1
0
Commit Graph

252 Commits

Author SHA1 Message Date
03bca77b33 tilelink2 Metadata: cannot assert data good when !valid 2016-11-18 17:16:12 -08:00
37a3c22639 rocketchip: move from using cde to config 2016-11-18 16:18:33 -08:00
94086f2270 [tl2] broadcast hub probe port width bugfix 2016-11-17 18:42:59 -08:00
960c2723ab [tl2] MemoryOpCategories: use def to supply Cat'd consts 2016-11-17 18:42:59 -08:00
dfc3a0dafb tilelink2: do not depend on obsolete TL1 configuration 2016-11-17 14:07:53 -08:00
24e3216fcf coreplex: allow zero interrupt sink/sources 2016-11-16 16:50:36 -08:00
479bc82f03 tilelink2 Broadcast: improve bufferless throughput 2016-11-16 16:50:36 -08:00
1f51564577 [rocket] dcache probe ack data bugfix 2016-11-16 14:25:21 -08:00
5d2e637a4a tilelink2 Legacy: uncached TL never needs manager_xact_id 2016-11-16 12:16:25 -08:00
10e459fedb rocket: change connection between rocketchip and coreplex
* rtc and dtm are now crossed half-and-half on the two sides
* groundtest no longer uses riscv platform traits
2016-11-15 18:27:52 -08:00
ab3dafb8bc Monitor: restore Probe&Acquire checks 2016-11-14 15:36:52 -08:00
c0efd247b0 [tl2] expand firstlast api and L1WB bugfix 2016-11-14 12:12:31 -08:00
b7730d66f2 WIP bugfixes: run until corrupted WB data (beats repeated) 2016-11-11 18:34:48 -08:00
afa1a6d549 WIP uncore and rocket changes compile 2016-11-10 15:57:29 -08:00
9d77e34bee tilelink2 Filter: make transfer cap robust against large filters 2016-11-04 13:35:36 -07:00
b8df59f43b tilelink2 Broadcast: support "bufferless" implementation 2016-11-04 13:35:36 -07:00
14800f8fb4 tilelink2 Broadcast: only support caching readable devices 2016-11-04 13:35:36 -07:00
0f3947bb86 tilelink2 Broadcast: add special case handling for 0 cached clients 2016-11-03 22:18:28 -07:00
ba3c83287f tilelink2 Xbar: merge the AddressSets of fractured managers 2016-11-03 22:18:28 -07:00
55326c29bb tilelink2: Filter adapter removes some of the address space 2016-11-03 22:18:23 -07:00
86ba94781b tilelink2: broadcast coherence manager 2016-11-03 14:37:19 -07:00
d067e87a7d tilelink2 Parameters: sinkId is per port, not per manager 2016-11-03 14:37:17 -07:00
ed4224dde4 tilelink2 AtomicAutomata: fix AccessAck on same cycle as PutFull
If we send out the PutFull portion of an AMO, the slave is allowed
to respond with AccessAck on the same cycle. In this case, we are
still in the AMO state, but must still match the D response.
2016-10-31 15:17:10 -07:00
ba529c3716 rocketchip: use TileLink2 interrupts 2016-10-31 11:42:47 -07:00
043ed48c8c tilelink2 HintHandler: delay answers to help TL1 legacy clients 2016-10-31 11:42:47 -07:00
015c3b862a diplomacy: print out bus widths on edges in agent graph 2016-10-31 11:42:47 -07:00
545154c1c3 groundtest: make it happy with TL2 addressing 2016-10-31 11:42:47 -07:00
9a26cb7ec7 Debug: mark the debug device executable 2016-10-31 11:42:47 -07:00
11121b6f4c rocket: convert scratchpad to TL2 2016-10-31 11:42:13 -07:00
650f6fb23f diplomacy: add BlindNodes for use as external ports 2016-10-31 11:41:18 -07:00
0edcd3304a diplomacy Nodes: leave flipping to the MixedNode implementation 2016-10-31 11:41:18 -07:00
e45b41b4b6 Don't rely on SeqMem output after read-enable is low 2016-10-27 23:44:10 -07:00
fee67c4abf diplomacy: add methods to find {out,in}ner-most common node 2016-10-25 15:58:09 -07:00
4d50733548 tilelink2 ToAXI4: use helper method for a_last (#418) 2016-10-25 10:16:42 -07:00
4c815f7958 tilelink2 Parameters: fix {contains,supports}Safe (#416)
When there is only one manager, you still want to know if the address
was wrong on the link to that manager!
2016-10-24 20:37:04 -07:00
72e5a97d40 tilelink2: factor out the OH1ToOH function 2016-10-16 22:04:01 -07:00
20288729b9 tilelink2 Isolation: cross the valid signals as well
Refactor the code to be less copy-pasty
2016-10-14 18:28:36 -07:00
ac0bb841da AsyncQueue: cope with far reset propagation delay 2016-10-14 18:05:35 -07:00
8f3c2ddfc3 tilelink2 Crossing: these asserts should be done by the AsyncQueue 2016-10-14 16:54:09 -07:00
a82cfb8306 tilelink2: replace addr_hi with address (#397)
When faced with ambiguous routing of wmask=0, we decided to include
all the address bits. Hopefully in most cases the low bits will be
optimized away anyway.
2016-10-14 14:09:39 -07:00
54b73aef57 tilelink2: WidthWidget and Fragmenter no longer erase latency 2016-10-13 17:02:18 -07:00
200cf3dd13 tilelink2 Nodes: include some options to test for conformance 2016-10-13 17:02:18 -07:00
5d5b5a66f4 tilelink2 RAMModel: fix a write-bad-data bug 2016-10-13 17:02:17 -07:00
e5a1483358 tilelink2 Fragmenter: eliminate most of the registers on A 2016-10-13 17:02:17 -07:00
99c7003d11 tilelink2: allow preemption of Fragmenter and WidthWidget 2016-10-13 17:02:17 -07:00
b42cfdc9dd tilelink2 Arbiter: there is only one winner 2016-10-13 17:02:17 -07:00
b6e9b0c558 tilelink2 Arbiter: allow preemption of first beat 2016-10-13 17:02:17 -07:00
0aebf9e341 tilelink2 ToAXI4: no arbitration path register needed 2016-10-13 17:02:17 -07:00
0e897b905f tilelink2 RegisterRouter: data path register is no longer required 2016-10-13 17:02:17 -07:00
c4eadd3ab3 tilelink2 Monitor: enforce stricter transaction ordering 2016-10-13 17:02:17 -07:00