1
0
Commit Graph

4286 Commits

Author SHA1 Message Date
Andrew Waterman
2888779422 Flush pipeline from WB stage, not MEM
Fixes sptbr write -> instruction translation hazard.
2017-03-24 16:39:52 -07:00
Andrew Waterman
44ca3b60ab Retime PTW response valid bits
It's not just to save the gate delay; it also reduces wire delay by
allowing the flops to be closer to their respective TLBs.
2017-03-24 16:39:52 -07:00
Andrew Waterman
a03556220c Default TLB size = 32
@davidbiancolin
2017-03-24 16:39:52 -07:00
Andrew Waterman
1875407316 Get TLB permission checks off D$ clock gating critical path 2017-03-24 16:39:52 -07:00
Andrew Waterman
a4164348b4 Expose MXR to S-mode 2017-03-24 16:39:52 -07:00
Andrew Waterman
0380aed329 PUM -> SUM 2017-03-24 16:39:52 -07:00
Andrew Waterman
2a413e4496 Remove fruitless debug() 2017-03-24 16:39:52 -07:00
Andrew Waterman
29414f3a23 Simplify interrupt-stack discipline
f2ed45b179
2017-03-24 16:39:52 -07:00
Andrew Waterman
723352c3e2 Mitigate some more PMP critical paths 2017-03-24 16:39:52 -07:00
Andrew Waterman
7484f27ed3 Don't gate exception-cause pipeline registers separately
They are too narrow to justify gating separately from the other pipeline
registers (and one of the clock gates was on the PMP critical path).
2017-03-24 16:39:52 -07:00
Andrew Waterman
3ea822c2cf Make blocking L1 D$ the default
The nonblocking cache is overdesigned for most Rocket-class cores, so
the blocking cache is the more appropriate default.
2017-03-24 16:39:52 -07:00
Andrew Waterman
487b8db5ef Address some PMP critical paths 2017-03-24 16:39:52 -07:00
Andrew Waterman
03fb334c4c Take mprv calculation off critical path 2017-03-24 16:39:52 -07:00
Andrew Waterman
f0796f0509 Pass correct access size information to PMP checker 2017-03-24 16:39:52 -07:00
Andrew Waterman
a6874c03f7 Remove DecoupledTLB 2017-03-24 16:39:52 -07:00
Andrew Waterman
78f9f6b9ef When SFENCE.VMA has rs2 != x0, don't flush global mappings 2017-03-24 16:39:52 -07:00
Andrew Waterman
1b950128e1 PTW should always use S-mode privilege
If an exception occurs while a page-table walk is coincidentally in
progress (e.g., an illegal instruction executes during data TLB refill),
then the processor might enter M-mode.  However, the PTW's accesses
should proceed without M privilege, to avoid bypassing PMPs.

Note, the same argument doesn't apply to the nonblocking cache's replay
queues, because those accesses have already been checked against the PMPs.
The cache correctly ignores access exceptions reported on replays,
provided no exceptions were reported on the initial access.
2017-03-24 16:39:52 -07:00
Andrew Waterman
aace526857 WIP on PMP 2017-03-24 16:39:52 -07:00
Andrew Waterman
b1b405404d Set PRV=M when entering debug mode
Debug mode mostly behaves like M-mode, so this approach avoids having
to check the debug bit in most permission checks.
2017-03-24 16:39:52 -07:00
Andrew Waterman
cf168e419b Support SFENCE.VMA rs1 argument
This one's a little invasive.  To flush a specific entry from the TLB, you
need to reuse its CAM port.  Since the TLB lookup can be on the critical
path, we wish to avoid muxing in another address.

This is simple on the data side, where the datapath already carries rs1 to
the TLB (it's the same path as the AMO address calculation).  It's trickier
for the I$, where the TLB lookup address comes from the fetch stage PC.
The trick is to temporarily redirect the PC to rs1, then redirect the PC
again to the instruction after SFENCE.VMA.
2017-03-24 16:39:52 -07:00
Henry Cook
797c18b8db Make some requirement failures more verbose (#608)
* tilelink: verbose requires in xbar

* diplomacy: verbose requires
2017-03-23 21:55:11 -07:00
Wesley W. Terpstra
bd08f10816 tilelink2: make sink ids optional (#607)
* tilelink2: make sink ids optional

