Feed Advanced Search

Today

bgamari updated the diff for D4870: base: Add missing instances for Data.Ord.Down.
  • Add missing @sinces
Mon, Jun 18, 4:33 PM
bgamari updated subscribers of D4870: base: Add missing instances for Data.Ord.Down.

That set of missing @since annotations gave me an idea: @sjakobi's :doc output could potentially be used by a linter of sorts to ensure such annotations aren't forgotten in newly introduced definitions.

Mon, Jun 18, 4:33 PM
bgamari updated the diff for D4870: base: Add missing instances for Data.Ord.Down.
  • update test output
Mon, Jun 18, 3:03 PM
bgamari accepted D4762: Fix gcc.exe: error: CreateProcess: No such file or directory.
In D4762#134268, @Phyx wrote:

@bgamari what's the timeline for updating the GCC toolchain for 8.6? I assume since you've already branched you're close to a release?

In that case I'd accept this patch for that branch and we can revert it for 8.8. the GCC patch has been merged into mingw-w64 however
they're having some trouble building GCC 8 due to some Ada build failures... and the Ada maintainers aren't being very helpful.

I've started looking at it as well but I'm stretched a bit thin here and don't know when it'll be solved.

It's far from ideal but at least @erikd et all would be able to (slowly) build their projects again. The flag is undocumented so we can remove it without issue I think.

Mon, Jun 18, 2:42 PM
bgamari added a comment to D4870: base: Add missing instances for Data.Ord.Down.

I think the current version ought to work, however.

Mon, Jun 18, 2:31 PM
bgamari added a comment to D4870: base: Add missing instances for Data.Ord.Down.

Sorry, @RyanGlScott, wasn't quite ready for review yet. I should have finished the patch before adding reviewers.

Mon, Jun 18, 2:31 PM
bgamari updated the diff for D4870: base: Add missing instances for Data.Ord.Down.

Really fix it

Mon, Jun 18, 2:31 PM
bgamari updated the diff for D4870: base: Add missing instances for Data.Ord.Down.

Fix it

Mon, Jun 18, 2:31 PM
bgamari created D4870: base: Add missing instances for Data.Ord.Down.
Mon, Jun 18, 1:51 PM
bgamari requested changes to D4787: Dump the NCG Cmm->Cmm optimization pass when -ddump-cmm-verbose is enabled.

I generally think it is best if flags remain orthogonal where possible. Let's just fix the users guide description.

Mon, Jun 18, 11:12 AM
bgamari created D4869: configure: Fix libnuma detection logic.
Mon, Jun 18, 11:02 AM
bgamari updated the diff for D4717: [WIP] CoreToStg: Try treating String unpackings as single-entry.

Handle top-level case as well

Mon, Jun 18, 8:56 AM
bgamari added inline comments to D4717: [WIP] CoreToStg: Try treating String unpackings as single-entry.
Mon, Jun 18, 8:40 AM
bgamari added a comment to D4868: Rewrite fasta in more idiomatic Haskell.

Out of curiosity, have you compare the performance of this to the previous version (with -fstate-hack, of course)?

Mon, Jun 18, 8:40 AM

Yesterday

bgamari committed rGHC04e9fe5c7d3c: Add -Werror=compat (authored by int-index).
Add -Werror=compat
Sun, Jun 17, 1:41 PM
bgamari closed D4860: Add -Werror=compat.
Sun, Jun 17, 1:41 PM
bgamari committed rGHC008ea12dd93b: Use __FILE__ for Cmm assertion locations, fix #8619 (authored by osa1).
Use __FILE__ for Cmm assertion locations, fix #8619
Sun, Jun 17, 1:41 PM
bgamari closed D4862: Use __FILE__ for Cmm assertion locations, fix #8619.
Sun, Jun 17, 1:41 PM
bgamari committed rGHCb8e349922b48: UNREG: fix CmmRegOff large offset handling on W64 platforms (authored by trofi).
UNREG: fix CmmRegOff large offset handling on W64 platforms
Sun, Jun 17, 1:41 PM
bgamari closed D4856: UNREG: fix CmmRegOff large offset handling on W64 platforms.
Sun, Jun 17, 1:40 PM
bgamari committed rGHCc63754118cf6: Provide a better error message for unpromotable data constructor contexts (authored by RyanGlScott).
Provide a better error message for unpromotable data constructor contexts
Sun, Jun 17, 1:40 PM
bgamari closed D4728: Provide a better error message for unpromotable data constructor contexts.
Sun, Jun 17, 1:40 PM
bgamari committed rGHC793902e6891c: Improve documentation of Eq, Ord instances for Float and Double (authored by Azel).
Improve documentation of Eq, Ord instances for Float and Double
Sun, Jun 17, 11:55 AM
bgamari closed D4736: Change to improve Eq, Ord, Float and Double documentation (trac issue #15078)..
Sun, Jun 17, 11:55 AM
bgamari committed rGHCdf0f148feae4: Improve error message when importing an unusable package (authored by sgillespie).
Improve error message when importing an unusable package
Sun, Jun 17, 11:55 AM
bgamari closed D4783: Improve error message when importing an unusable package (Trac #4806).
Sun, Jun 17, 11:55 AM
bgamari requested changes to D4780: a naive but simple fix to resolve trac ticket 15207, because currently stgcrun.c when compiled with gcc and then assembled with clang/apple as/llvm assembler fails..

