Removed alignment macros, and added functions instead.

src/bgfx.cpp

@@ -4107,7 +4107,7 @@ namespace bgfx
 	void allocInstanceDataBuffer(InstanceDataBuffer* _idb, uint32_t _num, uint16_t _stride)
 	{
 		BGFX_CHECK_CAPS(BGFX_CAPS_INSTANCING, "Instancing is not supported!");
-		BX_CHECK(_stride == BX_ALIGN_16(_stride), "Stride must be multiple of 16.");
+		BX_CHECK(bx::isAligned(_stride, 16), "Stride must be multiple of 16.");
 		BX_CHECK(0 < _num, "Requesting 0 instanced data vertices.");
 		s_ctx->allocInstanceDataBuffer(_idb, _num, _stride);
 		BX_CHECK(_num == _idb->size / _stride

src/bgfx_p.h

@@ -1335,7 +1335,7 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 		{
 			const uint32_t structSize = sizeof(UniformBuffer)-sizeof(UniformBuffer::m_buffer);
 
-			uint32_t size = BX_ALIGN_16(_size);
+			uint32_t size = bx::alignUp(_size, 16);
 			void*    data = BX_ALLOC(g_allocator, size+structSize);
 			return BX_PLACEMENT_NEW(data, UniformBuffer)(size);
 		}
@@ -1352,7 +1352,7 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 			if (_treshold >= uniformBuffer->m_size - uniformBuffer->m_pos)
 			{
 				const uint32_t structSize = sizeof(UniformBuffer)-sizeof(UniformBuffer::m_buffer);
-				uint32_t size = BX_ALIGN_16(uniformBuffer->m_size + _grow);
+				uint32_t size = bx::alignUp(uniformBuffer->m_size + _grow, 16);
 				void*    data = BX_REALLOC(g_allocator, uniformBuffer, size+structSize);
 				uniformBuffer = reinterpret_cast<UniformBuffer*>(data);
 				uniformBuffer->m_size = size;
@@ -3219,7 +3219,7 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 			}
 
 			const uint32_t indexSize = 0 == (_flags & BGFX_BUFFER_INDEX32) ? 2 : 4;
-			uint32_t size = BX_ALIGN_16(_num*indexSize);
+			const uint32_t size = bx::alignUp(_num*indexSize, 16);
 
 			uint64_t ptr = 0;
 			if (0 != (_flags & BGFX_BUFFER_COMPUTE_READ_WRITE) )
@@ -3583,9 +3583,12 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 				uint16_t flags = BGFX_BUFFER_NONE;
 				cmdbuf.write(flags);
 
-				const uint32_t size = BX_ALIGN_16(sizeof(TransientIndexBuffer) ) + BX_ALIGN_16(_size);
+				const uint32_t size = 0
+					+ bx::alignUp(sizeof(TransientIndexBuffer), 16)
+					+ bx::alignUp(_size, 16)
+					;
 				tib = (TransientIndexBuffer*)BX_ALIGNED_ALLOC(g_allocator, size, 16);
-				tib->data   = (uint8_t *)tib + BX_ALIGN_16(sizeof(TransientIndexBuffer) );
+				tib->data   = (uint8_t *)tib + bx::alignUp(sizeof(TransientIndexBuffer), 16);
 				tib->size   = _size;
 				tib->handle = handle;
 
@@ -3644,9 +3647,12 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 				uint16_t flags = BGFX_BUFFER_NONE;
 				cmdbuf.write(flags);
 
-				const uint32_t size = BX_ALIGN_16(sizeof(TransientVertexBuffer) ) + BX_ALIGN_16(_size);
+				const uint32_t size = 0
+					+ bx::alignUp(sizeof(TransientVertexBuffer), 16)
+					+ bx::alignUp(_size, 16)
+					;
 				tvb = (TransientVertexBuffer*)BX_ALIGNED_ALLOC(g_allocator, size, 16);
-				tvb->data = (uint8_t *)tvb + BX_ALIGN_16(sizeof(TransientVertexBuffer) );
+				tvb->data = (uint8_t *)tvb + bx::alignUp(sizeof(TransientVertexBuffer), 16);
 				tvb->size = _size;
 				tvb->startVertex = 0;
 				tvb->stride = stride;
@@ -3700,8 +3706,8 @@ constexpr uint64_t kSortKeyComputeProgramMask  = uint64_t(BGFX_CONFIG_MAX_PROGRA
 		{
 			BGFX_MUTEX_SCOPE(m_resourceApiLock);
 
