<feed xmlns='http://www.w3.org/2005/Atom'>
<title>delta/llvm.git/openmp/runtime/src, branch EmptyLineAfterFunctionDefinition</title>
<subtitle>github.com: llvm/llvm-project.git
</subtitle>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/'/>
<entry>
<title>[OpenMP][libomp] Enable HWLOC topology detection of multiple CPU kinds</title>
<updated>2021-11-17T22:30:18+00:00</updated>
<author>
<name>Peyton, Jonathan L</name>
<email>jonathan.l.peyton@intel.com</email>
</author>
<published>2021-10-21T19:07:15+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=a733b18bdbe28778cf34705447ee7691b2064b6e'/>
<id>a733b18bdbe28778cf34705447ee7691b2064b6e</id>
<content type='text'>
Teach the HWLOC topology method how to detect Atom and Core
types so hybrid CPUs are properly detected and represented when using
the HWLOC topology method.

Differential Revision: https://reviews.llvm.org/D112270
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Teach the HWLOC topology method how to detect Atom and Core
types so hybrid CPUs are properly detected and represented when using
the HWLOC topology method.

Differential Revision: https://reviews.llvm.org/D112270
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP][libomp] Improve Windows Processor Group handling within topology</title>
<updated>2021-11-17T22:29:01+00:00</updated>
<author>
<name>Peyton, Jonathan L</name>
<email>jonathan.l.peyton@intel.com</email>
</author>
<published>2021-10-21T21:28:00+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=286094af9bc0cfc89d00d8d645e8eaca3d78cd91'/>
<id>286094af9bc0cfc89d00d8d645e8eaca3d78cd91</id>
<content type='text'>
The current implementation of Windows Processor Groups has
a separate topology method to handle them. This patch deprecates
that specific method and uses the regular CPUID topology
method by default and inserts the Windows Processor Group objects
in the topology manually.

Notes:
* The preference for processor groups is lowered to a value less than
  socket so that the user will see sockets in the KMP_AFFINITY=verbose
  output instead of processor groups when sockets=processor groups.
* The topology's capacity is modified to handle additional topology layers
  without the need for reallocation.
* If a user asks for a granularity setting that is "above" the processor
  group layer, then the granularity is adjusted "down" to the processor
  group since this is the coarsest layer available for threads.

Differential Revision: https://reviews.llvm.org/D112273
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The current implementation of Windows Processor Groups has
a separate topology method to handle them. This patch deprecates
that specific method and uses the regular CPUID topology
method by default and inserts the Windows Processor Group objects
in the topology manually.

Notes:
* The preference for processor groups is lowered to a value less than
  socket so that the user will see sockets in the KMP_AFFINITY=verbose
  output instead of processor groups when sockets=processor groups.
* The topology's capacity is modified to handle additional topology layers
  without the need for reallocation.
* If a user asks for a granularity setting that is "above" the processor
  group layer, then the granularity is adjusted "down" to the processor
  group since this is the coarsest layer available for threads.

Differential Revision: https://reviews.llvm.org/D112273
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP][libomp] Add support for offline CPUs in Linux</title>
<updated>2021-11-17T22:28:01+00:00</updated>
<author>
<name>Peyton, Jonathan L</name>
<email>jonathan.l.peyton@intel.com</email>
</author>
<published>2021-10-21T21:43:54+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=1dd797168e9f8cb592c254c642f196dab3447092'/>
<id>1dd797168e9f8cb592c254c642f196dab3447092</id>
<content type='text'>
If some CPUs are offline, then make sure they are not included in the
fullMask even if norespect is given to KMP_AFFINITY.

Differential Revision: https://reviews.llvm.org/D112274
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If some CPUs are offline, then make sure they are not included in the
fullMask even if norespect is given to KMP_AFFINITY.

Differential Revision: https://reviews.llvm.org/D112274
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP][libomp] Allow users to specify KMP_HW_SUBSET in any order</title>
<updated>2021-11-17T21:27:37+00:00</updated>
<author>
<name>Peyton, Jonathan L</name>
<email>jonathan.l.peyton@intel.com</email>
</author>
<published>2021-10-18T18:58:38+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=a0afb9d0fc2b6b1a0e89c9cd8f0bd1d839e740ce'/>
<id>a0afb9d0fc2b6b1a0e89c9cd8f0bd1d839e740ce</id>
<content type='text'>
Remove restriction forcing users to specify the KMP_HW_SUBSET value in
topology order. This patch sorts the user KMP_HW_SUBSET value before
trying to apply it. For example: 1s,4c,2t is equivalent to 2t,1s,4c

