These platforms return to romstage from FSP only after
already having torn CAR down. A copy of the entire CAR
region is available and discoverable via HOB.
Previously, CBMEM console detected on-the-fly that CAR
migration had happened and relocated cbmem_console_p
accoringlin with car_sync_var(). However, if the CAR_GLOBAL
pointing to another object inside CAR is a relative offset
instead, we have a more generic solution that can be used
with timestamps code as well.
Change-Id: Ica877b47e68d56189e9d998b5630019d4328a419
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35140
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Tested on qemu-riscv.
Depends on OpenSBI integration and proper memory detection in qemu.
Boots into Linux until initrd should be loaded.
Tested on SiFive/unleashed:
Boots into Linux until earlycon terminates.
Change-Id: I5ebc6cc2cc9e328f36d70fba13555386bb8c29d6
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/30292
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Call OpenSBI in M-Mode and use it to set up SBI and to lockdown the
platform. It will also jump to the specified payload when done.
This behaviour is similar to BL31 on aarch31.
The payload is 41KiB in size on qemu.
Tested on qemu-riscv:
Required to boot a kernel as OpenSBI's instruction emulation feature
is required on that virtual machine.
Tested on SiFive/unleashed:
The earlycon is working. No console after regular serial driver
should take over, which might be related to kernel config.
Change-Id: I2a178595bd2aa2e1f114cbc69e8eadd46955b54d
Signed-off-by: Xiang Wang <merle@hardenedlinux.org>
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32394
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
There are only minimal differences between the architecture specific
stdint.h implementations, so let's tidy them up and merge them together
into a single file. In particular,
- Use 'unsigned long' for uintptr_t. This was already the case for x86
and riscv, while arm and mips used 'unsigned int', and arm64 and ppc64
used 'unsigned long long'. This change allows using a single integer
type for uintptr_t across all architectures, and brings it into
consistency with the rest of the code base, which generally uses
'unsigned long' for memory addresses anyway. This change required
fixing several assumptions about integer types in the arm code.
- Use _Bool as the boolean type. This is a specialized boolean type that
was introduced in C99, and is preferrable over hacking booleans
using integers. romcc sadly does not support _Bool, so for that we
stick with the old uint8_t.
- Drop the least and fast integer types. They aren't used
anywhere in the code base and are an unnecessary maintenance burden.
Using the standard fixed width types is essentially always better anyway.
- Drop the UINT64_C() macro. It also isn't used anywhere and doesn't
provide anything that a (uint64_t) cast doesn't.
- Implement the rest of the MIN and MAX numerical limits.
- Use static assertions to check that the integer widths are correct.
Change-Id: I6b52f37793151041b7bdee9ec3708bfad69617b2
Signed-off-by: Jacob Garber <jgarber1@ualberta.ca>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34075
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Also don't define the default as this results in spurious lines in the
.config.
The only difference in the generated config.h is that for most board
ARCH_RISCV_M goes from 1 to 0. This should not matter.
Change-Id: I3e8c1cc5696d621e243696a3b5e34f62ab69a688
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31311
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The printk has no effect as console_init() hasn't been called.
Also drop unused variables and headers.
Change-Id: I5bf5a8822c69bbcc3de1de460d19585b8330649f
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33786
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Philipp Hug <philipp@hug.cx>
On RISC-V the argument to a payload is always the hartid and a pointer to a FDT.
selfboot sets the coreboot tables as an argument, work around this here.
Change-Id: If6929897c7f12d8acb079eeebaef512ae506ca8b
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31477
Reviewed-by: ron minnich <rminnich@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Fixes a logic error that sets MPIE, but didn't use mret to return to the payload.
This left MIE set to an undefined value.
Now all modes are handled the same way:
- Trap vector base address point to the payload
- Disable Interrupt
- Return to payload using mret
TEST=Run an M-mode payload
Change-Id: Iaab595f916949c57104ec00f8b06ea047fe76bba
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33462
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Drop 'include <string.h>' when it is not used and
add it when it is missing.
Also extra lines removed, or added just before local includes.
Change-Id: Iccac4dbaa2dd4144fc347af36ecfc9747da3de20
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31966
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
This patch is a raw application of
find src/ -type f | xargs sed -i -e 's/IS_ENABLED\s*(CONFIG_/CONFIG(/g'
Change-Id: I6262d6d5c23cabe23c242b4f38d446b74fe16b88
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31774
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
MMIO operations are arch-agnostic so the include
path should not be arch/.
