Subject: Re: More on connect-resolveto...
From: Paul J. Menchini (mench@mench.com)
Date: Thu Jan 24 2002 - 06:30:21 PST
This from Kevin Cameron....
Forwarded message:
>>From owner-verilog-ams Thu Jan 24 00:10:38 2002
> Date: Thu, 24 Jan 2002 00:10:21 -0800 (PST)
> From: Kevin Cameron <edaorg@v-ms.com>
> Message-Id: <200201240810.AAA05916@grfx.com>
> To: verilog-ams@server.eda.org, schandra@asc.corp.mot.com
> Subject: Re: More on connect-resolveto...
> X-Sun-Charset: US-ASCII
>> Hi all,
>>
>> In connect-resolveto statement, for example:
>> connect cmos1 cmos2 cmos3 resolveto cmos1;
>>
>> Would cmos2 cmos3 (subset of the above list) also resolveto cmos1?
>>
>> If that is true what if there is a connect-resolveto statement that conflicts
>> with the above rule.
>> ie.
>> connect cmos1 cmos2 cmos3 resolveto cmos1;
>> connect cmos2 cmos3 resolveto cmos3;
>>
>> Is the above an error with conflicting resolutions?
>>
>> The LRM is currently not clear on the above point.
>>
>> cheers,
>> Sri
> It is difficult to find a good algorithm for this. The latter case
> above can be easily understood as the last "connect/resolveto" is a
> strict subset of the previous one and therefore more specific (and
> the prefered choice), if on the other hand you have:
> connect cmos1 cmos2 cmos3 resolveto cmosX;
> connect cmos0 cmos1 cmos2 resolveto cmosY;
> - and you are connecting cmos1 and cmos2 but there is no rule
> specific cmos1/cmos2 rule, which of cmosX or cmosY do you use.
> Some extra information may be required if you don't want to just
> take the first (or last) match.
> Kev.
This archive was generated by hypermail 2b28 : Thu Jan 24 2002 - 06:31:24 PST