When relocation is not required (i.e. _load_start == _link_start), we
can skip two stage wait for secondary HARTs. This means secondary HARTs
can skip the _wait_relocate_copy_done() loop and directly jump to the
_wait_for_boot_hart() loop when no relocation not required.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch extends fw_base to relocate to link address whenever
firmware load address is not equal to link address.
The relocation will not work when load start to load end overlap
link start to link end.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Tested-by: Lukas Auer <lukas.auer@aisec.fraunhofer.de>
The unpriv load/store instruction from M-mode can cause page/access
fault to M-mode if S-mode page table did not have mappings OR it did
not have PMP access permission.
To tackle this, we redirect trap back to S-mode if unpriv load/store
instruction traps in M-mode.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
This patch adds per-HART trap info pointer which can be used to
communicate trap information to sbi_trap_handler().
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
Currently, the unprivilege load/store functions are inline functions.
We will be extending these functions to track whether a page/access
fault occurs when we execute unprivilege load/store instruction.
To make things simpler and debugable, we reduce number of places which
can potentially generate a page/access fault by making all unprivilege
load/store functions as regular (non-inline) functions.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
This patch factor-out TLB management from IPI management to separate
sources sbi_tlb.c and sbi_tlb.h.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
We have extra space above scratch space (sbi_scratch) which we are
currently using to manage per-HART IPI data and TLB request management.
In future, more parts of OpenSBI will use the extra scratch space so
it will become difficult to manage extra scratch space using just
defines and macros.
This patch adds a simple brain-dead allocator to manage extra scratch
space. This allocator never expects anything to be free-ed hence it
keeps incrementing to next allocation offset until it runs-out of space.
In future, we can have more sophisticated allocator which will allow
us to re-claim free-ed space and also allows us to track owner of
allocated space.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
On latest Linux kernel (i.e. 5.2-rc1), we get large TLB flush
request for user space addresses (typically, start=x and end=-1).
This is caused by Linux kernel commit a21344dfc6 ("riscv: fix
sbi_remote_sfence_vma{,_asid}").
It's not practical to execute large number of sfence instructions
for a large TLB flush range because it takes too much time and
eventually causes CPU stall in Linux kernel.
This patch addresses above issue by upgrading TLB flush range to
TLB flush all whenever TLB flush range is greater than 1GB.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
This patch enables FW_DYNAMIC option (i.e. dynamic firmware) for
all appropriate/applicable platforms.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
This patch provides first-cut implementation of fw_dynamic firmware.
As compared to fw_jump and fw_payload, the fw_dynamic obtains next
address, next mode and OpenSBI options from struct fw_dynamic_info.
The previous booting stage can create struct fw_dynamic_info in memory
and pass address of struct fw_dynamic_info in 'a2' register. Also, the
struct fw_dynamic_info has versioning as well so changes to the struct
fw_dynamic_info can be done in a backward compatible manner.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
This patch extends existing firmwares (i.e. fw_jump and fw_payload)
to explicitly provide next mode and options to fw_base.
We also introduce fw_save_info() which is called by fw_base very
early on boot HART. This function can be used by existing firmwares
(i.e. fw_jump and fw_payload) to save information passed by previous
booting stage.
Overall, this is a preparatory patch for implementing fw_dynamic.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
We add CONTRIBUTORS.md file to provide a list of individuals and
organizations actively contributing to the OpenSBI project.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch extends our current build-system for building platform
sources which are not part of OpenSBI sources.
For example:
Let's say we have out-of-tree ABC platform sources. We place these
sources under <XYZ>/ABC directory along with its config.mk and
objects.mk.
To build out-of-tree ABC platform from OpenSBI directory:
$ make PLATFORM_DIR=<XYZ>/ABC
OR
$ make PLATFORM_DIR=<XYZ> PLATFORM=ABC
To build out-of-tree ABC platform from <XYZ>/ABC directory:
$ make PLATFORM_DIR=<XYZ>/ABC -C <path_to_opensbi>
OR
$ make PLATFORM_DIR=<XYZ> PLATFORM=ABC -C <path_to_opensbi>
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
This patch reduces memory consumed by struct sbi_fifo by droping
redundant "head" member and using u16 in-place of "unsigned long".
