1
0
Commit Graph

393 Commits

Author SHA1 Message Date
Andrew Waterman
81ff127dc3 Clean up TileLinkRecursiveInterconnect a bit 2016-04-27 14:53:11 -07:00
Andrew Waterman
87cecc336f Add new RTC as TileLink slave, not AXI master 2016-04-27 11:55:35 -07:00
Andrew Waterman
eb0b5ec61e Remove stats CSR 2016-04-27 00:16:21 -07:00
Andrew Waterman
9044a4a4b7 Replace NastiROM with ROMSlave, which uses TileLink
I'm not wedded to the name.
2016-04-27 00:15:30 -07:00
Andrew Waterman
356efe2fd5 Simplify TileLink Narrower
It's not necessary to use addr_beat to determine where to put the Grant
data.  Just stripe it across all lanes.

This also gets rid of a dependence on addr_beat in Grant.  If we move
towards a regime where TileLink is only narrowed, not widened, we may
be able to drop the field altogether.
2016-04-26 16:44:54 -07:00
Wei Song
f6e44b1348 avoid logical to physical header conversion overflow 2016-04-22 17:47:34 +01:00
Howard Mao
f9de99ed40 changes to match junctions no-mmio-base 2016-04-21 15:35:37 -07:00
Howard Mao
9b3faff5a5 add id field to write data channel in TL -> AXI converter 2016-04-19 09:46:31 -07:00
Howard Mao
152645b1bc use manager_id instead of client_id in GrantFromSrc and FinishToDst 2016-04-07 11:20:16 -07:00
Howard Mao
f88b6932ce don't add pending reads if data is already available 2016-04-06 15:43:21 -07:00
Howard Mao
31e145eaf0 fix BroadcastHub allocation and routing 2016-04-05 16:21:18 -07:00
Howard Mao
f68a7dabdf fix AXI -> TL converter 2016-04-04 19:42:25 -07:00
Howard Mao
f956d4edfb NASTI does not right-justify data; fix in converter 2016-04-01 20:55:00 -07:00
Henry Cook
c292a07ace Bugfix for merged voluntary releases in L2Cache.
Track pending release beats for voluntary releases that are merged by Acquire Trackers.
Closes #23 and #24.
2016-04-01 19:57:47 -07:00
Henry Cook
82bdf3afcb Fix LRSC starvation bug by punching Finish messages out to caching clients via a new TileLinkNetworkPort. 2016-04-01 16:17:27 -07:00
Andrew Waterman
8957b5e973 Improve simulation speed of BasicCrossbar 2016-04-01 13:28:11 -07:00
Howard Mao
3083bbca21 fix TileLink arbiters and add memory interconnect and memory selector 2016-03-31 18:15:51 -07:00
Howard Mao
cf363b1fe4 add TileLink interconnect generator 2016-03-31 14:12:55 -07:00
Howard Mao
d78066db5c chisel3 fix for split metadata 2016-03-30 22:11:19 -07:00
Howard Mao
3d990bdbef workaround for Chisel3 name-aliasing issue 2016-03-30 19:15:22 -07:00
Howard Mao
8e7f18084b switch RTC to use TileLink instead of AXI 2016-03-28 12:23:16 -07:00
Howard Mao
7f8f138d6a fix addPendingBitWhenPartialWritemask 2016-03-24 20:01:50 -07:00
Howard Mao
11bd15432a fix bug in RTC 2016-03-24 20:01:50 -07:00
Howard Mao
00b3908d92 git rid of reorder queue in narrower 2016-03-24 20:01:50 -07:00
Palmer Dabbelt
c9e1b72972 Don't assign SInt(-1) to a UInt 2016-03-23 16:24:27 -07:00
Palmer Dabbelt
aa22f175c3 Add cloneType methods for Chisel3 2016-03-21 13:35:02 -07:00
Palmer Dabbelt
1344d09cef Fix the SCR file for Chisel 3 2016-03-21 11:55:18 -07:00
Henry Cook
c13b8d243d BroadcastHub race on allocating VolWBs vs Acquires 2016-03-17 18:32:35 -07:00
Henry Cook
5f3d3a0b2d Bugfix for probe flags in L2BroadcastHub
Closes #25
2016-03-17 16:42:40 -07:00
Henry Cook
49d82864bf Fix StoreDataQueue allocation bug in BroadcastHub
Closes #27
2016-03-17 12:31:18 -07:00
Eric Love
8a47c3f346 Make sure there's enough xact id bits 2016-03-16 13:49:30 -07:00
Henry Cook
67e711844a index extraction bug 2016-03-10 17:37:40 -08:00
Palmer Dabbelt
e2185d40f6 Avoid right-shift by larger that the bit width
FIRRTL bails out on this.  There's an outstanding bug, this is just a
workaround.  See https://github.com/ucb-bar/firrtl/issues/69
2016-03-10 17:37:40 -08:00
Palmer Dabbelt
8c7e29eacd Avoid generating 0-width UInts
Chisel3 requires a 1-bit width to represent UInt(0).
2016-03-10 17:37:40 -08:00
Andrew Waterman
2eafc4c8f3 Extend AMOALU to support RV32 2016-03-10 17:32:23 -08:00
Andrew Waterman
c28d115b30 Chisel3 compatibility fix 2016-03-10 17:32:23 -08:00
Henry Cook
93773a4496 Refactor L2 transaction trackers to each be capable of processing Voluntary Writebacks.
To elide several races between reading and writing the metadata array for different types of transactions, all L2XactTrackers can now sink Voluntary Releases (writebacks from the L1 in the current implementation). These writebacks are merged with the ongoing transaction, and the merging tracker supplies an acknowledgment of the writeback in addition to its ongoing activities. This change involved another refactoring of the control logic for allocating new trackers and routing incoming Acquires and Releases. BroadcastHub uses the new routing logic, but still processes all voluntary releases through the VoluntaryReleaseTracker (not a problem because there are no metadata update races).