-			uint16_t stride = BX_ALIGN_16(_stride);
+			const uint16_t stride = bx::alignUp(_stride, 16);
-			uint32_t offset = m_submit->allocTransientVertexBuffer(_num, stride);
+			const uint32_t offset = m_submit->allocTransientVertexBuffer(_num, stride);
 
 			TransientVertexBuffer& dvb = *m_submit->m_transientVb;
 			_idb->data   = &dvb.data[offset];

src/renderer_d3d11.cpp

@@ -1903,7 +1903,7 @@ namespace bgfx { namespace d3d11
 				BX_FREE(g_allocator, m_uniforms[_handle.idx]);
 			}
 
-			uint32_t size = BX_ALIGN_16(g_uniformTypeSize[_type]*_num);
+			const uint32_t size = bx::alignUp(g_uniformTypeSize[_type]*_num, 16);
 			void* data = BX_ALLOC(g_allocator, size);
 			bx::memSet(data, 0, size);
 			m_uniforms[_handle.idx] = data;

src/renderer_d3d12.cpp

@@ -1789,7 +1789,7 @@ namespace bgfx { namespace d3d12
 				BX_FREE(g_allocator, m_uniforms[_handle.idx]);
 			}
 
-			uint32_t size = BX_ALIGN_16(g_uniformTypeSize[_type] * _num);
+			const uint32_t size = bx::alignUp(g_uniformTypeSize[_type] * _num, 16);
 			void* data = BX_ALLOC(g_allocator, size);
 			bx::memSet(data, 0, size);
 			m_uniforms[_handle.idx] = data;
@@ -3481,7 +3481,7 @@ namespace bgfx { namespace d3d12
 		_gpuAddress = m_gpuVA + m_pos;
 		void* data = &m_data[m_pos];
 
-		m_pos += BX_ALIGN_256(_size);
+		m_pos += bx::alignUp(_size, 256);
 
 //		D3D12_CONSTANT_BUFFER_VIEW_DESC desc;
 //		desc.BufferLocation = _gpuAddress;

src/renderer_d3d9.cpp

@@ -1213,7 +1213,7 @@ namespace bgfx { namespace d3d9
 				BX_FREE(g_allocator, m_uniforms[_handle.idx]);
 			}
 
-			uint32_t size = BX_ALIGN_16(g_uniformTypeSize[_type]*_num);
+			const uint32_t size = bx::alignUp(g_uniformTypeSize[_type]*_num, 16);
 			void* data = BX_ALLOC(g_allocator, size);
 			bx::memSet(data, 0, size);
 			m_uniforms[_handle.idx] = data;

src/renderer_mtl.h

@@ -913,9 +913,9 @@ namespace bgfx { namespace mtl
 		UniformBuffer* m_fshConstantBuffer;
 
 		uint32_t m_vshConstantBufferSize;
-		uint32_t m_vshConstantBufferAlignmentMask;
+		uint32_t m_vshConstantBufferAlignment;
 		uint32_t m_fshConstantBufferSize;
-		uint32_t m_fshConstantBufferAlignmentMask;
+		uint32_t m_fshConstantBufferAlignment;
 
 		enum
 		{

src/renderer_mtl.mm

@@ -540,12 +540,12 @@ namespace bgfx { namespace mtl
 				||  BX_ENABLED(BX_PLATFORM_OSX)
 				|| (BX_ENABLED(BX_PLATFORM_IOS) && iOSVersionEqualOrGreater("9.0.0") )
 				;
-			
+
 			m_hasStoreActionStoreAndMultisampleResolve = false
 			|| (BX_ENABLED(BX_PLATFORM_OSX) && macOSVersionEqualOrGreater(10,12,0))
 			|| (BX_ENABLED(BX_PLATFORM_IOS) && iOSVersionEqualOrGreater("10.0.0") )
 			;
-			
+
 			m_macOS11Runtime = true
 				&& BX_ENABLED(BX_PLATFORM_OSX)
 				&& macOSVersionEqualOrGreater(10,11,0)
@@ -967,7 +967,7 @@ namespace bgfx { namespace mtl
 				BX_FREE(g_allocator, m_uniforms[_handle.idx]);
 			}
 
