, LDO regulators needs to preserve previously set voltage
values.

Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
12 years agoTRATS: initialize panel_info data structure in board file
Donghwa Lee [Wed, 25 Apr 2012 13:29:39 +0000 (13:29 +0000)]
TRATS: initialize panel_info data structure in board file

panel_info data structure is gloable variable, so, I have initialized it
in board file. If it is initialized in init_panel_info() like existing,
it can't be used in drv_lcd_init() in common/lcd.c because
init_panel_info() is called after drv_lcd_init().

Signed-off-by: Donghwa Lee <dh09.lee@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
12 years agoExynos: fix cpuinfo and cpu detecting
Minkyu Kang [Thu, 26 Apr 2012 06:48:32 +0000 (15:48 +0900)]
Exynos: fix cpuinfo and cpu detecting

Since Exynos architecture have new SoCs,
need to fix cpuinfo correctly.

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Tested-by: Jaehoon Chung <jh80.chung@samsung.com>
Cc: Chander Kashyap <chander.kashyap@linaro.org>
12 years agocm-t35: print PCB revision information
Nikita Kiryanov [Thu, 24 May 2012 04:01:24 +0000 (04:01 +0000)]
cm-t35: print PCB revision information

Buffer the PCB revision to avoid multiple eeprom accesses
for the same data and print it as a part of board information.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tom Rini <trini@ti.com>
12 years agocm-t35: fix legacy board revision representation
Nikita Kiryanov [Thu, 24 May 2012 04:01:23 +0000 (04:01 +0000)]
cm-t35: fix legacy board revision representation

Legacy eeprom layout represents the revision number syntactically
(i.e. revision 1.00 is written as 0x100). This is inconsistent with
the representation in newer layouts, where it is defined semantically
(i.e. 0x64).

This patch fixes the issue by replacing the syntactic representation
with the semantic one.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
12 years agocm-t35: fix incorrect BOARD_REV_SIZE value
Nikita Kiryanov [Thu, 24 May 2012 04:01:22 +0000 (04:01 +0000)]
cm-t35: fix incorrect BOARD_REV_SIZE value

Non-legacy layouts have an extended revision field,
but only the first 2 bytes are the PCB revision.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
12 years agocm-t35: reduce the environment size
Igor Grinberg [Thu, 24 May 2012 04:01:21 +0000 (04:01 +0000)]
cm-t35: reduce the environment size

Reduce the environment size (128KB => 16KB) to improve the environment
operations time (e.g. reading, ecc calculation).
Also, remove the unused CONFIG_SYS_ENV_SECT_SIZE.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
12 years agoARM: OMAP4: Correct the lpddr2 io settings register value.
SRICHARAN R [Thu, 24 May 2012 00:30:25 +0000 (00:30 +0000)]
ARM: OMAP4: Correct the lpddr2 io settings register value.

To meet certain timing requirements on the lpddr2 cmd and data phy
interfaces ,lpddr iopads have to be configured as differential buffers
and a Vref has to be internally generated and provided to these buffers.

Correcting the above settings here.

Signed-off-by: R Sricharan <r.sricharan@ti.com>
12 years agoOMAP5: Change voltages for omap5432
Lokesh Vutla [Tue, 22 May 2012 00:03:27 +0000 (00:03 +0000)]
OMAP5: Change voltages for omap5432

Change voltages for OMAP5432

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: DPLL core lock for OMAP5432
Lokesh Vutla [Tue, 22 May 2012 00:03:26 +0000 (00:03 +0000)]
OMAP5: DPLL core lock for OMAP5432

No need to Unlock DPLL initially.
DDR3 can work at normal OPP from initialozation

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: EMIF: Add support for DDR3 device
Lokesh Vutla [Tue, 22 May 2012 00:03:25 +0000 (00:03 +0000)]
OMAP5: EMIF: Add support for DDR3 device

In OMAP5432 EMIF controlller supports DDR3 device.
This patch adds support for ddr3 device intialization and configuration.
Initialization sequence is done as specified in JEDEC specs.
This also adds support for ddr3 leveling.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: ADD precalculated timings for ddr3
Lokesh Vutla [Tue, 22 May 2012 00:03:24 +0000 (00:03 +0000)]
OMAP5: ADD precalculated timings for ddr3

