[PATCH] D29862: LSR: an alternative way to resolve complex solution

[Balaram Makam via llvm-commits]

Thanks Chad and Evgeny. I ran full performance tests on Kryo with D30552 and the performance results are still net negative despite recovering most of spec2006/hmmer loss.
Spec2006	-0.39% geomean
Spec2000	+0.04% geomean

-Balaram
-----Original Message-----
From: Chad Rosier via Phabricator [mailto:reviews at reviews.llvm.org] 
Sent: Friday, March 03, 2017 1:21 PM
To: evstupac at gmail.com; hfinkel at anl.gov; qcolombet at apple.com
Cc: bmakam at codeaurora.org; david.green at arm.com; llvm-commits at lists.llvm.org; farhana.a.aleen at intel.com; mzolotukhin at apple.com; wmi at google.com
Subject: [PATCH] D29862: LSR: an alternative way to resolve complex solution

mcrosier added a comment.

In https://reviews.llvm.org/D29862#691782, @evstupac wrote:

> Thanks for the data.
>  Sure, I'll set the option to false today.


Thank you!


Repository:
  rL LLVM

https://reviews.llvm.org/D29862