-			uint32_t size = BX_ALIGN_16(g_uniformTypeSize[_type]*_num);
+			const uint32_t size = bx::alignUp(g_uniformTypeSize[_type]*_num, 16);
 			void* data = BX_ALLOC(g_allocator, size);
 			bx::memSet(data, 0, size);
 			m_uniforms[_handle.idx] = data;
@@ -1159,7 +1159,10 @@ namespace bgfx { namespace mtl
 
 				if (vertexUniformBufferSize)
 				{
-					m_uniformBufferVertexOffset = BX_ALIGN_MASK(m_uniformBufferVertexOffset, pso->m_vshConstantBufferAlignmentMask);
+					m_uniformBufferVertexOffset = bx::alignUp(
+						  m_uniformBufferVertexOffset
+						, pso->m_vshConstantBufferAlignment
+						);
 					rce.setVertexBuffer(m_uniformBuffer, m_uniformBufferVertexOffset, 0);
 				}
 
@@ -1167,7 +1170,10 @@ namespace bgfx { namespace mtl
 
 				if (0 != fragmentUniformBufferSize)
 				{
-					m_uniformBufferFragmentOffset = BX_ALIGN_MASK(m_uniformBufferFragmentOffset, pso->m_fshConstantBufferAlignmentMask);
+					m_uniformBufferFragmentOffset = bx::alignUp(
+						  m_uniformBufferFragmentOffset
+						, pso->m_fshConstantBufferAlignment
+						);
 					rce.setFragmentBuffer(m_uniformBuffer, m_uniformBufferFragmentOffset, 0);
 				}
 
@@ -1541,14 +1547,20 @@ namespace bgfx { namespace mtl
 
 			if (0 != vertexUniformBufferSize)
 			{
-				m_uniformBufferVertexOffset = BX_ALIGN_MASK(m_uniformBufferVertexOffset, pso->m_vshConstantBufferAlignmentMask);
+				m_uniformBufferVertexOffset = bx::alignUp(
+					  m_uniformBufferVertexOffset
+					, pso->m_vshConstantBufferAlignment
+					);
 				m_renderCommandEncoder.setVertexBuffer(m_uniformBuffer, m_uniformBufferVertexOffset, 0);
 			}
 
 			m_uniformBufferFragmentOffset = m_uniformBufferVertexOffset + vertexUniformBufferSize;
 			if (fragmentUniformBufferSize)
 			{
-				m_uniformBufferFragmentOffset = BX_ALIGN_MASK(m_uniformBufferFragmentOffset, pso->m_fshConstantBufferAlignmentMask);
+				m_uniformBufferFragmentOffset = bx::alignUp(
+					  m_uniformBufferFragmentOffset
+					, pso->m_fshConstantBufferAlignment
+					);
 				m_renderCommandEncoder.setFragmentBuffer(m_uniformBuffer, m_uniformBufferFragmentOffset, 0);
 			}
 
@@ -1802,13 +1814,13 @@ namespace bgfx { namespace mtl
 							{
 								if (shaderType == 0)
 								{
-									ps->m_vshConstantBufferSize = (uint32_t)arg.bufferDataSize;
+									ps->m_vshConstantBufferSize = uint32_t(arg.bufferDataSize);
-									ps->m_vshConstantBufferAlignmentMask = (uint32_t)arg.bufferAlignment - 1;
+									ps->m_vshConstantBufferAlignment = uint32_t(arg.bufferAlignment);
 								}
 								else
 								{
-									ps->m_fshConstantBufferSize = (uint32_t)arg.bufferDataSize;
+									ps->m_fshConstantBufferSize = uint32_t(arg.bufferDataSize);
-									ps->m_fshConstantBufferAlignmentMask = (uint32_t)arg.bufferAlignment - 1;
+									ps->m_fshConstantBufferAlignment = uint32_t(arg.bufferAlignment);
 								}
 