Adding precalculated timings for ddr3 with 1cs
adding required registers for ddr3

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: Configure the io settings for omap5432 uevm board
Lokesh Vutla [Tue, 22 May 2012 00:03:23 +0000 (00:03 +0000)]
OMAP5: Configure the io settings for omap5432 uevm board

This patch adds the IO settings required for OMAP5432 uevm's DDR3 pads

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: ADD chip detection for OMAP5432 SOC
Lokesh Vutla [Tue, 22 May 2012 00:03:22 +0000 (00:03 +0000)]
OMAP5: ADD chip detection for OMAP5432 SOC

This patch adds chip detection for OMAP5432

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoOMAP5: Adding correct Control id code for OMAP5430
Lokesh Vutla [Tue, 22 May 2012 00:03:21 +0000 (00:03 +0000)]
OMAP5: Adding correct Control id code for OMAP5430

Control id code for omap5430 ES1.0 is hard coded with a wrong value.
This patch corrects the value

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
12 years agoomap3evm: Add CONFIG_CMD_ASKENV
Tom Rini [Mon, 21 May 2012 11:20:47 +0000 (11:20 +0000)]
omap3evm: Add CONFIG_CMD_ASKENV

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoomap3_beagle: Add CONFIG_CMD_ASKENV
Tom Rini [Mon, 21 May 2012 11:20:46 +0000 (11:20 +0000)]
omap3_beagle: Add CONFIG_CMD_ASKENV

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoomap3evm: Set BOOTDELAY to 3
Tom Rini [Mon, 21 May 2012 11:20:45 +0000 (11:20 +0000)]
omap3evm: Set BOOTDELAY to 3

We reduce the bootdelay from 10s to 3s to give users a short but usable
window to interrupt the boot process if needed.

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoOMAP3 Beagle: Set BOOTDELAY to 3
Tom Rini [Mon, 21 May 2012 11:20:44 +0000 (11:20 +0000)]
OMAP3 Beagle: Set BOOTDELAY to 3

We change the bootdelay to give users a little bit longer to break in if
needed.

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoomap24xx_i2c: Add AM33XX support
Tom Rini [Mon, 21 May 2012 06:46:33 +0000 (06:46 +0000)]
omap24xx_i2c: Add AM33XX support

The same places that check for CONFIG_OMAP44XX need to check for
CONFIG_AM33XX as we share the same i2c block.

Cc: Heiko Schocher <hs@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Heiko Schocher <hs@denx.de>
12 years agoam33xx: Fix i2c sampling rate typo
Tom Rini [Mon, 21 May 2012 06:46:32 +0000 (06:46 +0000)]
am33xx: Fix i2c sampling rate typo

Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Heiko Schocher <hs@denx.de>
12 years agoam33xx: Fill in more cm_wkuppll / cm_perpll
Tom Rini [Mon, 21 May 2012 06:46:31 +0000 (06:46 +0000)]
am33xx: Fill in more cm_wkuppll / cm_perpll

Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Heiko Schocher <hs@denx.de>
12 years agoam335x: Correct i2c sysc offset
Tom Rini [Mon, 21 May 2012 06:46:30 +0000 (06:46 +0000)]
am335x: Correct i2c sysc offset

Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Heiko Schocher <hs@denx.de>
12 years agoRevert "I2C: OMAP: detect more devices when probing an i2c bus"
Tom Rini [Mon, 21 May 2012 06:46:29 +0000 (06:46 +0000)]
Revert "I2C: OMAP: detect more devices when probing an i2c bus"

This reverts commit 0e57968a215d1b9d271f3fa5bebeddeaea0c8075.

