Merge pull request #500 from dbabokin/docs
Bringing docs/ispc.rst in sync with ispc.html at web site
This commit is contained in:
45
docs/ispc.rst
Normal file → Executable file
45
docs/ispc.rst
Normal file → Executable file
@@ -59,7 +59,7 @@ Contents:
|
||||
+ `Basic Command-line Options`_
|
||||
+ `Selecting The Compilation Target`_
|
||||
+ `Generating Generic C++ Output`_
|
||||
+ `Compiling For The Intel® Many Integrated Core Architecture`_
|
||||
+ `Compiling For The Intel® Xeon Phi™ Architecture`_
|
||||
+ `Selecting 32 or 64 Bit Addressing`_
|
||||
+ `The Preprocessor`_
|
||||
+ `Debugging`_
|
||||
@@ -560,16 +560,15 @@ C++ file; this can be used to easily include specific implementations of
|
||||
the vector types and functions.
|
||||
|
||||
|
||||
Compiling For The Intel® Many Integrated Core Architecture
|
||||
Compiling For The Intel® Xeon Phi™ Architecture
|
||||
----------------------------------------------------------
|
||||
|
||||
``ispc`` has beta-level support for compiling for the many-core Intel®
|
||||
Xeon Phi™ processor, part of the Intel® Many Integrated Core
|
||||
Architecture (MIC). For the first generation of processor, code named
|
||||
Knights Corner, this support is based on the "generic" C++ output,
|
||||
described in the previous section.
|
||||
Xeon Phi™ architecture (formerly, "Many Integrated Cores" / MIC).
|
||||
This support is based on the "generic" C++ output, described in the previous
|
||||
section.
|
||||
|
||||
To compile for Knight's Corner, first generate intermediate C++ code:
|
||||
To compile for Xeon Phi™, first generate intermediate C++ code:
|
||||
|
||||
::
|
||||
|
||||
@@ -578,8 +577,8 @@ To compile for Knight's Corner, first generate intermediate C++ code:
|
||||
|
||||
The ``ispc`` distribution now includes a header file,
|
||||
``examples/intrinsics/knc.h``, which maps from the generic C++ output
|
||||
to the corresponding intrinsic operations supported by Knights Corner. Thus,
|
||||
to generate an object file, use the Intel C Compiler (``icc``) compile
|
||||
to the corresponding intrinsic operations supported by Intel Xeon Phi™.
|
||||
Thus, to generate an object file, use the Intel C Compiler (``icc``) compile
|
||||
the C++ code generated by ``ispc``, setting the ``#include`` search
|
||||
path so that it can find the ``examples/intrinsics/knc.h`` header file
|
||||
in the ``ispc`` distribution.
|
||||
@@ -589,7 +588,7 @@ in the ``ispc`` distribution.
|
||||
icc -mmic -Iexamples/intrinsics/ foo.cpp -o foo.o
|
||||
|
||||
With the current beta implementation, complex ``ispc`` programs are able to
|
||||
run on Knights Corner, though there are a number of known limitations:
|
||||
run on Xeon Phi™, though there are a number of known limitations:
|
||||
|
||||
* The ``examples/intrinsics/knc.h`` header file isn't complete yet; for
|
||||
example, vector operations with ``int8`` and ``int16`` types aren't yet
|
||||
@@ -599,8 +598,8 @@ run on Knights Corner, though there are a number of known limitations:
|
||||
|
||||
* If you use the ``launch`` functionality to launch tasks across cores,
|
||||
note that the pthreads task system implemented in
|
||||
``examples/tasksys.cpp`` offers several implemenetations for Knights
|
||||
Corner. You will need to experiment to understand which one is most
|
||||
``examples/tasksys.cpp`` offers several implemenetations for Xeon Phi™.
|
||||
You will need to experiment to understand which one is most
|
||||
appropriate for your workload.
|
||||
|
||||
* The compiler currently emits unaligned memory accesses in many cases
|
||||
@@ -611,26 +610,18 @@ run on Knights Corner, though there are a number of known limitations:
|
||||
precision compliance (using ICC option ``-fp-model strict``) or
|
||||
accurate reporting of floating point exceptions (using ICC option
|
||||
``-fp-model except``) the compiler will generate code that uses the
|
||||
x87 unit rather than KNC's vector unit. For similar reasons, the
|
||||
x87 unit rather than Xeon Phi™'s vector unit. For similar reasons, the
|
||||
options ``–ansi`` and ``–fmath-errno`` may result in calls to math
|
||||
functions that are implemented in x87 rather than using Initial Many
|
||||
Core Instructions (Intel® IMCI) instructions. This will have a
|
||||
significant performance impact. See the ICC manual for details on
|
||||
these compiler options.
|
||||
functions that are implemented in x87 rather than using vector instructions.
|
||||
This will have a significant performance impact. See the ICC manual for
|
||||
details on these compiler options.
|
||||
|
||||
All of these issues are currently actively being addressed and will be
|
||||
fixed in future releases.
|
||||
|
||||
If you do use the current version of ``ispc`` on Knights Corner,
|
||||
If you do use the current version of ``ispc`` on Intel Xeon Phi™,
|
||||
please let us know of any bugs or unexpected results. (Also, any
|
||||
interesting results!). *Note that access to Intel® Xeon Phi™
|
||||
processor and public discussion of Intel® Xeon Phi™ processor
|
||||
performance is still governed by NDA*, so please send email to "matt
|
||||
dot pharr at intel dot com" for any issues that shouldn't be filed in
|
||||
the `public ispc bug tracker`_.
|
||||
|
||||
.. _public ispc bug tracker: https://github.com/ispc/ispc/issues
|
||||
|
||||
interesting results!).
|
||||
|
||||
Selecting 32 or 64 Bit Addressing
|
||||
---------------------------------
|
||||
@@ -4694,7 +4685,7 @@ countries.
|
||||
|
||||
* Other names and brands may be claimed as the property of others.
|
||||
|
||||
Copyright(C) 2011, Intel Corporation. All rights reserved.
|
||||
Copyright(C) 2011-2013, Intel Corporation. All rights reserved.
|
||||
|
||||
|
||||
Optimization Notice
|
||||
|
||||
Reference in New Issue
Block a user