Clean up README.md. Add another cmake option that feeds into

conditional compilation.

Signed-off-by: J. Eric Ivancich <[email protected]>

CMakeLists.txt

@@ -17,6 +17,10 @@ if(PROFILE)
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DPROFILE")
 endif()
 
+if(DO_NOT_DELAY_TAG_CALC)
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DDO_NOT_DELAY_TAG_CALC")
+endif()
+
 add_subdirectory(src)
 add_subdirectory(sim)
 

README.md

@@ -1,7 +1,7 @@
 # dmclock
 
 This repository contains C++ 11 code that implements the dmclock
-distributed quality of service algorithm. See __mClock: Hanling
+distributed quality of service algorithm. See __mClock: Handling
 Throughput Variability for Hypervisor IO Scheduling__ by Gulati,
 Merchant, and Varman for a description of the algorithm.
 
@@ -16,13 +16,18 @@ To turn on profiling, run cmake with an additional:
 
     -DPROFILE=yes
 
+An optimization/fix to the published algorithm has been added and is
+on by default. To disable this optimization/fix run cmake with:
+
+    -DDO_NOT_DELAY_TAG_CALC=yes
+
 ## Running make
 
 ### Building the dmclock library
 
 The `make` command builds a library libdmclock.a. That plus the header
-files in the src directory allow one to include the algorithm in their
-code.
+files in the src directory allow one to use the implementation in
+their code.
 
 ### Building unit tests
 
@@ -35,6 +40,6 @@ and *ssched_sim* -- which incorporate, respectively, the dmclock
 priority queue or a very simple scheduler for comparison. Other
 priority queue implementations could be added in the future.
 
-## Using dmclock
+## dmclock API
 
-To be written....
+To be written....