[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

aeubanks created this revision. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. A followup to D110201 . For example, we'd set OptimizationRemarkMissed's Regex to '.*' when encountering -Rpass. Th

[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

aeubanks added a comment. In D110673#3029253 , @dblaikie wrote: > It'd be good to understand/document (maybe document in the form of a test if > possible) how downstream users are relying on this - perhaps it's not a valid > reliance and we shouldn't ma

[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

aeubanks added a comment. In D110673#3029438 , @JamesNagurne wrote: > I'll take a quick look tomorrow, but the general idea is that on calling > ParseOptimizationRemark on line 1909 with a -cc1 command line containing > -Rpass=inline -Rno-pass, Opts.Op

[PATCH] D110665: [clang] Don't use the AST to display backend diagnostics

aeubanks updated this revision to Diff 375958. aeubanks added a comment. address comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110665/new/ https://reviews.llvm.org/D110665 Files: clang/include/clang/Basic/DiagnosticFrontendKinds.td cl

[PATCH] D110665: [clang] Don't use the AST to display backend diagnostics

aeubanks added inline comments. Comment at: clang/lib/CodeGen/CodeGenAction.cpp:336-337 + for (auto &F : getModule()->functions()) { +if (const Decl *FD = Gen->GetDeclForMangledName(F.getName())) { + auto Loc = FD->getASTContext().getFullLoc(FD->getLocatio

[PATCH] D110665: [clang] Don't use the AST to display backend diagnostics

aeubanks updated this revision to Diff 376024. aeubanks added a comment. add comments (including why it's a vector and not an actual map) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110665/new/ https://reviews.llvm.org/D110665 Files: clang/inc

[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

aeubanks updated this revision to Diff 376034. aeubanks added a comment. forcing the legacy PM plus -round-trip-args repros the issue. added a new PM equivalent RUN line (that does fail without this patch) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.o

[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

aeubanks added a comment. Thanks for investigating! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110673/new/ https://reviews.llvm.org/D110673 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http

[PATCH] D110673: [clang] Don't modify OptRemark if the argument is not relevant

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG76902079e429: [clang] Don't modify OptRemark if the argument is not relevant (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES

[PATCH] D110891: [inliner] Mandatory inlining decisions produce remarks

aeubanks added inline comments. Comment at: llvm/lib/Analysis/InlineAdvisor.cpp:72-89 + void recordUnsuccessfulInliningImpl(const InlineResult &Result) override { +if (IsInliningRecommended) + ORE.emit([&]() { +return OptimizationRemarkMissed(DEBUG_TYPE, "NotInl

[PATCH] D110665: [clang] Don't use the AST to display backend diagnostics

aeubanks added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110665/new/ https://reviews.llvm.org/D110665 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cg

[PATCH] D110891: [inliner] Mandatory inlining decisions produce remarks

aeubanks added inline comments. Comment at: llvm/lib/Analysis/InlineAdvisor.cpp:72-89 + void recordUnsuccessfulInliningImpl(const InlineResult &Result) override { +if (IsInliningRecommended) + ORE.emit([&]() { +return OptimizationRemarkMissed(DEBUG_TYPE, "NotInl

[PATCH] D110665: [clang] Don't use the AST to display backend diagnostics

This revision was automatically updated to reflect the committed changes. Closed by commit rG256828689231: [clang] Don't use the AST to display backend diagnostics (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110665/new/ htt

[PATCH] D111105: [clang] Add option to clear AST memory before running LLVM passes

aeubanks created this revision. Herald added subscribers: dexonsmith, dang. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This is to save memory for Clang compiles. Measuring building PassBuilder.cpp under /usr/bin/time, max rss

[PATCH] D96320: [ThinLTO, Sanitizers] Skip instrumentation by testing backend

aeubanks added a comment. This seems like the wrong fix to me. ThinLTO only runs the simplification pipeline pre-link, then runs it again plus the optimization pipeline. The optimization pipeline is the only place that OptimizerLastEPCallback should run (aside from the -O0 pipeline). Basically,

[PATCH] D96456: [ThinLTO, NewPM] Add Config::OptPassBuilderHook

aeubanks added a comment. This makes sense to me, we need to add the PassBuilder callbacks in both the prelink/normal and postlink steps, and those are separated. Comment at: llvm/include/llvm/LTO/Config.h:54 + /// For adding passes that run by opt. + std::function OptPassBu

