]> git.ipfire.org Git - thirdparty/u-boot.git/log
thirdparty/u-boot.git
13 years agoXilinx: ARM: Update NAND driver to match nand_scan_ident()
Joe Hershberger [Wed, 21 Mar 2012 18:47:18 +0000 (13:47 -0500)] 
Xilinx: ARM: Update NAND driver to match nand_scan_ident()

nand_scan_ident() has been updated to match Linux...
Remove special case.

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Only build the NAND controller driver if NAND is enabled
Joe Hershberger [Wed, 21 Mar 2012 00:35:07 +0000 (19:35 -0500)] 
Xilinx: ARM: Only build the NAND controller driver if NAND is enabled

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Fix memory corruption bug in pele timer_init
Joe Hershberger [Tue, 20 Mar 2012 22:38:48 +0000 (17:38 -0500)] 
Xilinx: ARM: Fix memory corruption bug in pele timer_init

Global variables (.bss) must not be accessed before relocation.
That memory is now overlayed with the .rel.dyn ELF relocation table.
Even if that was allowed, it would have no positive effect since
initialized value would not be relocated into RAM.  It would be 0
again.

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Define CONFIG_SYS_BOOTMAPSZ to enable fdt relocation
Joe Hershberger [Tue, 20 Mar 2012 23:14:10 +0000 (18:14 -0500)] 
Xilinx: ARM: Define CONFIG_SYS_BOOTMAPSZ to enable fdt relocation

This is needed until the fdt_high command is included (v2011.09)

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: INIT_SP is now needed before relocation
Joe Hershberger [Tue, 20 Mar 2012 23:19:37 +0000 (18:19 -0500)] 
Xilinx: ARM: INIT_SP is now needed before relocation

Define the offset and size of the INIT_RAM.
Base the INIT_SP on the INIT_RAM and reference the generated
GBL_DATA_SIZE.
Remove the explicit definition of the CONFIG_SYS_GBL_DATA_SIZE.

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: TEXT_BASE was renamed CONFIG_SYS_TEXT_BASE
Joe Hershberger [Tue, 20 Mar 2012 23:05:46 +0000 (18:05 -0500)] 
Xilinx: ARM: TEXT_BASE was renamed CONFIG_SYS_TEXT_BASE

The TEXT_BASE belongs in the config headers for the boards

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Remove redundant do_reset()
Joe Hershberger [Tue, 20 Mar 2012 23:04:04 +0000 (18:04 -0500)] 
Xilinx: ARM: Remove redundant do_reset()

do_reset() implementation was unified and is no longer needed here

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Fix crash when accessing uninit'd ptr
Joe Hershberger [Tue, 20 Mar 2012 23:01:48 +0000 (18:01 -0500)] 
Xilinx: ARM: Fix crash when accessing uninit'd ptr

The gd->bd pointer is no longer available in dram_init()

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Update to new build process
Joe Hershberger [Tue, 20 Mar 2012 23:00:13 +0000 (18:00 -0500)] 
Xilinx: ARM: Update to new build process

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoXilinx: ARM: Remove zynq boards from Makefile
Joe Hershberger [Wed, 21 Mar 2012 18:07:57 +0000 (13:07 -0500)] 
Xilinx: ARM: Remove zynq boards from Makefile

Board definitions belong in boards.cfg, not in Makefile

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
13 years agoMerge commit 'v2011.03' into xpele-upgrade-v2011.03
Joe Hershberger [Wed, 21 Mar 2012 18:04:49 +0000 (13:04 -0500)] 
Merge commit 'v2011.03' into xpele-upgrade-v2011.03

Conflicts:
arch/arm/lib/bootm.c
boards.cfg
drivers/mtd/cfi_flash.c
drivers/mtd/nand/nand_base.c

13 years agoMerge branch 'master-zc770'
John Linn [Mon, 5 Mar 2012 05:21:02 +0000 (21:21 -0800)] 
Merge branch 'master-zc770'

13 years agoXilinx: ARM: Adding more zynq board support
John Linn [Mon, 5 Mar 2012 02:04:45 +0000 (18:04 -0800)] 
Xilinx: ARM: Adding more zynq board support

New board configurations are needed to allow EP107 to
not be broken, and to allow each board to be configured
differently.

zynq_zc770_config and zynq_zc702_config are now new
make targets.

13 years agoXilinx: ARM: fix low level to include correct header file
John Linn [Mon, 5 Mar 2012 02:00:34 +0000 (18:00 -0800)] 
Xilinx: ARM: fix low level to include correct header file

The previous fix, including xpele.h, worked for EP107, but
doesn't work once there are more configuation header files
other than xpele.h.  This now works for new ones also.

