diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2008-09-04 22:59:58 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2008-09-04 22:59:58 +0000 |
| commit | 4a03775777785ef758cff8c0699a6bf571a1f2b9 (patch) | |
| tree | a3e8d527e079840bf5627f2e68f73f901728eb1e /test/CodeGen/X86/fastcc.ll | |
| parent | deafefabae915cf20884fd829d2542f9626786e8 (diff) | |
| download | llvm-4a03775777785ef758cff8c0699a6bf571a1f2b9.tar.gz llvm-4a03775777785ef758cff8c0699a6bf571a1f2b9.tar.bz2 llvm-4a03775777785ef758cff8c0699a6bf571a1f2b9.tar.xz | |
For whatever the reason, x86 CallingConv::Fast (i.e. fastcc) was not passing scalar arguments in registers. This patch defines a new fastcc CC which is slightly different from the FastCall CC. In addition to passing integer arguments in ECX and EDX, it also specify doubles are passed in 8-byte slots which are 8-byte aligned (instead of 4-byte aligned). This avoids a potential performance hazard where doubles span cacheline boundaries.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55807 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/fastcc.ll')
| -rw-r--r-- | test/CodeGen/X86/fastcc.ll | 19 |
1 files changed, 19 insertions, 0 deletions
diff --git a/test/CodeGen/X86/fastcc.ll b/test/CodeGen/X86/fastcc.ll new file mode 100644 index 0000000000..13068bada2 --- /dev/null +++ b/test/CodeGen/X86/fastcc.ll @@ -0,0 +1,19 @@ +; RUN: llvm-as < %s | llc -mtriple=i686-apple-darwin | grep mov | grep ecx | grep 0 +; RUN: llvm-as < %s | llc -mtriple=i686-apple-darwin | grep mov | grep xmm0 | grep 16 + +@d = external global double ; <double*> [#uses=1] +@c = external global double ; <double*> [#uses=1] +@b = external global double ; <double*> [#uses=1] +@a = external global double ; <double*> [#uses=1] + +define i32 @foo() nounwind { +entry: + %0 = load double* @d, align 8 ; <double> [#uses=1] + %1 = load double* @c, align 8 ; <double> [#uses=1] + %2 = load double* @b, align 8 ; <double> [#uses=1] + %3 = load double* @a, align 8 ; <double> [#uses=1] + tail call fastcc void @bar( i32 0, i32 1, i32 2, double 1.000000e+00, double %3, double %2, double %1, double %0 ) nounwind + ret i32 0 +} + +declare fastcc void @bar(i32, i32, i32, double, double, double, double, double) |