[PATCH] D96456: [ThinLTO, NewPM] Register sanitizers with OptimizerLastPassBuilderHook

aeubanks added inline comments. Comment at: clang/lib/CodeGen/BackendUtil.cpp:1070-1071 +// ThinLTOIndexFile is provideds so we must be in ThinLTO PostLink. +// For -O0 ThinLTO PreLink does basic optimization and triggers +// OptimizerLastEPCallbacks. Post

[PATCH] D96803: EntryExitInstrumenter: Move to a module pass and enable at all optimization levels (PR49143)

aeubanks added a comment. why is this now a module pass? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D96803/new/ https://reviews.llvm.org/D96803 ___ cfe-commits mailing list cfe-commits@lists.llvm.org h

[PATCH] D96803: EntryExitInstrumenter: Move to a module pass and enable at all optimization levels (PR49143)

aeubanks added a comment. In D96803#2568179 , @zatrazz wrote: > In D96803#2566322 , @aeubanks wrote: > >> why is this now a module pass? > > Mainly to avoid the default rule from new pass manager to *not* apply any

[PATCH] D96803: EntryExitInstrumenter: Move to a module pass and enable at all optimization levels (PR49143)

aeubanks added a comment. In D96803#2571316 , @zatrazz wrote: > In D96803#2569436 , @aeubanks wrote: > >> In D96803#2568179 , @zatrazz wrote: >> >>> In D96803#2566322

[PATCH] D96803: EntryExitInstrumenter: Enable at all optimization levels (PR49143)

aeubanks added a comment. code looks good, just some test nits Comment at: llvm/test/Transforms/EntryExitInstrumenter/mcount.ll:1 -; RUN: opt -passes="function(ee-instrument),cgscc(inline),function(post-inline-ee-instrument)" -S < %s | FileCheck %s +; RUN: opt --O0 --ee-instr

[PATCH] D96320: [ThinLTO, NewPM] Run OptimizerLastEPCallbacks from buildThinLTOPreLinkDefaultPipeline

aeubanks added a comment. Sorry, just got back from vacation, I'll try to take a look tomorrow Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D96320/new/ https://reviews.llvm.org/D96320 ___ cfe-commits mai

[PATCH] D96320: [ThinLTO, NewPM] Run OptimizerLastEPCallbacks from buildThinLTOPreLinkDefaultPipeline

aeubanks added a comment. I guess this is fine. I'm worried that various OptimizerLastEPCallbacks will only expect to be run once, and adding a special case of not adding those callbacks during thinlto post link will be confusing for other uses of OptimizerLastEPCallbacks, but maybe it won't be

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks created this revision. aeubanks added reviewers: rnk, thakis, amccarth. Herald added subscribers: usaxena95, kadircet. aeubanks requested review of this revision. Herald added subscribers: cfe-commits, ilya-biryukov. Herald added a project: clang. This fixes an issue where the toolchain d

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks added a comment. not exactly sure how to test this... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97437/new/ https://reviews.llvm.org/D97437 ___ cfe-commits mailing list cfe-commits@lists.llvm

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks added a comment. In `Linux.cpp` I only see uses of VFS, no uses of `llvm::sys::fs` so I think it should work there. And using clangd on vscode on Linux, it works with --sysroot. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97437/new/ htt

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks added a comment. Looks like -ivfsoverlay doesn't take effect in the driver. And -working-directory also changes the program's working directory since there's no vfs in the driver so it uses the actual fs. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://revie

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks reopened this revision. aeubanks added a comment. This revision is now accepted and ready to land. Herald added a subscriber: ormris. Using valgrind's massif, I profiled using clang build build PassBuilder.cpp (it's one of the longest LLVM files to compile), and at peak memory usage: wi

[PATCH] D99696: [clang] NRVO: Improvements and handling of more cases.

aeubanks added a comment. Crashes on a stage 2 build on Windows: ../rel/bin/clang-cl /nologo /showIncludes /Foobj/llvm/lib/MC/MCParser/MCParser.MasmParser.obj /c ../../llvm/lib/MC/MCParser/MasmParser.cpp -D_CRT_SECURE_NO_DEPRECATE -D_CRT_SECURE_NO_WARNINGS -D_CRT_NONSTDC_NO_DEPRECATE -D_CRT

