make[1]: *** [Makefile:24: mmix-arith.c] Error 20 make[1]: Leaving directory '/var/tmp/portage/dev-lang/mmix-20160804/work' make: *** [Makefile:30: mmix-arith.o] Error 2 * ERROR: dev-lang/mmix-20160804::gentoo failed (compile phase): * emake failed * ------------------------------------------------------------------- This is an unstable amd64 chroot image at a tinderbox (==build bot) name: 17.1-desktop_libressl-test_20190602-225504 ------------------------------------------------------------------- gcc-config -l: [1] x86_64-pc-linux-gnu-8.3.1 [2] x86_64-pc-linux-gnu-9.1.0 * Available Python interpreters, in order of preference: [1] python3.7 [2] python3.6 [3] python2.7 (fallback) Available Rust versions: [1] rust-bin-1.35.0 * [2] rust-1.35.0 java-config: The following VMs are available for generation-2: *) IcedTea JDK 3.12.0 [icedtea-bin-8] Available Java Virtual Machines: [1] icedtea-bin-8 system-vm emerge -qpvO dev-lang/mmix [ebuild N ] dev-lang/mmix-20160804 USE="-doc"
Created attachment 579224 [details] emerge-info.txt
Created attachment 579226 [details] dev-lang:mmix-20160804:20190608-162736.log
Created attachment 579228 [details] emerge-history.txt
Created attachment 579230 [details] environment
Created attachment 579232 [details] etc.portage.tbz2
Created attachment 579234 [details] logs.tbz2
That looks unusual. Probably a missing depend on fonts. Is it reproducible for you? >>> Compiling source in /var/tmp/portage/dev-lang/mmix-20160804/work ... make -j1 all 'CFLAGS=-O2 -pipe -march=native' CC=x86_64-pc-linux-gnu-gcc make mmix-arith.c make[1]: Entering directory '/var/tmp/portage/dev-lang/mmix-20160804/work' if test -r mmix-arith.ch; then ctangle mmix-arith.w mmix-arith.ch; else ctangle mmix-arith.w; fi warning: kpathsea: configuration file texmf.cnf not found in these directories: /usr/bin:/usr/bin/share/texmf-local/web2c:/usr/bin/share/texmf-dist/web2c:/usr/bin/share/texmf/web2c:/usr/bin/texmf-local/web2c:/usr/bin/texmf-dist/web2c:/usr/bin/texmf/web2c:/usr:/usr/share/texmf-local/web2c:/usr/share/texmf-dist/web2c:/usr/share/texmf/web2c:/usr/texmf-local/web2c:/usr/texmf-dist/web2c:/usr/texmf/web2c://texmf-local/web2c:/://share/texmf-local/web2c://share/texmf-dist/web2c://share/texmf/web2c://texmf-local/web2c://texmf-dist/web2c://texmf/web2c. ! Cannot open input file mmix-arith.w(That was a fatal error, my friend.) make[1]: *** [Makefile:24: mmix-arith.c] Error 20 make[1]: Leaving directory '/var/tmp/portage/dev-lang/mmix-20160804/work' make: *** [Makefile:30: mmix-arith.o] Error 2
(In reply to Sergei Trofimovich from comment #7) > That looks unusual. Probably a missing depend on fonts. Is it reproducible > for you? currently only at this particular tinderbox image.
I wonder why it complains about failure to open file: > ! Cannot open input file mmix-arith.w(That was a fatal error, my friend.) Could it be some out-of-disk-space condition?
