Maurice Lombardi wrote:
CBFalconer wrote:
Maurice Lombardi wrote:
Frank Heckenbach wrote:
da Silva, Joe wrote:
Well AFAIK, there is no "later than 2.95.3", that is suitable. From previous discussions, 3.XX is not currently supported by GPC and 2.96 is a broken version of GCC, so 2.95.3 is "it" (since this version for DJGPP is not listed below, by implication there were no problems with it, so I expect Maurice will have a build available shortly:-).
I haven't tested this particular configuration because I know that Maurice will do. :-) Though I don't expect problems here.
No problem indeed: 0 errors in the test suite. It is uploaded as usual under the fixed link
ftp://agnes.dida.physik.uni-essen.de/home/maurice/gpc2953b.zip
I assume that is the new version, even though it has the identical name to the older one? How are we to keep track :-)
It is always the last alpha I have compiled with zero error in running the test suite. The link points to the actual file in the same directory ( gpc-20011202-gcc-2953.i586-pc-msdosdjgpp.zip presently ). There is no reason not to download the last version successfully compiled and tested. A fixed link is handy for people having http pages which points to gpc
Woops - possible problem. I note the "i586" in that name. I am running on a 486. Does it have Pentium only opcodes embedded somewhere?
CBFalconer wrote:
Maurice Lombardi wrote:
Woops - possible problem. I note the "i586" in that name. I am running on a 486. Does it have Pentium only opcodes embedded somewhere?
I do not know. gcc, gpp, f77 and all other compilers of the suite available on simtelnet / djgpp are compiled with this target since gcc-2.95.2 at least. I suppose that it means only that it is better optimized for pentium. But I am not sure.
Maurice
Maurice Lombardi wrote:
CBFalconer wrote:
Woops - possible problem. I note the "i586" in that name. I am running on a 486. Does it have Pentium only opcodes embedded somewhere?
I do not know. gcc, gpp, f77 and all other compilers of the suite available on simtelnet / djgpp are compiled with this target since gcc-2.95.2 at least. I suppose that it means only that it is better optimized for pentium. But I am not sure.
From the GCC documentation:
: `-mcpu=CPU TYPE' : Assume the defaults for the machine type CPU TYPE when scheduling : instructions. The choices for CPU TYPE are: : : `i386' `i486' `i586' `i686' : `pentium' `pentiumpro' `k6' : : While picking a specific CPU TYPE will schedule things : appropriately for that particular chip, the compiler will not : generate any code that does not run on the i386 without the : `-march=CPU TYPE' option being used. `i586' is equivalent to : `pentium' and `i686' is equivalent to `pentiumpro'. `k6' is the : AMD chip as opposed to the Intel ones.
So as long as GPC is built with `-mcpu=i586' it should be ok, but not necessarily with `-march=i586'.
Frank
-
CBFalconer -
Frank Heckenbach -
Maurice Lombardi