6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-10-01 10:57:21 +00:00
Code Issues Projects Releases Wiki Activity

LegalizeDAG: allow libcalls for max/min atomic operations

Browse Source 
ARM processors without ldrex/strex need to be able to make libcalls for all
atomic operations, including the newer min/max versions.

The alternative would probably be expanding these operations in terms of
cmpxchg (as x86 does always), but in the configurations where this matters
code-size tends to be paramount so the libcall is more desirable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193398 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Tim Northover 2013-10-25 09:30:20 +00:00
parent abe3aa3520
commit 5a42ae81f7
4 changed files with 104 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
include/llvm/CodeGen/RuntimeLibcalls.h
View File

@ -361,6 +361,26 @@ namespace RTLIB {
SYNC_FETCH_AND_NAND_4,
SYNC_FETCH_AND_NAND_8,
SYNC_FETCH_AND_NAND_16,
SYNC_FETCH_AND_MAX_1,
SYNC_FETCH_AND_MAX_2,
SYNC_FETCH_AND_MAX_4,
SYNC_FETCH_AND_MAX_8,
SYNC_FETCH_AND_MAX_16,
SYNC_FETCH_AND_UMAX_1,
SYNC_FETCH_AND_UMAX_2,
SYNC_FETCH_AND_UMAX_4,
SYNC_FETCH_AND_UMAX_8,
SYNC_FETCH_AND_UMAX_16,
SYNC_FETCH_AND_MIN_1,
SYNC_FETCH_AND_MIN_2,
SYNC_FETCH_AND_MIN_4,
SYNC_FETCH_AND_MIN_8,
SYNC_FETCH_AND_MIN_16,
SYNC_FETCH_AND_UMIN_1,
SYNC_FETCH_AND_UMIN_2,
SYNC_FETCH_AND_UMIN_4,
SYNC_FETCH_AND_UMIN_8,
SYNC_FETCH_AND_UMIN_16,
// Stack Protector Fail.
STACKPROTECTOR_CHECK_FAIL,

40
lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
View File

@ -2734,6 +2734,46 @@ std::pair <SDValue, SDValue> SelectionDAGLegalize::ExpandAtomic(SDNode *Node) {
case MVT::i128:LC = RTLIB::SYNC_FETCH_AND_NAND_16;break;
}
break;
case ISD::ATOMIC_LOAD_MAX:
switch (VT.SimpleTy) {
default: llvm_unreachable("Unexpected value type for atomic!");
case MVT::i8: LC = RTLIB::SYNC_FETCH_AND_MAX_1; break;
case MVT::i16: LC = RTLIB::SYNC_FETCH_AND_MAX_2; break;
case MVT::i32: LC = RTLIB::SYNC_FETCH_AND_MAX_4; break;
case MVT::i64: LC = RTLIB::SYNC_FETCH_AND_MAX_8; break;
case MVT::i128:LC = RTLIB::SYNC_FETCH_AND_MAX_16;break;
}
break;
case ISD::ATOMIC_LOAD_UMAX:
switch (VT.SimpleTy) {
default: llvm_unreachable("Unexpected value type for atomic!");
case MVT::i8: LC = RTLIB::SYNC_FETCH_AND_UMAX_1; break;
case MVT::i16: LC = RTLIB::SYNC_FETCH_AND_UMAX_2; break;
case MVT::i32: LC = RTLIB::SYNC_FETCH_AND_UMAX_4; break;
case MVT::i64: LC = RTLIB::SYNC_FETCH_AND_UMAX_8; break;
case MVT::i128:LC = RTLIB::SYNC_FETCH_AND_UMAX_16;break;
}
break;
case ISD::ATOMIC_LOAD_MIN:
switch (VT.SimpleTy) {
default: llvm_unreachable("Unexpected value type for atomic!");
case MVT::i8: LC = RTLIB::SYNC_FETCH_AND_MIN_1; break;
case MVT::i16: LC = RTLIB::SYNC_FETCH_AND_MIN_2; break;
case MVT::i32: LC = RTLIB::SYNC_FETCH_AND_MIN_4; break;
case MVT::i64: LC = RTLIB::SYNC_FETCH_AND_MIN_8; break;
case MVT::i128:LC = RTLIB::SYNC_FETCH_AND_MIN_16;break;
}
break;
case ISD::ATOMIC_LOAD_UMIN:
switch (VT.SimpleTy) {
default: llvm_unreachable("Unexpected value type for atomic!");
case MVT::i8: LC = RTLIB::SYNC_FETCH_AND_UMIN_1; break;
case MVT::i16: LC = RTLIB::SYNC_FETCH_AND_UMIN_2; break;
case MVT::i32: LC = RTLIB::SYNC_FETCH_AND_UMIN_4; break;
case MVT::i64: LC = RTLIB::SYNC_FETCH_AND_UMIN_8; break;
case MVT::i128:LC = RTLIB::SYNC_FETCH_AND_UMIN_16;break;
}
break;
}
return ExpandChainLibCall(LC, Node, false);