13 years agoXilinx: ARM: fix NAND for EP107 and make EP107 the default build
John Linn [Sun, 4 Mar 2012 07:17:18 +0000 (23:17 -0800)] 
Xilinx: ARM: fix NAND for EP107 and make EP107 the default build

NAND was not being built for EP107. A new header file will
be added for Zynq so that xpele.h will remain for EP107.

13 years agoXilinx: ARM: fix EP107 XIP booting
John Linn [Sun, 4 Mar 2012 07:15:49 +0000 (23:15 -0800)] 
Xilinx: ARM: fix EP107 XIP booting

This fixes XIP (NOR and QSPI) booting. This problem
was caused by Zynq silicon integration.

13 years agoMerge branch 'master-zc770' of /home/linnj/sivlinnj/u-boot-xarm into master-zc770
GIT Repo [Wed, 18 Jan 2012 02:40:52 +0000 (18:40 -0800)] 
Merge branch 'master-zc770' of /home/linnj/sivlinnj/u-boot-xarm into master-zc770

13 years agoXilinx: ARM: prevent 1 Gb with GEM for now
John Linn [Wed, 18 Jan 2012 02:39:09 +0000 (18:39 -0800)] 
Xilinx: ARM: prevent 1 Gb with GEM for now

This commit is intended to be reverted once all the boards
are patched so that Gb works.  Until then, stay at 100 Mb.

This allowed the previous work to be checked in so that it's
ready.

13 years agoXilinx: ARM: Updated ethernet driver for EP107.
Andrei Simion [Thu, 12 Jan 2012 00:05:37 +0000 (16:05 -0800)] 
Xilinx: ARM: Updated ethernet driver for EP107.

On EP107, U-boot can now communicate over ethernet at 10 and 100 Mbps.

13 years agoXilinx: ARM: Clean-up of ethernet code.
Andrei Simion [Thu, 22 Dec 2011 23:04:24 +0000 (15:04 -0800)] 
Xilinx: ARM: Clean-up of ethernet code.

MAC/PHY driver has been cleaned up and commented thoroughly. Uneeded
code has been removed.

13 years agoXilinx: ARM: Update for gigabit ethernet support.
Andrei Simion [Thu, 22 Dec 2011 22:07:27 +0000 (14:07 -0800)] 
Xilinx: ARM: Update for gigabit ethernet support.

The PHY driver has been updated to establish a link at the highest
speed possible. The GEM divisors are modified to create an input
frequency that matches the link speed:
   2.5MHz for 10Mbps
   25MHz for 100Mbps
   125MHz for 1000Mbps

Error checking has also been included to prevent auto-negotiation
from stalling and for catching failed auto-negotiation attempts.

13 years agoMerge branch 'master-zc770' of /home/linnj/sivlinnj/u-boot-xarm into master-zc770
GIT Repo [Tue, 17 Jan 2012 22:15:49 +0000 (14:15 -0800)] 
Merge branch 'master-zc770' of /home/linnj/sivlinnj/u-boot-xarm into master-zc770

13 years agoXilinx: ARM: adding jtag boot mode
John Linn [Tue, 17 Jan 2012 22:12:50 +0000 (14:12 -0800)] 
Xilinx: ARM: adding jtag boot mode

To make it easier for developers in jtag mode, make it
automatically TFTP a kernel into memory and start it.

13 years agoXilinx: ARM: Stop GCC from generating unaligned accesses
John Linn [Fri, 6 Jan 2012 03:19:42 +0000 (19:19 -0800)] 
Xilinx: ARM: Stop GCC from generating unaligned accesses

With the newer GNU tools, GCC 4.5.2 and newer, we're
seeing data aborts. This solution may not be the right
long term solution, but works for now. This causes GCC
not to generate unaligned data.

13 years agoXilinx: ARM: Stop GCC from generating unaligned accesses
John Linn [Fri, 6 Jan 2012 03:19:42 +0000 (19:19 -0800)] 
Xilinx: ARM: Stop GCC from generating unaligned accesses

With the newer GNU tools, GCC 4.5.2 and newer, we're
seeing data aborts. This solution may not be the right
long term solution, but works for now. This causes GCC
not to generate unaligned data.

13 years agoXilinx: ARM: Stop GCC from generating unaligned accesses
John Linn [Fri, 6 Jan 2012 03:19:42 +0000 (19:19 -0800)] 
Xilinx: ARM: Stop GCC from generating unaligned accesses

With the newer GNU tools, GCC 4.5.2 and newer, we're
seeing data aborts. This solution may not be the right
long term solution, but works for now. This causes GCC
not to generate unaligned data.

