Mail Archives: djgpp-workers/1999/08/18/10:18:20
On 18 Aug 99, at 15:55, Eli Zaretskii wrote:
>
> On Mon, 16 Aug 1999 pavenis AT lanet DOT lv wrote:
>
> > At configure time some features of GAS is checked and if they are
> > present they are used also in compiler later. I built "official" port of
> > GCC-2.95 using binutils-2.9.1 which does not have this feature.
>
> May I suggest that GCC be configured with Binutils 2.8.1? v2.9.1 is not
> available in the DJGPP port yet, and who knows what features does 2.9.1
> add which GCC 2.95 will now try to use?
>
I haven't used binutils-2.8.1 for a long time. It's mean that also
egcs-1.1.2 were built using binutils-2.9.1 (I don't remeber about last update of
gcc-2.8.1)
Here is fragment from gcc/auto_host.h where the detected features appear
(generated by configure) from gcc-2.95.1
/* Define if your assembler supports specifying the maximum number
of bytes to skip when using the GAS .p2align command. */
#define HAVE_GAS_MAX_SKIP_P2ALIGN 1
/* Define if your assembler supports .balign and .p2align. */
#define HAVE_GAS_BALIGN_AND_P2ALIGN 1
/* Define if your assembler supports .subsection and .subsection -1 starts
emitting at the beginning of your section */
/* #undef HAVE_GAS_SUBSECTION_ORDERING */
/* Define if your assembler uses the old HImode fild and fist notation. */
/* #undef HAVE_GAS_FILDS_FISTS */
The problem with binutils snapshot was that configure defined
#define HAVE_GAS_FILDS_FISTS
which made problems with earlier binutils.
I think both
#define HAVE_GAS_MAX_SKIP_P2ALIGN 1
#define HAVE_GAS_BALIGN_AND_P2ALIGN 1
are not dangerous with binutils-2.8.1
That are all GAS related features detected by configure.
Andris
- Raw text -