Tom Rini [Fri, 1 May 2020 21:58:31 +0000 (17:58 -0400)]
Merge branch 'next' of git://git.denx.de/u-boot-sh
This is part 1 of big network cleanup / DM conversion.
The dc2114x/rtl8139/pcnet/tulip PCI adapter drivers received checkpatch
cleanups in preparation for DM conversion. The smc911x is converted to
DM completely. The dwc_eth_qos cache fixes are applied.
Marek Vasut [Fri, 1 May 2020 15:40:25 +0000 (17:40 +0200)]
mkimage: fit: Do not tail-pad fitImage with external data
There is no reason to tail-pad fitImage with external data to 4-bytes,
while fitImage without external data does not have any such padding and
is often unaligned. DT spec also does not mandate any such padding.
Moreover, the tail-pad fills the last few bytes with uninitialized data,
which could lead to a potential information leak.
Signed-off-by: Marek Vasut <marex@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Cc: Tom Rini <trini@konsulko.com>
common/board_r: arm: Merge initr_enable_interrupts into interrupts_init
initr_enable_interrupts() is an ARM-specific wrapper over
enable_interrupts(), which is run during the common init sequence. It can
be eliminated by moving the enable_interrupts() call to the end of
interrupt_init() function, in arch/arm/lib/interrupts*.c.
Signed-off-by: Ovidiu Panait <ovpanait@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Marek Bykowski [Wed, 29 Apr 2020 16:23:07 +0000 (18:23 +0200)]
malloc: dlmalloc: add an ability for the malloc to be re-init/init multiple times
Malloc gets initialized with a call to mem_malloc_init() with the address
the allocation starts to and its size. Currently it is not possible to
move the malloc from one memory area to another as the malloc would eventually
fail.
This patch adds in the ability to re-init the malloc with the updated
start address and the size.
One of the use cases of this feature is SPL U-Boot running from within
the static memory and calling to malloc init from within board_init_f():
Shortly after the DDR (main) memory is init and ready we call to malloc init
again but this time with the start address in the DDR memory and a much greater
size for moving the allocation off the static to the DDR memory:
Where CONFIG_SYS_SPL_MALLOC_START and CONFIG_SPL_MALLOC_OFFSET are the start
addresses of the malloc in the static and DDR memories respectively and
CONFIG_SYS_SPL_MALLOC_SIZE=SZ_16K and CONFIG_SPL_MALLOC_SIZE=SZ_2M are
the sizes of the mallocs in these memories. Note, now we have a much greater
memory, enlarging from 16K to 2M, available for allocation.
There is an alternative approach already existing in U-Boot with the use of
an early (simplified) malloc and the proper (dlamalloc) malloc however
necessitating managing the two mallocs whereas this approach proposes using
a single dlmalloc.
Signed-off-by: Marek Bykowski <marek.bykowski@gmail.com>
Philippe Reynes [Wed, 29 Apr 2020 13:26:17 +0000 (15:26 +0200)]
rsa: sig: fix config signature check for fit with padding
The signature check on config node is broken on fit with padding.
To compute the signature for config node, U-Boot compute the
signature on all properties of requested node for this config,
except for the property "data". But, when padding is used for
binary in a fit, there isn't a property "data" but two properties:
"data-offset" and "data-size". So to fix the check of signature,
we also don't use the properties "data-offset" and "data-size"
when checking the signature on config node.
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Sven Roederer [Mon, 27 Apr 2020 00:08:38 +0000 (02:08 +0200)]
tools/mkimage: fix handling long filenames
The cmdline for calling the dtc was cut-off when using long filenames (e.g.
245 bytes) for output-file and datafile of "-f" parameter.
For FIT-images cmd[MKIMAGE_MAX_DTC_CMDLINE_LEN] is declared (hardcoded 512 bytes),
and contains some static values, the path of a tmpfile and a datafile. tmpfile is
max MKIMAGE_MAX_TMPFILE_LEN (256) and datafile might be also this size. Having two
very long pathname results in a truncation os the executed shell command, as the
truncated datafile path will not be found.
