mention windows home dir

doc/chapters/advanced-examples.texi

@@ -317,8 +317,9 @@ few different sets of data input/output sizes. StarPU will then keep record of
 the average time of previous executions on the various processing units, and use
 it as an estimation. History is done per task size, by using a hash of the input
 and ouput sizes as an index.
-It will also save it in @code{~/.starpu/sampling/codelets}
-for further executions, and can be observed by using the
+It will also save it in @code{$HOME/.starpu/sampling/codelets}
+for further executions (@code{$USERPROFILE/.starpu/sampling/codelets} in windows
+environments), and can be observed by using the
 @code{starpu_perfmodel_display} command, or drawn by using
 the @code{starpu_perfmodel_plot} (@pxref{Performance model calibration}).  The
 models are indexed by machine name. To

doc/chapters/basic-api.texi

@@ -1934,7 +1934,7 @@ regression.
 @end deftp
 
 @deftypefun int starpu_load_history_debug ({const char} *@var{symbol}, {struct starpu_perfmodel} *@var{model})
-loads a given performance model. The @var{model} structure has to be completely zero, and will be filled with the information saved in @code{~/.starpu}.
+loads a given performance model. The @var{model} structure has to be completely zero, and will be filled with the information saved in @code{$HOME/.starpu} (@code{$USERPROFILE/.starpu} in windows environments).
 @end deftypefun
 
 @deftypefun void starpu_perfmodel_debugfilepath ({struct starpu_perfmodel} *@var{model}, {enum starpu_perf_archtype} @var{arch}, char *@var{path}, size_t @var{maxlen}, unsigned nimpl)
@@ -2347,7 +2347,8 @@ to the empty string.
 Compile the OpenCL kernel stored in the file @code{source_file_name}
 with the given options @code{build_options} and stores the result in
 the directory @code{$STARPU_HOME/.starpu/opencl} with the same
-filename as @code{source_file_name}. The compilation is done for every
+filename as @code{source_file_name} (@code{$USERPROFILE/.starpu/opencl} in
+windows environments). The compilation is done for every
 OpenCL device, and the filename is suffixed with the vendor id and the
 device id of the OpenCL device.
 @end deftypefun
@@ -2355,7 +2356,8 @@ device id of the OpenCL device.
 @deftypefun int starpu_opencl_compile_opencl_from_string ({const char *}@var{opencl_program_source}, {const char *}@var{file_name}, {const char* }@var{build_options})
 Compile the OpenCL kernel in the string @code{opencl_program_source}
 with the given options @code{build_options} and stores the result in
-the directory @code{$STARPU_HOME/.starpu/opencl} with the filename
+the directory @code{$STARPU_HOME/.starpu/opencl}
+(@code{$USERPROFILE/.starpu/opencl} in windows environments) with the filename
 @code{file_name}. The compilation is done for every
 OpenCL device, and the filename is suffixed with the vendor id and the
 device id of the OpenCL device.

doc/chapters/perf-optimization.texi

@@ -166,7 +166,7 @@ to configure a performance model for the codelets of the application (see
 @ref{Performance model example} for instance). History-based performance models
 use on-line calibration.  StarPU will automatically calibrate codelets
 which have never been calibrated yet, and save the result in
-@code{~/.starpu/sampling/codelets}.
+@code{~/.starpu/sampling/codelets} (@code{$USERPROFILE/.starpu/sampling/codelets} in windows environments)
 The models are indexed by machine name. To share the models between machines (e.g. for a homogeneous cluster), use @code{export STARPU_HOSTNAME=some_global_name}. To force continuing calibration, use
 @code{export STARPU_CALIBRATE=1} . This may be necessary if your application
 has not-so-stable performance. StarPU will force calibration (and thus ignore

doc/chapters/using.texi

@@ -60,7 +60,8 @@ AFTER: First element is 3.140000
 
 When StarPU is used for the first time, the directory
 @code{$STARPU_HOME/.starpu/} is created, performance models will be stored in
-that directory (@code{STARPU_HOME} defaults to @code{$HOME})
+that directory (@code{STARPU_HOME} defaults to @code{$HOME}, or @code{$USERPROFILE
+} in windows environments).
 
 Please note that buses are benchmarked when StarPU is launched for the
 first time. This may take a few minutes, or less if @code{hwloc} is