(In reply to Sergei Trofimovich from comment #9) > I wonder why it complains about failure to open file: > > ! Cannot open input file mmix-arith.w(That was a fatal error, my friend.) > > Could it be some out-of-disk-space condition? well, I#d say no, due to: 1.6 TB fs HDD, 16 GB tmpfs for /var/tmp/portage/
Can you check $WORKDIR for presence of mmix-arith.w? If you can try to run 'ctangle mmix-arith.w mmix-arith.ch' and see if it still fails. We might get some clue from strace log. You can get one as: $ strace -f -otrace.log ctangle mmix-arith.w mmix-arith.ch
(In reply to Sergei Trofimovich from comment #11) > Can you check $WORKDIR for presence of mmix-arith.w? > If you can try to run 'ctangle mmix-arith.w mmix-arith.ch' and see if it > still fails. > > We might get some clue from strace log. You can get one as: > $ strace -f -otrace.log ctangle mmix-arith.w mmix-arith.ch mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # find . -name mmix-arith.w ./mmix-arith.w mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # ls -l ./mmix-arith.w -rw-r--r-- 1 portage portage 59767 Aug 4 2015 ./mmix-arith.w mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # file ./mmix-arith.w ./mmix-arith.w: C source, ASCII text mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # strace -f -o /tmp/trace.log ctangle mmix-arith.w mmix-arith.ch warning: kpathsea: configuration file texmf.cnf not found in these directories: /usr/bin:/usr/bin/share/texmf-local/web2c:/usr/bin/share/texmf-dist/web2c:/usr/bin/share/texmf/web2c:/usr/bin/texmf-local/web2c:/usr/bin/texmf-dist/web2c:/usr/bin/texmf/web2c:/usr:/usr/share/texmf-local/web2c:/usr/share/texmf-dist/web2c:/usr/share/texmf/web2c:/usr/texmf-local/web2c:/usr/texmf-dist/web2c:/usr/texmf/web2c://texmf-local/web2c:/://share/texmf-local/web2c://share/texmf-dist/web2c://share/texmf/web2c://texmf-local/web2c://texmf-dist/web2c://texmf/web2c. ! Cannot open input file mmix-arith.w(That was a fatal error, my friend.) mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # ls -l /tmp/trace.log -rw-r--r-- 1 root root 6478 Jun 15 16:22 /tmp/trace.log mr-fox /var/tmp/portage/dev-lang/mmix-20160804/work # cat /tmp/trace.log 30257 execve("/usr/bin/ctangle", ["ctangle", "mmix-arith.w", "mmix-arith.ch"], 0x7ffcf64d1de8 /* 68 vars */) = 0 30257 brk(NULL) = 0x55c7ea876000 30257 access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory) 30257 openat(AT_FDCWD, "/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3 30257 fstat(3, {st_mode=S_IFREG|0644, st_size=310363, ...}) = 0 30257 mmap(NULL, 310363, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7ffa89aaa000 30257 close(3) = 0 30257 openat(AT_FDCWD, "/usr/lib64/libkpathsea.so.6", O_RDONLY|O_CLOEXEC) = 3 30257 read(3, "\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0\360I\0\0\0\0\0\0"..., 832) = 832 30257 fstat(3, {st_mode=S_IFREG|0755, st_size=97392, ...}) = 0 30257 mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7ffa89aa8000 30257 mmap(NULL, 113720, PROT_READ, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7ffa89a8c000 30257 mmap(0x7ffa89a90000, 57344, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x4000) = 0x7ffa89a90000 30257 mmap(0x7ffa89a9e000, 20480, PROT_READ, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x12000) = 0x7ffa89a9e000 30257 mmap(0x7ffa89aa3000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x16000) = 0x7ffa89aa3000 30257 mmap(0x7ffa89aa5000, 11320, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7ffa89aa5000 30257 close(3) = 0 30257 openat(AT_FDCWD, "/lib64/libc.so.6", O_RDONLY|O_CLOEXEC) = 3 30257 read(3, "\177ELF\2\1\1\3\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0000@\2\0\0\0\0\0"..., 832) = 832 30257 fstat(3, {st_mode=S_IFREG|0755, st_size=1873368, ...