Hｉ：
I generate IR by LLVM, and need compile the IR into the PTX, but, I get this error:
nvvmCompileProgram: Failed
test (2, 0): parse expected top-level entity
thanks for any reply.
this is my IR file:
; ModuleID = ‘test’
source_filename = “test”
target datalayout = “e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v16:16:16-v32:32:32-v64:64:64-v128:128:128-n16:32:64”
target triple = “nvptx64-nvidia-cuda”
; Function Attrs: nounwind readnone
declare i32 @llvm.nvvm.read.ptx.sreg.tid.x() #0
; Function Attrs: nounwind readnone
declare i32 @llvm.nvvm.read.ptx.sreg.ntid.x() #0
; Function Attrs: nounwind readnone
declare i32 @llvm.nvvm.read.ptx.sreg.ctaid() #0
; Function Attrs: inlinehint
define i32 @getThreadID() #1 {
get_thread_id_entry:
%0 = call i32 @llvm.nvvm.read.ptx.sreg.ntid.x()
%1 = call i32 @llvm.nvvm.read.ptx.sreg.ctaid()
%2 = call i32 @llvm.nvvm.read.ptx.sreg.tid.x()
%3 = mul i32 %0, %1
%4 = add i32 %2, %3
ret i32 %4
}
define void @kernel(double* %src, double* %dst, i32 %length) {
entry:
%0 = call i32 @getThreadID()
%1 = icmp slt i32 %0, %length
br i1 %1, label %LoopBB, label %EndBB
LoopBB: ; preds = %entry
%2 = sext i32 %0 to i64
%src_ptr = getelementptr double, double* %src, i64 %2
%src_val = load double, double* %src_ptr, align 8
%3 = fcmp ult double %src_val, 2.000000e+02
%cmp_val = uitofp i1 %3 to double
%dst_val = getelementptr double, double* %dst, i64 %2
store double %cmp_val, double* %dst_val, align 8
br label %EndBB
EndBB: ; preds = %LoopBB, %entry
ret void
}
attributes #0 = { nounwind readnone }
attributes #1 = { inlinehint }
!nvvm.annotations = !{!0}
!0 = !{void (double*, double*, i32)* @kernel, !“kernel”, i32 1}