<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
<html lang="en" style='--code-editor-font: var(--default-mono-font, "GitLab Mono"), JetBrains Mono, Menlo, DejaVu Sans Mono, Liberation Mono, Consolas, Ubuntu Mono, Courier New, andale mono, lucida console, monospace;'>
<head>
<meta content="text/html; charset=US-ASCII" http-equiv="Content-Type">
<title>
GitLab
</title>

<style data-premailer="ignore" type="text/css">
a { color: #1068bf; }
</style>


<style>img {
max-width: 100%; height: auto;
}
body {
font-size: 0.875rem;
}
body {
-webkit-text-shadow: rgba(255,255,255,0.01) 0 0 1px;
}
body {
font-family: var(--default-regular-font, "GitLab Sans"),-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,"Noto Sans",Ubuntu,Cantarell,"Helvetica Neue",sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji"; font-size: inherit;
}
</style>
</head>
<body style='font-size: inherit; -webkit-text-shadow: rgba(255,255,255,0.01) 0 0 1px; font-family: var(--default-regular-font, "GitLab Sans"),-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,"Noto Sans",Ubuntu,Cantarell,"Helvetica Neue",sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji";'>
<div class="content">

<h3 style="margin-top: 20px; margin-bottom: 10px;">
Simon Peyton Jones pushed to branch wip/simplifier-tweaks at <a href="https://gitlab.haskell.org/ghc/ghc">Glasgow Haskell Compiler / GHC</a>
</h3>
<h4 style="margin-top: 10px; margin-bottom: 10px;">
Commits:
</h4>
<ul>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/232301dea8f23f2af5e3dc3261ff2cf7c47c2f58">232301de</a></strong>
<div>
<span> by Simon Peyton Jones </span> <i> at 2024-01-22T13:28:08+00:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 14px; color: #333238; position: relative; font-family: var(--default-mono-font, "GitLab Mono"),"JetBrains Mono","Menlo","DejaVu Sans Mono","Liberation Mono","Consolas","Ubuntu Mono","Courier New","andale mono","lucida console",monospace; word-break: break-all; word-wrap: break-word; background-color: #fbfafd; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>Several improvements to the handling of coercions

* Make `mkSymCo` and `mkInstCo` smarter
  Fixes #23642

* Fix return role of `SelCo` in the coercion optimiser.
  Fixes #23617

* Make the coercion optimiser `opt_trans_rule` work better for newtypes
  Fixes #23619
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/badc5de420ddc839a3c00f65d7fe50d1949b2916">badc5de4</a></strong>
<div>
<span> by Simon Peyton Jones </span> <i> at 2024-01-22T13:28:08+00:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 14px; color: #333238; position: relative; font-family: var(--default-mono-font, "GitLab Mono"),"JetBrains Mono","Menlo","DejaVu Sans Mono","Liberation Mono","Consolas","Ubuntu Mono","Courier New","andale mono","lucida console",monospace; word-break: break-all; word-wrap: break-word; background-color: #fbfafd; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>Simplifier improvements

This MR started as: allow the simplifer to do more in one pass,
arising from places I could see the simplifier taking two iterations
where one would do.  But it turned into a larger project, because
these changes unexpectedly made inlining blow up, especially join
points in deeply-nested cases.

The net result is good: a 1.5% improvement in compile time.  The table
below shows changes over 1%.

The main changes are:

* The SimplEnv now has a seInlineDepth field, which says how deep
  in unfoldings we are.  See Note [Inline depth] in Simplify.Env

* Avoid repeatedly simplifying coercions.
  see Note [Avoid re-simplifying coercions] in Simplify.Iteration
  As you'll see from the Note, this makes use of the seInlineDepth.

* Allow Simplify.Utils.postInlineUnconditionally to inline variables
  that are used exactly once. See Note [Post-inline for single-use things].

* Allow Simplify.Iteration.simplAuxBind to inline used-once things.
  This is another part of Note [Post-inline for single-use things], and
  is really good for reducing simplifier iterations in situations like
      case K e of { K x -> blah }
  wher x is used once in blah.