}) = 0 30257 mmap(NULL, 1886056, PROT_READ, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7ffa898bf000 30257 mmap(0x7ffa898e1000, 1404928, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x22000) = 0x7ffa898e1000 30257 mmap(0x7ffa89a38000, 303104, PROT_READ, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x179000) = 0x7ffa89a38000 30257 mmap(0x7ffa89a82000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1c2000) = 0x7ffa89a82000 30257 mmap(0x7ffa89a88000, 14184, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7ffa89a88000 30257 close(3) = 0 30257 mmap(NULL, 12288, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7ffa898bc000 30257 arch_prctl(ARCH_SET_FS, 0x7ffa898bc740) = 0 30257 mprotect(0x7ffa89a82000, 16384, PROT_READ) = 0 30257 mprotect(0x7ffa89aa3000, 4096, PROT_READ) = 0 30257 mprotect(0x55c7e9701000, 4096, PROT_READ) = 0 30257 mprotect(0x7ffa89b1d000, 4096, PROT_READ) = 0 30257 munmap(0x7ffa89aaa000, 310363) = 0 30257 brk(NULL) = 0x55c7ea876000 30257 brk(0x55c7ea897000) = 0x55c7ea897000 30257 stat("/usr/lib/llvm/8/bin/ctangle", 0x7fff778ab290) = -1 ENOENT (No such file or directory) 30257 stat("/usr/local/sbin/ctangle", 0x7fff778ab290) = -1 ENOENT (No such file or directory) 30257 stat("/usr/local/bin/ctangle", 0x7fff778ab290) = -1 ENOENT (No such file or directory) 30257 stat("/usr/sbin/ctangle", 0x7fff778ab290) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/ctangle", {st_mode=S_IFREG|0755, st_size=52072, ...}) = 0 30257 lstat("/usr", {st_mode=S_IFDIR|0755, st_size=212, ...}) = 0 30257 lstat("/usr/bin", {st_mode=S_IFDIR|0755, st_size=130552, ...}) = 0 30257 lstat("/usr/bin/ctangle", {st_mode=S_IFREG|0755, st_size=52072, ...}) = 0 30257 stat("/usr/bin", {st_mode=S_IFDIR|0755, st_size=130552, ...}) = 0 30257 access("/usr/bin/texmf.cnf", R_OK) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/share/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/share/texmf-dist/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/share/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/texmf-dist/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/bin/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr", {st_mode=S_IFDIR|0755, st_size=212, ...}) = 0 30257 access("/usr/texmf.cnf", R_OK) = -1 ENOENT (No such file or directory) 30257 stat("/usr/share/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/share/texmf-dist/web2c", {st_mode=S_IFDIR|0755, st_size=688, ...}) = 0 30257 access("/usr/share/texmf-dist/web2c/texmf.cnf", R_OK) = -1 ENOENT (No such file or directory) 30257 stat("/usr/share/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/texmf-dist/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/usr/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/", {st_mode=S_IFDIR|0755, st_size=170, ...}) = 0 30257 access("/texmf.cnf", R_OK) = -1 ENOENT (No such file or directory) 30257 stat("/share/texmf-local/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/share/texmf-dist/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/share/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/texmf-dist/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 stat("/texmf/web2c", 0x7fff778ab120) = -1 ENOENT (No such file or directory) 30257 write(2, "warning: ", 9) = 9 30257 write(2, "kpathsea: configuration file tex"..., 538) = 538 30257 write(2, ".\n", 2) = 2 30257 stat("/nonesuch", 0x7fff778ab070) = -1 ENOENT (No such file or directory) 30257 mmap(NULL, 516096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7ffa8983e000 30257 munmap(0x7ffa8983e000, 516096) = 0 30257 openat(AT_FDCWD, "mmix-arith.ttp", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 3 30257 fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(0x88, 0x7), ...}) = 0 30257 write(1, "! Cannot open input file mmix-ar"..., 37) = 37 30257 write(1, "(That was a fatal error, my frie"..., 37) = 37 30257 close(3) = 0 30257 unlink("mmix-arith.ttp") = 0 30257 exit_group(20) = ? 30257 +++ exited with 20 +++
Aha, that is useful! I think our culprit is missing texmf.cnf: 30257 access("/usr/share/texmf-dist/web2c/texmf.cnf", R_OK) = -1 ENOENT (No such file or directory) I have it on my system: $ LANG=C ls -lh /usr/share/texmf-dist/web2c/texmf.cnf lrwxrwxrwx 1 root root 26 Jun 1 03:05 /usr/share/texmf-dist/web2c/texmf.cnf -> /etc/texmf/web2c/texmf.cnf That file shouldhave came from dev-libs/kpathsea. All ebuilds are supposed to install the symlink: TEXMF_PATH=/usr/share/texmf-dist ... src_install() { ... dosym /etc/texmf/web2c/texmf.cnf ${TEXMF_PATH}/web2c/texmf.cnf } Do you both that symlink and the file it points to? Other related note: emerge-history.log broke with new portage-utils: qlop: unrecognized option '--list' Usage: qlop [opts] <pkgname> : emerge log analyzer The change was intentional :(.