[PATCH] D104099: [NewPM] Remove SpeculateAroundPHIs pass from pipeline

aeubanks added a comment. Some backends don't run SimplifyCFG, e.g. X86. I believe the pass was originally created specifically for X86 (the header has some X86 examples) and may or may not extend to other targets (I'm not very familiar with the pass itself). I'm not opposed to landing this an

[PATCH] D104099: [NewPM] Remove SpeculateAroundPHIs pass from pipeline

aeubanks added a comment. Which pass that comes after SpeculateAroundPHIs in the X86 pipeline (either in the optimization or codegen) would undo its effects? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D104099/new/ https://reviews.llvm.org/D10409

[PATCH] D104099: [NewPM] Remove SpeculateAroundPHIs pass from pipeline

aeubanks accepted this revision. aeubanks added a comment. This revision is now accepted and ready to land. Ah, sorry I missed where you mentioned LSR. If this pass is causing regressions in multiple places, even on X86, then I think it does make sense to remove it. I've added some people that m

[PATCH] D104099: [NewPM] Remove SpeculateAroundPHIs pass

aeubanks accepted this revision. aeubanks added a comment. looks good, we can always revive the pass later if somebody wants it for some reason Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D104099/new/ https://reviews.llvm.org/D104099 ___

[PATCH] D86671: [clang-tidy] Add new case type to check variables with Hungarian notation

aeubanks added a comment. you can reopen the revision via "Add Action..." Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D86671/new/ https://reviews.llvm.org/D86671 ___ cfe-commits mailing list cfe-commits

[PATCH] D105877: [Coroutines] Run coroutine passes by default

aeubanks added a comment. it's probably because we're constructing the call graph every time at -O0 we want to skip all of that if there are no coroutine intrinsics in the module. we can't really express that currently. if we'd want to do something like that, we'd need a wrapper around the pass

[PATCH] D108013: [NFC] Rename AttributeList::has/getAttribute() -> has/getAttributeImpl()

aeubanks created this revision. Herald added subscribers: ormris, foad, dexonsmith, lxfind, okura, jdoerfert, kuter, kerbowa, pengfei, Jim, hiraditya, jgravelle-google, sbc100, nhaehnle, jvesely, dylanmckay, dschuff, arsenm, jholewinski. aeubanks added reviewers: rnk, efriedma. aeubanks published

[PATCH] D108003: [Clang] Extend -Wbool-operation to warn about bitwise and of bools with side effects

aeubanks added a comment. I ran this over Chrome and ran into a use case that looks legitimate. It seems like the pattern in LLVM where we want to run a bunch of transformations and get if any of them changed anything. https://source.chromium.org/chromium/chromium/src/+/main:third_party/webrtc/

[PATCH] D108013: [NFC] Rename AttributeList::has/getAttribute() -> has/getAttributeImpl()

aeubanks updated this revision to Diff 366380. aeubanks added a comment. rebase past cleanups Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108013/new/ https://reviews.llvm.org/D108013 Files: llvm/include/llvm/IR/Attributes.h llvm/include/llvm

[PATCH] D108013: [NFC] Rename AttributeList::has/getAttribute() -> has/getAttributeImpl()

aeubanks updated this revision to Diff 366381. aeubanks added a comment. one more place Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108013/new/ https://reviews.llvm.org/D108013 Files: llvm/include/llvm/IR/Attributes.h llvm/include/llvm/IR/Fu

[PATCH] D108003: [Clang] Extend -Wbool-operation to warn about bitwise and of bools with side effects

aeubanks added a comment. In D108003#2944714 , @xbolva00 wrote: > In D108003#2944178 , @aeubanks > wrote: > >> I ran this over Chrome and ran into a use case that looks legitimate. It >> seems like the pattern i

[PATCH] D105007: [NewPM] Make some sanitizer passes parameterized in the PassRegistry

aeubanks accepted this revision. aeubanks added a comment. This revision is now accepted and ready to land. lgtm Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D105007/new/ https://reviews.llvm.org/D105007 ___

[PATCH] D108003: [Clang] Extend -Wbool-operation to warn about bitwise and of bools with side effects

aeubanks added a comment. In D108003#2944178 , @aeubanks wrote: > I ran this over Chrome and ran into a use case that looks legitimate. It > seems like the pattern in LLVM where we want to run a bunch of > transformations and get if any of them changed