Redefine MKIMAGE_MAX_DTC_CMDLINE_LEN to "2 * MKIMAGE_MAX_TMPFILE_LEN + 35 for the
parameters.
This likely applies to the "-d" parameter, too.
Signed-off-by: Sven Roederer <devel-sven@geroedel.de>
Patrick Delaunay [Fri, 24 Apr 2020 18:20:17 +0000 (20:20 +0200)]
arm: caches: manage phys_addr_t overflow in mmu_set_region_dcache_behaviour
Solved the overflow on phys_addr_t type for start + size in
mmu_set_region_dcache_behaviour() function.
This overflow is avoided by dividing start and end by 2 before addition,
and we only expecting that start and size are even.
This patch doesn't change the current function behavior if the
parameters (start or size) are not aligned on MMU_SECTION_SIZE.
For example, this overflow occurs on ARM32 with:
start = 0xC0000000 and size = 0x40000000
then start + size = 0x100000000 and end = 0x0.
For information the function behavior change with risk of regression,
if we just shift start and size before the addition.
Example with 2MB section size:
MMU_SECTION_SIZE 0x200000 and MMU_SECTION_SHIFT = 21
with start = 0x1000000, size = 0x1000000,
- with the proposed patch, start = 0 and end = 0x1 as previously
- with the more simple patch:
end = (start >> MMU_SECTION_SHIFT) + (size >> MMU_SECTION_SHIFT)
the value of end change:
start >> 21 = 0, size >> 21 = 0 and end = 0x0 !!!
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Patrick Delaunay [Fri, 24 Apr 2020 18:20:16 +0000 (20:20 +0200)]
arm: caches: add DCACHE_DEFAULT_OPTION
Add the new flags DCACHE_DEFAULT_OPTION to define the default
option to use according the compilation flags
CONFIG_SYS_ARM_CACHE_*.
This new compilation flag allows to simplify dram_bank_mmu_setup()
and can be used as third parameter (option=dcache option to select)
of mmu_set_region_dcache_behaviour function.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Patrick Delaunay [Fri, 24 Apr 2020 18:20:15 +0000 (20:20 +0200)]
arm: caches: protect dram_bank_mmu_setup access to bi_dram
Add protection in dram_bank_mmu_setup() to avoid access to bd->bi_dram
before relocation.
This patch allow to use the generic weak function dram_bank_mmu_setup
to activate the MMU and the data cache in SPL or in U-Boot before
relocation, when bd->bi_dram is not yet initialized.
In this cases, the MMU must be initialized explicitly with
mmu_set_region_dcache_behaviour function.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Patrick Delaunay [Fri, 10 Apr 2020 14:02:02 +0000 (16:02 +0200)]
configs: migrate CONFIG_SYS_ARM_CACHE_* in Kconfig
Move CONFIG_SYS_ARM_CACHE_WRITETHROUGH and
CONFIG_SYS_ARM_CACHE_WRITEALLOC into Kconfig done by moveconfig.py.
Kconfig uses a choice between the 3 values supported in U-Boot,
including the new configuration CONFIG_SYS_ARM_CACHE_WRITEBACK
(the default configuration).
The patch also avoids to select simultaneously 2 configurations.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Marek Vasut [Sun, 15 Mar 2020 16:39:01 +0000 (17:39 +0100)]
net: smc911x: Add DM support
Add support for U-Boot DM and DT probing. Furthermore, build the
SMC911x standalone EEPROM example only for the non-DM case, as it
is not converted yet.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sun, 15 Mar 2020 14:36:09 +0000 (15:36 +0100)]
net: smc911x: Pass around driver private data
Introduce a private data structure for this driver with embedded
struct eth_device and pass it around. This prepares the driver to
work with both DM and non-DM systems.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sun, 15 Mar 2020 13:38:59 +0000 (14:38 +0100)]
net: smc911x: Drop weak alias from 32bit accessors
These accessors are not overridden by any board, and even if they were,
this is something which should be handled via DM now, so remove the
weak alias option. Moreover, drop the inline keyword, as the compiler
can decide better.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sun, 15 Mar 2020 14:03:07 +0000 (15:03 +0100)]
net: smc911x: Inline all functions from header file
Inline all the functions from the header file, as they are not used
outside of the driver or the standalone EEPROM example.
