<table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>Issue</th>
<td>
<a href=https://github.com/llvm/llvm-project/issues/79800>79800</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>
[LoopVectorize] Refine runtime memory check costs when there is an outer loop
</td>
</tr>
<tr>
<th>Labels</th>
<td>
performance,
release:backport,
vectorization
</td>
</tr>
<tr>
<th>Assignees</th>
<td>
</td>
</tr>
<tr>
<th>Reporter</th>
<td>
david-arm
</td>
</tr>
</table>
<pre>
We would like to backport https://github.com/llvm/llvm-project/pull/76034 to fix a major performance regression when using PGO to build certain applications/benchmarks. Specifically, it fixes a 25% regression in the SPEC2017 x264 benchmark.
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJyUksFunDAQQL_GXEZBZjCwHDikSemlUqNGas_GDDBZYyPbJNl-fbW0SXPtyZItz5PeGx0jz46oE9UnUd1nek-LD92on3m80WHNBj9eup8EL363I1g-EyQPgzbnzYcES0pbFOWtwF5gP3Na9iE3fhXYW_v8dtxswT-RSQL7bbdWYN_UslTXSRO_goZVP_kAG4XJh1U7QxBoDhQjewcvCznYI7sZHr58O_A72xEMhaTZgd42y0Yn9i4K7AdyZll1OMccHjcyPLHR1l4E3gGnK5AiaMBKYPURww7SQvD48PkOZdHAK9YK3ofl2diVY1u2OqOuaGRdYdFIlS3d0DbSoCr0aSpIniSWWjaFltKcCAdVZdyhRCULbGWrTqjyAgsaq7qWE9at0kooSatmm19d5T7MGce4U9e0Jykzqwey8QiE-EGRQBR4JxADWdKRRHn7luX96ZlM8oF_HXKut9V9FrojybDPUShpOab4D5w42WMZvnq__fj7m0R1D99pYkcQdpd4JVhp9eECZiFzBuNjin86pYUCAUfQDvyeKID1fsv2YLv_3pXDwjXpIeJ3AAAA__9EFOHq">