13 years agoXilinx: ARM: Adding board support for FMC daughter cards
John Linn [Tue, 20 Dec 2011 19:16:44 +0000 (11:16 -0800)] 
Xilinx: ARM: Adding board support for FMC daughter cards

The NAND working caused u-boot to lockup when there was
no NAND so now there's CONFIGs for each board permutation
of the ZC770. The default is for DC1 (XM010) card.

13 years agoXilinx: ARM: NAND: wait for device ready after device reset
John Linn [Mon, 19 Dec 2011 20:35:55 +0000 (12:35 -0800)] 
Xilinx: ARM: NAND: wait for device ready after device reset

When moving from PEEP to Zynq, the ID of the flash was not
being read at all, only zeroes. The driver was not waiting
for the device to be ready after resetting the device and
since Zynq is so much faster it was a problem.

13 years agoXilinx: ARM: adding SD boot mode
John Linn [Tue, 13 Dec 2011 20:58:04 +0000 (12:58 -0800)] 
Xilinx: ARM: adding SD boot mode

SD boot was not supported in the automatic boot mode, now
it is. Even though this loads a ramdisk you can just ignore
the ramdisk if the root file system is on SD in an EXT2
file system.

13 years agoXilinx: ARM: SD: change divisor to speed it up
John Linn [Tue, 13 Dec 2011 20:55:56 +0000 (12:55 -0800)] 
Xilinx: ARM: SD: change divisor to speed it up

In the former change, a divisor too low seems cause problems
with the lower speed cards. This is attempt to have something
that's tolerable even if it's not the optimal.

This allows the kernel and ramdisk to be loaded reasonably
quickly.

13 years agoRevert "Xilinx: ARM: SD: speed up SD by removing the clock divide"
John Linn [Tue, 13 Dec 2011 17:43:58 +0000 (09:43 -0800)] 
Revert "Xilinx: ARM: SD: speed up SD by removing the clock divide"

This reverts commit a0d4d7e736ac3fc721ea59a54e3d4c94c3853a2a.

13 years agoXilinx: ARM: BSP: adding hack to take USB out of reset
John Linn [Tue, 13 Dec 2011 16:25:40 +0000 (08:25 -0800)] 
Xilinx: ARM: BSP: adding hack to take USB out of reset

Temporary hack to take USB out of reset til Linux is fixed.

13 years agoXilinx: ARM: SD: speed up SD by removing the clock divide
John Linn [Tue, 13 Dec 2011 16:20:36 +0000 (08:20 -0800)] 
Xilinx: ARM: SD: speed up SD by removing the clock divide

It's not clear why it was was running so slow, but maybe
it was needed for EP107. Speed it up now.

14 years agoXilinx: ARM: adding ping to qspi boot mode
John Linn [Wed, 7 Dec 2011 23:41:00 +0000 (15:41 -0800)] 
Xilinx: ARM: adding ping to qspi boot mode

This is a hack to compensate for Linux not having the PHY updated
yet as u-boot needs to touch the PHY for Linux networking to work.

14 years agoXilinx: ARM: making u-boot match FSBL for zc770
John Linn [Sun, 4 Dec 2011 14:59:09 +0000 (06:59 -0800)] 
Xilinx: ARM: making u-boot match FSBL for zc770

This should allow u-boot to be built for the zc770 and
work with the default FSBL which setups up the DDR, PLLs,
and MIO for 800 MHz CPU, 533 MHz DDR.

UART and Ethernet are working at this point.

14 years agoXilinx: ARM: adding header file for compile error
John Linn [Fri, 2 Dec 2011 00:51:44 +0000 (16:51 -0800)] 
Xilinx: ARM: adding header file for compile error

This header file was needed for the commit (a ways back)
that added PLL init to the board.c file.

14 years agoXilinx: ARM: GEM: updated to work with ZC770 and DC1
John Linn [Wed, 30 Nov 2011 01:36:25 +0000 (17:36 -0800)] 
Xilinx: ARM: GEM: updated to work with ZC770 and DC1

Only minor changes to GEM assuming MIO and IO PL are
setup somewhere else (FSBL or a script).

The DC1 PHY needs some help to work on the ZC770 and
that was added.

Since Zynq is much faster than EP107, removed some of
the printing of "." which was too much.

