Merge pull request #1891 from IntelPython/contribution-to-fix-gh-1887

Contribution to fix gh 1887

Author: oleksandr-pavlyk

dpctl/tensor/libtensor/include/kernels/copy_as_contiguous.hpp

@@ -44,8 +44,8 @@ namespace copy_as_contig
 
 template <typename T,
           typename IndexerT,
-          int vec_sz = 4,
-          int n_vecs = 2,
+          std::uint32_t vec_sz = 4u,
+          std::uint32_t n_vecs = 2u,
           bool enable_sg_loadstore = true>
 class CopyAsCContigFunctor
 {
@@ -66,53 +66,63 @@ class CopyAsCContigFunctor
 
     void operator()(sycl::nd_item<1> ndit) const
     {
+        static_assert(vec_sz > 0);
+        static_assert(n_vecs > 0);
+        static_assert(vec_sz * n_vecs < (std::uint32_t(1) << 8));
+
+        constexpr std::uint8_t elems_per_wi =
+            static_cast<std::uint8_t>(vec_sz * n_vecs);
+
         using dpctl::tensor::type_utils::is_complex;
         if constexpr (!enable_sg_loadstore || is_complex<T>::value) {
-            const std::uint32_t sgSize =
+            const std::uint16_t sgSize =
                 ndit.get_sub_group().get_local_range()[0];
             const std::size_t gid = ndit.get_global_linear_id();
 
-            const std::size_t base =
-                (gid / sgSize) * sgSize * n_vecs * vec_sz + (gid % sgSize);
-            for (size_t offset = base;
-                 offset < std::min(nelems, base + sgSize * (n_vecs * vec_sz));
-                 offset += sgSize)
-            {
+            // base = (gid / sgSize) * sgSize * elems_per_wi + (gid % sgSize)
+            // gid % sgSize == gid - (gid / sgSize) * sgSize
+            const std::size_t elems_per_sg = sgSize * (elems_per_wi - 1);
+            const std::size_t base = (gid / sgSize) * elems_per_sg + gid;
+            const std::size_t offset_max =
+                std::min(nelems, base + sgSize * elems_per_wi);
+
+            for (size_t offset = base; offset < offset_max; offset += sgSize) {
                 auto src_offset = src_indexer(offset);
                 dst_p[offset] = src_p[src_offset];
             }
         }
         else {
             auto sg = ndit.get_sub_group();
-            const std::uint32_t sgSize = sg.get_local_range()[0];
-            const size_t base = n_vecs * vec_sz *
-                                (ndit.get_group(0) * ndit.get_local_range(0) +
-                                 sg.get_group_id()[0] * sgSize);
+            const std::uint16_t sgSize = sg.get_max_local_range()[0];
+            const size_t base =
+                elems_per_wi * (ndit.get_group(0) * ndit.get_local_range(0) +
+                                sg.get_group_id()[0] * sgSize);
 
-            if (base + n_vecs * vec_sz * sgSize < nelems) {
+            if (base + elems_per_wi * sgSize < nelems) {
                 sycl::vec<T, vec_sz> dst_vec;
 
 #pragma unroll
-                for (std::uint32_t it = 0; it < n_vecs * vec_sz; it += vec_sz) {
+                for (std::uint8_t it = 0; it < elems_per_wi; it += vec_sz) {
+                    const size_t block_start_id = base + it * sgSize;
                     auto dst_multi_ptr = sycl::address_space_cast<
                         sycl::access::address_space::global_space,
-                        sycl::access::decorated::yes>(
-                        &dst_p[base + it * sgSize]);
+                        sycl::access::decorated::yes>(&dst_p[block_start_id]);
 
+                    const size_t elem_id0 = block_start_id + sg.get_local_id();
 #pragma unroll
-                    for (std::uint32_t k = 0; k < vec_sz; k++) {
-                        ssize_t src_offset = src_indexer(
-                            base + (it + k) * sgSize + sg.get_local_id());
+                    for (std::uint8_t k = 0; k < vec_sz; k++) {
+                        const size_t elem_id = elem_id0 + k * sgSize;
+                        const ssize_t src_offset = src_indexer(elem_id);
                         dst_vec[k] = src_p[src_offset];
                     }
                     sg.store<vec_sz>(dst_multi_ptr, dst_vec);
                 }
             }
             else {
-                for (size_t k = base + sg.get_local_id()[0]; k < nelems;
-                     k += sgSize)
-                {
-                    ssize_t src_offset = src_indexer(k);
+                const size_t lane_id = sg.get_local_id()[0];
+                const size_t k0 = base + lane_id;
+                for (size_t k = k0; k < nelems; k += sgSize) {
+                    const ssize_t src_offset = src_indexer(k);
                     dst_p[k] = src_p[src_offset];
                 }
             }
@@ -122,8 +132,8 @@ class CopyAsCContigFunctor
 
