summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--libclc/amdgcn-amdhsa/lib/OVERRIDES3
-rw-r--r--libclc/amdgcn/lib/OVERRIDES_3.91
-rw-r--r--libclc/amdgcn/lib/OVERRIDES_4.04
-rw-r--r--libclc/amdgcn/lib/OVERRIDES_5.04
-rw-r--r--libclc/amdgcn/lib/OVERRIDES_6.04
-rw-r--r--libclc/amdgcn/lib/SOURCES1
-rw-r--r--libclc/amdgcn/lib/SOURCES_3.92
-rw-r--r--libclc/amdgcn/lib/SOURCES_4.05
-rw-r--r--libclc/amdgcn/lib/SOURCES_5.04
-rw-r--r--libclc/amdgcn/lib/SOURCES_6.04
-rw-r--r--libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll49
-rw-r--r--libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll2
-rw-r--r--libclc/amdgcn/lib/workitem/get_global_size.40.ll23
-rw-r--r--libclc/amdgcn/lib/workitem/get_global_size.ll2
-rw-r--r--libclc/amdgcn/lib/workitem/get_local_size.40.ll23
-rw-r--r--libclc/amdgcn/lib/workitem/get_local_size.ll2
-rw-r--r--libclc/amdgcn/lib/workitem/get_num_groups.40.ll23
-rw-r--r--libclc/amdgcn/lib/workitem/get_num_groups.ll2
18 files changed, 153 insertions, 5 deletions
diff --git a/libclc/amdgcn-amdhsa/lib/OVERRIDES b/libclc/amdgcn-amdhsa/lib/OVERRIDES
index c9bd69b76eb..18ce3f8ea9c 100644
--- a/libclc/amdgcn-amdhsa/lib/OVERRIDES
+++ b/libclc/amdgcn-amdhsa/lib/OVERRIDES
@@ -1 +1,4 @@
workitem/get_num_groups.ll
+workitem/get_num_groups.40.ll
+workitem/get_global_size.40.ll
+workitem/get_local_size.40.ll
diff --git a/libclc/amdgcn/lib/OVERRIDES_3.9 b/libclc/amdgcn/lib/OVERRIDES_3.9
index f26fbbe9934..3268f67e192 100644
--- a/libclc/amdgcn/lib/OVERRIDES_3.9
+++ b/libclc/amdgcn/lib/OVERRIDES_3.9
@@ -1,3 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.ll
workitem/get_global_size.ll
workitem/get_local_size.ll
workitem/get_num_groups.ll
diff --git a/libclc/amdgcn/lib/OVERRIDES_4.0 b/libclc/amdgcn/lib/OVERRIDES_4.0
new file mode 100644
index 00000000000..3268f67e192
--- /dev/null
+++ b/libclc/amdgcn/lib/OVERRIDES_4.0
@@ -0,0 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.ll
+workitem/get_global_size.ll
+workitem/get_local_size.ll
+workitem/get_num_groups.ll
diff --git a/libclc/amdgcn/lib/OVERRIDES_5.0 b/libclc/amdgcn/lib/OVERRIDES_5.0
new file mode 100644
index 00000000000..3268f67e192
--- /dev/null
+++ b/libclc/amdgcn/lib/OVERRIDES_5.0
@@ -0,0 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.ll
+workitem/get_global_size.ll
+workitem/get_local_size.ll
+workitem/get_num_groups.ll
diff --git a/libclc/amdgcn/lib/OVERRIDES_6.0 b/libclc/amdgcn/lib/OVERRIDES_6.0
new file mode 100644
index 00000000000..3268f67e192
--- /dev/null
+++ b/libclc/amdgcn/lib/OVERRIDES_6.0
@@ -0,0 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.ll
+workitem/get_global_size.ll
+workitem/get_local_size.ll
+workitem/get_num_groups.ll
diff --git a/libclc/amdgcn/lib/SOURCES b/libclc/amdgcn/lib/SOURCES
index 8c177bdd8ee..8e14ce2c6c9 100644
--- a/libclc/amdgcn/lib/SOURCES
+++ b/libclc/amdgcn/lib/SOURCES
@@ -1,7 +1,6 @@
cl_khr_int64_extended_atomics/minmax_helpers.ll
math/ldexp.cl
mem_fence/fence.cl
-mem_fence/waitcnt.ll
synchronization/barrier.cl
workitem/get_global_offset.cl
workitem/get_group_id.cl
diff --git a/libclc/amdgcn/lib/SOURCES_3.9 b/libclc/amdgcn/lib/SOURCES_3.9
index 6ee7cb8184f..3cecdb01cbe 100644
--- a/libclc/amdgcn/lib/SOURCES_3.9
+++ b/libclc/amdgcn/lib/SOURCES_3.9
@@ -1,3 +1,5 @@
+cl_khr_int64_extended_atomics/minmax_helpers.39.ll
+mem_fence/waitcnt.ll
workitem/get_global_size.39.ll
workitem/get_local_size.39.ll
workitem/get_num_groups.39.ll
diff --git a/libclc/amdgcn/lib/SOURCES_4.0 b/libclc/amdgcn/lib/SOURCES_4.0
new file mode 100644
index 00000000000..5ed1d7c037e
--- /dev/null
+++ b/libclc/amdgcn/lib/SOURCES_4.0
@@ -0,0 +1,5 @@
+cl_khr_int64_extended_atomics/minmax_helpers.39.ll
+mem_fence/waitcnt.ll
+workitem/get_global_size.40.ll
+workitem/get_local_size.40.ll
+workitem/get_num_groups.40.ll
diff --git a/libclc/amdgcn/lib/SOURCES_5.0 b/libclc/amdgcn/lib/SOURCES_5.0
new file mode 100644
index 00000000000..45c51ec3689
--- /dev/null
+++ b/libclc/amdgcn/lib/SOURCES_5.0