14 years agoXilinx: ARM: BSP: updated default to build for ZC770/DC1
John Linn [Wed, 30 Nov 2011 01:15:21 +0000 (17:15 -0800)] 
Xilinx: ARM: BSP: updated default to build for ZC770/DC1

The EP107 is now not the default build.

14 years agoXilinx: ARM: BSP: remove DDR init for zc770
John Linn [Wed, 30 Nov 2011 01:12:27 +0000 (17:12 -0800)] 
Xilinx: ARM: BSP: remove DDR init for zc770

This should not be needed as FSBL will be doing it. The DDR init
was causing a hang and was not needed anyway.

14 years agoXilinx: ARM: GEM driver cleanup of I/O functions
John Linn [Wed, 30 Nov 2011 00:54:52 +0000 (16:54 -0800)] 
Xilinx: ARM: GEM driver cleanup of I/O functions

The memory barriers appeared to be wrong, this moves the
barriers to be correct.

14 years agoXilinx: ARM: update uart clock input for ZC770/DC1GEM testing
John Linn [Wed, 30 Nov 2011 00:51:42 +0000 (16:51 -0800)] 
Xilinx: ARM: update uart clock input for ZC770/DC1GEM testing

The 100 Mb mode needs a 25 MHz clock input so this was needed
for this testing.

14 years agoXilinx: ARM: Hack for doing MIO and IO PLL for GEM testing
John Linn [Wed, 30 Nov 2011 00:48:45 +0000 (16:48 -0800)] 
Xilinx: ARM: Hack for doing MIO and IO PLL for GEM testing

This should not live on and is only temporary til FSBL is
working.

14 years agoRevert "Xilinx: ARM: GEM PHY address update for silicon."
GIT Repo [Mon, 21 Nov 2011 22:39:07 +0000 (14:39 -0800)] 
Revert "Xilinx: ARM: GEM PHY address update for silicon."

This reverts commit 90a036349c6ad4e69fd9c3b7cce19c037eefbd48.

The new board has only 2 choices for PHY addresses, 0x7 and
0x17 which are based on the jumpers. 0x17 was what was being
used on EP107 so assuming we set the jumpers we shouldn't
need this change which would not have worked (0x16 phy addr).

14 years agoXilinx: ARM: Use different xparameters for silicon and EP107.
Andrei Simion [Wed, 16 Nov 2011 19:34:21 +0000 (11:34 -0800)] 
Xilinx: ARM: Use different xparameters for silicon and EP107.

If CONFIG_EP107 is defined, include 'xparameters.h' and 'xparameters_pss.h',
otherwise include 'xparameters_zynq.h' and 'xparameters_pss_zynq.h'.

The Zynq variants of xparameters are subject to change. For now, the
only difference is XPAR_CPU_CORTEXA9_CORE_CLOCK_FREQ_HZ is
216664500 Hz instead of 12500000 Hz and XPAR_XUARTPSS_1_CLOCK_HZ is
13756480 Hz instead of 50000000 Hz to reflect silicon.

14 years agoXilinx: ARM: No config environment in NOR flash for silicon.
Andrei Simion [Wed, 16 Nov 2011 00:48:15 +0000 (16:48 -0800)] 
Xilinx: ARM: No config environment in NOR flash for silicon.

U-boot's configuration environment will reside in NOR flash only if
CONFIG_EP107 is defined.

14 years agoXilinx: ARM: GEM PHY address update for silicon.
Andrei Simion [Tue, 15 Nov 2011 23:55:36 +0000 (15:55 -0800)] 
Xilinx: ARM: GEM PHY address update for silicon.

GEM update to use a PHY address of 0x17 if CONFIG_EP107 is defined,
0x10 otherwise (for ZC770).

14 years agoXilinx: ARM: SCU timer input frequency update.
Andrei Simion [Thu, 10 Nov 2011 03:29:19 +0000 (19:29 -0800)] 
Xilinx: ARM: SCU timer input frequency update.

Modified the 'xpele.h' configuration such that the timer input
frequency is always half of the CPU core clock frequency from
'xparameters.h' (PERIPHCLK in the TRM) instead of hard-coded
to 5 MHz.

14 years agoXilinx: ARM: QSPI update to use single device.
Andrei Simion [Wed, 9 Nov 2011 23:26:52 +0000 (15:26 -0800)] 
Xilinx: ARM: QSPI update to use single device.

Modified the xpele config header to use single device for QSPI
instead of dual.

14 years agoXilinx: ARM: Use UART1 for silicon.
Andrei Simion [Wed, 9 Nov 2011 02:23:45 +0000 (18:23 -0800)] 
Xilinx: ARM: Use UART1 for silicon.