(In reply to Sergei Trofimovich from comment #13) > Aha, that is useful! > > I think our culprit is missing texmf.cnf: > > 30257 access("/usr/share/texmf-dist/web2c/texmf.cnf", R_OK) = -1 ENOENT (No > such file or directory) > > I have it on my system: > > $ LANG=C ls -lh /usr/share/texmf-dist/web2c/texmf.cnf > lrwxrwxrwx 1 root root 26 Jun 1 03:05 /usr/share/texmf-dist/web2c/texmf.cnf > -> /etc/texmf/web2c/texmf.cnf > > That file shouldhave came from dev-libs/kpathsea. All ebuilds are supposed > to install the symlink: > TEXMF_PATH=/usr/share/texmf-dist > ... > src_install() { > ... > dosym /etc/texmf/web2c/texmf.cnf ${TEXMF_PATH}/web2c/texmf.cnf > } > > Do you both that symlink and the file it points to? Ick, a broken symlink: mr-fox ~ # LANG=C ls -lh /usr/share/texmf-dist/web2c/texmf.cnf lrwxrwxrwx 1 root root 26 Jun 4 05:44 /usr/share/texmf-dist/web2c/texmf.cnf -> /etc/texmf/web2c/texmf.cnf mr-fox ~ # ls -l /etc/texmf/web2c/texmf.cnf ls: cannot access '/etc/texmf/web2c/texmf.cnf': No such file or directory but why? > > Other related note: emerge-history.log broke with new portage-utils: > qlop: unrecognized option '--list' > Usage: qlop [opts] <pkgname> : emerge log analyzer > The change was intentional :(. Yes, I badly handled stable+unstable version, hopefully fixed.
(In reply to Toralf Förster from comment #14) > (In reply to Sergei Trofimovich from comment #13) > > Aha, that is useful! > > > > I think our culprit is missing texmf.cnf: > > > > 30257 access("/usr/share/texmf-dist/web2c/texmf.cnf", R_OK) = -1 ENOENT (No > > such file or directory) > > > > I have it on my system: > > > > $ LANG=C ls -lh /usr/share/texmf-dist/web2c/texmf.cnf > > lrwxrwxrwx 1 root root 26 Jun 1 03:05 /usr/share/texmf-dist/web2c/texmf.cnf > > -> /etc/texmf/web2c/texmf.cnf > > > > That file shouldhave came from dev-libs/kpathsea. All ebuilds are supposed > > to install the symlink: > > TEXMF_PATH=/usr/share/texmf-dist > > ... > > src_install() { > > ... > > dosym /etc/texmf/web2c/texmf.cnf ${TEXMF_PATH}/web2c/texmf.cnf > > } > > > > Do you both that symlink and the file it points to? > > > Ick, a broken symlink: > > mr-fox ~ # LANG=C ls -lh /usr/share/texmf-dist/web2c/texmf.cnf > lrwxrwxrwx 1 root root 26 Jun 4 05:44 /usr/share/texmf-dist/web2c/texmf.cnf > -> /etc/texmf/web2c/texmf.cnf > mr-fox ~ # ls -l /etc/texmf/web2c/texmf.cnf > ls: cannot access '/etc/texmf/web2c/texmf.cnf': No such file or directory > > > but why? Probably a bug in dev-libs/kpathsea ebuild (or install process). Do you have a build.log of that package building? I wonder if just reinstalling it will repair things for you.
