transform/shader_io: Generate a wrapper function

This is a major reworking of this transform. The old transform code
was getting unwieldy, with part of the complication coming from the
handling of multiple return statements. By generating a wrapper
function instead, we can avoid a lot of this complexity.

The original entry point function is stripped of all shader IO
attributes (as well as `stage` and `workgroup_size`), but the body is
left unmodified. A new entry point wrapper function is introduced
which calls the original function, packing/unpacking the shader inputs
as necessary, and propagates the result to the corresponding shader
outputs.

The new code has been refactored to use a state object with the
different parts of the transform split into separate functions, which
makes it much more manageable.

Fixed: tint:1076
Bug: tint:920
Change-Id: I3490a0ea7a3509a4e198ce730e476516649d8d96
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60521
Auto-Submit: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: James Price <jrprice@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>

test/vk-gl-cts/non_robust_buffer_access/unexecuted_oob_overflow/0-opt.spvasm.expected.hlsl

@@ -60,10 +60,13 @@ struct tint_symbol_1 {
   uint3 gl_WorkGroupID_param : SV_GroupID;
 };
 
-[numthreads(4, 1, 1)]
-void main(tint_symbol_1 tint_symbol) {
-  const uint3 gl_WorkGroupID_param = tint_symbol.gl_WorkGroupID_param;
+void main_inner(uint3 gl_WorkGroupID_param) {
   gl_WorkGroupID = gl_WorkGroupID_param;
   main_1();
+}
+
+[numthreads(4, 1, 1)]
+void main(tint_symbol_1 tint_symbol) {
+  main_inner(tint_symbol.gl_WorkGroupID_param);
   return;
 }

test/vk-gl-cts/non_robust_buffer_access/unexecuted_oob_overflow/0-opt.spvasm.expected.msl

@@ -23,7 +23,7 @@ struct In0 {
   /* 0x0000 */ tint_array_wrapper_2 data_in0;
 };
 
-void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, thread uint3* const tint_symbol_2) {
+void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, thread uint3* const tint_symbol_1) {
   uint base_index_in = 0u;
   uint base_index_out = 0u;
   int index_in0 = 0;
@@ -34,9 +34,9 @@ void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, c
   int i = 0;
   int temp0 = 0;
   int temp1 = 0;
-  uint const x_58 = (*(tint_symbol_2)).x;
+  uint const x_58 = (*(tint_symbol_1)).x;
   base_index_in = (128u * x_58);
-  uint const x_61 = (*(tint_symbol_2)).x;
+  uint const x_61 = (*(tint_symbol_1)).x;
   base_index_out = (256u * x_61);
   index_in0 = 127;
   index_in1 = 383;
@@ -99,10 +99,14 @@ void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, c
   return;
 }
 
+void tint_symbol_inner(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, uint3 gl_WorkGroupID_param, thread uint3* const tint_symbol_2) {
+  *(tint_symbol_2) = gl_WorkGroupID_param;
+  main_1(x_13, x_17, x_15, x_19, tint_symbol_2);
+}
+
 kernel void tint_symbol(uint3 gl_WorkGroupID_param [[threadgroup_position_in_grid]], const device In2& x_13 [[buffer(2)]], const device In1& x_17 [[buffer(1)]], device Out0& x_15 [[buffer(3)]], const device In0& x_19 [[buffer(0)]]) {
   thread uint3 tint_symbol_3 = 0u;
-  tint_symbol_3 = gl_WorkGroupID_param;
-  main_1(x_13, x_17, x_15, x_19, &(tint_symbol_3));
+  tint_symbol_inner(x_13, x_17, x_15, x_19, gl_WorkGroupID_param, &(tint_symbol_3));
   return;
 }
 

test/vk-gl-cts/non_robust_buffer_access/unexecuted_oob_overflow/0-opt.wgsl.expected.hlsl

@@ -60,10 +60,13 @@ struct tint_symbol_1 {
   uint3 gl_WorkGroupID_param : SV_GroupID;
 };
 
-[numthreads(4, 1, 1)]
-void main(tint_symbol_1 tint_symbol) {
-  const uint3 gl_WorkGroupID_param = tint_symbol.gl_WorkGroupID_param;
+void main_inner(uint3 gl_WorkGroupID_param) {
   gl_WorkGroupID = gl_WorkGroupID_param;
   main_1();
+}
+
+[numthreads(4, 1, 1)]
+void main(tint_symbol_1 tint_symbol) {
+  main_inner(tint_symbol.gl_WorkGroupID_param);
   return;
 }

test/vk-gl-cts/non_robust_buffer_access/unexecuted_oob_overflow/0-opt.wgsl.expected.msl

@@ -23,7 +23,7 @@ struct In0 {
   /* 0x0000 */ tint_array_wrapper_2 data_in0;
 };
 
-void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, thread uint3* const tint_symbol_2) {
+void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, thread uint3* const tint_symbol_1) {
   uint base_index_in = 0u;
   uint base_index_out = 0u;
   int index_in0 = 0;
@@ -34,9 +34,9 @@ void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, c
   int i = 0;
   int temp0 = 0;
   int temp1 = 0;
-  uint const x_58 = (*(tint_symbol_2)).x;
+  uint const x_58 = (*(tint_symbol_1)).x;
   base_index_in = (128u * x_58);
-  uint const x_61 = (*(tint_symbol_2)).x;
+  uint const x_61 = (*(tint_symbol_1)).x;
   base_index_out = (256u * x_61);
   index_in0 = 127;
   index_in1 = 383;
@@ -99,10 +99,14 @@ void main_1(const device In2& x_13, const device In1& x_17, device Out0& x_15, c
   return;
 }
 
+void tint_symbol_inner(const device In2& x_13, const device In1& x_17, device Out0& x_15, const device In0& x_19, uint3 gl_WorkGroupID_param, thread uint3* const tint_symbol_2) {
+  *(tint_symbol_2) = gl_WorkGroupID_param;
+  main_1(x_13, x_17, x_15, x_19, tint_symbol_2);
+}
+
 kernel void tint_symbol(uint3 gl_WorkGroupID_param [[threadgroup_position_in_grid]], const device In2& x_13 [[buffer(2)]], const device In1& x_17 [[buffer(1)]], device Out0& x_15 [[buffer(3)]], const device In0& x_19 [[buffer(0)]]) {
   thread uint3 tint_symbol_3 = 0u;
-  tint_symbol_3 = gl_WorkGroupID_param;
-  main_1(x_13, x_17, x_15, x_19, &(tint_symbol_3));
+  tint_symbol_inner(x_13, x_17, x_15, x_19, gl_WorkGroupID_param, &(tint_symbol_3));
   return;
 }