20
lib/CodeGen/TargetLoweringBase.cpp
View File

@ -354,6 +354,26 @@ static void InitLibcallNames(const char **Names, const TargetMachine &TM) {
Names[RTLIB::SYNC_FETCH_AND_NAND_4] = "__sync_fetch_and_nand_4";
Names[RTLIB::SYNC_FETCH_AND_NAND_8] = "__sync_fetch_and_nand_8";
Names[RTLIB::SYNC_FETCH_AND_NAND_16] = "__sync_fetch_and_nand_16";
Names[RTLIB::SYNC_FETCH_AND_MAX_1] = "__sync_fetch_and_max_1";
Names[RTLIB::SYNC_FETCH_AND_MAX_2] = "__sync_fetch_and_max_2";
Names[RTLIB::SYNC_FETCH_AND_MAX_4] = "__sync_fetch_and_max_4";
Names[RTLIB::SYNC_FETCH_AND_MAX_8] = "__sync_fetch_and_max_8";
Names[RTLIB::SYNC_FETCH_AND_MAX_16] = "__sync_fetch_and_max_16";
Names[RTLIB::SYNC_FETCH_AND_UMAX_1] = "__sync_fetch_and_umax_1";
Names[RTLIB::SYNC_FETCH_AND_UMAX_2] = "__sync_fetch_and_umax_2";
Names[RTLIB::SYNC_FETCH_AND_UMAX_4] = "__sync_fetch_and_umax_4";
Names[RTLIB::SYNC_FETCH_AND_UMAX_8] = "__sync_fetch_and_umax_8";
Names[RTLIB::SYNC_FETCH_AND_UMAX_16] = "__sync_fetch_and_umax_16";
Names[RTLIB::SYNC_FETCH_AND_MIN_1] = "__sync_fetch_and_min_1";
Names[RTLIB::SYNC_FETCH_AND_MIN_2] = "__sync_fetch_and_min_2";
Names[RTLIB::SYNC_FETCH_AND_MIN_4] = "__sync_fetch_and_min_4";
Names[RTLIB::SYNC_FETCH_AND_MIN_8] = "__sync_fetch_and_min_8";
Names[RTLIB::SYNC_FETCH_AND_MIN_16] = "__sync_fetch_and_min_16";
Names[RTLIB::SYNC_FETCH_AND_UMIN_1] = "__sync_fetch_and_umin_1";
Names[RTLIB::SYNC_FETCH_AND_UMIN_2] = "__sync_fetch_and_umin_2";
Names[RTLIB::SYNC_FETCH_AND_UMIN_4] = "__sync_fetch_and_umin_4";
Names[RTLIB::SYNC_FETCH_AND_UMIN_8] = "__sync_fetch_and_umin_8";
Names[RTLIB::SYNC_FETCH_AND_UMIN_16] = "__sync_fetch_and_umin_16";
if (Triple(TM.getTargetTriple()).getEnvironment() == Triple::GNU) {
Names[RTLIB::SINCOS_F32] = "sincosf";

