It turns out that calling fsp_init_phase_pci() in arch_misc_init()
is subject to break pci device drivers as with driver model, when
the bus enumeration happens is not deterministic.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
With dm pci conversion, pci config read/write in unprotect_spi_flash()
silently fails as at that time dm pci is not ready and bus enumeration
is not done yet. Actually we don't need to do this in that early phase,
hence we delay this call to arch_misc_init().
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Add some comments in start.S for the fact that with FSP U-Boot
actually enters the code twice. Also change to use fsp_init()
and fsp_continue for accuracy.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
After fsp_init() returns, the stack has already been switched to a
place within system memory as defined by CONFIG_FSP_TEMP_RAM_ADDR.
Enlarge the size of malloc() pool before relocation since we have
plenty of memory now.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Move x86_fsp_init() call after initf_malloc() so that we can fix up
the gd->malloc_limit later.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
On some platforms pci devices behind bridge need to be probed (eg:
a pci uart on recent x86 chipset) before relocation. But we won't
bind all devices found during the enumeration. Only devices whose
driver with DM_FLAG_PRE_RELOC set will be bound. Any other generic
devices except bridges won't be bound.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
When ifdtool collates the microcode into one place it effectively creates
a copy of the 'data' properties in the device tree microcode nodes. This
is wasteful since we now have two copies of the microcode in the ROM.
To avoid this, remove the microcode data from the device tree and shrink it
down. This means that there is only one copy and the overall ROM space used
by the microcode does not increase.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
The Intel Firmware Support Package (FSP) requires that microcode be provided
very early before the device tree can be scanned. We already support adding
a pointer to the microcode data in a place where early init code can access.
However this just points into the device tree and can only point to a single
lot of microcode. For boards which may have different CPU types we must
support multiple microcodes and pass all of them to the FSP in one place.
Enhance ifdtool to scan all the microcode, place it together in the ROM and
update the microcode pointer to point there. This allows us to pass multiple
microcode blocks to the FSP using its existing API.
Enable the flag in the Makefile so that this feature is used by default for
all boards.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
The code to set up the microcode pointer in the ROM shares almost nothing
with the write_uboot() function.
Move it into its own function so it will be easier to extend.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Since U-Boot and its device tree can grow we should check that it does not
overlap the regions above it. Track the ROM offset that U-Boot reaches and
check that other regions (written after U-Boot) do not interfere.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Intel FSP has the capability to walk through the microcode blocks
which are passed as the TempRamInit() parameter from U-Boot and
finds the most appropriate microcode which is suitable for the cpu
on which it is running. Now we've seen several steppings for Intel
BayTrail series processors, adding those microcodes to the Intel
BayleyBay and MinnowMax board device tree files.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
This commit adds the microcode blob for BayTrail-I D0 stepping,
CPUID signature 30679h.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
This is incorrect since we require the -m parameter to the microcode tool.
Update the two examples to show this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
The doc has a misleading 'make menuconfig' when building the EFI
application and payload. Clarify this and also update information
on test with QEMU.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Currenlty we only set up video framebuffer when VIDEO_VESA driver is
used. With coreboot, VIDEO_COREBOOT driver is used instead. Since we
already saved VESA mode in the VIDEO_COREBOOT driver, now we can also
set up video framebuffer for coreboot before loading Linux kernel.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
When booting as a coreboot payload, the framebuffer details are
passed from coreboot via configuration tables. We save these
information into vesa_mode_info structure for future use.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
It looks that x86 chipset always contains a host bridge at pci
b.d.f 0.0.0, so enable this for all boards.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
When booting as a coreboot payload, we don't need write any
configuration tables as coreboot does that for us.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Some platforms may have >=4GiB memory, so we need make U-Boot report
such configuration correctly when booting as the coreboot payload.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Now that we have generic routine to calculate relocation address,
remove the x86 specific one which is now only used by coreboot.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
coreboot has some extensions (type 6 & 16) to the E820 types.
When we detect this, mark it as E820_RESERVED.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Increase lib_sysinfo memrange entry number to 32 to sync with coreboot.
This allows a complete E820 table to be reported to the kernel, as on
some platforms (eg: Bayley Bay) having only 16 entires does not cover
all the memory ranges.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
With recent EFI support, the entry point address of coreboot payload
was changed. Now we update the address to use _x86boot_start, which
is the same one for EFI.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
commit 6497e37 "net: e1000: Support 64-bit physical address" causes
compiler warnings on 32-bit U-Boot build below.
drivers/net/e1000.c: In function 'e1000_configure_tx':
drivers/net/e1000.c:4982:2: warning: right shift count >= width of type [enabled by default]
drivers/net/e1000.c: In function 'e1000_configure_rx':
drivers/net/e1000.c:5126:2: warning: right shift count >= width of type [enabled by default]
This commit fixes the build warnings.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Follow linux dma flow:
Before DMA read, be sure to invalidate the cache over the address
range of DMA buffer to prevent cache coherency problems.