* Make GHC.Core.SimpleOpt.exprIsConApp_maybe do some simple case
  elimination.  Note [Case elim in exprIsConApp_maybe]

* Many new or rewritten Notes.  E.g. Note [Avoiding simplifying repeatedly]

Join points
~~~~~~~~~~~
* Be very careful about inlining join points. See
  Note [Duplicating join points] in GHC.Core.Opt.Simplify.Iteration

* When making join points, don't do so if the join point is so small
  it will immediately be inlined; check uncondInlineJoin.

* When considering inlining a join point, never do so unless
  there is a positive gain: see (DJ5) in Note [Duplicating join points].

* Do not float join points at all, except to top level.
  See GHC.Core.Opt.SetLevels.dontFloatNonRec

* Do not add an unfolding to a join point at birth.  This is a tricky one
  and has a long Note [Do not add unfoldings to join points at birth]
  It shows up in two places
  - In `mkDupableAlt` do not add an inlining
  - (trickier) In `simplLetUnfolding` don't add an unfolding for a
    fresh join point
  I am not fully satisifed with this, but it works and is well documented.

* In GHC.Core.Unfold.sizeExpr, make jumps small, so that we don't penalise
  having a non-inlined join point.

* Use plan A for dataToTag and tagToEnum

I discovered that GHC.HsToCore.Pmc.Solver.Types.trvVarInfo was very
delicately balanced.  It's a small, heavily used, overloaded function
and it's important that it inlines. By a fluke it was before, but at
various times in my journey it stopped doing so.  So I added an INLINE
pragma to it.

    Metrics: compile_time/bytes allocated
    ------------------------------------------------
           CoOpt_Singletons(normal)   -8.2% GOOD
                LargeRecord(normal)  -22.7% GOOD
                  PmSeriesS(normal)   -4.1%
                  PmSeriesT(normal)   -3.1%
                  PmSeriesV(normal)   -1.6%
                     T11195(normal)   -1.9%
                     T12227(normal)  -19.9% GOOD
                     T12545(normal)   -5.4%
                     T12707(normal)   -2.1% GOOD
                 T13253-spj(normal)  -13.1% GOOD
                     T13386(normal)   -1.6%
                     T14766(normal)   -2.2% GOOD
                    T15630a(normal)          NEW
                     T15703(normal)  -13.5% GOOD
                     T16577(normal)   -4.3% GOOD
                     T17096(normal)   -4.4%
                     T17516(normal)   -0.2%
                     T18223(normal)  -16.3% GOOD
                     T18282(normal)   -5.3% GOOD
                     T18730(optasm)          NEW
                     T18923(normal)   -3.7% GOOD
                    T21839c(normal)   -2.3% GOOD
                      T3064(normal)   -1.3%
                      T5030(normal)  -16.1% GOOD
                   T5321Fun(normal)   -1.5%
                      T6048(optasm)  -11.8% GOOD
                       T783(normal)   -1.4%
                      T8095(normal)   -5.9% GOOD
                      T9630(normal)   -5.1% GOOD

                      T9020(optasm)   +1.5%
                    T18698a(normal)   +1.5%  BAD
                     T14683(normal)   +1.2%
        DsIncompleteRecSel3(normal)   +1.2%
MultiComponentModulesRecomp(normal)   +1.0%
    MultiLayerModulesRecomp(normal)   +1.9%
   MultiLayerModulesTH_Make(normal)   +1.4%
                     T10421(normal)   +1.9%  BAD
                    T10421a(normal)   +3.0%
                     T13056(optasm)   +1.1%
                     T13253(normal)   +1.0%
                      T1969(normal)   +1.1%  BAD
                     T15304(normal)   +1.7%
                      T9675(optasm)   +1.2%
