diff --git a/libuavcan/include/uavcan/marshal/char_array_formatter.hpp b/libuavcan/include/uavcan/marshal/char_array_formatter.hpp
index 27fbb84a12..40e2dc6a8b 100644
--- a/libuavcan/include/uavcan/marshal/char_array_formatter.hpp
+++ b/libuavcan/include/uavcan/marshal/char_array_formatter.hpp
@@ -126,7 +126,6 @@ public:
      * This version does not support more than one formatted argument, though it can be improved.
      * And it is unsafe.
      * There is typesafe version for C++11 above.
-     * TODO: make this version typesafe and add support for multiple args.
      */
     template <typename A>
     void write(const char* const format, const A value)
diff --git a/libuavcan/src/driver/uc_can.cpp b/libuavcan/src/driver/uc_can.cpp
index 5df37bf28b..974de88359 100644
--- a/libuavcan/src/driver/uc_can.cpp
+++ b/libuavcan/src/driver/uc_can.cpp
@@ -86,7 +86,6 @@ std::string CanFrame::toString(StringRepresentation mode) const
     }
     else if (id & FlagERR)
     {
-        // TODO: print error flags
         wpos += snprintf(wpos, epos - wpos, " ERR");
     }
     else
diff --git a/libuavcan/src/marshal/uc_float_spec.cpp b/libuavcan/src/marshal/uc_float_spec.cpp
index acfaaf65b2..e09e94b052 100644
--- a/libuavcan/src/marshal/uc_float_spec.cpp
+++ b/libuavcan/src/marshal/uc_float_spec.cpp
@@ -23,7 +23,6 @@ namespace uavcan
 /*
  * IEEE754Converter
  * Float16 conversion algorithm: http://half.sourceforge.net/ (MIT License)
- * TODO: Use conversion tables (conditional compilation - it would require something like 10Kb+ ROM).
  */
 template <typename T>
 static inline bool signbit(T arg)