[Clang][Manual] Mention -fdebug-info-for-profiling.

[mbenfield]

Summary

This option is necessary for effective AutoFDO, and it also seems that
the create_llvm_prof converter won't work if the binary is compiled
without it.

clang/docs/UsersManual.rst

 to determine what the most executed areas of the code are.
 
 Using the data from a sample profiler requires some changes in the way
 a program is built. Before the compiler can use profiling information,
 the code needs to execute under the profiler. The following is the
 usual build cycle when using sample profilers for optimization:
 
 1. Build the code with source line table information. You can use all the
-   usual build flags that you always build your application with. The only
-   requirement is that you add ``-gline-tables-only`` or ``-g`` to the
-   command line. This is important for the profiler to be able to map
-   instructions back to source line locations.
+   usual build flags that you always build your application with. There are two
+   flags you must use. The first is ``-gline-tables-only`` or ``-g``. This is
+   important for the profiler to be able to map instructions back to source line
+   locations. The second is ``-fdebug-info-for-profiling``. This will add
+   discriminators so that different expressions on the same source line can be
+   distinguished, as well as some other changes making profiling more effective.
 
    .. code-block:: console
 
-     $ clang++ -O2 -gline-tables-only code.cc -o code
+     $ clang++ -O2 -fdebug-info-for-profiling -gline-tables-only code.cc -o code
 
 2. Run the executable under a sampling profiler. The specific profiler
    you use does not really matter, as long as its output can be converted
    into the format that the LLVM optimizer understands. Currently, there
    exists a conversion tool for the Linux Perf profiler
    (https://perf.wiki.kernel.org/), so these examples assume that you
    are using Linux Perf to profile your code.