[PATCH] D95807: [Coroutines] Add the newly generated SCCs back to the CGSCC work queue after CoroSplit actually happened

aeubanks added a comment. this will run the function simplification pipeline twice on every single function when coroutines are enabled, I don't think that's the intention I thought the intention was to do all the the re-adding of SCCs inside CoroSplit.cpp, including the SCC with the function t

[PATCH] D95807: [Coroutines] Add the newly generated SCCs back to the CGSCC work queue after CoroSplit actually happened

aeubanks accepted this revision. aeubanks added a comment. The CoroSplit/PassBuilder changes lgtm Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D95807/new/ https://reviews.llvm.org/D95807 ___ cfe-commits

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks added a comment. Herald added a reviewer: ctetreau. I rebased this and there are still major memory regressions for tramp3d-v4: https://llvm-compile-time-tracker.com/compare.php?from=f63405f6e3d30f33e715ef5ad09136127535a3fb&to=aec1375e4afe90727f1c30dae1659d11a20d&stat=max-rss Perhaps

[PATCH] D112971: [NFC] Remove LinkAll*.h

aeubanks created this revision. Herald added a subscriber: hiraditya. aeubanks requested review of this revision. Herald added a reviewer: jdoerfert. Herald added a reviewer: jdoerfert. Herald added a reviewer: sstefan1. Herald added subscribers: llvm-commits, cfe-commits, sstefan1. Herald added a

[PATCH] D112971: [NFC] Remove LinkAll*.h

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGfe364e5dc78c: [NFC] Remove LinkAll*.h (authored by aeubanks). Herald added a subscriber: awarzynski. Changed prior to commit: https://reviews.llvm.

[PATCH] D112971: [NFC] Remove LinkAll*.h

aeubanks added a comment. Turns out the comments are misleading, they aren't to prevent specific functions from being removed, they're to force an odr usage of various functions, which in turn makes sure that globals in the corresponding translation units with the function definition are initia

[PATCH] D112732: [ASan] Process functions in Asan module pass

aeubanks added a comment. can we rename "asan-module" to "asan" after this change? and remove the extra "asan-pipeline"/"asan-function-pipeline" parsing callbacks in NewPMDriver.cpp? and previously we had `-passes=require` because a function pass couldn't invoke a module analysis, but that's no

[PATCH] D113210: [NewPM] Use the default AA pipeline by default

aeubanks created this revision. aeubanks added a reviewer: asbirlea. Herald added subscribers: ormris, steven_wu, hiraditya. Herald added a reviewer: ctetreau. Herald added a reviewer: ctetreau. aeubanks requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers

[PATCH] D113210: [NewPM] Use the default AA pipeline by default

aeubanks updated this revision to Diff 384842. aeubanks added a comment. update LTOBackend.cpp Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113210/new/ https://reviews.llvm.org/D113210 Files: clang/lib/CodeGen/BackendUtil.cpp llvm/lib/LTO/LTO

[PATCH] D113210: [NewPM] Use the default AA pipeline by default

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG13317286f829: [NewPM] Use the default AA pipeline by default (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTIO

[PATCH] D113210: [NewPM] Use the default AA pipeline by default

aeubanks added a comment. In D113210#3110780 , @Meinersbur wrote: > This change caused the Polly build to fail: > https://lab.llvm.org/buildbot/#/builders/10/builds/7501 > > opt: > /home/worker/buildbot-workers/polly-x86_64-gce1/rundir/llvm.src/llvm/

[PATCH] D113210: [NewPM] Use the default AA pipeline by default

aeubanks added a comment. fixed with 7f62759697762473 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113210/new/ https://reviews.llvm.org/D113210 ___

[PATCH] D113231: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks updated this revision to Diff 385122. aeubanks added a comment. Herald added subscribers: cfe-commits, wenlei, steven_wu. Herald added a project: clang. update tests Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113231/new/ https://reviews

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks created this revision. Herald added subscribers: ormris, wenlei, steven_wu, hiraditya, eraman. Herald added a reviewer: ctetreau. aeubanks requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. Previously, any change in