I would much rather we had a better understanding of what is going on here before we brush the issue under the rug; afterall, it may also break in other configurations.

Sun, Jun 17, 11:54 AM
bgamari accepted D4860: Add -Werror=compat.

Thanks, this looks good!

Sun, Jun 17, 11:54 AM
bgamari added inline comments to D4739: Fix "redundant constraint" warnings when meets functional dependencies.
Sun, Jun 17, 11:54 AM
bgamari updated the diff for D4717: [WIP] CoreToStg: Try treating String unpackings as single-entry.

Rebase

Sun, Jun 17, 11:54 AM
bgamari updated the Trac tickets for D4717: [WIP] CoreToStg: Try treating String unpackings as single-entry.
Sun, Jun 17, 11:54 AM
bgamari accepted D4728: Provide a better error message for unpromotable data constructor contexts.

Looks quite reasonable to me.

Sun, Jun 17, 11:54 AM
bgamari committed rGHCccd8ce405db8: Handle DuplicateRecordFields correctly in filterImports (fixes #14487) (authored by adamgundry).
Handle DuplicateRecordFields correctly in filterImports (fixes #14487)
Sun, Jun 17, 11:24 AM
bgamari closed D4805: Handle DuplicateRecordFields correctly in filterImports (fixes #14487).
Sun, Jun 17, 11:24 AM
bgamari committed rGHC9897440ec9fb: testsuite: Mark print022 as broken on 32-bit platforms (authored by bgamari).
testsuite: Mark print022 as broken on 32-bit platforms
Sun, Jun 17, 11:24 AM
bgamari committed rGHC749bc1a0b08c: testsuite: Mark T3001-2 as broken on 32-bit platforms (authored by bgamari).
testsuite: Mark T3001-2 as broken on 32-bit platforms
Sun, Jun 17, 11:24 AM
bgamari requested changes to D4713: Mark code related symbols as @function not @object.

Alright, I suppose this is reasonable. However, we should make this logic and its comment much more discoverable. Let's make the comment a Note; I'm not entirely sure where to refer to it from yet.

Sun, Jun 17, 11:23 AM
bgamari updated subscribers of D4849: Add cmpeq pack int instructions.
Sun, Jun 17, 10:45 AM
bgamari accepted D4862: Use __FILE__ for Cmm assertion locations, fix #8619.

Looks good to me assuming it validates.

Sun, Jun 17, 10:44 AM
bgamari accepted D4783: Improve error message when importing an unusable package (Trac #4806).

Thanks for the patch, @sgillespie! This looks quite reasonable to me.

Sun, Jun 17, 10:44 AM
bgamari accepted D4805: Handle DuplicateRecordFields correctly in filterImports (fixes #14487).

Looks good, thanks!

Sun, Jun 17, 10:44 AM
bgamari accepted D4817: rts/linker/{SymbolExtras,elf_got}.c: protect mmaped addresses from writes after being initially manipulated.

Great, thanks!

Sun, Jun 17, 10:15 AM
bgamari committed rGHCd1c7239c037e: configure: Fail when bootstrapping with GHC 8.2.1 (authored by bgamari).
configure: Fail when bootstrapping with GHC 8.2.1
Sun, Jun 17, 9:55 AM
bgamari committed rGHC0db05ad9e4d2: Bump process submodule (authored by bgamari).
Bump process submodule
Sun, Jun 17, 9:55 AM
bgamari committed rGHC76b343f5d2e9: Revert "rts: Use .cfi_{start|end}proc directives" (authored by bgamari).
Revert "rts: Use .cfi_{start|end}proc directives"
Sun, Jun 17, 9:55 AM
bgamari added a reverting change for rGHC86210b238b86: rts: Use .cfi_{start|end}proc directives: rGHC76b343f5d2e9: Revert "rts: Use .cfi_{start|end}proc directives".
Sun, Jun 17, 9:54 AM
bgamari abandoned D4863: configure: Fail when bootstrapping with GHC 8.2.1.

Merged with d1c7239c037e267873658160b5c290f08f0d6502.

