- ghc-devs - Haskell.org

Upcoming fork for GHC 10.0, new changelog system for GHC 10.1
by Zubin Duggal 11 Apr '26

11 Apr '26

Hi all, We plan to fork for GHC 10.0 next week on Thursday (April 9th). Please make sure any merge requests you would like to see merged are merged in time for the fork. If the merge request is particularly important, and you are confident it is ready for a merge, please let me know by responding to this email, and I will do my best in trying to land it before the fork. If a merge request is still incomplete but fixes a critical bug we should fix in GHC 10.0, mark it with the backport needed:10.0 label and we will consider it for backporting. After the GHC 10.0 fork, master will be bumped to GHC 10.1. Any MRs that land after this point will require changelog entries, according to the scheme implemented in https://gitlab.haskell.org/ghc/ghc/-/merge_requests/15830 This will require all MRs, unless tagged with "no-changelog" label (reserved for entirely non user facing changes like CI), to populate an entry in the changelog.d/ directory. You will need to create a file with a descriptive name, based on your changes, for instance (picking a patch that recently landed): changelog.d/fix-datakindsinfix: ``` section: language synopsis: Fix a bug where GHC would erroneously accept infix promoted data constructors without enabling :extension:`DataKinds`. issues: #26737 mrs: !15609 description: { As a result, you may need to enable :extension:`DataKinds` in code that did not previously require it. } ``` The `section` field controls what section of the release notes your entry ends up in, available sections are defined in changelog.d/config (https://gitlab.haskell.org/ghc/ghc/-/blob/05ddce3b88cfe9729811c4ce843c8ccbe…) The `synopsis` field is a brief description of the change. This is the main payload `issues` is a space seperated list of issues fixed by the change `mrs` is a space seperated list of mrs. If you are modifying a feature which already has a changelog entry, add your MR to the `mrs` list for the existing entry instead of creating a new changelog entry. `description` is an optional field where you can expand on the description of your change. It will be rendered on a new line following the `synopsis` entry in the changelog. Both `section` and `description` accept the existing RST syntax we use for release notes entries. A new CI job ensures that each patch (unless otherwise labeled) touches a properly formatted changelog file with an `mrs` field including the MR number of the patch. This job will not block subsequent build/validation jobs from running, but it will be needed to pass before Marge can pick up the MR for merging. If you have any questions or feedback about this system, now is the time to give it. It will be much easier adding any additional metadata we might want to track in changelog entries now, rather than after the fork. Thanks, Zubin.

1 3

Loss of debug output
by Simon Peyton Jones 04 Apr '26

04 Apr '26

Suddenly, after pulling today's patches, my standard workflow doesn't work. - -ddump-simpl no longer shows the simplified Core - -dverbose-core2core no longer shows the output of each pass What has happened?? Thanks Simon

1 1

If static001 fails on darwin, please rebase
by Magnus Viernickel 04 Apr '26

04 Apr '26

Dear GHC devs, I have recently updated the darwin (MacOS) runners to a newer OS and toolchain. That caused a test called static001 to fail. I have landed a fix. If your MR pipelines fail on darwin because of static001, you can fix this issue by rebasing on origin/master. I am sorry for the inconvenience caused. Thanks everyone for your patience. I am seeing a light at the end of the tunnel of bringing things up to date, you almost made it! Have a nice weekend. Best Magnus

