[PATCH] D53967: [XRay] Fix TSC and atomic custom/typed event accounting

[Marcus Boerger via Phabricator via llvm-commits]

mboerger added inline comments.


================
Comment at: compiler-rt/lib/xray/tests/unit/fdr_controller_test.cc:126
+
+  // Serialize the buffers then test to see we find the *no* records, because
+  // the function entry-exit comes under the cycle threshold.
----------------
to see if...


================
Comment at: compiler-rt/lib/xray/tests/unit/fdr_log_writer_test.cc:102
+
+  // THen we re-use the buffer, but only write one record.
+  ASSERT_EQ(Buffers.getBuffer(B), BufferQueue::ErrorCode::Ok);
----------------
s/THen/Then/
where is first?


================
Comment at: compiler-rt/lib/xray/tests/unit/fdr_log_writer_test.cc:112
+  ASSERT_EQ(Buffers.releaseBuffer(B), BufferQueue::ErrorCode::Ok);
+  ASSERT_EQ(B.Data, nullptr);
+  ASSERT_EQ(Buffers.finalize(), BufferQueue::ErrorCode::Ok);
----------------
ASSERT_THAT(B.Data, IsNull());


================
Comment at: compiler-rt/lib/xray/xray_fdr_log_writer.h:116
+    // We write the metadata record and the custom event data into the buffer
+    // first before we atomically update the extents for the buffer. This allows
+    // us to ensure that any threads reading the extents of the buffer will only
----------------
s/first before/first, before/


https://reviews.llvm.org/D53967