Andrew Waterman
5df8f0d1ea
Add L2 TLB miss counter
2017-07-25 15:19:16 -07:00
Henry Cook
01ca3efc2b
Combine Coreplex and System Module Hierarchies ( #875 )
...
* coreplex collapse: peripherals now in coreplex
* coreplex: better factoring of TLBusWrapper attachement points
* diplomacy: allow monitorless :*= and :=*
* rocket: don't connect monitors to tile tim slave ports
* rename chip package to system
* coreplex: only sbus has a splitter
* TLFragmenter: Continuing my spot battles on requires without explanatory strings
* pbus: toFixedWidthSingleBeatSlave
* tilelink: more verbose requires
* use the new system package for regression
* sbus: add more explicit FIFO attachment points
* delete leftover top-level utils
* cleanup ResetVector and RTC
2017-07-23 08:31:04 -07:00
Wesley W. Terpstra
4eface8a9e
rocket: do not require FIFO order for memory-like regions
2017-07-12 17:39:00 -07:00
Henry Cook
4c595d175c
Refactor package hierarchy and remove legacy bus protocol implementations ( #845 )
...
* Refactors package hierarchy.
Additionally:
- Removes legacy ground tests and configs
- Removes legacy bus protocol implementations
- Removes NTiles
- Adds devices package
- Adds more functions to util package
2017-07-07 10:48:16 -07:00
Andrew Waterman
a0cbc376b4
Merge pull request #849 from freechipsproject/l2-tlb
...
L1 memory system improvements
2017-07-06 13:03:06 -07:00
Andrew Waterman
e1cc0a0a0e
Mask debug interrupts similarly to other interrupts ( #847 )
...
This makes single-step exceptions higher-priority than debug interrupts.
2017-07-06 12:03:24 -07:00
Andrew Waterman
b2351c5fbf
Use consistent casing
2017-07-06 11:16:56 -07:00
Andrew Waterman
be4eceec0d
Fix stupid D$ probe bug
2017-07-06 01:20:47 -07:00
Andrew Waterman
90a7d6a343
Add L2 TLB option
2017-07-06 01:19:18 -07:00
Andrew Waterman
438abc76d2
Handle TL errors in L1 I$
...
Cache the error bit in the tag array; report precisely on access.
2017-07-06 01:02:11 -07:00
Andrew Waterman
0ef45fac9b
Add tag ECC to D$
2017-07-03 18:16:37 -07:00
Andrew Waterman
e9752f76ae
Improve probe state machine
...
- Reduce reliance on s2_prb_ack_data due to future ECC changes
- Shave a cycle off valid, but clean, probes
- Code cleanup
2017-07-03 16:25:04 -07:00
Richard Xia
5b46350bc3
Make sure that DCache s2_xcpt data scratchpad case is assigned to after initial assignment.
2017-06-30 17:44:16 -07:00
Wesley W. Terpstra
5edc4546e3
rocket: add dtim and itim refs to cpus
2017-06-28 23:10:58 -07:00
Wesley W. Terpstra
7d6f8d48f2
Revert "rocket: link dtim to its cpu"
...
This reverts commit e6c2d446cc
.
2017-06-28 23:10:57 -07:00
Wesley W. Terpstra
fbcd6f0eb2
Revert "rocket: link itim to its cpu"
...
This reverts commit 48390ed604
.
2017-06-28 23:10:57 -07:00
Henry Cook
6e5a4c687f
diplomacy: a type of connect that always disables monitors ( #828 )
2017-06-28 21:48:10 -07:00
Megan Wachs
992b480c74
Merge pull request #825 from freechipsproject/debug_wfi
...
Debug + WFI Interactions
2017-06-28 21:28:51 -07:00
Wesley W. Terpstra
ca3030cba3
dcache: fix a gender inversion bug introduced in #826
2017-06-28 15:38:53 -07:00
Wesley W. Terpstra
48390ed604
rocket: link itim to its cpu
2017-06-28 15:06:19 -07:00
Wesley W. Terpstra
e6c2d446cc
rocket: link dtim to its cpu
2017-06-28 15:06:19 -07:00
Wesley W. Terpstra
3f6d5110cd
rocket: dtim is not a dcache
2017-06-28 15:06:19 -07:00
Wesley W. Terpstra
84dc23c215
devices: add reg-names to most devices
2017-06-28 15:06:16 -07:00
Wesley W. Terpstra
852f03282f
rocket: give itim and dtim a compatible field for drivers to match
2017-06-28 14:26:55 -07:00
Andrew Waterman
b9a934ae28
Support eccBytes > 1
2017-06-28 02:09:18 -07:00
Andrew Waterman
8e4be40efc
Propagate wb_reg_rs2 for sfence ASID
...
