Sagar Karandikar
93aa370b87
yunsup's fix for dgemm-opt assertion failure
2015-12-03 14:03:10 -08:00
Andrew Waterman
e52685f2e9
Fix LoadGen zero flag
2015-11-25 20:52:30 -08:00
Andrew Waterman
27df04354f
Add ROM with NASTI interface
2015-11-25 20:04:31 -08:00
Andrew Waterman
57e82442a1
Make LoadGen and StoreGen generic
2015-11-24 18:12:42 -08:00
Howard Mao
ee6514e4f4
make sure WritebackUnit sends correct probe addresses
2015-11-21 15:55:11 -08:00
Howard Mao
04383a31f5
Revert "make sure L2MetadataArray assigns unoccupied way if available"
...
This reverts commit 1857f36c1e6f2b2859c724eea6ae3cfb2618f81b.
2015-11-21 10:35:40 -08:00
Howard Mao
3c95afebc6
Shift set index for multi-bank configurations
...
Prior to this commit, the L2 cache banks used the lower bits of the
block address as the set index. However, the lower bits are also used to
route addresses to different banks. As a result, in multi-bank
configurations, only a fraction of the sets in each bank could be
accessed. This commit fixes that problem by using the bits ahead of the
bank index as the set index, so that all sets in the cache can be
accessed.
2015-11-20 23:24:57 -08:00
Howard Mao
55a85cc67a
make sure wmask is passed for PutBlock in broadcast hub
2015-11-20 14:09:24 -08:00
Howard Mao
941b64cd62
make partial write-masking PutBlock constructor always set alloc bit
2015-11-20 13:34:07 -08:00
Howard Mao
24f7b9f472
make sure L2MetadataArray assigns unoccupied way if available
2015-11-19 10:45:54 -08:00
Howard Mao
e50c7ad306
add NASTI error assertions back in
2015-11-18 17:05:54 -08:00
Henry Cook
2b977325e3
Make prefetch type available in a_type, issue probeInvalidates for putPrefetches
2015-11-16 23:26:13 -08:00
Andrew Waterman
d426ecee78
Remove IPI network
...
This is now provided via MMIO.
2015-11-16 21:52:24 -08:00
Henry Cook
0290635454
amo_shift_bits -> amo_shift_bytes
2015-11-16 19:07:58 -08:00
Henry Cook
64aaf71b06
L2AcquireTracker refactor to support merging Gets and Puts into Prefetches of the correct type.
...
Transaction metadata for primary and seconday misses now stored in the secondary miss queue.
Added BuiltInAcquireBuilder factory.
2015-11-16 18:10:09 -08:00
Henry Cook
03fa06e6e7
fix prefetch lockup on L2 hit
2015-11-15 12:51:34 -08:00
Howard Mao
e12efab423
skip meta_write state if no meta write pending
2015-11-13 13:50:35 -08:00
Howard Mao
7e7d688a01
make sure L2 passes no-alloc acquires through to outer memory
2015-11-12 15:40:58 -08:00
Howard Mao
b3865c370a
make sure correct addr_beat is sent for Get response by narrower/converter
2015-11-12 15:40:38 -08:00
Howard Mao
f397d61033
add alloc option to Put constructor
2015-11-12 11:39:59 -08:00
Howard Mao
7733fbe6a3
make sure no-alloc write still updates data array if there is a cache hit
2015-11-12 11:39:36 -08:00
Howard Mao
b59ce5fed4
make sure L2 waits for outer grant before sending grant for write request
2015-11-10 16:06:14 -08:00
Howard Mao
42d3d09d7a
add a ClientTileLinkEnqueuer to complement the TileLinkEnqueuer
2015-11-09 11:49:19 -08:00
Howard Mao
7942be4e01
make sure outerTL method is idempotent
2015-11-09 11:10:02 -08:00
Henry Cook
e3efc09b5b
remove unnecessary UInt encode/decode on releaseMatches path
2015-11-05 17:20:03 -08:00
Henry Cook
3698153535
OHToUInt instead of PriorityEncoder on Acq/RelMatches signals in L2Bank
2015-11-03 14:31:35 -08:00
Howard Mao
baa2544651
Fix some more issues with narrower
2015-10-31 19:36:30 -07:00
Howard Mao
812c5bcc55
make sure narrower can handle sub-block level requests correctly
2015-10-31 15:58:36 -07:00
Howard Mao
d4b8653002
fix too strict assertion in broadcast hub
2015-10-31 15:58:10 -07:00
Howard Mao
c10870a87c
make sure ID width requirement in TL -> NASTI converter is correct
2015-10-27 13:25:29 -07:00
Howard Mao
9fa4541916
get rid of unused full signal in ReorderQueue
2015-10-26 12:17:25 -07:00
Howard Mao
6403f27fbe
fix bug in ReorderQueue breaking TileLink Unwrapper
2015-10-22 15:52:55 -07:00
Henry Cook
f8594da1d3
depend on external cde library
2015-10-21 18:17:17 -07:00
Howard Mao
02d113b39f
outerDataBits / innerDataBits should be per beat, not per block
2015-10-21 11:31:13 -07:00
Howard Mao
baf95533a4
fix combinational loop in TileLink Unwrapper
2015-10-20 23:26:11 -07:00
Howard Mao
ffe7df2fed
make sure TL -> NASTI converter acquire ready not dependent on valid
2015-10-20 22:09:22 -07:00
Howard Mao
1c135c1628
fix ready-valid mixup in TileLink unwrapper
2015-10-20 21:07:42 -07:00
Henry Cook
4389b9edb0
tilelink parameter tweak: addrBits now a constant
2015-10-20 15:00:30 -07:00
Howard Mao
d12403e7dc
fix up and simplify TL -> NASTI converter logic
2015-10-19 13:47:13 -07:00
Henry Cook
d391f97953
Minor refactor of StoreGen/AMOALU. Bugfix for 32b ops in L2's AMOALU.
