This patch fixes a problem on systems where the NOR flash is attached
to a 64 bit bus. The toggle bit detection in flash_toggle() is based
on the assumption that the same flash address is read twice without
any other interjacent flash accesses. However, on 32 bit systems the
function flash_read64() [as currently implemented] does not perform
an atomic 64 bit read - instead, this is broken down into two 32 bit
read accesses on addresses "addr" and "addr + 4". So instead of
reading a 64 bit value twice from "addr", we see a sequence of 4 32
bit reads from "addr", "addr + 4", "addr", and "addr + 4". The
consequence is that flash_toggle() fails to work.
This patch implements a simple, but somewhat ugly solution, as it
avoids the use of flash_read64() in this critical place (by breaking
it down manually into 32 bit read operations) instead of rewriting
flash_read64() such to perform atomic 64 bit reads as one could
expect. However, such a rewrite would require the use of floating
point load operations, which becomes pretty complex:
save MSR;
set Floating Point Enable bit in MSR;
use "lfd" instruction to perform atomic 64 bit read;
use "stfd" to store value to temporary variable on stack;
load u64 value from temporary variable;
restore saved MSR;
return u64 value;
The benefit-cost ratio of such an implementation was considered too
bad to actually attempt this, especially as we can expect that such
an implementation would not only have a bigger memory footprint but
also cause a performance degradation.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
The patch is following the commit 3924384060
mpc86xx: use r4 instead of r2 in lock_ram_in_cache and unlock_ram_in_cache
This is needed in unlock_ram_in_cache() because it is called from C and
will corrupt the small data area anchor that is kept in R2.
lock_ram_in_cache() is modified similarly as good coding practice, but
is not called from C.
Signed-off-by: Nick Spence <nick.spence@freescale.com>
also, the r2 is used as global data pointer.
Signed-off-by: Dave Liu <daveliu@freescale.com>
The current calculation will fail to cover all memory if
its size is not a power of two.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
The commit 67256678f0 add
the another global data pointer save, but in fact the
global data pointer will be initialized in the board_init_r,
so remove it such as the 85xx/83xx family.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Acked-by: Kumar Gala <kumar.gala@freescale.com>
I believe these code was copied from 74xx family, but for
86xx, it is unused.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Acked-by: Kumar Gala <kumar.gala@freescale.com>
The DDR1 LAW will precedence the DDR2 LAW, so remove
the second DDR LAW.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Acked-by: Becky Bruce <becky.bruce@freescale.com>
The DDR controller of 86xx processors have the ECC data init
feature, and the new DDR code is using the feature, we don't
need the way with DMA to init memory again.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Acked-by: Kumar Gala <kumar.gala@freescale.com>
We must invalidate TLBs before MMU turn on, but
currently the code is not, if there are some stale
TLB entry valid in the TLBs, it will cause strange
issue.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Acked-by: Becky Bruce <becky.bruce@freescale.com>
As reported by Ilko Iliev <iliev@ronetix.at>, the "nand erase clean"
command is currently broken, and among other things causes all blocks
to be marked bad.
This implements it properly using MTD_OOB_AUTO, along with some
indentation fixes.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Hardware expects ECCM 0 for small page and ECCM 1 for large page
when booting from NAND, so use those defaults.
Signed-off-by: Scott Wood <scottwood@freescale.com>
This patch defines 1M TLB&LAW size for NAND on MPC8572DS, assigns
0xffa00000 for CONFIG_SYS_NAND_BASE and adds other NAND supports in
config file.
It also moves environment(CONFIG_ENV_ADDR) outside of u-boot image, to
make room for the increased code size with NAND enabled.
Signed-off-by: Jason Jin <Jason.Jin@freescale.com>
Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
- Rename lbus83xx_t to fsl_lbus_t and move it to asm/fsl_lbc.h so that it
can be shared by both 83xx and 85xx
- Remove lbus83xx_t and replace it with fsl_lbus_t in all 83xx boards
files which use lbus83xx_t.
- Move FMR, FIR, FCR, FPAR, LTESR from mpc83xx.h to asm/fsl_lbc.h so that
85xx can share them.
Signed-off-by: Jason Jin <Jason.Jin@freescale.com>
Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
Some chips require a RESET after power-up (e.g. Micron MT29FxGxxxxx).
The first command sent is NAND_CMD_READID.
Issue a NAND_CMD_RESET in nand_scan_ident before reading the device id.
Tested with an MT29F4G08AAC.
Signed-off-by: Karl Beldan <karl.beldan@gmail.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
This brings the core NAND code up to date with the Linux kernel.
Since there were several drivers in Linux as of the last update that are
not in u-boot, I'm not bringing over new drivers that have been added
since in the absence of an interested party.
I did not update OneNAND since it was recently synced by Kyungmin Park,
and I'm not sure exactly what the common ancestor is.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Add the ability to break the steps of the bootm command into several
subcommands: start, loados, ramdisk, fdt, bdt, cmdline, prep, go.
This allows us to do things like manipulate device trees before
they are passed to a booting kernel or setup memory for a secondary
core in multicore situations.
Not all OS types support all subcommands (currently only start, loados,
ramdisk, fdt, and go are supported).
