Github_Repos/cvw
1
0
Fork 1
mirror of https://github.com/openhwgroup/cvw synced 2025-02-05 11:15:19 +00:00
Code Issues Packages Projects Releases Wiki Activity
1,867 Commits 4 Branches 2 Tags 53 MiB
694b3fbb6f
Commit Graph

150 Commits

Author SHA1 Message Date
Ross Thompson
20a04d8cee 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
25533bdc49 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
89a7b38f79 Removed 1 cycle delay on store miss. 
Changed some logic to partially support atomics.
 2021-07-30 14:00:51 -05:00
Ross Thompson
c60a1fed69 Fixed bug which caused stores to take an extra clock cycle. 2021-07-26 12:22:53 -05:00
Ross Thompson
0291d987da Added additional interlock for itlb miss -> instrPageFault with concurrent memory operation. 2021-07-25 23:14:28 -05:00
Ross Thompson
32ec457e09 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
Ross Thompson
007812dbdc 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
936e034be9 Fixed bug with the itlb fault not dcache ptw ready state to ready state. 2021-07-22 14:04:56 -05:00
Ross Thompson
42fe5ceee3 Cleaned up icache and dcache. 2021-07-22 11:06:44 -05:00
Ross Thompson
e907d57340 Tested all numbers of ways for dcache 1, 2, 4, and 8. 2021-07-22 10:38:07 -05:00
Ross Thompson
511c36fb1b 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
abe57e3fd0 Added comment about better muxing. 2021-07-21 14:40:14 -05:00
Ross Thompson
3d79dc51bb 4 way set associative is now working. 2021-07-21 14:01:14 -05:00
Ross Thompson
39fc9278ba Fixed remaining bugs in 2 way set associative dcache. 2021-07-21 10:35:23 -05:00
Ross Thompson
ba3aed8760 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
Ross Thompson
8d0a552b5b Partially working 2 way set associative d cache. 2021-07-20 17:51:42 -05:00
Ross Thompson
bb5b5e71b1 Replaced FinalReadDataM with ReadDataM in dcache. 2021-07-20 13:27:29 -05:00
David Harris
c117356432 Parameterized I$/D$ configurations and added sanity check assertions in testbench 2021-07-20 08:57:13 -04:00
Ross Thompson
ae2371f2ce Added performance counters for dcache access and dcache miss. 2021-07-19 22:12:20 -05:00
Ross Thompson
5edd513f8c Furture simplification of the dcache ReadDataW update. 2021-07-19 12:46:31 -05:00
Ross Thompson
2ee97efb9c 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
6ccbdc372d 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
986b7a8252 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
Ross Thompson
d0ed6e250a Fixed LRSC in 64bit version. 32bit version is broken. 2021-07-17 20:58:49 -05:00
Ross Thompson
053e9593af Before returning to the ready state the dcache must set SelAdr = 0 on the cycle before. 2021-07-17 18:26:29 -05:00
Ross Thompson
e9649eb1f5 Made furture progress in the mmu tests. 2021-07-16 15:56:06 -05:00
Ross Thompson
abce241f68 Also changed the shadow ram's dcache copy widths. 
Merge branch 'dcache' into main
 2021-07-16 14:21:09 -05:00
Ross Thompson
d3715acf2d Fixed walker fault interaction with dcache. 2021-07-16 12:22:13 -05:00
Ross Thompson
5ca7dc619c Updated the ptw, lsuarb and dcache to hopefully solve the interlock issues. 2021-07-16 11:12:57 -05:00
Kip Macsai-Goren
ba5bb12e26 Still broken, midway through fixing understanding of how ptw and datacache interact in time especially wrt adrE, adrM, faults, and tlb interaction. 2021-07-15 18:30:29 -04:00
Ross Thompson
c39a228266 Fixed how the dcache and page table walker stall cpu so that once a tlb miss occurs the CPU is always stalled until the walk is complete, the tlb updated, and the dcache fetched and hits. 2021-07-15 11:00:42 -05:00
Ross Thompson
c954fb510b Renamed DCacheStall to LSUStall in hart and hazard. 
Added missing logic in lsu.
 2021-07-15 10:16:16 -05:00
Ross Thompson
f234875779 dcache STATE_CPU_BUSY needs to assert CommittedM. This is required to ensure a completed memory operation is not bound to an interrupt. ie. MEPC should not be PCM when committed. 2021-07-14 23:08:07 -05:00
Ross Thompson
6163629204 Finally have the ptw correctly walking through the dcache to update the itlb. 
Still not working fully.
