I've run into this before on the Assembler List. Yes, it should logically work, and I guess it's just one of the edge cases that HLASM hasn't sorted out yet. I switched to LARL; it works, and isn't any less efficient or straightforward really.
Generally, I think HLASM needs to overhaul the way it processes the immediates... it's easy to e.g. L 15,=AL1(B'00000000',(0*16),(68),B'00000011'), and nearly impossible with IILF, LLILF or LGFI, without performing unnatural acts. sas On Wed, Dec 28, 2016 at 10:51 AM, John McKown <[email protected]> wrote: > On Tue, Dec 27, 2016 at 7:01 PM, Charles Mills <[email protected]> wrote: > > > Why won't this assemble? > > > > SOME CSECT > > LLILF 2,FOO > > OTHER CSECT > > FOO EQU * > > > > Why do I get ASMA032E Relocatable value or unresolved symbol found when > > absolute value required - FOO? > > > > Can't the assembler generate a relocatable address constant for the > > immediate operand of LLILF? Or am I doing something stupid that I am not > > seeing? > > > > DC A(FOO) at the same spot assembles with no problem. I could presumably > > code DC X'C02F',AL4(FOO) and it would execute correctly. > > > > Charles > > > > > I would go with: > > LLILF 2,0 > ORG *-4 > DC VL4(FOO) > > At least it assembles. > > > -- > Heisenberg may have been here. > > http://xkcd.com/1770/ > > Maranatha! <>< > John McKown > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- sas ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