1 0

Boot libraries for GHC 10.0
by Zubin Duggal 02 Apr '26

02 Apr '26

Hi all, The GHC 10.0 branch is scheduled to be cut next week. You are receiving this email because you are listed as a maintainer on a boot library or have uploaded a version to hackage recently. Below are the current versions of all boot libraries and submodules in the GHC tree. Libraries behind Hackage: - file-io: 0.1.5 in tree, 0.2.0 on Hackage - filepath: 1.5.4.0 in tree, 1.5.5.0 on Hackage - os-string: 2.0.8 in tree, 2.0.10 on Hackage - process: 1.6.26.1 in tree, 1.6.27.0 on Hackage Unless there are objections, we will attempt to bump these four libraries to their latest Hackage versions before the fork. Libraries currently at the latest Hackage version: - Cabal 3.16.1.0 - Cabal-syntax 3.16.1.0 - Win32 2.14.2.1 - array 0.5.8.0 - binary 0.8.9.3 - bytestring 0.12.2.0 - containers 0.8 - deepseq 1.5.2.0 - directory 1.3.10.1 - exceptions 0.10.12 - haskeline 0.8.4.1 - hpc 0.7.0.2 - hsc2hs 0.68.10 - libffi-clib 3.5.2 - mtl 2.3.2 - parsec 3.1.18.0 - pretty 1.1.3.6 - semaphore-compat 1.0.0 - stm 2.5.3.1 - template-haskell-lift 0.1.0.0 - template-haskell-quasiquoter 0.1.0.0 - terminfo 0.4.1.7 - text 2.1.4 - time 1.15 - transformers 0.6.3.0 - unix 2.8.8.0 - xhtml 3000.4.0.0 I presume there will also be a corresponding Cabal 3.18 release, which we will coordinate with the Cabal developlers. Otherwise, if you would like a different version of any library to ship with GHC 10.0, please open a GHC ticket at https://gitlab.haskell.org/ghc/ghc/-/issues/new so we can track it. Thanks, Zubin

1 0

Recent Windows CI issues
by Cheng Shao 01 Apr '26

01 Apr '26

Hi GHC devs, Me and Magnus have brought up a new temporary windows runner to address the recent GHC CI traffic jam, as a part of our joint work to make CI faster. If you see your windows CI job fails with T17912, before retrying the job, please rebase your branch against the latest master branch, so that it includes latest fixes and can reliably pass on the new runner. If you spot other spurious windows CI failures, please open a new issue and tag us. Cheers, Cheng Shao

1 0

GHC 9.12.4 is now available
by Zubin Duggal 27 Mar '26

27 Mar '26

The GHC developers are very pleased to announce the release of GHC 9.12.4. Binary distributions, source distributions, and documentation are available at [downloads.haskell.org][] and via [GHCup](https://www.haskell.org/ghcup/). GHC 9.12.4 is a bug-fix release fixing many issues of a variety of severities and scopes, including: * Fixed a critical code generation regression where sub-word division produced incorrect results (#26711, #26668), similar to the bug fixed in 9.12.2 * Numerous fixes for register allocation bugs, preventing data corruption when spilling and reloading registers (#26411, #26526, #26537, #26542, #26550) * Fixes for several compiler crashes, including issues with CSE (#25468), and the simplifier(#26681), implicit parameters (#26451), and the type-class specialiser (#26682) * Fixed cast worker/wrapper incorrectly firing on INLINE functions (#26903) * Fixed LLVM backend miscompilation of bit manipulation operations (#20645, #26065, #26109) * Fixed associated type family and data family instance changes not triggering recompilation (#26183, #26705) * Fixed negative type literals causing the compiler to hang (#26861) * Improvements to determinism of compiler output (#26846, #26858) * Fixes for eventlog shutdown deadlocks (#26573) and lost wakeups in the RTS (#26324) * Fixed split sections support on Windows (#26696, #26494) and the LLVM backend (#26770) * Fixes for the bytecode compiler, PPC native code generator, and Wasm backend * The runtime linker now supports COMMON symbols (#6107) * Improved backtrace support: backtraces for `error` exceptions are now evaluated at throw time * `NamedDefaults` now correctly requires the class to be standard or have an in-scope default declaration, and handles poly-kinded classes (#25775, #25778, #25882) * ... and many more A full accounting of these fixes can be found in the [release notes][]. As always, GHC's release status, including planned future releases, can be found on the GHC Wiki [status][]. GHC development is sponsored by: * [Juspay](https://juspay.com/) * [QBayLogic](https://qbaylogic.com/) * [Channable](https://www.channable.com/) * [Haskell Foundation](https://haskell.foundation/) * [Serokell](https://serokell.io/) * [Well-Typed](https://well-typed.com/) * [Tweag](https://www.tweag.io/) * [Dotcom-Monitor](https://www.dotcom-monitor.com/) * [LoadView](https://www.loadview-testing.com/) * [Web Hosting Buddy](https://webhostingbuddy.com/) * [Find My Electric](https://www.findmyelectric.com/) * [Standard Chartered](https://www.sc.com) * [UpCloud](https://upcloud.com) * [Mercury](https://mercury.com) We would like to thank these sponsors and other anonymous contributors whose on-going financial and in-kind support has facilitated GHC maintenance and release management over the years. Finally, this release would not have been possible without the hundreds of open-source contributors whose work comprise this release. As always, do give this release a try and open a [ticket][] if you see anything amiss. [release notes]: https://gitlab.haskell.org/ghc/ghc/-/blob/ghc-9.12/docs/users_guide/9.12.4-… [status]: https://gitlab.haskell.org/ghc/ghc/-/wikis/GHC-status [downloads.haskell.org] https://downloads.haskell.org/ghc/9.12.4 [ticket]: https://gitlab.haskell.org/ghc/ghc/-/issues/new