                      T9961(normal)   +2.4%  BAD

                          geo. mean   -1.5%
                          minimum    -22.7%
                          maximum     +3.0%
Metric Decrease:
    CoOpt_Singletons
    LargeRecord
    T12227
    T12707
    T12990
    T13253-spj
    T13536a
    T14766
    T15703
    T16577
    T18223
    T18282
    T18923
    T21839c
    T5030
    T6048
    T8095
    T9630
Metric Increase:
    T10421
    T18698a
    T1969
    T9961
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/a7b153492e9462720511b310cbffce9e0121df6a">a7b15349</a></strong>
<div>
<span> by Simon Peyton Jones </span> <i> at 2024-01-22T13:28:08+00:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 14px; color: #333238; position: relative; font-family: var(--default-mono-font, "GitLab Mono"),"JetBrains Mono","Menlo","DejaVu Sans Mono","Liberation Mono","Consolas","Ubuntu Mono","Courier New","andale mono","lucida console",monospace; word-break: break-all; word-wrap: break-word; background-color: #fbfafd; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>Improve postInlineUnconditionally

This commit adds two things to postInlineUnconditionally:

1. Do not postInlineUnconditionally join point, ever.
   Doing so does not reduce allocation, which is the main point,
   and with join points that are used a lot it can bloat code.
   See point (1) of Note [Duplicating join points] in
   GHC.Core.Opt.Simplify.Iteration.

2. Do not postInlineUnconditionally a strict (demanded) binding.
   It will not allocate a thunk (it'll turn into a case instead)
   so again the main point of inlining it doesn't hold.  Better
   to check per-call-site.
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/9f5fa612e0068d10ddf383a06fba11d090765c29">9f5fa612</a></strong>
<div>
<span> by Simon Peyton Jones </span> <i> at 2024-01-23T12:26:02+00:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 14px; color: #333238; position: relative; font-family: var(--default-mono-font, "GitLab Mono"),"JetBrains Mono","Menlo","DejaVu Sans Mono","Liberation Mono","Consolas","Ubuntu Mono","Courier New","andale mono","lucida console",monospace; word-break: break-all; word-wrap: break-word; background-color: #fbfafd; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>More wibbles

* Inline join points whose RHS just calls another join point
* Don't float join points at all (SetLevels)
* Ensure that WorkWrap preserves lambda binders, in case of join points
</pre>
</li>
</ul>
<h4 style="margin-top: 10px; margin-bottom: 10px;">
30 changed files:
</h4>
<ul>
<li class="file-stats">
<a href="#783e5dae6e86931f06700fc088fb7d48c8a07386">
compiler/GHC/Core/Coercion.hs
</a>
</li>
<li class="file-stats">
<a href="#975dc08a8e7942b32d621f617d5a9c1b668601dd">
compiler/GHC/Core/Coercion/Opt.hs
</a>
</li>
<li class="file-stats">
<a href="#c3967bb9d3e8f5aae2dd111b5a335b48c21c1999">
compiler/GHC/Core/Opt/Arity.hs
</a>
</li>
<li class="file-stats">
<a href="#cc763cdd1967f5d921161a32f64136cfcf0332c5">
compiler/GHC/Core/Opt/SetLevels.hs
</a>
</li>
<li class="file-stats">
<a href="#2f46b19cb85e3f7b4e72305644bc50015628c41d">
compiler/GHC/Core/Opt/Simplify/Env.hs
</a>
</li>
<li class="file-stats">
<a href="#8664fd2691eea70244c0a1bae2f51f2ad8a36eef">
compiler/GHC/Core/Opt/Simplify/Inline.hs
</a>
</li>
<li class="file-stats">
<a href="#ae6d91a5d028418bbf1431347d659e744e0a3128">
compiler/GHC/Core/Opt/Simplify/Iteration.hs
</a>
</li>
<li class="file-stats">
<a href="#37879a759bc2ca5cc8f1f35235fe96e3e55b57c9">
compiler/GHC/Core/Opt/Simplify/Monad.hs
</a>
</li>
<li class="file-stats">
<a href="#48fbb5cdea308650de5756521feb28ec68819b9b">
