The microchip_icicle_kit, sifive_u, spike and virt boards are now doing
the same steps when '-kernel' is used:
- execute load_kernel()
- load init_rd()
- write kernel_cmdline
Let's fold everything inside riscv_load_kernel() to avoid code
repetition. To not change the behavior of boards that aren't calling
riscv_load_init(), add an 'load_initrd' flag to riscv_load_kernel() and
allow these boards to opt out from initrd loading.
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230206140022.2748401-3-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Next patch will move all calls to riscv_load_initrd() to
riscv_load_kernel(). Machines that want to load initrd will be able to
do via an extra flag to riscv_load_kernel().
This change will expose a sign-extend behavior that is happening in
load_elf_ram_sym() when running 32 bit guests [1]. This is currently
obscured by the fact that riscv_load_initrd() is using the return of
riscv_load_kernel(), defined as target_ulong, and this return type will
crop the higher 32 bits that would be padded with 1s by the sign
extension when running in 32 bit targets. The changes to be done will
force riscv_load_initrd() to use an uint64_t instead, exposing it to the
padding when dealing with 32 bit CPUs.
There is a discussion about whether load_elf_ram_sym() should or should
not sign extend the value returned by 'lowaddr'. What we can do is to
prevent the behavior change that the next patch will end up doing.
riscv_load_initrd() wasn't dealing with 64 bit kernel entries when
running 32 bit CPUs, and we want to keep it that way.
One way of doing it is to use target_ulong in 'kernel_entry' in
riscv_load_kernel() and rely on the fact that this var will not be sign
extended for 32 bit targets. Another way is to explictly clear the
higher 32 bits when running 32 bit CPUs for all possibilities of
kernel_entry.
We opted for the later. This will allow us to be clear about the design
choices made in the function, while also allowing us to add a small
comment about what load_elf_ram_sym() is doing. With this change, the
consolation patch can do its job without worrying about unintended
behavioral changes.
[1] https://lists.gnu.org/archive/html/qemu-devel/2023-01/msg02281.html
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230206140022.2748401-2-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
As it is now, riscv_compute_fdt_addr() is receiving a dram_base, a
mem_size (which is defaulted to MachineState::ram_size in all boards)
and the FDT pointer. And it makes a very important assumption: the DRAM
interval dram_base + mem_size is contiguous. This is indeed the case for
most boards that use a FDT.
The Icicle Kit board works with 2 distinct RAM banks that are separated
by a gap. We have a lower bank with 1GiB size, a gap follows, then at
64GiB the high memory starts. MachineClass::default_ram_size for this
board is set to 1.5Gb, and machine_init() is enforcing it as minimal RAM
size, meaning that there we'll always have at least 512 MiB in the Hi
RAM area.
Using riscv_compute_fdt_addr() in this board is weird because not only
the board has sparse RAM, and it's calling it using the base address of
the Lo RAM area, but it's also using a mem_size that we have guarantees
that it will go up to the Hi RAM. All the function assumptions doesn't
work for this board.
In fact, what makes the function works at all in this case is a
coincidence. Commit 1a475d39ef introduced a 3GB boundary for the FDT,
down from 4Gb, that is enforced if dram_base is lower than 3072 MiB. For
the Icicle Kit board, memmap[MICROCHIP_PFSOC_DRAM_LO].base is 0x80000000
(2 Gb) and it has a 1Gb size, so it will fall in the conditions to put
the FDT under a 3Gb address, which happens to be exactly at the end of
DRAM_LO. If the base address of the Lo area started later than 3Gb this
function would be unusable by the board. Changing any assumptions inside
riscv_compute_fdt_addr() can also break it by accident as well.
Let's change riscv_compute_fdt_addr() semantics to be appropriate to the
Icicle Kit board and for future boards that might have sparse RAM
topologies to worry about:
- relieve the condition that the dram_base + mem_size area is contiguous,
since this is already not the case today;
- receive an extra 'dram_size' size attribute that refers to a contiguous
RAM block that the board wants the FDT to reside on.
Together with 'mem_size' and 'fdt', which are now now being consumed by a
MachineState pointer, we're able to make clear assumptions based on the
DRAM block and total mem_size available to ensure that the FDT will be put
in a valid RAM address.
