From 4639af8b0e5d6ddad7461d1a4155430a2e2f20dd Mon Sep 17 00:00:00 2001
From: Andy McFadden <thefadden@gmail.com>
Date: Fri, 5 Oct 2018 11:10:03 -0700
Subject: [PATCH] Strip single-byte format items on single-byte instructions

It's possible to have format descriptors on instructions that are
left over from when the bytes were treated as data.  Single-byte
formats were being allowed on single-byte instructions, which
confused things later when the code tried to apply the format to
an instruction with no operand.
---
 SourceGen/DisasmProject.cs | 17 +++++++++++++----
 SourceGen/DisplayList.cs   |  5 ++++-
 2 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/SourceGen/DisasmProject.cs b/SourceGen/DisasmProject.cs
index e73f15a..b32b77d 100644
--- a/SourceGen/DisasmProject.cs
+++ b/SourceGen/DisasmProject.cs
@@ -532,20 +532,29 @@ namespace SourceGen {
                 }
 
                 if (mAnattribs[offset].IsInstructionStart) {
-                    // Check length for instruction formatters.
+                    // Check length for instruction formatters.  This can happen if you format
+                    // a bunch of bytes as single-byte data items and then add a code entry
+                    // point.
                     if (kvp.Value.Length != mAnattribs[offset].Length) {
-                        genLog.LogW("Unexpected length on instr format descriptor (" +
+                        genLog.LogW("+" + offset.ToString("x6") +
+                            ": unexpected length on instr format descriptor (" +
                             kvp.Value.Length + " vs " + mAnattribs[offset].Length + ")");
                         continue;       // ignore this one
                     }
+                    if (kvp.Value.Length == 1) {
+                        // No operand to format!
+                        genLog.LogW("+" + offset.ToString("x6") +
+                            ": unexpected format descriptor on single-byte op");
+                        continue;       // ignore this one
+                    }
                     if (!kvp.Value.IsValidForInstruction) {
                         genLog.LogW("Descriptor not valid for instruction: " + kvp.Value);
                         continue;       // ignore this one
                     }
                 } else if (mAnattribs[offset].IsInstruction) {
                     // Mid-instruction format.
-                    genLog.LogW("Unexpected mid-instruction format descriptor at +" +
-                        offset.ToString("x6"));
+                    genLog.LogW("+" + offset.ToString("x6") +
+                        ": unexpected mid-instruction format descriptor");
                     continue;       // ignore this one
                 }
 
diff --git a/SourceGen/DisplayList.cs b/SourceGen/DisplayList.cs
index 4d2858e..ba0086f 100644
--- a/SourceGen/DisplayList.cs
+++ b/SourceGen/DisplayList.cs
@@ -1096,8 +1096,11 @@ namespace SourceGen {
                 operandForSymbol = attr.OperandAddress;
             }
 
-            // Check Length to watch for bogus descriptors (?)
+            // Check Length to watch for bogus descriptors.  ApplyFormatDescriptors() should
+            // have discarded anything appropriate, so we might be able to eliminate this test.
             if (attr.DataDescriptor != null && attr.Length == attr.DataDescriptor.Length) {
+                Debug.Assert(operandLen > 0);
+
                 // Format operand as directed.
                 if (op.AddrMode == OpDef.AddressMode.BlockMove) {
                     // Special handling for the double-operand block move.