Luke Wren
|
1b0e205f87
|
Fix bad AMO asserts. Fix hwdata instability during stalled AMO write dphase, which meant AMOs were fundamentally broken following yesterday's datapath origami
|
2021-12-18 14:51:46 +00:00 |
|
Luke Wren
|
6b8d4913ee
|
Remove unnecessary mux of mw_result -> m_result
|
2021-12-18 01:34:25 +00:00 |
|
Luke Wren
|
79fec3a2f5
|
Overload mw_result register for capturing AMO read data. Save some LCs.
|
2021-12-18 01:24:26 +00:00 |
|
Luke Wren
|
28b53ef7b5
|
Delete the AMO ALU. Save around 80 LCs vs original implementation, maybe enables some more savings.
|
2021-12-18 00:35:13 +00:00 |
|
Luke Wren
|
7485269ddf
|
Use the branch target adder for load/store addresses. Preparing for AMO ALU deletion
|
2021-12-17 22:36:40 +00:00 |
|
Luke Wren
|
a35739baf1
|
Fix AMO failing to loop on global monitor write fail
|
2021-12-17 17:04:22 +00:00 |
|
Luke Wren
|
b0d28447ab
|
New license headers: DWTFPL -> Apache 2.0
|
2021-12-13 23:23:40 +00:00 |
|
Luke Wren
|
f1cda26bcc
|
Oops, try a little harder this time to ensure local monitor is cleared by sc.w errors _always_, even if lined up with trap entry stalls etc
|
2021-12-12 23:32:01 +00:00 |
|
Luke Wren
|
25b44d04cf
|
Add more properties wrt AMO. Fix awful bugs in interaction between AMOs and prior lr/sc, prior load/store exception, or IRQ which is asserted then deasserted. Fix sc with HRESP=1 HEXOKAY=1 not clearing the local monitor (not a bug, but unfriendly).
|
2021-12-12 23:24:25 +00:00 |
|
Luke Wren
|
88fea7acfa
|
Fix lockup on misaligned AMO. Add tests for misaligned/faulting AMOs.
|
2021-12-12 18:28:23 +00:00 |
|
Luke Wren
|
9460b3cd04
|
Add load/store and lr/sc bus fault tests. Fix lr.w not clearing local monitor when HRESP=1 HEXOKAY=1.
|
2021-12-11 15:52:34 +00:00 |
|
Luke Wren
|
3d2c912b4f
|
Add test script to make it easier to add software testcases
|
2021-12-09 22:25:18 +00:00 |
|
Luke Wren
|
7d2fa6a049
|
Fix width warnings in A instruction decode, add don't-care to bus rdata picking logic
|
2021-12-09 06:26:31 +00:00 |
|
Luke Wren
|
116e34df49
|
Fix commented out frontend properties which relied on non-constant past reset values
|
2021-12-07 20:24:29 +00:00 |
|
Luke Wren
|
ed22d502fd
|
Fix bus stall getting stuck high when a bus fault exception isn't immediately accepted by the frontend
|
2021-12-06 19:28:21 +00:00 |
|
Luke Wren
|
50d3d5d3b3
|
Maintain AMO IRQ holdoff when transitioning from data phase to error phase, to prevent error possibly being flushed
|
2021-12-06 19:27:20 +00:00 |
|
Luke Wren
|
29c5c8ca7f
|
Fix AMO stall falling through when write data phase should proceed to error phase
|
2021-12-06 18:28:56 +00:00 |
|
Luke Wren
|
8bfc089660
|
Slightly more strict holdoff of IRQs on AMO
|
2021-12-06 18:13:43 +00:00 |
|
Luke Wren
|
4c532240f8
|
Hold off IRQ when AMO is past the point of no return
|
2021-12-06 07:45:13 +00:00 |
|
Luke Wren
|
cc38f46848
|
Fix AMO wdata valid left high when entering trap at just the right time
|
2021-12-06 07:28:50 +00:00 |
|
Luke Wren
|
9b9120960d
|
Fix missing RAW stall on sc.w succes result. Closing laptop again.