 2021-07-14 22:26:07 -05:00
Ross Thompson
d41c9d5ad9 Added d cache StallW checks for any time the cache wants to go to STATE_READY. 2021-07-14 17:25:50 -05:00
Ross Thompson
d3a1a2c90a Fixed d cache not honoring StallW for uncache writes and reads. 2021-07-14 17:23:28 -05:00
Ross Thompson
1d7aa27316 Fixed a bug where the dcache did not update the read data if the CPU was stalled, but the memory not stalled. 2021-07-14 15:47:38 -05:00
Ross Thompson
e17de4eb11 Separated interruptM into PendingInterruptM and InterruptM. The d cache now takes in both exceptions and PendingInterrupts. 
This solves the committedM issue.
 2021-07-14 15:00:33 -05:00
Ross Thompson
ef598d0e79 Implemented uncached reads. 2021-07-13 23:03:09 -05:00
Ross Thompson
b6e5670bc3 Added CommitedM to data cache output. 2021-07-13 22:43:42 -05:00
Ross Thompson
278bbfbe3c Partially working changes to support uncached memory access. Not sure what CommitedM is. 2021-07-13 17:24:59 -05:00
Ross Thompson
b780e471b4 Fixed interaction between icache stall and dcache. On hit dcache needs to enter a cpu busy state when the cpu is stalled. 2021-07-13 14:51:42 -05:00
Ross Thompson
51249a0e04 Fixed the fetch buffer accidental overwrite on eviction. 2021-07-13 14:21:29 -05:00
Ross Thompson
2034a6584f Dcache AHB address generation was wrong. Needed to zero the offset. 2021-07-13 14:19:04 -05:00
Ross Thompson
ee09fa5f58 Moved StoreStall into the hazard unit instead of in the d cache. 2021-07-13 13:20:50 -05:00
Ross Thompson
2004b2e044 Fixed back to back store issue. 
Note there is a bug in the lsuarb which needs to arbitrate a few execution stage signals.
 2021-07-13 12:46:20 -05:00
Ross Thompson
40922cf064 Fixed subword write. subword read should not feed into subword write. 2021-07-13 11:21:44 -05:00
Ross Thompson
9fe6190763 Team work on solving the dcache data inconsistency problem. 2021-07-12 23:46:32 -05:00
Ross Thompson
8ca8b9075d Progress towards the test bench flush. 2021-07-12 14:22:13 -05:00
Ross Thompson
282bde7205 Fixed the spurious AHB requests to address 0. Somehow by not having a default 
(else) in the fsm branch selection for STATE_READY in the d cache it was
possible to take an invalid branch through the fsm.
 2021-07-10 22:34:47 -05:00
Ross Thompson
d9fa3af94d Loads are working. 
There is a bug when the icache stalls 1 cycle before the d cache.
 2021-07-10 22:15:44 -05:00
Ross Thompson
a82c4c99c2 Actually writes the correct data now on stores. 2021-07-10 17:48:47 -05:00
Ross Thompson
ee72178eec Write miss with eviction works. 2021-07-10 15:17:40 -05:00
Ross Thompson
0a6c86af94 Write Hits and Write Misses without eviction are working correctly! The next 
step is to add eviction of dirty lines.
 2021-07-10 10:56:25 -05:00
Ross Thompson
94b29ec418 Loads in modelsim, but the first store double does not function correctly. The write address is wrong so the cache is updated using the wrong address. 
I think this is do to the cycle latency of stores.  We probably need extra muxes to select between MemPAdrM and MemPAdrW depending on if the write is a
full cache block or a word write from the CPU.
 2021-07-09 17:14:54 -05:00
Ross Thompson
7e98610651 Design loads in modelsim, but trap is an X. 2021-07-09 15:37:16 -05:00
Ross Thompson
6abd23a61d Lint passes, but I only hope to have loads working. Stores, lr/sc, atomic, are not fully implemented. 
Also faults and the dcache ptw interlock are not implemented.
 2021-07-09 15:16:38 -05:00
Ross Thompson
6041aef263 completed read miss branch through dcache fsm. 
The challenge now is to connect to ahb and lsu.
 2021-07-08 17:53:08 -05:00
Ross Thompson
4c5aee3042 This d cache fsm is getting complex. 2021-07-08 15:26:16 -05:00
Ross Thompson
adcc7afffa Partial implementation of the data cache. Missing the fsm. 2021-07-07 17:52:16 -05:00
David Harris
d2e3e14cbc Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main 2021-07-04 18:55:24 -04:00
David Harris
57e1111df3 Gave names to for loops in generate blocks for ease of reference 2021-07-04 18:52:16 -04:00
bbracker
11606e96f1 ICacheCntrl now reacts differently to InstrPageFaultF vs ITLBWriteF 2021-07-04 18:17:06 -04:00
Ben Bracker
66692af57c src/cache/ICacheCntrl.sv 2021-07-03 11:24:41 -05:00
Ben Bracker
d6c7dc02ed fix ICache indenting 2021-07-03 11:11:07 -05:00
Ross Thompson
386193de00 added page table walker fault exit for icache. 2021-07-01 17:59:55 -05:00
Ross Thompson
3dae02818c OMG. It's working! 2021-07-01 17:37:53 -05:00
Ross Thompson
4530e43df6 The icache ptw interlock is actually correct now. There needed to be a 1 cycle delay. 2021-06-30 17:02:36 -05:00
Ross Thompson
2598f08782 Page table walker now walks the table. 
