<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
<html lang="en">
<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: -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: -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;">
Marge Bot pushed to branch wip/marge_bot_batch_merge_job at <a href="https://gitlab.haskell.org/ghc/ghc" style="color: #1068bf;">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/d4134e920a79ddfe7abb291964614e4f428c1a24" style="color: #1068bf;">d4134e92</a></strong>
<div>
<span> by Cheng Shao </span> <i> at 2022-11-28T23:48:14-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>compiler: remove unused MO_U_MulMayOflo

We actually only emit MO_S_MulMayOflo and never emit MO_U_MulMayOflo anywhere.
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/b777c34dc8ce302b47cf092a80b307d2116ce39f" style="color: #1068bf;">b777c34d</a></strong>
<div>
<span> by Apoorv Ingle </span> <i> at 2022-11-29T00:19:11-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>Killing cc_fundeps, streamlining kind equality orientation, and type equality processing order

Fixes: #217093
Associated to #19415

This change
* Flips the orientation of the the generated kind equality coercion in canEqLHSHetero;
* Removes `cc_fundeps` in CDictCan as the check was incomplete;
* Changes `canDecomposableTyConAppOk` to ensure we process kind equalities before type equalities and avoiding a call to `canEqLHSHetero` while processing wanted TyConApp equalities
* Adds 2 new tests for validating the change
   - testsuites/typecheck/should_compile/T21703.hs and
   - testsuites/typecheck/should_fail/T19415b.hs (a simpler version of T19415.hs)
* Misc: Due to the change in the equality direction some error messages now have flipped type mismatch errors
* Changes in Notes:
  - Note [Fundeps with instances, and equality orientation] supercedes Note [Fundeps with instances]
  - Added Note [Kind Equality Orientation] to visualize the kind flipping
  - Added Note [Decomposing Dependent TyCons and Processing Wanted Equalties]
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/0ffbe59b7fcad332740cdf2b4999f2948bd237d7" style="color: #1068bf;">0ffbe59b</a></strong>
<div>
<span> by Krzysztof Gogolewski </span> <i> at 2022-11-29T00:19:19-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>Change printing of sized literals to match the proposal

Literals in Core were printed as e.g. 0xFF#16 :: Int16#.
The proposal 451 now specifies syntax 0xFF#Int16.
This change affects the Core printer only - more to be done later.

Part of #21422.
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/a1202ae1bef5f1cf57e1f8a77759ac355e9b9ef5" style="color: #1068bf;">a1202ae1</a></strong>
<div>
<span> by Simon Peyton Jones </span> <i> at 2022-11-29T00:19:19-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>Be a bit more selective about floating bottoming expressions

This MR arranges to float a bottoming expression to the top
only if it escapes a value lambda.

See #22494 and Note [Floating to the top] in SetLevels.

This has a generally beneficial effect in nofib

+-------------------------------++----------+
|                               ||tsv (rel) |
+===============================++==========+
|           imaginary/paraffins ||   -0.93% |
|                imaginary/rfib ||   -0.05% |
|                      real/fem ||   -0.03% |
|                    real/fluid ||   -0.01% |
|                   real/fulsom ||   +0.05% |
|                   real/gamteb ||   -0.27% |
|                       real/gg ||   -0.10% |
|                   real/hidden ||   -0.01% |
|                      real/hpg ||   -0.03% |
|                      real/scs ||  -11.13% |
|         shootout/k-nucleotide ||   -0.01% |
|               shootout/n-body ||   -0.08% |
|   shootout/reverse-complement ||   -0.00% |
|        shootout/spectral-norm ||   -0.02% |
|             spectral/fibheaps ||   -0.20% |
|           spectral/hartel/fft ||   -1.04% |
|         spectral/hartel/solid ||   +0.33% |
|     spectral/hartel/wave4main ||   -0.35% |
|                 spectral/mate ||   +0.76% |
+===============================++==========+
|                     geom mean ||   -0.12% |

