[turboprop] Fix DCHECK in SetOptimizedCode

With --always-opt it is possible that we set optimized code even
when the marker is LogFirstExecution.

Bug=v8:9684
TBR=rmcilroy@chromium.org

Change-Id: Ic45db4a6344a7345bee7ffd075b390d442c540fa
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2505710Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70844}

src/objects/feedback-vector.cc

@@ -391,12 +391,9 @@ void FeedbackVector::SetOptimizedCode(Handle<FeedbackVector> vector,
   // TODO(mythria): We could see a CompileOptimized marker here either from
   // tests that use %OptimizeFunctionOnNextCall or because we re-mark the
   // function for non-concurrent optimization after an OSR. We should avoid
-  // these cases and remove the CompileOptimized expectation from this DCHECK.
-  DCHECK(vector->optimization_marker() == OptimizationMarker::kNone ||
-         vector->optimization_marker() ==
-             OptimizationMarker::kInOptimizationQueue ||
-         vector->optimization_marker() ==
-             OptimizationMarker::kCompileOptimized);
+  // these cases and also check that marker isn't kCompileOptimized.
+  DCHECK(vector->optimization_marker() !=
+         OptimizationMarker::kCompileOptimizedConcurrent);
   vector->set_maybe_optimized_code(HeapObjectReference::Weak(*code));
   int32_t state = vector->flags();
   state = OptimizationTierBits::update(state, GetTierForCodeKind(code->kind()));