From 2c80f2dbdb92d89bf1168c1d4de9c838884170b2 Mon Sep 17 00:00:00 2001 From: Peter Wemm Date: Sat, 26 Oct 1996 17:00:50 +0000 Subject: [PATCH] Apparently this patch fixes the objc backend on the i386 platforms. Without this, compiled programs die with FP errors. This is originally credited to: jlemon@netcom.com (Jonathan Lemon), and has been forwarded to me by quite a few of people. --- contrib/gcc/config/i386/i386.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/contrib/gcc/config/i386/i386.md b/contrib/gcc/config/i386/i386.md index ff0119631cb4..090baa13ef26 100644 --- a/contrib/gcc/config/i386/i386.md +++ b/contrib/gcc/config/i386/i386.md @@ -5312,10 +5312,17 @@ coprocessor registers as containing a possible return value, simply pretend the untyped call returns a complex long double value. */ +#if 1 + /* this may be part of (set (reg: ..) (call_insn ...)), and we can't + directly set a fp register from the call. so we revert to the + old behavior - jlemon@netcom.com (Jonathan Lemon) */ + emit_call_insn (gen_call (operands[0], const0_rtx, NULL, const0_rtx)); +#else emit_call_insn (TARGET_80387 ? gen_call_value (gen_rtx (REG, XCmode, FIRST_FLOAT_REG), operands[0], const0_rtx) : gen_call (operands[0], const0_rtx)); +#endif for (i = 0; i < XVECLEN (operands[2], 0); i++) {