Florent Kermarrec [Thu, 10 Oct 2019 19:40:29 +0000 (21:40 +0200)]
cpu/rocket: rename ibus/dbus to mem_wb/mmio_wb and add size suffix
Florent Kermarrec [Thu, 10 Oct 2019 19:35:06 +0000 (21:35 +0200)]
cpu: add buses list and use it in soc_core to add bus masters
Florent Kermarrec [Thu, 10 Oct 2019 19:15:49 +0000 (21:15 +0200)]
integration: move soc constants to soc.h of csr.h
software retro-compat with soc.h included in csr.h
Florent Kermarrec [Thu, 10 Oct 2019 17:39:33 +0000 (19:39 +0200)]
build/generic_platform: only add sources if language is not None
Florent Kermarrec [Thu, 10 Oct 2019 17:36:17 +0000 (19:36 +0200)]
xilinx/vivado: replace "xy" == language with language == "xy"
enjoy-digital [Thu, 10 Oct 2019 17:31:09 +0000 (19:31 +0200)]
Merge pull request #277 from railnova/feature/vivado_sysverilog_support
[feature] Add SystemVerilog support for the Vivado builder
Florent Kermarrec [Thu, 10 Oct 2019 17:23:01 +0000 (19:23 +0200)]
integration/soc_zynq: shadow_base no longer recommended (replace with io_regions)
Florent Kermarrec [Thu, 10 Oct 2019 17:21:32 +0000 (19:21 +0200)]
bios/main: use same banner than README (MiSoC cited in README/LICENSE)
Florent Kermarrec [Thu, 10 Oct 2019 17:18:28 +0000 (19:18 +0200)]
software/bios: don't show peripherals init banner if nothing to init, add Ethernet init printf
Martin Cornil [Thu, 10 Oct 2019 12:06:37 +0000 (14:06 +0200)]
Add system Verilog support for the Vivado builder
enjoy-digital [Wed, 9 Oct 2019 19:25:57 +0000 (21:25 +0200)]
Merge pull request #276 from gsomlo/gls-rocket-map
cpu/rocket: swap main_mem and io regions
Gabriel Somlo [Wed, 9 Oct 2019 18:25:41 +0000 (14:25 -0400)]
cpu/rocket: swap main_mem and io regions
The total size of RAM (main_mem) can be expected to vary significantly,
and often exceed the size needed for MMIO allocations by a large margin.
As such, place Rocket's MMIO (io regions) below 0x8000_0000, and start
the RAM (main_mem) at 0x8000_0000, with nothing above it to limit its
future growth.
Also, bump the pre-built Rocket verilog submodule to an updated version,
which also comes with matching changes to the way MMIO and RAM accesses
are mapped and routed to their respective AXI interfaces.
Signed-off-by: Gabriel Somlo <gsomlo@gmail.com>
Florent Kermarrec [Wed, 9 Oct 2019 10:05:10 +0000 (12:05 +0200)]
cpu: add default io_regions to CPUNone (all address range can be used as IO)
enjoy-digital [Wed, 9 Oct 2019 09:20:50 +0000 (11:20 +0200)]
Merge pull request #275 from pcotret/patch-1
Update README (related to issue #273)
Florent Kermarrec [Wed, 9 Oct 2019 08:47:19 +0000 (10:47 +0200)]
soc_core: improve check_io_region error message
Florent Kermarrec [Wed, 9 Oct 2019 08:38:22 +0000 (10:38 +0200)]
targets/sim: switch from shadow_base to io_regions
Florent Kermarrec [Wed, 9 Oct 2019 08:24:01 +0000 (10:24 +0200)]
cpu/rocket: move csr to IO region
Florent Kermarrec [Wed, 9 Oct 2019 08:19:18 +0000 (10:19 +0200)]
build/xilinx/vivado: fix default synth-mode
Florent Kermarrec [Wed, 9 Oct 2019 08:14:14 +0000 (10:14 +0200)]
soc_core/cpu: add io_regions and deprecate shadow_base (with API retro-compat)
The shadow_base parameter has always been difficult to apprehend, replace it with
io_regions (uncached regions) defined user or the CPU.
The equivalent of a shadow_base parameter of 0x80000000 in the old API is:
io_regions = {0x80000000: 0x80000000} # origin, length
It's still possible to use shadow_base with retro-compat, but user is encouraged
to update and features will be removed in the future.
