[ASan] Remove sanity checks during annotation of contiguous container

[AdvenamTacet]

Summary

This revision removes sanity checks in
__sanitizer_annotate_contiguous_container.
(Changed them to DCHECK_EQ.)
Those checks may be problematic, if someone manually unpoisoned memory block.
Manual unpoisoning may be used if part of the program is not
instrumented.

Those checks are helpful while confirming correctness of ASan annotations
implementation.

Originally suggested here: https://reviews.llvm.org/D136765#4174546

[vitalybuka]

compiler-rt/lib/asan/asan_poisoning.cpp

454

DCHECK?

[vitalybuka]

compiler-rt/lib/asan/asan_poisoning.cpp

454

DCHECK?

Another option could be full repainting of the container.

[AdvenamTacet]

Changed to DCHECK.

[AdvenamTacet]

compiler-rt/lib/asan/asan_poisoning.cpp

454

DCHECK is for sure good when writing annotations for libc++ containers. Repainting (if that check fails) may be a nice option as well, but I don't have a strong opinion.

[AdvenamTacet]

@vitalybuka could you land that revision as I don't have commiter rights?

[AdvenamTacet]

compiler-rt/lib/asan/asan_poisoning.cpp

454

Repainting may have terrible performance in some cases, so it would require some way to turn it off. After looking at it closer, I don't really like it Yes, in 95% of cases it's ok solution, but if someone was able to unpoison memory area, should be able to poison as well. Let's not force it.

[AdvenamTacet]

Rebase.

compiler-rt/lib/asan/asan_poisoning.cpp

void __sanitizer_annotate_contiguous_container(const void *beg_p,
   // uptr d2 = RoundUpTo(old_mid, granularity);
   // Currently we should be in this state:
   // [a, d1) is good, [d2, c) is bad, [d1, d2) is partially good.
   // Make a quick sanity check that we are indeed in this state.
   //
   // FIXME: Two of these three checks are disabled until we fix
   // https://github.com/google/sanitizers/issues/258.
   // if (d1 != d2)
-  //  CHECK_EQ(*(u8*)MemToShadow(d1), old_mid - d1);
+  //  DCHECK_EQ(*(u8*)MemToShadow(d1), old_mid - d1);
   if (a + granularity <= d1)
-    CHECK_EQ(*(u8 *)MemToShadow(a), 0);
+    DCHECK_EQ(*(u8 *)MemToShadow(a), 0);
   // if (d2 + granularity <= c && c <= end)
-  //   CHECK_EQ(*(u8 *)MemToShadow(c - granularity),
+  //   DCHECK_EQ(*(u8 *)MemToShadow(c - granularity),
   //            kAsanContiguousContainerOOBMagic);
 
   uptr b1 = RoundDownTo(new_end, granularity);
   uptr b2 = RoundUpTo(new_end, granularity);
   // New state:
   // [a, b1) is good, [b2, c) is bad, [b1, b2) is partially good.
   if (b1 > a)
     PoisonShadow(a, b1 - a, 0);