<feed xmlns='http://www.w3.org/2005/Atom'>
<title>llvm-project.git/llvm/test/Transforms/PhaseOrdering/AArch64/loopflatten.ll, branch users/chapuni/cov/single/switch</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>[PhaseOrdering] Update test RUN lines to use `-passes="default&lt;O3&gt;"` to allow evaluation by DOS batch scripts. NFC.</title>
<updated>2024-12-31T15:25:12+00:00</updated>
<author>
<name>Simon Pilgrim</name>
<email>llvm-dev@redking.me.uk</email>
</author>
<published>2024-12-31T15:25:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=d1e5e6735a845f1281f11389da1e5a55a0d2e87a'/>
<id>d1e5e6735a845f1281f11389da1e5a55a0d2e87a</id>
<content type='text'>
`-passes='default&lt;O3&gt;'` isn't correctly parsed on DOS, so when update_test_checks.py runs a system call on the opt RUN line, it fails to evaluate properly - use `-passes="default&lt;O3&gt;"` instead.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
`-passes='default&lt;O3&gt;'` isn't correctly parsed on DOS, so when update_test_checks.py runs a system call on the opt RUN line, it fails to evaluate properly - use `-passes="default&lt;O3&gt;"` instead.
</pre>
</div>
</content>
</entry>
<entry>
<title>[InstCombine] Infer nusw + nneg -&gt; nuw for getelementptr (#111144)</title>
<updated>2024-12-05T13:36:40+00:00</updated>
<author>
<name>Nikita Popov</name>
<email>npopov@redhat.com</email>
</author>
<published>2024-12-05T13:36:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=462cb3cd6cecd0511ecaf0e3ebcaba455ece587d'/>
<id>462cb3cd6cecd0511ecaf0e3ebcaba455ece587d</id>
<content type='text'>
If the gep is nusw (usually via inbounds) and the offset is
non-negative, we can infer nuw.

Proof: https://alive2.llvm.org/ce/z/ihztLy</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If the gep is nusw (usually via inbounds) and the offset is
non-negative, we can infer nuw.

Proof: https://alive2.llvm.org/ce/z/ihztLy</pre>
</div>
</content>
</entry>
<entry>
<title>[IndVars] Preserve flags of narrow IV inc if replacing with wider inc. (#80446)</title>
<updated>2024-02-10T18:11:17+00:00</updated>
<author>
<name>Florian Hahn</name>
<email>flo@fhahn.com</email>
</author>
<published>2024-02-10T18:11:17+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=dce77a357948709e335910ddc07f9c3f2eb2ac4b'/>
<id>dce77a357948709e335910ddc07f9c3f2eb2ac4b</id>
<content type='text'>
We are replacing a narrow IV increment with a wider one. If the original
(narrow) increment did not wrap, the wider one should not wrap either.
Set the flags to be the union of both wide increment and original
increment; this ensures we preserve flags SCEV could infer for the wider
increment.

Fixes https://github.com/llvm/llvm-project/issues/71517.</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
We are replacing a narrow IV increment with a wider one. If the original
(narrow) increment did not wrap, the wider one should not wrap either.
Set the flags to be the union of both wide increment and original
increment; this ensures we preserve flags SCEV could infer for the wider
increment.