The short version of the original commit is that some i2c devices cannot
be probed via read as they NAK the first cycle, so try and probe via a
write that we abort before it writes to the device.  This however is not
allowed by the TRM for any of these parts.  The section on I2C_CON
(table 17-35 I2C_CON for am/dm37x for example) says you must not change
the register while STT has been set.  On these parts, the unpredictable
behavior that the chip exhibits is not problematic.  On OMAP4 however it
results in the chip being in a bad state:
Panda # i2c probe
Valid chip addresses: 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12
13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20 21 22 23 24 25 26 27 28 29 2A
2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40 41 42
43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A
5B 5C 5D 5E 5F 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72
73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
Panda # i2c md 50 0
timed out in wait_for_pin: I2C_STAT=0
I2C read: I/O error
Error reading the chip.

We must revert the original behavior to bring probe back into line with
the TRM.

Cc: Nick Thompson <nick.thompson@ge.com>
Cc: Heiko Schocher <hs@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Heiko Schocher <hs@denx.de>
12 years agoDaVinci: fix ddr2 vtp i/o calibration
Troy Kisky [Fri, 18 May 2012 13:21:59 +0000 (13:21 +0000)]
DaVinci: fix ddr2 vtp i/o calibration

Previously, only the low 5 bits (NCH) were being transfered
from DDRVTPR to DDRVTPIOCR, the bits 5-9 where zeroed.

VTP_RECAL should be bit 15, not 18.

The only mainline board affected by this change is davinci_sonata.
The other Davinci boards define CONFIG_SKIP_LOWLEVEL_INIT.

However, if the program that loads u-boot on these boards
copied the code from u-boot, they will need fixed as well.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Please get tested by acks before applying, where tested by
means an overnight memory test.

Thanks
Troy

12 years agoARM: OMAP5: Correct the DRAM_ADDR_SPACE_END macro.
SRICHARAN R [Thu, 17 May 2012 00:12:09 +0000 (00:12 +0000)]
ARM: OMAP5: Correct the DRAM_ADDR_SPACE_END macro.

OMAP5 evm board has 2GB of memory. So correct the
macro to take in to account of the full dram size.

Signed-off-by: R Sricharan <r.sricharan@ti.com>
12 years agoARM: OMAP5: Align memory used for testing to the power of 2
SRICHARAN R [Thu, 17 May 2012 00:12:08 +0000 (00:12 +0000)]
ARM: OMAP5: Align memory used for testing to the power of 2

get_ram_size checks the given memory range for valid ram,
but expects the size of memory to be aligned to the power
of 2. In case of OMAP5 evm board the memory available is
2GB - 16MB(used for TRAP section) = 2032MB.

So always ensure that the size of memory used for testing is
aligned to the power of 2.

Signed-off-by: R Sricharan <r.sricharan@ti.com>
12 years agoARM: OMAP5: dmm: Create a tiler trap section.
SRICHARAN R [Thu, 17 May 2012 00:12:07 +0000 (00:12 +0000)]
ARM: OMAP5: dmm: Create a tiler trap section.

The unmapped entries in tiler space are set with
values 0xFF. So creating a DMM section of
size 16MB at 0xFF000000 with ADDRSPACE set to 0x2.

This way all the unmapped entry accesses to tiler
will be trapped by the EMIF and a error response
is sent to the L3 interconnect. L3 errors are
inturn reported to MPU.

Note that here the tiler trap section is overlapping
with the actual ddr physical space and we lose 16MB
out of the total 2GB.

Signed-off-by: R Sricharan <r.sricharan@ti.com>
12 years agoARM: OMAP4+: dmm: Take care of overlapping dmm and trap sections.
SRICHARAN R [Thu, 17 May 2012 00:12:06 +0000 (00:12 +0000)]
ARM: OMAP4+: dmm: Take care of overlapping dmm and trap sections.

The DMM sections can be overlapping with each other, with
sections 3 to 0 having the highest to lowest priority in that
order. There could also be a section that is used trap the
unmapped Tiler entries and this trap section could be
overlapping with the actual sdram area.

So take care of the above scenarios while calculating the
size of the actual ram.