compiler/GHC/Core/Opt/Simplify/Utils.hs
</a>
</li>
<li class="file-stats">
<a href="#b6a5ba32bafb8fbda933538b3007e755fef6f101">
compiler/GHC/Core/Opt/SpecConstr.hs
</a>
</li>
<li class="file-stats">
<a href="#a2c999a27f983737a4e40710396b1c7aaa5d4015">
compiler/GHC/Core/Opt/Stats.hs
</a>
</li>
<li class="file-stats">
<a href="#5b215b0e109123e4bef5434854f18e2f61c45b88">
compiler/GHC/Core/Opt/WorkWrap.hs
</a>
</li>
<li class="file-stats">
<a href="#11ffe98a94d798427bc600e4fcfe899407536346">
compiler/GHC/Core/SimpleOpt.hs
</a>
</li>
<li class="file-stats">
<a href="#578c7c3857d66d963736ff6742f3433a0e8d01b7">
compiler/GHC/Core/TyCo/Rep.hs
</a>
</li>
<li class="file-stats">
<a href="#2811a7297b8aa206197ac1f5dabd0818e3c7ec5a">
compiler/GHC/Core/Unfold.hs
</a>
</li>
<li class="file-stats">
<a href="#940913dd549c6b1c334daafbc4b7eef29c94d924">
compiler/GHC/Core/Unfold/Make.hs
</a>
</li>
<li class="file-stats">
<a href="#1a7aba0daeafab195716dd25432479804a55ab60">
compiler/GHC/Core/Utils.hs
</a>
</li>
<li class="file-stats">
<a href="#6983b6dc1a4f225d4e72180ad45d8484d9d32931">
compiler/GHC/HsToCore/Pmc/Solver/Types.hs
</a>
</li>
<li class="file-stats">
<a href="#f2b2b84d138b362920ab9f144f58bf16254aab64">
compiler/GHC/Iface/Type.hs
</a>
</li>
<li class="file-stats">
<a href="#06ff7bac58fd8cfe0c17b81963b03f4cce86a065">
compiler/GHC/IfaceToCore.hs
</a>
</li>
<li class="file-stats">
<a href="#03ec4f8741105271af29a61fdb47c8feba9bbcfa">
compiler/GHC/Tc/Solver/Equality.hs
</a>
</li>
<li class="file-stats">
<a href="#20aaf8344f379f354fe31dd0c1c4db4ddc5b17aa">
compiler/GHC/Types/Id.hs
</a>
</li>
<li class="file-stats">
<a href="#930ade98b040b89136455bb644c3cee18c23ffdc">
compiler/GHC/Types/Tickish.hs
</a>
</li>
<li class="file-stats">
<a href="#a65c5dfcf1c6d1e05003fb0b52f8ffef3e577e74">
testsuite/tests/driver/inline-check.stderr
</a>
</li>
<li class="file-stats">
<a href="#40f72248511536d518e207be3b158d8c1d12e740">
testsuite/tests/numeric/should_compile/T19641.stderr
</a>
</li>
<li class="file-stats">
<a href="#852d4ab9cf8a80734f7f06cc119356a8eeb2f3ed">
testsuite/tests/perf/compiler/T15630.hs
</a>
</li>
<li class="file-stats">
<a href="#af27a4a12a18c81f10b65d7b524c008b4b76d59a">
<span class="new-file">
+
testsuite/tests/perf/compiler/T15630a.hs
</span>
</a>
</li>
<li class="file-stats">
<a href="#0832112831f7f82257c7088886bf1627bdfe4ff7">
testsuite/tests/simplCore/should_compile/T18730.hs

testsuite/tests/perf/compiler/T18730.hs
</a>
</li>
<li class="file-stats">
<a href="#57916c7e301a76bcc59b8156c32f917000882a4e">
testsuite/tests/simplCore/should_compile/T18730_A.hs

testsuite/tests/perf/compiler/T18730_A.hs
</a>
</li>
<li class="file-stats">
<a href="#fd4a0b65d74153aed8b33b01a04eb3ff7442cd45">
testsuite/tests/perf/compiler/all.T
</a>
</li>
</ul>
<h5 style="margin-top: 10px; margin-bottom: 10px; font-size: 0.875rem;">
The diff was not included because it is too large.
</h5>

</div>
<div class="footer" style="margin-top: 10px;">
<p style="font-size: small; color: #737278;">

<br>
<a href="https://gitlab.haskell.org/ghc/ghc/-/compare/e4ff310c8e4b75049add18c2025f32657bc2e0af...9f5fa612e0068d10ddf383a06fba11d090765c29">View it on GitLab</a>.
<br>
You're receiving this email because of your account on <a target="_blank" rel="noopener noreferrer" href="https://gitlab.haskell.org">gitlab.haskell.org</a>. <a href="https://gitlab.haskell.org/-/profile/notifications" target="_blank" rel="noopener noreferrer" class="mng-notif-link">Manage all notifications</a> · <a href="https://gitlab.haskell.org/help" target="_blank" rel="noopener noreferrer" class="help-link">Help</a>



</p>
</div>
</body>
</html>