|
2021-12-05 01:05:01 +00:00 |
|
Luke Wren
|
df658d86ff
|
First plausibly working AMOs. Add AMOs to instruction timings list
|
2021-12-04 23:44:22 +00:00 |
|
Luke Wren
|
5c098866f2
|
Sketch in AMO support
|
2021-12-04 20:46:39 +00:00 |
|
Luke Wren
|
a8933c332d
|
Fix illegal issue of pipelined exclusives on the bus, and document correct timings
|
2021-12-04 18:23:01 +00:00 |
|
Luke Wren
|
5e17bb805e
|
Add basic support for lr/sc instructions from the A extension
|
2021-12-04 15:02:31 +00:00 |
|
Luke Wren
|
607147f280
|
Rewrite byte pick/sign-extend logic, preparing to handle more memops
|
2021-12-04 12:08:54 +00:00 |
|
Luke Wren
|
a988adfec8
|
Add RISC-V opcodes and memory operation codes for atomics
|
2021-12-04 11:16:24 +00:00 |
|
Luke Wren
|
c8afb4ac33
|
Add option for fast high-half multiplies
|
2021-11-29 18:48:02 +00:00 |
|
Luke Wren
|
58c20a39d0
|
First pass at implementing bitmanip. Breaks CXXRTL. Ooop
|
2021-11-25 23:30:35 +00:00 |
Luke Wren
|
ed6b6a3660
|
Cleanup order of declaration/use of a couple of wires
|
2021-11-25 15:16:59 +00:00 |
|
Luke Wren
|
e05e9a4109
|
Add default_nettype none at top of every file, and default_nettype wire at bottom
|
2021-11-23 22:10:39 +00:00 |
|
Luke Wren
|
0b9b706e81
|
Safer logic for load/store blocked by preceding WFI
|
2021-11-23 22:01:14 +00:00 |
|
Luke Wren
|
cc6a6c09ba
|
Vaguely implement wfi
|
2021-11-05 18:48:42 +00:00 |
|
Luke Wren
|
cfe16caf41
|
Remove some old todos
|
2021-09-05 22:20:40 +01:00 |
|
Luke Wren
|
e9fccffca0
|
Fix break and single-step on a load/store access fault dropping the exception. Better fix for halt request on definitely-excepting stage M instruction giving unreachable next-instruction DPC.
|
2021-09-05 04:45:38 +01:00 |
|
Luke Wren
|
d03a82a826
|
Add instruction fetch faults
|
2021-09-04 02:57:39 +01:00 |
|
Luke Wren
|
8e3dc62b97
|
Fiddly handshake changes for hardware single-stepping. Can now step correctly through illegal instructions
|
2021-07-16 20:43:24 +01:00 |
|
Luke Wren
|
5aca6be572
|
Implement data0 inside of DM instead of core, so it gets correct reset. Add dummy tselect CSR.
|
2021-07-16 18:28:30 +01:00 |
|
Luke Wren
|
307955c810
|
Make CPU regfile nonresettable when FPGA symbol is defined, to support BRAM inference
|
2021-07-13 01:10:55 +01:00 |
|
Luke Wren
|
63d661af63
|
Start hacking in debug support to the core -- seems to work as well as before adding debug!
|
2021-07-10 18:53:48 +01:00 |
|
Luke Wren
|
af684c4e82
|
Some cleanup; correctly decode 16-bit EBREAK
|
2021-06-03 20:03:43 +01:00 |
|
Luke Wren
|
5f8d217395
|
Implement new IRQ behaviour, and change mip.meip to be masked by individual enables in meip0
|
2021-05-31 17:54:12 +01:00 |
|
Luke Wren
|
ea5db61582
|
Fix exception being passed to M when X is stalled but M is not (a load which had an unaligned address spuriously during a RAW stall on the result register)
|
2021-05-29 22:52:50 +01:00 |
|
Luke Wren
|
4b9a3c2c78
|
Fix wrong reg readaddr when D starvation ends during a downstream load/store dphase stall (was using garbage from D instead of new fetch data predecoded in F)
|
2021-05-29 19:32:12 +01:00 |
|
Luke Wren
|
f23ec3f941
|
Don't hold back instruction in M when an IRQ entry is stalled (but do for exception entry). Now pass add and lw checks in riscv-formal with depth 15, so getting somewhere
|
2021-05-29 18:57:43 +01:00 |
|
Luke Wren
|
65075df0e5
|
More work on traps, delaying IRQs which arrive whilst a load/store address phase is stalled to avoid deassertion on the bus
|
2021-05-29 18:00:43 +01:00 |
|
Luke Wren
|
1b252d4bda
|
Significant overhaul of trap handling. Exceptions now taken from stage 3 instead of stage 2
|
2021-05-23 11:59:46 +01:00 |
|
Luke Wren
|
5e61c9f9ac
|
Use branch target adder for JALR target, and use ALU for JAL/JALR linkaddr instead of muxing in next_pc
|
2021-05-23 09:12:50 +01:00 |
|
Luke Wren
|
7a3ce494e4
|
Fix a couple issues with trap exit, can now run add check with traps enabled (at low depth)
|
2021-05-23 06:40:44 +01:00 |
|
Luke Wren
|
dec78a728d
|
Fix a few things that were obviously wrong, and the first signs of a plausible RVFI bridge circuit
|
2021-05-22 15:35:52 +01:00 |
|
Luke Wren
|
08e986912c
|
Also fix RAW stall on JALR instructions, oops. Runs CoreMark and Dhrystone now
|
2021-05-22 11:18:56 +01:00 |
|
Luke Wren
|
6692c1f26d
|
Fix premature taking of branches with RAW data dependencies on the previous instruction
|
2021-05-22 10:18:47 +01:00 |
|
Luke Wren
|
692abbad8b
|
Merge stages D and X, and bring all branch resolution into X. Passes RV32I compliance
|
2021-05-22 07:55:13 +01:00 |
|
Luke Wren
|
844fa8f97f
|
Rename hazard5 -> hazard3
|
2021-05-21 03:46:29 +01:00 |