From 0ca87a4807b4d26baef3b7813e1b948b88dddf79 Mon Sep 17 00:00:00 2001
From: David Guillen Fandos
Date: Mon, 5 Jul 2021 18:19:19 +0200
Subject: Fix conditional ARM instructions at the end of a translation block

This fixes issue #133

The explanation is as follows. Most blocks end on an inconditional
jump/branch, but there's two cases where this doesn't happen:
translation gates and when we hit MAX_EXITS. These are very uncommon
cases and therefore more prone to hidden bugs.

When this happens, the last instruction emits a conditional jump (via
arm_conditional_block_header macro) which is patched by a later
instruction via generate_branch_patch_conditional. Typically the last
unconditional branch will trigger the patching condition (which is
aproximately condition != last_condition), but in these two cases it
might not happen, leaving an unpatched branch. This makes x86 and ARM
dynarecs crash in interesting ways (although it might not crash
depending on $stuff and make the bug even harder to track).
---
 cpu_threaded.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/cpu_threaded.c b/cpu_threaded.c
index 557b197..2d61b4a 100644
--- a/cpu_threaded.c
+++ b/cpu_threaded.c
@@ -3115,6 +3115,12 @@ s32 translate_block_arm(u32 pc, translation_region_type
       generate_cycle_update();
     }
   }
+
+  /* This can happen if the last instruction is *not* inconditional */
+  if ((last_condition & 0x0F) != 0x0E) {
+    generate_branch_patch_conditional(backpatch_address, translation_ptr);
+  }
+
   for(i = 0; i < translation_gate_targets; i++)
   {
     if(pc == translation_gate_target_pc[i])
-- 
cgit v1.2.3