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch does following improvements to sbi_fifo:
1. Use valid SBI_Exxxx error codes instead of -1
2. The sbi_fifo_is_full() and sbi_fifo_is_empty() did
not acquire qlock before accessing head and tail
hence fixed it
3. Added avail member for ease in debugging and simplifying
head/tail updates.
Due to above changes size of sbi_fifo changes from 48 bytes
to 56 bytes.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We now have an OpenSBI mailing list available for development
and discussiong so update contributing.md accordingly. The
Github PR based review and issue tracking will also continue
to exist.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The mstatus parameter of get_insn() is used to return MSTATUS CSR
value which get_insn() saw. Most of the get_insn() callers don't
use the value returned in mstatus so this patch makes mstatus
parameter optional for get_insn().
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The sbi_unpriv.h has quite a few load_xyz() and store_xyz() helper
routines based on RISC-V inline assembly for unpriviledged accesses
from M-mode. These helper routines are similar to helper routines
present in riscv_locks.h, riscv_io.h, and riscv_atomic.h so let's
rename sbi_unpriv.h to riscv_unpriv.h.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch adds scripts/create-binary-archive.sh which helps
us create a common tarball containing headers, static library
and firmware ELFs for all platforms.
This script can be used for release purpose OR for compile
testing all platforms.
Example usage commands are as follows:
1) Binary release archive for 32bit systems
./scripts/create-binary-archive.sh -s "bin" -x 32 -d
2) Binary release archive for 64bit systems
./scripts/create-binary-archive.sh -s "bin" -d
3) Compile test for 32bit systems
./scripts/create-binary-archive.sh -x 32 -t
4) Compile test for 64bit systems
./scripts/create-binary-archive.sh -t
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The compile_ld() is actually used to create ELF files so
the name is misleading hence this patch renames it to
compile_elf(). We also rename LDFLAGS to ELFFLAGS because
these will be used for ELF creation only.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch extends our low-level trap handler in fw_base.S for
handling M-mode to M-mode traps without overwritting stack.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The unsigned long is always machine word size. This means it is
4 bytes on 32bit system and 8 bytes on 64bit system.
This patch adds PRILX define for sbi_printf() which will help us
print unsigned long without worrying whether it is 32bit or 64bit
system.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch replaces use of awk with sed in top-level makefile
to parse OpenSBI version from include/sbi/sbi_version.h.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We get following error evaluating OPENSBI_CC_XLEN on
Ubuntu-18.04:
/bin/sh: 1: Bad substitution
This patch fixes above error by using "awk" in string
assigned to OPENSBI_CC_XLEN and it also moves "Setup
compilation commands" before OPENSBI_CC_XLEN.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The mask shift in for-loop of sbi_ipi_send_many() is
broken with commit 918c1354b7
("lib: Improve delivery of SBI_IPI_EVENT_HALT")
This patch fix it.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch introduces following optional PLATFORM options:
PLATFORM_RISCV_XLEN -> RISC-V register width
PLATFORM_RISCV_ABI -> RISC-V GCC ABI
PLATFORM_RISCV_ISA -> RISC-V GCC ISA string
PLATFORM_RISCV_CODE_MODEL -> RISC-V GCC Code Model
If the above options are not provided by platform config.mk
or by command-line parameters then:
1. PLATFORM_RISCV_XLEN will be determined using toolchain
capability
2. PLATFORM_RISCV_ABI, PLATFORM_RISCV_ISA, and
PLATFORM_RISCV_CODE_MODEL is set to value best suited for
generic libsbi.a
As a result of these optional PLATFORM options, the
platform-cflags-y and platform-asflags-y is further
simplified for platform config.mk.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The current 4MB aligned FW_JUMP_ADDR and FW_PAYLOAD_OFFSET breaks
U-Boot on QEMU virt and sifive_u machines.
Instead of using 4MB aligned for both 32bit and 64bit systems, we
use different values based compiler XLEN. Another advantage of this
approach will be that our fw_payload.bin will smaller for 64bit
systems.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Both FW_JUMP_ADDR and FW_PAYLOAD_OFFSET, should be:
1. 4MB aligned for 32bit system
2. 2MB aligned for 64bit system
Explicitly specify the above details in config.mk comments
for template platform.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch updates doxygen config file to include
docs/platform_guide.md and docs/library_usage.md in
generated PDF document.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Currently, the OpenSBI version is in top-level Makefile so
firmware linking to OpenSBI static library have no-way to
know OpenSBI version.