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks added inline comments. Comment at: llvm/lib/Transforms/IPO/FunctionAttrs.cpp:1845 + if (auto *Call = dyn_cast(U)) +FAM.invalidate(*Call->getParent()->getParent(), FuncPA); +} nikic wrote: > Do we need to worry about indirect references h

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks updated this revision to Diff 385539. aeubanks added a comment. address comments [argpromo] only invalidate direct callers that call the function, not if the function is a parameter Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113304/new/

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks added a comment. In D113304#3112937 , @nikic wrote: > I don't think I fully understand the interaction this has with eager > invalidation. I would have expected that if we eagerly invalidate, then this > fine-grained invalidation wouldn't make

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks updated this revision to Diff 385642. aeubanks added a comment. rebase Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113304/new/ https://reviews.llvm.org/D113304 Files: clang/test/CodeGen/thinlto-distributed-newpm.ll llvm/lib/Analysis

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

aeubanks added inline comments. Comment at: llvm/lib/Transforms/IPO/Inliner.cpp:1017 +// invalidate analyses for all functions in this SCC later. +FAM.invalidate(F, PreservedAnalyses::none()); } mtrofin wrote: > Should we do this if !Changed? Actually,

[PATCH] D112971: [NFC] Remove LinkAll*.h

aeubanks added a comment. the comments were misleading, I updated them in D113074 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D112971/new/ https://reviews.llvm.org/D112971 __

[PATCH] D112935: [NFC] Rename GVN -> GVNPass and SROA -> SROAPass

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG1d8750c3dad4: [NFC] Rename GVN -> GVNPass and SROA -> SROAPass (authored by aeubanks). Herald added a project: clang. Herald added a subscriber: cfe-

[PATCH] D113738: [LTO] Allow passing -Os/-Oz as the optimization level

aeubanks created this revision. Herald added subscribers: ormris, steven_wu, hiraditya, arichardson, inglorion, sbc100, mgorny, emaste. Herald added a reviewer: MaskRay. Herald added a reviewer: gkm. Herald added a project: lld-macho. Herald added a reviewer: lld-macho. aeubanks requested review o

[PATCH] D113738: [LTO] Allow passing -Os/-Oz as the optimization level

aeubanks updated this revision to Diff 386748. aeubanks added a comment. update Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113738/new/ https://reviews.llvm.org/D113738 Files: clang/lib/CodeGen/BackendUtil.cpp lld/COFF/Config.h lld/COFF/Dr

[PATCH] D113738: [LTO] Allow passing -Os/-Oz as the optimization level

aeubanks updated this revision to Diff 386882. aeubanks added a comment. add more tests Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113738/new/ https://reviews.llvm.org/D113738 Files: clang/lib/CodeGen/BackendUtil.cpp lld/COFF/Config.h lld

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks updated this revision to Diff 326455. aeubanks added a comment. remove extra parameter Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97437/new/ https://reviews.llvm.org/D97437 Files: clang/lib/Driver/ToolChains/MSVC.cpp Index: clang/li

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

aeubanks added a comment. We don't use a VFS for registry searching because we just take the value given by the registry and pass it on, rather than checking for the existence of specific files/directories. But yeah they're probably all absolute paths so it shouldn't matter in those cases. Re

[PATCH] D97437: Rewrite MSVC toolchain discovery with VFS

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG6d52c4819294: Rewrite MSVC toolchain discovery with VFS (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION h

[PATCH] D97608: Move EntryExitInstrumentation pass location

aeubanks created this revision. Herald added subscribers: nikic, pengfei, hiraditya. aeubanks requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. This seems to be more of a Clang thing rather than a generic LLVM thing, so this

[PATCH] D96803: EntryExitInstrumenter: Enable at all optimization levels (PR49143)

aeubanks added a comment. Alternatively: https://reviews.llvm.org/D97608 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D96803/new/ https://reviews.llvm.org/D96803 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/c

[PATCH] D97608: Move EntryExitInstrumentation pass location

aeubanks updated this revision to Diff 327164. aeubanks added a comment. Herald added subscribers: kerbowa, atanasyan, jrtc27, nhaehnle, jvesely, nemanjai, sdardis. missed non-x86 tests Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97608/new/ http

[PATCH] D97608: Move EntryExitInstrumentation pass location

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG040c1b49d7a7: Move EntryExitInstrumentation pass location (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D97816: [clang] Note that -debug-pass only works with legacy PM when using new PM

aeubanks created this revision. aeubanks added a reviewer: ychen. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Follow-up to https://reviews.llvm.org/D97810. Repository: rG LLVM Github Monorepo https://reviews.llvm.org/D9781