Signed-off-by: R Sricharan <r.sricharan@ti.com>
12 years agoarm,davinci: update for enbw_cmc board
Heiko Schocher [Mon, 14 May 2012 20:24:14 +0000 (20:24 +0000)]
arm,davinci: update for enbw_cmc board

- change gpio pin settings:

  - gpio pin 6[13] (PLC reset) default value low
  - gpio pin 6[0] (TPM reset) default value low
  - 4 new GPIO pins
      pin  i/o   name
    - 3[9] input Board Type
    - 2[7] input HW-ID0
    - 2[6] input HW-ID1
    - 2[3] input HW-ID2

- read board type and hw id from gpio pins on the enbw_cmc board,
  and use board type for setting up different gpio pin settings.

- do not pass "davinci_mmc.use_dma=0" to linux, as MMC now
  works with DMA.

- update logbuf support:
  store post word in RTC scratch register

- add support for configuring KSZ8864RMN switch through
  a config file on u-boot startup. For more infos see:
  doc/README.switch_config

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Tom Rini <tom.rini@gmail.com>
Cc: Christian Riesch <christian.riesch@omicron.at>
Cc: Sandeep Paulraj <s-paulraj@ti.com>
12 years agoam33xx: Do not call init_timer twice
Tom Rini [Mon, 14 May 2012 12:38:18 +0000 (12:38 +0000)]
am33xx: Do not call init_timer twice

We do not need to call init_timer both in SPL and U-Boot itself, just
SPL needs to initialize the timer.

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoomap4/5: Use CPUDIR for .lds script
Thomas Weber [Mon, 14 May 2012 10:28:54 +0000 (10:28 +0000)]
omap4/5: Use CPUDIR for .lds script

Signed-off-by: Thomas Weber <thomas@tomweber.eu>
12 years agotegra: trimslice: fix a couple typos
Stephen Warren [Wed, 30 May 2012 06:45:50 +0000 (06:45 +0000)]
tegra: trimslice: fix a couple typos

Fix the .dts file USB unit addresses not to duplicate each-other.

Fix the board name string to indicate the vendor is Compulab not NVIDIA.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
12 years agoarm: Tegra: Use ODMDATA from BCT in IRAM
Tom Warren [Wed, 30 May 2012 21:06:09 +0000 (14:06 -0700)]
arm: Tegra: Use ODMDATA from BCT in IRAM

Walk the BIT and BCT to find the ODMDATA word in the
CustomerData field and put it into Scratch20 reg for
use by kernel, etc.

Built all Tegra builds OK; Booted on Seaboard and saw
ODMDATA in PMC scratch20 was the same as the value in my
burn-u-boot.sh file (0x300D8011). NOTE: All flash utilities
will have to specify the odmdata (nvflash --odmdata n) on
the command line or via a cfg file, or built in to their
BCT.

Signed-off-by: Tom Warren <twarren@nvidia.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
12 years agotegra: seaboard: disable SPI, move environment to eMMC
Stephen Warren [Thu, 24 May 2012 11:38:39 +0000 (11:38 +0000)]
tegra: seaboard: disable SPI, move environment to eMMC

The SPI hardware on Seaboard is too broken to use; it is muxed with the
console UART and requires evil interactions between the SPI and UART
drivers to work even partially. The current code in U-Boot is not
sufficient to make this work correctly; auto boot is aborted due to
corruption in the UART RX channel interrupting it.

Instead, move the environment to eMMC, at the end of the second boot
sector. This should not conflict with any other eMMC usage, irrespective
of whether the board boots from SPI, NAND, or eMMC: if U-Boot is stored
in eMMC, it will be stored well below this location. The kernel only
uses the general area of the eMMC once booted, not the boot sectors.

Boards that are derivatives of Seaboard don't have the muxing issue,
and should/could have a separate U-Boot configuration file that does
enable SPI if desired.