* CacheCork: add a special-case for 1 sink id
2017-03-23 18:19:04 -07:00
Wesley W. Terpstra
19eb9b6906 l1tol2: put a flow Q on the exits (#606)
This Xbar connects the largest components in the design; the cores
and the L2 banks. We already have a full buffer on the core side.
However, the valid path going to the L2 comes back as a ready path.
Putting a flow Q also on the outputs of the l1tol2 cuts this path
in half at no cost to IPC.
2017-03-23 16:28:32 -07:00
Henry Cook
055b8ba1f0 rocket: avoid LinkedHashMap.keys to preserve traversal order (#603) 2017-03-22 14:38:33 -07:00
Wesley W. Terpstra
4f78eafbdf Merge pull request #602 from ucb-bar/tl-mmio-pipeline
TL pipeline MMIO
2017-03-21 14:59:11 -07:00
Andrew Waterman
76f083b469 FIFOFixer: Not all D-channel messages are A-channel responses 2017-03-21 14:17:38 -07:00
Andrew Waterman
3609254e4a There's no structural hazard on MMIO store responses
So don't stall as though there were.
2017-03-21 14:17:32 -07:00
Yunsup Lee
5eae7e1da4 make DCache s1_nack less conservative for pipelined MMIO requests 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
4c00066746 rocket: describe dcache as two clients (fifo+cached) 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
81d717e82f coreplex: guarantee FIFO for those tiles that need it 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
198afddb4b tilelink2: add the FIFOFixer 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
c33f31dd3c tilelink2 RAMModel: weaken fifo requirement check 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
930438adba tilelink2 SourceShrinker: destroy FIFO behaviour 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
fd521c56a6 tilelink2: add client-side FIFO parameterization 2017-03-21 11:16:51 -07:00
Wesley W. Terpstra
d4c9c13fb4 Merge pull request #600 from ucb-bar/monitor-spec
Update monitor spec
2017-03-20 15:23:34 -07:00
Wesley W. Terpstra
4eef317e84 RegisterRouter: support devices with gaps 2017-03-20 14:49:22 -07:00
Wesley W. Terpstra
431cb41e27 tilelink2 Parameters: clarify client minLatency is B=>C, not D=>E 2017-03-20 14:49:22 -07:00
Wesley W. Terpstra
04892fea01 Monitor: support early ack 2017-03-20 14:49:19 -07:00
Wesley W. Terpstra
278f6fea24 tilelink2: define is{Request,Response} based on spec 2017-03-20 13:41:02 -07:00
Wesley W. Terpstra
778e189bba Monitor: ProbeAckData and ReleaseData may carry an error 2017-03-20 11:44:13 -07:00
Wesley W. Terpstra
48c7aed4e1 Monitor: any probe supported by the client is legal 2017-03-20 11:34:19 -07:00
Wesley W. Terpstra
5a50acfd9d Merge pull request #595 from ucb-bar/ignore-tl-c
Ignore TL-C
2017-03-19 18:49:11 -07:00
Wesley W. Terpstra
0c92283a61 rocket icache: tie off b ready 2017-03-19 17:18:50 -07:00
Wesley W. Terpstra
c9459fe4eb tilelink2 Xbar: don't use unnecessary ports 2017-03-19 17:02:24 -07:00
Wesley W. Terpstra
7971947d6c tilelink2 Monitor: don't inspect bits if valid is forbidden 2017-03-19 16:34:23 -07:00
Wesley W. Terpstra
a4ca424a22 AHBToTL: finally get the error signal right? (#594) 2017-03-18 22:24:20 -07:00
Wesley W. Terpstra
d4272db067 travis: only run 4 jobs at once (#593)
We can only run 4 at a time; 5 causes the test time to double.
In the past we had a 50minute build deadline, but that's fixed.
2017-03-18 04:14:50 -07:00
Wesley W. Terpstra
f6daa782d3 AHBToTL: fix the order of updates to d_pause (#592) 2017-03-17 19:34:40 -07:00
Megan Wachs
dcc9827ab4 Rename Prci.scala to Clint.scala (#591)
The internals of this were renamed to CoreplexLocalInterrupter, so changing the top level name to match.
2017-03-17 15:36:10 -07:00
Wesley W. Terpstra
db55a1d755 Fragmenter: fix a bug when underlying device supports larger bursts (#589) 2017-03-17 11:00:49 -07:00