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230201171212.1219375-4-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
A common trend in other archs is to calculate the fdt address, which is
usually straightforward, and then calling a function that loads the
fdt/dtb by using that address.
riscv_load_fdt() is doing a bit too much in comparison. It's calculating
the fdt address via an elaborated heuristic to put the FDT at the bottom
of DRAM, and "bottom of DRAM" will vary across boards and
configurations, then it's actually loading the fdt, and finally it's
returning the fdt address used to the caller.
Reduce the existing complexity of riscv_load_fdt() by splitting its code
into a new function, riscv_compute_fdt_addr(), that will take care of
all fdt address logic. riscv_load_fdt() can then be a simple function
that just loads a fdt at the given fdt address.
We're also taken the opportunity to clarify the intentions and
assumptions made by these functions. riscv_load_fdt() is now receiving a
hwaddr as fdt_addr because there is no restriction of having to load the
fdt in higher addresses that doesn't fit in an uint32_t.
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Message-Id: <20230201171212.1219375-3-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
All callers are using kernel_filename as machine->kernel_filename.
This will also simplify the changes in riscv_load_kernel() that we're
going to do next.
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230102115241.25733-10-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
'filename', 'mem_size' and 'fdt' from riscv_load_initrd() can all be
retrieved by the MachineState object for all callers.
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230102115241.25733-9-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
riscv_load_initrd() returns the initrd end addr while also writing a
'start' var to mark the addr start. These informations are being used
just to write the initrd FDT node. Every existing caller of
riscv_load_initrd() is writing the FDT in the same manner.
We can simplify things by writing the FDT inside riscv_load_initrd(),
sparing callers from having to manage start/end addrs to write the FDT
themselves.
An 'if (fdt)' check is already inserted at the end of the function
because we'll end up using it later on with other boards that doesn´t
have a FDT.
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230102115241.25733-7-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The system controller on PolarFire SoC is access via a mailbox. The
control registers for this mailbox lie in the "IOSCB" region & the
interrupt is cleared via write to the "SYSREG" region. It also has a
QSPI controller, usually connected to a flash chip, that is used for
storing FPGA bitstreams and used for In-Application Programming (IAP).
Linux has an implementation of the system controller, through which the
hwrng is accessed, leading to load/store access faults.
Add the QSPI as unimplemented and a very basic (effectively
unimplemented) version of the system controller's mailbox. Rather than
purely marking the regions as unimplemented, service the mailbox
requests by reporting failures and raising the interrupt so a guest can
better handle the lack of support.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221117225518.4102575-4-conor@kernel.org>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The Fabric Interconnect Controllers provide interfaces between the FPGA
fabric and the core complex. There are 5 FICs on PolarFire SoC, numbered
0 through 4. FIC2 is an AXI4 slave interface from the FPGA fabric and
does not show up on the MSS memory map. FIC4 is dedicated to the User
Crypto Processor and does not show up on the MSS memory map either.
FIC 0, 1 & 3 do show up in the MSS memory map and neither FICs 0 or 1
are represented in QEMU, leading to load access violations while booting
Linux for Icicle if PCIe is enabled as the root port is connected via
either FIC 0 or 1.
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Message-Id: <20221117225518.4102575-3-conor@kernel.org>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Booting using "Direct Kernel Boot" for PolarFire SoC & skipping u-boot
entirely is probably not advisable, but it does at least show signs of
life. Recent Linux kernel versions make use of peripherals that are
missing definitions in QEMU and lead to kernel panics. These issues
almost certain rear their head for other methods of booting, but I was
unable to figure out a suitable HSS version that is recent enough to
support these peripherals & works with QEMU.
With these peripherals added, booting a kernel with the following hangs
hangs waiting for the system controller's hwrng, but the kernel no
longer panics. With the Linux driver for hwrng disabled, it boots to
console.
qemu-system-riscv64 -M microchip-icicle-kit \
-m 2G -smp 5 \
-kernel $(vmlinux_bin) \
-dtb $(dtb)\
-initrd $(initramfs) \
-display none -serial null \
-serial stdio
More peripherals are added than strictly required to fix the panics in
the hopes of avoiding a replication of this problem in the future.