Change-Id: I0fd70f5aeca02e98e96b980c3aca0819f5c44b98
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/31691
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* Adding separate targets for 32bit and 64bit qemu
* Using the riscv64 toolchain for 32bit builds requires setting -m elf32lriscv
* rv32/rv64 is currently configured with ARCH_RISCV_RV32/RV64 and not per stage.
This should probably be changed later.
TEST=Boots to "Payload not loaded." on 32bit qemu using the following commands:
util/riscv/make-spike-elf.sh build/coreboot.rom build/coreboot.elf
qemu-system-riscv32 -M virt -m 1024M -nographic -kernel build/coreboot.elf
Change-Id: I35e59b459d1770df10b51fe9e77dcc474d7c75a0
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/c/31253
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: ron minnich <rminnich@gmail.com>
Also show hart id in trap information for easier debugging.
Change-Id: I20acf86e1af111600c158295ae03b2167838d127
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/c/31201
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: ron minnich <rminnich@gmail.com>
1. Simplify payload code and convert it to C
2. Save the FDT pointer to HLS (hart-local storage).
3. Don't use mscratch to pass FDT pointer as it is used for exception handling.
Change-Id: I32bf2a99e07a65358a7f19b899259f0816eb45e8
Signed-off-by: Xiang Wang <wxjstz@126.com>
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/c/31179
Reviewed-by: ron minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
ARCH_RISCV_RV{32,64} will now select ARCH_RISCV.
Change-Id: Ia7a1a8f0bfab20e91b8429dd6dd3e9a4180a0a5b
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/c/31042
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Hug <philipp@hug.cx>
RISCV parts can be created with any one of four CPU modes enabled,
with or without PMP, and with either 32 or 64 bit XLEN.
In anticipation of parts to come, create the Kconfig variables for these
architecture attributes.
Change-Id: I32ee51b2a469c7684a2f1b477bdac040e972e253
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/c/30348
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Quoting from the RISC-V Privileged Architecture manual version 1.10,
chapter 3.1.11:
The FS and XS fields use the same status encoding as shown in Table
3.3, with the four possible status values being Off, Initial, Clean,
and Dirty.
Status FS Meaning XS Meaning
0 Off All off
1 Initial None dirty of clean, some on
2 Clean None dirty, some clean
3 Dirty Some dirty
Change-Id: If0225044ed52215ce64ea979d120014e02d4ce37
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/c/28987
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
They are hopefully stable enough by now.
TEST=Building with for emulation/spike-riscv with BUILD_TIMELESS,
with and without this patch, results in the same coreboot.rom.
Change-Id: Ie6747c7eeea6cd8fd2138c5ba535a08c5add9038
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/c/30164
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Use CONFIG_CPU_MAX which defaults to 1 instead of CONFIG_RISCV_HART_NUM.
The default value of CONFIG_RISCV_HART_NUM was 0 and cause a jump to address 0.
Add a die() call to fail gracefully.
Change-Id: I4e3aa09b787ae0f26a4aae375f4e5fcd745a0a1e
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/c/29993
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Xiang Wang <wxjstz@126.com>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Each stage performs some basic initialization (stack, HLS etc) and then
call smp_pause to enter the single-threaded state. The main work of each
stage is executed in a single-threaded state, and the multi-threaded
state is restored by call smp_resume while booting the next stage.
Change-Id: I8d508c3d0f65a022010e74f8edad7ad2cfdc7dee
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/29024
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Philipp Hug <philipp@hug.cx>
See https://doc.coreboot.org/arch/riscv/ we know that we need to execute
smp_pause at the start of each stage and smp_resume at the end of each
stage.
Change-Id: I6f8159637bfb15f54f0abeb335de2ba6e9cf82fb
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/29023
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Just disable the timer interrupt and notify supervisor.
To receive another timer interrupt just set timecmp and
enable machine mode timer interrupt again.
TEST=Run linux on sifive unleashed
Change-Id: I5d693f872bd492c9d0017b514882a4cebd5ccadd
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/29340
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Pointer to opcode increases by unit uint16_t not byte.
Change-Id: I2986ca5402ad86d80e0eb955478bfbdc5d50e1f5
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/29339
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Bounce buffers used to be used in those cases where the payload
might overlap coreboot.
Bounce buffers are a problem for rampayloads as they need malloc.
They are also an artifact of our x86 past before we had relocatable
ramstage; only x86, out of the 5 architectures we support, needs them;
currently they only seem to matter on the following chipsets:
src/northbridge/amd/amdfam10/Kconfig
src/northbridge/amd/lx/Kconfig
src/northbridge/via/vx900/Kconfig
src/soc/intel/fsp_baytrail/Kconfig
src/soc/intel/fsp_broadwell_de/Kconfig
The first three are obsolete or at least could be changed
to avoid the need to have bounce buffers.
