https://sourceware.org/bugzilla/show_bug.cgi?id=33267
--- Comment #4 from H.J. Lu <hjl.tools at gmail dot com> --- (In reply to Sam James from comment #2) > (In reply to Sam James from comment #1) > > Some notes: > > * I am new to S-Records. Need to check what they are. > > * Need to check if "Build pr17689 with PIE without -fPIE" (etc) failure is > > OK. Maybe a better error could be given when we see the property in some > > objects. > > BTW (I didn't investigate it yet), but I saw this when building cross > bpf-toolchain (it's a packaged "easy" way of getting bpf-unknown-none): > ``` > /usr/lib/gcc/x86_64-pc-linux-gnu/16/../../../../x86_64-pc-linux-gnu/bin/ld: > /var/tmp/portage/sys-devel/bpf-toolchain-15.2.0_p1/temp/ccSG8m3J.ltrans3. > ltrans.o: warning: relocation against `__environ@@GLIBC_2.2.5' in read-only > section `.text' > /usr/lib/gcc/x86_64-pc-linux-gnu/16/../../../../x86_64-pc-linux-gnu/bin/ld: > read-only segment has dynamic relocations > collect2: error: ld returned 1 exit status > ``` Can you open a separate linker or gcc bug with a testcase? > I think it's because of no --enable-host-pie in that packaging. But it's an > example of where maybe the error could be better. We can see that libc.so > has the property, so we can warn when non-PIE? -- You are receiving this mail because: You are on the CC list for the bug.