Some of the peripherals which are in the device tree for recent kernels
are implemented in the FPGA fabric. The eMMC/SD mux, which exists as
an unimplemented device is replaced by a wider entry. This updated
entry covers both the mux & the remainder of the FPGA fabric connected
to the MSS using Fabric Interrconnect (FIC) 3.
Link: https://github.com/polarfire-soc/icicle-kit-reference-design#fabric-memory-map
Link: https://ww1.microchip.com/downloads/aemDocuments/documents/FPGA/ProductDocuments/SupportingCollateral/V1_4_Register_Map.zip
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220813135127.2971754-1-mail@conchuod.ie>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The 'fdt' param is not being used in riscv_setup_rom_reset_vec().
Simplify the API by removing it. While we're at it, remove the redundant
'return' statement at the end of function.
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Alistair Francis <alistair.francis@wdc.com>
Cc: Bin Meng <bin.meng@windriver.com>
Cc: Vijai Kumar K <vijai@behindbytes.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220728181926.2123771-1-danielhb413@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Commit 7c28f4da20 ("RISC-V: Don't add NULL bootargs to device-tree")
tried to avoid adding *NULL* bootargs to device tree, but unfortunately
the changes were entirely useless, due to MachineState::kernel_cmdline
can't be NULL at all as the default value is given as an empty string.
(see hw/core/machine.c::machine_initfn()).
Note the wording of *NULL* bootargs is wrong. It can't be NULL otherwise
a segfault had already been observed by dereferencing the NULL pointer.
It should be worded as *empty" bootargs.
Fixes: 7c28f4da20 ("RISC-V: Don't add NULL bootargs to device-tree")
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220421055629.1177285-2-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
When realising the SoC use error_fatal instead of error_abort as the
process can fail and report useful information to the user.
Currently a user can see this:
$ ../qemu/bld/qemu-system-riscv64 -M sifive_u -S -monitor stdio -display none -drive if=pflash
QEMU 6.1.93 monitor - type 'help' for more information
(qemu) Unexpected error in sifive_u_otp_realize() at ../hw/misc/sifive_u_otp.c:229:
qemu-system-riscv64: OTP drive size < 16K
Aborted (core dumped)
Which this patch addresses
Reported-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Message-Id: <20220105213937.1113508-8-alistair.francis@opensource.wdc.com>
drive_get_next() is basically a bad idea. It returns the "next" block
backend of a certain interface type. "Next" means bus=0,unit=N, where
subsequent calls count N up from zero, per interface type.
This lets you define unit numbers implicitly by execution order. If the
order changes, or new calls appear "in the middle", unit numbers change.
ABI break. Hard to spot in review.
A number of machines connect just one backend with drive_get_next().
Change them to use drive_get() directly. This makes the (zero) unit
number explicit in the code.
Cc: Beniamino Galvani <b.galvani@gmail.com>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Subbaraya Sundeep <sundeep.lkml@gmail.com>
Cc: Niek Linnenbank <nieklinnenbank@gmail.com>
Cc: Andrew Baumann <Andrew.Baumann@microsoft.com>
Cc: "Philippe Mathieu-Daudé" <f4bug@amsat.org>
Cc: Jean-Christophe Dubois <jcd@tribudubois.net>
Cc: Alistair Francis <Alistair.Francis@wdc.com>
Cc: Bin Meng <bin.meng@windriver.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Artyom Tarasenko <atar4qemu@gmail.com>
Cc: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Cc: Michael Tokarev <mjt@tls.msk.ru>
Cc: Laurent Vivier <laurent@vivier.eu>
Cc: qemu-arm@nongnu.org
Cc: qemu-riscv@nongnu.org
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20211117163409.3587705-3-armbru@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Message-id: 20211022060133.3045020-4-alistair.francis@opensource.wdc.com
Using memory_region_init_ram(), which can't possibly handle vhost-user,
and can't work as expected with '-numa node,memdev' options.
Use MachineState::ram instead of manually initializing RAM memory
region, as well as by providing MachineClass::default_ram_id to
opt in to memdev scheme.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 20211020014112.7336-2-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The RISC-V ACLINT is more modular and backward compatible with
original SiFive CLINT so instead of duplicating the original
SiFive CLINT implementation we upgrade the current SiFive CLINT
implementation to RISC-V ACLINT implementation.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Message-id: 20210831110603.338681-3-anup.patel@wdc.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
We will be upgrading SiFive CLINT implementation into RISC-V ACLINT
implementation so let's first rename the sources.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Message-id: 20210831110603.338681-2-anup.patel@wdc.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Instead of using riscv_cpu_update_mip() let's instead use the new RISC-V
CPU GPIO lines to set the external MIP bits.
