Add clang driver test and update one more SYCL runtime test.

Author: bader

clang/test/Driver/sycl-device-optimizations.cpp

@@ -0,0 +1,14 @@
+/// Check that optimizations for sycl device are enabled by default:
+// RUN:   %clang -### -fsycl %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK-DEFAULT %s
+// RUN:   %clang -### -fsycl -fsycl-device-only %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK-DEFAULT %s
+// CHECK-DEFAULT-NOT: "-fno-sycl-std-optimizations"
+// CHECK-DEFAULT-NOT: "-disable-llvm-passes"
+
+/// Check "-fno-sycl-std-optimizations" is passed to the front-end:
+// RUN:   %clang -### -fsycl -fno-sycl-std-optimizations %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK-NO-SYCL-STD-OPTS %s
+// RUN:   %clang -### -fsycl -fsycl-device-only -fno-sycl-std-optimizations %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK-NO-SYCL-STD-OPTS %s
+// CHECK-NO-SYCL-STD-OPTS: "-fno-sycl-std-optimizations"

sycl/test/sub_group/generic-shuffle.cpp

@@ -2,7 +2,7 @@
 // CUDA compilation and runtime do not yet support sub-groups.
 // #2245 failed on OpenCL CPU (2020.10.7.0.15) with avx2 instruction set
 //
-// RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple %s -o %t.out
+// RUN: %clangxx -fsycl -fno-sycl-std-optimizations -fsycl-targets=%sycl_triple %s -o %t.out
 // RUN: env SYCL_DEVICE_TYPE=HOST %t.out
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 // RUN: %GPU_RUN_PLACEHOLDER %t.out