| Summary: | dev-lisp/clozurecl-1.11.5: fails to build on ~x86 in 17.0 profile | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Andrey Grozin <grozin> |
| Component: | Current packages | Assignee: | Common Lisp Bugs <common-lisp> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | slyfox |
| Priority: | Normal | Keywords: | PATCH |
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 642232 | ||
| Attachments: | clozurecl-1.11.5-no-pie-32.patch | ||
In the 13.0 profile, clozurecl-1.11.5 compiled fine on the same computer, and could successfully compile maxima and fricas. On a ~amd64 computer with 13.0 profile, clozurecl-1.11.5 compiles fine and can successfully compile maxima and fricas. I have no ~amd64 computer with 17.0 profile, and cannot test what happens. But from bug #665364 I see that clozurecl-1.11.5 can be emerged, but the resulting lisp cannot compile maxima. This is the reply I got from the upstream: Date: Sat, 7 Jul 2018 11:14:10 +0200 From: R. Matthew Emerson <rme@acm.org> To: Andrey G. Grozin <A.G.Grozin@inp.nsk.su> Cc: openmcl-devel Devel <openmcl-devel@clozure.com> Subject: Re: [Openmcl-devel] cannot build ccl-1.11.5 from sources on a 32-bit Gentoo linux box https://github.com/Clozure/ccl/issues/126 is an issue about Gentoo (apparently a Gentoo project that seeks to improve security has a rule that programs mustn’t use absolute addressing). I don’t know if that issue is related, but when I saw the linker warning about the text relocation, I thought of that issue. There’s another 32-bit-only Linux problem that I ran into also, but in this case, the lisp crashes. That’s issue https://github.com/Clozure/ccl/issues/85. Created attachment 557034 [details, diff]
clozurecl-1.11.5-no-pie-32.patch
I suggest fixing CFLAGS/LDFLAGS passing in this package to allow for easier playing with parameters. Attached tapch is enough to make interpreter not spin on ~x86. I think it means that clozurecl has a bug where it does not account for unexpected load address caused by PIE binary. Perhaps it's image embeds absolute addresses? The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=01277b43fb99ba50bf59849c19305ba94783cfe2 commit 01277b43fb99ba50bf59849c19305ba94783cfe2 Author: Andrey Grozin <grozin@gentoo.org> AuthorDate: 2018-12-09 13:35:59 +0000 Commit: Andrey Grozin <grozin@gentoo.org> CommitDate: 2018-12-09 13:35:59 +0000 dev-lisp/clozurecl: fix compilation on x86 with 17.0 profile Suggested-by: Sergei Trofimovich <slyfox@gentoo.org> Closes: https://bugs.gentoo.org/672454 Signed-off-by: Andrey Grozin <grozin@gentoo.org> Package-Manager: Portage-2.3.41, Repoman-2.3.9 dev-lisp/clozurecl/clozurecl-1.11.5.ebuild | 2 ++ dev-lisp/clozurecl/files/clozurecl-1.11.5-no-pie-32.patch | 7 +++++++ 2 files changed, 9 insertions(+) |
make -j3 -C lisp-kernel/linuxx8632 clean make: Entering directory '/var/tmp/portage/dev-lisp/clozurecl-1.11.5/work/ccl/lisp-kernel/linuxx8632' /bin/rm -f pmcl-kernel.o gc-common.o x86-gc.o bits.o x86-exceptions.o x86-utils.o image.o thread_manager.o lisp-debug.o memory.o unix-calls.o x86-asmutils32.o imports.o lispdcmd.o plprint.o plsym.o xlbt.o x86_print.o ../../lx86cl /bin/rm -f pad.o x86-spjump32.o x86-spentry32.o x86-subprims32.o make: Leaving directory '/var/tmp/portage/dev-lisp/clozurecl-1.11.5/work/ccl/lisp-kernel/linuxx8632' make -j3 -C lisp-kernel/linuxx8632 all CC=i686-pc-linux-gnu-gcc make: Entering directory '/var/tmp/portage/dev-lisp/clozurecl-1.11.5/work/ccl/lisp-kernel/linuxx8632' m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../pad.s | as --32 -o pad.o m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../x86-spjump32.s | as --32 -o x86-spjump32.o m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../x86-spentry32.s | as --32 -o x86-spentry32.o m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../x86-subprims32.s | as --32 -o x86-subprims32.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../pmcl-kernel.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o pmcl-kernel.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../gc-common.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o gc-common.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../x86-gc.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o x86-gc.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../bits.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o bits.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../x86-exceptions.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o x86-exceptions.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../x86-utils.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o x86-utils.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../image.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o image.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../thread_manager.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o thread_manager.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../lisp-debug.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o lisp-debug.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../memory.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o memory.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../unix-calls.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o unix-calls.o m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../x86-asmutils32.s | as --32 -o x86-asmutils32.o m4 -DLINUX -DX86 -DX8632 -DHAVE_TLS -I../ ../imports.s | as --32 -o imports.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../lispdcmd.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o lispdcmd.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../plprint.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o plprint.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../plsym.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o plsym.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../xlbt.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o xlbt.o i686-pc-linux-gnu-gcc -include ../platform-linuxx8632.h -c ../x86_print.c -DLINUX -D_REENTRANT -DX86 -DX8632 -D_GNU_SOURCE -DHAVE_TLS -DVC_REVISION="v1.11.5-dirty" -g -O2 -Wno-format -m32 -o x86_print.o i686-pc-linux-gnu-gcc -m32 -g -Wl,--export-dynamic "-Wl,--hash-style=sysv" -o ../../lx86cl pad.o x86-spjump32.o x86-spentry32.o x86-subprims32.o pmcl-kernel.o gc-common.o x86-gc.o bits.o x86-exceptions.o x86-utils.o image.o thread_manager.o lisp-debug.o memory.o unix-calls.o x86-asmutils32.o imports.o lispdcmd.o plprint.o plsym.o xlbt.o x86_print.o -Wl,--no-as-needed -ldl -lm -lpthread -lrt /usr/lib/gcc/i686-pc-linux-gnu/7.3.0/../../../../i686-pc-linux-gnu/bin/ld: x86-spentry32.o: warning: relocation in read-only section `.text' /usr/lib/gcc/i686-pc-linux-gnu/7.3.0/../../../../i686-pc-linux-gnu/bin/ld: warning: creating a DT_TEXTREL in object. make: Leaving directory '/var/tmp/portage/dev-lisp/clozurecl-1.11.5/work/ccl/lisp-kernel/linuxx8632' After that point, lx86cl just runs forever, consuming 100% CPU: PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 15887 portage 20 0 1917616 20048 4032 R 99.7 1.0 5:13.03 lx86cl