Eddie Hung
c0bb1d22e8
Remove newline
2019-09-26 10:31:55 -07:00
Eddie Hung
781dda6175
select once
2019-09-26 10:15:05 -07:00
Eddie Hung
27e5bf5aad
Stop trying to be too smart by prematurely optimising
2019-09-26 09:57:11 -07:00
Eddie Hung
35aaa8d73a
mul2dsp.v slice names
2019-09-25 22:58:55 -07:00
Eddie Hung
f1de93edf5
Do not die if DSP48E1.P has no users (would otherwise get 'clean'-ed)
2019-09-25 22:58:03 -07:00
Eddie Hung
cd8a640989
Reject if (* init *) present
2019-09-25 18:21:08 -07:00
Eddie Hung
34aa3532fb
Remove unnecessary check for A_SIGNED != B_SIGNED; be more explicit
2019-09-25 17:26:47 -07:00
Eddie Hung
a4238637ac
Revert "Remove _TECHMAP_CELLTYPE_ check since all $mul"
...
This reverts commit 234738b103
2019-09-25 17:25:44 -07:00
Eddie Hung
f4387e817c
Revert "No need for $__mul anymore?"
...
This reverts commit 1d875ac76a
2019-09-25 17:24:11 -07:00
Eddie Hung
aeb1539818
Rework xilinx_dsp postAdd for new wreduce call
2019-09-25 17:22:30 -07:00
Eddie Hung
63940913d2
Only wreduce on t:$add
2019-09-25 17:22:04 -07:00
Eddie Hung
234738b103
Remove _TECHMAP_CELLTYPE_ check since all $mul
2019-09-25 16:51:31 -07:00
Eddie Hung
5f8917c984
Fix memory issue since SigSpec& could be invalidated
2019-09-25 16:45:51 -07:00
Eddie Hung
a009314597
Merge pull request #1401 from SergeyDegtyar/SergeyDegtyar/ice40
...
ICE40 tests. adffs test update (equiv_opt -multiclock).
2019-09-25 16:43:24 -07:00
Eddie Hung
1d875ac76a
No need for $__mul anymore?
2019-09-25 14:06:21 -07:00
Eddie Hung
486dd7c483
unextend only used in init
2019-09-25 14:05:59 -07:00
Eddie Hung
53ea5daa42
Call 'wreduce' after mul2dsp to avoid unextend()
2019-09-25 14:04:36 -07:00
Eddie Hung
93363c94a2
Oops. Actually use __NAME__ in ABC_DSP48E1 macro
2019-09-25 10:33:16 -07:00
SergeyDegtyar
b66364ada2
Change sync controls to async.
2019-09-25 14:43:26 +03:00
Clifford Wolf
739c621330
Merge pull request #1402 from YosysHQ/clifford/portlist
...
Add "portlist" command
2019-09-25 09:20:54 +02:00
Clifford Wolf
b432c9b44b
Improve "portlist" command
...
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2019-09-25 09:20:38 +02:00
Clifford Wolf
6c427d36dd
Add "portlist" command
...
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2019-09-24 18:08:59 +02:00
SergeyDegtyar
fc6ebf8268
adffs test update (equiv_opt -multiclock).
2019-09-24 14:55:32 +03:00
Eddie Hung
b41d2fb4e4
Add (* techmap_autopurge *) to abc_unmap.v too
2019-09-23 22:02:22 -07:00
Eddie Hung
44374b1b2b
"abc_padding" attr for blackbox outputs that were padded, remove them later
2019-09-23 21:58:40 -07:00
Eddie Hung
c340fbfab2
Force $inout.out ports to begin with '$' to indicate internal
2019-09-23 21:58:04 -07:00
Eddie Hung
11ac37733d
Add techmap_autopurge to outputs in abc_map.v too
2019-09-23 21:56:28 -07:00
Eddie Hung
27167848f4
Revert "Add a xilinx_finalise pass"
...
