rmfowler@texrex.uucp (Rex Fowler) (05/04/91)
How do I tell the GNU assembler (gas) to assemble 68010 code? I thought that I knew at one time but I can't get anything to work now. I got the binary from OSU and I don't seem to have any docs. What I'm wanting to do is see if gas is able to assemble teval.s from the perl 4.0.0.3 distribution. eval.c compiled fine but the tainted version (teval.c) won't (ie. Branch offset too remote) If you've ever tried to compile perl, you know what I'm talking about. perl 3.0 pl44 wasn't much problem but 4.0 has got me stumped. I'm hoping that gas will allow larger branch offsets than the stock assembler. -- Rex Fowler <rmfowler@texrex.cirr.com> UUCP: egsner!texrex!rmfowler
tkacik@kyzyl.mi.org (Tom Tkacik) (05/05/91)
In article <1991May4.040336.10403@texrex.uucp>, rmfowler@texrex.uucp (Rex Fowler) writes: > What I'm wanting to do is see if gas is able to assemble teval.s > from the perl 4.0.0.3 distribution. eval.c compiled fine but the > tainted version (teval.c) won't (ie. Branch offset too remote) > > If you've ever tried to compile perl, you know what I'm talking about. > perl 3.0 pl44 wasn't much problem but 4.0 has got me stumped. I just took a quick look at the man page for as(1). There is a -j switch to as(1) that "invokes the long-jump assembler." This looks like it should solve your problem with the branch offsets being too long. I have not tried this, so let us know if it works. -- Tom Tkacik | tkacik@kyzyl.mi.org | To rent this space, call 1-800-555-QUIP. ...!rphroy!kyzyl!tkacik |
rmfowler@texrex.uucp (Rex Fowler) (05/08/91)
In article <334@kyzyl.mi.org> tkacik@kyzyl.mi.org (Tom Tkacik) writes: >In article <1991May4.040336.10403@texrex.uucp>, rmfowler@texrex.uucp writes: >> What I'm wanting to do is see if gas is able to assemble teval.s >> from the perl 4.0.0.3 distribution. eval.c compiled fine but the >> tainted version (teval.c) won't (ie. Branch offset too remote) [ I'm trying to compile with gcc but as(1) won't assemble teval.s ] >> > >I just took a quick look at the man page for as(1). There is a -j switch >to as(1) that "invokes the long-jump assembler." This looks like it >should solve your problem with the branch offsets being too long. > >I have not tried this, so let us know if it works. I tried it, it doesn't help. > >-- >Tom Tkacik | >tkacik@kyzyl.mi.org | To rent this space, call 1-800-555-QUIP. >...!rphroy!kyzyl!tkacik | By the way, I did succeed in building perl 4.003 with setuid/setgid emulation with the stock cc. Also, I haven't rec'd any info on assembling with gas yet, so if you're waiting on the other guy to tell me, wait no longer. -- Rex Fowler <rmfowler@texrex.cirr.com> UUCP: egsner!texrex!rmfowler
tkacik@hobbes.cs.gmr.com (Tom Tkacik CS/50) (05/09/91)
In article <1991May8.035216.15851@texrex.uucp>, rmfowler@texrex.uucp (Rex Fowler) writes: |> In article <334@kyzyl.mi.org> tkacik@kyzyl.mi.org (Tom Tkacik) writes: |> >In article <1991May4.040336.10403@texrex.uucp>, rmfowler@texrex.uucp writes: |> >> What I'm wanting to do is see if gas is able to assemble teval.s |> >> from the perl 4.0.0.3 distribution. eval.c compiled fine but the |> >> tainted version (teval.c) won't (ie. Branch offset too remote) |> |> [ I'm trying to compile with gcc but as(1) won't assemble teval.s ] Last night I compiled perl 4.003. If you read my other message on the subject you will notice I am having some problems (though different than yours). (Anybody compile perl with dbm?) I compiled perl with gcc, except for eval.c (teval.c), and toke.c (ttoke.c). Gcc ran out of virtual memory, so I fell back on cc for those files. I did not have a problem with as(1) not being able to assemble any of the resulting assembler files. (Note that teval.c was compiled with cc.) I do not know why you ran into difficulty. |> By the way, I did succeed in building perl 4.003 with setuid/setgid emulation |> with the stock cc. What did you do to solve your problem? Enquiring minds want to know. :-) -- Tom Tkacik GM Research Labs tkacik@hobbes.cs.gmr.com tkacik@kyzyl.mi.org
rmfowler@texrex.uucp (Rex Fowler) (05/09/91)
In article <52765@rphroy.UUCP> tkacik@hobbes.cs.gmr.com (Tom Tkacik CS/50) writes: >In article <1991May8.035216.15851@texrex.uucp>, rmfowler@texrex.uucp >(Rex Fowler) writes: >|> In article <334@kyzyl.mi.org> tkacik@kyzyl.mi.org (Tom Tkacik) writes: >|> >In article <1991May4.040336.10403@texrex.uucp>, rmfowler@texrex.uucp >writes: >|> [ I'm trying to compile with gcc but as(1) won't assemble teval.s ] > > >I compiled perl with gcc, except for eval.c (teval.c), and toke.c (ttoke.c). >Gcc ran out of virtual memory, so I fell back on cc for those files. >I did not have a problem with as(1) not being able to assemble any of the >resulting assembler files. (Note that teval.c was compiled with cc.) I do not >know why you ran into difficulty. > >|> By the way, I did succeed in building perl 4.003 with setuid/setgid >emulation >|> with the stock cc. > >What did you do to solve your problem? Enquiring minds want to know. :-) Actually I'm not quite sure but I noticed that perly.fixer ran whereas before it didn't. It assumed that my machine was an Interactive UNIX. I've never went and poked around to see what made it run this time and not before. Anyway, it compiled without a hitch. Optimization was turned on for ALL files (as was -DDEBUGGING). The perly.fixer allowed for dynamic stack allocation in yacc(or something like that, I don't remember what it said exactly.) > >-- >Tom Tkacik >GM Research Labs >tkacik@hobbes.cs.gmr.com >tkacik@kyzyl.mi.org -- Rex Fowler <rmfowler@texrex.cirr.com> UUCP: egsner!texrex!rmfowler