Activity
From 02/09/2018 to 03/10/2018
03/10/2018
- 10:50 PM Issue #1290 (Confirmed): scr1 test suite: assert properties don't work
- John, thanks for looking.
- 06:29 PM Issue #1290: scr1 test suite: assert properties don't work
- Thanks. I discovered that concurrent asserts are always clocked in verilog.
So this is a dup of 785.
- 06:22 PM Issue #1290: scr1 test suite: assert properties don't work
- You're right - the scr1 tests are all clocked. My example was a bit too simplified!
Great news that you're working... - 06:13 PM Issue #1290: scr1 test suite: assert properties don't work
- For that matter, does verilog even permit unclocked 'assert property' statements? I'm no language lawyer, and I don't...
- 05:55 PM Issue #1290: scr1 test suite: assert properties don't work
- Hi Joel,
I'm working on issue 785 which could be considered a dup:
https://www.veripool.org/issues/785-Verilato... - 04:18 AM Issue #1290 (Closed): scr1 test suite: assert properties don't work
- The following SystemVerilog code fails to build......
- 08:32 PM Issue #1288: scr1 test suite: In some cases mixed assignment to struct member fails
- Perfect- thansk!
- 07:17 PM Issue #1288: scr1 test suite: In some cases mixed assignment to struct member fails
- Good point, similar suggested comments added to git.
- 07:00 PM Issue #1288: scr1 test suite: In some cases mixed assignment to struct member fails
- Ok -- now I feel silly.
My feedback is that - expected that -Wno-XX would only affect warnings, not errors. Would ... - 05:51 PM Issue #1288: scr1 test suite: In some cases mixed assignment to struct member fails
- Use -Wno-BLKANDNBLK
- 01:43 AM Issue #1288: scr1 test suite: In some cases mixed assignment to struct member fails
- Am I being obtuse? I can't see a way to disable the error at run-time.
- 06:33 PM Issue #785: Support for SystemVerilog assertions
- UPDATE: there IS a fundamental difficulty supporting unclocked combinational concurrent asserts -- they're not allowe...
- 05:50 PM Issue #1291 (WillNotFix): scr1 test suite: delayed always blocks are unsupported
- Correct, clocks must come from the C side. At least until events are supported which is long off.
Note you can tur... - 04:16 PM Issue #1291: scr1 test suite: delayed always blocks are unsupported
- RTFM.... I see this isn't going to work
- 05:17 AM Issue #1291 (WillNotFix): scr1 test suite: delayed always blocks are unsupported
- This code generates a warning:...
- 04:21 AM Issue #1286: scr1 test suite: processing passes get stuck, and allocate huge amounts of system RA...
- The block was 16Mb in size. Fortunately nothing in the test suite needs this much memory, so I reduced it to 64kB, an...
03/09/2018
- 09:46 PM Issue #785: Support for SystemVerilog assertions
- I'll have a fix for this shortly.
Still NOT supported will be:
* "|->" syntax
* "##" syntax
* referenc... - 08:05 PM Issue #1287 (WillNotFix): scr1 test suite: SystemVerilog nested @ blocks are not supported
- Turn it into a FSM, e.g....
- 05:06 PM Issue #1287: scr1 test suite: SystemVerilog nested @ blocks are not supported
- Makes sense.
So here is the relevant code: https://github.com/syntacore/scr1/blob/master/src/tb/scr1_top_tb_ahb.sv... - 04:50 AM Issue #1287 (AskedReporter): scr1 test suite: SystemVerilog nested @ blocks are not supported
- Keeping open in case I'm misunderstanding what is needed...
Verilator is not an event based simulator and requires... - 05:20 PM Issue #1285: scr1 test suite: built-in system functions only work when the return value is assigned
- Nice! Thanks for fixing it so quickly
- 04:42 AM Issue #1285 (Resolved): scr1 test suite: built-in system functions only work when the return valu...
- Simple enough, mapped all system tasks to be allowed as functions. Thanks for the report.
Fixed in git towards 3.9... - 05:10 PM Issue #1286: scr1 test suite: processing passes get stuck, and allocate huge amounts of system RA...
- It would be nice to get an error or a warning. I only figured out the source of the problem by reading the file & lin...
- 04:45 AM Issue #1286 (Confirmed): scr1 test suite: processing passes get stuck, and allocate huge amounts ...
- Verilator needs enough memory to fit the entire model, times some inefficiency factor.
The fix could be to throw a... - 05:02 PM Issue #1289: scr1 test suite: string.hextoa is not implemented
- Sure, I don't mind having a look.
- 04:57 AM Issue #1289 (Confirmed): scr1 test suite: string.hextoa is not implemented
- 04:57 AM Issue #1289: scr1 test suite: string.hextoa is not implemented
- Perhaps you would like to attempt a patch? First make a test case (see internals.txt) which passes on another simula...
- 05:07 AM Issue #1288 (Confirmed): scr1 test suite: In some cases mixed assignment to struct member fails
- Added this to the documentation of this error:
"It is generally safe to disable this error when one of the assig... - 04:58 AM Issue #1276 (AskedReporter): vcd trace splits packed data type when it comes through a typedef
- Still awaiting a test case... Thanks
03/08/2018
- 11:43 PM Issue #1289 (Closed): scr1 test suite: string.hextoa is not implemented
- The follow code SystemVerilog code fails to build......
- 11:35 PM Issue #1288 (Confirmed): scr1 test suite: In some cases mixed assignment to struct member fails
- This SystemVerilog code......
