en Copyright 2015 Java http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#d648aa1b [BOLT][TEST] Use double dash flags in testsReplace a single dash with a double dash for options that have morethan a single letter.llvm-bolt-wrapper.py has special treatment for output options such as"-o" and "-w" causing issues when a single dash is used, e.g. for"-write-dwp". The wrapper can be fixed as well, but using a double dashhas other advantages as well.Reviewed By: rafaulerDifferential Revision: https://reviews.llvm.org/D127538 List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Fri, 10 Jun 2022 21:10:58 +0000 Maksim Panchenko <maks@fb.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#38fb7d56 [BOLT][TEST] Replace cache+ option with ext-tspReplace "cache+" with "ext-tsp" in all BOLT testsTest Plan:```ninja check-boltgrep -rnw . -e "cache+"```no more tests containing "cache+""cache+" and "ext-tsp" are aliasesReviewed By: rafaulerDifferential Revision: https://reviews.llvm.org/D126714 List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Wed, 01 Jun 2022 20:59:28 +0000 Huan Nguyen <nhuhuan@yahoo.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#f0f5d19a [BOLT][test] Fix X86 cross-platform testsUse target-specific flags for building X86 non-runnable tests.Reviewed By: AmirDifferential Revision: https://reviews.llvm.org/D123072 List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Tue, 05 Apr 2022 01:01:10 +0000 Maksim Panchenko <maks@fb.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#4609f60e [BOLT] Avoid pointless loop rotationIt seems the earlier implementation does not follow the descriptionin LoopRotationPass.h: It rotates loops even if they are already laid outcorrectly. The diff adjusts the behaviour.Given that the impact of LoopInversionPass is minor, this change won'tyield significant perf differences. Tested on clang-10: there seems to be a0.1%-0.3% cpu win and a small reduction of branch misses.**Before:**BOLT-INFO: 120 Functions were reordered by LoopInversionPass**After:**BOLT-INFO: 79 Functions were reordered by LoopInversionPassReviewed By: yota9Differential Revision: https://reviews.llvm.org/D121921 List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Thu, 17 Mar 2022 16:51:07 +0000 spupyrev <spupyrev@fb.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#d4fdc981 [BOLT][TEST] Remove dependence on host_cc and host_cxxSummary: Add dependency on clang and clangxx instead.(cherry picked from FBD31128140) List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Wed, 22 Sep 2021 22:53:38 +0000 Amir Ayupov <aaupov@fb.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s#79807d99 [PR] Introduce loop inversion passSummary:This patch introduces LoopInversionPass. Its main purpose is to ensurethat the loop layout is optimal depending on the profile information. Soif profile information shows that the loop is used, the unconditionaljump instruction must be executed only once and vice-versa. Please takea look to the pass header file and test for more details.Also change link_fdata script a bit, to be able to change FDATA prefix,like FileCheck does.Vladislav Khmelevsky,Advanced Software Technology Lab, HuaweiPR facebookincubator/BOLT#153(cherry picked from FBD28391811) List of files: /llvm-project-15.0.7/bolt/test/X86/loop-inversion-pass.s Tue, 11 May 2021 17:59:13 +0000 Vladislav Khmelevsky <Vladislav.Khmelevskyi@huawei.com>