Commit Graph

152 Commits

Author SHA1 Message Date
Florent Kermarrec 74467e3b38 test/test_axi/test_axi_width_converter: Switch to DUT_ref (To avoid breaking CI).
We'll switch back to DUT when AXI Converter will be fixed.
2022-07-25 12:34:38 +02:00
Ilia Sergachev 982f94ba8d test: add axi 64b to 32b conversion test 2022-07-25 00:20:48 +02:00
Ilia Sergachev 65d5161408 test/axi_lite: parametrize address and data width in another test; add another test call with 64b data width 2022-07-20 02:44:57 +02:00
Ilia Sergachev bffd59726c test/axi_lite: rename a test for clarity; parametrize address and data width; add another test call with 64b data width 2022-07-20 02:43:43 +02:00
Florent Kermarrec 9c3663f3d2 test/test_cpu: Re-enable cv32e40p/marocchino. 2022-06-29 11:15:48 +02:00
enjoy-digital ec9d1c4fd0
CI: Disable more CPUs. 2022-06-27 22:43:01 +02:00
Florent Kermarrec f898423390 test/test_cpu: Diable mor1kx/picorv32 for now due to issue with newer Verilator. 2022-06-27 19:54:50 +02:00
Robert Szczepanski cbd873a33e test: FifoSyncMacro: Use F4PGA instead of deprecated Symbiflow 2022-06-17 16:27:25 +02:00
Florent Kermarrec 3f8cafeb58 test/test_cpu: Disable cva5 and enable marocchino/mor1kx. 2022-06-03 17:14:52 +02:00
Florent Kermarrec c0fc342823 test/test_cpu: Simplify using subTest/lists, test more RISC-V CPUs and comments for untested CPUs.
Also use --opt-level=O0 to reduce compilation time (execution is a bit slower but since we are only
executing the BIOS here, total test time is still reduced).
2022-06-03 16:16:21 +02:00
Florent Kermarrec 69451fad09 test/test_cpu: Disable test_cva6 for now since seems to be failing. 2022-05-25 09:32:30 +02:00
Massimiliano Giacometti c95ddbbff8 UART_POLLING 2022-05-19 15:07:46 +02:00
Robert Szczepanski 22abe1d543 Add tests for FIFOSyncMacro 2022-04-27 10:53:52 +02:00
Rafal Kolucki 8c1bc139ab soc/interconnect/wishbone: Cleanup in burst cycles support logic 2022-04-12 15:32:29 +02:00
Rafal Kolucki ad46a57403 test/test_wishbone: Add test for Wishbone SRAM constant address burst cycle 2022-04-12 14:06:22 +02:00
Rafal Kolucki cdd216f692 test/test_wishbone: Add basic test for SRAM with burst cycles support
Tests incrementing address burst cycle with linear and wrapped increments.
Only 4-beat wrap burst is tested in `test_sram_burst_wrap` test.
2022-04-12 14:06:22 +02:00
Florent Kermarrec ed6a6a83a9 litex_setup: Switch to manual install for Amaranth/Minerva (No longer supporting Python 3.6).
We could revert when upgrading LiteX python requirement.
2022-04-04 15:39:05 +02:00
Florent Kermarrec d39c3ed626 soc/cores/led: Review/Rework #1265.
- Split FSM in Main FSM/Xfer FSM to decouple Led data read from bit xfer and do read during xfer.
- Only keep optimization that are easily to understand.
- Default to new WS2812 revision (Since also works on old revision).
- Test 75/50/25MHz sys_clk_freq.
2022-04-04 15:24:54 +02:00
Wolfgang Nagele 67369403a9 Improve WS2812 timings and add different hardware revision support 2022-04-03 17:09:56 +02:00
Florent Kermarrec 6ef96b17bc soc/interconnect/csr: Fix CSRConstant read method (And add test_csr_constant to test_csr). 2022-03-21 15:21:08 +01:00
Florent Kermarrec dbde036162 soc/cores: Re-integrated generic/portable HyperBus/HyperRAM core from LiteHyperBus.
The generic version of the HyperRAM core is simple enough to be directly integrated in LiteX
which avoid an additional dependency.
2022-03-01 09:11:55 +01:00
Florent Kermarrec f62eca77e3 test/test_axi: Minor cleanups. 2022-02-17 15:13:05 +01:00
Florent Kermarrec 77c6cdd78e cores/clocks/lattice_ecp5: Rename ECP5Delay to ECP5DynamicDelay and adapt style for consistency. 2022-01-25 11:09:15 +01:00
Florent Kermarrec ea6bb3dd80 test/test_clock: Add minimal ECP5Delay test (syntax), rename tests with underscore. 2022-01-25 10:49:33 +01:00
Florent Kermarrec e50ff33c6e test/test_cpu: Disable Minerva test for now. 2021-12-13 16:51:23 +01:00
Florent Kermarrec 12d53790a9 test/test_cpu: Prepare microwatt/lm32 test.
microwatt: Still requires Yoys/GHDL-Synth installation.
lm32: Still requires prebuilt toolchain.
2021-10-26 16:37:38 +02:00
Florent Kermarrec e617f52e34 test/test_cpu: Comment test_mor1kx for now (test work but issue seems related to the pre-built toolchain). 2021-10-26 16:30:02 +02:00
Florent Kermarrec cb9f0fb1b0 ci/test_cpu: Install OpenRISC GCC toolchain in CI and add mor1kx to test_cpu. 2021-10-26 15:45:43 +02:00
Florent Kermarrec 71a91eac15 test: Rename test_boot.py to test_cpu.py. 2021-10-26 08:35:16 +02:00
Florent Kermarrec 47b3c9bc08 soc/interconnect/packet: Remove last_be support in LiteX, specialized Packetizer/Depacketizer have been moved to LiteEth to simplify development and avoid eventual regresion on others cores.
As seen during the last LiteEth developments, last_be data qualifier is not easy to handle correctly and should be replaced by a simpler data qualifier (similar to AXI's tkeep/tstrb). It will
be easier to do so by having a local copy of Packetizer/Depacketizer directly in LiteEth (still with last_be support) and work on the simpler data qualifier in LiteX (and test it on LitePCIe).
2021-10-25 11:17:36 +02:00
Navaneeth Bhardwaj 2886fe1701
Add bios test mode for CI (#1076)
* Add bios test mode for CI

