This fixes a naming bug for at91rm9200 lowlevel init code:
NOR boot flash is on chipselect 0, not chipselect 2. This
makes code use the register name from chip datasheets.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Provide a board_lmb_reserve helper function to ensure we reserve
the page of memory we are using for the boot page translation code.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The following changes are needed to be inline with ePAPR v0.81:
* r4, r5 and now always set to 0 on boot release
* r7 is used to pass the size of the initial map area (IMA)
* EPAPR_MAGIC value changed for book-e processors
* changes in the spin table layout
* spin table supports a 64-bit physical release address
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
get_ddr_freq() and get_bus_freq() used get_sys_info() each time they were
called. However, get_sys_info() recalculates extraneous information when
called each time. Have get_ddr_freq() and get_bus_freq() return memoized
values from global_data instead.
Signed-off-by: James Yang <James.Yang@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Show the DDR memory data rate in addition to the memory clock
frequency. For DDR/DDR2 memories the memory data rate is 2x the
memory clock.
Signed-off-by: James Yang <James.Yang@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Speed up get_tbclk() by referencing pre-computed bus clock
frequency value from global data instead of sys_info_t. Fix
rounding of result to nearest; previously it was rounding
upwards.
Signed-off-by: James Yang <James.Yang@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
FSL has taken to using SVR[16:23] as an SOC sub-version field. This
is used to distinguish certain variants within an SOC family. To
account for this, we add the SVR_SOC_VER() macro, and update the SVR_*
constants to reflect the larger value. We also add SVR numbers for all
of the current variants. Finally, to make things neater, rather than
use an enormous switch statement to print out the CPU type, we create
and array of SVR/name pairs (using a macro), and print out the CPU name
that matches the SVR SOC version.
Signed-off-by: Andy Fleming <afleming@freescale.com>
Added the cpu command that provides a generic mechanism to get status,
reset, and release secondary cores in multicore processors.
Added support for using the ePAPR defined spin-table mechanism on 85xx.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Andy Fleming <afleming@freescale.com>
Added the cpu command that provides a generic mechanism to get status,
reset, and release secondary cores in multicore processors.
Added support for using the ePAPR defined spin-table mechanism on 85xx.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
When we go to 36-bit physical addresses we need to keep the concept of
the physical CCSRBAR address seperate from the virtual one.
For the majority of boards CFG_CCSBAR_PHYS == CFG_CCSRBAR
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
device_type = "soc" is being deprecated, newer device trees will use
"fsl,soc" and/or "fsl,immr" for the soc nodes.
This patch also adds clock-frequency property for soc nodes (the same
value as bus-frequency).
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
The following changes are based on kernel UCC ethernet performance:
1. Make the CSB bus pipeline depth as 4, and enable the repeat mode
2. Optimize transactions between QE and CSB. Added CFG_SPCR_OPT
switch to enable this setting.
The following changes are based on the App Note AN3369 and
verified to improve memory latency using LMbench:
3. CS0_CONFIG[AP_n_EN] is changed from 1 to 0
4. CS0_CONFIG[ODT_WR_CONFIG] set to 1. Was a reserved setting
previously.
5. TIMING_CFG_1[WRREC] is changed from 3clks to 2clks (based on
Twr=15ns, and this was already the setting in DDR_MODE)
6. TIMING_CFG_1[PRETOACT] is changed from 3clks to 2clks. (based on
Trp=15ns)
7. TIMING_CFG_1[ACTTOPRE] is changed from 9clks to 6clks. (based on
Tras=40ns)
8. TIMING_CFG_1[ACTTORW] is changed from 3clks to 2clks. (based on
Trcd=15ns)
9. TIMING_CFG_1[REFREC] changed from 21 clks to 11clks. (based on
Trfc=75ns)
10. TIMING_CFG_2[FOUR_ACT] is changed from 10 clks to 7clks. (based
on Tfaw=50ns)
11. TIMING_CFG_2[ADD_LAT] and DDR_MODE[AL] changed from 0 to 1 (based
on CL=3 and WL=2).
Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com>
Acked-by: Kim Phillips <kim.phillips@freescale.com>
Include FEC specific nodes in ft_cpu_setup only if CONFIG_MPC5xxx_FEC is
defined. Systems without FEC, i.e. no FEC node in DTB, should be possible.
Signed-off-by: Andre Schwarz <andre.schwarz@matrix-vision.de>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Add setup_c0_status from Linux. For the moment we disable interrupts, set
CU0, mark the kernel mode, and clear ERL and EXL. This is good enough for
reset-time configuration and will work well across most processors.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
Without this change, we'll be suffering from deffered WATCH exception
once Status.EXL is cleared. Make sure Cause.WP is cleared.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
We do Hit_Writeback_Inv_D and Hit_Invalidate_I. You might think that you
don't need to do Hit_Invalidate_I, but flush_cache() needs it since this
function is used not only in U-Boot specfic programs but also at loading
target binaries.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
Currently we do 1) Index_Store_Tag_I, 2) Fill and 3) Index_Store_Tag_I
again per a loop for I-cache initialization. But according to 'See MIPS
Run', we're encouraged to use three separate loops rather than combining
them *for both I- and D-cache*. This patch tries to fix this.
In accordance with fixing above, mips_init_[id]cache are separated from
mips_cache_reset(), and rewrite cache loops are completely rewritten with
useful macros.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
This patch replaces the current function definitions with NESTED, LEAF
and END macro. They specify some more additional information about the
function; an alignment of symbol, type of symbol, stack frame usage, etc.
