AMDGPU/SI: Add support for 32-bit immediate SMRD offsets on CI
Reviewers: arsenm
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D11604
llvm-svn: 244254
diff --git a/llvm/test/CodeGen/AMDGPU/smrd.ll b/llvm/test/CodeGen/AMDGPU/smrd.ll
index 167bbe3..1d6bb9e 100644
--- a/llvm/test/CodeGen/AMDGPU/smrd.ll
+++ b/llvm/test/CodeGen/AMDGPU/smrd.ll
@@ -1,9 +1,10 @@
-; RUN: llc < %s -march=amdgcn -mcpu=SI -show-mc-encoding -verify-machineinstrs | FileCheck --check-prefix=SI --check-prefix=GCN %s
-; RUN: llc < %s -march=amdgcn -mcpu=tonga -show-mc-encoding -verify-machineinstrs | FileCheck --check-prefix=VI --check-prefix=GCN %s
+; RUN: llc < %s -march=amdgcn -mcpu=SI -show-mc-encoding -verify-machineinstrs | FileCheck --check-prefix=SI --check-prefix=GCN --check-prefix=SIVI %s
+; RUN: llc < %s -march=amdgcn -mcpu=bonaire -show-mc-encoding -verify-machineinstrs | FileCheck --check-prefix=CI --check-prefix=GCN %s
+; RUN: llc < %s -march=amdgcn -mcpu=tonga -show-mc-encoding -verify-machineinstrs | FileCheck --check-prefix=VI --check-prefix=GCN --check-prefix=SIVI %s
; SMRD load with an immediate offset.
; GCN-LABEL: {{^}}smrd0:
-; SI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x1 ; encoding: [0x01
+; SICI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x1 ; encoding: [0x01
; VI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x4
define void @smrd0(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
entry:
@@ -15,7 +16,7 @@
; SMRD load with the largest possible immediate offset.
; GCN-LABEL: {{^}}smrd1:
-; SI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff
+; SICI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff,0x{{[0-9]+[137]}}
; VI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x3fc
define void @smrd1(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
entry:
@@ -29,6 +30,7 @@
; GCN-LABEL: {{^}}smrd2:
; SI: s_movk_i32 s[[OFFSET:[0-9]]], 0x400
; SI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], s[[OFFSET]] ; encoding: [0x0[[OFFSET]]
+; CI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x100
; VI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x400
; GCN: s_endpgm
define void @smrd2(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
@@ -58,6 +60,7 @@
; GCN-LABEL: {{^}}smrd4:
; SI: s_mov_b32 [[OFFSET:s[0-9]+]], 0xffffc
; SI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; CI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x3ffff
; VI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xffffc
define void @smrd4(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
entry:
@@ -69,9 +72,9 @@
; SMRD load with an offset greater than the largest possible immediate on VI
; GCN-LABEL: {{^}}smrd5:
-; GCN: s_mov_b32 [[OFFSET:s[0-9]+]], 0x100000
-; SI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
-; VI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; SIVI: s_mov_b32 [[OFFSET:s[0-9]+]], 0x100000
+; SIVI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; CI: s_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x40000
; GCN: s_endpgm
define void @smrd5(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
entry:
@@ -83,7 +86,7 @@
; SMRD load using the load.const intrinsic with an immediate offset
; GCN-LABEL: {{^}}smrd_load_const0:
-; SI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x4 ; encoding: [0x04
+; SICI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x4 ; encoding: [0x04
; VI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x10
define void @smrd_load_const0(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
main_body:
@@ -97,7 +100,7 @@
; SMRD load using the load.const intrinsic with the largest possible immediate
; offset.
; GCN-LABEL: {{^}}smrd_load_const1:
-; SI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff
+; SICI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff
; VI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x3fc
define void @smrd_load_const1(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
main_body:
@@ -113,6 +116,7 @@
; GCN-LABEL: {{^}}smrd_load_const2:
; SI: s_movk_i32 s[[OFFSET:[0-9]]], 0x400
; SI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], s[[OFFSET]] ; encoding: [0x0[[OFFSET]]
+; CI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x100
; VI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x400
define void @smrd_load_const2(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
main_body:
@@ -127,6 +131,7 @@