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Message-id: 0364190bfa935058a845c0fa1ecf650328840ad5.1630301632.git.alistair.francis@wdc.com
At present the Microchip Icicle Kit machine only supports using
'-bios' to load the HSS, and does not support '-kernel' for direct
kernel booting just like other RISC-V machines do. One has to use
U-Boot which is chain-loaded by HSS, to load a kernel for testing.
This is not so convenient.
Adding '-kernel' support together with the existing '-bios', we
follow the following table to select which payload we execute:
-bios | -kernel | payload
------+------------+--------
N | N | HSS
Y | don't care | HSS
N | Y | kernel
This ensures backwards compatibility with how we used to expose
'-bios' to users. When '-kernel' is used for direct boot, '-dtb'
must be present to provide a valid device tree for the board,
as we don't generate device tree.
When direct kernel boot is used, the OpenSBI fw_dynamic BIOS image
is used to boot a payload like U-Boot or OS kernel directly.
Documentation is updated to describe the direct kernel boot. Note
as of today there is still no PolarFire SoC support in the upstream
Linux kernel hence the document does not include instructions for
that. It will be updated in the future.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20210430071302.1489082-8-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Many files include qemu/log.h without needing it. Remove the superfluous
include statements.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20210328054833.2351597-1-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Many files include hw/irq.h without needing it. Remove the superfluous
include statements.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20210327050236.2232347-1-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Since HSS commit c20a89f8dcac, the Icicle Kit reference design has
been updated to use a register mapped at 0x4f000000 instead of a
GPIO to control whether eMMC or SD card is to be used. With this
support the same HSS image can be used for both eMMC and SD card
boot flow, while previously two different board configurations were
used. This is undocumented but one can take a look at the HSS code
HSS_MMCInit() in services/mmc/mmc_api.c.
With this commit, HSS image built from 2020.12 release boots again.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20210322075248.136255-1-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
There is already a MemMapEntry type defined in hwaddr.h. Let's drop
the RISC-V defined `struct MemmapEntry` and use the existing one.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20210220144807.819-2-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The latest SD card image [1] released by Microchip ships a Linux
kernel with built-in PolarFire SoC I2C driver support. The device
tree file includes the description for the I2C1 node hence kernel
tries to probe the I2C1 device during boot.
It is enough to create an unimplemented device for I2C1 to allow
the kernel to continue booting to the shell.
[1] ftp://ftpsoc.microsemi.com/outgoing/core-image-minimal-dev-icicle-kit-es-sd-20201009141623.rootfs.wic.gz
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-11-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
When system memory is larger than 1 GiB (high memory), PolarFire SoC
maps it at address 0x10_0000_0000. Address 0xC000_0000 and above is
aliased to the same 1 GiB low memory with different cache attributes.
At present QEMU maps the system memory contiguously from 0x8000_0000.
This corrects the wrong QEMU logic. Note address 0x14_0000_0000 is
the alias to the high memory, and even physical memory is only 1 GiB,
the HSS codes still tries to probe the high memory alias address.
It seems there is no issue on the real hardware, so we will have to
take that into the consideration in our emulation. Due to this, we
we increase the default system memory size to 1537 MiB (the minimum
required high memory size by HSS) so that user gets notified an error
when less than 1537 MiB is specified.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20201101170538.3732-1-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Somehow HSS needs to access address 0 [1] for the DDR calibration data
which is in the chipset's reserved memory. Let's map it.