(In reply to Sergei Trofimovich from comment #15) > Probably a bug in dev-libs/kpathsea ebuild (or install process). Do you have > a build.log of that package building? I wonder if just reinstalling it will > repair things for you. indeed: tinderbox@mr-fox ~ $ chr img/17.1-desktop_libressl-test_20190602-225504 mr-fox ~ # emerge -qpvO dev-libs/kpathsea [ebuild R ] dev-libs/kpathsea-6.3.1_p20190410 USE="-doc -source -static-libs" mr-fox ~ # qcheck dev-libs/kpathsea Checking dev-libs/kpathsea-6.3.1_p20190410 ... * 134 out of 134 files are good mr-fox ~ # ls /var/log/portage/*kpathsea* /var/log/portage/dev-libs:kpathsea-6.3.1_p20190410:20190604-034323.log mr-fox ~ # grep texmf.cnf /var/log/portage/dev-libs:kpathsea-6.3.1_p20190410:20190604-034323.log echo "/* paths.h: Generated from texmf.cnf. */" >paths.tmp gawk -f ./bsnl.awk ./texmf.cnf \ /usr/lib/portage/python3.6/ebuild-helpers/xattr/install -c -m 644 './texmf.cnf' '/var/tmp/portage/dev-libs/kpathsea-6.3.1_p20190410/image/usr/share/texmf-dist/web2c/texmf.cnf' * QA Notice: Symbolic link /usr/share/texmf-dist/web2c/texmf.cnf points to /etc/texmf/web2c/texmf.cnf which does not exist. >>> /usr/share/texmf-dist/web2c/texmf.cnf -> /etc/texmf/web2c/texmf.cnf mr-fox ~ # exitexitz -su: exitexitz: command not found mr-fox ~ # exit logout tinderbox@mr-fox ~ $
Let's reassign to dev-libs/kpathsea maintainers. Maybe it's a known problem. Also +dev-portage@ if it's remnant of some temporary broken emerge behaviour.
(In reply to Sergei Trofimovich from comment #17) > Let's reassign to dev-libs/kpathsea maintainers. Maybe it's a known problem. > Also +dev-portage@ if it's remnant of some temporary broken emerge behaviour. This is expected behaviour as kpathsea does not install texmf.cnf file itself, we ship texmf.cnf files logically devided by parts[1] and then texmf.cnf is being generated on next `texmf-update` call [1] - https://gitweb.gentoo.org/proj/tex.git/tree/2019/texmf.d
(In reply to Mikle Kolyada from comment #18) > (In reply to Sergei Trofimovich from comment #17) > > Let's reassign to dev-libs/kpathsea maintainers. Maybe it's a known problem. > > Also +dev-portage@ if it's remnant of some temporary broken emerge behaviour. > > This is expected behaviour as kpathsea does not install texmf.cnf file > itself, we ship texmf.cnf files logically devided by parts[1] and then > texmf.cnf is being generated on next `texmf-update` call > > [1] - https://gitweb.gentoo.org/proj/tex.git/tree/2019/texmf.d What is expected behaviour? kpathsea installing dangling symlink that is not satisfied after emerge succeeds? Please don't drop maintainer from CC. Why reassign to me? What action do you expect from dev-lang/mmix maintainer?
my personal thoughts: there is the deptree: dev-lang/mmix -> virtual/tex-base -> app-text/texlive-core -> dev-libs/kpathsea | calls etexmf-update which brings texmf.cnf to live So I think the bug is invalid as by the time dev-lang/mmix compilation is in action all should be in place.
