From 6f681279d99e72e717bb2b91763b80e570ae725b Mon Sep 17 00:00:00 2001
From: Tim Foley <tfoleyNV@users.noreply.github.com>
Date: Wed, 20 Dec 2017 17:35:10 -0800
Subject: IR: fixes for subscript accessors (#322)

* IR: fixes for subscript accessors

Fixes #320

This is a bunch of fixes for handling of `__subscript` operations on builtin types (notably `RWStructuredBuffer` and `StructuredBuffer` at this point).

- Automatically add a `GetterDecl` to any subscript decalratio was declithout any accessors. This avoids hitting a null- dereference in the emit logic.

- Add a notion of a `RefAccessor` (declared with `ref`) as a peer to getters and setters. The idea is that a `ref` accessor returns a pointer to the element data, so that it can be used for both getting and setting values. This is closer to the behavior of `RWStructuredBuffer` element access in HLSL.

- Fixes for dealing with "access chains" where there might be a combination of a subscript (where the is a `get` and `set` but no `ref`) and member access, so that we have to read the base value into a temp, modify it, and then write it back.
  - This logic is still a bit of a mess, so we will eventually want to take a more consistent pass over this to deal with how we "materialize" values for setters.

- Update `RWStructuredBuffer` to have a `ref` accessor, and then fix up the IR tests to handle the new opcode that I added for it.
  - Note: I didn't handle this as an intrinsic simply because the `tests/ir/*` tests aren't really set up to handle builtins with ugly mangled names.

* Fixup: type error in VM for buffer element ref

I was using the result type of the op as the element type for computing the element address, but the result type is a pointer to the real element type.
This caused test failures on 64-bit platforms, where the stride of the buffer in the `ir/factorial` test needs to be 4.

The fix is to assume the result type is a pointer, and extract the pointed-to type out of that.
---
 tests/compute/write-structured-buffer-field.slang  | 19 +++++++++++++
 ...rite-structured-buffer-field.slang.expected.txt | 32 ++++++++++++++++++++++
 2 files changed, 51 insertions(+)
 create mode 100644 tests/compute/write-structured-buffer-field.slang
 create mode 100644 tests/compute/write-structured-buffer-field.slang.expected.txt

(limited to 'tests/compute')

diff --git a/tests/compute/write-structured-buffer-field.slang b/tests/compute/write-structured-buffer-field.slang
new file mode 100644
index 000000000..b824644b2
--- /dev/null
+++ b/tests/compute/write-structured-buffer-field.slang
@@ -0,0 +1,19 @@
+//TEST(compute):COMPARE_COMPUTE:-xslang -use-ir
+
+//TEST_INPUT:ubuffer(data=[1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32], stride=32):dxbinding(0),glbinding(0),out
+
+struct S
+{
+	int4 a;
+	int4 b;
+};
+
+RWStructuredBuffer<S> outputBuffer;
+
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+	uint tid = dispatchThreadID.x;
+
+	outputBuffer[tid].b.y = outputBuffer[0].a.x;
+}
\ No newline at end of file
diff --git a/tests/compute/write-structured-buffer-field.slang.expected.txt b/tests/compute/write-structured-buffer-field.slang.expected.txt
new file mode 100644
index 000000000..475b4f578
--- /dev/null
+++ b/tests/compute/write-structured-buffer-field.slang.expected.txt
@@ -0,0 +1,32 @@
+1
+2
+3
+4
+5
+1
+7
+8
+9
+A
+B
+C
+D
+1
+F
+10
+11
+12
+13
+14
+15
+1
+17
+18
+19
+1A
+1B
+1C
+1D
+1
+1F
+20
-- 
cgit v1.2.3