Alternatively, the environment could be stored in NAND flash, but we
currently have no driver for that controller.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: paz00: store environment in eMMC
Stephen Warren [Thu, 24 May 2012 11:38:38 +0000 (11:38 +0000)]
tegra: paz00: store environment in eMMC

Store the environment in eMMC, at the end of the second boot sector.
This should not conflict with any other eMMC usage: U-Boot is stored
well below this location, and the kernel only uses the general area
of the eMMC once booted, not the boot sectors.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: trimslice: store environment in SPI flash
Stephen Warren [Thu, 24 May 2012 11:38:37 +0000 (11:38 +0000)]
tegra: trimslice: store environment in SPI flash

The chosen flash offset matches Compulab's downstream U-Boot.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agogpio: tegra2: rename tegra2_gpio.* to tegra_gpio.*
Tom Warren [Tue, 22 May 2012 12:19:25 +0000 (12:19 +0000)]
gpio: tegra2: rename tegra2_gpio.* to tegra_gpio.*

In anticipation of Tegra3 support, continue removing/renaming
Tegra2-specific files. No functional changes (yet).
Updated copyrights to 2012.

Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agommc: tegra2: rename tegra2_mmc.* to tegra_mmc.*
Tom Warren [Tue, 22 May 2012 11:44:48 +0000 (11:44 +0000)]
mmc: tegra2: rename tegra2_mmc.* to tegra_mmc.*

In anticipation of Tegra3 support, continue removing/renaming
Tegra2-specific files. No functional changes (yet).
Updated copyrights to 2012.

Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agospi: tegra2: rename tegra2_spi.* to tegra_spi.*
Tom Warren [Tue, 22 May 2012 07:33:47 +0000 (07:33 +0000)]
spi: tegra2: rename tegra2_spi.* to tegra_spi.*

In anticipation of Tegra3 support, start removing/renaming
Tegra2-specific files. No functional changes (yet).
Also updated copyright to 2012.

Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: whistler: store environment in eMMC
Stephen Warren [Thu, 24 May 2012 11:38:36 +0000 (11:38 +0000)]
tegra: whistler: store environment in eMMC

Store the environment in eMMC, at the end of the second boot sector.
This should not conflict with any other eMMC usage: U-Boot is stored
well below this location, and the kernel only uses the general area
of the eMMC once booted, not the boot sectors.

Note: This assumes the user plugged the standard 8MB MoviNAND card into
J29/HSMMC/POP. If they didn't, the boot sector layout may be different.
However, use of that particular card is standard practice as far as I
know.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: ventana: store environment in eMMC
Stephen Warren [Thu, 24 May 2012 11:38:35 +0000 (11:38 +0000)]
tegra: ventana: store environment in eMMC

Store the environment in eMMC, at the end of the second boot sector.
This should not conflict with any other eMMC usage: U-Boot is stored
well below this location, and the kernel only uses the general area
of the eMMC once booted, not the boot sectors.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agosf: winbond: Add support for the Winbond W25Q80BL
Stephen Warren [Thu, 24 May 2012 11:38:34 +0000 (11:38 +0000)]
sf: winbond: Add support for the Winbond W25Q80BL

This chip is present on the Compulab TrimSlice.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agoenv_mmc: align buffers using ALLOC_CACHE_ALIGN_BUFFER
Stephen Warren [Thu, 24 May 2012 11:38:33 +0000 (11:38 +0000)]
env_mmc: align buffers using ALLOC_CACHE_ALIGN_BUFFER

This allows MMC drivers to perform cache flusing on the bufffers
without issue.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Cc: Andy Fleming <afleming@gmail.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: ventana: add own device tree, enable USB
Stephen Warren [Wed, 23 May 2012 07:46:15 +0000 (07:46 +0000)]
tegra: ventana: add own device tree, enable USB

Add a device tree for Ventana; the Seaboard file no longer represents
the HW present on Ventana.

Enable USB on Ventana.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: remove CONFIG_USB_ETHER_SMSC95XX from boards without it
Stephen Warren [Wed, 23 May 2012 07:29:29 +0000 (07:29 +0000)]
tegra: remove CONFIG_USB_ETHER_SMSC95XX from boards without it

The SMSC95xx series may exist either directly on a main board, or as a USB
to Ethernet dongle. However, dongles containing these chips are very rare.
Hence, remove this config option, except on Harmony where such a chip is
actually present on the board.