Note that this does introduce considerable amount of duplication in
the standalone EEPROM example, however that one has to be rewritten
anyway, roughly such that the SMC911x driver would expose DM EEPROM
interface and the standalone example would use that.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sat, 21 Mar 2020 16:25:41 +0000 (17:25 +0100)]
net: smc911x: Pull MII registration into separate function
Pull the MII interface registration into separate function to avoid the
ifdeffery in smc911x_initialize(). Moreover, adjust the fail path such
that we use goto labels.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sun, 15 Mar 2020 14:57:14 +0000 (15:57 +0100)]
net: smc911x: Fix potential memleak() in init fail path
Fix memleak in the init fail path, where if allocation or registration
of MDIO bus fails, then ethernet interface is not unregistered and the
private data are not freed, yet the probe function reports a failure.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Marek Vasut [Sun, 12 Apr 2020 21:49:25 +0000 (23:49 +0200)]
net: rtl8139: Fill in SPDX tag
The rtl8139 driver is derived from Etherboot driver, which is in
turn derived from Linux 8139too driver added in Linux 2.2.18pre14.
An undocumented change in Linux 2.4.10.2 added a new field to the
driver, MODULE_LICENSE("GPL"); . According to current Linux kernel
licensing rules, Documentation/process/license-rules.rst, quote:
"GPL" Module is licensed under GPL version 2. This
does not express any distinction between
GPL-2.0-only or GPL-2.0-or-later. The exact
license information can only be determined
via the license information in the
corresponding source files.
And since the code does not contain any "future" clause, the tag
is therefore GPL-2.0 only.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 12 Apr 2020 21:01:45 +0000 (23:01 +0200)]
net: rtl8139: Finish cleanup
Finish the checkpatch cleanup of the driver, fix the remaining issues
in probe and init function and in global variables, rename the probe
function to rtl8139_init(), no functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 12 Apr 2020 20:58:27 +0000 (22:58 +0200)]
net: rtl8139: Factor out hardware reset
This hardware reset and reset-wait implementation was twice in the
driver, factor it out into a separate function. This really should
use wait_for_bit() eventually and return -ETIMEDOUT, but thus far,
handling of any of this is missing from the driver. This must be
added later. Thus far, no functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 12 Apr 2020 20:40:45 +0000 (22:40 +0200)]
net: rtl8139: Minor cleanup of rtl_transmit()
Rename the function to rtl8139_send(), clean up the TX polling, the TX
OK condition at the end, overuse of typecasts, and various line-over-80
problems. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 12 Apr 2020 19:20:31 +0000 (21:20 +0200)]
net: rtl8139: Rework eeprom_delay() macro
The macro assumes ee_addr variable to be present when it's being
used. Rework the macro into a function instead and pass it an
argument specifying the register base address, to make it future
proof for DM conversion.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 12 Apr 2020 18:47:26 +0000 (20:47 +0200)]
net: rtl8139: Register macro cleanup
Clean up the horrible register definitions in the RTL8139 driver.
This does create a couple of checkpatch errors, but the driver is
full of them anyway, and those will be cleaned up later.
