1
0
Commit Graph

5193 Commits

Author SHA1 Message Date
Henry Cook
60934ac622 coreplex: TilePortParams use BasicBusBlockers 2017-10-11 13:36:46 -07:00
Henry Cook
2dbe882e58 tilelink: add BasicBusBlocker device 2017-10-11 13:36:42 -07:00
Henry Cook
9f8e3d8879 tilelink: BusBypass can be sent to DeadlockDevice 2017-10-11 12:45:36 -07:00
Henry Cook
ec056535dc tilelink: add DeadlockDevice 2017-10-11 12:44:23 -07:00
Henry Cook
329a5c35d4 tilelink: unsafe cache cork discards outer d.sink 2017-10-11 00:30:51 -07:00
Henry Cook
1240cb275c coreplex: TilePortParams formatting 2017-10-11 00:29:11 -07:00
Henry Cook
1867a5b226 rocket: only cache when AcquireT is possible 2017-10-10 18:06:58 -07:00
Henry Cook
37406706b4 coreplex: move CacheCork in front of SBus
Continue to not allow caches to cache ROMs.
Update TinyConfig and WithStatelessBridge.
2017-10-10 16:24:32 -07:00
Henry Cook
8f5f80f958 coreplex: TileSlavePortParams inject adapters into PBus 2017-10-10 15:25:08 -07:00
Henry Cook
660355004e coreplex: TileMasterPortParams inject adapters into SBus 2017-10-10 15:02:50 -07:00
Henry Cook
9026646459 coreplex: first cut at using RocketCrossingParams 2017-10-10 12:02:04 -07:00
Wesley W. Terpstra
d6766a8c68 RocketTile: make sure 'hartid' is available for traits (#1037) 2017-10-09 21:03:18 -07:00
Andrew Waterman
a9686ab883 Merge pull request #1035 from freechipsproject/big-paddr
Fix paddrBits < xLen && paddrBits == vaddrBits case
2017-10-09 20:59:21 -07:00
Andrew Waterman
2c4009a138 Fix paddrBits < xLen && paddrBits == vaddrBits case
Require and/or force vaddrBits to be bigger than paddrBits so there's
room to zero-extend a physical address by 1 bit, so that when the virtual
address is sign-extended, the sign is zero.
2017-10-09 16:48:04 -07:00
Megan Wachs
0e6aa7ae9d Merge pull request #1024 from freechipsproject/jtag_coverage
Add Coverage points for JTAG TAP
2017-10-09 12:29:18 -07:00
Andrew Waterman
d78ad857ee Merge pull request #1034 from freechipsproject/base-tile
Generalize Tile/Coreplex hierarchy
2017-10-09 11:42:20 -07:00
Megan Wachs
0916cf1bdd JTAG Coverage: Correct jtag_reset case 2017-10-09 09:54:15 -07:00
Megan Wachs
9efe1c448e Merge remote-tracking branch 'origin/master' into HEAD 2017-10-09 09:48:38 -07:00
Andrew Waterman
986cbfb6b1 For Rockets without VM, widen vaddrBits to paddrBits
This supports addressing a >39-bit physical address space.
2017-10-08 01:21:47 -07:00
Andrew Waterman
a0e5a20b60 Don't route branch comparison result through ALU output mux
This potentially mitigates a critical path, and makes the ALU usable
in processors that have dedicated branch comparators.
2017-10-07 17:36:24 -07:00
Andrew Waterman
36c39d01e4 Factor out most of HasRocketTiles into HasTiles 2017-10-07 17:36:24 -07:00
Andrew Waterman
70a4127cb8 Factor out some of HaveRocketTiles into HaveTiles 2017-10-07 17:36:24 -07:00
Andrew Waterman
34e96c03b1 Move HCF to BaseTile 2017-10-07 17:36:24 -07:00
Andrew Waterman
71205b70cc Make RocketTileWrapper a BaseTile 2017-10-07 17:36:24 -07:00
Andrew Waterman
4645b61fd3 Decouple BaseTile from HasTileLinkMasterPort 2017-10-07 17:36:24 -07:00
Andrew Waterman
86a1953287 Merge pull request #1032 from freechipsproject/fpga_pipeline_fpu_master
FPU FMA FPGA retiming assist
2017-10-05 20:11:34 -07:00
Henry Styles
5498468743 FPU : simplify pipeline register generation in FMA 2017-10-05 15:18:19 -07:00
Henry Styles
7a46715cbc FPU : to assist retiming move upto first 2 register stages of into FMA 2017-10-05 15:18:04 -07:00
Wesley W. Terpstra
bd045a3b95 tilelink: split Acquire into Acquire{Block,Perm} (#1030)
We had planned for a while to add an 'Overwrite' message which obtains
permissions without requiring retrieval of data. This is useful whenever
a master knows it will completely replace the contents of a cache block.