 template <typename T,
           typename IndexerT,
-          std::uint32_t n_vecs,
           std::uint32_t vec_sz,
+          std::uint32_t n_vecs,
           bool enable_sg_load,
           typename KernelName>
 sycl::event submit_c_contiguous_copy(sycl::queue &exec_q,
@@ -133,6 +143,10 @@ sycl::event submit_c_contiguous_copy(sycl::queue &exec_q,
                                      const IndexerT &src_indexer,
                                      const std::vector<sycl::event> &depends)
 {
+    static_assert(vec_sz > 0);
+    static_assert(n_vecs > 0);
+    static_assert(vec_sz * n_vecs < (std::uint32_t(1) << 8));
+
     constexpr std::size_t preferred_lws = 256;
 
     const auto &kernel_id = sycl::get_kernel_id<KernelName>();
@@ -150,9 +164,11 @@ sycl::event submit_c_contiguous_copy(sycl::queue &exec_q,
     const std::size_t lws =
         ((preferred_lws + max_sg_size - 1) / max_sg_size) * max_sg_size;
 
-    constexpr std::uint32_t nelems_per_wi = n_vecs * vec_sz;
-    size_t n_groups =
-        (nelems + nelems_per_wi * lws - 1) / (nelems_per_wi * lws);
+    constexpr std::uint8_t nelems_per_wi = n_vecs * vec_sz;
+
+    const size_t nelems_per_group = nelems_per_wi * lws;
+    const size_t n_groups =
+        (nelems + nelems_per_group - 1) / (nelems_per_group);
 
     sycl::event copy_ev = exec_q.submit([&](sycl::handler &cgh) {
         cgh.depends_on(depends);
@@ -171,8 +187,8 @@ sycl::event submit_c_contiguous_copy(sycl::queue &exec_q,
 
 template <typename T,
           typename IndexT,
-          int vec_sz,
-          int n_vecs,
+          std::uint32_t vec_sz,
+          std::uint32_t n_vecs,
           bool enable_sgload>
 class as_contig_krn;
 
@@ -194,8 +210,8 @@ as_c_contiguous_array_generic_impl(sycl::queue &exec_q,
     using IndexerT = dpctl::tensor::offset_utils::StridedIndexer;
     const IndexerT src_indexer(nd, ssize_t(0), shape_and_strides);
 
-    constexpr std::uint32_t n_vecs = 2;
-    constexpr std::uint32_t vec_sz = 4;
+    constexpr std::uint32_t vec_sz = 4u;
+    constexpr std::uint32_t n_vecs = 2u;
 
     using dpctl::tensor::kernels::alignment_utils::
         disabled_sg_loadstore_wrapper_krn;
@@ -207,7 +223,7 @@ as_c_contiguous_array_generic_impl(sycl::queue &exec_q,
         constexpr bool enable_sg_load = true;
         using KernelName =
             as_contig_krn<T, IndexerT, vec_sz, n_vecs, enable_sg_load>;
-        copy_ev = submit_c_contiguous_copy<T, IndexerT, n_vecs, vec_sz,
+        copy_ev = submit_c_contiguous_copy<T, IndexerT, vec_sz, n_vecs,
                                            enable_sg_load, KernelName>(
             exec_q, nelems, src_tp, dst_tp, src_indexer, depends);
     }
@@ -216,7 +232,7 @@ as_c_contiguous_array_generic_impl(sycl::queue &exec_q,
         using InnerKernelName =
             as_contig_krn<T, IndexerT, vec_sz, n_vecs, disable_sg_load>;
         using KernelName = disabled_sg_loadstore_wrapper_krn<InnerKernelName>;
-        copy_ev = submit_c_contiguous_copy<T, IndexerT, n_vecs, vec_sz,
+        copy_ev = submit_c_contiguous_copy<T, IndexerT, vec_sz, n_vecs,
                                            disable_sg_load, KernelName>(
             exec_q, nelems, src_tp, dst_tp, src_indexer, depends);
     }