[scudo] Always exclude Tag 0

[vitalybuka]

Summary

prepareTaggedChunk uses Tag 0 for header.

Android already PR_MTE_TAG_MASK to 0xfffe,
but with the patch we will not need to deppend
on the system configuration.

[pcc]

There are probably going to be more callers of selectRandomTag and I don't think we want to have to deal with tag 0 in any of them. Could we just or 1 into the mask in selectRandomTag?

[vitalybuka]

Could we just or 1 into the mask in selectRandomTag?

done

[vitalybuka]

ping

[pcc]

LGTM

compiler-rt/lib/scudo/standalone/combined.h

#ifdef HAVE_ANDROID_UNSAFE_FRAME_POINTER_CHASE
     return Depot.insert(Stack + Min<uptr>(DiscardFrames, Size), Stack + Size);
 #else
     return 0;
 #endif
   }
 
   uptr computeOddEvenMaskForPointerMaybe(Options Options, uptr Ptr,
                                          uptr ClassId) {
+    const uptr Tag = 1; // Always exclude Tag 0 is may be used for headers.
     if (!Options.get(OptionBit::UseOddEvenTags))
-      return 0;
+      return Tag;
 
     // If a chunk's tag is odd, we want the tags of the surrounding blocks to be
     // even, and vice versa. Blocks are laid out Size bytes apart, and adding
     // Size to Ptr will flip the least significant set bit of Size in Ptr, so
     // that bit will have the pattern 010101... for consecutive blocks, which we
     // can use to determine which tag mask to use.
-    return 0x5555U << ((Ptr >> SizeClassMap::getSizeLSBByClassId(ClassId)) & 1);
+    return Tag |
+           (0x5555U << ((Ptr >> SizeClassMap::getSizeLSBByClassId(ClassId)) &
+                        1));
   }
 
   NOINLINE void *allocate(uptr Size, Chunk::Origin Origin,
                           uptr Alignment = MinAlignment,
                           bool ZeroContents = false) {
     initThreadMaybe();
 
     const Options Options = Primary.Options.load();