Summary: | sys-devel/gcc-6.5.0-r3 - make[2]: *** [Makefile:19716: compare] Error 1` | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Jeroen Roovers (RETIRED) <jer> |
Component: | Current packages | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
Status: | RESOLVED WORKSFORME | ||
Severity: | normal | CC: | hppa, slyfox |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | HPPA | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | sys-devel:gcc-6.5.0-r3:20200716-103433.log.xz |
Description
Jeroen Roovers (RETIRED)
![]() Stage comparison failure is bad. That means stage1 and stage2 don't generate identical code with debugging options flipped. If you still have compiled object files around can you upload them as well? For example the 'gcc/tree-if-conv.o' file. It should be present in a few copies across the build tree. (In reply to Sergei Trofimovich from comment #1) > If you still have compiled object files around can you upload them as well? > For example the 'gcc/tree-if-conv.o' file. It should be present in a few > copies across the build tree. I *just* started another run to see if something random happened the first time, so those files are now lost. I'll see if it happens again and upload those files if it does. OK, so the full story is that during the first sys-devel/gcc-6.5.0-r3 build, the kernel suffered a meltdown somewhere in the SCSI code. The second build attempt that is recorded here merely suffered an internal build discrepancy. The third build attempt finished swimmingly, so I guess the three different kernels I was running may have been affecting the outcome in each case. |