Ross Thompson
2ccf479354
Moved more logic inside the dcache memory.
2021-08-25 13:17:07 -05:00
Ross Thompson
35e57a7c61
partial dcache reorg.
2021-08-25 12:42:05 -05:00
David Harris
7d24ed3c51
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-08-25 06:47:20 -04:00
David Harris
3fa55a01f4
simplified or_rows generation and renamed oneHotDecoder to onehotdecoder
2021-08-25 06:46:41 -04:00
Ross Thompson
0cc47f3daf
Modified the preformance counter's InstRet to include ECALL and EBREAK by changing the hazard logic so these instructions don't self flush the W stage.
2021-08-23 15:46:17 -05:00
Ross Thompson
c31b7b4dc5
Wally previously was overcounting retired instructions when they were flushed.
...
InstrValidM was used to control when the counter was updated. However this is
not suppress the counter when the instruction is flushed in the M stage.
2021-08-23 12:24:03 -05:00
Ross Thompson
2825074114
Confirmed David's changes to the interrupt code.
...
When a timer interrupt occurs it should be routed to the machine interrupt
pending MTIP even if MIDELEG[5] = 1 when the current privilege mode is
Machine. This is true for all the interrupts. The interrupt should not be
masked even though it is delegated to a lower privilege. Since the CPU
is currently in machine mode the interrupt must be taken if MIE.
Additionally added a new qemu script which pipes together all the parsing and
post processing scripts to produce the singular all.txt trace without the
massivie intermediate files.
2021-08-22 21:36:31 -05:00
David Harris
4677b4bb38
possible interrupt code
2021-08-22 17:02:40 -04:00
Ross Thompson
65870877c3
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-08-17 16:06:54 -05:00
Ross Thompson
91b51c698e
Minor changes to dcache.
2021-08-17 15:22:10 -05:00
Katherine Parry
c8847b27e8
all conversions go through the execute stage result mux
2021-08-16 13:06:09 -04:00
Ross Thompson
a70d51f4c9
Modified the hptw's simulation error message so that synthesis does not attempt to include this statement.
2021-08-16 10:02:29 -05:00
Ross Thompson
36761d9155
Fixed syntax errors in some floating point modules. This came up in
...
Xilinx synthesis.
2021-08-15 16:48:49 -05:00
Ross Thompson
766c829d31
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-08-13 17:23:04 -05:00
Ross Thompson
55fda4de62
Switched ExceptionM to dcache to be just exceptions.
...
Added test bench logic to hold forces until the W stage is unstalled.
2021-08-13 15:53:50 -05:00
Katherine Parry
aedd71d570
move some FPU select muxs to execute stage
2021-08-13 14:41:22 -04:00
Ross Thompson
6a6d5e9b15
Added documentation about how the dcache and ptw interact.
2021-08-12 18:05:36 -05:00
Ross Thompson
814fd80b0f
Optimized subwordread to reduce critical path from 8 muxes to 5 muxes + 1 AND gate.
2021-08-12 13:36:33 -05:00
Ross Thompson
565c01709d
Removed unused states from dcache fsm.
2021-08-11 17:06:09 -05:00
Ross Thompson
2be625d8b9
Modified invalid plic reads to return 0 rather than deadbeaf.
2021-08-11 16:56:22 -05:00
Ross Thompson
4b25fed6d8
Simplified Dcache by sharing the read data mux with the victim selection mux.
2021-08-11 16:55:55 -05:00
Ross Thompson
22f274c51e
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-08-10 13:36:29 -05:00
Ross Thompson
67c1028862
Dcache and LSU clean up.
2021-08-10 13:36:21 -05:00
Katherine Parry
e00f181bcf
LZA added to FMA and attemting a merged FMA and adder in synthesis
2021-08-10 13:57:16 -04:00
Ross Thompson
cce0571925
Fixed another bug with the atomic instrucitons implemention in the dcache.
2021-08-08 22:50:31 -05:00
Ross Thompson
d3be04b7de
Fixed another bug with AMO. If the CPU stalled as an AMO was finishing, the write to the
...
cache's SRAM would occur. Then in the next cycle the SRAM would be reread while stalled
providing the new update dated rather than the correct older value.
2021-08-08 11:42:10 -05:00
Ross Thompson
fc7016eea6
Fixed the AMO dcache bug. The subword write needs to occur before the AMO logic.
...
Fixed logic for trace update in the M and W stages. The M stage should not update if there
is an instruction fault.
2021-08-08 00:28:18 -05:00
Ross Thompson
c749d08542
fixed the read timer issue but we still have problems with interrupts and i/o devices.
2021-08-06 10:16:06 -05:00
Ross Thompson
37ba6b19e5
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-30 17:57:13 -05:00
Ross Thompson
7b9e53fbe5
Removed 1 cycle delay on store miss.
...
Changed some logic to partially support atomics.
