https://bugs.kde.org/show_bug.cgi?id=468575
--- Comment #76 from Mark Wielaard <m...@klomp.org> --- Created attachment 176828 --> https://bugs.kde.org/attachment.cgi?id=176828&action=edit integer.c: replace zero by a1 Simplest seems to be to replace the usage of "zero" in the branch instruction tests by "a1". It seems that does test something similar that isn't tested before. With this patch the integer testcase compiles, links (*) and the passes regtest. Does anybody knowing riscv assembly/instructions know whether the tests against the zero register should also work? And if so, if there is a way to write them in gas assembly? (*) links with binutils ld >= 2.43, with 2.42 or 2.41 it produces: /home/mark/valgrind/none/tests/riscv64/integer.c:81:(.text+0x22cd8): dangerous relocation: The addend isn't allowed for R_RISCV_GOT_HI20 Similar to compress with older binutils. (It also does compiles, links and passes with binutils 2.43.1) -- You are receiving this mail because: You are watching all bug changes.