Pascal Cotret [Wed, 9 Oct 2019 07:48:32 +0000 (09:48 +0200)]
Update README (related to issue #273)
Following my problems with the quick start guide (issue #273), I suggest a few modifications to have a "real" quick start guide.
enjoy-digital [Tue, 8 Oct 2019 19:15:54 +0000 (21:15 +0200)]
Merge pull request #274 from gsomlo/gls-shadow-base
builder: use the SoC's existing shadow base with get_csr_header()
Gabriel Somlo [Tue, 8 Oct 2019 18:28:50 +0000 (14:28 -0400)]
builder: use the SoC's existing shadow base with get_csr_header()
Both the SoC and get_csr_header() have independently set defaults
for the value of 'shadow_base'. If the SoC's value was modified,
ensure that get_csr_header() uses the modified value instead of
its own default.
Signed-off-by: Gabriel Somlo <somlo@cmu.edu>
Florent Kermarrec [Mon, 7 Oct 2019 08:38:26 +0000 (10:38 +0200)]
targets/arty: allow setting synth-mode to yosys with command line: --synth-mode=yosys
Florent Kermarrec [Mon, 7 Oct 2019 08:37:16 +0000 (10:37 +0200)]
build/xilinx/vivado: add vivado_build_args/vivado_build_argdict for yosys synthesis mode
Florent Kermarrec [Mon, 7 Oct 2019 08:36:32 +0000 (10:36 +0200)]
xilinx/common: be sure language is not vhdl when yosys synthesis is used
Florent Kermarrec [Mon, 7 Oct 2019 06:49:32 +0000 (08:49 +0200)]
cpu/vexriscv: use specific mem_map for linux variant
enjoy-digital [Sun, 6 Oct 2019 12:55:44 +0000 (14:55 +0200)]
Merge pull request #271 from gsomlo/gls-yosys-nowidelut
RFC: optional '-nowidelut' flag for yosys synth_ecp5
Gabriel L. Somlo [Thu, 8 Aug 2019 22:55:14 +0000 (18:55 -0400)]
trellis, versa_ecp5: optional '-nowidelut' flag for yosys synth_ecp5
Passing '-nowidelut' to yosys' synth_ecp5 command improves area utilization
to the point where a (linux variant) rocket-chip based design will fit on a
versa_ecp5 board. Usually '-nowidelut' incurs a timing penalty, but that is
then mitigated by using DSP inference (enabled by default from yosys commit
8474c5b3).
Off by default, this flag can be enabled by adding '--yosys-nowidelut=True'
to the litex/boards/targets/versa_ecp5.py command line.
Signed-off-by: Gabriel Somlo <gsomlo@gmail.com>
enjoy-digital [Sun, 6 Oct 2019 10:10:19 +0000 (12:10 +0200)]
Merge pull request #272 from sergachev/fix-comments
fix comments in icap.py
Ilia Sergachev [Sun, 6 Oct 2019 08:47:28 +0000 (10:47 +0200)]
fix comments
Florent Kermarrec [Fri, 4 Oct 2019 08:00:45 +0000 (10:00 +0200)]
litex_setup: add litejesd204b
enjoy-digital [Tue, 1 Oct 2019 19:40:56 +0000 (21:40 +0200)]
Merge pull request #270 from gsomlo/gls-csr-upper
soc/integration: ensure CSR constants are in uppercase
Florent Kermarrec [Tue, 1 Oct 2019 19:30:14 +0000 (21:30 +0200)]
soc/cores/icap: simplify ICAPBitstream (untested)
Florent Kermarrec [Tue, 1 Oct 2019 19:04:49 +0000 (21:04 +0200)]
soc/cores/icap: rename ICAP to ICAPBistream and revert old ICAP
enjoy-digital [Tue, 1 Oct 2019 18:55:28 +0000 (20:55 +0200)]
Merge pull request #269 from antmicro/rework_icap
soc: cores: support sending custom bitstream to ICAP
Gabriel Somlo [Tue, 1 Oct 2019 16:14:33 +0000 (12:14 -0400)]
soc/integration: ensure CSR constants are in uppercase
Fixup over commit
8be5824e.
Signed-off-by: Gabriel Somlo <gsomlo@gmail.com>
Jan Kowalewski [Tue, 17 Sep 2019 19:35:53 +0000 (21:35 +0200)]
soc: cores: support sending custom bitstream to ICAP
This adds FIFO that can be used to send any
sequence of commands to the ICAP controller.
Florent Kermarrec [Mon, 30 Sep 2019 21:31:34 +0000 (23:31 +0200)]
soc/interconnect: rename stream_packet to packet & cleanup (with retro-compat)
Florent Kermarrec [Mon, 30 Sep 2019 21:18:39 +0000 (23:18 +0200)]
soc/interconnect/stream: add separators, mode Actor modules just after Endpoint
Florent Kermarrec [Mon, 30 Sep 2019 14:00:11 +0000 (16:00 +0200)]
soc_zynq: update get_csr_header
Florent Kermarrec [Mon, 30 Sep 2019 09:32:07 +0000 (11:32 +0200)]
soc/integration: move cpu_interface retro-compatibility to litex/__init__
Florent Kermarrec [Mon, 30 Sep 2019 08:59:36 +0000 (10:59 +0200)]
soc/integration: use dicts for constants/mem_regions/csr_regions to cleanup/simplify iterations on theses
Florent Kermarrec [Mon, 30 Sep 2019 06:30:45 +0000 (08:30 +0200)]
cpu: remove initial SERV support (we'll work in a branch to experiment with it)
Florent Kermarrec [Mon, 30 Sep 2019 06:26:38 +0000 (08:26 +0200)]
soc_core: fix cpu_type=None case and add test for it
Florent Kermarrec [Sun, 29 Sep 2019 15:33:16 +0000 (17:33 +0200)]
soc_sdram: remove axi usecase, this was only useful to do some preliminary axi tests.
Proper AXI support will be added in the future for SoCs.
Florent Kermarrec [Sun, 29 Sep 2019 15:31:37 +0000 (17:31 +0200)]
soc_core: cleanup/re-align
Florent Kermarrec [Sun, 29 Sep 2019 15:23:01 +0000 (17:23 +0200)]
soc/integration: rename cpu_interface to export (with retro-compat), re-arrange a bit, add separators
Florent Kermarrec [Sun, 29 Sep 2019 15:12:15 +0000 (17:12 +0200)]
test/test_targets: update cpu-type to mor1kx
Florent Kermarrec [Sun, 29 Sep 2019 15:04:21 +0000 (17:04 +0200)]
soc/cores: uniformize (continue)
Florent Kermarrec [Sun, 29 Sep 2019 14:10:44 +0000 (16:10 +0200)]
soc/cores/gpio: uniformize with others cores
Florent Kermarrec [Sun, 29 Sep 2019 14:08:39 +0000 (16:08 +0200)]
soc/cores: rename frequency_meter to freqmeter and uniformize with others cores
Florent Kermarrec [Sun, 29 Sep 2019 14:02:04 +0000 (16:02 +0200)]
soc/cores/ecc: improve readibility, uniformize with others cores
Florent Kermarrec [Sun, 29 Sep 2019 13:58:22 +0000 (15:58 +0200)]
soc/cores/clocks: improve readibility
Florent Kermarrec [Sun, 29 Sep 2019 13:47:10 +0000 (15:47 +0200)]
soc_core: use cpu.data_width to compute csr_alignment (and remove Rocket workaround)
Florent Kermarrec [Sun, 29 Sep 2019 13:41:18 +0000 (15:41 +0200)]
soc/cores/cpus: improve ident/align, uniformize between cpus
Florent Kermarrec [Sun, 29 Sep 2019 13:05:29 +0000 (15:05 +0200)]
soc/cores/cpu: add CPU class and make all CPU inheritate from it
Also rename reserved_interrupts to interrupts (empty dict is no reserved interrupts)
Florent Kermarrec [Sun, 29 Sep 2019 12:38:07 +0000 (14:38 +0200)]
soc_sdram: move ControllerInjector to LiteDRAM (LiteDRAMCore)
Florent Kermarrec [Sun, 29 Sep 2019 12:22:26 +0000 (14:22 +0200)]
soc/integration: add common.py and move helpers from soc_core to it
Florent Kermarrec [Sat, 28 Sep 2019 17:04:38 +0000 (19:04 +0200)]
soc_core: avoid manual listing of support CPUs, just use CPU.keys()
Florent Kermarrec [Sat, 28 Sep 2019 17:01:41 +0000 (19:01 +0200)]
soc_core: remove add_cpu_or_bridge retro-compatibility (most of the designs have been updated since the change)
Florent Kermarrec [Sat, 28 Sep 2019 16:59:30 +0000 (18:59 +0200)]
integration/soc_core: remove csr_map_update (no longer used)
Florent Kermarrec [Sat, 28 Sep 2019 12:13:39 +0000 (14:13 +0200)]
soc_core/cpu: move memory map override to CPUs, select reset_address after eventual memory map has override been done
Florent Kermarrec [Sat, 28 Sep 2019 10:35:41 +0000 (12:35 +0200)]
soc/cores/cpu: add set_reset_address method and use it instead of passing reset_address as a parameter
Florent Kermarrec [Sat, 28 Sep 2019 10:09:55 +0000 (12:09 +0200)]
soc/cores/cpu: do instance in do_finalize for all cpus (allow updating parameters until the design is generated)
Florent Kermarrec [Fri, 27 Sep 2019 22:55:08 +0000 (00:55 +0200)]
cores/cpu: define CPUS and simplify instance
Florent Kermarrec [Fri, 27 Sep 2019 22:42:00 +0000 (00:42 +0200)]
soc_core/serv: use UART_POLLING (no interrupt support)
Florent Kermarrec [Fri, 27 Sep 2019 22:41:28 +0000 (00:41 +0200)]
add SERV submodule
Florent Kermarrec [Fri, 27 Sep 2019 22:35:26 +0000 (00:35 +0200)]
software/libbase/uart: add polling mode
Florent Kermarrec [Fri, 27 Sep 2019 22:17:00 +0000 (00:17 +0200)]
add SERV CPU initial support (not working)
Florent Kermarrec [Wed, 25 Sep 2019 12:09:44 +0000 (14:09 +0200)]
targets/ulx3s: revert to cl=2
Florent Kermarrec [Wed, 25 Sep 2019 12:07:28 +0000 (14:07 +0200)]
boards/netv2: switch to MVP, add spiflashx4 and hdmi in/out
Florent Kermarrec [Tue, 24 Sep 2019 15:55:29 +0000 (17:55 +0200)]
wishbone2csr: refactor using FSM, reduce latency (make it asynchronous) and set csr.adr only when access is done (allow use of CSR/CSRBase we signal)
Making it asynchronous does not seem to deteriorate timing or resource usage, if it's the case for some designs, we'll add a register parameter.
Florent Kermarrec [Tue, 24 Sep 2019 15:51:06 +0000 (17:51 +0200)]
csr: add we signal to CSR, CSRStatus
Doing actions on register read is generally not a good design practice (it's
better to do separate register write to trigger actions) but in some very
specific cases being able to know that register has been read can solve cases
that are difficult to do with the recommended practives and that can justify
doing an exception.
This commit add a we signal to CSR, CSRStatus and this allow the logic to know
when the CSR, CSRStatus is read.
Florent Kermarrec [Tue, 24 Sep 2019 12:40:48 +0000 (14:40 +0200)]
build/xilinx/programmer: fix vivado_cmd
Florent Kermarrec [Tue, 24 Sep 2019 08:11:31 +0000 (10:11 +0200)]
soc/integration/doc: replace "== None" by "is None"
enjoy-digital [Tue, 24 Sep 2019 08:09:22 +0000 (10:09 +0200)]
Merge pull request #266 from xobs/add-moduledoc-autodoc
Add ModuleDoc and AutoDoc
Florent Kermarrec [Tue, 24 Sep 2019 06:49:00 +0000 (08:49 +0200)]
tools/litex_read_verilog: also delete yosys_v2j.ys
Benjamin Herrenschmidt [Tue, 24 Sep 2019 06:40:22 +0000 (08:40 +0200)]
soc_core/sdram: Don't blow up if _wb_sdram_ifs or _csr_masters are empty
For example a standalone controller with no exposed CSRs (probably not
a very useful configuration but I really don't like python backtraces)
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Sean Cross [Tue, 24 Sep 2019 06:34:41 +0000 (14:34 +0800)]
timer: inherit ModuleDoc
With the new ModuleDoc class, we can inherit `ModuleDoc` and
automatically get module-level documentation.
This patch also corrects a typo in `timer` that causes an error in
sphinx.
Signed-off-by: Sean Cross <sean@xobs.io>
Sean Cross [Tue, 24 Sep 2019 06:30:28 +0000 (14:30 +0800)]
integration: add ModuleDoc and AutoDoc
It is important to be able to document modules other than CSRs.
This patch adds ModuleDoc and AutoDoc, both of which can be used
together to document modules.
ModuleDoc can be used to transform the __doc__ string of a class into a
reference-manual section. Alternately, it can be used to add additional
sections to a module.
AutoDoc is used to gather all submodule ModuleDoc objects in order to
traverse the tree of documentation.