(In reply to Sergei Trofimovich from comment #19) > (In reply to Mikle Kolyada from comment #18) > > (In reply to Sergei Trofimovich from comment #17) > > > Let's reassign to dev-libs/kpathsea maintainers. Maybe it's a known problem. > > > Also +dev-portage@ if it's remnant of some temporary broken emerge behaviour. > > > > This is expected behaviour as kpathsea does not install texmf.cnf file > > itself, we ship texmf.cnf files logically devided by parts[1] and then > > texmf.cnf is being generated on next `texmf-update` call > > > > [1] - https://gitweb.gentoo.org/proj/tex.git/tree/2019/texmf.d > > What is expected behaviour? kpathsea installing dangling symlink that is not > satisfied after emerge succeeds? > > Please don't drop maintainer from CC. > > Why reassign to me? What action do you expect from dev-lang/mmix maintainer? please check the tex alias.
(In reply to Sergei Trofimovich from comment #19) > (In reply to Mikle Kolyada from comment #18) > > (In reply to Sergei Trofimovich from comment #17) > > > Let's reassign to dev-libs/kpathsea maintainers. Maybe it's a known problem. > > > Also +dev-portage@ if it's remnant of some temporary broken emerge behaviour. > > > > This is expected behaviour as kpathsea does not install texmf.cnf file > > itself, we ship texmf.cnf files logically devided by parts[1] and then > > texmf.cnf is being generated on next `texmf-update` call > > > > [1] - https://gitweb.gentoo.org/proj/tex.git/tree/2019/texmf.d > > What is expected behaviour? kpathsea installing dangling symlink that is not > satisfied after emerge succeeds? > > Please don't drop maintainer from CC. > > Why reassign to me? What action do you expect from dev-lang/mmix maintainer? kpathsea itself is irrelevant, this is being installed before texlive-core, and all the generation happens after kpathsea and symlink are installed but before dev-lang/mmix
(In reply to Mikle Kolyada from comment #20) > my personal thoughts: > > there is the deptree: > > dev-lang/mmix -> virtual/tex-base -> app-text/texlive-core -> > dev-libs/kpathsea > | > calls etexmf-update > which brings texmf.cnf > to live > > So I think the bug is invalid Once again, what action do you expect from dev-lang/mmix maintainer by reassigning it to me? I am very confused. > as by the time dev-lang/mmix compilation is in action all should be in place. Clearly it's not what happened in the original bug report. #comment12 has debugging details. Whichever package was responsible to generation of incorrect end state failed should be tracked down and fixed. If you disagree please state it explicitly. Not just reassign it to me for the third time. > > What is expected behaviour? kpathsea installing dangling symlink that is not > > satisfied after emerge succeeds? > > > > Please don't drop maintainer from CC. > > > > Why reassign to me? What action do you expect from dev-lang/mmix maintainer? > > please check the tex alias. Please specify which package's metadata.xml you are talking about. dev-libs/kpathsea lists aballier@gentoo.org.