Sun, Jun 17, 9:54 AM
bgamari requested changes to D4861: Introduce flag -keep-hspp-files (#10869).

Thanks for your contribution @RolandSenn!

Sun, Jun 17, 9:54 AM
bgamari updated the diff for D4863: configure: Fail when bootstrapping with GHC 8.2.1.

Fix syntax

Sun, Jun 17, 9:05 AM
bgamari created D4863: configure: Fail when bootstrapping with GHC 8.2.1.
Sun, Jun 17, 9:05 AM
bgamari retitled D4861: Introduce flag -keep-hspp-files (#10869) from Docu for flag: keep-hspp-files (#10869) to Introduce flag -keep-hspp-files (#10869).
Sun, Jun 17, 8:49 AM

Sat, Jun 16

bgamari committed rGHC8df24474d019: Warn about implicit kind variables with -Wcompat (authored by int-index).
Warn about implicit kind variables with -Wcompat
Sat, Jun 16, 10:52 PM
bgamari committed rGHC4cd552184cbc: base: Add default implementation for Data.Bits.bitSize (authored by bgamari).
base: Add default implementation for Data.Bits.bitSize
Sat, Jun 16, 10:52 PM
bgamari committed rGHCd55035f5fe13: Revert "Amend configure script to support lndir build tree" (authored by bgamari).
Revert "Amend configure script to support lndir build tree"
Sat, Jun 16, 10:51 PM
bgamari added a reverting change for rGHC8ee9c574a6d2: Amend configure script to support lndir build tree: rGHCd55035f5fe13: Revert "Amend configure script to support lndir build tree".
Sat, Jun 16, 10:51 PM
bgamari committed rGHC60e4bb4d305b: Enhanced constant folding (authored by Sylvain Henry <hsyl20@gmail.com>).
Enhanced constant folding
Sat, Jun 16, 10:51 PM
bgamari closed D4857: base: Add default implementation for Data.Bits.bitSize.
Sat, Jun 16, 10:51 PM
bgamari closed D4834: Warn about implicit kind variables with -Wcompat.
Sat, Jun 16, 10:51 PM
bgamari accepted D4834: Warn about implicit kind variables with -Wcompat.

Looks good to me. Thanks!

Sat, Jun 16, 10:51 PM
bgamari accepted rGHCfea04defa648: Enhanced constant folding.
Sat, Jun 16, 10:51 PM
bgamari added a comment to rGHCfea04defa648: Enhanced constant folding.

Builders upgraded and re-applied in 60e4bb4d305bc1a65457ee79b1e69c11b9ed747d.

Sat, Jun 16, 10:51 PM
bgamari raised a concern with rGHC8ee9c574a6d2: Amend configure script to support lndir build tree.

Strangely this appears to break the OS X build. Reverting for now.