'xpele.h' configuration header updated so that U-boot uses UART1 if
CONFIG_ZYNQ is defined, UART0 otherwise (conditional compilation).

14 years agoXilinx: ARM: Flexible baud rate calculation.
Andrei Simion [Wed, 9 Nov 2011 02:16:51 +0000 (18:16 -0800)] 
Xilinx: ARM: Flexible baud rate calculation.

Values used in U-boot's baud rate calculation are chosen in such
a way to keep the error between the target and calculated baud
rate under 3%.

The header was changed to avoid hardcoding values such as the ones
used in the baud rate calculation. The UART input frequency uses
the value defined in 'xparameters.h'.

14 years agoXilinx: ARM: Changed environment to load device trees
John Linn [Fri, 23 Sep 2011 18:05:11 +0000 (12:05 -0600)] 
Xilinx: ARM: Changed environment to load device trees

The Linux kernel is now defaulting to device tree mode such
that a device tree blob is to be loaded into memory when
loading Linux from each of the flash memory types.

This only changed the commands in the environment for
each flash to also load the device tree blob.

I found that the commands can get too long such that
u-boot gives an error so they are not as verbose as
previously.

14 years agoXilinx: ARM: SD: reset interrupt status after operation
John Linn [Mon, 25 Jul 2011 21:05:46 +0000 (15:05 -0600)] 
Xilinx: ARM: SD: reset interrupt status after operation

The interrupt status of the SD was not being cleared after the
operation was completed which was causing problems for other
applications that were loaded.

14 years agoXilinx: ARM: Support new boot rom header.
Brian Hill [Fri, 1 Jul 2011 16:51:11 +0000 (10:51 -0600)] 
Xilinx: ARM: Support new boot rom header.

14 years agoXilinx: ARM: Added support for Dual Die NAND
Naveen Mamindlapalli [Fri, 24 Jun 2011 11:11:21 +0000 (16:41 +0530)] 
Xilinx: ARM: Added support for Dual Die NAND

Added support for detecting Dual Die NAND with OnDie ECC support.

Signed-Off-by: Naveen Mamindlapalli<naveenm@xilinx.com>
14 years agoXilinx: ARM: Increased malloc size
Naveen Mamindlapalli [Thu, 16 Jun 2011 13:44:29 +0000 (19:14 +0530)] 
Xilinx: ARM: Increased malloc size

Increased malloc size while testing Clear NAND flash.

Signed-Off-by: Naveen Mamindlapalli<naveenm@xilinx.com>
14 years agoXilinx: ARM: Added ClearNAND ECC support
Naveen Mamindlapalli [Fri, 3 Jun 2011 13:11:05 +0000 (18:41 +0530)] 
Xilinx: ARM: Added ClearNAND ECC support

14 years agoXilinx: ARM: Update modeboot to match Boot ROM (yet again)
Brian Hill [Tue, 21 Jun 2011 17:40:04 +0000 (11:40 -0600)] 
Xilinx: ARM: Update modeboot to match Boot ROM (yet again)

14 years agoXilinx: ARM: DT 614252 Set baudrate from command line.
Brian Hill [Fri, 17 Jun 2011 19:50:02 +0000 (13:50 -0600)] 
Xilinx: ARM: DT 614252 Set baudrate from command line.

14 years agoXilinx: ARM: Update modeboot to match the latest changes to Boot ROM
Brian Hill [Fri, 17 Jun 2011 18:51:57 +0000 (12:51 -0600)] 
Xilinx: ARM: Update modeboot to match the latest changes to Boot ROM

14 years agoXilinx: ARM: Adjust CONFIG_SYS_HZ to 1000.
Brian Hill [Tue, 24 May 2011 19:00:37 +0000 (13:00 -0600)] 
Xilinx: ARM: Adjust CONFIG_SYS_HZ to 1000.

Adjust CONFIG_SYS_HZ to 1000. Parts of the networking code assume that
the clock tick is natively expressed in milliseconds (and don't adjust
the values by using CONFIG_SYS_HZ).

14 years agoXilinx: ARM: Default baud rate set to 115200.
Brian Hill [Mon, 23 May 2011 16:08:29 +0000 (10:08 -0600)] 
Xilinx: ARM: Default baud rate set to 115200.

14 years agoXilinx: ARM: Correct timekeeping (DT 584643)
Brian Hill [Thu, 19 May 2011 20:24:55 +0000 (14:24 -0600)] 
Xilinx: ARM: Correct timekeeping (DT 584643)

14 years agoXilinx: U-boot: Added support for On-Die ECC
Naveen Mamindlapalli [Wed, 11 May 2011 15:07:53 +0000 (20:37 +0530)] 
Xilinx: U-boot: Added support for On-Die ECC

Added support for On-Die ECC supported by Micron flash.