2021-07-30 14:00:51 -05:00
Katherine Parry
d60e394ef9
all fpu units use the unpacking unit
2021-07-28 23:49:21 -04:00
Ross Thompson
915d8136e5
Fixed bug which caused stores to take an extra clock cycle.
2021-07-26 12:22:53 -05:00
Ross Thompson
79ebc53977
Fixed bug with the compressed immediate generation. Several formats should zero extend.
2021-07-26 11:55:31 -05:00
Ross Thompson
ef55b30e99
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
...
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-26 11:55:00 -05:00
Ross Thompson
60177b92a6
Added additional interlock for itlb miss -> instrPageFault with concurrent memory operation.
2021-07-25 23:14:28 -05:00
Katherine Parry
30ac22edff
fixed some fpu lint errors
2021-07-24 16:41:12 -04:00
Katherine Parry
6c4aa624a5
fpu cleanup
2021-07-24 15:00:56 -04:00
Katherine Parry
ef28679721
fpu cleanup
2021-07-24 14:59:57 -04:00
kipmacsaigoren
f3579032bd
Cleaned up priority thermometer verilog. passses regression, ideally shortens critical path through pmp's
2021-07-23 11:57:58 -05:00
David Harris
5d2b30e332
Removed LEVELx states from HPTW
2021-07-23 08:11:15 -04:00
Ross Thompson
9939c66a1f
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-22 19:42:32 -05:00
Ross Thompson
3e916da36e
Removed the hardware page table walker fault state from the icache so that the icache will only unstall CPU for 1 cycle.
...
In the dcache we added a register to save the load read data in the event an itlb miss occurs concurrently with
the load in the memory stage. Under this situation we need to record the load ReadDataM into a temporary register,
SavedReadDataM. At this time the CPU is stall; however the walker is going to change the address in the dcache
which destroys this data. When leaving the PTW_READY state via a walker instruction fault or ITLB write we select
this SavedReadDataM so that the CPU can capture it.
2021-07-22 19:42:19 -05:00
David Harris
98660e0d19
Minor unpacking cleanup
2021-07-22 17:52:37 -04:00
Ross Thompson
551e3491af
Moved the ReadDataW register into the datapath.
...
The StallW from the hazard unit controls this.
Previously it was in the dcache and controlled by both the HPTW and hazard unit.
This caused an issue when the CPU expected the data to stay constant while
stalled, but the HPTW was causing the data to be modified.
2021-07-22 14:52:03 -05:00
Ross Thompson
fbbfc799b9
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-22 14:05:08 -05:00
Ross Thompson
9c90b4bdf7
Fixed bug with the itlb fault not dcache ptw ready state to ready state.
2021-07-22 14:04:56 -05:00
David Harris
c9890afb7f
Move Z sign swapping out of unpacker
2021-07-22 14:32:38 -04:00
David Harris
31be570461
Move Z=0 mux out of unpacker.
2021-07-22 14:28:55 -04:00
David Harris
63718cef8f
Move Z=0 mux out of unpacker.
2021-07-22 14:22:28 -04:00
David Harris
21a65f45cd
Partial work on Unpacking exponents to larger word size. FCVT and FMA are presently broken.
2021-07-22 14:18:27 -04:00
David Harris
b53eb6d030
Simplify unpacker
2021-07-22 13:42:16 -04:00
David Harris
19dac66264
Simplify unpacker
2021-07-22 13:40:42 -04:00
David Harris
44141047ef
Removed Assumed1 from FPU interface
2021-07-22 13:04:47 -04:00
David Harris
3ad2170ffd
Simplified interface to fclassify and fsgn (fixed)
2021-07-22 12:33:38 -04:00
David Harris
5e155e4fd1
Simplified interface to fclassify and fsgn
2021-07-22 12:30:46 -04:00
Ross Thompson
b4029a2848
Cleaned up icache and dcache.
2021-07-22 11:06:44 -05:00
Ross Thompson
3dd89a7e62
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-22 10:38:24 -05:00
Ross Thompson
25a8920a69
Tested all numbers of ways for dcache 1, 2, 4, and 8.
2021-07-22 10:38:07 -05:00
bbracker
d3059dd04c
fix UART RX FIFO bug where tail pointer can overtake head pointer
2021-07-22 02:09:41 -04:00
Ross Thompson
dac93bb366
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-21 16:44:32 -05:00
Kip Macsai-Goren
c69a5dc8a6
fixed issue with tlbflush remaining high during a stalled sfence instruction
2021-07-21 17:43:36 -04:00
Ross Thompson
71375ba655
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-21 16:39:07 -05:00
Ross Thompson
7785401281
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-07-21 14:56:30 -05:00
Ross Thompson
313bc5255c
Improved address bus names and usages in the walker, dcache, and tlbs.
...
Merge branch 'walkerEnhance' into main
2021-07-21 14:55:09 -05:00
Ross Thompson
310b454fa1
Added comment about better muxing.
2021-07-21 14:40:14 -05:00
Ross Thompson
5860f147d4
4 way set associative is now working.