(In reply to Sergei Trofimovich from comment #23) > (In reply to Mikle Kolyada from comment #20) > > my personal thoughts: > > > > there is the deptree: > > > > dev-lang/mmix -> virtual/tex-base -> app-text/texlive-core -> > > dev-libs/kpathsea > > | > > calls etexmf-update > > which brings texmf.cnf > > to live > > > > So I think the bug is invalid > > Once again, what action do you expect from dev-lang/mmix maintainer by > reassigning it to me? I am very confused. You are a package maintainer and there is nothing the tex team can help you with, so this is up to you to decide what to do. As I said, this looks invalid to me. > > > as by the time dev-lang/mmix compilation is in action all should be in place. > > Clearly it's not what happened in the original bug report. #comment12 has > debugging details. Whichever package was responsible to generation of > incorrect end state failed should be tracked down and fixed. well, I explained how texmf.cnf appeared in c#18, you can make sure yourself. Even more dev-lang/mmix-20160804 is in the tree since 2017-01-26 which is 2,5 years for now (almost), kpathsea behaviour regarding configuration has not been a subject of change since then, I believe you yorself would catch this bug earlier if this were the case. > > If you disagree please state it explicitly. Not just reassign it to me for > the third time. > > > > What is expected behaviour? kpathsea installing dangling symlink that is not > > > satisfied after emerge succeeds? > > > > > > Please don't drop maintainer from CC. > > > > > > Why reassign to me? What action do you expect from dev-lang/mmix maintainer? > > > > please check the tex alias. > > Please specify which package's metadata.xml you are talking about. > dev-libs/kpathsea lists aballier@gentoo.org. well, abalier is also tex@ alias member just exists in metadata explicitly, but as you wish :)
(In reply to Mikle Kolyada from comment #24) > (In reply to Sergei Trofimovich from comment #23) > > (In reply to Mikle Kolyada from comment #20) > > > my personal thoughts: > > > > > > there is the deptree: > > > > > > dev-lang/mmix -> virtual/tex-base -> app-text/texlive-core -> > > > dev-libs/kpathsea > > > | > > > calls etexmf-update > > > which brings texmf.cnf > > > to live > > > > > > So I think the bug is invalid > > > > Once again, what action do you expect from dev-lang/mmix maintainer by > > reassigning it to me? I am very confused. > > You are a package maintainer and there is nothing the tex team can help you > with, so this is up to you to decide what to do. As I said, this looks > invalid to me. Ok, that's clear. > > > as by the time dev-lang/mmix compilation is in action all should be in place. > > > > Clearly it's not what happened in the original bug report. #comment12 has > > debugging details. Whichever package was responsible to generation of > > incorrect end state failed should be tracked down and fixed. > > well, I explained how texmf.cnf appeared in c#18, you can make sure yourself. > Even more dev-lang/mmix-20160804 is in the tree since 2017-01-26 which is > 2,5 years for now (almost), kpathsea behaviour regarding configuration has > not been a subject of change since then, I believe you yorself would catch > this bug earlier if this were the case. I don't do fresh installs regularly. I expect this bug has special affinity to break on first install. Toralf, assuming app-text/texlive-core was supposed to work and did not do you have build/install logs of initial install of app-text/texlive-core on that chroot?
Created attachment 580410 [details] app-text:texlive-core-2019-r1:20190604-060016.log.gz -rw-rw---- 1 portage portage 2013115 Jun 4 08:17 img/17.1-desktop_libressl-test_20190602-225504/var/log/portage/app-text:texlive-core-2019-r1:20190604-060016.log
I have just taken clear chroot and compiled mmix-20160804 with tl-core-2019, no failures ocurred.
(In reply to Toralf Förster from comment #26) > Created attachment 580410 [details] > app-text:texlive-core-2019-r1:20190604-060016.log.gz > > -rw-rw---- 1 portage portage 2013115 Jun 4 08:17 > img/17.1-desktop_libressl-test_20190602-225504/var/log/portage/app-text: > texlive-core-2019-r1:20190604-060016.log literally > * Cannot run texmf-update for some reason. > * Your texmf tree might be inconsistent with your configuration > * Please try to figure what has happened This is the problem. Also you may want to emerge --sync. The latest tl-core version is 2019-r4 as this got some fixes, (including eclass fixes).
So it was an actual bug in texlive-common.eclass and was fixed as https://gitweb.gentoo.org/repo/gentoo.git/commit/eclass/texlive-common.eclass?id=c321399add5dac4278bb7f7d52102dc34828036f