[1] See the config_copy() calls in various places in ddr_setup() in
the HSS source codes.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-9-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Previously SYSREG was created as an unimplemented device. Now that
we have a simple SYSREG module, connect it.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-8-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Previously IOSCB_CFG was created as an unimplemented device. With
the new IOSCB model, its memory range is already covered by the
IOSCB hence remove the previous unimplemented device creation in
the SoC codes.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-6-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Connect DDR SGMII PHY module and CFG module to the PolarFire SoC.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-4-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
It is not easy to find out the memory map for a specific component
in the PolarFire SoC as the information is scattered in different
documents. Add some comments so that people can know where to get
such information from the Microchip website.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1603863010-15807-2-git-send-email-bmeng.cn@gmail.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
This is an effort to clean up the hw/riscv directory. Ideally it
should only contain the RISC-V SoC / machine codes plus generic
codes. Let's move sifive_plic model to hw/intc directory.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1599129623-68957-7-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
This is an effort to clean up the hw/riscv directory. Ideally it
should only contain the RISC-V SoC / machine codes plus generic
codes. Let's move sifive_clint model to hw/intc directory.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1599129623-68957-6-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
At present the CLINT timestamp is using a hard-coded timebase
frequency value SIFIVE_CLINT_TIMEBASE_FREQ. This might not be
true for all boards.
Add a new 'timebase-freq' property to the CLINT device, and
update various functions to accept this as a parameter.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1598924352-89526-16-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Microchip PolarFire SoC integrates 3 GPIOs controllers. It seems
enough to create unimplemented devices to cover their register
spaces at this point.
With this commit, QEMU can boot to U-Boot (2nd stage bootloader)
all the way to the Linux shell login prompt, with a modified HSS
(1st stage bootloader).
For detailed instructions on how to create images for the Icicle
Kit board, please check QEMU RISC-V WiKi page at:
https://wiki.qemu.org/Documentation/Platforms/RISCV
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <1598924352-89526-15-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Microchip PolarFire SoC integrates 2 Candence GEMs to provide
IEEE 802.3 standard-compliant 10/100/1000 Mbps ethernet interface.
On the Icicle Kit board, GEM0 connects to a PHY at address 8 while
GEM1 connects to a PHY at address 9.
The 2nd stage bootloader (U-Boot) is using GEM1 by default, so we
must specify 2 '-nic' options from the command line in order to get
a working ethernet.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <1598924352-89526-14-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
On the Icicle Kit board, the HSS firmware utilizes the on-chip DMA
controller to move the 2nd stage bootloader in the system memory.
Let's connect a DMA controller to Microchip PolarFire SoC.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1598924352-89526-11-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Microchip PolarFire SoC integrates one Cadence SDHCI controller.
On the Icicle Kit board, one eMMC chip and an external SD card
connect to this controller depending on different configuration.
As QEMU does not support eMMC yet, we just emulate the SD card
configuration. To test this, the Hart Software Services (HSS)
should choose the SD card configuration:
$ cp boards/icicle-kit-es/def_config.sdcard .config
$ make BOARD=icicle-kit-es
The SD card image can be built from the Yocto BSP at:
https://github.com/polarfire-soc/meta-polarfire-soc-yocto-bsp
Note the generated SD card image should be resized before use:
$ qemu-img resize /path/to/sdcard.img 4G
Launch QEMU with the following command:
$ qemu-system-riscv64 -nographic -M microchip-icicle-kit -sd sdcard.img
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1598924352-89526-9-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Microchip PolarFire SoC has 5 MMUARTs, and the Icicle Kit board
wires 4 of them out. Let's connect all 5 MMUARTs.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1598924352-89526-7-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
This is an initial support for Microchip PolarFire SoC Icicle Kit.
The Icicle Kit board integrates a PolarFire SoC, with one SiFive's
E51 plus four U54 cores and many on-chip peripherals and an FPGA.
For more details about Microchip PolarFire Soc, please see:
https://www.microsemi.com/product-directory/soc-fpgas/5498-polarfire-soc-fpga
Unlike SiFive FU540, the RISC-V core resect vector is at 0x20220000.
The following perepherals are created as an unimplemented device:
- Bus Error Uint 0/1/2/3/4
- L2 cache controller
- SYSREG
- MPUCFG
- IOSCBCFG
More devices will be added later.
The BIOS image used by this machine is hss.bin, aka Hart Software
Services, which can be built from:
https://github.com/polarfire-soc/hart-software-services
To launch this machine:
$ qemu-system-riscv64 -nographic -M microchip-icicle-kit
The memory is set to 1 GiB by default to match the hardware.
A sanity check on ram size is performed in the machine init routine
to prompt user to increase the RAM size to > 1 GiB when less than
1 GiB ram is detected.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <1598924352-89526-5-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