The effect on compile time is generally slightly beneficial

Metrics: compile_time/bytes allocated
----------------------------------------------
MultiLayerModulesTH_OneShot(normal)  +0.3%
                  PmSeriesG(normal)  -0.2%
                  PmSeriesT(normal)  -0.1%
                     T10421(normal)  -0.1%
                    T10421a(normal)  -0.1%
                     T10858(normal)  -0.1%
                     T11276(normal)  -0.1%
                    T11303b(normal)  -0.2%
                     T11545(normal)  -0.1%
                     T11822(normal)  -0.1%
                     T12150(optasm)  -0.1%
                     T12234(optasm)  -0.3%
                     T13035(normal)  -0.2%
                     T16190(normal)  -0.1%
                     T16875(normal)  -0.4%
                    T17836b(normal)  -0.2%
                     T17977(normal)  -0.2%
                    T17977b(normal)  -0.2%
                     T18140(normal)  -0.1%
                     T18282(normal)  -0.1%
                     T18304(normal)  -0.2%
                    T18698a(normal)  -0.1%
                     T18923(normal)  -0.1%
                     T20049(normal)  -0.1%
                    T21839r(normal)  -0.1%
                      T5837(normal)  -0.4%
                      T6048(optasm)  +3.2% BAD
                      T9198(normal)  -0.2%
                      T9630(normal)  -0.1%
       TcPlugin_RewritePerf(normal)  -0.4%
             hard_hole_fits(normal)  -0.1%

                          geo. mean  -0.0%
                          minimum    -0.4%
                          maximum    +3.2%

The T6048 outlier is hard to pin down, but it may be the effect of
reading in more interface files definitions. It's a small program for
which compile time is very short, so I'm not bothered about it.

Metric Increase:
    T6048
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/6e78da35e4b40c18f775f04ce7d9bb3d649bb7c0" style="color: #1068bf;">6e78da35</a></strong>
<div>
<span> by Ben Gamari </span> <i> at 2022-11-29T00:19:20-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>testsuite: Mark unpack_sums_6 as fragile due to #22504

This test is explicitly dependent upon runtime, which is generally not
appropriate given that the testsuite is run in parallel and generally
saturates the CPU.
</pre>
</li>
<li>
<strong style="font-weight: bold;"><a href="https://gitlab.haskell.org/ghc/ghc/-/commit/26823b77246d7fceaff74798af1fa92138afeab4" style="color: #1068bf;">26823b77</a></strong>
<div>
<span> by Ben Gamari </span> <i> at 2022-11-29T00:19:20-05:00 </i>
</div>
<pre class="commit-message" style='white-space: pre-wrap; display: block; font-size: 0.8125rem; color: #303030; position: relative; font-family: "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: #fafafa; border-radius: 2px; margin: 0; padding: 8px 12px; border: 1px solid #dbdbdb;'>testsuite: Don't use grep -q in unpack_sums_7