No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Mon, 23 Mar 2020 01:11:46 +0000 (02:11 +0100)]
net: dwc_eth_qos: Prevent DMA from writing updated RX DMA descriptor
The DMA may attempt to write a DMA descriptor in the ring while it is
being updated. By writing the DMA descriptor buffer address to 0, it
is assured the DMA will not use such a buffer and the buffer can be
updated without any interference.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Mon, 23 Mar 2020 01:09:55 +0000 (02:09 +0100)]
net: dwc_eth_qos: Invalidate RX packet DMA buffer
This patch prevents an issue where the RX packet might have been
accessed by the CPU, which now has cached data from the packet in
the caches and possibly various write buffers, and these data may
be evicted from the caches into the DRAM while the buffer is also
written by the DMA.
By invalidating the buffer after the CPU accessed it and before the
DMA populates the buffer, it is assured that the buffer will not be
corrupted.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Mon, 23 Mar 2020 01:09:21 +0000 (02:09 +0100)]
net: dwc_eth_qos: Invalidate RX descriptor before reading
The current code polls the RX desciptor ring for new packets by reading
the RX descriptor status. This works by accident, as the RX descriptors
are often in non-cacheable memory. However, the driver does support use
of RX descriptors in cacheable memory.
This patch adds a missing RX descriptor invalidation, which assures the
CPU will read a fresh copy of the RX descriptor instead of a cached one.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Mon, 23 Mar 2020 01:09:01 +0000 (02:09 +0100)]
net: dwc_eth_qos: Flush the RX descriptors on init
Currently the code only flushes the first RX descriptor, not every entry
in the RX descriptor ring. Fix this, to make sure the DMA engine can pick
the RX descriptors correctly.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Mon, 23 Mar 2020 01:03:50 +0000 (02:03 +0100)]
net: dwc_eth_qos: Correctly wrap around TX descriptor tail pointer
This code programs the next descriptor in the TX descriptor ring into
the hardware as the last valid TX descriptor. The problem is that if
the currenty descriptor is the last one in the array, the code will
not wrap around correctly and use TX descriptor 0 again, but instead
will use TX descriptor at address right past the TX descriptor ring,
which is the first descriptor in the RX ring.
Fix this by adding the necessary wrap-around.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Mon, 23 Mar 2020 01:02:57 +0000 (02:02 +0100)]
net: dwc_eth_qos: Fully rewrite RX descriptor field 3
The RX descriptor field 3 should contain only OWN and BUF1V bits before
being used for receiving data by the DMA engine. However, right now, if
the descriptor was already used for receiving data and is being cleared,
the field 3 is only modified and the aforementioned two bits are ORRed
into the field. This could lead to a residual dirty bits being left in
the field 3 from previous transfer, and it generally does. Fully set the
field 3 instead to clear those residual dirty bits.
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Joe Hershberger <joe.hershberger@ni.com> Cc: Patrice Chotard <patrice.chotard@st.com> Cc: Patrick Delaunay <patrick.delaunay@st.com> Cc: Ramon Fried <rfried.dev@gmail.com> Cc: Stephen Warren <swarren@nvidia.com>
Marek Vasut [Sun, 19 Apr 2020 01:40:03 +0000 (03:40 +0200)]
net: dc2114x: Clean up INL/OUTL functions
Rename these functions to dc2114x_{inl,outl}(), use u32 values in
them instead of plain signed integers as all those values are in
fact register values and the driver code does bitwise operations
on them. No functional change intended.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 19 Apr 2020 01:36:46 +0000 (03:36 +0200)]
net: dc2114x: Clean up DE4X5 macros
Replace these macros with static functions to permit the compiler to
do type checking on the functions. The INL()/OUTL() functions have to
be moved in this patch as well, as those DE4X5 macros are using them.