This would have been a bug if we supported ASIDs.
2017-06-28 02:09:18 -07:00
Andrew Waterman
2077e4190b
Make log more sensible for long-latency operations
...
Show only one write to the destination register, not two.
2017-06-28 02:09:18 -07:00
Andrew Waterman
6f8fdff762
Basic L1 D$ ECC support
...
Only supports ECC on data, not tags; only supports byte granularity.
2017-06-28 02:09:18 -07:00
Andrew Waterman
6100600179
Minor D$ code cleanup
2017-06-28 02:09:18 -07:00
Andrew Waterman
3e04a99f61
Refactor frontend exception passing
...
Bundle them, and leverage regularity, so that if we have to add more
exceptions in the future, we don't need to change so much code.
2017-06-28 02:09:18 -07:00
Andrew Waterman
cc2f87c214
Forbid S-mode execution from user memory
...
285c81746f
2017-06-28 02:09:18 -07:00
Andrew Waterman
8aa16a11f3
Reduce D$ access energy when refill width > access width
2017-06-28 02:09:18 -07:00
Andrew Waterman
25f585f2a9
Remove unused signal from TLB interface
2017-06-28 02:09:18 -07:00
Andrew Waterman
d5f80df0ae
Allow speculative I$ refill to cacheable regions
...
Backpedaling on 27b143013f
. Shaving
four cycles off of I$ miss penalty is obviously worth the HW cost.
2017-06-28 02:09:18 -07:00
Megan Wachs
e1fe0f245b
debug: Don't reset debugint register, as none of the interrupt registers are.
2017-06-27 14:10:13 -07:00
Megan Wachs
136e4b6c27
debug: use consistent coding style (Reg(init ... ) vs RegInit)
2017-06-27 13:42:38 -07:00
Megan Wachs
3b9550ede3
debug: correctly declare reg_debugint
2017-06-27 13:42:38 -07:00
Megan Wachs
56839b2c62
debug: Remove DebugInterrupt from DCSR (it is no longer part of V13 spec)
2017-06-27 13:42:38 -07:00
Megan Wachs
665c2a349c
Correct Debug + WFI Interactions
...
1) Debug interrupt should end WFI
2) WFI should end immedately during single-step
3) WFI should act like NOP during Debug Mode
2017-06-27 13:42:38 -07:00
Zihao Yu
c9cfe46604
rocket,Rocket: fix type mismatch ( #819 )
2017-06-27 11:22:08 -07:00
Colin Schmidt
aced18b3bb
Move RoCC interface to Diplomacy and TL2 ( #807 )
...
* Move RoCC interface to Diplomacy and TL2
* guard rocc arbiter to prevent zero-width wires
2017-06-22 12:07:09 -07:00
Henry Cook
5552f23294
tims: explictly name them for generated address map
2017-06-20 17:18:29 -07:00
Henry Cook
6b79842e66
dcache: just left shift by untagbits to get tag
...
Always safe because of the requirement on coreplex/RocketTiles.scala:126
2017-06-20 16:35:28 -07:00
Andrew Waterman
f396b4142d
Merge pull request #806 from freechipsproject/mulh
...
Improve integer mul/div
2017-06-20 13:01:16 -07:00
Colin Schmidt
675f183dd2
refactor ICache to be reusable by other frontends ( #808 )
...
* refactor ICache to be reusable by other frontends
specifically one that would like to change the fetch width and number of
bytes in an instruction
2017-06-20 08:21:01 -07:00
Andrew Waterman
a6d9884cc0
Improve integer mul/div
...
- Signed integer multiplication latency is now deterministic (before,
it would take an extra cycle if the multiplier was negative).
- High-part multiplication is now one cycle faster.
- RV64 MULW now takes half as many cycles as MUL.
- Positive remainders are now one cycle faster.
2017-06-19 12:09:21 -07:00
Richard Xia
61c39da475
Check for rvc before declaring illegal instruction after an ebreak.
2017-06-16 10:49:36 -07:00
Andrew Waterman
8552c77972
Fix I$ reset regression FU-357
...
Can't rely on s2 TLB response, so mask using s2_valid.
2017-06-09 00:48:24 -07:00
Andrew Waterman
16ecbdd5b2
Reduce fanout on critical I$ miss signal
2017-06-02 20:45:50 -07:00
Andrew Waterman
27b143013f
Improve ITLB QoR
...
- No need to check cacheability
- Remove a gate delay from PMP path
2017-06-02 20:45:50 -07:00