- ghc-devs - Haskell.org

Build problem
by Simon Peyton Jones 03 Jul '26

03 Jul '26

Friends My GHC builds have started falling over with | Copy file: _build/stage1/bin/runghc => _build/bindist/ghc-10.1.20260702-x86_64-unknown-linux/bin/runghc-10.1.20260702 | Copy directory: _build/stage1/lib => _build/bindist/ghc-10.1.20260702-x86_64-unknown-linux ]0;Running for 9m50s [36386/36406], predicted 1s (99%)(/home/simonpj/code/HEAD) ]0;Running for 9m55s [36386/36406], predicted 1s (99%)(/home/simonpj/code/HEAD) | Copy directory: _build/stage1/lib/x86_64-linux-ghc-10.1.20260702/rts-1.0.3/include => _build/bindist/ghc-10.1.20260702-x86_64-unknown-linux # ghc-pkg (for binary-dist-dir) ]0;Finished in 10m00s(/home/simonpj/code/HEAD) Non optional builder 'sphinx-build' is not specified in 'hadrian/cfg/system.config' file. Build failed. What should I do? Thanks Simon

3 2

GHC 9.12.5 RC2 is now available
by Magnus Viernickel 24 Jun '26

24 Jun '26

Dear GHC users, distributors and devs, I'm pleased to announce that the second release candidate for GHC 9.12.5 is now available from downloads.haskell.org and on GHCup. GHC 9.12.5 is a bug fix release. For a long form announcement, check out https://discourse.haskell.org/t/ghc-9-12-5-rc2-is-now-available/14313/1. Mind that the very important overhaul of the semaphore-compat library to v2 is included in this release and that the semaphore feature will only start working with cabal-install 3.18 again, which is sadly a breaking but necessary change. Your testing of the release is much appreciated. Many thanks to the nixpkgs and stackage package set distributors for testing the first release candidate. If you find any issues with the candidate, please open an issue on the ghc issue tracker. If there are no critical regressions are discovered within this release candidate, it will be published in about 2 weeks of time (the week of 8 July 2026). Best Magnus

2 1

GHC Status Report 2026
by Andreas Klebinger 02 Jun '26

02 Jun '26

Hello devs, I'm currently in the process of preparing the Status report (to be given at Zurihac/HIW). If you you have a feature that made it into GHC during the last year or something big you are working on you think we should highlight during the talk let me know. I will go through the release notes to look for things but it would be a shame if something slips through the cracks! Andreas

4 3

Boot libraries shipping with 9.12.5
by Magnus Viernickel 29 May '26

29 May '26

Dear boot library maintainers, We are preparing releases for GHC 9.12.5. As part of the release process, we are notifying you about the boot library versions that we are planning to include. As always, please reply swiftly if you need us to include commits that are not included in the release we are planning to ship. The boot library versions for ghc 9.12.5 are as follows: - Cabal: v3.14.2.0 - binary: v0.8.9.3 - bytestring: v0.12.2.0 - containers: v0.7 - directory: v1.3.10.1 - exceptions: v0.10.12 - file-io: v0.1.6 - filepath: v1.5.5.0 - haskeline: v0.8.4.1 - mtl: v2.3.2 - os-string: v2.0.10 - parsec: v3.1.18.0 - pretty: v1.1.3.6 - terminfo: v0.4.1.7 - text: v2.1.4 - time: v1.14 - transformers: v0.6.3.0 - unix: v2.8.8.0 - process: v1.6.29.0 - Win32: v2.14.2.1 - array: v0.5.8.0 - deepseq: v1.5.1.0-1 - hpc: v0.7.0.2 - semaphore-compat: v2.0.0 - stm: v2.5.3.1-8 - xhtml: v3000.2.2.1 - hpc: v0.69 - hsc2hs: v0.68.10 Thank you for notifying me timely if this doesn't look right to you. Best Magnus

2 2

MacOS and perf runner currently offline
by Magnus Viernickel 27 May '26

27 May '26

Dear GHC devs, it appears that our perf runner which at the same time is the machine that serves as a gateway for our macos runners is offline. That means darwin and perf CI is currently stuck. I unfortunately don't have access to that machine and thus cannot resolve the issue for now but have contacted persons who do have access. I will notify you as soon as the issue is resolved. Best Magnus

