Update docs for v0.1 (#109)

* Add Aliza as contributor, remove CPU/Mem usage from upcoming features

* Add more Upcoming features

* Remove AOP terms from Basic Tutorial page as discussed, and add reference to slide deck (waiting for IM).

* Formatted

* require python 3.8

* Add Dan and Jae-Seung to Contributors

* Add slide deck with IM, update link in BasicTutorial

* minor updates, add Hari as contributor

---------

Co-authored-by: Stephanie Brink <[email protected]>

Author: tpatki

docs/BasicTutorial.rst

@@ -9,29 +9,13 @@
  Basic Tutorial
 ################
 
-PerfFlowAspect is based on Aspect-Oriented Programming (AOP). Some common terms
-used in AOP are described below.
-
--  **Join Points**: A point in a program's execution in which the behavior can
-   be modified by AOP. The candidate points are method invocations.
-
--  **Pointcuts**: An expression used to match join points (creating a set of
-   join points), allows you to specify to AOP where and when in the code to make
-   modifications.
-
--  **Advice**: Defines the additional behavior or action that will be inserted
-   into the code, specifically at each join point matched by the pointcut.
-
--  **Aspect**: The collection of the pointcut expression and the advice.
-
--  **Weaver**: Applies aspects into the code, modifying code at join points with
-   matching pointcuts and associated advices. The combining of aspects and code
-   enables execution of cross-cutting concerns.
-
-PerfFlowAspect relies on annotated functions in the user's source code and can
-invoke specific performance-analysis actions, a piece of tracing code, etc. on
-those points of execution. In AOP, these trigger points are called join points
-in the source code, and the functionality invoked is called advice.
+PerfFlowAspect is based on Aspect-Oriented Programming (AOP). PerfFlowAspect
+relies on annotated functions in the user's source code and can invoke specific
+performance-analysis actions, a piece of tracing code, etc. on those points of
+execution. In AOP, these trigger points are called join points in the source
+code, and the functionality invoked is called `advice`. To learn more about AOP
+and associated terminology, please refer to our presentation slides `here
+<ppts/PerfFlowAspectTutorialSlides.pdf>`_.
 
 The python package ``perfflowaspect`` contains the PerfFlowAspect tool for the
 Python language. The file ``src/python/perfflowaspect/aspect.py`` contains a key
@@ -123,7 +107,7 @@ variable. Separate multiple variables with a colon as follows:
 **********************************************
 
 There are two types of logging allowed in PerfFlowAspect trace files which are
-``verbose`` and ``compact``. Either can be anabled by setting
+``verbose`` and ``compact``. Either can be enabled by setting
 ``PERFFLOW_OPTIONS="log-event="`` to ``compact`` or ``verbose``, respectively.
 The logging is ``verbose`` by default. ``Verbose`` logging uses B (begin) and E
 (end) events in the trace file as shown below:

docs/BuildingPerfFlowAspect.rst

@@ -13,7 +13,7 @@
  Python Install
 ****************
 
-The minimum Python version needed is 3.6. You can get PerfFlowAspect from its
+The minimum Python version needed is 3.8. You can get PerfFlowAspect from its
 GitHub repository using this command:
 
 .. code:: bash

docs/UpcomingFeatures.rst

@@ -9,8 +9,9 @@
  Upcoming Features
 ###################
 
-Upcoming features include the ability to enable and disable PerfFlowAspect with
-an environment variable, and also allow for collection of other statistics, such
-as CPU usage, memory usage, and GPU utilization. Please follow our GitHub issues
-page for learning about upcoming features, as well as for suggesting new
-features for PerfFlowAspect.
+Upcoming features include the ability to specify categories while tracing, a
+connector for Hatchet, and allow for collection of other statistics, such as GPU
+utilization. Additionally, the team plans to provide examples of how various
+benchmarks and workflows have been annotated with PerfFlowAspect. Please follow
+our GitHub issues page for learning about upcoming features, as well as for
+suggesting new features for PerfFlowAspect.

docs/index.rst

@@ -60,8 +60,12 @@ uniformity as to how performance is measured and controlled.
 -  Stephanie Brink
 -  James Corbett
 -  Stephen Herbein (NVIDIA)
+-  Aliza Lisan (University of Oregon)
+-  Daniel Milroy
 -  Francesco Di Natale (NVIDIA)
 -  Tapasya Patki
+-  Jae-Seung Yeom
+-  Hariharan Devarajan
 
 ******************************
  PerfFlowAspect Documentation