![]() ![]() ![]() Jim Wilson had suggested me that I should improve this, and also need toĬhange the ifunc checking rules in the riscv_elf_check_relocs. Has only R_AARCH64_CALL26 for calls rather than two seperate relocations. RISC-V lld had made the same improvement, and so had GNU AARCH64, they But in fact they can (should) be handled identically. For R_RISCV_CALL and R_RISCV_CALL_PLT, GNU GCC and Binutils handle There should be two potential issues currently,ġ. R_RISCV_CALL_PLT, R_RISCV_PCREL_HI20, R_RISCV_GOT_HI20, and Please see the details in the first patch comments. Subject: RISC-V: Enable ifunc attribute by default for RISC-V.ġ file changed, 1 insertion(+), 1 deletion(-)ĭiff -git a/gcc/config.gcc b/gcc/config.gcc However, We have tested and passed the toolchain regressions, including gccĪnd binutils testsuites, and glibc testcases for ifunc support.įrom c7ce689c15cc8b4609abcad476ea330eaff4308f Mon Sep 17 00:00:00 2001 The gcc check will be sent by another patch to gcc mail list. Me that we need to check the bintuils ifunc feature in gcc, in case someoneĪctually uses the ifunc attribute, but they don't have good enough configureĬhecks. Patch should be enough to enable the ifunc supports. The binutils' patch accepted and upstreamed. Vincent Chen have finished the related works, and he is waiting for This patch is used to support GNU ifunc for RISC-V in binutils.Īnd we also have the ifunc supports in the gcc and glibc. To: binutils, jimw, andrew, palmer +Cc: vincent.chen, kito.cheng, i, nelson.chu 18:45 ` Jim Wilson 0 siblings, 2 replies 9+ messages in thread public inbox for help / color / mirror / Atom feed * RISC-V: Support GNU indirect functions. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |