浏览代码

Add tasks_size_overhead output example

Samuel Thibault 12 年之前
父节点
当前提交
b8f821babe

+ 3 - 1
doc/doxygen/Makefile.am

@@ -1,6 +1,6 @@
 # StarPU --- Runtime system for heterogeneous multicore architectures.
 # StarPU --- Runtime system for heterogeneous multicore architectures.
 #
 #
-# Copyright (C) 2009, 2011  Université de Bordeaux 1
+# Copyright (C) 2009, 2011, 2013  Université de Bordeaux 1
 # Copyright (C) 2010, 2011, 2012, 2013  Centre National de la Recherche Scientifique
 # Copyright (C) 2010, 2011, 2012, 2013  Centre National de la Recherche Scientifique
 #
 #
 # Permission is granted to copy, distribute and/or modify this document
 # Permission is granted to copy, distribute and/or modify this document
@@ -138,6 +138,8 @@ EXTRA_DIST	= 		\
 	$(chapters) 		\
 	$(chapters) 		\
 	chapters/version.sty	\
 	chapters/version.sty	\
 	chapters/version.html	\
 	chapters/version.html	\
+	chapters/tasks_size_overhead.png	\
+	chapters/tasks_size_overhead.eps	\
 	doxygen.cfg 		\
 	doxygen.cfg 		\
 	refman.tex
 	refman.tex
 
 

+ 5 - 2
doc/doxygen/chapters/building.doxy

@@ -266,11 +266,14 @@ schedulers, for instance <c>STARPU_SCHED=dmda</c>.
 
 
 \subsection TaskSizeOverhead Task Size Overhead
 \subsection TaskSizeOverhead Task Size Overhead
 
 
-This benchmark gives a glimpse into how big a size should be for StarPU overhead
-to be low enough.  Run <c>tasks_size_overhead.sh</c>, it will generate a plot
+This benchmark gives a glimpse into how long a task should be (in µs) for StarPU overhead
+to be low enough to keep efficiency.  Run <c>tasks_size_overhead.sh</c>, it will generate a plot
 of the speedup of tasks of various sizes, depending on the number of CPUs being
 of the speedup of tasks of various sizes, depending on the number of CPUs being
 used.
 used.
 
 
+\image html tasks_size_overhead.png
+\image latex tasks_size_overhead.eps "" width=\textwidth
+
 \subsection DataTransferLatency Data Transfer Latency
 \subsection DataTransferLatency Data Transfer Latency
 
 
 <c>local_pingpong</c> performs a ping-pong between the first two CUDA nodes, and
 <c>local_pingpong</c> performs a ping-pong between the first two CUDA nodes, and

文件差异内容过多而无法显示
+ 1253 - 0
doc/doxygen/chapters/tasks_size_overhead.eps


二进制
doc/doxygen/chapters/tasks_size_overhead.png


+ 1 - 0
doc/doxygen/doxygen-config.cfg.in

@@ -63,3 +63,4 @@ INPUT_FILTER           = @top_builddir@/doc/doxygen/doxygen_filter.sh
 
 
 LATEX_HEADER           = @top_srcdir@/doc/doxygen/refman.tex
 LATEX_HEADER           = @top_srcdir@/doc/doxygen/refman.tex
 
 
+IMAGE_PATH             = @top_srcdir@/doc/doxygen/chapters

+ 2 - 1
doc/doxygen/doxygen.cfg

@@ -774,7 +774,8 @@ EXAMPLE_RECURSIVE      = NO
 # directories that contain image that are included in the documentation (see
 # directories that contain image that are included in the documentation (see
 # the \image command).
 # the \image command).
 
 
-IMAGE_PATH             =
+# From @INCLUDE, above
+#IMAGE_PATH             =
 
 
 # The INPUT_FILTER tag can be used to specify a program that doxygen should
 # The INPUT_FILTER tag can be used to specify a program that doxygen should
 # invoke to filter for each input file. Doxygen will invoke the filter program
 # invoke to filter for each input file. Doxygen will invoke the filter program