@@ -0,0 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.39.ll
+workitem/get_global_size.40.ll
+workitem/get_local_size.40.ll
+workitem/get_num_groups.40.ll
diff --git a/libclc/amdgcn/lib/SOURCES_6.0 b/libclc/amdgcn/lib/SOURCES_6.0
new file mode 100644
index 00000000000..45c51ec3689
--- /dev/null
+++ b/libclc/amdgcn/lib/SOURCES_6.0
@@ -0,0 +1,4 @@
+cl_khr_int64_extended_atomics/minmax_helpers.39.ll
+workitem/get_global_size.40.ll
+workitem/get_local_size.40.ll
+workitem/get_num_groups.40.ll
diff --git a/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll b/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll
new file mode 100644
index 00000000000..ec14179c243
--- /dev/null
+++ b/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll
@@ -0,0 +1,49 @@
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+
+define i64 @__clc__sync_fetch_and_min_global_8(i64 addrspace(1)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile min i64 addrspace(1)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_umin_global_8(i64 addrspace(1)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile umin i64 addrspace(1)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_min_local_8(i64 addrspace(3)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile min i64 addrspace(3)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_umin_local_8(i64 addrspace(3)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile umin i64 addrspace(3)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_max_global_8(i64 addrspace(1)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile max i64 addrspace(1)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_umax_global_8(i64 addrspace(1)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile umax i64 addrspace(1)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_max_local_8(i64 addrspace(3)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile max i64 addrspace(3)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
+
+define i64 @__clc__sync_fetch_and_umax_local_8(i64 addrspace(3)* nocapture %ptr, i64 %value) nounwind alwaysinline {
+entry:
+ %0 = atomicrmw volatile umax i64 addrspace(3)* %ptr, i64 %value seq_cst
+ ret i64 %0
+}
diff --git a/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll b/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll
index ec14179c243..509d535ec41 100644
--- a/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll
+++ b/libclc/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll
@@ -1,4 +1,4 @@
-target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target datalayout = "e-p:64:64-p1:64:64-p2:64:64-p3:32:32-p4:32:32-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-A5"
define i64 @__clc__sync_fetch_and_min_global_8(i64 addrspace(1)* nocapture %ptr, i64 %value) nounwind alwaysinline {
entry:
diff --git a/libclc/amdgcn/lib/workitem/get_global_size.40.ll b/libclc/amdgcn/lib/workitem/get_global_size.40.ll
new file mode 100644
index 00000000000..3d26d2fc373
--- /dev/null
+++ b/libclc/amdgcn/lib/workitem/get_global_size.40.ll
@@ -0,0 +1,23 @@
+declare i32 @llvm.r600.read.global.size.x() nounwind readnone
+declare i32 @llvm.r600.read.global.size.y() nounwind readnone
+declare i32 @llvm.r600.read.global.size.z() nounwind readnone
+
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+
+define i64 @get_global_size(i32 %dim) nounwind readnone alwaysinline {
+ switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
+x_dim:
+ %x = call i32 @llvm.r600.read.global.size.x()
+ %x.ext = zext i32 %x to i64
+ ret i64 %x.ext
+y_dim:
+ %y = call i32 @llvm.r600.read.global.size.y()
+ %y.ext = zext i32 %y to i64
+ ret i64 %y.ext
+z_dim:
+ %z = call i32 @llvm.r600.read.global.size.z()
+ %z.ext = zext i32 %z to i64
+ ret i64 %z.ext
+default:
+ ret i64 1
+}
diff --git a/libclc/amdgcn/lib/workitem/get_global_size.ll b/libclc/amdgcn/lib/workitem/get_global_size.ll
index 3d26d2fc373..3292af5fca6 100644
--- a/libclc/amdgcn/lib/workitem/get_global_size.ll
+++ b/libclc/amdgcn/lib/workitem/get_global_size.ll