Instead of calling it Overwrite, we decided to split the Acquire type.

If you AcquirePerm, you MUST Release and ProbeAck with Data.
2017-10-05 12:49:49 -07:00
Wei Song (宋威)
81b9ac42a3 add comments to diplomacy resource. (#913) 2017-10-05 12:45:56 -07:00
Henry Cook
9040d921b5 Merge pull request #1031 from freechipsproject/non-contiguous-hartids
Miscellaneous multicore cleanup
2017-10-05 12:44:31 -07:00
Henry Cook
8da7aabd51 tile: supply hartid from RocketTileParams
make WithNCores partial configs override rather than append more tiles
2017-10-05 00:31:53 -07:00
Henry Cook
45581e60f0 Revert "Merge pull request #1027 from freechipsproject/dont-touch-hartid"
This reverts commit 5232a29d7d, reversing
changes made to a2dc13669a.
2017-10-05 00:26:44 -07:00
Andrew Waterman
5a84564203 Merge pull request #1023 from freechipsproject/csr-cleanup
Generalize CSR file to support simpler cores
2017-10-04 14:04:59 -07:00
Andrew Waterman
32fda51a2c Get rid of paddrBits from SystemBus (#1029) 2017-10-04 12:11:37 -07:00
Andrew Waterman
7bcf28c585 Define fetchBytes in HasCoreParams, not Frontend
It is more generally useful.
2017-10-03 17:34:18 -07:00
Andrew Waterman
2786e42d99 Don't register interrupts in CSRFile
They are usually registered outside the tile in a CDC.
2017-10-03 17:34:18 -07:00
Andrew Waterman
5cfe070932 Add option to make misa read-only 2017-10-03 17:34:18 -07:00
Andrew Waterman
09468a272b Add option to remove basic counters (mcycle/minstret) 2017-10-03 17:34:18 -07:00
Andrew Waterman
ab0821f25b Move microarchitecture-neutral params from Rocket to Core
This makes some of the units more reusable.
2017-10-03 17:34:18 -07:00
Andrew Waterman
190d5c50d9 Remove deprecated custom-CSR support 2017-10-03 17:34:18 -07:00
Henry Cook
5232a29d7d Merge pull request #1027 from freechipsproject/dont-touch-hartid
Make use of the new DontTouch annotation
2017-10-03 12:55:34 -07:00
Henry Cook
d33737802a util: add DontTouch trait with dontTouchPorts method 2017-10-02 19:36:34 -07:00
Henry Cook
aa3a18222c HellaCache: users like to peep resp.data and resp.addr 2017-10-02 19:36:30 -07:00
Henry Cook
cedfb0e784 coreplex: dontTouch the rocket_tile_inputs wire
which contains hartid.
2017-10-02 19:36:10 -07:00
Wesley W. Terpstra
a2dc13669a Error grants (#1025)
* CacheCork: an error Grant still says 'toT' even though it is transient

Grants with errors must be handled by a client as though no actual
permissions were obtained, so that two clients do not both end up believing
that they own a block which is only temporarily offline. However, the
Grant MESSAGE should still match the request; ie. if you did Acquire.NtoT,
the response must be Grant.toT, even though the 'error' bit signals that
the Grant actually grants no permissions.

This keeps the implementation of request-response tracking in interstitial
adapters and FSMs simple, consistent with the way multibeat errors must
include all their beats.

* Error: handle permissions properly
2017-10-02 14:49:25 -07:00
Megan Wachs
9c9cb68462 JTAG Coverage: Add reset coverage points 2017-10-02 11:08:13 -07:00
Megan Wachs
a8ab06d572 JTAG: Add coverage points to the JTAG Tap 2017-10-02 11:08:13 -07:00
Jack Koenig
723af5e6b6 Merge pull request #971 from freechipsproject/bump-chisel-firrtl
Bump chisel3 and firrtl, update plugin versions
2017-09-29 17:24:12 -07:00
Jack Koenig
8891bf1b64 Bump chisel3 and firrtl, update plugin versions
And update chisel3 code
2017-09-29 15:44:27 -07:00