2015-10-16 19:11:06 -07:00
Henry Cook
e1f573918d
simplify TileLinkParameters with Option
2015-10-16 18:24:38 -07:00
Howard Mao
49667aa4b0
make sure broadcast acquire tracker doesn't try to send requests back-to-back
2015-10-14 18:56:13 -07:00
Howard Mao
1d362d6d3a
make sure correct parameters are used for TileLink constructors
2015-10-14 17:58:54 -07:00
Henry Cook
7fa3eb95e3
refactor tilelink params
2015-10-14 12:13:37 -07:00
Henry Cook
66ea39638e
GlobalAddrMap
2015-10-14 00:23:28 -07:00
Henry Cook
31be6407ec
Removed all traces of params
2015-10-14 00:23:28 -07:00
Henry Cook
908922c1a4
refactor NASTI to not use param
2015-10-14 00:23:28 -07:00
Howard Mao
47da284e56
TileLinkNarrower should do nothing if interfaces are the same width
2015-10-13 13:28:47 -07:00
Howard Mao
83df05cb6a
add TileLink data narrower
2015-10-13 12:45:39 -07:00
Howard Mao
993ed86198
move ReorderQueue to utils.scala
2015-10-13 09:49:22 -07:00
Andrew Waterman
0fe16ac1c0
Chisel3 compatibility fixes
2015-09-30 14:37:00 -07:00
Howard Mao
1e7f656527
get release block address from inner release
2015-09-28 15:02:51 -07:00
Andrew Waterman
3b1da4c57e
Revert "replace remaining uses of Vec.fill"
...
This reverts commit b6bb4e42127d1ed42b55ec8b859a4e074b347d47.
2015-09-25 17:06:57 -07:00
Andrew Waterman
20b7a82ab6
Use Vec.fill, not Vec.apply, when making Vec literals
2015-09-25 17:06:52 -07:00
Andrew Waterman
2179cb64ae
Let isRead be true for store-conditional
...
This works around a deadlock bug in the L1 D$, and is arguably true.
2015-09-25 15:28:02 -07:00
Howard Mao
308022210a
use updated NASTI channel constructors
2015-09-25 12:07:27 -07:00
Howard Mao
8c4ac0f4f3
make sure CSR/SCR data width matches xLen
2015-09-25 12:07:03 -07:00
Howard Mao
d1f2d40a90
replace remaining uses of Vec.fill
2015-09-24 17:50:09 -07:00
Howard Mao
3ff830e118
ReorderQueue uses Vec of Bools instead of Bits for roq_free
2015-09-24 17:43:53 -07:00
Howard Mao
83740dfaa5
Merge branch 'master' of github.com:ucb-bar/uncore
2015-09-24 17:10:09 -07:00
Howard Mao
3b86790c3f
replace NASTIMasterIO and NASTISlaveIO with NASTIIO
2015-09-24 16:58:20 -07:00
ducky
ee6754daca
Fix clone -> cloneType
2015-09-24 16:18:25 -07:00
Howard Mao
b4d21148ec
get rid of NASTI error assertion
2015-09-22 09:43:42 -07:00
Howard Mao
8b2341b1b1
use reorder queue instead of extra tag bit to determine TL g_type in NASTI -> TL converter
2015-09-18 09:41:37 -07:00
Howard Mao
bd536d8832
make HTIFModuleIO an anonymous bundle
2015-09-14 12:58:44 -07:00
Howard Mao
9d89d2a558
get rid of MemIO -> TileLink converters
2015-09-14 12:58:44 -07:00
Howard Mao
f9965648f2
fix up some things in tilelink.scala
2015-09-14 12:57:54 -07:00
Howard Mao
64717706a9
get rid of non-NASTI RTC module
2015-09-14 12:57:54 -07:00
Howard Mao
6ee6ea4f1e
use Put/Get/PutBlock/GetBlock constructors in broadcast hub
2015-09-14 12:57:54 -07:00
Howard Mao
ae3d96013a
make TL -> NASTI converter ingest ClientUncachedTileLinkIO and move functionality to Unwrapper
2015-09-14 12:57:54 -07:00
Howard Mao
21f96f382c
split off SCR functionality from HTIF
2015-09-14 12:57:54 -07:00
Howard Mao
bdc6972a8d
separate RTC updates from HTIF