Sat, Jun 16, 10:35 PM
bgamari committed rGHCf998947fe93a: circleci: Add a reference to the documentation on the Wiki (authored by bgamari).
circleci: Add a reference to the documentation on the Wiki
Sat, Jun 16, 3:40 PM
bgamari committed rGHC819d8efd568e: circleci: Bump fedora docker image tag (authored by bgamari).
circleci: Bump fedora docker image tag
Sat, Jun 16, 3:40 PM
bgamari committed rGHCde34a71460a9: rts: Remove use of __USE_MINGW_ANSI_STDIO (authored by bgamari).
rts: Remove use of __USE_MINGW_ANSI_STDIO
Sat, Jun 16, 3:40 PM
bgamari committed rGHCdbc8c0f8a9a5: base: Improve the documentation of the enumFrom series of functions (authored by Azel).
base: Improve the documentation of the enumFrom series of functions
Sat, Jun 16, 3:40 PM
bgamari closed D4737: Change to improve the documentation of the enumFrom series of functions (trac issue #15134)..
Sat, Jun 16, 3:39 PM
bgamari committed rGHCa81b99d0b14b: Bump nofib submodule (authored by bgamari).
Bump nofib submodule
Sat, Jun 16, 3:39 PM
bgamari committed rNOFIBa3b0f2b2cff8: Set -O2 via NoFibHcOpts instead of individual Makefiles. (authored by AndreasK).
Set -O2 via NoFibHcOpts instead of individual Makefiles.
Sat, Jun 16, 3:39 PM
bgamari closed D4829: Set -O2 via NoFibHcOpts instead of individual Makefiles..
Sat, Jun 16, 3:39 PM
bgamari committed rNOFIB22240ca8034d: Eliminate trailing whitespace (authored by mpickering).
Eliminate trailing whitespace
Sat, Jun 16, 3:39 PM
bgamari closed D4045: Eliminate trailing whitespace.
Sat, Jun 16, 3:39 PM
bgamari committed rNOFIBcf032a225ce9: Several fixes to work with ghc-head (authored by Douglas Wilson <douglas.wilson@gmail.com>).
Several fixes to work with ghc-head
Sat, Jun 16, 3:39 PM
bgamari closed D4391: Several fixes to work with ghc-head.
Sat, Jun 16, 3:39 PM
bgamari committed rNOFIB3ef116aa7ed5: Don't use binary output for real/eff. (authored by AndreasK).
Don't use binary output for real/eff.
Sat, Jun 16, 3:39 PM
bgamari closed D4705: Don't use binary output for real/eff..
Sat, Jun 16, 3:39 PM
bgamari committed rNOFIBf7de4c92475a: Use /usr/bin/env to get perl path instead of using a hardcoded path. (authored by AndreasK).
Use /usr/bin/env to get perl path instead of using a hardcoded path.
Sat, Jun 16, 3:39 PM
bgamari closed D4756: Use /usr/bin/env to get perl path instead of using a hardcoded path..
Sat, Jun 16, 3:39 PM
bgamari created D4859: [WIP] testsuite: Add tests for #15270.
Sat, Jun 16, 3:25 PM
bgamari created D4858: [TESTING] Try applying "Simplify callSiteInline a little" again.
Sat, Jun 16, 3:25 PM
bgamari added a reverting change for rGHC9dbf66d74e65: Revert "Simplify callSiteInline a little": D4858: [TESTING] Try applying "Simplify callSiteInline a little" again.
Sat, Jun 16, 3:25 PM
bgamari committed rGHCcd95c2ffdc51: Preserve parenthesis in function application in typechecker (authored by wz1000).
Preserve parenthesis in function application in typechecker
Sat, Jun 16, 1:09 PM
bgamari closed D4822: Preserve parenthesis in function application in typechecker.
Sat, Jun 16, 1:09 PM
bgamari committed rGHC86210b238b86: rts: Use .cfi_{start|end}proc directives (authored by bgamari).
rts: Use .cfi_{start|end}proc directives
Sat, Jun 16, 1:08 PM
bgamari closed D4781: rts: Use .cfi_{start|end}proc directives.
Sat, Jun 16, 1:08 PM
bgamari committed rGHC1f2ed994de55: testsuite: Mark overflow1 as broken on 32-bit platforms due to #15255 (authored by bgamari).
testsuite: Mark overflow1 as broken on 32-bit platforms due to #15255
Sat, Jun 16, 1:08 PM
bgamari closed D4855: testsuite: Mark overflow1 as broken on 32-bit platforms due to #15255.
Sat, Jun 16, 1:08 PM
bgamari committed rGHC1ab2dcb04d96: testsuite: Mark num009 as broken due to #15062 (authored by bgamari).
testsuite: Mark num009 as broken due to #15062
Sat, Jun 16, 1:08 PM
bgamari closed D4854: testsuite: Mark num009 as broken due to #15062.
Sat, Jun 16, 1:08 PM
bgamari committed rGHC8ee9c574a6d2: Amend configure script to support lndir build tree (authored by adamgundry).
Amend configure script to support lndir build tree
Sat, Jun 16, 1:08 PM
bgamari closed D4853: Amend configure script to support lndir build tree.
Sat, Jun 16, 1:08 PM
bgamari created D4857: base: Add default implementation for Data.Bits.bitSize.
Sat, Jun 16, 12:48 PM
bgamari accepted D4829: Set -O2 via NoFibHcOpts instead of individual Makefiles..

Seems reasonable to me.

Sat, Jun 16, 11:43 AM
bgamari requested changes to D4852: --show-iface: Qualify all non-local names.

Requesting note.

Sat, Jun 16, 11:43 AM
bgamari accepted D4856: UNREG: fix CmmRegOff large offset handling on W64 platforms.

Looks reasonable to me.

Sat, Jun 16, 10:33 AM

Fri, Jun 15

bgamari created D4855: testsuite: Mark overflow1 as broken on 32-bit platforms due to #15255.
Fri, Jun 15, 4:42 PM
bgamari created D4854: testsuite: Mark num009 as broken due to #15062.
Fri, Jun 15, 4:32 PM
bgamari committed rGHCfe770c211631: Built-in Natural literals in Core (authored by Sylvain Henry <hsyl20@gmail.com>).
Built-in Natural literals in Core
Fri, Jun 15, 4:02 PM
bgamari closed D4212: Built-in Natural literals in Core.
Fri, Jun 15, 4:02 PM
bgamari accepted D4736: Change to improve Eq, Ord, Float and Double documentation (trac issue #15078)..

Looks good to me.

Fri, Jun 15, 3:58 PM