Vivado parses wicked slow

[rickman]

On 8/7/2016 10:20 AM, already5chosen@yahoo.com wrote:

On Friday, August 5, 2016 at 12:04:52 AM UTC+3, rickman wrote:

I know that the simulators are intentionally speed crippled to encourage
users to upgrade to paid versions. I don't know if they do the same
thing with synthesis tools or not.

--

Rick C

We are talking about X&A own integrated synthesis. I don't believe that there is a difference between "paid" and "free" tools except that "free" tools can't target certain devices.
Anyway, all my Quartus measurements were "paid".
Vivado measurements were sort of paid too - license came due to voucher that was attached to Zync Eval. board.

I didn't realize Vivado was a Xilinx product. I haven't done much with
Xilinx in some time. Diamond does include a Lattice synthesis tool, but
I always use the Synopsis tool which is third party.

I seem to recall Xilinx stopped offering third party synthesis with the
free tools some time back, but I seem to recall they had something
different. But then many years ago a Xilinx person pointed out to me
they spend more money on software development than they do software, or
maybe they used more people or something like that. The point was they
are a software company that collects revenue from selling the hardware
that the software supports.

--

Rick C

 

[Kevin Neilson]

Interesting. I used your files and got the same results: 15s for synth, 40s for PAR. I wonder why yours is so much faster. I reduced the clock period in yours to 2.7ns, and the synth time went to 35s, but the PAR was still about 45s. I'm using Verilog and a different part. Curious. I also have a bunch of other files in my project, but they're not in the build hierarchy, so that shouldn't make a difference.