Created attachment 1390035 [details] preprocessed source Description of problem: We see an ICE when build java-1.8.0-openjdk in Rawhide, https://koji.fedoraproject.org/koji/taskinfo?taskID=24621185 is the failed jon ... // /usr/libexec/gcc/ppc64le-redhat-linux/8/cc1 -quiet -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/include -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/include/linux -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/javavm/export -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/solaris/javavm/export -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/native/common -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/solaris/native/common -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/gensrc_headers -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/native/sun/awt/medialib -I /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/solaris/native/sun/awt/medialib -MMD /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/objs/libmlib_image/mlib_c_ImageThresh1_U8.d -MF /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/objs/libmlib_image/mlib_c_ImageThresh1_U8.d -MQ /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/objs/libmlib_image/mlib_c_ImageThresh1_U8.o -D__unix__ -D__gnu_linux__ -D__linux__ -Dunix -D__unix -Dlinux -D__linux -Asystem=linux -Asystem=unix -Asystem=posix -D _GNU_SOURCE -D _REENTRANT -D _LARGEFILE64_SOURCE -D _LP64=1 -D _LITTLE_ENDIAN -D ABI_ELFv2 -D LINUX -D NDEBUG -D ARCH="ppc64le" -D ppc64le -D RELEASE="1.8.0_161" -D __USE_J2D_NAMES -D __MEDIALIB_OLD_NAMES -D MLIB_NO_LIBSUNMATH -D MLIB_OS64BIT -D THIS_FILE="mlib_c_ImageThresh1_U8.c" -D_FORTIFY_SOURCE=2 -D_GLIBCXX_ASSERTIONS /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/native/sun/awt/medialib/mlib_c_ImageThresh1_U8.c -msecure-plt -quiet -dumpbase mlib_c_ImageThresh1_U8.c -m64 -mcpu=power8 -mtune=power8 -auxbase-strip /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/build/jdk8.build/jdk/objs/libmlib_image/mlib_c_ImageThresh1_U8.o -g -grecord-gcc-switches -g -O3 -Wextra -Wall -Wno-unused -Wno-parentheses -Wformat=1 -Wno-cpp -Werror=format-security -Wno-error -std=gnu++98 -fstack-protector-strong -fstack-clash-protection -fno-strict-aliasing -fno-delete-null-pointer-checks -fno-lifetime-dse -fPIC -iplugindir=/usr/lib/gcc/ppc64le-redhat-linux/8/plugin -fplugin=annobin -o - -frandom-seed=0 -fdump-noaddr // /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/native/sun/awt/medialib/mlib_c_ImageThresh1_U8.c: In function 'mlib_c_ImageThresh1_U83': // /home/sharkcz/java-1.8.0-openjdk/java-1.8.0-openjdk-1.8.0.161-5.b14.fc28.ppc64le/openjdk/jdk/src/share/native/sun/awt/medialib/mlib_c_ImageThresh1_U8.c:270:1: internal compiler error: in lra_set_insn_recog_data, at lra.c:998 // } // ^ // Please submit a full bug report, // with preprocessed source if appropriate. Version-Release number of selected component (if applicable): gcc-8.0.1-0.6.fc28.ppc64le How reproducible: 100% Steps to Reproduce: 1. fedpkg co java-1.8.0-openjdk 2. fedpkg local Additional info: It looks to me as https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83399 but the fix is supposed to be included in gcc-8.0.1-0.6.
Seems it's not a problem with gcc-8.0.1-0.9.fc28.ppc64le any more, although the upstream bug is not mentioned in the changelog.
(In reply to Dan Horák from comment #1) > Seems it's not a problem with gcc-8.0.1-0.9.fc28.ppc64le any more, although > the upstream bug is not mentioned in the changelog. Thus closing based on this comment.