From: Peter Dyballa <Peter_Dyballa@Web.DE>
To: Segher Boessenkool <segher@kernel.crashing.org>
Cc: Jonathan Wakely <jwakely.gcc@gmail.com>, gcc-help@gcc.gnu.org
Subject: Re: Configuring GCC 10.3 on PPC Mac OS X 10.4.11/Tiger for build reveals flaws in configure scripts
Date: Thu, 1 Dec 2022 00:09:50 +0100 [thread overview]
Message-ID: <D1043E80-0574-456C-A248-537540227820@Web.DE> (raw)
In-Reply-To: <20221125155607.GI25951@gate.crashing.org>
Hello!
Here is my evaluation from building GCC 10.4 on PPC Mac OS X 10.4.11, Tiger, with modern tools with 40 hours. I am using in the first, leftmost column, line numbers from compile log in GNU Emacs which quantify when one of the many steps is taken. It might help classify and explain some differences between building GCC 10.3 and GCC 10.4 I encountered.
10:Configuring stage 1 in ./intl
134:Configuring stage 1 in ./gcc
620:Configuring stage 1 in ./libiberty
configure: target_header_dir = <=== no title here
1367:Configuring in build-powerpc-apple-darwin8.11.0/libiberty
configure: target_header_dir = <=== no title here
2110:Configuring in build-powerpc-apple-darwin8.11.0/fixincludes
2232:Configuring in build-powerpc-apple-darwin8.11.0/libcpp
2397:Configuring stage 1 in ./libbacktrace
checking for readelf... readelf
checking whether objcopy supports debuglink... ..../gcc-10.4.0/libbacktrace/configure: line 13493: --add-gnu-debuglink=x: command not found
no
checking whether tests can run... yes
2598:Configuring stage 1 in ./libcpp
2766:Configuring stage 1 in ./libdecnumber
checking target system type... powerpc-apple-darwin8.11.0
checking for decimal floating point... configure: WARNING: decimal float is not supported for this target, ignored
==> dpd
checking whether byte ordering is bigendian... yes
4128:Configuring stage 1 in powerpc-apple-darwin8.11.0/libgcc
5263:Configuring stage 1 in powerpc-apple-darwin8.11.0/libgomp
5588:Configuring stage 1 in powerpc-apple-darwin8.11.0/libstdc++-v3
configure: Debug build flags set to -gdwarf-4 -g3 -O0 -D_GLIBCXX_ASSERTIONS
...
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no2
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
...
Running configure in multilib subdir ppc64
...
configure: Debug build flags set to -gdwarf-4 -g3 -O0 -D_GLIBCXX_ASSERTIONS /* for orientation */
...
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
8500:Configuring stage 2 in ./intl
8626:Configuring stage 2 in ./gcc
checking __stack_chk_fail in target C library... checking for __stack_chk_fail... no
no __stack_chk_fail on this target
==> no
checking sys/sdt.h in the target C library... no
...
checking for otool64... otool64
checking for -single_module linker flag... warning: no debug map in executable (-arch ppc)
yes
checking for -exported_symbols_list linker flag... yes
checking for -force_load linker flag... warning: no debug map in executable (-arch ppc)
no
checking for dlfcn.h... yes
9116:Configuring stage 2 in ./libiberty
checking whether to install libiberty headers and static library... no
configure: target_header_dir =
checking for powerpc-apple-darwin8.11.0-gcc... ..../feddich/./prev-gcc/xgcc -B..../feddich/./prev-gcc/ -B/usr/local/powerpc-apple-darwin8.11.0/bin/ -B/usr/local/powerpc-apple-darwin8.11.0/bin/ -B/usr/local/powerpc-apple-darwin8.11.0/lib/ -isystem /usr/local/powerpc-apple-darwin8.11.0/include -isystem /usr/local/powerpc-apple-darwin8.11.0/sys-include -fno-checking
9888:Configuring stage 2 in ./libbacktrace
checking whether objcopy supports debuglink... ..../gcc-10.4.0/libbacktrace/configure: line 13493: --add-gnu-debuglink=x: command not found
no
checking whether tests can run... yes
10096:Configuring stage 2 in ./libcpp
10266:Configuring stage 2 in ./libdecnumber
checking target system type... powerpc-apple-darwin8.11.0
checking for decimal floating point... configure: WARNING: decimal float is not supported for this target, ignored
==> dpd
checking whether byte ordering is bigendian... yes
34397:Configuring stage 2 in powerpc-apple-darwin8.11.0/libgcc
35532:Configuring stage 2 in powerpc-apple-darwin8.11.0/libgomp
35857:Configuring stage 2 in powerpc-apple-darwin8.11.0/libstdc++-v3
configure: Debug build flags set to -gdwarf-4 -g3 -O0 -D_GLIBCXX_ASSERTIONS
...
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
...
Running configure in multilib subdir ppc64
...
configure: Debug build flags set to -gdwarf-4 -g3 -O0 -D_GLIBCXX_ASSERTIONS
...