24
test/CodeGen/ARM/atomic-op.ll
View File

@ -1,5 +1,6 @@
; RUN: llc < %s -mtriple=armv7-apple-ios -verify-machineinstrs | FileCheck %s
; RUN: llc < %s -mtriple=thumbv7-apple-ios -verify-machineinstrs | FileCheck %s
; RUN: llc < %s -mtriple=thumbv6-apple-ios -verify-machineinstrs | FileCheck %s --check-prefix=CHECK-T1
define void @func(i32 %argc, i8** %argv) nounwind {
entry:
@ -24,78 +25,93 @@ entry:
; CHECK: ldrex
; CHECK: add
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_add_4
%0 = atomicrmw add i32* %val1, i32 %tmp monotonic
store i32 %0, i32* %old
; CHECK: ldrex
; CHECK: sub
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_sub_4
%1 = atomicrmw sub i32* %val2, i32 30 monotonic
store i32 %1, i32* %old
; CHECK: ldrex
; CHECK: add
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_add_4
%2 = atomicrmw add i32* %val2, i32 1 monotonic
store i32 %2, i32* %old
; CHECK: ldrex
; CHECK: sub
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_sub_4
%3 = atomicrmw sub i32* %val2, i32 1 monotonic
store i32 %3, i32* %old
; CHECK: ldrex
; CHECK: and
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_and_4
%4 = atomicrmw and i32* %andt, i32 4080 monotonic
store i32 %4, i32* %old
; CHECK: ldrex
; CHECK: or
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_or_4
%5 = atomicrmw or i32* %ort, i32 4080 monotonic
store i32 %5, i32* %old
; CHECK: ldrex
; CHECK: eor
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_xor_4
%6 = atomicrmw xor i32* %xort, i32 4080 monotonic
store i32 %6, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_min_4
%7 = atomicrmw min i32* %val2, i32 16 monotonic
store i32 %7, i32* %old
%neg = sub i32 0, 1
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_min_4
%8 = atomicrmw min i32* %val2, i32 %neg monotonic
store i32 %8, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_max_4
%9 = atomicrmw max i32* %val2, i32 1 monotonic
store i32 %9, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_max_4
%10 = atomicrmw max i32* %val2, i32 0 monotonic
store i32 %10, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_4
%11 = atomicrmw umin i32* %val2, i32 16 monotonic
store i32 %11, i32* %old
%uneg = sub i32 0, 1
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_4
%12 = atomicrmw umin i32* %val2, i32 %uneg monotonic
store i32 %12, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_4
%13 = atomicrmw umax i32* %val2, i32 1 monotonic
store i32 %13, i32* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_4
%14 = atomicrmw umax i32* %val2, i32 0 monotonic
store i32 %14, i32* %old
@ -110,22 +126,26 @@ entry:
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_2
%0 = atomicrmw umin i16* %val, i16 16 monotonic
store i16 %0, i16* %old
%uneg = sub i16 0, 1
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_2
%1 = atomicrmw umin i16* %val, i16 %uneg monotonic
store i16 %1, i16* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_2
%2 = atomicrmw umax i16* %val, i16 1 monotonic
store i16 %2, i16* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_2
%3 = atomicrmw umax i16* %val, i16 0 monotonic
store i16 %3, i16* %old
ret void
@ -139,22 +159,26 @@ entry:
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_1
%0 = atomicrmw umin i8* %val, i8 16 monotonic
store i8 %0, i8* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umin_1
%uneg = sub i8 0, 1
%1 = atomicrmw umin i8* %val, i8 %uneg monotonic
store i8 %1, i8* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_1
%2 = atomicrmw umax i8* %val, i8 1 monotonic
store i8 %2, i8* %old
; CHECK: ldrex
; CHECK: cmp
; CHECK: strex
; CHECK-T1: blx ___sync_fetch_and_umax_1
%3 = atomicrmw umax i8* %val, i8 0 monotonic
store i8 %3, i8* %old
ret void