2021-07-21 14:01:14 -05:00
Kip Macsai-Goren
4eaf95de60
Fixed TLB parameterization and valid bit flop to correctly do instr page faults
2021-07-21 14:44:43 -04:00
Katherine Parry
01f0b4e5df
FDIV and FSQRT work
2021-07-21 14:08:14 -04:00
Ross Thompson
e0990535e1
Fixed remaining bugs in 2 way set associative dcache.
2021-07-21 10:35:23 -05:00
Ross Thompson
3f780f012a
Finally fixed bug with the set associative design. The issue was not in the LRU but instead in the way selection mux.
...
Also forgot to include cacheLRU.sv file.
2021-07-20 23:17:42 -05:00
Katherine Parry
b9081e514c
FMA parameterized
2021-07-20 22:04:21 -04:00
Ross Thompson
14e949d6e3
Partially working 2 way set associative d cache.
2021-07-20 17:51:42 -05:00
bbracker
f9b6bd91f5
fix PC checking during InstrPageFault; fix order of S-mode CSR checking; rename peripheral scopes to not be genblk
2021-07-20 17:55:44 -04:00
David Harris
e5e3f5abe6
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-20 14:46:58 -04:00
David Harris
1f3dfa20f6
flag for optional boottim
2021-07-20 14:46:37 -04:00
Ross Thompson
00081ebc68
Replaced FinalReadDataM with ReadDataM in dcache.
2021-07-20 13:27:29 -05:00
David Harris
e1a1a8395e
Parameterized I$/D$ configurations and added sanity check assertions in testbench
2021-07-20 08:57:13 -04:00
James E. Stine
12e09a7ace
slight mod to fpdiv - still bug in batch vs. non-batch
2021-07-20 01:47:46 -04:00
Ross Thompson
365485bd8b
Added performance counters for dcache access and dcache miss.
2021-07-19 22:12:20 -05:00
David Harris
23b76a724d
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-19 18:19:59 -04:00
Kip Macsai-Goren
5880cbafe4
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-19 16:46:46 -04:00
bbracker
bc5222e721
put MTIMECMP's reset value back to 0 because the reset value of -1 broke the MCAUSE tests
2021-07-19 16:19:24 -04:00
bbracker
64e0fe4c5a
whoops MTIMECMP is always 64 bits
2021-07-19 15:40:53 -04:00
bbracker
bdb1ece183
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-19 15:13:14 -04:00
bbracker
cd469035be
make testbench check the same CSRs that QEMU logs; change CLINT to reset MTIMECMP to -1 so that we don't instantly get a timer interrupt upon reset
2021-07-19 15:13:03 -04:00
Kip Macsai-Goren
2614df627e
added changes to priority encoders from synthesis branch (correctly this time I hope)
2021-07-19 15:06:14 -04:00
Ross Thompson
bf3ca50a9a
Furture simplification of the dcache ReadDataW update.
2021-07-19 12:46:31 -05:00
Ross Thompson
b61dad4b83
Fixed a complex bug in the dcache, where back to back loads would lose data on the load before a stall occurred. The solution was to modify the logic for SelAdrM in the dcache so that a stall would cause the SRAM to reread the address in the Memory stage rather than Execution stage. This also required updating the ReadDataWEn control so it is always enabled on ~StallW.
2021-07-19 12:32:16 -05:00
Ross Thompson
4d53b9002f
Broken.
...
Possible change to walker, dcache, tlb addressing.
Improves the naming of address signals.
But has a problem when the walker finishes the dcache does not get the correct
address on the cycle the DTLB is updated. This leads to incorrect index
selection in the dcache.
2021-07-19 10:33:27 -05:00
bbracker
67eb1f5c6b
change sram1rw to have a small delay so that we don't have signals changing on clock edges
2021-07-19 11:30:07 -04:00
David Harris
2ed6285a3d
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-19 10:34:18 -04:00
James Stine
7d571f27a6
delete sbtm_a4 and sbtm_a5 as they are not needed
2021-07-19 08:06:00 -05:00
James Stine
186b5dee69
remove sbtm3.sv - not needed
2021-07-19 08:00:53 -05:00
James Stine
5b1f9797f5
update part I on sbtm change
2021-07-19 07:59:27 -05:00
David Harris
8e01007d1c
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-19 00:25:06 -04:00
Katherine Parry
c9180f4ebd
FDIV and FSQRT passes when simulating in modelsim
2021-07-18 23:00:04 -04:00
David Harris
46ab609498
Updated FMA1 with parameterized size
2021-07-18 20:40:49 -04:00
David Harris
ed64d37e65
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-07-18 17:36:29 -04:00
David Harris
4f8f52f283
Added FLEN, NE, NF to config and started using these in FMA1
2021-07-18 17:28:25 -04:00
Katherine Parry
60dabb9094
fdivsqrt inegrated, but not completley working
2021-07-18 14:03:37 -04:00