After DMA read, invalidate dcache again.
Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
Acked-by: Marek Vasut <marex@denx.de>
Acked-by: Tim Harvey <tharvey@gateworks.com>
If ecc chunk data size is 512 and oobsize is bigger than 512, there is
a chance that block_mark_bit_offset conflicts with bch ecc area.
The following graph is modified from kernel gpmi-nand.c driver with
each data block 512 bytes. We can see that Block Mark conflicts with
ecc area from bch view. We can enlarge the ecc chunk size to avoid
this problem to those oobsize which is larger than 512.
| P |
|<----------------------------------------------------------------->|
| |
| (Block Mark) |
| P' | | | |
|<--------------------------------------------------->| D | | O'|
| |<--------->| |<->|
V V V V V
+---+--------------+-+--------------+-+--------------+-+----------+-+---+
| M | data |E| data |E| data |E| data |E| |
+---+--------------+-+--------------+-+--------------+-+----------+-+---+
^ ^
| O |
|<---------------->|
P : the page size for BCH module.
E : The ECC strength.
G : the length of Galois Field.
N : The chunk count of per page.
M : the metasize of per page.
C : the ecc chunk size, aka the "data" above.
P': the nand chip's page size.
O : the nand chip's oob size.
O': the free oob.
Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Tested-By: Tim Harvey <tharvey@gateworks.com>
This patch enables NAND support on the Marvell Armada XP
DB-MV784MP-GP eval board.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Peter Morrow <peter@senient.com>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Cloned from the Linux driver v4.2.0-rc2. Plus some patches from
Antoine Tenart enabling controller initialization and ONFI timing
support:
http://lists.infradead.org/pipermail/linux-mtd/2015-July/060197.html
Please note that this driver needs the Linux NAND subsystem sync to v4.1
from Scott to be applied:
https://www.mail-archive.com/u-boot@lists.denx.de/msg175762.html
Otherwise it will not compile.
Tested on the Marvell Armada XP DB-MV784MP-GP eval board.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Antoine Tenart <antoine.tenart@free-electrons.com>
Cc: Ezeguil Garcia <ezequiel.garcia@free-electrons.com>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Cc: Scott Wood <scottwood@freescale.com>
Update the NAND code to match Linux v4.1. The previous sync was
from Linux v3.15 in commit 4e67c57125.
CONFIG_SYS_NAND_RESET_CNT is removed, as the upstream Linux code now
has its own timeout. Plus, CONFIG_SYS_NAND_RESET_CNT was undocumented
and not selected by any board.
Signed-off-by: Scott Wood <scottwood@freescale.com>
In addition to mtd_block_isbad(), which checks if a block is bad or
reserved, it's needed to check if a block is reserved only (but not
bad). This commit adds an MTD interface for it, in a similar fashion to
mtd_block_isbad().
While here, fix mtd_block_isbad() so the out-of-bounds checking is done
before the callback check.
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
Tested-by: Pekon Gupta <pekon@ti.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
[scottwood: Cherry-picked from Linux 8471bb73ba10ed67]
Signed-off-by: Scott Wood <scottwood@freescale.com>
Add optional LCD support. Note that depending on the toolchain used
one might have to drop some other features to stay within the 0x40000
size limit.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Looks like the define CONFIG_SYS_LCD_PXA_NO_L_BIAS is not used anywhere
else throughout the U-Boot sources any more. Drop it.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
This patch moves pxa to the common timer functions added in commit
8dfafdd - Introduce common timer functions <Rob Herring>
The (removed) pxa timer code (specifically __udelay()) could deadlock at
the 32-bit boundary of get_ticks(). get_ticks() returned a 32-bit value
cast up to a 64-bit value. If get_ticks() + tmo in __udelay() crossed
the 32-bit boundary, the while condition became unconditionally true and
locked the processor. Rather than patch the specific pxa issues, simply
move everything over to the common code.
Signed-off-by: Andrew Ruder <andrew.ruder@elecsyscorp.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Tom Rini <trini@konsuiko.com>
Since commit 3ff46cc42b ("arm: relocate the exception vectors")
pxa does not boot anymore.
Add a specific relocate_vectors macro that skips the vector relocation,
as the pxa SoC does not provide RAM at the high vectors address
(0xFFFF0000), and (0x00000000) maps to ROM.
This allows pxa to boot again.
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Andrew Ruder <andrew.ruder@elecsyscorp.com>
Conflicts:
configs/socfpga_arria5_defconfig
configs/socfpga_cyclone5_defconfig
configs/socfpga_socrates_defconfig
Merged these by hand and re-ran savedefconfig on them.
Signed-off-by: Tom Rini <trini@konsulko.com>
Bin Meng
Simon Glass
Tom Rini
Peng Fan
Stefan Roese
Siva Durga Prasad Paladugu
Scott Wood
Ezequiel Garcia
Marcel Ziswiler
Andrew Ruder