Signed-off-by: Naveen Mamindlapalli<naveenm@xilinx.com>
14 years agoXilinx: ARM: Modify serial driver so that putc(\n) will actually output \r\n
Brian Hill [Tue, 10 May 2011 20:50:08 +0000 (14:50 -0600)] 
Xilinx: ARM: Modify serial driver so that putc(\n) will actually output \r\n

14 years agoXilinx: ARM: Update DDR initialization to PEEP11.
Brian Hill [Tue, 10 May 2011 18:52:44 +0000 (12:52 -0600)] 
Xilinx: ARM: Update DDR initialization to PEEP11.

14 years agoXilinx: ARM: Include EXT2 support in Pele build.
Brian Hill [Thu, 28 Apr 2011 21:37:42 +0000 (15:37 -0600)] 
Xilinx: ARM: Include EXT2 support in Pele build.

14 years agoXilinx: ARM: Make use of boot mode register possible with u-boot
Brian Hill [Tue, 26 Apr 2011 14:54:27 +0000 (08:54 -0600)] 
Xilinx: ARM: Make use of boot mode register possible with u-boot

The environment variable modeboot is set to something appropriate
based on the contents of the boot mode register.
EXAMPLE:
For bootmode 2, modeboot will be set to "run norboot".
This method of booting can be utilized via bootcmd = "run modeboot"
or not, if the user prefers that u-boot and the kernel come from
separate storage devices.

14 years agoXilinx: ARM: Update SD driver to handle MMC_RSP_R2 properly.
Brian Hill [Thu, 21 Apr 2011 15:34:46 +0000 (09:34 -0600)] 
Xilinx: ARM: Update SD driver to handle MMC_RSP_R2 properly.

 Update SD driver to handle MMC_RSP_R2 properly. 136 bit response was
 not being copied properly.  This prevented some SD cards from being
 detected correctly.

14 years agoXilinx: ARM: Update DDR initialization for peep10
Brian Hill [Tue, 5 Apr 2011 14:38:26 +0000 (08:38 -0600)] 
Xilinx: ARM: Update DDR initialization for peep10

14 years agoPrepare v2011.03 v2011.03
Wolfgang Denk [Thu, 31 Mar 2011 21:45:36 +0000 (23:45 +0200)] 
Prepare v2011.03

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoFix build issues cause by LDFLAGS_FINAL changes
Wolfgang Denk [Thu, 31 Mar 2011 21:26:29 +0000 (16:26 -0500)] 
Fix build issues cause by LDFLAGS_FINAL changes

Commit 6dc1ece "Introduce a new linker flag LDFLAGS_FINAL" modified a
number of Makefiles in a way that broke out-of-tree builds.  The
problem was that $(nandobj) was used before it got defined.

Fix this.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Scott Wood <scottwood@freescale.com>
14 years agoXilinx: ARM: Update to support PEEP 9 NOR flash changes.
Brian Hill [Thu, 31 Mar 2011 21:34:48 +0000 (15:34 -0600)] 
Xilinx: ARM: Update to support PEEP 9 NOR flash changes.

 NOR flash is now at 0xE2000000, not 0xE4000000.

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Thu, 31 Mar 2011 07:01:36 +0000 (09:01 +0200)] 
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-cfi-flash
Wolfgang Denk [Thu, 31 Mar 2011 06:59:32 +0000 (08:59 +0200)] 
Merge branch 'master' of git://git.denx.de/u-boot-cfi-flash

14 years agoFix build problems caused by "_end" -> "__bss_end__" rename
Wolfgang Denk [Tue, 29 Mar 2011 12:34:50 +0000 (14:34 +0200)] 
Fix build problems caused by "_end" -> "__bss_end__" rename

Commit 44c6e65 "rename _end to __bss_end__ broke building of a large
number of systems (at least all PowerPC?):

libstubs.o: In function `app_startup':
examples/standalone/stubs.c:197: undefined reference to `__bss_end__'

The rename should not be done for the files in the
examples/standalone/ directory, as these are not using the code from
start.S, but do their own BSS clearing, and either use their own
linker scripts or the ones provided by the compilers.

Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agopowerpc/85xx: Handle PCIe initialization requires for P1021 class SoCs
Prabhakar Kushwaha [Tue, 1 Feb 2011 15:55:58 +0000 (15:55 +0000)] 
powerpc/85xx: Handle PCIe initialization requires for P1021 class SoCs

The P1011, P1012, P1015, P1016, P1020, P1021, P1024, & P1025 SoCs require
that we initialize the SERDES registers if the lanes are configured for
PCIe.  Additionally these devices PCIe controller do not support ASPM
and we have to explicitly disable it.

Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agocfi_flash: fix bug with flash banks with different sector numbers
Martin Krause [Mon, 21 Mar 2011 17:07:56 +0000 (18:07 +0100)] 
cfi_flash: fix bug with flash banks with different sector numbers

The function find_sector() does not take into account if the flash bank
has changed since the last call. This could lead to illegal accesses inside
and beyond the flash_info_t info strcture. For example if the current
flash bank has less sectors than the last used flash bank.

This patch adds two cheks. One that insures, that the current sector does
not exceed the allowed maximum (which is always a good idea). And one that
checks if the current access is to the same flash bank as the last access.
If not, the search loop will start with sector 0.

Signed-off-by: Martin Krause <martin.krause@tqs.de>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agopowerpc/85xx: Enable various errata on P1022/P1013 SoCs
Jiang Yutang [Sun, 30 Jan 2011 23:06:20 +0000 (17:06 -0600)] 
powerpc/85xx: Enable various errata on P1022/P1013 SoCs

Enable workaround for errata ELBC A001, ESDHC 111 & SATA A001 on
P1022/P1013 SoCs.

Also updated P1022DS config to properly enable CONFIG_FSL_SATA_V2.

Signed-off-by: Jiang Yutang <b14898@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoPrepare v v2011.03-rc2
Wolfgang Denk [Sun, 27 Mar 2011 19:50:07 +0000 (21:50 +0200)] 
Prepare v

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoCoding Style cleanup: remove trailing empty lines
Wolfgang Denk [Sun, 27 Mar 2011 19:48:08 +0000 (21:48 +0200)] 
Coding Style cleanup: remove trailing empty lines

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Sun, 27 Mar 2011 19:20:29 +0000 (21:20 +0200)] 
Merge branch 'master' of git://git.denx.de/u-boot-arm

14 years agoarm: Tegra2: Change mach-type to MACH_TYPE_SEABOARD due to mach-types.h update
Tom Warren [Wed, 23 Feb 2011 09:54:31 +0000 (09:54 +0000)] 
arm: Tegra2: Change mach-type to MACH_TYPE_SEABOARD due to mach-types.h update

Signed-off-by: Tom Warren <twarren@nvidia.com>
14 years agoS5P: mmc: Resolved interrupt error during mmc_init
Chander Kashyap [Tue, 22 Mar 2011 01:40:50 +0000 (01:40 +0000)] 
S5P: mmc: Resolved interrupt error during mmc_init

Blocksize was hardcoded to 512 bytes. But the blocksize varies
depeding on various mmc subsystem commands (between 8 and 512).
This hardcoding was resulting in interrupt error during data
transfer.

It is now calculated based upon the request sent by mmc subsystem.

Signed-off-by: Chander Kashyap <chander.kashyap@linaro.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoARMV7: S5P: Fixed register offset in mmc.h
Chander Kashyap [Tue, 22 Mar 2011 01:29:38 +0000 (01:29 +0000)] 
ARMV7: S5P: Fixed register offset in mmc.h

The MMC registers are accessed through struct s5p_mmc member
variables. MMC controller "control4" register offset is set
to 0x8C as per data sheet. The size of struct s5p_mmc is also
corrected.

Signed-off-by: Chander Kashyap <chander.kashyap@linaro.org>
Signed-off-by: Tushar Behera <tushar.behera@linaro.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoS5P: timer: replace bss variable by gd
Minkyu Kang [Wed, 16 Mar 2011 11:09:20 +0000 (20:09 +0900)] 
S5P: timer: replace bss variable by gd

Use the global data instead of bss variable, replace as follow.

count_value -> removed
timestamp -> tbl
lastdec -> lastinc

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Acked-by: Albert ARIBAUD <albert.aribaud@free.fr>
14 years agoS5P: universal: Enable the pwm driver
Minkyu Kang [Thu, 10 Mar 2011 11:10:38 +0000 (20:10 +0900)] 
S5P: universal: Enable the pwm driver

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoS5P: goni: Enable the pwm driver
Minkyu Kang [Thu, 10 Mar 2011 11:10:19 +0000 (20:10 +0900)] 
S5P: goni: Enable the pwm driver

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoS5P: smdkc100: Enable the pwm driver
Minkyu Kang [Thu, 10 Mar 2011 11:09:43 +0000 (20:09 +0900)] 
S5P: smdkc100: Enable the pwm driver

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoS5P: timer: Use pwm functions
Minkyu Kang [Thu, 10 Mar 2011 11:05:58 +0000 (20:05 +0900)] 
S5P: timer: Use pwm functions

Use pwm functions for timer that is PWM timer 4.