The asix option remains, since it's a popular chip, and I actively use a
dongle containing this.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
12 years agotegra: override compiler flags for low level init code
amartin@nvidia.com [Mon, 14 May 2012 13:14:39 +0000 (13:14 +0000)]
tegra: override compiler flags for low level init code

Override -march setting for tegra to -march=armv4t for files that are
necessary for low level init on tegra.

The recent change to use -march=armv7-a for armv7 caused a regression
on tegra because tegra starts boot on a arm7tdmi processor before
transferring control to the cortex-a9.  While still executing on the
arm7tdmi there are calls to getenv_ulong() and memset() that cause an
illegal instruction exception if compiled for armv7.

Signed-off-by: Allen Martin <amartin@nvidia.com>
Tested-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: Correct PLL access in ap20.c and clock.c
Simon Glass [Thu, 19 Apr 2012 08:04:39 +0000 (08:04 +0000)]
tegra: Correct PLL access in ap20.c and clock.c

Correct this warning seen by Albert:

ap20.c:44:18: warning: array subscript is above array bounds

There is a subtle bug here which currently causes no errors, but might
in future if people use PCI or the 32KHz clock. So take the opportunity
to correct the logic now.

Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: paz00: add device tree support
Stephen Warren [Mon, 21 May 2012 10:04:37 +0000 (10:04 +0000)]
tegra: paz00: add device tree support

... to enable USB host support, which enables Ethernet support.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: harmony: add device tree support
Stephen Warren [Mon, 21 May 2012 10:04:27 +0000 (10:04 +0000)]
tegra: harmony: add device tree support

... to enable USB host support, which enables Ethernet support.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: Compulab TrimSlice board support
Stephen Warren [Wed, 16 May 2012 13:54:07 +0000 (13:54 +0000)]
tegra: Compulab TrimSlice board support

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: add SDMMC1 on SDIO1 funcmux entry
Stephen Warren [Wed, 16 May 2012 13:54:06 +0000 (13:54 +0000)]
tegra: add SDMMC1 on SDIO1 funcmux entry

This will be used on TrimSlice.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: add SDIO1 funcmux entry for UARTA
Lucas Stach [Wed, 16 May 2012 08:21:02 +0000 (08:21 +0000)]
tegra: add SDIO1 funcmux entry for UARTA

This is based on top of:
tegra: add alternate UART1 funcmux entry
tegra: add UART1 on GPU funcmux entry

v2: remove enum change

Signed-off-by: Lucas Stach <dev@lynxeye.de>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
CC: Stephen Warren <swarren@wwwdotorg.org>
CC: Tom Warren <twarren@nvidia.com>
CC: Marek Vasut <marex@denx.de>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: sync SDIO1 pingroup enum name with TRM
Lucas Stach [Wed, 16 May 2012 08:21:01 +0000 (08:21 +0000)]
tegra: sync SDIO1 pingroup enum name with TRM

Signed-off-by: Lucas Stach <dev@lynxeye.de>
CC: Tom Warren <twarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: add UART1 on GPU funcmux entry
Stephen Warren [Wed, 16 May 2012 05:59:59 +0000 (05:59 +0000)]
tegra: add UART1 on GPU funcmux entry

TrimSlice uses UART1 on the GPU pingroup.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: seaboard: add support for USB networking
Stephen Warren [Wed, 16 May 2012 06:36:12 +0000 (06:36 +0000)]
tegra: seaboard: add support for USB networking

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: whistler: reduce and comment network cfg options
Stephen Warren [Wed, 16 May 2012 06:36:11 +0000 (06:36 +0000)]
tegra: whistler: reduce and comment network cfg options

CONFIG_CMD_PING/NFS aren't required for Whistler to boot.

Add some comments.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: flesh out bootcmd
Stephen Warren [Wed, 16 May 2012 06:21:00 +0000 (06:21 +0000)]
tegra: flesh out bootcmd

This implements a useful bootcmd for Tegra. The boot order is:

* If USB enabled, USB storage
* Internal MMC (SD card or eMMC)
* If networking is enabled, BOOTP/TFTP

When booting from USB or MMC, the boot script is assumed to be in
partition 1 (although this may be overridden via the rootpart variable),
both ext2 and FAT filesystems are supported, the boot script may exist
in either / or /boot, and the boot script may be named boot.scr.uimg or
boot.scr.