2015-09-14 12:56:44 -07:00
Howard Mao
24f3fac90a
fix broadcast hub and TL -> NASTI converter to support subblock operations
2015-09-14 12:56:44 -07:00
Andrew Waterman
24389a5257
Chisel3 compatibility fixes
2015-09-11 15:41:39 -07:00
Andrew Waterman
350d530766
Use Vec.fill, not Vec.apply, for Vec literals
2015-08-27 10:00:43 -07:00
Andrew Waterman
94287fed90
Avoid type-unsafe assignments
2015-08-27 09:57:36 -07:00
Andrew Waterman
05d311c517
Use Vec.apply, not Vec.fill, for type nodes
2015-08-27 09:47:02 -07:00
Henry Cook
005752e2a6
use the parameters used to create the original object
2015-08-10 14:43:17 -07:00
Andrew Waterman
01fc61ba96
Don't construct so many Vecs
2015-08-05 18:43:59 -07:00
Howard Mao
a551a12d70
add missing Wire wrap in BasicCrossbar
2015-08-05 17:05:31 -07:00
Andrew Waterman
eb6583d607
use cloneType in PhysicalNetworkIO
2015-08-05 16:47:49 -07:00
Andrew Waterman
798ddeb5f5
Chisel3 compatibility: use >>Int instead of >>UInt
...
The latter doesn't contract widths anymore.
2015-08-04 13:15:17 -07:00
Andrew Waterman
77cf26aeba
Chisel3: Flip order of := and <>
2015-08-03 18:53:39 -07:00
Andrew Waterman
121e4fb511
Flip direction of some bulk connects
2015-08-03 18:01:14 -07:00
Andrew Waterman
a21979a2fa
Bits -> UInt
2015-08-03 18:01:06 -07:00
Andrew Waterman
9c7a41e8d3
Chisel3: bulk connect is not commutative
...
We haven't decided if this is a FIRRTL limitation that we should relax,
or a backwards incompatibility we're forced to live with. Should make
for lively debate.
2015-08-01 21:09:00 -07:00
Andrew Waterman
6fc807f069
Chisel3: Avoid subword assignment
2015-08-01 21:08:35 -07:00
Andrew Waterman
6d574f8c1b
Fix incompatible assignment
2015-07-31 00:59:34 -07:00
Andrew Waterman
377e17e811
Add Wire() wrap
2015-07-31 00:32:02 -07:00
Andrew Waterman
0686bdbe28
Avoid cross-module references
...
You can't instantiate a Vec in one module and use it in another.
An idiosyncrasy of the Chisel2 implementation let this one slip by.
In this case, it's just a matter of using def instead of val.
2015-07-30 23:49:06 -07:00
Andrew Waterman
8f7b390353
UInt-> Bits; avoid mixed UInt/SInt code
2015-07-30 23:49:06 -07:00
Andrew Waterman
6c391e3b37
Use UInt(0), not UInt(width=0), for constant 0
2015-07-30 23:49:06 -07:00
Henry Cook
c70b495f6d
moved buses to junctions repo
2015-07-29 18:04:30 -07:00
Andrew Waterman
a69c749249
Fix compilation with scala 2.11.6
...
We forgot to specify return types on overloaded methods, and a previous
version of the scala compiler failed to flag this as an error.
2015-07-28 16:24:45 -07:00
Andrew Waterman
f8ec6d6393
Chisel3 compatibility: use BitPat for don't-cares
...
Also, call the UInt factory instead of the Bits one, for good measure.
2015-07-28 02:46:23 -07:00
Andrew Waterman
0e06c941df
Chisel3 compatibility fixes
2015-07-23 14:58:46 -07:00
Andrew Waterman
3c0475e08b
Add Wire() wrap
2015-07-15 20:24:03 -07:00
Andrew Waterman
2d6b3b2331
Don't use clone
2015-07-15 18:06:27 -07:00
Andrew Waterman
276f53b652
Delete BigMem; it's not used anymore
2015-07-15 17:41:47 -07:00
Andrew Waterman
15cec0eab7
Vec(Reg) -> Reg(Vec)
2015-07-15 12:44:54 -07:00