Differential Revision: https://reviews.llvm.org/D112027
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Remove restriction forcing users to specify the KMP_HW_SUBSET value in
topology order. This patch sorts the user KMP_HW_SUBSET value before
trying to apply it. For example: 1s,4c,2t is equivalent to 2t,1s,4c

Differential Revision: https://reviews.llvm.org/D112027
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP][libomp][NFC] Remove non-ASCII apostrophe in comment</title>
<updated>2021-11-17T20:46:40+00:00</updated>
<author>
<name>Jonathan Peyton</name>
<email>jonathan.l.peyton@intel.com</email>
</author>
<published>2021-11-17T20:46:23+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=c46becf500df2a7fb4b4fce16178a036c344315a'/>
<id>c46becf500df2a7fb4b4fce16178a036c344315a</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP] Silence build warnings when built with MinGW</title>
<updated>2021-11-17T16:51:18+00:00</updated>
<author>
<name>Martin Storsjö</name>
<email>martin@martin.st</email>
</author>
<published>2021-11-17T08:49:44+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=9b2b54983765abd365f6e43428c6a2697c8e4c03'/>
<id>9b2b54983765abd365f6e43428c6a2697c8e4c03</id>
<content type='text'>
There's an attempt to upstream this change in
https://github.com/intel/ittapi/pull/25 too.

Differential Revision: https://reviews.llvm.org/D114069
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
There's an attempt to upstream this change in
https://github.com/intel/ittapi/pull/25 too.

Differential Revision: https://reviews.llvm.org/D114069
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP] Set default blocktime to 0 for hybrid cpu</title>
<updated>2021-11-12T18:05:35+00:00</updated>
<author>
<name>Nawrin Sultana</name>
<email>nawrin.sultana@intel.com</email>
</author>
<published>2021-10-25T19:01:56+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=7a5680233e03c8ac65d4a9bd97c59b3a3b683526'/>
<id>7a5680233e03c8ac65d4a9bd97c59b3a3b683526</id>
<content type='text'>
Differential Revision:https://reviews.llvm.org/D113012
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Differential Revision:https://reviews.llvm.org/D113012
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP] Allow building libomp using Microsoft Visual C++ naming scheme</title>
<updated>2021-11-11T19:11:56+00:00</updated>
<author>
<name>Bran Hagger</name>
<email>brhagger@microsoft.com</email>
</author>
<published>2021-11-11T19:08:25+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=9f15cacc2ecd2eb1dda8ca0b8ae984286fb4135f'/>
<id>9f15cacc2ecd2eb1dda8ca0b8ae984286fb4135f</id>
<content type='text'>
Differential Revision: https://reviews.llvm.org/D110354
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Differential Revision: https://reviews.llvm.org/D110354
</pre>
</div>
</content>
</entry>
<entry>
<title>Revert "[openmp] Add OMPT initialization in libomptarget"</title>
<updated>2021-11-10T11:44:25+00:00</updated>
<author>
<name>Joachim Protze</name>
<email>protze@itc.rwth-aachen.de</email>
</author>
<published>2021-11-10T11:44:25+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=52da6f562e7d45f03ca4c5e291987f4ebdacccf1'/>
<id>52da6f562e7d45f03ca4c5e291987f4ebdacccf1</id>
<content type='text'>
Reverting initial OMPT for target implementation in favor of a
different implementation.

This reverts commit 3bc8ce5dd718beef0031bf4b070ac4026e6910d7.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Reverting initial OMPT for target implementation in favor of a
different implementation.

This reverts commit 3bc8ce5dd718beef0031bf4b070ac4026e6910d7.
</pre>
</div>
</content>
</entry>
<entry>
<title>[OpenMP] libomp: Fix handling of barrier pattern environment variables</title>
<updated>2021-11-08T12:01:26+00:00</updated>
<author>
<name>@t-msn</name>
<email>misono.tomohiro@fujitsu.com</email>
</author>
<published>2021-11-08T12:01:19+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/llvm.git/commit/?id=0808d956c43e10646bc3f79ef5f9ccf12fc829f3'/>
<id>0808d956c43e10646bc3f79ef5f9ccf12fc829f3</id>
<content type='text'>
It is better to set all barrier patterns to use "dist" when at least
one environment variable specifies "dist". Otherwise if only one
environment is set to "dist" and others left blank inadvertently,
it would result in mixing dist barrier with default hyper barrier
pattern.

Differential Revision: https://reviews.llvm.org/D112597
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It is better to set all barrier patterns to use "dist" when at least
one environment variable specifies "dist". Otherwise if only one
environment is set to "dist" and others left blank inadvertently,
it would result in mixing dist barrier with default hyper barrier
pattern.

Differential Revision: https://reviews.llvm.org/D112597
</pre>
</div>
</content>
</entry>
</feed>