; GCN-LABEL: {{^}}smrd_load_const3:
; SI: s_mov_b32 [[OFFSET:s[0-9]+]], 0xffffc
; SI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; CI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x3ffff
; VI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xffffc
define void @smrd_load_const3(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
main_body:
@@ -139,9 +144,9 @@
; SMRD load with an offset greater than the largest possible immediate on VI
; GCN-LABEL: {{^}}smrd_load_const4:
-; GCN: s_mov_b32 [[OFFSET:s[0-9]+]], 0x100000
-; SI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
-; VI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; SIVI: s_mov_b32 [[OFFSET:s[0-9]+]], 0x100000
+; SIVI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], [[OFFSET]]
+; CI: s_buffer_load_dword s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x40000
; GCN: s_endpgm
define void @smrd_load_const4(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
main_body:
diff --git a/llvm/test/MC/AMDGPU/smrd.s b/llvm/test/MC/AMDGPU/smrd.s
index b67abf7..2ef73a1 100644
--- a/llvm/test/MC/AMDGPU/smrd.s
+++ b/llvm/test/MC/AMDGPU/smrd.s
@@ -1,32 +1,53 @@
-// RUN: llvm-mc -arch=amdgcn -show-encoding %s | FileCheck %s
-// RUN: llvm-mc -arch=amdgcn -mcpu=SI -show-encoding %s | FileCheck %s
+// RUN: not llvm-mc -arch=amdgcn -show-encoding %s | FileCheck --check-prefix=GCN --check-prefix=SI %s
+// RUN: not llvm-mc -arch=amdgcn -mcpu=SI -show-encoding %s | FileCheck --check-prefix=GCN --check-prefix=SI %s
+// RUN: llvm-mc -arch=amdgcn -mcpu=bonaire -show-encoding %s | FileCheck --check-prefix=GCN --check-prefix=CI %s
+
+// RUN: not llvm-mc -arch=amdgcn %s 2>&1 | FileCheck %s --check-prefix=NOSI
+// RUN: not llvm-mc -arch=amdgcn -mcpu=SI %s 2>&1 | FileCheck %s --check-prefix=NOSI
+//===----------------------------------------------------------------------===//
+// Offset Handling
+//===----------------------------------------------------------------------===//
+
+s_load_dword s1, s[2:3], 0xfc
+// GCN: s_load_dword s1, s[2:3], 0xfc ; encoding: [0xfc,0x83,0x00,0xc0]
+
+s_load_dword s1, s[2:3], 0xff
+// GCN: s_load_dword s1, s[2:3], 0xff ; encoding: [0xff,0x83,0x00,0xc0]
+
+s_load_dword s1, s[2:3], 0x100
+// NOSI: error: instruction not supported on this GPU
+// CI: s_load_dword s1, s[2:3], 0x100 ; encoding: [0xff,0x82,0x00,0xc0,0x00,0x01,0x00,0x00]
+
+//===----------------------------------------------------------------------===//
+// Instructions
+//===----------------------------------------------------------------------===//
s_load_dword s1, s[2:3], 1
-// CHECK: s_load_dword s1, s[2:3], 0x1 ; encoding: [0x01,0x83,0x00,0xc0]
+// GCN: s_load_dword s1, s[2:3], 0x1 ; encoding: [0x01,0x83,0x00,0xc0]
s_load_dword s1, s[2:3], s4
-// CHECK: s_load_dword s1, s[2:3], s4 ; encoding: [0x04,0x82,0x00,0xc0]
+// GCN: s_load_dword s1, s[2:3], s4 ; encoding: [0x04,0x82,0x00,0xc0]
s_load_dwordx2 s[2:3], s[2:3], 1
-// CHECK: s_load_dwordx2 s[2:3], s[2:3], 0x1 ; encoding: [0x01,0x03,0x41,0xc0]
+// GCN: s_load_dwordx2 s[2:3], s[2:3], 0x1 ; encoding: [0x01,0x03,0x41,0xc0]
s_load_dwordx2 s[2:3], s[2:3], s4
-// CHECK: s_load_dwordx2 s[2:3], s[2:3], s4 ; encoding: [0x04,0x02,0x41,0xc0]
+// GCN: s_load_dwordx2 s[2:3], s[2:3], s4 ; encoding: [0x04,0x02,0x41,0xc0]
s_load_dwordx4 s[4:7], s[2:3], 1
-// CHECK: s_load_dwordx4 s[4:7], s[2:3], 0x1 ; encoding: [0x01,0x03,0x82,0xc0]
+// GCN: s_load_dwordx4 s[4:7], s[2:3], 0x1 ; encoding: [0x01,0x03,0x82,0xc0]
s_load_dwordx4 s[4:7], s[2:3], s4
-// CHECK: s_load_dwordx4 s[4:7], s[2:3], s4 ; encoding: [0x04,0x02,0x82,0xc0]
+// GCN: s_load_dwordx4 s[4:7], s[2:3], s4 ; encoding: [0x04,0x02,0x82,0xc0]
s_load_dwordx8 s[8:15], s[2:3], 1
-// CHECK: s_load_dwordx8 s[8:15], s[2:3], 0x1 ; encoding: [0x01,0x03,0xc4,0xc0]
+// GCN: s_load_dwordx8 s[8:15], s[2:3], 0x1 ; encoding: [0x01,0x03,0xc4,0xc0]
s_load_dwordx8 s[8:15], s[2:3], s4
-// CHECK: s_load_dwordx8 s[8:15], s[2:3], s4 ; encoding: [0x04,0x02,0xc4,0xc0]
+// GCN: s_load_dwordx8 s[8:15], s[2:3], s4 ; encoding: [0x04,0x02,0xc4,0xc0]
s_load_dwordx16 s[16:31], s[2:3], 1
-// CHECK: s_load_dwordx16 s[16:31], s[2:3], 0x1 ; encoding: [0x01,0x03,0x08,0xc1]
+// GCN: s_load_dwordx16 s[16:31], s[2:3], 0x1 ; encoding: [0x01,0x03,0x08,0xc1]
s_load_dwordx16 s[16:31], s[2:3], s4
-// CHECK: s_load_dwordx16 s[16:31], s[2:3], s4 ; encoding: [0x04,0x02,0x08,0xc1]
+// GCN: s_load_dwordx16 s[16:31], s[2:3], s4 ; encoding: [0x04,0x02,0x08,0xc1]