- 11:32 PM Issue #1287 (WillNotFix): scr1 test suite: SystemVerilog nested @ blocks are not supported
- The follow SystemVerilog code fails to parse......
- 11:30 PM Issue #1286 (Confirmed): scr1 test suite: processing passes get stuck, and allocate huge amounts ...
- Verilator gets stuck allocating huge amounts of memory when processing SystemVerilog code like this......
- 11:26 PM Issue #1285 (Closed): scr1 test suite: built-in system functions only work when the return value ...
- Building this test code fails......
- 12:54 AM Issue #1284 (Resolved): Unsupported --annotate-min command line argument for verilator_coverage tool
- Thanks for the great patch. Pushed to git towards 3.922.
03/07/2018
- 11:16 PM Issue #1284 (Closed): Unsupported --annotate-min command line argument for verilator_coverage tool
- Hi,
*--annotate-min* command line argument for *verilator_coverage* tool is not supported. Missing implementation ...
03/04/2018
- 12:27 PM Issue #1282 (Assigned): False UNOPTFLAT warning when using 2 interfaces with the same name on dif...
- This hits a bug in the module inlining code. I'll take a look.
03/02/2018
- 04:38 PM Issue #1282: False UNOPTFLAT warning when using 2 interfaces with the same name on different hier...
- I accidentally uploaded the test-case after the fix. Please use the updated one.
- 04:30 PM Issue #1282 (): False UNOPTFLAT warning when using 2 interfaces with the same name on different h...
- I ran into false UNOPTFLAT warning while adding an interface to my design. After investigating it a little bit, I fou...
02/28/2018
- 12:01 PM Issue #1244 (Resolved): V3Split not splitting
- John Coiner provided an excellent patch which is merged into git develop-v4 branch, towards version 4.000.
Thanks ...
02/26/2018
- 09:26 AM Issue #1281 (Resolved): Trig functions
- Pushed patch to git (with minor space changes) towards 3.922.
Thanks again for your effort.
- 09:05 AM Issue #1278 (Confirmed): Unsupported LHS tristate construct: ARRAYSEL
- Due to how the code currently works supporting arrays will be painful, so is unlikely to be improved in the short ter...
02/25/2018
- 02:31 PM Issue #1278: Unsupported LHS tristate construct: ARRAYSEL
- You can use packed arrays instead:
inout [NumBusses-1:0][7:0]p_busses ;
02/21/2018
- 04:51 PM Issue #1281: Trig functions
- Do you mean the C++ math functions? None of the @<cmath>@ functions raise exceptions, they just return @NaN@ or @+-in...
- 04:06 PM Issue #1281: Trig functions
- Nicely done. Do any if these return exceptions with e.g. 0 inputs? If so can you please have the test check that, and...
- 04:22 AM Issue #1281: Trig functions
- Sorry, that should be: "https://github.com/patstew/verilator/commit/f3fa47a6791815619b7bf749be2be92d9cc985ac":https:/...
- 04:18 AM Issue #1281 (Closed): Trig functions
- I've added the missing system trig functions ($sin() etc) here: http://github.com/patstew/verilator/commit/ae1139d73e...
02/15/2018
- 05:58 PM Issue #1278: Unsupported LHS tristate construct: ARRAYSEL
- What I provided was a trifle example, as you probably know, I am writing software that generates RTL dynamically. Al...
- 12:18 PM Issue #1278: Unsupported LHS tristate construct: ARRAYSEL
- Any message with "Unsupported" generally means the language supports something but Verilator does not.
In this cas...
02/14/2018
- 08:12 PM Issue #1278 (Confirmed): Unsupported LHS tristate construct: ARRAYSEL
- I am wondering what this actually means, I was under the impression that this is legal SystemVerilog. I am trying to...
- 12:25 PM Issue #1276: vcd trace splits packed data type when it comes through a typedef
- I don't see this behavior. Please add to test_regress/t/t_trace_complex.v to show the problem.
- 12:17 PM Issue #1277 (WillNotFix): instantiate SystemC model in Verilog dut
- You cannot have SystemC underneath Verilog, unless you use DPI/VPI to make some sort of harness similar to what you w...
- 05:48 AM Issue #1277 (WillNotFix): instantiate SystemC model in Verilog dut
- Hello,
I'm trying to verilate a system with the following hierarchy:
sim_main.cpp as a testbench instantiating ...
02/13/2018
- 07:59 PM Issue #1276 (NotEnoughInfo): vcd trace splits packed data type when it comes through a typedef
- In our verilog code, we have something similar to the below typedef:...
- 07:50 PM Issue #1275: genvar in always_comb doesn't fail or lint warn during compile when not in a generat...
- I might look into it after I get some downtime at work. I'll reference the original issue for what you may have alre...
02/12/2018
- 03:06 AM Issue #1275 (Duplicate): genvar in always_comb doesn't fail or lint warn during compile when not ...
- This is a duplicate of bug408. If you're interested in looking at fixes, please update there, otherwise perhaps I've...
02/11/2018
- 11:53 PM Issue #1275 (Duplicate): genvar in always_comb doesn't fail or lint warn during compile when not ...
- I think the language spec only allows for genvar variable usage in always_comb and assign when under a generate scope...
- 12:11 AM Issue #1274 (Resolved): Pullup / Pulldown on arrays causes internal crash
- Two problems, first it should have printed an unsupported error. Then I presume you want it supported.
Both fixed ...
Also available in: Atom