1 |
On Thu, 5 Feb 2004 22:55:36 -0500 |
2 |
Aron Griffis <agriffis@g.o> wrote: |
3 |
|
4 |
> Marc Giger wrote: [Wed Feb 04 2004, 06:13:30PM EST] |
5 |
> > Can someone of you comment why the following patch is needed? |
6 |
> > |
7 |
> > diff -ruN glibc-2.3.2.orig/include/features.h |
8 |
> > glibc-2.3.2/include/features.h--- |
9 |
> > glibc-2.3.2.orig/include/features.h 2003-06-14 00:28:23.000000000 |
10 |
> > +0100+++ glibc-2.3.2/include/features.h |
11 |
> > 2003-06-14 00:58:57.000000000 +0100@@ -285,7 +285,8 @@ |
12 |
> > #if defined __GNUC__ \ |
13 |
> > || (defined __PGI && defined __i386__ ) \ |
14 |
> > || (defined __INTEL_COMPILER && (defined __i386__ || defined |
15 |
> > __ia64__))\- || (defined __STDC_VERSION__ && __STDC_VERSION__ >= |
16 |
> > 199901L)+ || (defined __STDC_VERSION__ && __STDC_VERSION__ >= |
17 |
> > 199901L)\+ &&!(defined(__DECC) || defined(__DECCXX)) |
18 |
> > # define__GLIBC_HAVE_LONG_LONG 1 |
19 |
> > #endif |
20 |
> > |
21 |
> > The Compaq C Compiler knows the "long long" datatype like gcc does. |
22 |
> > It's also of the same size on both compilers |
23 |
> > (long == long int == long long == 8bytes). |
24 |
> |
25 |
> You're right, this looks broken to me. |
26 |
|
27 |
Yeah, it looks like:-) but I was wrong... I removed this check and ran |
28 |
into troubles. The Macro __GLIBC_HAVE_LONG_LONG must not be defined |
29 |
with ccc. So this check is totally ok! It has nothing to do with |
30 |
"long long". Sorry... I was searching for the reason why I couldn't |
31 |
compile python with ccc. |
32 |
After some time I found it! The attached patch solves the problem. Now |
33 |
I'm running python 2.3 compiled with ccc:-) |
34 |
|
35 |
|
36 |
> |
37 |
> > printf '%s %s %s\n' \ |
38 |
> > "-L${gcc_libs_path}" \ |
39 |
> > " -SysIncDir |
40 |
> > /usr/lib/compaq/ccc-${ccc_release}/alpha-linux/include" \ |
41 |
> > |
42 |
> > " -SysIncDir /usr/include/linux" \ > |
43 |
> > usr/lib/compaq/ccc-${ccc_release}/alpha-linux/bin/comp.config |
44 |
> |
45 |
> So do you know how this should be changed exactly? |
46 |
|
47 |
Hehe, no! |
48 |
This seems not to be the right place neither. I think this is useless |
49 |
anyway. The right place to fix it is in "create-comp-config.sh". I tried |
50 |
to change it but it seems that a shell-script and regex guru is |
51 |
needed;-) The same should be changed by cxx (preventing). |
52 |
|
53 |
With these two fixes, I'm able to compile a lot more packages with the |
54 |
compaq compiler. |
55 |
|
56 |
|
57 |
Oh, while we are at it. What do you think if we move the libots libs to |
58 |
/lib instead of /usr/lib ? |
59 |
|
60 |
Lastly, I ran into troubles as I compiled gawk with ccc. |
61 |
Could you imagine what happens? I've made a RAID array with several |
62 |
partitions on it. One partition was /usr. |
63 |
gawk is needed in the init scripts, but libots.so is located in |
64 |
/usr/lib, which is not mounted yet. Too bad. |
65 |
|
66 |
greets |
67 |
|
68 |
Marc |