[Libclc-dev] [PATCH 1/4] amdgcn: Fix datalayout after clang r324101
Jan Vesely via Libclc-dev
libclc-dev at lists.llvm.org
Thu Feb 22 18:43:29 PST 2018
On Wed, 2018-02-21 at 08:31 -0600, Aaron Watry via Libclc-dev wrote:
> On Mon, Feb 19, 2018 at 3:22 PM, Jan Vesely via Libclc-dev
> <libclc-dev at lists.llvm.org> wrote:
> > On Tue, 2018-02-13 at 20:30 -0500, Jan Vesely wrote:
> > > r324101 switched around AS numbering
> > >
> > > Signed-off-by: Jan Vesely <jan.vesely at rutgers.edu>
> > > ---
> > > amdgcn-amdhsa/lib/OVERRIDES | 3 ++
> > > amdgcn/lib/OVERRIDES_3.9 | 1 +
> > > amdgcn/lib/OVERRIDES_4.0 | 4 ++
> > > amdgcn/lib/OVERRIDES_5.0 | 4 ++
> > > amdgcn/lib/OVERRIDES_6.0 | 4 ++
> > > amdgcn/lib/SOURCES | 1 -
> > > amdgcn/lib/SOURCES_3.9 | 2 +
> > > amdgcn/lib/SOURCES_4.0 | 5 +++
> > > amdgcn/lib/SOURCES_5.0 | 4 ++
> > > amdgcn/lib/SOURCES_6.0 | 4 ++
> > > .../minmax_helpers.39.ll | 49 ++++++++++++++++++++++
> > > .../minmax_helpers.ll | 2 +-
> > > amdgcn/lib/workitem/get_global_size.40.ll | 23 ++++++++++
> > > amdgcn/lib/workitem/get_global_size.ll | 2 +-
> > > amdgcn/lib/workitem/get_local_size.40.ll | 23 ++++++++++
> > > amdgcn/lib/workitem/get_local_size.ll | 2 +-
> > > amdgcn/lib/workitem/get_num_groups.40.ll | 23 ++++++++++
> > > amdgcn/lib/workitem/get_num_groups.ll | 2 +-
> > > 18 files changed, 153 insertions(+), 5 deletions(-)
> > > create mode 100644 amdgcn/lib/OVERRIDES_4.0
> > > create mode 100644 amdgcn/lib/OVERRIDES_5.0
> > > create mode 100644 amdgcn/lib/OVERRIDES_6.0
> > > create mode 100644 amdgcn/lib/SOURCES_4.0
> > > create mode 100644 amdgcn/lib/SOURCES_5.0
> > > create mode 100644 amdgcn/lib/SOURCES_6.0
> > > create mode 100644 amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll
> > > create mode 100644 amdgcn/lib/workitem/get_global_size.40.ll
> > > create mode 100644 amdgcn/lib/workitem/get_local_size.40.ll
> > > create mode 100644 amdgcn/lib/workitem/get_num_groups.40.ll
> > >
> > > diff --git a/amdgcn-amdhsa/lib/OVERRIDES b/amdgcn-amdhsa/lib/OVERRIDES
> > > index c9bd69b..18ce3f8 100644
> > > --- a/amdgcn-amdhsa/lib/OVERRIDES
> > > +++ b/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/amdgcn/lib/OVERRIDES_3.9 b/amdgcn/lib/OVERRIDES_3.9
> > > index f26fbbe..3268f67 100644
> > > --- a/amdgcn/lib/OVERRIDES_3.9
> > > +++ b/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/amdgcn/lib/OVERRIDES_4.0 b/amdgcn/lib/OVERRIDES_4.0
> > > new file mode 100644
> > > index 0000000..3268f67
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/OVERRIDES_5.0 b/amdgcn/lib/OVERRIDES_5.0
> > > new file mode 100644
> > > index 0000000..3268f67
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/OVERRIDES_6.0 b/amdgcn/lib/OVERRIDES_6.0
> > > new file mode 100644
> > > index 0000000..3268f67
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/SOURCES b/amdgcn/lib/SOURCES
> > > index 8c177bd..8e14ce2 100644
> > > --- a/amdgcn/lib/SOURCES
> > > +++ b/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/amdgcn/lib/SOURCES_3.9 b/amdgcn/lib/SOURCES_3.9
> > > index 6ee7cb8..3cecdb0 100644
> > > --- a/amdgcn/lib/SOURCES_3.9
> > > +++ b/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/amdgcn/lib/SOURCES_4.0 b/amdgcn/lib/SOURCES_4.0
> > > new file mode 100644
> > > index 0000000..5ed1d7c
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/SOURCES_5.0 b/amdgcn/lib/SOURCES_5.0
> > > new file mode 100644
> > > index 0000000..45c51ec
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/SOURCES_6.0 b/amdgcn/lib/SOURCES_6.0
> > > new file mode 100644
> > > index 0000000..45c51ec
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll b/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.39.ll
> > > new file mode 100644
> > > index 0000000..ec14179
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll b/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll
> > > index ec14179..509d535 100644
> > > --- a/amdgcn/lib/cl_khr_int64_extended_atomics/minmax_helpers.ll
> > > +++ b/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/amdgcn/lib/workitem/get_global_size.40.ll b/amdgcn/lib/workitem/get_global_size.40.ll
> > > new file mode 100644
> > > index 0000000..3d26d2f
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/workitem/get_global_size.ll b/amdgcn/lib/workitem/get_global_size.ll
> > > index 3d26d2f..3292af5 100644
> > > --- a/amdgcn/lib/workitem/get_global_size.ll
> > > +++ b/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/amdgcn/lib/workitem/get_local_size.40.ll b/amdgcn/lib/workitem/get_local_size.40.ll
> > > new file mode 100644
> > > index 0000000..36141f9
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/workitem/get_local_size.ll b/amdgcn/lib/workitem/get_local_size.ll
> > > index 36141f9..1b5fec1 100644
> > > --- a/amdgcn/lib/workitem/get_local_size.ll
> > > +++ b/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/amdgcn/lib/workitem/get_num_groups.40.ll b/amdgcn/lib/workitem/get_num_groups.40.ll
> > > new file mode 100644
> > > index 0000000..12ec8ea
> > > --- /dev/null
> > > +++ b/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/amdgcn/lib/workitem/get_num_groups.ll b/amdgcn/lib/workitem/get_num_groups.ll
> > > index 12ec8ea..8fb55ed 100644
> > > --- a/amdgcn/lib/workitem/get_num_groups.ll
> > > +++ b/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]
> >
> > ping
>
> I've at least verified that this series fixes the build for me. I
> haven't had a chance to test the applicable CTS tests yet to verify
> functionality. If I don't get back to you in the next 24 hours or so,
> feel free to add my Acked-By on these, otherwise I'll at least try to
> upgrade to a Tested-by. I'm not familiar enough with the datalayout
> stuff to say what is right/wrong, but I can at least try to see if it
> works :).
Thanks.
I assume the time is up and will push this tomorrow.
Jan
>
> --Aaron
>
> > _______________________________________________
> > Libclc-dev mailing list
> > Libclc-dev at lists.llvm.org
> > http://lists.llvm.org/cgi-bin/mailman/listinfo/libclc-dev
> >
>
> _______________________________________________
> Libclc-dev mailing list
> Libclc-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/libclc-dev
--
Jan Vesely <jan.vesely at rutgers.edu>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: This is a digitally signed message part
URL: <http://lists.llvm.org/pipermail/libclc-dev/attachments/20180222/70a6b4cf/attachment.sig>
More information about the Libclc-dev
mailing list