Added interlock so the icache stalls.
Page table walker not walking correctly, goes to fault state.
 2021-06-29 22:33:57 -05:00
Ross Thompson
d5063bee7d Updated icache to abhlite to use pa_bits length and moved F/D stage instr register to ifu from icache. 2021-06-23 15:13:56 -05:00
Ross Thompson
03084a4128 Icache now uses physical lenght bits rather than XLEN. 2021-06-21 16:41:09 -05:00
Ross Thompson
8ec5b0c4f1 Improved some names in icache. 2021-06-21 16:40:37 -05:00
Ross Thompson
bb756849a7 Revert "Icache now uses physical lenght bits rather than XLEN." 
This reverts commit d4de8a54a2.
 2021-06-19 08:58:34 -05:00
Ross Thompson
e4c932265d Revert "Improved some names in icache." 
This reverts commit 22ea801edb.
 2021-06-19 08:58:32 -05:00
Ross Thompson
22ea801edb Improved some names in icache. 2021-06-18 12:22:41 -05:00
Ross Thompson
d4de8a54a2 Icache now uses physical lenght bits rather than XLEN. 2021-06-18 12:02:59 -05:00
David Harris
5e01f71c52 disabled Verilator WIDTH warnings in ICCacheCntrl 2021-06-12 19:50:06 -04:00
David Harris
e231fc6b00 More verilator fixes, but bpred is broken 2021-06-09 21:03:03 -04:00
Ross Thompson
e200b4b5a4 Continued I-Cache cleanup. 
Removed strange mux on InstrRawD along with
the select logic.
 2021-06-04 15:14:05 -05:00
Ross Thompson
35afdecda2 Moved I-Cache offset selection mux to icache.sv (top level). 
When we switch to set associative this is will be more efficient.
 2021-06-04 13:49:33 -05:00
Ross Thompson
fdc7c673dd Cleaned up the I-Cache memory. 2021-06-04 13:36:06 -05:00
Ross Thompson
2c16591396 Reorganized the icache names. 2021-06-04 12:53:42 -05:00
Ross Thompson
147be536f1 Relocated the icache to the cache directoy. 2021-06-04 12:23:46 -05:00
Ross Thompson
7185905f7b Reduced icache to 1 port memory. 2021-05-03 14:47:49 -05:00
Ross Thompson
12b978fec2 Eliminated extra register and fixed ports to icache. 
Still need to support physical tag check and write in icache memory.
Still need to reduce to 1 port SRAM in icache.
I would like to refactor the icache code.
 2021-05-03 12:04:54 -05:00
Ross Thompson
50e893eec9 Fixed for the instruction spills. 2021-04-21 16:47:05 -05:00
Ross Thompson
daa1ab9261 Progress on icache. Fixed some issues aligning the PC with instruction. Still broken. 2021-04-20 21:19:53 -05:00
Jarred Allen
3868a82932 dcache lints 2021-04-15 21:13:56 -04:00
Jarred Allen
c32fe09056 More icache bugfixes 2021-04-14 19:03:33 -04:00
Jarred Allen
757b64e487 Merge branch 'main' into cache 
Conflicts:
	wally-pipelined/src/cache/dmapped.sv
	wally-pipelined/src/cache/line.sv
	wally-pipelined/src/ifu/icache.sv
 2021-04-14 18:24:32 -04:00
Teo Ene
0bffac2c74 Various code syntax changes to bring HDL to a synthesizable level 2021-04-13 11:27:12 -05:00
Jarred Allen
357aed75ee A few more cache fixes 2021-04-13 01:07:40 -04:00
Jarred Allen
6ce4d44ae1 Merge from branch 'main' 2021-04-08 17:19:34 -04:00
Jarred Allen
8dc57a7706 Begin changes to direct-mapped cache 2021-04-01 13:55:21 -04:00
ushakya22
ba01d57767 Merge branch 'main' of https://github.com/davidharrishmc/riscv-wally into main 2021-03-30 15:25:07 -04:00
Jarred Allen
e8e4e1bee2 rv64i linear control flow now working 2021-03-25 13:02:26 -04:00
Jarred Allen
5a86225e1c Fix bug in cache line 2021-03-25 12:59:30 -04:00
Jarred Allen
0776127c75 Give some cache mem inputs a better name 2021-03-24 12:31:50 -04:00
Jarred Allen
abf9f3b3cb Fix compile errors from const not actually being constant (why does Verilog do this) 2021-03-24 00:58:56 -04:00
Jarred Allen
ebd2c60b74 Begin work on direct-mapped cache 2021-03-23 17:03:02 -04:00