Ross Thompson
a15d6c1c96
Slight modification to wave file.
2021-09-08 10:40:46 -05:00
Ross Thompson
05455f8392
Changed name of memory in icache.
2021-09-06 20:54:52 -05:00
Ross Thompson
2968623f9a
Partial multiway set associative icache.
2021-08-30 10:49:24 -05:00
Ross Thompson
96cbd8e785
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
d6ff89b7e6
Swapped out the icachemem for cacheway. cacheway is modified to optionally support dirty bits.
2021-08-26 15:43:02 -05:00
Ross Thompson
e4bbd3bbc7
Converted the icache type from logic to state type.
2021-08-26 10:41:42 -05:00
Ross Thompson
0530047f53
Moved dcache fsm to separate module.
2021-08-25 21:37:10 -05:00
Ross Thompson
d23b860c96
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
e9a1dc90f6
Removed generate around the dcache memories.
2021-08-25 13:27:26 -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
Ross Thompson
e166cc84ee
Patched up changes for wally-pipelined.do and wally-buildroot.do to support moved common testbench files.
2021-07-30 14:24:50 -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
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
25a8920a69
Tested all numbers of ways for dcache 1, 2, 4, and 8.
2021-07-22 10:38:07 -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
e0990535e1
Fixed remaining bugs in 2 way set associative dcache.
2021-07-21 10:35:23 -05:00
Ross Thompson
14e949d6e3
Partially working 2 way set associative d cache.
2021-07-20 17:51:42 -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
Ross Thompson
6521d2b468
Also changed the shadow ram's dcache copy widths.
...
Merge branch 'dcache' into main
2021-07-16 14:21:09 -05:00
Ross Thompson
b3bf04d474
Updated wave file.
2021-07-16 12:34:37 -05:00
Ross Thompson
46bce70e42
Fixed walker fault interaction with dcache.
2021-07-16 12:22:13 -05:00
Ross Thompson
e5d624c1fa
Found bug in the PMA such that invalid addresses were sent to the tim. Once addressing this issue the sv48 test fails early with a pma access fault.
2021-07-15 11:56:35 -05:00
Ross Thompson
fd1de6b047
Updated wave file.
2021-07-15 11:04:49 -05:00
Ross Thompson
704f4f724e
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
ba1e1ec231
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
2c946a282f
Fixed d cache not honoring StallW for uncache writes and reads.
2021-07-14 17:23:28 -05:00
Ross Thompson
e91501985c
Routed CommittedM and PendingInterruptM through the lsu arb.
2021-07-14 16:18:09 -05:00
Ross Thompson
9b756d6a94
Implemented uncached reads.
2021-07-13 23:03:09 -05:00
Ross Thompson
3e57c899a2
Partially working changes to support uncached memory access. Not sure what CommitedM is.
2021-07-13 17:24:59 -05:00
Ross Thompson
baa2b5d15f
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
3c1a717399
Fixed the fetch buffer accidental overwrite on eviction.
2021-07-13 14:21:29 -05:00
Ross Thompson
32f27cfecf
Dcache AHB address generation was wrong. Needed to zero the offset.
2021-07-13 14:19:04 -05:00
Ross Thompson
afc1bc9c38
Moved StoreStall into the hazard unit instead of in the d cache.
2021-07-13 13:20:50 -05:00
Ross Thompson
e594eb540d
Got the shadow ram cache flush working.
2021-07-13 10:03:47 -05:00
Ross Thompson
49f6eec579
Team work on solving the dcache data inconsistency problem.
2021-07-12 23:46:32 -05:00
Ross Thompson
1cc258ade1
Progress towards the test bench flush.
2021-07-12 14:22:13 -05:00
Ross Thompson
60ed023734
Actually writes the correct data now on stores.
2021-07-10 17:48:47 -05:00
Ross Thompson
6e7e318396
Fixed bug in the LSU pagetable walker interlock.
2021-07-06 10:41:36 -05:00
Ross Thompson
a252416535
Removed the TranslationVAdrQ as it is not necessary.
2021-07-04 16:49:34 -05:00
Ross Thompson
5b70eb86b0
relocated lsuarb and pagetable walker inside the lsu. Does not pass busybear or buildroot, but passes rv32ic and rv64ic.
2021-07-04 13:49:38 -05:00
Ross Thompson
61027f650c
OMG. It's working!
2021-07-01 17:37:53 -05:00
Ross Thompson
2dc349ea6f
Fixed the wrong virtual address write into the dtlb.
2021-07-01 16:55:16 -05:00
Ross Thompson
88a18496cf
Got some stores working in virtual memory.
2021-07-01 12:49:09 -05:00
Ross Thompson
002c32d2ad
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
9ec624702d
Major rewrite of ptw to remove combo loop.
2021-06-30 16:25:03 -05:00
Ross Thompson
b2d8ba6742
The icache now correctly interlocks with the PTW on TLB miss.
2021-06-30 11:24:26 -05:00
Ross Thompson
dd84f2958e
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
bc9c944ba0
Don't use this branch walker still broken.
2021-06-28 17:26:11 -05:00
Ross Thompson
d80ebab941
AMO and LR/SC instructions now working correctly.
...
Page table walking is not working.
2021-06-25 15:42:07 -05:00