No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 19 Apr 2020 01:09:26 +0000 (03:09 +0200)]
net: dc2114x: Clean up remaining driver code
Clean up the remaining driver code, macro space alignment, function
declaration indent, replace __attribute__((aligned(32))) with plain
__aligned(32). No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 19 Apr 2020 01:10:50 +0000 (03:10 +0200)]
net: dc2114x: Clean up send_setup_frame()
Clean up the send_setup_frame() to bring it up to standards with
U-Boot coding style, invert the loops where applicable to cut
down the level of indent. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sun, 19 Apr 2020 01:10:14 +0000 (03:10 +0200)]
net: dc2114x: Clean up dc21x4x_send()
Clean up the driver send code to bring it up to standards with
U-Boot coding style, invert the loops where applicable to cut
down the level of indent. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sat, 18 Apr 2020 03:11:05 +0000 (05:11 +0200)]
net: pcnet: Switch to PCI memory access
Replace the PCI IO access with PCI memory access, the card
supports both, but the former does not work with QEMU SH4.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sat, 18 Apr 2020 00:32:19 +0000 (02:32 +0200)]
net: pcnet: Replace mips-specific accessors
Replace mips-specific UNCACHED_SDRAM() macro with standard
map_physmem(), which permits the driver to work on other
systems than mips.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Fri, 17 Apr 2020 23:56:51 +0000 (01:56 +0200)]
net: tulip: Remove CONFIG_TULIP_*
These macros are not used by any board, remove them to simplify
the driver. The EEPROM accessors are still retained however, as
those might still be useful.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Fri, 17 Apr 2020 23:44:12 +0000 (01:44 +0200)]
net: pcnet: Remove CONFIG_PCNET_79C97x
These macros guard one switch-case statement, which grows mips malta
by some 20 bytes if debug is enabled, and even less if it is not. To
make the code simpler, just support all the NICs and be done with it.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Marek Vasut [Sat, 4 Apr 2020 10:21:57 +0000 (12:21 +0200)]
net: Fix warning when including netdev.h on DM systems
If the DM_ETH is enabled and netdev.h is included somewhere, the
struct eth_device may not be defined, yet it is used in the header
file as an argument to fecmxc_register_mii_postcall. Add forward
declaration to remove the warning.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Joe Hershberger <joe.hershberger@ni.com>
Add U-Boot specific dts file for hifive-unleashed-a00, this
would help to add u-boot specific properties and other node
changes without touching the base dts(i) files which are easy
to sync from Linux.
Added spi2 alias for qspi2 as an initial u-boot specific
property change.
spi probing in current dm model is very much rely on aliases
numbering. Even though the qspi2 can't come under any associated
spi nor flash it would require to specify the same to make proper
binding happen for other spi slaves.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Tested-by: Sagar Kadam <sagar.kadam@sifive.com>
Jagan Teki [Thu, 23 Apr 2020 17:00:55 +0000 (22:30 +0530)]
spi: sifive: Fix QPP transfer
The guessed reason is that the existing logic of filling
tx fifo with data, rx fifo with NULL for tx transfer and
filling rx fifo with data, tx fifo with NULL for rx transfer
is not clear enough to support the Quad Page Program.
SiFive SPI controllers have specific sets of watermark
registers and SPI I/O directions bits in order to program
SPI controllers clear enough to support all sets of operating
modes.
Here is the exact programing sequence that would follow on this
patch and tested via SPI-NOR and MMC_SPI.
- set the frame format proto, endian
- set the frame format dir, set it for tx and clear it for rx
- TX transfer:
fill tx fifo with data.
wait for TX watermark bit to clear.
- RX transfer:
fill tx fifo with 0xff.
write nbytes to rx watermark register
wait for rx watermark bit to clear.
read the rx fifo data.
So, this patch adopts this program sequence and fixes the existing
I/O direction bit.
Cc: Vignesh R <vigneshr@ti.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Tested-by: Sagar Kadam <sagar.kadam@sifive.com>
Jagan Teki [Thu, 23 Apr 2020 17:00:54 +0000 (22:30 +0530)]
spi: sifive: Fix format register proto field
SiFive SPI controller has a proto bit field in frame format
register which would be used to configure the SPI I/O protocol
lines used on specific transfer.
Right now the driver is configuring this proto using slave->mode,
for all types of transctions. This makes the driver unable to
function since the proto needs to configure dynamically for
each and every transaction separately at runtime.