This reverts commit 23d90e0439
2019-09-23 19:52:55 -07:00
Eddie Hung
0f53893104
Revert "Remove (* techmap_autopurge *) from abc_unmap.v since no effect"
...
This reverts commit 67c2db3486
2019-09-23 19:52:55 -07:00
Eddie Hung
29db96fa1f
Revert "Vivado does not like zero width port connections"
...
This reverts commit 895e2befa7
2019-09-23 19:52:54 -07:00
Eddie Hung
895e2befa7
Vivado does not like zero width port connections
2019-09-23 19:04:07 -07:00
Eddie Hung
67c2db3486
Remove (* techmap_autopurge *) from abc_unmap.v since no effect
2019-09-23 18:56:18 -07:00
Eddie Hung
23d90e0439
Add a xilinx_finalise pass
2019-09-23 18:56:02 -07:00
Eddie Hung
e556d48d45
Set [AB]CASCREG to legal values
2019-09-23 16:00:11 -07:00
Eddie Hung
b824a56cde
Comment to explain separating CREG packing
2019-09-23 13:58:10 -07:00
Eddie Hung
15dfbc8125
Separate out CREG packing into new pattern, to avoid conflict with PREG
2019-09-23 13:27:10 -07:00
Eddie Hung
26a6c55665
Move log_debug("\n") later
2019-09-23 13:27:00 -07:00
Eddie Hung
d0dbbc2605
Move unextend initialisation later
2019-09-23 13:26:34 -07:00
Eddie Hung
a67af3d5e5
Use new port() overload once more
2019-09-23 13:00:44 -07:00
Miodrag Milanović
057dae4f78
Merge pull request #1399 from nakengelhardt/fix-show-macos
...
fix show command for macos
2019-09-23 20:06:40 +02:00
Eddie Hung
bcee87a457
Merge remote-tracking branch 'origin/master' into xc7dsp
2019-09-23 10:58:28 -07:00
N. Engelhardt
2b81ce5648
add xdot dependency to Brewfile
2019-09-23 18:25:04 +02:00
N. Engelhardt
3bed4cb18a
fix show command for macos
2019-09-23 17:47:05 +02:00
SergeyDegtyar
1070f2e90b
Add new tests for Efinix architecture.
...
Problems/questions:
- fsm.ys. equiv_opt -assert failed because of unproven cells;
- latches.ys,tribuf.ys - internal cells present;
- memory.ys - sat called with -verify and proof did fail.
2019-09-23 15:51:41 +03:00
Benedikt Tutzer
f39269805d
Generate Python wrappers for inline constructors
...
Fixes : #1353
2019-09-23 13:17:59 +02:00
SergeyDegtyar
27377c4663
Add new tests for Anlogic architecture
...
Problems/questions:
- memory.ys: ERROR: Failed to import cell gate.mem.0.0.0 (type
EG_LOGIC_DRAM16X4) to SAT database.
Why EG_LOGIC_DRAM16X4, not AL_LOGIC_BRAM?
- Internal cell type $_TBUF_ is present.
2019-09-23 12:12:02 +03:00
whitequark
4f426c2ac4
write_verilog: do not print (*init*) attributes on regs.
...
If an init value is emitted for a reg, an (*init*) attribute is never
necessary, since it is exactly equivalent. On the other hand, some
tools that consume Verilog (ISE, Vivado, Quartus) complain about
(*init*) attributes because their interpretation differs from Yosys.
All (*init*) attributes that would not become reg init values anyway
are emitted as before.
2019-09-22 16:52:06 +00:00
Clifford Wolf
0a2d8db793
Merge pull request #1392 from YosysHQ/eddie/fix1391
...
(* techmap_autopurge *) fixes when ports aren't consistently-sized
2019-09-21 11:25:36 +02:00
Eddie Hung
7c8de1dd18
Hell let's add the original #1381 testcase too
2019-09-20 17:58:51 -07:00
Eddie Hung
ec08a031b5
Revert abc9.cc
2019-09-20 17:52:23 -07:00
