https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112970
--- Comment #1 from Xi Ruoyao <xry111 at gcc dot gnu.org> --- With -mexplicit-relocs=auto the generated code is sub-optimal as well: pcalau12i $r12,%pc_hi20(.LANCHOR0) addi.d $r12,$r12,%pc_lo12(.LANCHOR0) ld.wu $r5,$r12,4 addi.d $r4,$r12,4