1 0

Re: Boot libraries shipping with 9.12.5
by Oleg Grenrus 27 May '26

27 May '26

What is? - hpc: v0.69 - hpc: v0.7.0.2 ??? Also sorting the list for the next time would be a small but nice improvemt - Oleg On 5/26/26 12:50, Magnus Viernickel wrote: > Dear boot library maintainers, > > We are preparing releases for GHC 9.12.5. As part of the release > process, we are notifying you about the boot library versions that we > are planning to include. As always, please reply swiftly if you need > us to include commits that are not included in the release we are > planning to ship. > > The boot library versions for ghc 9.12.5 are as follows: > > - Cabal: v3.14.2.0 > - binary: v0.8.9.3 > - bytestring: v0.12.2.0 > - containers: v0.7 > - directory: v1.3.10.1 > - exceptions: v0.10.12 > - file-io: v0.1.6 > - filepath: v1.5.5.0 > - haskeline: v0.8.4.1 > - mtl: v2.3.2 > - os-string: v2.0.10 > - parsec: v3.1.18.0 > - pretty: v1.1.3.6 > - terminfo: v0.4.1.7 > - text: v2.1.4 > - time: v1.14 > - transformers: v0.6.3.0 > - unix: v2.8.8.0 > - process: v1.6.29.0 > - Win32: v2.14.2.1 > - array: v0.5.8.0 > - deepseq: v1.5.1.0-1 > - hpc: v0.7.0.2 > - semaphore-compat: v2.0.0 > - stm: v2.5.3.1-8 > - xhtml: v3000.2.2.1 > - hpc: v0.69 > - hsc2hs: v0.68.10 > > Thank you for notifying me timely if this doesn't look right to you. > > Best > > Magnus > > >

2 1

release progress GHC 9.12.5
by Magnus Viernickel 27 May '26

27 May '26

Dear GHC release stakeholders, We are approaching a release candidate for GHC 9.12.5. Currently there's only one blocking patch (https://gitlab.haskell.org/ghc/ghc/-/merge_requests/15640) that still needs to be backported and the test-suite seems to be doing well. Here are some important notes for this release of GHC 9.12. 1. Our Darwin CI is now running on MacOS version 26 for the aarch64 runners and on MacOS version 15 for the x86_64 runner. The SDK version on the MacOS 15 runner is 26.2, the SDK version on the aarch64 runners is 26.4. Note that this may affect bindists. 2. See https://gitlab.haskell.org/ghc/ghc/-/issues/27286 for an exhaustive list of MRs that have been backported. 3. If you are a distributor of haskell packages like stackage or nixpkgs, I would appreciate testing an RC with you, as soon as it's there. Please reach out if that's the case. 4. In ncoordinatio with the cabal developers, the GHC team has decided to stop maintaining the V1 ghc semaphore protocol that had some fundamental architectural flaws. While this is technically a (performance) regression (cabal versions that only support v1 will fall back to -j1 even when passed -jsem with 9.12.5) and adds a significant feature, it is the most pragmatic approach to fixing the glaring issues with the previous semaphore implementation without having to maintain a fundamentally broken design. Please check out https://gitlab.haskell.org/ghc/ghc/-/merge_requests/15729 for further information. We expect that `-jsem` will work out of the box with the next released cabal version and 9.12.5. I expect a release candidate to appear this week, depending on progress of the blocking patch. Please don't hesitate to reach out if you have any questions or remarks. Best Magnus

1 0

Please rebase if you see an x86_64-darwin failure on your branches
by Magnus Viernickel 26 May '26

26 May '26

Hi GHC Devs, after merging https://gitlab.haskell.org/ghc/ghc/-/merge_requests/16096, I will reactivate our x86_64 darwin runner to run x86_darwin-m1 jobs on the pipelines. This will mainly impact full-ci. If you see a failure on x86_64 darwin testing a process program that exits with a segfault, then you probably will want to rebase. I hope you're having a great day. Best Magnus

1 0

Namespace-specific warning pragma example in user guide?
by Viktor Dukhovni 24 May '26

24 May '26

In section 6.3.4.3 of the GHC 9.14.1 user guide (the same text dates back to earlier versions), I am puzzled by the example of a namespace-specific warning: https://downloads.haskell.org/ghc/9.14.1/docs/users_guide/exts/explicit_nam… type family Head xs where Head (x : _) = x pattern Head x <- (head -> x) {-# WARNING in "x-partial" data Head "this is a partial type synonym" #-} The text string refers to a type synonym but the namespace tag is "data". Which is it, and what's an example in which the intended warning would be issued? -- Viktor. 🇺🇦 Слава Україні!

1 0

CI "skipped"
by Simon Peyton Jones 22 May '26

22 May '26

Friends In !15175, I can't get CI to run! It just says "skipped". Can anyone tell me how to fix this? Thanks Simon [image: image.png]

2 2