Re: GCC 4.1.1 RC1 -- bootstrap error sparc-sun-solaris2.8
This is probably very low priority, maybe a release note? In the dim-and-musty-systems department, using cc: Sun WorkShop 6 update 1 C 5.2 2000/09/11 One gets: cc -c -g -DENABLE_CHECKING -DENABLE_ASSERT_CHECKING -DIN_GCC -DHAVE_CONFIG_H -I. -I. -I../../../gcc-4.1.1-20060517/gcc -I../../../gcc-4.1.1-20060517/gcc/. -I../../../gcc-4.1.1-20060517/gcc/../include -I./../intl -I../../../gcc-4.1.1-20060517/gcc/../libcpp/include ../../../gcc-4.1.1-20060517/gcc/tree-ssa-structalias.c -o tree-ssa-structalias.o "../../../gcc-4.1.1-20060517/gcc/tree-ssa-structalias.c", line 1615: warning: end-of-loop code not reached cc: Fatal error in /opt/SUNWspro/bin/../WS6U1/bin/acomp Status 138 gmake[3]: *** [tree-ssa-structalias.o] Error 138 Building tree-ssa-structalias.o with an older gcc works fine. I'll try to look at it more this afternoon and see if I can put in a patch. Marc Mengel <[EMAIL PROTECTED]>
Re: GCC 4.1.1 RC1 -- bootstrap error alpha-dec-osf4.0
This is probably very low priority, maybe a release note? In the dim-and-musty-systems department, using Compaq C V6.1-120 on Digital UNIX V4.0G (Rev. 1530) We get: cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/alloca.c, line 162: In this declaration, the type of "C_alloca" is not compatible with the type of a previous declaration of "C_alloca" at line number 559 in file /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/../include/libiberty.h. (notcompat) C_alloca (size_t size) ^ gmake[1]: *** [alloca.o] Error 1 So it looks like include/libiberty.h should define it as a PTR also, since ansidecl.h declares it differently if you're an old-school C compiler... Marc
Re: GCC 4.1.1 RC1 -- second bootstrap error alpha-dec-osf4.0
This is probably very low priority, maybe a release note? In the dim-and-musty-systems department, using Compaq C V6.1-120 on Digital UNIX V4.0G (Rev. 1530) We get: c -c -DHAVE_CONFIG_H -g -I. -I/tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/../include /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/concat.c -o concat.o cc: Warning: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/concat.c, line 105: Too few actual parameters in macro call. (toofewactuals) VA_OPEN (args, first); --^ cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/concat.c, line 105: Identifier expected but not found. (idexpected) VA_OPEN (args, first); --^ cc: Warning: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/concat.c, line 120: Too few actual parameters in macro call. (toofewactuals) VA_OPEN (args, first); --^ Not seeing the bug just yet, something funky with varargs... Compiles with older gcc just fine... Marc
Re: GCC 4.1.1 RC1 -- third bootstrap error alpha-dec-osf4.0
This is probably very low priority, maybe a release note? In the dim-and-musty-systems department, using Compaq C V6.1-120 on Digital UNIX V4.0G (Rev. 1530) We get: cc -c -DHAVE_CONFIG_H -g -I. -I/tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/../include /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c -o hashtab.o cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 197: In this declaration, parameter 1 has a different type than specified in an earlier declaration of this function. (mismatparam) hash_pointer (const PTR p) ^ cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 197: In this declaration, the type of "hash_pointer" is not compatible with the type of a previous declaration of "hash_pointer" at line number 71 in file /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c. (notcompat) hash_pointer (const PTR p) ^ cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 205: In this declaration, parameter 1 has a different type than specified in an earlier declaration of this function. (mismatparam) eq_pointer (const PTR p1, const PTR p2) ^ It goes on like this for quite a while. LOTS of inconsisencies of using PTR one place and void * in another... Marc Mengel <[EMAIL PROTECTED]>
Re: Re: GCC 4.1.1 RC1 -- third bootstrap error alpha-dec-osf4.0
So basically on alpha-dec-osf4.0 with Compaq C v6.1-120, I just had to use an older gcc to build libiberty. Everything chokes with PTR==(char *) vs (void *) all over the place... Marc W. Mengel wrote: This is probably very low priority, maybe a release note? In the dim-and-musty-systems department, using Compaq C V6.1-120 on Digital UNIX V4.0G (Rev. 1530) We get: cc -c -DHAVE_CONFIG_H -g -I. -I/tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/../include /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c -o hashtab.o cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 197: In this declaration, parameter 1 has a different type than specified in an earlier declaration of this function. (mismatparam) hash_pointer (const PTR p) ^ cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 197: In this declaration, the type of "hash_pointer" is not compatible with the type of a previous declaration of "hash_pointer" at line number 71 in file /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c. (notcompat) hash_pointer (const PTR p) ^ cc: Error: /tmp/build-gcc-v4_1_1/src/gcc-4.1.1-20060517/libiberty/hashtab.c, line 205: In this declaration, parameter 1 has a different type than specified in an earlier declaration of this function. (mismatparam) eq_pointer (const PTR p1, const PTR p2) ^ It goes on like this for quite a while. LOTS of inconsisencies of using PTR one place and void * in another... Marc Mengel <[EMAIL PROTECTED]>