The last two should change to no longer need them.
In any event they can be fixed or pegged to a release which supports
them.
For these five chipsets we change CONFIG_RAMBASE from 0x100000 (the
value needed in 1999 for the 32-bit Linux kernel, the original ramstage)
to 0xe00000 (14 Mib) which will put the non-relocatable x86
ramstage out of the way of any reasonable payload until we can
get rid of it for good.
14 MiB was chosen after some discussion, but it does fit well:
o Fits in the 16 MiB cacheable range coreboot sets up by default
o Most small payloads are well under 14 MiB (even kernels!)
o Most large payloads get loaded at 16 MiB (especially kernels!)
With this change in place coreboot correctly still loads a bzImage payload.
Werner reports that the 0xe00000 setting works on his broadwell systems.
Change-Id: I602feb32f35e8af1d0dc4ea9f25464872c9b824c
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/28647
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
These codes are written by me based on the privileged instruction set.
I tested it by qemu/riscv-probe.
Change-Id: I2e9e0c94e6518f63ade7680a3ce68bacfae219d4
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/28569
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Its spreading copies got out of sync. And as it is not a standard header
but used in commonlib code, it belongs into commonlib. While we are at
it, always include it via GCC's `-include` switch.
Some Windows and BSD quirk handling went into the util copies. We always
guard from redefinitions now to prevent further issues.
Change-Id: I850414e6db1d799dce71ff2dc044e6a000ad2552
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/28927
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
coreboot does not set up virtual memory anymore.
Change-Id: I231af07b2988e8362d1cdd606ce889fb31136ff1
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/28831
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Clang doesn't understand -march=riscv64imac and -mcmodel=medany, so
don't use them when running the clang static analyzer. On the other
hand, __riscv and __riscv_xlen need to be defined in order to select
some macros in src/arch/riscv/include/arch/encoding.h. __riscv_flen
selects the floating-point paths in src/arch/riscv/misaligned.c.
-mabi is moved with -march for consistency.
A complete list of preprocessor definitions on RISC-V can be found at
https://github.com/riscv/riscv-toolchain-conventions#cc-preprocessor-definitions
With this commit, scan-build produces a useful result on RISC-V.
Change-Id: Ia2eb8c3c2f7eb5ddd47db24b8e5fcd6eaf6c5589
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/28713
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
After emulating an instruction in the misaligned load/store handler, we
need to increment the program counter by the size of instruction.
Otherwise the same instruction is executed (and emulated) again and again.
While were at it: Also return early in the unlikely case that the
faulting instruction is not 16 or 32 bits long, and be more explicit
about the return values of fetch_*bit_instruction.
Tested by Philipp Hug, using the linuxcheck payload.
Fixes: cda59b56ba ("riscv: update misaligned memory access exception handling")
Change-Id: Ie2dc0083835809971143cd6ab89fe4f7acd2a845
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/28617
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
XS is a read-only field of mstatus. Unable to be write. So remove this code.
Change-Id: I3ad6b0029900124ac7cce062e668a0ea5a8b2c0e
Signed-off-by: Xiang Wang <wxjstz@126.com>
Reviewed-on: https://review.coreboot.org/28357
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
On the FU540 the bootblock runs on a core without lesser privilege
modes, so the medeleg/mideleg CSRs are not implemented on that core,
leading to a CPU exception when these CSRs are accessed.
Configure medeleg/mideleg only if the misa register indicates that
S-mode is implemented on the executing RISC-V core.
Change-Id: Idad97e42bac2ff438dd233a5d125f93594505d63
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/25791
Reviewed-by: Xiang Wang <wxjstz@126.com>
Reviewed-by: Philipp Hug <philipp@hug.cx>
Reviewed-by: Johanna Schander <coreboot@mimoja.de>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Only execute coreboot on hart 0 until synchronisation between hart's is ready.
Change-Id: I2181e79572fbb9cc7bee39a3c2298c0dae6c1658
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/28605
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The RISC-V Privileged Architecture specification defines the Machine
Time Registers (mtime and mtimecmp) in section 3.1.15.
Makes it possible to use the generic udelay.
The timer is enabled using RISCV_USE_ARCH_TIMER for the lowrisc,
sifive and ucb soc.
Change-Id: I5139601226e6f89da69e302a10f2fb56b4b24f38
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/27434
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Make it uniform as other architectures also include it in io.h
Change-Id: I62c2d909c703f01cdaabdaaba344f82b6746f094
Signed-off-by: Philipp Hug <philipp@hug.cx>
Reviewed-on: https://review.coreboot.org/28601
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Elyes HAOUAS