samperi@marob.MASA.COM (Dominick Samperi) (09/19/88)
It appears that the assembler that is supplied with SCO Xenix (386) cannot
assemble code that is generated by the SCO Xenix C compiler. That is, code
that is generated via "cc -S -c prog.c" will causes masm to generate error
messages about the COMM directive, and some other messages about illegal
ORG arguments, etc. The generated assembly code _will_ compile under
MSDOS, using MASM 5.0+! Unfortunately, I don't think the object code file
generated can be used under XENIX (?).
Is there some information I'm missing here? Any tips will be greatly
appreciated.
--
Dominick Samperi, NYC
samperi@acf8.NYU.EDU samperi@marob.MASA.COM
cmcl2!phri!marob uunet!hombre!samperi
(^ ell)hudson@vsedev.VSE.COM (C Hudson Hendren III) (09/20/88)
In article <416@marob.MASA.COM> samperi@marob.MASA.COM (Dominick Samperi) writes: >It appears that the assembler that is supplied with SCO Xenix (386) cannot >assemble code that is generated by the SCO Xenix C compiler. That is, code >that is generated via "cc -S -c prog.c" will causes masm to generate error >messages about the COMM directive, and some other messages about illegal >ORG arguments, etc. >Is there some information I'm missing here? Any tips will be greatly >appreciated. It says right here in the SCO XENIX Programmer's Reference Manual under the man page for cc that the -S option to cc creates an assembly listing that is not suitable for assembly. The option provides code for reading only. Can it possibly be stated any more clearly than that or didn't you RTFM? -- ==> ..!uunet!vsedev!hudson [hudson@vsedev.vse.com] (C Hudson Hendren III) <== ==> These are my opinions and are not necessarily those of VSE Corporation. <== ==> MS-DOS was created to keep idiots away from UNIX computers <==
karlh@ico.ISC.COM (Karl Hanzel) (09/20/88)
In article <416@marob.MASA.COM> samperi@marob.MASA.COM (Dominick Samperi) writes: >It appears that the assembler that is supplied with SCO Xenix (386) cannot >assemble code that is generated by the SCO Xenix C compiler. That is, code >that is generated via "cc -S -c prog.c" will causes masm to generate error >messages about the COMM directive, and some other messages about illegal >ORG arguments, etc. The generated assembly code _will_ compile under >MSDOS, using MASM 5.0+! Unfortunately, I don't think the object code file >generated can be used under XENIX (?). > >Is there some information I'm missing here? Any tips will be greatly >appreciated. I've seen this with Microsoft Xenix-286 for my Intel System 310,... using the 'as' assembler. For one, the assembler uses '|' as a delimeter for comments, whereas the compiler 'cc -S' produces ';'s. But beyond that 'as' still doesn't work to produce valid dot-o's. Ugh. pax, Karl *----->
samperi@marob.MASA.COM (Dominick Samperi) (09/20/88)
In article <1183@vsedev.VSE.COM> hudson@vsedev.VSE.COM (C Hudson Hendren III) writes: >It says right here in the SCO XENIX Programmer's Reference Manual under the >man page for cc that the -S option to cc creates an assembly listing that is >not suitable for assembly. The option provides code for reading only. This is rediculous. How does one get a _valid_ assembly listing? -- Dominick Samperi, NYC samperi@acf8.NYU.EDU samperi@marob.MASA.COM cmcl2!phri!marob uunet!hombre!samperi (^ ell)
uhclem@trsvax.UUCP (09/24/88)
R2>This is rediculous. How does one get a _valid_ assembly listing? Beats me. Also, have you ever tried to look at the actual input to the compiler (post cpp) by running the provided cpp? Suprise! It is not what the compiler uses and lots of things don't work like long symbols. We used to ask Microsoft about the compiler output, cpp and lint and for cpp and lint we got "you don't need them, the C compiler has all of that built-in." We never got a good answer about the valid assembler output, but since they don't run the code through masm, I guess they could say you don't need that either. "Thank you, Uh Clem." Frank Durda IV @ <trsvax!uhclem> ...decvax!microsoft!trsvax!uhclem ...sys1!hal6000!trsvax!uhclem I've come to praise ihnp4, not to send mail through him. <I'm not a letnI-hater, I'm just coded that way...>