[PATCH] D97872: [clang] Don't assert in EmitAggregateCopy on trivial_abi types

aeubanks created this revision. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Fixes PR42961. Repository: rG LLVM Github Monorepo https://reviews.llvm.org/D97872 Files: clang/lib/CodeGen/CGExprAgg.cpp clang/test/CodeGenC

[PATCH] D97872: [clang] Don't assert in EmitAggregateCopy on trivial_abi types

aeubanks updated this revision to Diff 329165. aeubanks added a comment. check IR Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97872/new/ https://reviews.llvm.org/D97872 Files: clang/lib/CodeGen/CGExprAgg.cpp clang/test/CodeGenCXX/trivial_abi

[PATCH] D97872: [clang] Don't assert in EmitAggregateCopy on trivial_abi types

This revision was automatically updated to reflect the committed changes. Closed by commit rGc8227f06b335: [clang] Don't assert in EmitAggregateCopy on trivial_abi types (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97872/new/

[PATCH] D98440: [NPM][CGSCC] FunctionAnalysisManagerCGSCCProxy: do not clear immutable function passes

aeubanks added a comment. This doesn't break the pipeline tests in llvm/test/Other? Running check-llvm with expensive checks is probably a good idea to see if there are any weird issues. Otherwise seems fine to me. Comment at: llvm/unittests/Analysis/CGSCCPassManagerTest.cpp:

[PATCH] D113304: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on eagerly invalidate analyses

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG19867de9e793: [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on… (authored by aeubanks). Changed prior to commit: htt

[PATCH] D114908: [clang] Don't call inheritDefaultTemplateArguments() on CXXDeductionGuideDecl's template parameters

aeubanks created this revision. aeubanks added a reviewer: rsmith. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. A CXXDeductionGuideDecl references its class's template parameters (rather than cloning them). This causes us to cur

[PATCH] D106585: Fix clang debug info irgen of i128 enums

aeubanks added a comment. In D106585#3169898 , @rnk wrote: > This usage of isSameValue seems suspicious: > https://github.com/llvm/llvm-project/blob/main/llvm/lib/IR/LLVMContextImpl.h#L389 > > It seems to allow the possibility that APInts of differing bi

[PATCH] D106585: Fix clang debug info irgen of i128 enums

aeubanks added a comment. In D106585#3170019 , @aeubanks wrote: > In D106585#3169898 , @rnk wrote: > >> This usage of isSameValue seems suspicious: >> https://github.com/llvm/llvm-project/blob/main/llvm/lib/IR/LL

[PATCH] D136474: [CodeView][clang] Add flag to disable emitting command line into CodeView

aeubanks created this revision. Herald added a project: All. aeubanks requested review of this revision. Herald added subscribers: cfe-commits, MaskRay. Herald added a project: clang. In https://reviews.llvm.org/D80833, there were concerns about determinism emitting the commandline into CodeView.

[PATCH] D136474: [CodeView][clang] Add flag to disable emitting command line into CodeView

aeubanks updated this revision to Diff 469733. aeubanks added a comment. add driver test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136474/new/ https://reviews.llvm.org/D136474 Files: clang/include/clang/Basic/CodeGenOptions.def clang/inclu

[PATCH] D136474: [CodeView][clang] Add flag to disable emitting command line into CodeView

aeubanks added a comment. In D136474#3875442 , @aganea wrote: > Thanks for the patch @aeubanks and apologies for this oversight. I am > wondering however if it wouldn’t make more sense to just strip this flag > (-fmessage-length) from the emitted cmd-li

[PATCH] D136474: [CodeView][clang] Add flag to disable emitting command line into CodeView

aeubanks updated this revision to Diff 469954. aeubanks added a comment. default to off Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136474/new/ https://reviews.llvm.org/D136474 Files: clang/include/clang/Basic/CodeGenOptions.def clang/includ

[PATCH] D136474: [CodeView][clang] Add flag to disable emitting command line into CodeView

aeubanks added inline comments. Comment at: clang/lib/Driver/ToolChains/Clang.cpp:4354 + +// Emit codeview command line if requested. +if (Args.hasFlag(options::OPT_gcodeview_command_line, aganea wrote: > aeubanks wrote: > > MaskRay wrote: > > > This need