@@ -2,7 +2,7 @@ declare i32 @llvm.r600.read.global.size.x() nounwind readnone
declare i32 @llvm.r600.read.global.size.y() nounwind readnone
declare i32 @llvm.r600.read.global.size.z() nounwind readnone
-target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target datalayout = "e-p:64:64-p1:64:64-p2:64:64-p3:32:32-p4:32:32-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-A5"
define i64 @get_global_size(i32 %dim) nounwind readnone alwaysinline {
switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
diff --git a/libclc/amdgcn/lib/workitem/get_local_size.40.ll b/libclc/amdgcn/lib/workitem/get_local_size.40.ll
new file mode 100644
index 00000000000..36141f91263
--- /dev/null
+++ b/libclc/amdgcn/lib/workitem/get_local_size.40.ll
@@ -0,0 +1,23 @@
+declare i32 @llvm.r600.read.local.size.x() nounwind readnone
+declare i32 @llvm.r600.read.local.size.y() nounwind readnone
+declare i32 @llvm.r600.read.local.size.z() nounwind readnone
+
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+
+define i64 @get_local_size(i32 %dim) nounwind readnone alwaysinline {
+ switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
+x_dim:
+ %x = call i32 @llvm.r600.read.local.size.x()
+ %x.ext = zext i32 %x to i64
+ ret i64 %x.ext
+y_dim:
+ %y = call i32 @llvm.r600.read.local.size.y()
+ %y.ext = zext i32 %y to i64
+ ret i64 %y.ext
+z_dim:
+ %z = call i32 @llvm.r600.read.local.size.z()
+ %z.ext = zext i32 %z to i64
+ ret i64 %z.ext
+default:
+ ret i64 1
+}
diff --git a/libclc/amdgcn/lib/workitem/get_local_size.ll b/libclc/amdgcn/lib/workitem/get_local_size.ll
index 36141f91263..1b5fec13806 100644
--- a/libclc/amdgcn/lib/workitem/get_local_size.ll
+++ b/libclc/amdgcn/lib/workitem/get_local_size.ll
@@ -2,7 +2,7 @@ declare i32 @llvm.r600.read.local.size.x() nounwind readnone
declare i32 @llvm.r600.read.local.size.y() nounwind readnone
declare i32 @llvm.r600.read.local.size.z() nounwind readnone
-target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target datalayout = "e-p:64:64-p1:64:64-p2:64:64-p3:32:32-p4:32:32-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-A5"
define i64 @get_local_size(i32 %dim) nounwind readnone alwaysinline {
switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
diff --git a/libclc/amdgcn/lib/workitem/get_num_groups.40.ll b/libclc/amdgcn/lib/workitem/get_num_groups.40.ll
new file mode 100644
index 00000000000..12ec8eacda1
--- /dev/null
+++ b/libclc/amdgcn/lib/workitem/get_num_groups.40.ll
@@ -0,0 +1,23 @@
+declare i32 @llvm.r600.read.ngroups.x() nounwind readnone
+declare i32 @llvm.r600.read.ngroups.y() nounwind readnone
+declare i32 @llvm.r600.read.ngroups.z() nounwind readnone
+
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+
+define i64 @get_num_groups(i32 %dim) nounwind readnone alwaysinline {
+ switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
+x_dim:
+ %x = call i32 @llvm.r600.read.ngroups.x()
+ %x.ext = zext i32 %x to i64
+ ret i64 %x.ext
+y_dim:
+ %y = call i32 @llvm.r600.read.ngroups.y()
+ %y.ext = zext i32 %y to i64
+ ret i64 %y.ext
+z_dim:
+ %z = call i32 @llvm.r600.read.ngroups.z()
+ %z.ext = zext i32 %z to i64
+ ret i64 %z.ext
+default:
+ ret i64 1
+}
diff --git a/libclc/amdgcn/lib/workitem/get_num_groups.ll b/libclc/amdgcn/lib/workitem/get_num_groups.ll
index 12ec8eacda1..8fb55eda7f5 100644
--- a/libclc/amdgcn/lib/workitem/get_num_groups.ll
+++ b/libclc/amdgcn/lib/workitem/get_num_groups.ll
@@ -2,7 +2,7 @@ declare i32 @llvm.r600.read.ngroups.x() nounwind readnone
declare i32 @llvm.r600.read.ngroups.y() nounwind readnone
declare i32 @llvm.r600.read.ngroups.z() nounwind readnone
-target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target datalayout = "e-p:64:64-p1:64:64-p2:64:64-p3:32:32-p4:32:32-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-A5"
define i64 @get_num_groups(i32 %dim) nounwind readnone alwaysinline {
switch i32 %dim, label %default [i32 0, label %x_dim i32 1, label %y_dim i32 2, label %z_dim]
OpenPOWER on IntegriCloud