Andrew Waterman
87ad1a5703
More control cleanup
2015-01-04 19:46:01 -08:00
Andrew Waterman
2aee85cb11
Flush pipeline from MEM stage
...
This means we no longer have to rely on the instruction behind a serializing
instruction being valid, simplifying the control. But we have to be a
little more cautious when flusing the I$/ITLB/BTB.
2015-01-04 16:40:16 -08:00
Andrew Waterman
94b75c7cb1
Continue refactoring control
2015-01-04 15:32:05 -08:00
Andrew Waterman
6181de4cc9
Much refactor, so control
2015-01-03 13:34:38 -08:00
Henry Cook
1cb65d5ec1
%s/master/manager/g
2014-12-29 22:56:18 -08:00
Henry Cook
77e5e6b561
refill bug
2014-12-17 19:29:28 -08:00
Henry Cook
08dcf4c6ca
refactor cache params
2014-12-17 14:28:05 -08:00
Henry Cook
d29793d1f7
cleanup CoherenceMetadata and coherence params
2014-12-15 19:23:38 -08:00
Henry Cook
c9320862ae
add l2 dmem signal to rocc
2014-12-12 16:55:08 -08:00
Henry Cook
72ea24283b
multibeat TL; passes all tests
2014-12-12 16:54:33 -08:00
Henry Cook
b7b2923bff
Cleanup MSHR internal bundles
2014-11-11 18:18:35 -08:00
Henry Cook
c9e7874818
Major tilelink revision for uncached message types
2014-11-11 17:36:48 -08:00
Henry Cook
bf901e4bca
Remove master_xact_id from Release
2014-11-06 12:09:45 -08:00
Andrew Waterman
7bb7299018
Don't pollute BTB with PC+4 target predictions
2014-10-14 17:28:37 -07:00
Andrew Waterman
cde7c9d869
simplify CSR decoding code
2014-10-03 14:31:26 -07:00
Christopher Celio
9cc35dee9a
Returned history update to fetch.
...
- Global history only contains branches.
- Only update BHT and history on BTB hits.
- Gate off speculative update on stall or icmiss.
- Fixed bug where BHT updates were delayed a cycle.
2014-09-29 21:41:07 -07:00
Christopher Celio
8ccd07cfeb
Moved updating global history from fetch to decode.
...
- No longer update global history in fetch stage.
- Only update global history when instruction is a branch.
- Does allow for the possibility of back-to-back branches to see
slightly different histories on subsequent executions.
2014-09-28 05:16:36 -07:00
Christopher Celio
681b43f398
Bug fixes with global history register.
...
- Updated in fetch speculatively.
* Updates gated off by cpu.resp.fire().
* BTB direction factored into history update.
- All branches update the BHT.
- Each instruction carries history; index into BHT is recomputed by
passing in mem_reg_pc.
2014-09-26 10:39:57 -07:00
Christopher Celio
a71bdbbc54
Update history register in fetch speculatively
2014-09-26 05:42:08 -07:00
Christopher Celio
f917810061
Removed RocketCoreParameters from use.
...
- The nbdache (among others?) use CoreParameters, which has nothing to do with RetireWidth requirements.
- This conflicts with other cores which uses nbdcache.
- RocketCoreParameters may be unneccessary, and the require() check can be moved deeper into Rocket.
2014-09-26 05:14:50 -07:00
Christopher Celio
868e747656
Factored out Rocket specifics from CoreParameters
...
- Added new RocketCoreParameters
- Other cores using Rocket as a library will no longer conflict against
Rocket's requires().
2014-09-25 18:52:58 -07:00
Henry Cook
8eb64205f5
bug fix for nbdcache s2_data
2014-09-25 12:00:20 -07:00
Henry Cook
b55c38cdc7
Remove spurious vec consts
2014-09-25 12:00:20 -07:00
Adam Izraelevitz
3e256439c9
Add abstract class Tile
2014-09-24 13:04:20 -07:00
Yunsup Lee
238f7761f6
update README
2014-09-17 11:23:25 -07:00
Yunsup Lee
8abf62fae3
add LICENSE
2014-09-12 18:06:41 -07:00
Yunsup Lee
25180b71f7
add LICENSE
2014-09-12 15:36:42 -07:00
Andrew Waterman
a999c055ed
Don't take an interrupt when EX stage PC is invalid
...
It was possible to take an interrupt on the instruction in the shadow of
a short forward branch. EPC would thus get the wrong value, and so
a wrong-path instruction would be executed upon return from interrupt.
h/t Yunsup
2014-09-11 01:46:52 -07:00
Henry Cook
5eb5e9eaf5
Standardize ()=>Module(...) top-level Parameters
2014-09-07 17:54:41 -07:00
Henry Cook
5e2f98747f
Merge branch 'dse'
2014-09-06 06:10:15 -07:00
Scott Beamer
600c5d50a9
better fix with explanation of sbt issue
2014-09-02 15:14:56 -07:00
Scott Beamer
f9922a106b
fixes sbt error during first run
2014-09-02 14:34:36 -07:00
Henry Cook
b42a2ab40a
Final parameter refactor
2014-09-01 13:28:58 -07:00
Adam Izraelevitz
2d6aafc32e
Merge branch 'dse' of github.com:ucb-bar/rocket-staging into HEAD
2014-09-01 11:23:50 -07:00
Sagar Karandikar
83c6c2c9e2
rename refs to zynq-fpga to fpga-zynq
2014-08-29 10:26:48 -07:00
Henry Cook
6a4193cf90
minor cache param cleanup
2014-08-19 11:38:46 -07:00
Henry Cook
2de268b3b1
Cache utility traits. Completely compiles, asm tests hang.
2014-08-19 11:38:20 -07:00
Henry Cook
ca5f38ff26
a few more fixes. some param lookups fail (here() in Alter blocks)
2014-08-19 11:38:11 -07:00
Henry Cook
0dac9a7467
Full conversion to params. Compiles but does not elaborate.
2014-08-19 11:38:02 -07:00
Adam Izraelevitz
4e6d69892d
Added initial brainstorm for parameter hierarchical flattening, does not compile ;)
2014-08-19 11:37:50 -07:00
Adam Izraelevitz
812353bace
Ported FPU parameters to new Chisel Parameters
2014-08-19 11:37:27 -07:00
Yunsup Lee
4ac8e59b1f
add .gitignore
2014-08-18 19:27:50 -07:00
Yunsup Lee
d520846638
add README and sbt files
2014-08-18 19:23:10 -07:00
Andrew Waterman
7bffc6c586
rename Unsigned.size to Unsigned.clog2
2014-06-14 13:58:07 -07:00
Andrew Waterman
3828c628c3
Remove vestigial control signals
2014-06-14 13:58:07 -07:00
Andrew Waterman
04593d433e
clean up Int <-> Boolean conversion stuff
2014-06-14 13:58:07 -07:00
Andrew Waterman
ac88ded35a
Use ROMs to reduce node count and improve QoR a bit
2014-06-14 13:58:07 -07:00
Andrew Waterman
88899eafe0
Reduce node count a bit
2014-06-14 13:58:07 -07:00
Jim Lawson
0c93567dea
Replace needWidth() with getWidth.
2014-06-13 14:58:52 -07:00
Jim Lawson
de32595fba
Quick change to work with new Width class.
2014-06-13 12:00:50 -07:00