These information explicitly tells the assembler and the debugger about
the types of code we want to generate.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
The initial intension of having mips_cache_lock() was to use the cache
as memory for temporary stack use so that a C environment can be set up
as early as possible.
But now mips_cache_lock() follow lowlevel_init(). We've already have the
real memory initilaized at this point, therefore we could/should use it.
No reason to lock at all.
Other problems:
Cache locking is not consistent across MIPS implementaions. Some imple-
mentations don't support locking at all. The style of locking varies -
some support per line locking, others per way, etc. Some parts use bits
in status registers instead of cache ops. Current mips_cache_lock() is
not necessarily general-purpose.
And this is worthy of special mention; once U-Boot/MIPS locks the lines,
they are never get unlocked, so the code relies on whatever gets loaded
after U-Boot to re-initialize the cache and clear the locks. We're sup-
posed to have CFG_INIT_RAM_LOCK and unlock_ram_in_cache() implemented,
but leave the situation as it is for a long time.
For these reasons, I proposed the removal of mips_cache_lock() from the
global start-up code.
This patch adds CFG_INIT_RAM_LOCK_MIPS to make existing users aware that
*things have changed*. If he wants the same behavior as before, he needs
to have CFG_INIT_RAM_LOCK_MIPS in his config file.
If we don't have any regression report through several releases, then
we'll remove codes entirely.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
Acked-by: Andrew Dyer <amdyer@gmail.com>
Modify the RTC API to provide one a status for the time reported by
the rtc_get() function:
0 - a reliable time is guaranteed,
< 0 - a reliable time isn't guaranteed (power fault, clock issues,
and so on).
The RTC chip drivers are responsible for providing this info if the
corresponding chip supports such functionality. If not - always
report that the time is reliable.
The POST RTC test was modified to detect the RTC faults utilizing
this new rtc_get() feature.
Signed-off-by: Yuri Tikhonov <yur@emcraft.com>
460EX doesn't support a fixed bootstrap option to boot from 512 byte page
NAND devices. The only bootstrap option for NAND booting is option F for
2k page devices. So to boot from a 512 bype page device, the I2C bootstrap
EEPROM needs to be programmed accordingly.
This patch adds basic NAND booting support for the AMCC Canyonlands aval
board and also adds support to the "bootstrap" command, to enable NAND
booting I2C setting.
Tested with 512 byte page NAND device (32MByte) on Canyonlands.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch is a rework of the 4xx interrupt handling done while
adding the 460EX/GT support. Interrupts are needed on 4xx for the
EMAC driver.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch changes the physical addess parameter from 32bit to 64bit.
This is needed for 36bit 4xx platforms to access areas located
beyond the 4GB border, like SoC peripherals (EBC etc.).
Signed-off-by: Stefan Roese <sr@denx.de>
Every now and then a Sequoia board (or equivalent hardware) had
problems connecting to a Gigabit capable network interface.
There were differences in the PHY setup between Linux and U-Boot.
This patch fixes the problem. Apparently "remote fault" is being set,
which signals to some devices (on the other end of the cable) that a
fault has occurred, while other devices ignore it. I believe the RF bit
was causing the issue, but I removed T4 also, to match up with Linux.
Signed-off-by: Mike Nuss <mike@terascala.com>
The 405EZ only supports 512 bytes of rx-/tx-fifo EMAC sizes. But
currently 4k/2k is configured. This patch fixes this issue.
Thanks to Thomas Kindler <tkindler@lenord.de> for pointing this out.
Signed-off-by: Stefan Roese <sr@denx.de>
Adds PCI support for MPC5121
Tested with drivers/net/rtl8139.c
Support is conditional since PCI on old silicon does not work.
ads5121_PCI_config turns on PCI
In this version, condition compilation of PCI code has been moved
from ifdef in board/ads5121/pci.c to board/ads5121/Makefile as
suggested by Jean-Christophe PLAGNIOL-VILLARD
Signed-off-by: John Rigby <jrigby@freescale.com>
U-Boot crashes in the net loop if CONFIG_4xx_DCACHE is
enabled. To reproduce the problem ensure that 'ethrotate'
environment variable isn't set to "no" and then run
"tftp 200000 not_existent_file".
This patch tries to fix the issue.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
There is no reason to icbi when invalidating the temporary stack in
the d-cache. Its impossible on e500 to have the i-cache contain
any addresses in the temp stack and it can be problematic in generating
transactions on the bus to non-valid addresses.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The source vector for the ECM was being set to 2,
but that's what the source vector for DDR was being
set to. Change it to 1.
Signed-off-by: Andy Fleming <afleming@freescale.com>
Commit 0db37dc... (and some others) changed the INIT_RAM TLB
mappings to be unguarded. This collided with an existing "bug"
where the mappings for the INIT_RAM were being kept around.
This meant that speculative loads to those addresses were
succeeding in the TLB, and going out to the bus, where they
were causing an exception (there's nothing at that address). The
Flash code was coincidentally causing such a speculative load.
Rather than go back to mapping the INIT RAM as guarded, we fix
it so that the entries for the INIT_RAM are invalidated. Thus
the speculative loads will fail in the TLB, and have no effect.
Signed-off-by: Andy Fleming <afleming@freescale.com>