1 0

Re: Strange "Unknown flag" message
by Simon Peyton Jones 26 Mar '26

26 Mar '26

> > I think to avoid this you will need to be careful to only apply these > flags when the BOOTSTRAP_TH flag is set. > In particular, I think what's happening is that we try to find the > GHC.KnownKey module and then start pulling in all of ghc-internal. Ah yes. The command line does indeed invoke the *bootstrap *compiler, which of course does now know these flags. *But what business does the bootstrap compiler have with compiling GHC.Internal.Classes? * I would be entirely unsurprised if the bootstrap compiler stumbled on compiling a module that is so closely coupled to GHC's internals. Simon On Wed, 25 Mar 2026 at 13:02, Teo Camarasu <teo(a)informal.codes> wrote: > Hi Simon, > > Note that this is occurring while compiling `ghc-boot-th-next`. This > package vendors the TemplateHaskell interfaces from ghc-internal and is > built by the boot compiler. You will find a bunch of CPP in the > TemplateHaskell modules like `GHC.Internal.TH.Syntax` that switches out > `ghc-internal` imports for `base` imports when bootstrapping so we can > compile with the boot compiler. For more context see `Note [Bootstrapping > Template Haskell]`. > > I think to avoid this you will need to be careful to only apply these > flags when the BOOTSTRAP_TH flag is set. > In particular, I think what's happening is that we try to find the > GHC.KnownKey module and then start pulling in all of ghc-internal. > > Cheers, > Teo > > On Wed, 25 Mar 2026, at 12:56 PM, Brandon Allbery via ghc-devs wrote: > > I'd suspect it's your boot compiler producing that error, and that what > changed might have been the file being compiled, causing a recompile with > the wrong compiler? > > On Wed, Mar 25, 2026 at 8:53 AM Simon Peyton Jones via ghc-devs < > ghc-devs(a)haskell.org> wrote: > > Friends > > After 10 days of working on !15719 I have suddenly started getting > > /home/simonpj/code/HEAD-19/libraries/ghc-boot-th-next/../ghc-internal/src/GHC/Internal/Classes.hs:10:17: > error: [GHC-04924] > Unknown flag in {-# OPTIONS_GHC #-} pragma: -fdefines-known-key-names > | > 10 | {-# OPTIONS_GHC -fdefines-known-key-names #-} > | ^^^^^^^^^^^^^^^^^^^^^^^^^ > > I have no idea why this was fine until 10am this morning, and has suddenly > started happening now. > > *How does GHC decide what runtime flags it allows? * > `-fdefines-known-key-names` is in GHC.Driver.Session.fFlagsDeps. > > > I'm boggled that it worked until 10am and then abruptly stopped. > > Any advice? > > Simon > > _______________________________________________ > ghc-devs mailing list -- ghc-devs(a)haskell.org > To unsubscribe send an email to ghc-devs-leave(a)haskell.org > > > > -- > brandon s allbery kf8nh > allbery.b(a)gmail.com > _______________________________________________ > ghc-devs mailing list -- ghc-devs(a)haskell.org > To unsubscribe send an email to ghc-devs-leave(a)haskell.org > > >

3 7

Strange "Unknown flag" message
by Simon Peyton Jones 25 Mar '26

25 Mar '26

Friends After 10 days of working on !15719 I have suddenly started getting /home/simonpj/code/HEAD-19/libraries/ghc-boot-th-next/../ghc-internal/src/GHC/Internal/Classes.hs:10:17: error: [GHC-04924] Unknown flag in {-# OPTIONS_GHC #-} pragma: -fdefines-known-key-names | 10 | {-# OPTIONS_GHC -fdefines-known-key-names #-} | ^^^^^^^^^^^^^^^^^^^^^^^^^ I have no idea why this was fine until 10am this morning, and has suddenly started happening now. *How does GHC decide what runtime flags it allows? * `-fdefines-known-key-names` is in GHC.Driver.Session.fFlagsDeps. I'm boggled that it worked until 10am and then abruptly stopped. Any advice? Simon