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoARM: S5P: pwm driver support
Donghwa Lee [Mon, 7 Mar 2011 21:11:42 +0000 (21:11 +0000)] 
ARM: S5P: pwm driver support

This is common pwm driver of S5P.

Signed-off-by: Donghwa Lee <dh09.lee@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Fixup dram_init for relocation support
seedshope [Sat, 22 Jan 2011 10:06:14 +0000 (10:06 +0000)] 
SMDK6400: Fixup dram_init for relocation support

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Disable LED function in start.s on the nand booting
seedshope [Sat, 22 Jan 2011 10:06:13 +0000 (10:06 +0000)] 
SMDK6400: Disable LED function in start.s on the nand booting

Since nand boot have some limit for the first 4KB, We only
disable the LED function to reduce the code space. At the
same time, Fix the compile error for LED function undefined
in the compile time of nand_spl.

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Add some labels to u-boot.lds to support nand_spl
seedshope [Sat, 22 Jan 2011 10:06:12 +0000 (10:06 +0000)] 
SMDK6400: Add some labels to u-boot.lds to support nand_spl

In the nand_spl feature of SMDK6400. Add some relocation symbols to
nand_spl/board/samsung/smdk6400/u-boot.lds to fix the compile error.

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Fix the mutiple link error
seedshope [Sat, 22 Jan 2011 10:06:11 +0000 (10:06 +0000)] 
SMDK6400: Fix the mutiple link error

The first, the cpu_init.o have already been link for cmd_link_o_target
atfer compile, But, The link script re-link the point file. So the link
machine will generate multiple definition error information.

The second, Since the first 4kB of nand boot featue code move to nand_spl,
So It is not necessary to force the cpu_init.o in non-nand boot.

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Fix some label undefined in build error
seedshope [Sat, 22 Jan 2011 10:06:10 +0000 (10:06 +0000)] 
SMDK6400: Fix some label undefined in build error

Modify Makefile for cpu_init.c and Start.s use some label,this defined
u-boot.lds of arch/arm/cpu/arm1176. But SMDK6400 use the link script
board/samsung/smdk6400/u-boot-nand.lds. So add some label form u-boot.lds
to u-boot-nand.lds

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoSMDK6400: Fix CONFIG_SYS_INIT_SP_ADDR undefined
seedshope [Sat, 22 Jan 2011 10:06:09 +0000 (10:06 +0000)] 
SMDK6400: Fix CONFIG_SYS_INIT_SP_ADDR undefined

Fix CONFIG_SYS_INIT_SP_ADDR undefined issue.

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
14 years agoarm: fix incorrect monitor protection region in FLASH
Po-Yu Chuang [Tue, 1 Mar 2011 23:02:04 +0000 (23:02 +0000)] 
arm: fix incorrect monitor protection region in FLASH

Monitor protection region in FLASH did not cover .rel.dyn
and .dynsym sections, because it uses __bss_start to compute
monitor_flash_len. Use _end instead.

Add _end to linker scripts for end of u-boot image
Add _end_ofs to all the start.S.

Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
14 years agorename _end to __bss_end__
Po-Yu Chuang [Tue, 1 Mar 2011 22:59:59 +0000 (22:59 +0000)] 
rename _end to __bss_end__

Currently, _end is used for end of BSS section.  We want _end to mean
end of u-boot image, so we rename _end to __bss_end__ first.

Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
14 years agopowerpc/85xx: Fix PCI memory map setup on P1_P2_RDB
Prabhakar Kushwaha [Wed, 23 Mar 2011 09:21:13 +0000 (04:21 -0500)] 
powerpc/85xx: Fix PCI memory map setup on P1_P2_RDB

Update the PCIe address map to match standard FSL memory map.
Additionally, fix the TLBs so the cover the PCIe address space properly
so cards plugged in like an e1000 work correctly.

Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopowerpc/mpc8xxx: fix workaround for errata DDR111 and DDR134
York Sun [Thu, 17 Mar 2011 18:18:13 +0000 (11:18 -0700)] 
powerpc/mpc8xxx: fix workaround for errata DDR111 and DDR134

The fix for errata workaround is to avoid covering physical address
0xff000000 to 0xffffffff during the implementation.

Signed-off-by: York Sun <yorksun@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>