Now, the controller driver supports per transfer via spi-mem
exec_opo, so add the fmt_proto flag and fill the per transfer
buswidth so that the controller configures the proto bit at
runtime.
This patch fixes the SPI controller works with SPI NOR flash
on quad read with page program.
Cc: Vignesh R <vigneshr@ti.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Tested-by: Sagar Kadam <sagar.kadam@sifive.com>
Jagan Teki [Thu, 23 Apr 2020 17:00:53 +0000 (22:30 +0530)]
spi: sifive: Add spi-mem exec op
SiFive SPI controller is responsible to handle the
slave devices like mmc spi and spi nor flash.
The controller is designed such a way that it would
handle the slave transactions based on the I/O protocol
numbers, example if spi nor slave send quad write opcode
it has to send alone with I/O protocol number of 4 and
if it try to send data it has to send I/O protocol number
along with 4 line data.
But the current spi-xfer code from spi-mem is combining
the opcode and address in a single transaction, so the
SPI controller will be unable to identify the I/O protocol
number of opcode vs address.
So, add the spi-mem exec_op with spi-xfer of opcode, address
and data as a separate transaction. This doesn't remove
the .xfer of dm_spi_ops since mmc spi will make use of it.
Note: This code might have moved to the spi-mem core area
once we have done the dedicated tests on other controllers
and have real reason to move.
Cc: Vignesh R <vigneshr@ti.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Tested-by: Sagar Kadam <sagar.kadam@sifive.com>
Jagan Teki [Fri, 10 Apr 2020 18:26:31 +0000 (23:56 +0530)]
rk3399: Enable SF distro bootcmd
Enable SPI flash(SF) distro boot command in rk3399.
This distro boot will read the boot script at specific
location at the flash and start sourcing the same.
Included the SF device at the last of the target devices
list since all the rest of the devices on the list have
more possibility to boot the distribution due to the
size of the SPI flash is concern.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
- Add DM_ETH support for DPAA1, DPAA2 based RDB platforms: ls1046ardb,
ls1043ardb, lx2160ardb, ls2088ardb, ls1088ardb.
- Add GICv3 support for ls1028a, ls2088a, ls1088a.
- Add lpuart support on ls1028aqds.
- Few bug fixes and updates on ls2088a, ls1012a, ls1046a, ls1021a based
platforms.
Simon Glass [Sun, 26 Apr 2020 15:12:58 +0000 (09:12 -0600)]
x86: Move coreboot-table detection into common code
To support detecting booting from coreboot, move the code which locates
the coreboot tables into a common place. Adjust the algorithm slightly to
use a word comparison instead of string, since it is faster.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
[bmeng: correct the comments to 960KB] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sun, 26 Apr 2020 15:12:56 +0000 (09:12 -0600)]
pci: Avoid auto-config when chain loading
When U-Boot is not the first-stage bootloader we don't want to
re-configure the PCI devices, since this has already been done. Add a
check to avoid this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sun, 26 Apr 2020 15:12:55 +0000 (09:12 -0600)]
x86: cpu: Skip init code when chain loading
When U-Boot is not the first-stage bootloader the interrupt and cache init
must be skipped, as well as init for various peripherals. Update the code
to add checks for this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Sun, 26 Apr 2020 15:12:53 +0000 (09:12 -0600)]
x86: fsp: Allow skipping init code when chain loading
It is useful to be able to boot the same x86 image on a device with or
without a first-stage bootloader. For example, with chromebook_coral, it
is helpful for testing to be able to boot the same U-Boot (complete with
FSP) on bare metal and from coreboot. It allows checking of things like
CPU speed, comparing registers, ACPI tables and the like.
When U-Boot is not the first-stage bootloader much of this code is not
needed and can break booting. Add checks for this to the FSP code.
Rather than checking for the amount of available SDRAM, just use 1GB in
this situation, which should be safe. Using 2GB may run into a memory
hole on some SoCs.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
projects / u-boot.git / log