This enables to test the booting of each CPU configurations with the bios in Verilator simulation.
2021-10-24 12:08:58 +02:00
Florent Kermarrec 8e448592f0 interconnect/packet: Revert old last/ready logic handling (new one breaks test_packet) and comment out test_packet2 tests (does not seems to be working with previous last/ready handling). 2021-10-23 18:21:47 +02:00
Florent Kermarrec 59fd2d31c7 test/test_packet2: Fix imports. 2021-10-23 17:54:00 +02:00
Florent Kermarrec 32bb2554bc test: Rename new test_packet/stream to test_packet2/stream2 and revert old tests.
Old and new tests are complementary and would need to be merged.
2021-10-23 17:40:41 +02:00
enjoy-digital 434b3a3654
Merge pull request #1008 from lschuermann/dev/packetizer-last_be-fix
{Dep,P}acketizer: properly handle last_be wraparound
2021-10-23 17:33:29 +02:00
Florent Kermarrec 99f3498f2d cores/icap/ICAP: Add Register read capability.
Useful to get some internal status, ex the IDCode or know if the executed bistream
in a multiboot configuration is the operational or fallback one.
2021-10-04 17:22:57 +02:00
Florent Kermarrec 3504904c09 cores/icap/ICAP: Rewrite using with an FSM instead of Timeline (will be easier to extend). 2021-10-04 15:06:03 +02:00
Florent Kermarrec 9416e30249 test/test_icap: Add IPROG sequence check. 2021-10-04 14:41:38 +02:00
Florent Kermarrec cb2f2d7021 cores/icap/ICAP: Rewrite using constants and cleanup. 2021-10-04 14:25:40 +02:00
George Hilliard 6733a3e3e6 clock/lattice_ecp5/ECP5PLL: ensure feedback path selected before exiting search 2021-09-15 00:07:43 -05:00
Leon Schuermann a568b7e26a test_packet.py: test randomly generated headers
Also includes some fixes for the non-last_be test path such that the
expected behavior is the one currently oberserved with the Packetizer
and Depacketizer.
2021-09-10 15:30:05 +02:00
Leon Schuermann 958bcaad2e test_packet.py: add last_be tests 2021-09-10 12:30:04 +02:00
Leon Schuermann 037294dc3b test_packet.py: support passing debug_print parameter 2021-09-09 16:37:26 +02:00
Leon Schuermann 6bda383178 test_packet.py: support {Dep,P}acketizer behavior without last_be 2021-09-09 16:37:26 +02:00
Leon Schuermann a08271b83a test_packet.py: utilize generic stream_inserter/colletor interface 2021-09-09 16:08:12 +02:00
Leon Schuermann ca50cba986 Rewrite test_stream.py with reusable stream_inserter/collector 2021-09-09 16:08:12 +02:00
Florent Kermarrec 79ac09316a interconnect/axi/AXIBurst2Beat: Fix BURST_WRAP case. 2021-08-06 16:41:58 +02:00
Florent Kermarrec 1ce48a973b clock/lattice_ecp5: Fix and rework 4-output solver implementation.
The implementation was causing regressions on actual designs, rework done:
- Only keep a common iteration loop as before.
- Add iteration on CLKO dividers (to fall in the VCO range).
- Do the iterations as before, if while doing it we find a clock suitable for feedback: just use it.
- If no feedback clock has been found: create it (if at least one free output available, if not raise an error).
2021-07-26 14:00:00 +02:00
George Hilliard 8954041a93 clock/lattice_ecp5/ECP5PLL: Only consider non-dpa clocks as feedback
Dynamically adjusting the phase of a feedback will cause it to unlock.
The phase adjust ports are shared by all the outputs, so there is no
technical way to prevent this.  Allow the user to indicate that they
will not adjust a clock when requesting an output by setting
uses_dpa=False, and only consider those that the user has promised not
to use.
2021-07-08 08:14:14 -05:00
Florent Kermarrec f6b2135cc9 test/test_timer: Update. 2021-05-27 19:37:51 +02:00