[SYCL][CUDA] Allow varying program metadata in CUDA backend (#7555)

Recently the program metadata produced for reqd_work_group_size changed
to allow between 1 and 3 values instead of guaranteing 3 values. This
commit makes the CUDA backend correctly parse reqd_work_group_size
metadata with variable number of elements.

Signed-off-by: Larsen, Steffen <steffen.larsen@intel.com>

Author: steffenlarsen

sycl/plugins/cuda/pi_cuda.cpp

@@ -685,13 +685,20 @@ pi_result _pi_program::set_metadata(const pi_device_binary_property *metadata,
     if (get_kernel_metadata(metadataElementName,
                             __SYCL_PI_PROGRAM_METADATA_TAG_REQD_WORK_GROUP_SIZE,
                             kernelName)) {
-      assert(metadataElement->ValSize ==
-                 sizeof(std::uint64_t) + sizeof(std::uint32_t) * 3 &&
+      size_t MDElemsSize = metadataElement->ValSize - sizeof(std::uint64_t);
+
+      // Expect between 1 and 3 32-bit integer values.
+      assert(MDElemsSize >= sizeof(std::uint32_t) &&
+             MDElemsSize <= sizeof(std::uint32_t) * 3 &&
              "Unexpected size for reqd_work_group_size metadata");
 
       // Get pointer to data, skipping 64-bit size at the start of the data.
-      const auto *reqdWorkGroupElements =
-          reinterpret_cast<const std::uint32_t *>(metadataElement->ValAddr) + 2;
+      const char *ValuePtr =
+          reinterpret_cast<const char *>(metadataElement->ValAddr) +
+          sizeof(std::uint64_t);
+      // Read values and pad with 1's for values not present.
+      std::uint32_t reqdWorkGroupElements[] = {1, 1, 1};
+      std::memcpy(reqdWorkGroupElements, ValuePtr, MDElemsSize);
       kernelReqdWorkGroupSizeMD_[kernelName] =
           std::make_tuple(reqdWorkGroupElements[0], reqdWorkGroupElements[1],
                           reqdWorkGroupElements[2]);