This patch moves OpenSBI version from top-level Makefile to
sbi/sbi_version.h header which provides OPENSBI_VERSION_MAJOR
and OPENSBI_VERSION_MINOR defines.
NOTE: the SBI spec (or SBI ecall interface) version is
different. The SBI spec version is provided by functions
sbi_ecall_version_major() and sbi_ecall_version_minor().
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch updates copyright header in all files as follows:
1. Makes "SPDX-License-Identifier: BSD-2-Clause" as first line
2. Change copyright year to 2019 for Western Digital
Signed-off-by: Anup Patel <anup.patel@wdc.com>
For better naming, we rename ipi_inject() to ipi_send() in
struct sbi_platform. We also replace term "inject" with
"send" in all related places.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
It is not necessary that platform has MMIO-based timer value
register. It can also have some custom (implementation specific)
CSR for timer value.
This patch renames SBI_PLATFORM_HAS_MMIO_TIMER_VALUE to
SBI_PLATFORM_HAS_TIMER_VALUE to imply "platform timer value"
instead of "mmio timer value".
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The target_hart and hartid paramter of TIMER callbacks is not
required because it always current hartid which can be obtained
using sbi_current_hartid() API.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The hartid parameter in IRQCHIP callbacks of sbi_platform
is not required because current hartid can be determined
using sbi_current_hartid() API.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The source_hart and hartid parameter is really not required in
IPI callbacks of sbi_platform because current hartid can always
be obtained by calling sbi_current_hartid() API.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch adds doxygen style documenation for struct sbi_platform
and related functions/macros/defines.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We don't need to pre-enable MSIP in MIE CSR when
calling sbi_init() from firmware. This patch updates
documentation accordingly.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We don't need to install generic headers separately for every
platforn under <install_dir>/platform. It is unnecessary duplication
of files hence updating install_libplatsbi in top-level makefile.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The PDF document generated by doxygen should have OpenSBI version
hence this patch adds OpenSBI version to doxygen.cfg.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We should use AMO instructions whenever __riscv_atomic is
defined (i.e. atomics are supported). We use LR/SC only when
__riscv_atomic is not defined.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Avoid using standard includes namely stdint.h, string.h, stdlib.h, etc.
All standard include except stddef.h give compilation error due
to differences in RISC-V toolchain configuration. Typically, the
compilation error is related to "gnu-stubs-lp64.h".
This patch fixes compile error caused by standard includes by
providing substitutes of definetions provided by standard includes
wherever required.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch adds doxygen style documentation for struct sbi_trap_regs
and related macros/defines/functions.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
There are quite a few noreturn functions in OpenSBI hence
we add commong __noreturn define for noreturn functin attribute.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch extends top-level makefile to build and install
documentation. The 'docs' make target is for building the
documentation PDF whereas 'install_docs' make target is for
installing the documentation PDF.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The '?=' will not assign value if the target make variable is
overriden via command-line or is set in environment variable.
On other hand, '=' will not assign value only if variable is
overriden via command-line parameter.
It is quite common to have CC, AS, CPP, LD, etc to be set as
environment variables pointing to native compiler and binutils.
If '-rR' option is not set in MAKEFLAGS then this results in
compile error because '?=' will use the native compiler and
binutils. If '-rR' option is set in MAKEFLAGS then this again
results in compile error because CC, AS, CPP, etc are set to
empty strings which causes '?=' to use empty string as compiler
and binutils.
To handle this, we use '?=' only when CROSS_COMPILE is not
defined and we use '=' when CROSS_COMPILE is defined.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The tmp0 member was added in struct sbi_scratch to assist
register save/restore at time of trap handling. This tmp0
is not unsed any more hence removing it.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The struct sbi_trap_regs related defines RISCV_TRAP_REGS_xyz should
be in sbi_trap.h so that we can keep these defines in-sync with changes
in struct sbi_trap_regs.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The struct sbi_platform related defines RISCV_PLATFORM_xyz should
be in sbi_platform.h so that we can keep these defines in-sync
with changes in struct sbi_platform.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The struct sbi_scratch related defines RISCV_SCRATCH_xyz should
be in sbi_scratch.h so that we can keep these defines in-sync
with changes in struct sbi_scratch.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We replace #csr with __ASM_STR(csr) in all csr_xyz() macros
so that we can pass CSR number instead of CSR name when GCC
is not aware of CSR name.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The patch removes redundant csr_read_n() and csr_write_n()
because same thing can be achieved by using __ASM_STR()
macro in csr_read() and csr_write() macros.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Currently, we fail with error SBI_ENOTSUPP when we are not able
to handle illegal instruction trap. Instead, we should just
redirect illegal instruction trap to S-mode when not handled.
