Re: gEDA-user: net labels question

[Stuart Brorson <sdb@xxxxxxxxxx>]

>>> Does everyone understand a trailing underscore as negative logic?
>>
>> No.
>>
>> In my experience, a negatively asserted signal can be flagged in
>> several different ways,
>
> A suffix of SIG_N seems good -- it can either signify Not() or
> Negative logic or Negative half of a pair...

That's true.

OTOH, I have also seen folks use SIG_H and SIG_L for diff pairs, and
this naming convention can be interpreted similarly.

As my personal rule, I have settled on using DIFF_P and DIFF_N for
diff pairs, and SIG_L for single-ended, low-asserted signals (like a
reset).  That way I can distinguish between a net in a diff pair and a
normal logic signal.

(Unless I am working on something where there is a pre-defined
convention, that is....)

YMMV.  There's no standard way of doing it.  I think most people just
make up a convention and then stick to it throughout their working
lives.

Slightly off-topic -- It gets hairier:  When I was doing high-speed
telecom boards, I had lots of signals of different logic families
running around.  I adoped the convention of first giving the logic
family, then the netname (i.e. the net's functionality), and then the
logic sense (if a diff pair or if asserted low).  For examples:

LVPECL_LINE_IN_P
LVPECL_LINE_IN_N
MRESET_L
CML_OPTO_OUT_P
CML_OPTO_OUT_N
LATCH_L
TTL3P3_DATA[7:0]
LVDS_DATA[15:0]_P
LVDS_DATA[15:0]_N

I did this because we used a spreadsheet program to attach routing
attributes (e.g. diff pair length tolerances) to the nets, and being
able to sort on logic family prior to attaching attributes was very
useful.

We can't do this yet in gEDA.   :-(

Stuart


_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user