[llvm] [BOLT] Check if symbol is in data area of function (PR #160143)

[Asher Dobrescu via llvm-commits]

https://github.com/Asher8118 created https://github.com/llvm/llvm-project/pull/160143

There are cases in which `getEntryIDForSymbol` is called, where the given Symbol is in a constant island, and so BOLT can not find its function. This causes BOLT to reach `llvm_unreachable("symbol not found")` and crash. This patch adds a check that avoids this crash and gives a warning to the user.

>From 36823a0c80776a97b6579baeadbee0e762377720 Mon Sep 17 00:00:00 2001
From: Ash Dobrescu <ash.dobrescu at arm.com>
Date: Mon, 22 Sep 2025 16:14:39 +0000
Subject: [PATCH] [BOLT] Check if symbol is in data area of function

There are cases in which `getEntryIDForSymbol` is called,
where the given Symbol is in a constant island, and so BOLT
can not find its function. This causes BOLT to reach
`llvm_unreachable("symbol not found")` and crash. This patch adds
a check that avoid this crash and gives a warning to the user.
---
 bolt/lib/Core/BinaryContext.cpp       | 10 +++++-
 bolt/test/AArch64/check-symbol-area.s | 49 +++++++++++++++++++++++++++
 2 files changed, 58 insertions(+), 1 deletion(-)
 create mode 100644 bolt/test/AArch64/check-symbol-area.s

diff --git a/bolt/lib/Core/BinaryContext.cpp b/bolt/lib/Core/BinaryContext.cpp
index 72c72bbaf4a65..5f3b3c0e57152 100644
--- a/bolt/lib/Core/BinaryContext.cpp
+++ b/bolt/lib/Core/BinaryContext.cpp
@@ -2374,8 +2374,16 @@ BinaryFunction *BinaryContext::getFunctionForSymbol(const MCSymbol *Symbol,
     return nullptr;
 
   BinaryFunction *BF = BFI->second;
-  if (EntryDesc)
+  if (EntryDesc) {
+    const uint64_t Address = BF->getAddress() + Symbol->getOffset();
+    if (BF->isInConstantIsland(Address)) {
+      this->outs() << "BOLT-WARNING: symbol " << Symbol->getName()
+                   << " is in data region of function 0x"
+                   << Twine::utohexstr(Address) << ".\n";
+      return nullptr;
+    }
     *EntryDesc = BF->getEntryIDForSymbol(Symbol);
+  }
 
   return BF;
 }
diff --git a/bolt/test/AArch64/check-symbol-area.s b/bolt/test/AArch64/check-symbol-area.s
new file mode 100644
index 0000000000000..43ce25e16181f
--- /dev/null
+++ b/bolt/test/AArch64/check-symbol-area.s
@@ -0,0 +1,49 @@
+// This test checks that when looking for a function
+// correspnding to a symbol, BOLT is not looking 
+// through a data area (constant island).
+
+# RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown %s -o %t.o
+# RUN: %clang %cflags %t.o -o %t.exe -Wl,-q
+# RUN: llvm-bolt %t.exe -o %t.bolt 2>&1 | FileCheck %s
+
+// Before adding a check for constant islands, BOLT would exit with an error
+// of the form: "symbol not found" and throw an LLVM UNREACHABLE error.
+# CHECK-NOT: symbol not found
+# CHECK-NOT: UNREACHABLE
+
+// Now BOLT throws a warning and does not crash.
+# CHECK: BOLT-WARNING: symbol [[SYM:.*]]  is in data region of function 0x{{.*}}.
+
+.text
+.global main
+main:
+        stp     x14, x15, [sp, -8]!
+        mov     x14, sp
+        adrp    x1, .test
+        add     x0, x1, :lo12:.test
+        bl      first_block
+        ret
+
+.global first_block
+$d:
+first_block:
+        stp     x14, x15, [sp, -8]!
+        mov     x14, sp
+        bl      second_block
+        ret
+second_block:
+        stp     x14, x15, [sp, -8]!
+        mov     x14, sp
+        bl      third_block
+        ret
+$x:
+third_block:
+        stp     x14, x15, [sp, -8]!
+        mov     x14, sp
+        adrp    x1, .data
+        add     x0, x1, :lo12:.test
+        ret
+
+.data
+.test:
+        .string "test"