This redirection of illegal instruction trap will help lazy
save/restore of floating point registers to work correctly in
Linux kernel.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The mtval CSR is very useful information when debugging hence
print it upon trap error in sbi_trap_error().
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We rename target_hart parameter of PMP platform callbacks to
hartid for clarity because the parameter represents current
hartid.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Instead of having separate timer_init() hooks for cold and
warm boot, this patch updates struct sbi_platform to have just
one timer_init() hook. The type of boot (cold or warm) is now
a boolean flag parameter for the updated timer_init() hook.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Instead of having separate ipi_init() hooks for cold and warm boot,
this patch updates struct sbi_platform to have just one ipi_init()
hook. The type of boot (cold or warm) is now a boolean flag parameter
for the updated ipi_init() hook.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Instead of having separate irqchip_init() hooks for cold and
warm boot, this patch updates struct sbi_platform to have just
one irqchip_init() hook. The type of boot (cold or warm) is now
a boolean flag parameter for the updated irqchip_init() hook.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Instead of having separate early_init() and final_init() hooks
for cold and warm boot, this patch updates struct sbi_platform
to have just one early_init() and one final_init() hook. The
type of boot (cold or warm) is now a boolean flag parameter for
the updated early_init() and final_init() hooks.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The fw_common.S is the base firmware extendend by fw_jump and
fw_payload. This patch renames fw_common.S to fw_base.S to
have more clear/intutive name for base firmware.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This improves 'clean' and 'distclean' makefile target as follows:
1. Remove only .o, .a, .elf, and .bin files for 'clean'
2. Remove .dep in-addition to what 'clean' does for 'distclean'
3. Remove default build and install directory for 'distclean'
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We now have payloads sub-directory under firmware directory which
means payload BINs and ELFs should be installed under firmware/payloads.
This patch improves inst_file_list() to handle payloads sub-directory
under firmware installation directory.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Currently, the default payload for FW_PAYLOAD is embedded
fw_payload.S itself. This means people have to hack fw_payload.S
if they want to have some temporary S-mode test code.
This patch adds a separate dummy payload for FW_PAYLOAD which
can be easily hacked for some S-mode testing.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
This patch moves all ECALL defines to sbi_ecall_interface.h so
that it can be shared with firmware payloads.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We generally don't get misaligned load/store traps from Linux/U-Boot
compiled using GCC 8.2 or higher but this is not true with older
GCC toolchains. To tackle this we add misaligned load/store trap
handling adopted from BBL sources but much more simpler.
(Note: BBL sources can be found at https://github.com/riscv/riscv-pk.git)
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The name CROSS_COMPILE is an industry standard used across
open-source project to specify cross-compiler prefix.
In fact, distro build systems such as buildroot and yocto
assume CROSS_COMPILE to be supported by projects added
as packages to these build systems.
This patch rename back CROSS_COMPILE_PREFIX to CROSS_COMPILE.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The platform specific options will be PLATFORM_<xyz> and not
PLAT_<xyz> hence this patch fixes typo in top-level README.md.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The PLAT_HART_COUNT and PLAT_HART_STACK_SIZE are only used
by platform code so no need of exposing these to everyone
by adding it to platform-cppflags-y.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The hart_count and hart_stack_size information is already available
in "struct sbi_platform" so we use that instead of depending on
PLAT_HART_COUNT and PLAT_HART_STACK_SIZE.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
We don't need a separate stack section for per-HART stack
instead we create per-HART stack at the end of firmware
(i.e. after _fw_end symbol).
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Pass S-Mode and M-mode context id separately to common warm init.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Unleashed doesn't have S mode for hart 0.
PLIC register base calculation is current wrong because of that.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The math.h is not required anymore and it also gives
compile error for certain toolchains hence thsi patch
removes math.h include from kendryte/k210/sysctl.c.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
The mideleg and medeleg are already programmed in delegate_traps()
so no need to set it here.
Any CSR setup in our reference firmware becomes a requirement
for bootloader linking to libsbi.a so we should have minimum
possible CSR setup in our reference firmware.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Anup Patel