[PATCH] D136615: [Instrumentation] Remove legacy passes

aeubanks created this revision. aeubanks added a reviewer: asbirlea. Herald added a reviewer: deadalnix. Herald added subscribers: Enna1, hiraditya. Herald added a reviewer: ctetreau. Herald added a project: All. aeubanks requested review of this revision. Herald added projects: clang, LLVM. Herald

[PATCH] D136615: [Instrumentation] Remove legacy passes

aeubanks added a comment. did you mean to point to a different patch? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136615/new/ https://reviews.llvm.org/D136615 ___ cfe-commits mailing list cfe-commits@l

[PATCH] D136615: [Instrumentation] Remove legacy passes

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGef37504879ee: [Instrumentation] Remove legacy passes (authored by aeubanks). Changed prior to commit: https://reviews.llvm.org/D136615?vs=470185&i

[PATCH] D136789: [clang] Remove no-op -fexperimental-new-pass-manager/-fno-legacy-pass-manager flags

aeubanks created this revision. aeubanks added a reviewer: MaskRay. Herald added a subscriber: StephenFan. Herald added a reviewer: ctetreau. Herald added a project: All. aeubanks requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. These have b

[PATCH] D136789: [clang] Remove no-op -fexperimental-new-pass-manager/-fno-legacy-pass-manager flags

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG69b2b7282e92: [clang] Remove no-op -fexperimental-new-pass-manager/-fno-legacy-pass-manager… (authored by aeubanks). Changed prior to commit: htt

[PATCH] D136959: clang: Improve errors for DiagnosticInfoResourceLimit

aeubanks added a comment. see https://reviews.llvm.org/D110665 for how we get line/col from a function. basically, we have a map from function name hashes to source location available in clang if we don't have clang source information for functions available, there's not much more we can do ot

[PATCH] D137116: [AggressiveInstCombine] Remove legacy PM pass

aeubanks created this revision. aeubanks added a reviewer: asbirlea. Herald added a reviewer: deadalnix. Herald added subscribers: ormris, hiraditya. Herald added a reviewer: ctetreau. Herald added a reviewer: ctetreau. Herald added a reviewer: ctetreau. Herald added a reviewer: ctetreau. Herald ad

[PATCH] D136474: [CodeView][clang] Disable emitting command line into CodeView by default and add flag to enable

aeubanks updated this revision to Diff 472351. aeubanks added a comment. on by default Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136474/new/ https://reviews.llvm.org/D136474 Files: clang/include/clang/Basic/CodeGenOptions.def clang/include

[PATCH] D131910: [Windows] Put init_seg(compiler/lib) in llvm.global_ctors

aeubanks created this revision. aeubanks added reviewers: rnk, hans. Herald added subscribers: pengfei, hiraditya. Herald added a project: All. aeubanks requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. Currently we treat in

[PATCH] D131910: [Windows] Put init_seg(compiler/lib) in llvm.global_ctors

aeubanks updated this revision to Diff 452798. aeubanks added a comment. Herald added a reviewer: aaron.ballman. use 400 for lib, update documentation Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D131910/new/ https://reviews.llvm.org/D131910 Files

[PATCH] D131910: [Windows] Put init_seg(compiler/lib) in llvm.global_ctors

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG9181ce623fd8: [Windows] Put init_seg(compiler/lib) in llvm.global_ctors (authored by aeubanks). Repository: rG LLVM Github Monorepo CHANGES SINCE

[PATCH] D131872: [Intrinsics] Add initial support for NonNull attribute

aeubanks accepted this revision. aeubanks added a comment. lgtm Commit message nit: Saying Test plan: ninja check-all isn't super useful, just mentioning the desired changes is good enough (e.g. your code snippet) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https:

[PATCH] D131980: [Passes] Don't run tail-call-elim in -O1

aeubanks created this revision. Herald added subscribers: kosarev, pmatos, asb, ormris, wenlei, kerbowa, arphaman, steven_wu, hiraditya, sbc100, jvesely. Herald added a project: All. aeubanks requested review of this revision. Herald added subscribers: llvm-commits, cfe-commits, aheejin. Herald ad

<    1   2   3   4   5   6   7   8   >