When booting over the network, it is assumed that boot.scr.uimg exists
on the TFTP server. There is less flexibility here since those setting
up network booting are expected to need less hand-holding.

In all cases, it is expected that the initial file loaded is a U-Boot
image containing a script that will load the kernel, load any required
initrd, load any required DTB, and finally bootm the kernel.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: remove some cruft from CONFIG_EXTRA_ENV_SETTINGS
Stephen Warren [Wed, 16 May 2012 06:20:59 +0000 (06:20 +0000)]
tegra: remove some cruft from CONFIG_EXTRA_ENV_SETTINGS

console isn't used by anything, and the kernel should be set appropriately
by whatever script is booting the kernel, not imposed by the bootloader.

mem might be useful, but the current value is pretty bogus, since it
includes nvmem options that make no sense for an upstream kernel, and
equally should not be required for any downstream kernel. Either way, this
is also best left to the kernel boot script.

smpflag isn't used by anything, and again was probably intended to be a
kernel command-line option better set by the kernel boot script.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: Whistler board support
Stephen Warren [Tue, 15 May 2012 06:45:28 +0000 (06:45 +0000)]
tegra: Whistler board support

Whistler is a highly configurable Tegra evaluation and development board.
This change adds support for the following specific configuration:

E1120 motherboard
E1108 CPU board
E1116 PMU board

The motherboard configuration switches are set as follows:
SW1=0 SW2=0 SW3=5
S1/S2/S3/S4 all on, except S3 7/8 are off.

Other combinations of daugher boards may work to varying degrees, but will
likely require some SW adjustment.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: add alternate UART1 funcmux entry
Stephen Warren [Mon, 14 May 2012 13:13:45 +0000 (13:13 +0000)]
tegra: add alternate UART1 funcmux entry

(In at least some configurations) Whistler uses UART1 on pingroups
UAA, UAB.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agotegra: paz00: fix typo in SD slot CD detect GPIO
Stephen Warren [Tue, 15 May 2012 11:58:11 +0000 (11:58 +0000)]
tegra: paz00: fix typo in SD slot CD detect GPIO

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agospi: Tegra2: Seaboard: enable SPI/UART corruption fix
Tom Warren [Tue, 15 May 2012 21:38:03 +0000 (14:38 -0700)]
spi: Tegra2: Seaboard: enable SPI/UART corruption fix

Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agospi: Tegra2: Seaboard: fix UART corruption during SPI transactions
Tom Warren [Tue, 15 May 2012 21:32:40 +0000 (14:32 -0700)]
spi: Tegra2: Seaboard: fix UART corruption during SPI transactions

Simon Glass's proposal to fix this on Seaboard was NAK'd, so I
removed his NS16550 references and added a small delay before
SPI/UART muxing. Tested on my Seaboard with large SPI reads/writes
and saw no corruption (crc's matched) and no spurious comm chars.

Signed-off-by: Tom Warren <twarren@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Jimmy Zhang <jimmzhang@nvidia.com>
12 years agotools/mkenvimage.c: fix basename(3) usage
Andreas Bießmann [Thu, 28 Jun 2012 06:01:58 +0000 (08:01 +0200)]
tools/mkenvimage.c: fix basename(3) usage

Use the POSIX variant of basename due to BSD systems (e.g. OS X) do not provide
GNU version of basename(3). It is save to use the POSIX variant here cause we do
never use argv[0] later on which may be modified by the basename(3) POSIX
variant.
On systems providing GNU variant the GNU variant should be used since string.h
is included before libgen.h. Therefore let the _GNU_SOURCE as is.

