[PATCH] D42319: [CUDA] CUDA has no device-side library builtins.

Artem Belevich via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Fri Jan 19 14:40:50 PST 2018

tra created this revision.
tra added a reviewer: jlebar.
Herald added a subscriber: sanjoy.

We should (almost) never consider a device-side declaration to match a
builtin.  If we do, the un-inlined device-side functions provided by
CUDA headers that ship with clang may be ignored. We may end up emitting
as a call to a llvm intrinsic which would typically be lowered as
an external library call. This results in a back-end error because NVPTX
back-end does not support it.



Index: clang/test/CodeGenCUDA/library-builtin.cu
--- /dev/null
+++ clang/test/CodeGenCUDA/library-builtin.cu
@@ -0,0 +1,22 @@
+// REQUIRES: x86-registered-target
+// REQUIRES: nvptx-registered-target
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -o - %s | \
+// RUN:  FileCheck --check-prefixes=HOST,BOTH %s
+// RUN: %clang_cc1 -fcuda-is-device -triple nvptx64-nvidia-cuda \
+// RUN:   -emit-llvm -o - %s | FileCheck %s --check-prefixes=DEVICE,BOTH
+// BOTH-LABEL: define float @logf(float
+// logf() should be calling itself recursively as we don't have any standard
+// library on device side.
+// DEVICE: call float @logf(float
+extern "C" __attribute__((device)) float logf(float __x) { return logf(__x); }
+// NOTE: this case is to illustrate the expected differences in behavior between
+// the host and device. In general we do not mess with host-side standard
+// library.
+// Host is assumed to have standerd library, so logf() calls LLVM intrinsic.
+// HOST: call float @llvm.log.f32(float
+extern "C" float logf(float __x) { return logf(__x); }
Index: clang/lib/AST/Decl.cpp
--- clang/lib/AST/Decl.cpp
+++ clang/lib/AST/Decl.cpp
@@ -2901,6 +2901,13 @@
     return 0;
+  // CUDA does not have device-side standard library. printf and malloc are the
+  // only special cases that are supported by device-side runtime.
+  if (Context.getLangOpts().CUDA && hasAttr<CUDADeviceAttr>() &&
+      !hasAttr<CUDAHostAttr>() &&
+      !(BuiltinID == Builtin::BIprintf || BuiltinID == Builtin::BImalloc))
+    return 0;
   return BuiltinID;

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D42319.130697.patch
Type: text/x-patch
Size: 1778 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20180119/ed074cc3/attachment.bin>

More information about the cfe-commits mailing list