Signed-off-by: Sean Cross <sean@xobs.io>
enjoy-digital [Mon, 23 Sep 2019 21:19:45 +0000 (23:19 +0200)]
Merge pull request #264 from antmicro/mor1kx_linux
Enable to run Linux on mork1x
Florent Kermarrec [Mon, 23 Sep 2019 13:57:14 +0000 (15:57 +0200)]
soc_core: set csr to 0x00000000 when there is no wishbone
Florent Kermarrec [Mon, 23 Sep 2019 13:53:07 +0000 (15:53 +0200)]
soc_sdram: Don't add the L2 Cache when there's no wishbone bus
Filip Kokosinski [Thu, 19 Sep 2019 10:23:05 +0000 (12:23 +0200)]
soc_core: adapt memory map for mainline Linux with mor1kx
Mainline Linux expects it to be loaded at the physical address of 0x0.
Change the MAIN_RAM base address to 0x0 and update exception vector
during the booting process.
Filip Kokosinski [Mon, 23 Sep 2019 11:45:46 +0000 (13:45 +0200)]
boards/targets: increase integrated ROM size if EthernetSoC is used
Currently section '.rodata' of the LiteX BIOS doesn't fit in the 'rom'
region if mor1kx is used with EthernetSoC. Increase the integrated ROM
size from 0x8000 to 0x10000 in EthernetSoC.
Florent Kermarrec [Mon, 23 Sep 2019 10:53:37 +0000 (12:53 +0200)]
soc_core: revert wishbone2csr to __init__ but add with_wishbone parameter
Florent Kermarrec [Mon, 23 Sep 2019 08:15:27 +0000 (10:15 +0200)]
soc_sdram: change l2_size checks order
Florent Kermarrec [Mon, 23 Sep 2019 07:58:47 +0000 (09:58 +0200)]
soc_core: move CSR bridge to finalize (only generate it if there is a wishbone master), revert default parameter when cpu_type is None (we have systems with cpu_type=None but that are using these peripherals)
Florent Kermarrec [Mon, 23 Sep 2019 07:26:47 +0000 (09:26 +0200)]
integration/builder: avoid specific _generate_standalone_includes
Benjamin Herrenschmidt [Mon, 23 Sep 2019 06:39:25 +0000 (08:39 +0200)]
This will allow it to be built for microwatt out of tree
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Mon, 23 Sep 2019 06:33:35 +0000 (08:33 +0200)]
soc_core: When cpu_type is "None", let's not generate useless UART, timer, ROMs, wishbone to CSR bridge etc...
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Mon, 23 Sep 2019 06:30:01 +0000 (08:30 +0200)]
integration/builder: When the CPU is "None", we used to not generate any code.
With this change, we will now generate csr.h and sdram_phy.h, which
will be needed by the initialization code running on the host CPU.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
enjoy-digital [Fri, 20 Sep 2019 06:28:19 +0000 (08:28 +0200)]
Merge pull request #263 from xobs/spi-flash-csrfield
spi_flash: document register fields
Sean Cross [Fri, 20 Sep 2019 04:11:59 +0000 (12:11 +0800)]
spi_flash: document register fields
Document the various fields present in the SPI flash bitbang interface.
This adds documentation for the Single and DualQuad modules.
Signed-off-by: Sean Cross <sean@xobs.io>
enjoy-digital [Fri, 20 Sep 2019 04:25:57 +0000 (06:25 +0200)]
Merge pull request #262 from jersey99/master
vivado just needs to be in the path for the programmer as well
Vamsi K Vytla [Fri, 20 Sep 2019 03:35:55 +0000 (20:35 -0700)]
vivado just needs to be in the path for the programmer as well
enjoy-digital [Thu, 19 Sep 2019 09:40:55 +0000 (11:40 +0200)]
Merge pull request #261 from xobs/event-documentation
csr_eventmanager: add `name` and `description` args
Sean Cross [Thu, 19 Sep 2019 09:23:03 +0000 (17:23 +0800)]
csr_eventmanager: add `name` and `description` args
Add `name` and `description` as optional arguments to the various
EventSource types. These default to `None`, so this should be a
backwards-compatible change.
Use the same trick as CSRs, where we default the `name` to be the
instantiated object name as read from the Migen `get_obj_var_name()`
call.
Signed-off-by: Sean Cross <sean@xobs.io>
Florent Kermarrec [Thu, 19 Sep 2019 07:18:16 +0000 (09:18 +0200)]
cores/timer: add general documentation on Timer implementation and behavior.
Florent Kermarrec [Thu, 19 Sep 2019 03:16:01 +0000 (05:16 +0200)]
soc_sdram: improve readibility and convert l2_size to minimal allowed if provided l2_size is lower