en Copyright 2015 Java http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#fa2421392802025e842454a31ed50dbfab643130 [MemCpyOpt] Forward `memcpy` based on the actual copy memory location. (#87190)Fixes #85560. We can forward `memcpy` as long as the actual memory location being copied have not been altered. alive2: https://alive2.llvm.org/ce/z/q9JaHV List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Fri, 12 Jul 2024 14:58:28 +0000 DianQK <dianqk@dianqk.net> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#a11faeed446882a81e79d780125d93e7199df645 [test] Switch to use -passes syntax in various test cases List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Thu, 01 Dec 2022 20:00:24 +0000 Bjorn Pettersson <bjorn.a.pettersson@ericsson.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#4d33cf4166f9a4438cb36ab60502eef7affd7df2 [MemCpyOpt] Avoid moving lifetime marker above def (PR58903)This is unlikely to happen with opaque pointers, so just bail outof the transform, rather than trying to move bitcasts/etc as well.Fixes https://github.com/llvm/llvm-project/issues/58903. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Fri, 11 Nov 2022 14:05:11 +0000 Nikita Popov <npopov@redhat.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#9a45e4beede24fdcd6b7d95416d0447ba154651c [MemCpyOpt] Move lifetime marker before call to enable call slot optimizationCurrently call slot optimization may be prevented because thelifetime markers for the destination only start after the call.In this case, rather than aborting the transform, we should movethe lifetime.start before the call to enable the transform.Differential Revision: https://reviews.llvm.org/D135886 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Thu, 13 Oct 2022 14:45:37 +0000 Nikita Popov <npopov@redhat.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#f386f7690d3aa49994f4645c6e30463c1e221bc6 [MemCpyOpt] Add additional tests with lifetime intrinsics (NFC) List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Thu, 13 Oct 2022 15:29:37 +0000 Nikita Popov <npopov@redhat.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#19aa1aab2edd6e2d1d41f0f99a823bbbddca549a [MemCpyOpt] Don't run full pipeline in test (NFC)Just memcpyopt is enough for this test. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Thu, 13 Oct 2022 14:47:04 +0000 Nikita Popov <npopov@redhat.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#42d98d80fb86b0eb7058a7288e7a4c900f04c556 [MemCpyOpt] Convert tests to opaque pointers (NFC)Converted using the script athttps://gist.github.com/nikic/98357b71fd67756b0f064c9517b62a34. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Wed, 05 Oct 2022 12:26:57 +0000 Nikita Popov <npopov@redhat.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#88003cea1c6eb3ebcb18b4f1b9d0e2052dd76fe4 [MemCpyOpt] Remove MemDepAnalysis-based implementationThe MemorySSA-based implementation has been enabled for a few months(since D94376). This patch drops the old MDA-based implementationentirely.I've kept this to only the basic cleanup of dropping variousconditions -- the code could be further cleaned up now that thereis only one implementation.Differential Revision: https://reviews.llvm.org/D102113 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Sat, 08 May 2021 18:39:54 +0000 Nikita Popov <nikita.ppv@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#cd54c5791988ca5e80f0c08c4a27baef8dce205c Reland "[Libcalls, Attrs] Annotate libcalls with noundef"Fixed Clang tests. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Sat, 20 Feb 2021 03:19:03 +0000 Dávid Bolvanský <david.bolvansky@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#94d034fb86584ff2e95dd5c62dec46c8b4c383a9 Revert "[Libcalls, Attrs] Annotate libcalls with noundef"This reverts commit 33b0c63775ce58014c55e285671e3315104a6076. Bots are failing. Some Clang tests need to be updated too. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Sat, 20 Feb 2021 03:17:44 +0000 Dávid Bolvanský <david.bolvansky@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#33b0c63775ce58014c55e285671e3315104a6076 [Libcalls, Attrs] Annotate libcalls with noundefI think we can use here same logic as for nonnull.strlen(X) - X must be noundef => valid pointer.for libcalls with size arg, we add noundef only if size is known and greater than 0 - so pointers must be noundef (valid ones)Reviewed By: jdoerfert, aqjuneDifferential Revision: https://reviews.llvm.org/D95122 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Sat, 20 Feb 2021 03:08:50 +0000 Dávid Bolvanský <david.bolvansky@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#be9889b350f47319b795a4a4826b4b0bc190807a [MemorySSA] Don't treat lifetime.end as NoAliasMemorySSA currently treats lifetime.end intrinsics as not aliasinganything. This breaks MemorySSA-based MemCpyOpt, because we'll happilymove a read of a pointer below a lifetime.end intrinsic, as no clobberis reported.I think the MemorySSA modelling here isn't correct: lifetime.end(p)has approximately the same effect as doing a memcpy(p, undef), andshould be treated as a clobber.This patch removes the special handling of lifetime.end, leavingalias analysis to handle it appropriately.Differential Revision: https://reviews.llvm.org/D95763 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Sun, 31 Jan 2021 16:55:24 +0000 Nikita Popov <nikita.ppv@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#6e52eebc2a68f34aa19a2b80f1b763aac692d542 [MemCpyOpt] Add test for incorrect optimization across lifetime (NFC)This only affects the MemorySSA-based implementation. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Fri, 29 Jan 2021 11:56:23 +0000 Nikita Popov <nikita.ppv@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#d2f8961b7b8149848a78321c03ae9c74c45a2d5d [MemCpyOpt] Remove unused check-prefixesJust use default CHECK List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Mon, 09 Nov 2020 11:38:39 +0000 Simon Pilgrim <llvm-dev@redking.me.uk> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#cd6f40f4326c8251ebb628122bf092ca5536a634 [MemCpyOpt] Add test scaffolding for MSSA based MemCpyOptThis adds an -enable-memcpyopt-memoryssa option that currently doesnothing apart from requiring MSSA as a dependency. The tests aresplit to run both with the option disabled and enabled. I went withthis rather than the separate directory DSE uses, as I found itconvenient to have a direct side-by-side comparison of differences.Differential Revision: https://reviews.llvm.org/D89206 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Fri, 02 Oct 2020 18:48:39 +0000 Nikita Popov <nikita.ppv@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#64c54c5459cfae8478ce28710784f36b0d94fb2f [MemCpyOpt] Regnerate test checks (NFC) List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Thu, 01 Oct 2020 20:42:14 +0000 Nikita Popov <nikita.ppv@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#fd39b1bb20cec32c310ae9b6f1b4603c17a5f832 Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there.""This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4Original commit message:As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there.This change doesn't include any change to move from selection dag to fast iseland that will come with other numbers that should help inform that decision.There also haven't been any real debuggability studies with this pipeline yet,this is just the initial start done so that people could see it and we could starttweaking after.Test updates: Outside of the newpm tests most of the updates are coming from eitheroptimization passes not run anymore (and without a compelling argument at the moment)that were largely used for canonicalization in clang.Original post:http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.htmlTags: #llvmDifferential Revision: https://reviews.llvm.org/D65410This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d. List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Wed, 27 Nov 2019 04:28:52 +0000 Eric Christopher <echristo@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."This reverts commit 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4.This commit introduced 9 new failures on lldb buildbot host at http://lab.llvm.org:8014/builders/lldb-aarch64-ubuntuFollowing tests were failing: lldb-api :: functionalities/tail_call_frames/ambiguous_tail_call_seq1/TestAmbiguousTailCallSeq1.py lldb-api :: functionalities/tail_call_frames/ambiguous_tail_call_seq2/TestAmbiguousTailCallSeq2.py lldb-api :: functionalities/tail_call_frames/disambiguate_call_site/TestDisambiguateCallSite.py lldb-api :: functionalities/tail_call_frames/disambiguate_paths_to_common_sink/TestDisambiguatePathsToCommonSink.py lldb-api :: functionalities/tail_call_frames/disambiguate_tail_call_seq/TestDisambiguateTailCallSeq.py lldb-api :: functionalities/tail_call_frames/inlining_and_tail_calls/TestInliningAndTailCalls.py lldb-api :: functionalities/tail_call_frames/sbapi_support/TestTailCallFrameSBAPI.py lldb-api :: functionalities/tail_call_frames/thread_step_out_message/TestArtificialFrameStepOutMessage.py lldb-api :: functionalities/tail_call_frames/thread_step_out_or_return/TestSteppingOutWithArtificialFrames.py lldb-api :: functionalities/tail_call_frames/unambiguous_sequence/TestUnambiguousTailCalls.pyTags: #llvmDifferential Revision: https://reviews.llvm.org/D65410 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Tue, 26 Nov 2019 04:32:13 +0000 Muhammad Omair Javaid <omair.javaid@linaro.org> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there.This change doesn't include any change to move from selection dag to fast iseland that will come with other numbers that should help inform that decision.There also haven't been any real debuggability studies with this pipeline yet,this is just the initial start done so that people could see it and we could starttweaking after.Test updates: Outside of the newpm tests most of the updates are coming from eitheroptimization passes not run anymore (and without a compelling argument at the moment)that were largely used for canonicalization in clang.Original post:http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.htmlTags: #llvmDifferential Revision: https://reviews.llvm.org/D65410 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Tue, 26 Nov 2019 00:33:19 +0000 Eric Christopher <echristo@gmail.com> http://src.rcs.uwaterloo.ca:8080/source/history/llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll#cee313d288a4faf0355d76fb6e0e927e211d08a5 Revert "Temporarily Revert "Add basic loop fusion pass.""The reversion apparently deleted the test/Transforms directory.Will be re-reverting again.llvm-svn: 358552 List of files: /llvm-project/llvm/test/Transforms/MemCpyOpt/lifetime.ll Wed, 17 Apr 2019 04:52:47 +0000 Eric Christopher <echristo@gmail.com>