David Harris
cb624fe679
Lint cleaning, riscv-arch-test testing
2021-09-09 11:05:12 -04:00
David Harris
a31828e925
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-09-08 16:00:12 -04:00
David Harris
30e2ec3987
Added testbench-arch for riscv-arch-test suite
2021-09-08 15:59:40 -04:00
Ross Thompson
8141a515bb
Changed configs to support 4 ways set associative caches.
2021-09-08 12:52:49 -05:00
Ross Thompson
6606eea27e
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-09-08 12:47:03 -05:00
Ross Thompson
5bc90ef32f
Slight modification to wave file.
2021-09-08 10:40:46 -05:00
bbracker
5e9a39e755
fixed bug where M mode was sensitive to S mode traps
2021-09-07 19:14:39 -04:00
bbracker
b3f00f2682
make testbench successfully deactivate TimerIntM so as to create a nice pulse
2021-09-07 15:36:47 -04:00
Ross Thompson
150a73d6cf
Set associate icache working, but way 0 is never written.
2021-09-07 12:46:16 -05:00
bbracker
28fed18421
No longer forcing CSRReadValM because that can feedback to corrupt some CSRs
2021-09-06 22:59:54 -04:00
Ross Thompson
00f50184d8
Changed name of memory in icache.
2021-09-06 20:54:52 -05:00
bbracker
0646bf2b90
help in case a script is run accidentally
2021-09-06 16:23:45 -04:00
bbracker
a13b561759
modified testbench to not allow Wally to generate its own interrupts (because of fundamental interrupt imprecision limitations)
2021-09-04 19:49:26 -04:00
bbracker
58d478eb23
restore functionality of being able to turn on waves at a certain instruction count; restore linux-waves.do because wave.do seems to be in disrepair
2021-09-04 19:45:04 -04:00
bbracker
0004f647ec
switching over to hopefully more consistent QEMU simulated clock
2021-09-04 19:43:39 -04:00
bbracker
6155716de4
replace triple gdb breakpoint continue with a double breakpoint ignore in hopes of improving parsing
2021-09-04 19:41:55 -04:00
James E. Stine
5bc3569b0e
Not sure I understand the Misaligned hptw - seems like a bug and should be L1_ADR instead of L0_ADR
2021-09-03 10:26:38 -05:00
bbracker
4a938e493e
output trace to linux-testvectors folder
2021-09-01 17:37:46 -04:00
Ross Thompson
5c2deab4e4
Partial multiway set associative icache.
2021-08-30 10:49:24 -05:00
Katherine Parry
7607adc951
FMA cleanup
2021-08-28 10:53:35 -04:00
Ross Thompson
4b0344898b
Fixed bugs I introduced to the icache.
2021-08-27 15:00:40 -05:00
Ross Thompson
2dff72d9e9
Renamed PCMux (icache) to SelAdr to match dcache.
...
Removed unused cache files.
2021-08-27 11:14:10 -05:00
Ross Thompson
de9e234ffa
Modified icache to no longer need StallF in the PCMux logic. Instead this is handled in the icachefsm.
...
One downside is it increases the icache complexity. However it also fixes an untested bug. If a region
was uncacheable it would have been possible for the request to be made multiple times. Now that is
not possible. Additionally spills were oscillating between the spill hit states without this change.
The impact was 'benign' as the final spilled instruction always had the correct upper 16 bits.
2021-08-27 11:03:36 -05:00
Ross Thompson
62d91e9ea1
Renamed ICacheCntrl to icachefsm.
2021-08-26 15:57:17 -05:00
Ross Thompson
cbb47956cb
Swapped out the icachemem for cacheway. cacheway is modified to optionally support dirty bits.
2021-08-26 15:43:02 -05:00
Ross Thompson
b230d4daec
Finished moving data path logic from the ICacheCntrl.sv to icache.sv.
2021-08-26 13:06:24 -05:00
Ross Thompson
b3849d8abb
Moved data path logic from icacheCntrl to icache.
2021-08-26 10:58:19 -05:00
Ross Thompson
c83f0a2e99
Removed unused logic in icache.
2021-08-26 10:49:54 -05:00
Ross Thompson
642efbb563
Converted the icache type from logic to state type.
2021-08-26 10:41:42 -05:00
Ross Thompson
b5d6c4fb46
Additional cleanup of ahblite.
2021-08-25 22:53:20 -05:00
Ross Thompson
bf312bb37c
Removed amo logic from ahblite. Removed many unused signals from ahblite.
2021-08-25 22:45:13 -05:00
Ross Thompson
939ff663a5
Forgot to include a few files in the last few commits.
...
Also reorganized the dcache by read cpu path, write cpu path, and bus interface path.
Changed i/o names on subwordread to match signals in dcache.
2021-08-25 22:30:05 -05:00
Ross Thompson
d2b3b7345e
Moved dcache fsm to separate module.
2021-08-25 21:37:10 -05:00
Ross Thompson
7be0a73db1
Moved LRU and storage for the LRU into a single module. Also found a subtle bug with the update address used to write the cache's memory.
...
This was correct for the LRU but incorrect for the data, tag, valid, and dirty storage.
2021-08-25 21:09:42 -05:00
Ross Thompson
b5eba44417
Replaced dcache generate ORing with or_rows.
2021-08-25 13:46:36 -05:00
Ross Thompson
83cc0266b2
Rename of DCacheMem to cacheway.
...
simplified dcache names.
2021-08-25 13:33:15 -05:00
Ross Thompson
c48556836b
Removed generate around the dcache memories.
2021-08-25 13:27:26 -05:00
Ross Thompson
7139279e50
Moved more logic inside the dcache memory.
2021-08-25 13:17:07 -05:00
Ross Thompson
a99b5f648b
partial dcache reorg.
2021-08-25 12:42:05 -05:00
Ross Thompson
699053bab0
Updated linux test bench documenation and scripts.
2021-08-25 10:54:47 -05:00
David Harris
cb13e36d20
Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main
2021-08-25 06:47:20 -04:00
David Harris
cf1e458ccf
simplified or_rows generation and renamed oneHotDecoder to onehotdecoder
2021-08-25 06:46:41 -04:00
Ross Thompson
b7972eafeb
Added function tracking to linux test bench.
2021-08-24 11:08:46 -05:00
Ross Thompson
bb3e94d68a
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
97653e1aea
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
f006655bdc
Renamed output of qemu trace.
2021-08-22 22:56:34 -05:00
Ross Thompson
b6e2710f5d
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
696be3ff68
possible interrupt code
2021-08-22 17:02:40 -04:00
Ross Thompson
c0667f30bb
Fixed bug with coremark do file. When I moved the testbench to have a common set of files i forgot to remove the old path reference to function_radix.sv in wally-coremark_bare.do.
2021-08-19 10:33:11 -05:00
Ross Thompson
95f5ebaf30
Merge branch 'main' of github.com:davidharrishmc/riscv-wally into main
2021-08-17 16:06:54 -05:00
