On Thursday 05 June 2008 22:01:15 Tristan Gingold wrote: > On Thu, Jun 05, 2008 at 12:13:19PM +0200, [EMAIL PROTECTED] wrote: > > > this is slightly too vague. I tried a few cases and all were > > > sucessfull. What is the type of xyz ? > > > > The type was a subtype to integer. > > > > The reason is I like to use 'high 'low 'range etc... rather to hardcode > > limits in the software. That allows to change the limits ( or take > > values from a generic ) without reviewing the code, and to set all > > parameters on the begining of the architecture. > > Yes, this is a very common use of for generate. > Can you send me a reproducer ? That would be very useful.
I don't know if this is what the OP meant, but here is an example of where 'high and 'low work, but 'range doesn't. What I'm not clear on, is if this is a bug in GHDL or is standard behavior. entity bug is end entity; architecture works of bug is subtype t1 is integer range 1 to 2; begin gen : for i in t1'low to t1'high generate assert false report integer'image(i) severity note; end generate; end architecture; $ ghdl -a bug.vhdl $ ghdl -m bug elaborate bug $ ./bug test.vhdl:9:5:@0ms:(assertion note): 1 test.vhdl:9:5:@0ms:(assertion note): 2 On the other hand: entity bug is end entity; architecture does_not_work of bug is subtype t1 is integer range 1 to 2; begin gen : for i in t1'range generate assert false report integer'image(i) severity note; end generate; end architecture; $ ghdl -a bug.vhdl bug.vhdl:8:21: prefix of range attribute must be an array type or object bug.vhdl:8:3: range must be a static discrete range bug.vhdl:9:39: can't match 'i' with type integer /usr/lib/ghdl/bin/ghdl: compilation error Again, I don't know if this a bug or a standard VHDL "feature", but this may be what the OP meant. -- Wesley J. Landaker <[EMAIL PROTECTED]> <xmpp:[EMAIL PROTECTED]> OpenPGP FP: 4135 2A3B 4726 ACC5 9094 0097 F0A9 8A4C 4CD6 E3D2
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Ghdl-discuss mailing list Ghdl-discuss@gna.org https://mail.gna.org/listinfo/ghdl-discuss