Ross Thompson
1e2180ef98
Updated HPTW to route access faults generated by the HPTW to the original access type either instruction access fault, load access fault or store access fault.
2022-11-29 17:19:31 -06:00
Ross Thompson
ac0f6ddb7b
I found the issue with the cache changes. FlushW is not asserted for all TrapM. Ecall and Ebreak don't flush the W stage. However the ifu's bus controllable must disable the BusRW for all traps.
2022-11-16 15:38:37 -06:00
Ross Thompson
42c0a10d07
Removed TrapM from the LSU and IFU. TrapM is replaced with FlushW for both. (Don't like this for the IFU).
...
FlushW prevents writting the cache, dtim, and bus state. FlushW still gates HTRANS.
FlushW does not impact the mealy outputs of the cache and bus FSMs and hazard is updated to
not stall W if we get a trap.
2022-11-07 15:50:55 -06:00
David Harris
a5a922d048
Removed unnecessary configuration conditions from subwordread sign extension/NaN boxing
2022-10-10 07:12:37 -07:00
cturek
c72e2e5d49
Added integer inputs and flags to divsqrt
2022-09-29 23:08:27 +00:00
Ross Thompson
638e506d0b
Hmm. the icache and ifu didn't have a CommittedF signals going back to the privileged unit. They probably should. If an interrupt occurred during the middle of an instruction fetch icache miss I think it would corrupt the icache.
2022-09-28 17:39:51 -05:00
Ross Thompson
f24b0feeed
renamed ahbmulticontroller to ebu.
2022-09-26 14:37:18 -05:00
Ross Thompson
cc1ba84637
Found the ahb burst bug.
...
We had instruction fetches fixed HSIZE = 2 (4 bytes) for all requests. It should be HSIZE = 3 (8 bytes) for cache fetches and 4 for uncached reads. The reason this worked for non burst is the DDR4 memory controller returns the full double word even for 4 byte reads. In burst mode the second beat ending up pointing to the next 4 bytes rather than the next 8 bytes.
2022-09-17 20:30:01 -05:00
Ross Thompson
db56a326c9
renamed multimanager to multicontroller.
2022-09-14 14:03:37 -05:00
Ross Thompson
1c248e5164
Removed old signals.
2022-08-31 09:50:39 -05:00
Ross Thompson
5eb1fff27d
Have a rough working multi manager!
2022-08-29 17:11:27 -05:00
Ross Thompson
4f40bd07c3
Modified rv32e configuration to use a true ahb bus interface in the lsu and ifu.
2022-08-29 17:04:53 -05:00
Ross Thompson
4d7b905806
Part way through the updated bus fsm for direct AHB in lsu/ifu + multi-manager.
2022-08-29 13:01:24 -05:00
David Harris
352bf88ac0
FIxed wallypipelinedsoc merge conflict
2022-08-25 15:36:47 -07:00
David Harris
b96942e84c
Removed delayed AHB signals from top level
2022-08-25 15:34:14 -07:00
Ross Thompson
109bcd470e
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2022-08-25 16:01:02 -05:00
David Harris
6222e15946
Extended HADDR to PA_BITS
2022-08-25 13:11:36 -07:00
Ross Thompson
502eb0f5d1
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2022-08-25 14:40:52 -05:00
David Harris
5dc4fb757a
Continued busdp/ebu simplification
2022-08-25 10:20:02 -07:00
David Harris
24ce72f0a2
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2022-08-25 09:52:49 -07:00
David Harris
89860588b8
Renamed AHB signals coming out of LSU to LSH_<AHBNAME>
2022-08-25 09:52:08 -07:00
Ross Thompson
bd9401179d
BROKEN. Don't use this commit.
...
Issue running cacheless with bus.
2022-08-25 11:02:46 -05:00
Ross Thompson
1e1646da90
Added generate around ebu.
2022-08-25 09:24:13 -05:00
David Harris
8d48ff4e63
Fixed FPU-IEU forwarding stall
2022-08-23 14:14:41 -07:00
David Harris
113258a0d0
Cleaned up fcvt selection control to IEU and FPUIllegalInst signals
2022-08-23 12:17:19 -07:00
David Harris
24a05c35d9
Renamed signals for LSU - FPU interface
2022-08-22 13:47:56 -07:00
David Harris
7151befd04
Removed FStore2 and simplified HPTW
2022-08-22 13:29:54 -07:00
Ross Thompson
5d9dab6149
pulled swbbytemask out of subword write.
2022-08-01 20:48:45 -05:00
Katherine Parry
ca4fe08fd9
renamed FLoad2 to FStore2
2022-07-09 00:26:45 +00:00
David Harris
d73645944f
APB CLINT passing regression
2022-07-05 15:51:35 +00:00
Katherine Parry
6baded9121
added rv32 double precision stores - untested
2022-06-28 21:33:31 +00:00
Katherine Parry
254ebf478e
added fld in rv32 - needs testing
2022-06-20 22:53:13 +00:00
Katherine Parry
31fd8772cf
postprocessing unit created and passing all tests
2022-06-13 22:47:51 +00:00
slmnemo
054cf5f7b0
Reworked AHB fsm to support one cycle latency read and writes, renamed key signals to better reflect their use, and fixed HTRANS errors
2022-06-08 15:03:15 -07:00
slmnemo
284e0395a0
Merge branch 'main' into cacheburstmode
2022-06-08 02:21:33 +00:00
slmnemo
2d76953d42
Added lock signal to ensure AHB speaks with the right bus
2022-06-08 02:19:21 +00:00
slmnemo
73e0c1c07f
Reworked bus to handle burst interfacing
2022-06-07 11:22:53 +00:00
David Harris
1d8bc2dc1b
Added stalls for pending SFENCE.VMA and FENCE.I in hazard unit
2022-06-02 09:37:59 -07:00
David Harris
c7ec9282fe
Provided sfencevmaM to hazard unit and renamed TLBFlush signals to sfencevma going into LSU/IFU. Preparing for SFENCE.VMA to flush the pipeline, but that is not yet working.
2022-06-02 14:18:55 +00:00
slmnemo
847c7930c4
added LSUBurstDone signal to signal when a burst has finished
2022-05-26 16:29:13 -07:00
slmnemo
08430a1e85
added burst size signals to the IFU, EBU, LSU, and busdp
2022-05-25 18:02:50 -07:00
David Harris
ce24c080d5
More unused signal cleanup
2022-05-12 15:26:08 +00:00
David Harris
5670f77de2
More unused signal cleanup
2022-05-12 15:21:09 +00:00
David Harris
e2e63ca9a8
Clean up unused signals
2022-05-12 14:49:58 +00:00
David Harris
8166fd772e
Added M prefix for MTimerInt and MSwInt to distinguish from future supervisor SwInt
2022-05-11 15:08:33 +00:00
David Harris
137b411bea
Removed M suffix from interrupts because they are generated asynchronously to pipeline
2022-05-11 14:41:55 +00:00
David Harris
4f1b0fdc64
Preliminary support for big endian modes. Regression passes but no big endian tests written yet.
2022-05-08 06:46:35 +00:00
David Harris
a516f89f22
WFI terminates when an interrupt is pending even if interrupts are globally disabled
2022-05-08 04:30:46 +00:00
David Harris
412d4656ed
Zero'd wfiM when ZICSR not supported to fix hang in E tests
2022-05-05 15:32:13 +00:00
David Harris
1a7599ce94
Changed WFI to stall pipeline in memory stage
2022-05-05 02:03:44 +00:00
