<feed xmlns='http://www.w3.org/2005/Atom'>
<title>llvm-project.git/llvm/lib/Target/AMDGPU/AMDGPUAnnotateUniformValues.cpp, branch users/nico/python-2</title>
<subtitle>Unnamed repository; edit this file 'description' to name the repository.
</subtitle>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/'/>
<entry>
<title>[AMDGPU] Preserve all analyses if nothing changed (#117994)</title>
<updated>2024-11-28T14:33:05+00:00</updated>
<author>
<name>Jay Foad</name>
<email>jay.foad@amd.com</email>
</author>
<published>2024-11-28T14:33:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=3923e0451a4b02c4fe7c394e0f0ec90708ebca7a'/>
<id>3923e0451a4b02c4fe7c394e0f0ec90708ebca7a</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Move AMDGPUMemoryUtils out of Utils. NFC. (#104930)</title>
<updated>2024-08-20T15:15:46+00:00</updated>
<author>
<name>Jay Foad</name>
<email>jay.foad@amd.com</email>
</author>
<published>2024-08-20T15:15:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=55d744eea36194809c4d16437db08aaa91c74292'/>
<id>55d744eea36194809c4d16437db08aaa91c74292</id>
<content type='text'>
It is only used by CodeGen so does not need to be shared with the
assembler/disassembler.</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It is only used by CodeGen so does not need to be shared with the
assembler/disassembler.</pre>
</div>
</content>
</entry>
<entry>
<title>AMDGPU/NewPM: Initialize class member</title>
<updated>2024-08-10T04:54:55+00:00</updated>
<author>
<name>Vitaly Buka</name>
<email>vitalybuka@google.com</email>
</author>
<published>2024-08-10T04:54:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=109f2f04bcf8f0010380d235a03799520dd4c37c'/>
<id>109f2f04bcf8f0010380d235a03799520dd4c37c</id>
<content type='text'>
After #102654
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
After #102654
</pre>
</div>
</content>
</entry>
<entry>
<title>AMDGPU/NewPM: Port AMDGPUAnnotateUniformValues to new pass manager (#102654)</title>
<updated>2024-08-10T03:06:08+00:00</updated>
<author>
<name>Matt Arsenault</name>
<email>Matthew.Arsenault@amd.com</email>
</author>
<published>2024-08-10T03:06:08+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=77e68fbdd3fdc205bafa042fcb87014011dc7799'/>
<id>77e68fbdd3fdc205bafa042fcb87014011dc7799</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Replace LegacyDA with Uniformity Analysis in AnnotateUniformValues</title>
<updated>2023-02-28T20:05:38+00:00</updated>
<author>
<name>Anshil Gandhi</name>
<email>gandhi21299@gmail.com</email>
</author>
<published>2023-02-28T20:03:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=a78301560d8d921d1980de8a1e1820f678d109f1'/>
<id>a78301560d8d921d1980de8a1e1820f678d109f1</id>
<content type='text'>
Reviewed By: sameerds

Differential Revision: https://reviews.llvm.org/D144162
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Reviewed By: sameerds

Differential Revision: https://reviews.llvm.org/D144162
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Return better Changed status from AMDGPUAnnotateUniformValues</title>
<updated>2022-02-17T09:31:42+00:00</updated>
<author>
<name>Jay Foad</name>
<email>jay.foad@amd.com</email>
</author>
<published>2022-02-16T10:59:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=77e793d0255b67454176257d8306c932704ebd94'/>
<id>77e793d0255b67454176257d8306c932704ebd94</id>
<content type='text'>
Differential Revision: https://reviews.llvm.org/D119943
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Differential Revision: https://reviews.llvm.org/D119943
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Improve clobbering checks in the kernel argument promotion</title>
<updated>2022-02-10T22:51:47+00:00</updated>
<author>
<name>Stanislav Mekhanoshin</name>
<email>Stanislav.Mekhanoshin@amd.com</email>
</author>
<published>2022-02-10T21:02:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=290e5722e83e9c7480d64c049a14b74e30b6af4a'/>
<id>290e5722e83e9c7480d64c049a14b74e30b6af4a</id>
<content type='text'>
Use same MSSA clobbering checks as in the AMDGPUAnnotateUniformValues.
Kernel argument promotion needs exactly the same information so factor
out utility function isClobberedInFunction.