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for shared library run path origin... done
checking for iconv... no, consider installing GNU libiconv <=== libiconv and /opt/local/bin/iconv are installed, other tests (stage3 stdc++) find it
looking for a compliant stdint.h in stdint.h, checking for uintmax_t... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
38769:Configuring stage 3 in ./intl
38895:Configuring stage 3 in ./gcc
checking __stack_chk_fail in target C library... checking for __stack_chk_fail... no
no __stack_chk_fail on this target
==> no
checking sys/sdt.h in the target C library... no
39383:Configuring stage 3 in ./libiberty
40155:Configuring stage 3 in ./libbacktrace
checking output filetype...
configure: WARNING: could not determine output file type <=== happens three times during build
looking for a compliant stdint.h in stdint.h, checking for uintmax_t... yes
...
checking for readelf... readelf
checking whether objcopy supports debuglink... ..../gcc-10.4.0/libbacktrace/configure: line 13493: --add-gnu-debuglink=x: command not found
no
checking whether tests can run... yes
40361:Configuring stage 3 in ./libcpp
40531:Configuring stage 3 in ./libdecnumber
checking target system type... powerpc-apple-darwin8.11.0
checking for decimal floating point... configure: WARNING: decimal float is not supported for this target, ignored
==> dpd
checking whether byte ordering is bigendian... yes
64662:Configuring stage 3 in powerpc-apple-darwin8.11.0/libgcc
65797:Configuring stage 3 in powerpc-apple-darwin8.11.0/libgomp
66122:Configuring stage 3 in powerpc-apple-darwin8.11.0/libstdc++-v3
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
...
Running configure in multilib subdir ppc64
...
configure: Debug build flags set to -gdwarf-4 -g3 -O0 -D_GLIBCXX_ASSERTIONS
...
checking uchar.h usability... no
checking uchar.h presence... no
checking for uchar.h... no
==> no
checking for int64_t... yes
checking for int64_t as long... checking for int64_t as long long... yes
...
checking for local stylesheet directory... yes
configure: <=== no comment here
checking for docbook stylesheets for documentation creation... yes
69046:Configuring in ./fixincludes
69175:Configuring in ./libcc1
checking for gethostbyname... yes
<=== empty line
checking for exported symbols... yes
69307:Configuring in powerpc-apple-darwin8.11.0/libssp
69784:Configuring in powerpc-apple-darwin8.11.0/libquadmath
Running configure in multilib subdir ppc64
checking for egrep... /opt/local/bin/ggrep -E
checking for ANSI C header files... rm: conftest.dSYM: is a directory
==> rm: conftest.dSYM: is a directory <=== complains twice
yes
checking for sys/types.h... yes
70158:Configuring in powerpc-apple-darwin8.11.0/libobjc
70851:Configuring in powerpc-apple-darwin8.11.0/libitm
71440:Configuring in powerpc-apple-darwin8.11.0/libatomic
GCC 10.3 was built with old system's tools and utilities, while for GCC 10.4 I could use modernised versions. This might explain a few differences. Particularly the following:
134:Configuring stage 1 in ./gcc
524:checking assembler for -mmacosx-version-min option... no
537:checking assembler for compressed debug sections... 0
549:checking linker for compressed debug sections... 0
Presumingly the old as is better integrated in Mac OS X and configure finds in GCC 10.3:
checking assembler for -mmacosx-version-min option... yes
checking assembler for compressed debug sections... 2
checking linker for compressed debug sections... 0
Is this difference intended?
Three times configure finds:
2509:configure: WARNING: could not determine output file type
10005:configure: WARNING: could not determine output file type
40270:configure: WARNING: could not determine output file type
while it finds 45 times:
checking for C compiler default output file name... a.out
Is this difference intended?
Another (intended?) difference is this:
10.3:
checking whether a program can dlopen itself... yes
checking whether a statically linked program can dlopen itself... yes
checking whether a program can dlopen itself... cross
checking whether a program can dlopen itself... yes
checking whether a statically linked program can dlopen itself... yes
checking whether a program can dlopen itself... cross
checking whether a program can dlopen itself... yes
checking whether a statically linked program can dlopen itself... yes
checking whether a program can dlopen itself... cross
10.4:
5673:checking whether a program can dlopen itself... no
6244:checking whether a program can dlopen itself... cross
35942:checking whether a program can dlopen itself... no
36513:checking whether a program can dlopen itself... cross
66207:checking whether a program can dlopen itself... no
66778:checking whether a program can dlopen itself... cross
A bit disturbing is this, finding yes and no for the same item, __int128 (because libquadmath supports this data type?):
5698:checking for ISO/IEC TR 24733 ... no
5699:checking for __int128... no
5700:checking for __float128... no
6269:checking for ISO/IEC TR 24733 ... no
6270:checking for __int128... yes
6271:checking for __float128... no
35967:checking for ISO/IEC TR 24733 ... no
35968:checking for __int128... no
35969:checking for __float128... no
36538:checking for ISO/IEC TR 24733 ... no
36539:checking for __int128... yes
36540:checking for __float128... no
66232:checking for ISO/IEC TR 24733 ... no
66233:checking for __int128... no
66234:checking for __float128... no
66803:checking for ISO/IEC TR 24733 ... no
66804:checking for __int128... yes
66805:checking for __float128... no
(Disk space used after building: almost 5 GB.)
--
Greetings
Pete
A designer knows he has arrived at perfection not when there is no longer anything to add, but when there is no longer anything to take away.
– Antoine de Saint-Exupéry
next prev parent reply other threads:[~2022-11-30 23:09 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-25 10:19 Peter Dyballa
2022-11-25 11:22 ` Jonathan Wakely
2022-11-25 11:29 ` Jonathan Wakely
2022-11-25 11:39 ` Jonathan Wakely
2022-11-25 11:30 ` Peter Dyballa
2022-11-25 15:56 ` Segher Boessenkool
2022-11-25 16:50 ` Peter Dyballa
2022-11-25 19:43 ` Segher Boessenkool
2022-11-28 22:05 ` Peter Dyballa
2022-11-30 23:09 ` Peter Dyballa [this message]
2022-12-01 8:22 ` Jonathan Wakely
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=D1043E80-0574-456C-A248-537540227820@Web.DE \
--to=peter_dyballa@web.de \
--cc=gcc-help@gcc.gnu.org \
--cc=jwakely.gcc@gmail.com \
--cc=segher@kernel.crashing.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).