 								for (MTLStructMember* uniform in arg.bufferStructType.members )
@@ -3081,7 +3093,7 @@ namespace bgfx { namespace mtl
 							[contentView setLayer:m_metalLayer];
 						}
 					};
-					
+
 					if ([NSThread isMainThread])
 					{
 						setLayer();
@@ -3090,7 +3102,7 @@ namespace bgfx { namespace mtl
 					{
 						bx::Semaphore semaphore;
 						bx::Semaphore* psemaphore = &semaphore;
-						
+
 						CFRunLoopPerformBlock([[NSRunLoop mainRunLoop] getCFRunLoop],
 											  kCFRunLoopCommonModes,
 											  ^{
@@ -4075,7 +4087,10 @@ namespace bgfx { namespace mtl
 
 						if (0 != vertexUniformBufferSize)
 						{
-							m_uniformBufferVertexOffset = BX_ALIGN_MASK(m_uniformBufferVertexOffset, currentPso->m_vshConstantBufferAlignmentMask);
+							m_uniformBufferVertexOffset = bx::alignUp(
+								  m_uniformBufferVertexOffset
+								, currentPso->m_vshConstantBufferAlignment
+								);
 							m_computeCommandEncoder.setBuffer(m_uniformBuffer, m_uniformBufferVertexOffset, 0);
 						}
 
@@ -4440,14 +4455,20 @@ namespace bgfx { namespace mtl
 
 					if (0 != vertexUniformBufferSize)
 					{
-						m_uniformBufferVertexOffset = BX_ALIGN_MASK(m_uniformBufferVertexOffset, currentPso->m_vshConstantBufferAlignmentMask);
+						m_uniformBufferVertexOffset = bx::alignUp(
+							  m_uniformBufferVertexOffset
+							, currentPso->m_vshConstantBufferAlignment
+							);
 						rce.setVertexBuffer(m_uniformBuffer, m_uniformBufferVertexOffset, 0);
 					}
 
 					m_uniformBufferFragmentOffset = m_uniformBufferVertexOffset + vertexUniformBufferSize;
 					if (0 != fragmentUniformBufferSize)
 					{
-						m_uniformBufferFragmentOffset = BX_ALIGN_MASK(m_uniformBufferFragmentOffset, currentPso->m_fshConstantBufferAlignmentMask);
+						m_uniformBufferFragmentOffset = bx::alignUp(
+							  m_uniformBufferFragmentOffset
+							, currentPso->m_fshConstantBufferAlignment
+							);
 						rce.setFragmentBuffer(m_uniformBuffer, m_uniformBufferFragmentOffset, 0);
 					}
 

src/renderer_vk.cpp

@@ -2502,7 +2502,7 @@ VK_IMPORT_DEVICE
 				BX_FREE(g_allocator, m_uniforms[_handle.idx]);
 			}
 
-			uint32_t size = BX_ALIGN_16(g_uniformTypeSize[_type] * _num);
+			const uint32_t size = bx::alignUp(g_uniformTypeSize[_type] * _num, 16);
 			void* data = BX_ALLOC(g_allocator, size);
 			bx::memSet(data, 0, size);
 			m_uniforms[_handle.idx] = data;

tools/shaderc/shaderc_hlsl.cpp

@@ -479,7 +479,7 @@ namespace bgfx { namespace hlsl
 								un.type = uniformType;
 								un.num = uint8_t(constDesc.Elements);
 								un.regIndex = uint16_t(varDesc.StartOffset);
-								un.regCount = BX_ALIGN_16(varDesc.Size) / 16;
+								un.regCount = bx::alignUp(varDesc.Size, 16) / 16;
 								_uniforms.push_back(un);
 
 								BX_TRACE("\t%s, %d, size %d, flags 0x%08x, %d (used)"