Differential Revision: https://reviews.llvm.org/D119480
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Use same MSSA clobbering checks as in the AMDGPUAnnotateUniformValues.
Kernel argument promotion needs exactly the same information so factor
out utility function isClobberedInFunction.

Differential Revision: https://reviews.llvm.org/D119480
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Simplify AMDGPUAnnotateUniformValues::visitLoadInst</title>
<updated>2022-02-03T16:27:48+00:00</updated>
<author>
<name>Jay Foad</name>
<email>jay.foad@amd.com</email>
</author>
<published>2022-02-03T15:27:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=b9cf52bc3d29fc5a28015ee642d2b63354391c41'/>
<id>b9cf52bc3d29fc5a28015ee642d2b63354391c41</id>
<content type='text'>
Always set uniform metadata on the pointer if it is an instruction, but
otherwise do not bother to create a trivial getelementptr instruction,
because AMDGPUInstrInfo::isUniformMMO can already detect that various
non-instruction pointers are uniform.

Most of the test case churn is from tests that used undef as a pointer,
which AMDGPUInstrInfo::isUniformMMO treats as uniform.

Differential Revision: https://reviews.llvm.org/D118909
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Always set uniform metadata on the pointer if it is an instruction, but
otherwise do not bother to create a trivial getelementptr instruction,
because AMDGPUInstrInfo::isUniformMMO can already detect that various
non-instruction pointers are uniform.

Most of the test case churn is from tests that used undef as a pointer,
which AMDGPUInstrInfo::isUniformMMO treats as uniform.

Differential Revision: https://reviews.llvm.org/D118909
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Use new target MMO flag MONoClobber</title>
<updated>2022-02-02T17:12:36+00:00</updated>
<author>
<name>Jay Foad</name>
<email>jay.foad@amd.com</email>
</author>
<published>2022-02-02T11:35:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=ddd3807e6952dedfd842b3b10076a614f75c1979'/>
<id>ddd3807e6952dedfd842b3b10076a614f75c1979</id>
<content type='text'>
This allows us to set the noclobber flag on (the MMO of) a load
instruction instead of on the pointer. This fixes a bug where noclobber
was being applied to all loads from the same pointer, even if some of
them were clobbered.

Differential Revision: https://reviews.llvm.org/D118775
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This allows us to set the noclobber flag on (the MMO of) a load
instruction instead of on the pointer. This fixes a bug where noclobber
was being applied to all loads from the same pointer, even if some of
them were clobbered.

Differential Revision: https://reviews.llvm.org/D118775
</pre>
</div>
</content>
</entry>
<entry>
<title>[AMDGPU] Check atomics aliasing in the clobbering annotation</title>
<updated>2022-02-01T20:33:25+00:00</updated>
<author>
<name>Stanislav Mekhanoshin</name>
<email>Stanislav.Mekhanoshin@amd.com</email>
</author>
<published>2022-01-31T23:10:08+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=79606ee85c5384819e34f20c7ebd28d3f8ba306b'/>
<id>79606ee85c5384819e34f20c7ebd28d3f8ba306b</id>
<content type='text'>
MemorySSA considers any atomic a def to any operation it dominates
just like a barrier or fence. That is correct from memory state
perspective, but not required for the no-clobber metadata since
we are not using it for reordering. Skip such atomics during the
scan just like a barrier if it does not alias with the load.

Differential Revision: https://reviews.llvm.org/D118661
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
MemorySSA considers any atomic a def to any operation it dominates
just like a barrier or fence. That is correct from memory state
perspective, but not required for the no-clobber metadata since
we are not using it for reordering. Skip such atomics during the
scan just like a barrier if it does not alias with the load.

Differential Revision: https://reviews.llvm.org/D118661
</pre>
</div>
</content>
</entry>
</feed>