3 2

GHC 9.12.4-rc1 is now available
by Zubin Duggal 19 Mar '26

19 Mar '26

The GHC developers are very pleased to announce the availability of the release candidate for GHC 9.12.4. Binary distributions, source distributions, and documentation are available at [downloads.haskell.org][] and via [GHCup](https://www.haskell.org/ghcup/). GHC 9.12.4 is a bug-fix release fixing many issues of a variety of severities and scopes, including: * Fixed a critical code generation regression where sub-word division produced incorrect results (#26711, #26668), similar to the bug fixed in 9.12.2 * Numerous fixes for register allocation bugs, preventing data corruption when spilling and reloading registers (#26411, #26526, #26537, #26542, #26550) * Fixes for several compiler crashes, including issues with CSE (#25468), SetLevels (#26681), implicit parameters (#26451), and the type-class specialiser (#26682) * Fixed cast worker/wrapper incorrectly firing on INLINE functions (#26903) * Fixed LLVM backend miscompilation of bit manipulation operations (#20645, #26065, #26109) * Fixed associated type family and data family instance changes not triggering recompilation (#26183, #26705) * Fixed negative type literals causing the compiler to hang (#26861) * Improvements to determinism of compiler output (#26846, #26858) * Fixes for eventlog shutdown deadlocks (#26573) and lost wakeups in the RTS (#26324) * Fixed split sections support on Windows (#26696, #26494) and the LLVM backend (#26770) * Fixes for the bytecode compiler, PPC native code generator, and Wasm backend * The runtime linker now supports COMMON symbols (#6107) * Improved backtrace support: backtraces for `error` exceptions are now evaluated at throw time * `NamedDefaults` now correctly requires the class to be standard or have an in-scope default declaration, and handles poly-kinded classes (#25775, #25778, #25882) * ... and many more A full accounting of these fixes can be found in the [release notes][]. As always, GHC's release status, including planned future releases, can be found on the GHC Wiki [status][]. This release candidate will have a two-week testing period. If all goes well the final release will be available the week of 26 March 2026. GHC development is sponsored by: * [Juspay](https://juspay.com/) * [QBayLogic](https://qbaylogic.com/) * [Channable](https://www.channable.com/) * [Haskell Foundation](https://haskell.foundation/) * [Serokell](https://serokell.io/) * [Well-Typed](https://well-typed.com/) * [Tweag](https://www.tweag.io/) * [Dotcom-Monitor](https://www.dotcom-monitor.com/) * [LoadView](https://www.loadview-testing.com/) * [Web Hosting Buddy](https://webhostingbuddy.com/) * [Find My Electric](https://www.findmyelectric.com/) * [Standard Chartered](https://www.sc.com) * [UpCloud](https://upcloud.com) * [Mercury](https://mercury.com) We would like to thank these sponsors and other anonymous contributors whose on-going financial and in-kind support has facilitated GHC maintenance and release management over the years. Finally, this release would not have been possible without the hundreds of open-source contributors whose work comprise this release. As always, do give this release a try and open a [ticket][] if you see anything amiss. [release notes]: https://gitlab.haskell.org/ghc/ghc/-/blob/ghc-9.12/docs/users_guide/9.12.4-… [status]: https://gitlab.haskell.org/ghc/ghc/-/wikis/GHC-status [downloads.haskell.org] https://downloads.haskell.org/ghc/9.12.4-rc1 [ticket]: https://gitlab.haskell.org/ghc/ghc/-/issues/new

2 4

Wrapping up GHC work next week
by Matthew Pickering 17 Mar '26

17 Mar '26

Hi all, Since my last day working on GHC will be next Tuesday (24th March), I'll be unassigning myself from all the issues I'm currently assigned to and closing my open PRs. Thank you all for the collaboration over the last 11 years of working on the project. My first commit was all the way back in March 2015 and since then I've contributed over 1000 commits to the compiler, it's been quite a journey! This also means that I’m unlikely to be available for reviewing patches or commenting on issues after next week. If there is anything high priority that would particularly benefit from my input before then, please let me know. Cheers, Matt

2 1