[llvm] [SHT_LLVM_BB_ADDR_MAP] Adds test for when PGO exists but feature bit is off. (PR #77366)

[Micah Weston via llvm-commits]

================
@@ -897,6 +897,34 @@ TEST(ELFObjectFileTest, InvalidDecodePGOAnalysisMap) {
 
   DoCheck(MissingBrProb, "unable to decode LEB128 at offset 0x00000017: "
                          "malformed uleb128, extends past end");
+
+  // Check that we fail when pgo data exists but the feature bits are disabled.
+  SmallString<128> ZeroFeatureButWithAnalyses(CommonYamlString);
+  ZeroFeatureButWithAnalyses += R"(
+        Version: 2
+        Feature: 0x00
+        BBEntries:
+          - ID:            1
+            AddressOffset: 0x0
+            Size:          0x1
+            Metadata:      0x6
+          - ID:            2
+            AddressOffset: 0x1
+            Size:          0x1
+            Metadata:      0x2
+      - Address: 0x11111
+        Version: 2
+        Feature: 0x00
+        BBEntries: []
+    PGOAnalyses:
+      - PGOBBEntries:
+         - BBFreq:        1000
+         - BBFreq:        1000
+      - PGOBBEntries: []
+)";
+
+  DoCheck(ZeroFeatureButWithAnalyses,
+          "unsupported SHT_LLVM_BB_ADDR_MAP version: 232");
----------------
red1bluelost wrote:

That's only the case if `len(PGOAnalyses) != len(Entries)` which isn't the case here.

I could remove those returns in ELFEmitter so that it just emits Entries and PGOAnalyses interspersed until the shorter one runs out then just emit the longer one until the end. Do you think this is preferable for testing?

https://github.com/llvm/llvm-project/pull/77366