On Katmai the complete auto-calibration somehow doesn't seem to
produce the best results, meaning optimal values for RQFD/RFFD.
This was discovered by GDA using a high bandwidth scope,
analyzing the DDR2 signals. GDA provided a fixed value for RQFD,
so now on Katmai "only" RFFD is auto-calibrated.
This patch also adds RDCC calibration as mentioned on page 7 of
the AMCC PowerPC440SP/SPe DDR2 application note:
"DDR1/DDR2 Initialization Sequence and Dynamic Tuning"
Signed-off-by: Stefan Roese <sr@denx.de>
On Yucca & Katmai, the inbound memory map pointed to 0x4.0000.0000, which
is the internal SRAM. Since I now ported and tested this endpoint mode
on Kilauea successfully to map to 0 (SDRAM), I also changed this for
Katmai.
Yucca will stay at internal SRAM for now. Not sure if somebody relies on
this setup.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch adds support for dynamic configuration of PCIe ports for the
AMCC PPC4xx boards equipped with PCIe interfaces. These are the PPC440SPe
boards Yucca & Katmai and the 405EX board Kilauea.
This dynamic configuration is done via the "pcie_mode" environement
variable. This variable can be set to "EP" or "RP" for endpoint or
rootpoint mode. Multiple values can be joined via the ":" delimiter.
Here an example:
pcie_mode=RP:EP:EP
This way, PCIe port 0 will be configured as rootpoint, PCIe port 1 and 2
as endpoint.
Per default Yucca will be configured as:
pcie_mode=RP:EP:EP
Per default Katmai will be configured as:
pcie_mode=RP:RP:REP
Per default Kilauea will be configured as:
pcie_mode=RP:RP
Signed-off-by: Tirumala R Marri <tmarri@amcc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
128MB seems to be the smallest possible value for the memory size
for on PCIe port. With this change now the BAR's of the PCIe cards
are accessible under U-Boot.
One big note: This only works for PCIe port 0 & 1. For port 2 this
currently doesn't work, since the base address is now 0xc0000000
(0xb0000000 + 2 * 0x08000000), and this is already occupied by
CFG_PCIE0_CFGBASE. But solving this issue for port 2 would mean
to change the base addresses completely and this change would have
too much impact right now.
This patch adds debug output to the 4xx pcie driver too.
Signed-off-by: Stefan Roese <sr@denx.de>
- 'pciconfighost' is set by default in order to be able to scan bridges
behind the primary host/PCIe
- 'pciscandelay' env variable is recognized to allow for user-controlled
delay before the PCIe bus enumeration; some peripheral devices require a
significant delay before they can be scanned (e.g. LSI8408E); without the
delay they are not detected
Signed-off-by: Grzegorz Bernacki <gjb@semihalf.com>
- correct configuration space mapping
- correct bus numbering
- better access to config space
Prior to this patch, the 440SPe host/PCIe bridge was able to configure only the
first device on the first bus. We now allow to configure up to 16 buses;
also, scanning for devices behind the PCIe-PCIe bridge is supported, so
peripheral devices farther in hierarchy can be identified.
Signed-off-by: Grzegorz Bernacki <gjb@semihalf.com>
platforms wishing to display RAM diagnostics in addition to size,
can do so, on one line, in their own board_add_ram_info()
implementation.
this consequently eliminates CONFIG_ADD_RAM_INFO.
Thanks to Stefan for the hint.
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Explicitly add in default CONFIG_BOOTP_* options where cmd_confdefs.h
used to be included but CONFIG_BOOTP_MASK was not defined.
Remove lingering references to CFG_CMD_* symbols.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
This patch removes the CFG_PCI_PRE_INIT option completely, since
it's not needed anymore with the patch from Matthias Fuchs with
the "weak" pci_pre_init() implementation.
Signed-off-by: Stefan Roese <sr@denx.de>
- Introduced dedicated switches for building 440 and 405 images required
for 440-specific machine instructions like 'rfmci' etc.
- Exception vectors moved to the proper location (_start moved away from
the critical exception handler space, which it occupied)
- CriticalInput now serviced (with default handler)
- MachineCheck properly serviced (added a dedicated handler and return
subroutine)
- Overall cleanup of exceptions declared with STD_EXCEPTION macro (unused,
unhandled and those not relevant for 4xx were eliminated)
- Eliminated Linux leftovers, removed dead code
Signed-off-by: Grzegorz Bernacki <gjb@semihalf.com>
Signed-off-by: Rafal Jaworowski <raj@semihalf.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Previous versions used full wait states for the chip select #1 which
is connected to the Xilinix SystemACE controller on the AMCC Katmai
evaluation board. This leads to really slow access and therefore low
performance. This patch now sets up the chip select a lot faster
resulting in much better read/write performance of the Linux driver.
Signed-off-by: Stefan Roese <sr@denx.de>
With this new base address of the Xilinx SystemACE controller
the Linux driver will be easier to adapt, since it can now be
mapped via the "normal" ioremap() call.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch updates the recently added Katmai board support. The biggest
change is the support of ECC DIMM modules in the 440SP(e) SPD DDR2
driver.
Please note, that still some problems are left with some memory
configurations. See the driver for more details.
Signed-off-by: Stefan Roese <sr@denx.de>