Fixes https://github.com/llvm/llvm-project/issues/71517.</pre>
</div>
</content>
</entry>
<entry>
<title>Revert "[PM] Execute IndVarSimplifyPass precede RessociatePass" (#71617)</title>
<updated>2023-11-08T01:22:55+00:00</updated>
<author>
<name>dewen</name>
<email>923406109@qq.com</email>
</author>
<published>2023-11-08T01:22:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=3b82336188354b70acaf4bca1da246f36a38c78f'/>
<id>3b82336188354b70acaf4bca1da246f36a38c78f</id>
<content type='text'>
Reverts llvm/llvm-project#71054</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Reverts llvm/llvm-project#71054</pre>
</div>
</content>
</entry>
<entry>
<title>[PM] Execute IndVarSimplifyPass precede RessociatePass (#71054)</title>
<updated>2023-11-08T01:21:17+00:00</updated>
<author>
<name>dewen</name>
<email>923406109@qq.com</email>
</author>
<published>2023-11-08T01:21:17+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=e4d27d7f320e06333195aded11b080c52c1a3bfe'/>
<id>e4d27d7f320e06333195aded11b080c52c1a3bfe</id>
<content type='text'>
ReassociatePass may clear nsw/nuw flags of some instructions, which may
have side effects on optimizations in IndVarSimplifyPass.</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
ReassociatePass may clear nsw/nuw flags of some instructions, which may
have side effects on optimizations in IndVarSimplifyPass.</pre>
</div>
</content>
</entry>
<entry>
<title>[CVP] Infer nneg on zext when forming from non-negative sext. (#70715)</title>
<updated>2023-10-30T20:48:27+00:00</updated>
<author>
<name>Craig Topper</name>
<email>craig.topper@sifive.com</email>
</author>
<published>2023-10-30T20:48:27+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=55c9f24344a49cd1deb86af1d79d4dc3a798c6fb'/>
<id>55c9f24344a49cd1deb86af1d79d4dc3a798c6fb</id>
<content type='text'>
Builds on #67982 which recently introduced the nneg flag on a zext
instruction.</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Builds on #67982 which recently introduced the nneg flag on a zext
instruction.</pre>
</div>
</content>
</entry>
<entry>
<title>PhaseOrdering: Convert tests to opaque pointers</title>
<updated>2022-11-28T02:26:41+00:00</updated>
<author>
<name>Matt Arsenault</name>
<email>Matthew.Arsenault@amd.com</email>
</author>
<published>2022-11-26T21:56:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=1c55cc600e99a963233d6f467373c8f16a1b8826'/>
<id>1c55cc600e99a963233d6f467373c8f16a1b8826</id>
<content type='text'>
Required manually running update_test_checks:
  AArch64/hoisting-sinking-required-for-vectorization.ll
  AArch64/peel-multiple-unreachable-exits-for-vectorization.ll
  ARM/arm_mult_q15.ll
  X86/hoist-load-of-baseptr.ll
  X86/spurious-peeling.ll
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Required manually running update_test_checks:
  AArch64/hoisting-sinking-required-for-vectorization.ll
  AArch64/peel-multiple-unreachable-exits-for-vectorization.ll
  ARM/arm_mult_q15.ll
  X86/hoist-load-of-baseptr.ll
  X86/spurious-peeling.ll
</pre>
</div>
</content>
</entry>
<entry>
<title>Revert "[NFCI] Regenerate PhaseOrdering test checks"</title>
<updated>2022-04-04T10:30:57+00:00</updated>
<author>
<name>Muhammad Omair Javaid</name>
<email>omair.javaid@linaro.org</email>
</author>
<published>2022-04-04T10:30:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=a96638e50ef5932d53c70d052bef73a5f1f9cba9'/>
<id>a96638e50ef5932d53c70d052bef73a5f1f9cba9</id>
<content type='text'>
This reverts commit e91fe08999d5f5d7e7777837c529bac692d06c1b.

Breaks following buildbots: https://lab.llvm.org/buildbot/#/builders/171
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This reverts commit e91fe08999d5f5d7e7777837c529bac692d06c1b.

Breaks following buildbots: https://lab.llvm.org/buildbot/#/builders/171
</pre>
</div>
</content>
</entry>
<entry>
<title>[NFCI] Regenerate PhaseOrdering test checks</title>
<updated>2022-04-03T22:28:57+00:00</updated>
<author>
<name>Dávid Bolvanský</name>
<email>david.bolvansky@gmail.com</email>
</author>
<published>2022-04-03T22:28:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=e91fe08999d5f5d7e7777837c529bac692d06c1b'/>
<id>e91fe08999d5f5d7e7777837c529bac692d06c1b</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>[LoopFlatten] Move it from LPM2 to LPM1</title>
<updated>2022-01-19T14:38:05+00:00</updated>
<author>
<name>Sjoerd Meijer</name>
<email>sjoerd.meijer@arm.com</email>
</author>
<published>2022-01-19T14:09:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.belthelziquor.com/llvm-project.git/commit/?id=f269ec230e2b65df165009b85f1e8cae2aeb1bba'/>
<id>f269ec230e2b65df165009b85f1e8cae2aeb1bba</id>
<content type='text'>
In D110057 we moved LoopFlatten to a LoopPassManager. This caused a performance
regression for our 64-bit targets (the 32-bit were unaffected), the pass is no
longer triggering for a motivating example. The reason is that the IR is just
very different than expected; we try to match loop statements and particular
uses of induction variables. The easiest is to just move LoopFlatten to a place
in the pipeline where the IR is as expected, which is just before
IndVarSimplify. This means we move it from LPM2 to LPM1, so that it actually
runs just a bit earlier from where it was running before. IndVarSimplify is
responsible for significant rewrites that are difficult to "look through" in
LoopFlatten.

Differential Revision: https://reviews.llvm.org/D116612
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In D110057 we moved LoopFlatten to a LoopPassManager. This caused a performance
regression for our 64-bit targets (the 32-bit were unaffected), the pass is no
longer triggering for a motivating example. The reason is that the IR is just
very different than expected; we try to match loop statements and particular
uses of induction variables. The easiest is to just move LoopFlatten to a place
in the pipeline where the IR is as expected, which is just before
IndVarSimplify. This means we move it from LPM2 to LPM1, so that it actually
runs just a bit earlier from where it was running before. IndVarSimplify is
responsible for significant rewrites that are difficult to "look through" in
LoopFlatten.

Differential Revision: https://reviews.llvm.org/D116612
</pre>
</div>
</content>
</entry>
</feed>
