| Summary: | sys-boot/sgibootcd does not respect LDFLAGS | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Agostino Sarubbo <ago> |
| Component: | Current packages | Assignee: | MIPS Porters <mips> |
| Status: | RESOLVED INVALID | ||
| Severity: | normal | CC: | esigra |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 331933 | ||
| Attachments: | build.log | ||
Created attachment 642300 [details]
build.log
build log and emerge --info
This package is for an extremely small utility only used on SGI Boot CDs, which behave more like disk partitions burned to optical media than classic ISO9660 format. The code is known to have stylistic issues and needs to be refactored, but it's far down on my TODO list to whenever I get around to attempting bootable SGI media again. Doubled back on this bug and changing the resolution to INVALID because the current ebuild only invokes the C compiler to compile a single .c file straight to an executable. The external linker is not involved at all, except as an internal stage within gcc's invocation. The only way to pass any linker flags would be via -Wl,* syntax, and I don't know of a way to cleanly translate existing LDFLAGS to CFLAGS -Wl,* equivalents. I did work on a 0.13 version that fixes the style issues and adds missing includes to clear up the compiler warnings, however. I'll push that out in the next few days. Though eventually, I may tackle re-writing the code entirely. |
This is an auto-filed bug because sys-boot/sgibootcd does not respect LDFLAGS. The issue was originally discovered on amd64, but it may be reproducible on other arches as well. If you think that a different summary clarifies the issue better, feel free to change it. Attached build log and emerge --info. NOTE: To reproduce this issue you may want to set LDFLAGS="${LDFLAGS} -Wl,--defsym=__gentoo_check_ldflags__=0". If this is not something about c/c++ context you may want to see this bug as an hint to hide the QA warning where is not possible to respect LDFLAGS.