[PATCH] D64665: [MachinePipeliner] Refactor schedule emission logic

[James Molloy via llvm-commits]

Sorry!

Thanks for checking, I will sort this tomorrow. Very embarrassing.

On Thu, 1 Aug 2019, 18:24 Jinsong Ji, <jji at us.ibm.com> wrote:

> Looks like you did upload the wrong version.
>
>
> *https://reviews.llvm.org/D64665?vs=212147&id=212800&whitespace=ignore-most#toc*
> <https://reviews.llvm.org/D64665?vs=212147&id=212800&whitespace=ignore-most#toc>
> shows nothing new comparing to 212147 (July 29).
>
>
>
> Best,
>
> Jinsong Ji (纪金松), PhD.
>
> XL/LLVM on Power Compiler Development
> E-mail: jji at us.ibm.com
>
> [image: Inactive hide details for James Molloy ---08/01/2019 11:53:45
> AM---Hi, Did you test the latest version? Those errors were all f]James
> Molloy ---08/01/2019 11:53:45 AM---Hi, Did you test the latest version?
> Those errors were all fixed just before I
>
> From: James Molloy <jmolloy at google.com>
> To: reviews+D64665+public+6971cdf24f5c9ab4 at reviews.llvm.org
> Cc: bcahoon at quicinc.com, jji at us.ibm.com, Matthew.Arsenault at amd.com,
> hfinkel at anl.gov, llvm-commits at lists.llvm.org, jun.l at samsung.com,
> florian_hahn at apple.com
> Date: 08/01/2019 11:53 AM
> Subject: [EXTERNAL] Re: [PATCH] D64665: [MachinePipeliner] Refactor
> schedule emission logic
> ------------------------------
>
>
>
> Hi,
>
> Did you test the latest version? Those errors were all fixed just before I
> sent the last email (unless I somehow managed to upload the wrong
> version...)
>
> (The compiler options I use are slightly more permissive which is why
> these errors didn't show during development)
>
>
> James
>
> On Thu, 1 Aug 2019, 15:47 Jinsong Ji via Phabricator, <
> *reviews at reviews.llvm.org* <reviews at reviews.llvm.org>> wrote:
>
>    jsji added a comment.
>
>    How do you build and test? I can't config & build it with clang.
>
>
>
>    ================
>    Comment at: lib/CodeGen/MachineLoopUtils.cpp:1
>    +//=- MachineLoopUtils.cpp - Functions for manipulating loops
>    ----------------=//
>    +//
>    ----------------
>    New file, you need to update `llvm/lib/CodeGen/CMakeLists.txt` as well.
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4287
>    +  SmallVector<int, 4> LiveStages;
>    +  for (int I = 0; I < NumPrologs; ++I) {
>    +    LiveStages.push_back(I);
>    ----------------
>    `comparison of integers of different signs: 'int' and 'unsigned int'`
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4296
>    +  // stage NumStages-1, Epilog 1 performs stage NumStages-2 THEN
>    NumStages-1.
>    +  for (int I = 0; I < NumPrologs; ++I) {
>    +    int StartStage = NumStages - 1 - I;
>    ----------------
>    `comparison of integers of different signs: 'int' and 'unsigned int'`
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4299
>    +    CGBlocks.insert(std::next(Kernel), Kernel->peel(LPD_Back));
>    +    for (int Stage = StartStage + 1; Stage < NumStages; ++Stage) {
>    +      CGBlocks.insert(std::next(Kernel), Kernel->peel(LPD_Back));
>    ----------------
>    `comparison of integers of different signs: 'int' and 'unsigned int'`
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4304
>    +  auto EBI = std::next(Kernel);
>    +  for (int I = 0; I < NumPrologs; ++I) {
>    +    int StartStage = NumStages - 1 - I;
>    ----------------
>    `comparison of integers of different signs: 'int' and 'unsigned int'`
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4308
>    +    EBI++->setLiveStages({StartStage});
>    +    for (int Stage = StartStage + 1; Stage < NumStages; ++Stage) {
>    +      EBI++->setLiveStages({Stage});
>    ----------------
>    `comparison of integers of different signs: 'int' and 'unsigned int'`
>
>
>    ================
>    Comment at: lib/CodeGen/MachinePipeliner.cpp:4417
>    +      auto *MI = SU->getInstr();
>    +      InstrInfos[MI] = {Schedule.stageScheduled(SU), Cycle, Index++};
>    +      if (MI->isPHI())
>    ----------------
>    `.../llvm-git/llvm-project/llvm/lib/CodeGen/MachinePipeliner.cpp:4417:22:
>    error: no viable overloaded '='
>          InstrInfos[MI] = {Schedule.stageScheduled(SU), Cycle, Index++};
>          ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>    .../llvm-git/llvm-project/llvm/lib/CodeGen/MachinePipeliner.cpp:4190:10:
>    note: candidate function (the implicit copy assignment operator) not
>    viable: cannot convert initializer list argument to 'const (anonymous
>    namespace)::CGBlock::InstrInfo'
>      struct InstrInfo {
>             ^
>    .../llvm-git/llvm-project/llvm/lib/CodeGen/MachinePipeliner.cpp:4190:10:
>    note: candidate function (the implicit move assignment operator) not
>    viable: cannot convert initializer list argument to '(anonymous
>    namespace)::CGBlock::InstrInfo'
>      struct InstrInfo {
>    `
>
>
>    Repository:
>      rL LLVM
>
>    CHANGES SINCE LAST ACTION
>      *https://reviews.llvm.org/D64665/new/*
>    <https://reviews.llvm.org/D64665/new/>
>
> *https://reviews.llvm.org/D64665* <https://reviews.llvm.org/D64665>
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190801/9f59bf9b/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graycol.gif
Type: image/gif
Size: 105 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190801/9f59bf9b/attachment-0001.gif>