#11073: remove the spkg/base repo?
---------------------------+------------------------------------------------
Reporter: jhpalmieri | Owner: GeorgSWeber
Type: enhancement | Status: new
Priority: blocker | Milestone: sage-4.7.1
Component: build | Keywords: scripts base hg
Author: | Upstream: N/A
Reviewer: | Merged:
Work_issues: |
---------------------------+------------------------------------------------
Comment(by drkirkby):
Replying to [comment:8 jdemeyer]:
> Replying to [comment:7 drkirkby]:
> > It's pretty dumb, as it forces a long to be 4 bytes with:
> >
> > {{{
> > typedef unsigned long int_fast32_t;
> > }}}
> >
> > which is obviously invalid on 64-bit builds.
>
> It is invalid, but not for the reason you state.
>
> The problem is that `int_fast32_t` is supposed to be '''signed''', while
`unsigned long` is '''unsigned'''. As for size, the type `int_fast32_t`
is required to be ''at least'' 32 bits, not ''exactly'' 32 bits, so the
following would be correct, both on 32-bit and 64-bit machines:
> {{{
> typedef long int_fast32_t;
> }}}
On Solaris the int_fast_32 remains at 32-bits, even on 64-bit code, so
what actually caused Sage to fail to compile was certainly the size rather
than the sign.
That little bit of header file was included in Sage well after Sage built
reliably 32-bit. It was only when I tried a 64-bit build of Sage that this
became a problem. (I think a 64-bit Sage on Solaris is looking further
away now, as we don't have a good solution for making sure modules are
initialized properly before anything is imported from them.)
== Test program ==
{{{
drkirkby@laptop:~$ cat test.c
#include <stdio.h>
#include <stdlib.h>
#include <sys/int_types.h>
int main() {
printf("sizeof(int)=%d\n",sizeof(int));
printf("sizeof(long)=%d\n",sizeof(long));
printf("sizeof(int_fast32_t)=%d\n",sizeof(int_fast32_t));
exit(0);
}
}}}
== Create a 32-bit executable and execute it ==
{{{
drkirkby@laptop:~$ gcc test.c
drkirkby@laptop:~$ ./a.out
sizeof(int)=4
sizeof(long)=4
sizeof(int_fast32_t)=4
}}}
== Create a 64-bit executable and execute it ==
{{{
drkirkby@laptop:~$ gcc -m64 test.c
drkirkby@laptop:~$ ./a.out
sizeof(int)=4
sizeof(long)=8
sizeof(int_fast32_t)=4
}}}
> In fact, on my 64-bit Linux laptop, the types `int_fast16_t` and
`int_fast32_t` are `typedef`ed to be 64-bit `long`s.
As you can see, that's not so on Solaris, int_fast32_t remains 32-bits,
even in 64-bit code.
I recall using a Cray Y-MP with Unicos, where
{{{
sizeof(char)=1
sizeof(short)=8
sizeof(int)=8
sizeof(long)=8
}}}
The Cray is the only system I've come across where shorts are 64-bit. That
tripped me up, as some code I'd written assumed shorts were 2 bytes.
Dave
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11073#comment:13>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