`grep -q` closes stdin as soon as it finds the pattern it is looking
for, resulting in #22484.
</pre>
</li>
</ul>
<h4 style="margin-top: 10px; margin-bottom: 10px;">
30 changed files:
</h4>
<ul>
<li class="file-stats">
<a href="#c898e00d01234ab22d3b485be68db3645f52f220" style="color: #1068bf;">
compiler/GHC/Cmm/MachOp.hs
</a>
</li>
<li class="file-stats">
<a href="#71e696f452eb493722d70306c6f304fc9b2f6a95" style="color: #1068bf;">
compiler/GHC/Cmm/Parser.y
</a>
</li>
<li class="file-stats">
<a href="#6997a0cd37829d3d08db1c48185aaa3d7dfc3e6b" style="color: #1068bf;">
compiler/GHC/CmmToAsm/AArch64/CodeGen.hs
</a>
</li>
<li class="file-stats">
<a href="#94b86a17d123043012c2097c7c9559de28d86ad6" style="color: #1068bf;">
compiler/GHC/CmmToAsm/Wasm/FromCmm.hs
</a>
</li>
<li class="file-stats">
<a href="#76664ab267df4fc0bec2465efd78bf0afacfe3a7" style="color: #1068bf;">
compiler/GHC/CmmToC.hs
</a>
</li>
<li class="file-stats">
<a href="#a49dbda5c8a9c380f638f55cf5ade791db0017cc" style="color: #1068bf;">
compiler/GHC/CmmToLlvm/CodeGen.hs
</a>
</li>
<li class="file-stats">
<a href="#cc763cdd1967f5d921161a32f64136cfcf0332c5" style="color: #1068bf;">
compiler/GHC/Core/Opt/SetLevels.hs
</a>
</li>
<li class="file-stats">
<a href="#f018e77421f495c6f75733fee4a511cfe57bebe8" style="color: #1068bf;">
compiler/GHC/Tc/Solver/Canonical.hs
</a>
</li>
<li class="file-stats">
<a href="#f639b0b31b7245cb7d34ac190b798f5ec1c7da2e" style="color: #1068bf;">
compiler/GHC/Tc/Solver/Interact.hs
</a>
</li>
<li class="file-stats">
<a href="#3b6d78a50c8462304f843b952a88e531718dbbd2" style="color: #1068bf;">
compiler/GHC/Tc/Types/Constraint.hs
</a>
</li>
<li class="file-stats">
<a href="#74295edeac870ee7f2f3484ebaffab5c948b735a" style="color: #1068bf;">
compiler/GHC/Tc/Utils/Unify.hs
</a>
</li>
<li class="file-stats">
<a href="#d3cf57b6a5ebc234dc7689db38ef9b6cf2259c0f" style="color: #1068bf;">
compiler/GHC/Utils/Outputable.hs
</a>
</li>
<li class="file-stats">
<a href="#71247aaad504d89a31c398d6edd6cfc090975b86" style="color: #1068bf;">
testsuite/tests/deSugar/should_compile/T2431.stderr
</a>
</li>
<li class="file-stats">
<a href="#15c8e4e356746b542d334e3e9ccf3c5bc00672e8" style="color: #1068bf;">
testsuite/tests/indexed-types/should_fail/T3330c.stderr
</a>
</li>
<li class="file-stats">
<a href="#194c4c9f540e9ba088bfd59f6a251e87778dd805" style="color: #1068bf;">
testsuite/tests/indexed-types/should_fail/T9662.stderr
</a>
</li>
<li class="file-stats">
<a href="#a3f0c42c513830f3bd1f5243034c7ea4f0861fe4" style="color: #1068bf;">
testsuite/tests/numeric/should_compile/T16402.stderr-ws-32
</a>
</li>
<li class="file-stats">
<a href="#44924fe766978373c43b92417d06bc7838e6e08e" style="color: #1068bf;">
testsuite/tests/numeric/should_compile/T16402.stderr-ws-64
</a>
</li>
<li class="file-stats">
<a href="#3a9fe864a83ed22e050b52131e9f514a91c6fa91" style="color: #1068bf;">
testsuite/tests/patsyn/should_compile/T21531.stderr
</a>
</li>
<li class="file-stats">
<a href="#b37208b88586421af1e9bff92752a43407edcdad" style="color: #1068bf;">
testsuite/tests/polykinds/T11399.stderr
</a>
</li>
<li class="file-stats">
<a href="#f6f3fe949a2cbc72ebd9c2a22b3067595b1831a9" style="color: #1068bf;">
testsuite/tests/polykinds/T14172.stderr
</a>
</li>
<li class="file-stats">
<a href="#d3b20e0cf613360a3142cd563ed6b2148235d81d" style="color: #1068bf;">
testsuite/tests/rep-poly/T13929.stderr
</a>
</li>
<li class="file-stats">
<a href="#0e552c88e6ad6dda61e457558d769f9f7fc08caa" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles1.stderr
</a>
</li>
<li class="file-stats">
<a href="#11d454d9734f1432f982c1261b8ddad6ac9c1e02" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles13.stderr
</a>
</li>
<li class="file-stats">
<a href="#c5d8728cba431493a7f9249d50fa059478ebeedd" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles14.stderr
</a>
</li>
<li class="file-stats">
<a href="#41639098249315c642bf184008c3cd0451a669fa" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles2.stderr
</a>
</li>
<li class="file-stats">
<a href="#abbf4fb3aa1f761971354189e6537b8113d2e6ce" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles3.stderr
</a>
</li>
<li class="file-stats">
<a href="#9f6dfb765b6863e5abfa795af35b9d5712639a5e" style="color: #1068bf;">
testsuite/tests/roles/should_compile/Roles4.stderr
</a>
</li>
<li class="file-stats">
<a href="#ea5c6f85b1fed60ee6ee6de85ef4d558e1112fd5" style="color: #1068bf;">
testsuite/tests/roles/should_compile/T8958.stderr
</a>
</li>
<li class="file-stats">
<a href="#11ca6c36a465280185664c54bed1de3cf206b723" style="color: #1068bf;">
testsuite/tests/simplCore/should_compile/Makefile
</a>
</li>
<li class="file-stats">
<a href="#305c31113fe82345545d1f66d39c8551712ce112" style="color: #1068bf;">
testsuite/tests/simplCore/should_compile/OpaqueNoCastWW.stderr
</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: #666;">

<br>
<a href="https://gitlab.haskell.org/ghc/ghc/-/compare/ac928c5e512b519185959030b79de1bf3a5ef8eb...26823b77246d7fceaff74798af1fa92138afeab4" style="color: #1068bf;">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" style="color: #1068bf;">gitlab.haskell.org</a>. <a href="https://gitlab.haskell.org/-/profile/notifications" target="_blank" rel="noopener noreferrer" class="mng-notif-link" style="color: #1068bf;">Manage all notifications</a> · <a href="https://gitlab.haskell.org/help" target="_blank" rel="noopener noreferrer" class="help-link" style="color: #1068bf;">Help</a>



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