fix workgroup sizes in mutable_command_global_size test

test MutableDispatchGlobalSize was failing (with an implementation that
doesn't support non-uniform WGs), because the update_global_size = 3
was not a multiple of the local work-size.

Fixed by increasing the global work-size to 256K and update size to 16K.
This should work with all devices that have max_work_group_size <= 16K.

Updates also MutableDispatchWorkGroups which had an out-of-bounds access,
because it hardcoded the value of global work-size.

Author: franz

test_conformance/extensions/cl_khr_command_buffer/cl_khr_command_buffer_mutable_dispatch/mutable_command_basic.h

@@ -126,7 +126,7 @@ struct BasicMutableCommandBufferTest : BasicCommandBufferTest
     clUpdateMutableCommandsKHR_fn clUpdateMutableCommandsKHR = nullptr;
 
     const char* kernelString = "__kernel void empty() {}";
-    const size_t global_work_size = 4 * 16;
+    const size_t global_work_size = 256 * 1024;
 };
 
 struct InfoMutableCommandBufferTest : BasicMutableCommandBufferTest

test_conformance/extensions/cl_khr_command_buffer/cl_khr_command_buffer_mutable_dispatch/mutable_command_global_size.cpp

@@ -153,9 +153,9 @@ struct MutableDispatchGlobalSize : public InfoMutableCommandBufferTest
     }
 
     size_t info_global_size = 0;
-    const size_t update_global_size = 3;
-    const size_t sizeToAllocate = global_work_size;
-    const size_t num_elements = sizeToAllocate / sizeof(cl_int);
+    const size_t update_global_size = 16 * 1024;
+    const size_t sizeToAllocate = global_work_size * sizeof(cl_int);
+    const size_t num_elements = global_work_size;
     cl_mutable_command_khr command = nullptr;
 };
 

test_conformance/extensions/cl_khr_command_buffer/cl_khr_command_buffer_mutable_dispatch/mutable_command_work_groups.cpp

@@ -278,10 +278,9 @@ struct MutableDispatchWorkGroups : public BasicMutableCommandBufferTest
     Configuration config;
 
     size_t info_global_size = 0;
-    static constexpr size_t test_global_work_size = 64;
-    static constexpr size_t update_global_size = 16;
+    static constexpr size_t update_global_size = 16 * 1024;
     const size_t local_work_size = 8;
-    const size_t sizeToAllocate = 64 * sizeof(cl_int);
+    const size_t sizeToAllocate = global_work_size * sizeof(cl_int);
 };
 
 int test_command_buffer_with_no_additional_work_groups(cl_device_id device,