Closes #18
Closes #20
2016-03-10 17:14:34 -08:00
Andrew Waterman
36f2e6504c Fix width of NastiROM rows, preventing out-of-range extraction 2016-03-03 16:57:16 -08:00
Henry Cook
7eef3393f1 fix bug resulting in different g_types on tail beats in L2CacheBank.io.inner.grant 2016-03-02 14:11:45 -08:00
Henry Cook
57370bdf49 first and last on HasTileLinkData 2016-03-02 14:11:39 -08:00
Palmer Dabbelt
4acdc67485 Add an assertion in the NastiIOTileLink converter
This uses an reorder queue but doesn't check to ensure that the data it fetches
from the queue is actually in the queue before using it.  It seems that during
correct operation this never breaks, but I'm trying to get the backup memory
port working again and this assertion fails with it enabled (without the
assertion the core just gets a bogus data beat dies).

Closes #16
2016-03-01 12:23:32 -08:00
Albert Magyar
ab30983aa9 Add support for per-way cache metadata
Exposes new parameter field SplitMetadata to determine whether the metadata array in a particular cache is stored in a single SeqMem or with one SeqMem per way.

Closes #14
2016-03-01 12:19:42 -08:00
Howard Mao
6d984273b7 finally fix all release assertions ... hopefully 2016-02-29 15:22:24 -08:00
John Wright
6095e7361e Move N_CORES and MMIO_BASE to SCRFile instance in RocketChip 2016-02-27 16:19:25 -08:00
Palmer Dabbelt
ebffd69b8e Provide both __OFFSET and __PADDR for SCR entries
This was recently changed to write out physical addresses for SCR file entries,
but to bring up the chip we need SCR offsets so we can write the uncore SCR
file over HTIF.  This changes the map generator to generate both.

Without this change things happened to work anyway because the high bits were
getting dropped by the SCR file.
2016-02-25 21:48:32 -08:00
John Wright
19420cd5df add utility overloads of SCRIO.attach, pass base address so that generated c header is correct, and print debug messages/header in hex instead of decimal 2016-02-24 15:26:55 -08:00
Howard Mao
8873222e42 fix cache release assertion 2016-02-23 16:03:51 -08:00
Henry Cook
929d8e31f7 refactor ready/valid logic for routing release messages in the l2 2016-02-19 16:30:26 -08:00
Howard Mao
f97bd70df5 add NastiIO to HostIO converter 2016-02-19 11:21:23 -08:00
Palmer Dabbelt
1ac9f59b31 Allow SCR files to be enumerated in C headers
Right now there's no way to ensure that SCR addresses don't conflict within
RocketChip.  Since upstream only has one of them this isn't a big deal, but we
want to add a whole bunch more to control all the IP on Hurricane.

This patch adds some Scala code to allocate registers inside the SCR file,
ensure they don't conflict, to provide names for SCRs, attach registers to the
SCR file, and generate a C header file that contains the addresses of every SCR
on a chip.

With this patch we'll be able to get rid of that constant in the testbench.
This also allows us to kill one of the Raven diffs, which is does pretty much
the same thing (just in a second SCR file, and hacked in).
2016-02-17 14:21:12 -08:00
Howard Mao
53ad8387cc add NASTI to TL converter 2016-02-10 11:06:52 -08:00
Howard Mao
2825b2d645 make sure TL to NASTI converter handles MT_WU 2016-02-10 11:06:41 -08:00
Howard Mao
66e9cc8c82 make sure CSR width is parameterizable 2016-02-02 12:49:58 -08:00
Howard Mao
adaec18bec add TL manager for MMIO requests 2016-02-02 12:49:58 -08:00
Howard Mao
c1fe188c81 some fixes to RTC 2016-02-02 12:49:58 -08:00
Howard Mao
ba94010928 DMA requests should go through MMIO 2016-02-02 12:49:58 -08:00
Howard Mao
0dc8cd5b11 move ReorderQueue and DecoupledHelper to junctions 2016-01-21 15:36:22 -08:00
Andrew Waterman
2946bc928e Avoid muxing between bundles of different size 2016-01-16 19:01:24 -08:00
Howard Mao
4ff1aea288 fix more Chisel3 deprecations 2016-01-14 14:55:45 -08:00
Andrew Waterman
0b90b8fe5f Avoid zero-width wire case :-/ 2016-01-12 15:32:29 -08:00
Andrew Waterman
a953ff384a Chisel3 compatibility: use more concrete types 2016-01-12 15:32:14 -08:00
Howard Mao
c81745eb8e lowercase SMI to Smi 2016-01-11 16:18:44 -08:00
Howard Mao
d0a14c6de9 separate TileLink converter/wrapper/unwrapper/narrower into separate file 2016-01-11 16:14:56 -08:00
Howard Mao
46069ea13b implement streaming DMA functionality 2016-01-06 21:37:56 -08:00
Howard Mao
872b162e1b implement DMA engine 2015-12-16 21:27:31 -08:00
Howard Mao
8a61177224 generalize TwoWayCounter 2015-12-16 21:07:30 -08:00
Howard Mao
a48237f36d get rid of the rest of the PutBlock special casing in L2 2015-12-16 20:56:29 -08:00
Albert Magyar
922b1adc9c Add optional PLRU replacement to the L2 2015-12-16 10:00:56 -08:00
Howard Mao
ddc79674f9 fix some issues with cache request merging 2015-12-15 21:31:02 -08:00
Howard Mao
e71293e2ae fix bug in narrower logic 2015-12-06 02:58:12 -08:00
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