bpf: add " ll" in the LD_IMM64 asmstring

This partially revert previous fix in commit f5858045aa0b
("bpf: proper print imm64 expression in inst printer").

In that commit, the original suffix "ll" is removed from
LD_IMM64 asmstring. In the customer print method, the "ll"
suffix is printed if the rhs is an immediate. For example,
"r2 = 5ll" => "r2 = 5ll", and "r3 = varll" => "r3 = var".

This has an issue though for assembler. Since assembler
relies on asmstring to do pattern matching, it will not
be able to distiguish between "mov r2, 5" and
"ld_imm64 r2, 5" since both asmstring is "r2 = 5".
In such cases, the assembler uses 64bit load for all
"r = <val>" asm insts.

This patch adds back " ll" suffix for ld_imm64 with one
additional space for "#reg = #global_var" case.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
llvm-svn: 312978

llvm/lib/Target/BPF/BPFInstrInfo.td

@@ -249,7 +249,7 @@ class MOV_RI<string OpcodeStr>
 
 class LD_IMM64<bits<4> Pseudo, string OpcodeStr>
     : InstBPF<(outs GPR:$dst), (ins u64imm:$imm),
-              "$dst "#OpcodeStr#" ${imm}",
+              "$dst "#OpcodeStr#" ${imm} ll",
               [(set GPR:$dst, (i64 imm:$imm))]> {
 
   bits<3> mode;

llvm/lib/Target/BPF/InstPrinter/BPFInstPrinter.cpp

@@ -88,7 +88,7 @@ void BPFInstPrinter::printImm64Operand(const MCInst *MI, unsigned OpNo,
                                        raw_ostream &O) {
   const MCOperand &Op = MI->getOperand(OpNo);
   if (Op.isImm())
-    O << (uint64_t)Op.getImm() << "ll";
+    O << (uint64_t)Op.getImm();
   else if (Op.isExpr())
     printExpr(Op.getExpr(), O);
   else

llvm/test/CodeGen/BPF/cc_args.ll

@@ -12,7 +12,7 @@ entry:
 ; CHECK: call f_i32
   call void @f_i32(i32 12345678)
 
-; CHECK: r1 = 72623859790382856ll # encoding: [0x18,0x01,0x00,0x00,0x08,0x07,0x06,0x05,0x00,0x00,0x00,0x00,0x04,0x03,0x02,0x01]
+; CHECK: r1 = 72623859790382856 ll # encoding: [0x18,0x01,0x00,0x00,0x08,0x07,0x06,0x05,0x00,0x00,0x00,0x00,0x04,0x03,0x02,0x01]
 ; CHECK: call f_i64
   call void @f_i64(i64 72623859790382856)
 
@@ -28,7 +28,7 @@ entry:
   call void @f_i16_i32_i16(i16 2, i32 3, i16 4)
 
 ; CHECK: r1 = 5
-; CHECK: r2 = 7262385979038285ll
+; CHECK: r2 = 7262385979038285 ll
 ; CHECK: r3 = 6
 ; CHECK: call f_i16_i64_i16
   call void @f_i16_i64_i16(i16 5, i64 7262385979038285, i16 6)

llvm/test/CodeGen/BPF/cc_args_be.ll

@@ -13,7 +13,7 @@ entry:
 ; CHECK: call f_i32
   call void @f_i32(i32 12345678)
 
-; CHECK: r1 = 72623859790382856ll # encoding: [0x18,0x10,0x00,0x00,0x05,0x06,0x07,0x08,0x00,0x00,0x00,0x00,0x01,0x02,0x03,0x04]
+; CHECK: r1 = 72623859790382856 ll # encoding: [0x18,0x10,0x00,0x00,0x05,0x06,0x07,0x08,0x00,0x00,0x00,0x00,0x01,0x02,0x03,0x04]
 ; CHECK: call f_i64
   call void @f_i64(i64 72623859790382856)
 
@@ -29,7 +29,7 @@ entry:
   call void @f_i16_i32_i16(i16 2, i32 3, i16 4)
 
 ; CHECK: r1 = 5
-; CHECK: r2 = 7262385979038285ll
+; CHECK: r2 = 7262385979038285 ll
 ; CHECK: r3 = 6
 ; CHECK: call f_i16_i64_i16
   call void @f_i16_i64_i16(i16 5, i64 7262385979038285, i16 6)

llvm/test/CodeGen/BPF/cc_ret.ll

@@ -42,7 +42,7 @@ define i32 @f_i32() #0 {
 
 define i64 @f_i64() #0 {
 ; CHECK: f_i64:
-; CHECK: r0 = 72623859790382856ll
+; CHECK: r0 = 72623859790382856 ll
 ; CHECK: exit
   ret i64 72623859790382856
 }

llvm/test/CodeGen/BPF/ex1.ll

@@ -32,7 +32,7 @@ define i32 @bpf_prog1(%struct.bpf_context* nocapture %ctx) #0 section "events/ne
 ; CHECK: call 9
 ; CHECK: if r0 != 0
 ; CHECK: r1 = 622884453
-; CHECK: r1 = 7214898703899978611ll
+; CHECK: r1 = 7214898703899978611 ll
 ; CHECK: call 11
 ; CHECK: r0 = 0
 ; CHECK: exit

llvm/test/CodeGen/BPF/intrinsics.ll

@@ -55,7 +55,7 @@ entry:
   tail call void inttoptr (i64 4 to void (i64, i32)*)(i64 %call, i32 4) #2
   ret i32 0
 ; CHECK-LABEL: ld_pseudo:
-; CHECK: ld_pseudo r1, 2, 3ll # encoding: [0x18,0x21,0x00,0x00,0x03,0x00
+; CHECK: ld_pseudo r1, 2, 3 # encoding: [0x18,0x21,0x00,0x00,0x03,0x00
 }
 
 declare i64 @llvm.bpf.pseudo(i64, i64) #2

llvm/test/CodeGen/BPF/mem_offset_be.ll

@@ -2,7 +2,7 @@
 
 ; Function Attrs: nounwind
 define i32 @bpf_prog1(i8* nocapture readnone) local_unnamed_addr #0 {
-; CHECK: r1 = 590618314553ll   # encoding: [0x18,0x10,0x00,0x00,0x83,0x98,0x47,0x39,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x89]
+; CHECK: r1 = 590618314553 ll   # encoding: [0x18,0x10,0x00,0x00,0x83,0x98,0x47,0x39,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x89]
 ; CHECK: r1 += -1879113726     # encoding: [0x07,0x10,0x00,0x00,0x8f,0xff,0x00,0x02]
 ; CHECK: r0 = *(u64 *)(r1 + 0) # encoding: [0x79,0x01,0x00,0x00,0x00,0x00,0x00,0x00]
   %2 = alloca i64, align 8

llvm/test/CodeGen/BPF/sanity.ll

@@ -105,7 +105,7 @@ define void @foo_printf() #1 {
   %1 = getelementptr inbounds [9 x i8], [9 x i8]* %fmt, i64 0, i64 0
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* getelementptr inbounds ([9 x i8], [9 x i8]* @foo_printf.fmt, i64 0, i64 0), i64 9, i32 1, i1 false)
 ; CHECK-LABEL: foo_printf:
-; CHECK: r1 = 729618802566522216ll
+; CHECK: r1 = 729618802566522216 ll
   %2 = call i32 (i8*, ...) @printf(i8* %1) #3
   ret void
 }