This patch fixes following warning (on OS X):
---8<---
mkenvimage.c: In function ‘main’:
mkenvimage.c:105: warning: implicit declaration of function ‘basename’
mkenvimage.c:105: warning: assignment makes pointer from integer without a cast
--->8---

Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
cc: Keith Mok <ek9852@gmail.com>

12 years agosh: ap_sh4a_4a: Fix wrong register initialization value
Nobuhiro Iwamatsu [Thu, 28 Jun 2012 01:22:57 +0000 (10:22 +0900)]
sh: ap_sh4a_4a: Fix wrong register initialization value

The value of ET0_ERXD6 and GPSR1_INIT was wrong.
This fixes them.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
12 years agoext2load: increase read speed
u-boot@lakedaemon.net [Wed, 28 Mar 2012 04:37:11 +0000 (04:37 +0000)]
ext2load: increase read speed

This patch dramatically drops the amount of time u-boot needs to read a
file from an ext2 partition.  On a typical 2 to 5 MB file (kernels and
initrds) it goes from tens of seconds to a couple seconds.

All we are doing here is grouping contiguous blocks into one read.

Boot tested on Globalscale Technologies Dreamplug (Kirkwood ARM SoC)
with three different files.  sha1sums were calculated in Linux
userspace, and then confirmed after ext2load.

Signed-off-by: Jason Cooper <u-boot@lakedaemon.net>
Tested-by: Eric Nelson <eric.nelson@boundarydevices.com>
Tested-by: Thierry Reding <thierry.reding@avionic-design.de>
12 years agopxe: add support for parsing local syslinux files
Rob Herring [Fri, 25 May 2012 10:47:39 +0000 (10:47 +0000)]
pxe: add support for parsing local syslinux files

Add a new command "sysboot" which parses syslinux menu files and boots
using kernel and initrd specified by menu files. The operation is similar
to "pxe boot" except local files on ext2 or fat filesystem are parsed.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agopxe: parse initrd file from append string
Rob Herring [Wed, 28 Mar 2012 05:51:37 +0000 (05:51 +0000)]
pxe: parse initrd file from append string

For syslinux, the initrd can be set in the append string as
"initrd=<file>", so try to find it there if we haven't already set the
initrd.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agopxe: support absolute paths
Rob Herring [Wed, 28 Mar 2012 05:51:36 +0000 (05:51 +0000)]
pxe: support absolute paths

If the file path starts with a '/', then don't pre-pend the bootfile path.
This fixes a problem with running 'pxe boot' multiple times where the
bootfile path gets pre-pended to itself each time.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agopxe: support linux entries for labels
Rob Herring [Wed, 28 Mar 2012 05:51:35 +0000 (05:51 +0000)]
pxe: support linux entries for labels

Kernels can be specified using "linux" or "kernel" entry. The difference
is kernel is supposed to detect the type of file, but for u-boot both are
treated the same.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agopxe: add support for label menu text
Rob Herring [Wed, 28 Mar 2012 05:51:34 +0000 (05:51 +0000)]
pxe: add support for label menu text

Use a menu string if present, otherwise use the kernel string.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agopxe: support include files at top-level
Rob Herring [Fri, 25 May 2012 10:43:16 +0000 (10:43 +0000)]
pxe: support include files at top-level

Include files outside of a menu were not getting included and parsed.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agomenu: only timeout when menu is displayed
Rob Herring [Wed, 28 Mar 2012 05:51:32 +0000 (05:51 +0000)]
menu: only timeout when menu is displayed

Make the menu timeout apply only when prompt flag is set and after the
menu is displayed. This allows auto boot to work no matter whether prompt
is set or cleared. Use the default selection if the menu times out.

This also fixes the timeout value given to readline_into_buffer to be
seconds instead of 10th of seconds.

Old behavior:

if prompt
  display menu and wait for choice
else
  wait for timeout
  if key pressed
    display menu and wait for choice
  else
    exit command

New behavior:

if prompt
  display menu
  if key pressed
    wait for choice
  else
    boot default entry on timeout
else
  boot default entry

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
12 years agocheckpatch: add check for whitespace before semicolon at end-of-line
Eric Nelson [Thu, 3 May 2012 03:32:18 +0000 (20:32 -0700)]
checkpatch: add check for whitespace before semicolon at end-of-line

This tests for a bad habits of mine like this:

return 0 ;

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>