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This removes a bit of code and makes it easier for the upcoming sub bootm
command support to call into the proper OS specific handler.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Renamed show_boot_progress in assembler init phase to
show_boot_progress_asm to avoid link conflicts with C version
Signed-off-by: Graeme Russ <graeme.russ@gmail.com>
The MPC8572 has a 4-bit wide PORDEVSR IO_SEL field. Other MPC85xx
processors have a 3-bit wide IO_SEL field but have the most
significant bit is wired to 0 so this change should not affect
them.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
The existing code has a few errors that need to be fixed in
order to support large RAM sizes. Fix those, and add a
comment to make it clearer.
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
The information displayed when CONFIG_LCD_INFO is set is inherently
board-specific, so it should be done by the board code. The current code
dealing with this only handles two cases, and is already a horrible mess
of #ifdeffery.
Yes, this duplicates some code, but it also allows boards to print more
board-specific information; this used to be very difficult.
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
This allows the logo/info rendering routines to use the regular
lcd_putc/lcd_puts/lcd_printf calls.
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
lcd_printf() has a prototype in include/lcd.h but no implementation. Fix
this by borrowing the lcd_printf() implementation from the cogent board
code (which appears to use its own LCD framework.)
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
If the board _didn't_ request INVLINE_INVERTED, we set INVLINE_INVERTED,
otherwise we don't. WTF?
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
atmel_lcdfb doesn't actually need anything from asm/arch/hardware.h. It
includes a file that does, asm/arch/gpio.h, but this file doesn't
include <asm/arch/hardware.h> like it's supposed to.
Add the missing include to asm/arch/gpio.h and remove the workaround
from the atmel_lcdfb driver. This makes the driver compile on avr32.
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Converted MPC8610HCPD, MPC8641HPCN, and SBC8641D to use
fsl_pci_setup_inbound_windows() and ft_fsl_pci_setup().
With these changes the board code is a bit smaller and we get dma-ranges
set in the device tree for these boards.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Acked-by: Jon Loeliger <jdl@freescale.com>
Converted ATUM8548, MPC8536DS, MPC8544DS, MPC8548CDS, MPC8568MDS,
MPC8572DS, TQM85xx, and SBC8548 to use fsl_pci_setup_inbound_windows()
and ft_fsl_pci_setup().
With these changes the board code is a bit smaller and we get dma-ranges
set in the device tree for these boards.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Add a common setup function that determines the pci_region(s) based
on how much memory we have in the system.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
* PCI Inbound window was setup incorrectly. The PCI address and system
address were swapped. The PCI address should be setting piwar/piwbear
and the system address should be setting pitar.
* Removed masking of addresses to allow for system address to support
system address & PCI address >32-bits
* Set PIWBEAR & POTEAR to allow for full 64-bit PCI addresses
* Respect the PCI_REGION_PREFETCH for inbound windows
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Added fdt_pci_dma_ranges() that parses the pci_region info from the
struct pci_controller and populates the dma-ranges based on it.
The max # of windws/dma-ranges we support is 3 since on embedded
PowerPC based systems this is the max number of windows.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Add helper functions to return find a node and return it's property
or a default value.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Acked-by: Gerald Van Baren <vanbaren@cideas.com>
PCI bus is inherently 64-bit. While not all system require access to
the full 64-bit PCI address range some do. This allows those systems
to enable the full PCI address width via CONFIG_SYS_PCI_64BIT.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andrew Fleming-AFLEMING <afleming@freescale.com>
Acked-by: Wolfgang Denk <wd@denx.de>
The 8572 DDR erratum1:
DDR controller may enter an illegal state when operating
in 32-bit bus mode with 4-beat bursts.
Description:
When operating with a 32-bit bus, it is recommended that
DDR_SDRAM_CFG[8_BE] is cleared when DDR2 memories are used.
This forces the DDR controller to use 4-beat bursts when
communicating to the DRAMs. However, an issue exists that
could lead to data corruption when the DDR controller is
in 32-bit bus mode while using 4-beat bursts.
Projected Impact:
If the DDR controller is operating in 32-bit bus mode with
4-beat bursts, then the controller may enter into a bad state.
All subsequent reads from memory is corrupted.
Four-beat bursts with a 32-bit bus only is used with DDR2 memories.
Therefore, this erratum does not affect DDR3 mode.
Work Arounds:
To work around this issue, software must set DEBUG_1[31] in
DDR memory mapped space (CCSRBAR offset + 0x2f00 for DDR_1
and CCSRBAR offset + 0x6f00 for DDR_2).
Currenlty, the code is using incorrect register DDR_SDRAM_CFG_2
as condition, but it should be DDR_SDRAM_CFG register.
Signed-off-by: Dave Liu <daveliu@freescale.com>
Introduce CONFIG_E500MC to deal with the minor differences between
e500v2 and e500mc.
* Certain fields of HID0/1 don't exist anymore on e500mc
* Cache line size is 64-bytes on e500mc
* reset value of PIR is different
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Using CONFIG_SYS_CACHELINE_SIZE instead of 31 means we can handle
e500mc's 64-byte cacheline properly when it gets added.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This patch adds support for the avnet fx12 minimodul.
It needs the "ppc4xx: Generic architecture for xilinx ppc405"
patch from Ricardo.
Signed-off-by: Georg Schardt <schardt@team-ctech.de>
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@uam.es>
Signed-off-by: Stefan Roese <sr@denx.de>
