Timeline


and

Mar 23, 2011:

9:17 PM Ticket #5043 (Compiling bytestring-mmap with -fvia-C yields broken .s file, two .size ...) created by jeffwheeler
I tried to install the bytestring-mmap package, which specifies in its …
9:16 PM Changeset in ghc [4a2a508b]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Ian Lynagh <igloo@…>
Document new hsc2hs flags, and use the --cross-safe flag when we're compiling Patch from Brian Bloniarz
7:23 PM Ticket #5042 (STM, newArray, and a stack overflow) created by basvandijk
Ketil Malde …
3:31 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
multiple reloads (diff)
3:12 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
incidence (diff)
3:04 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
(diff)
1:57 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
Assignment of stack locations to temps (diff)
1:54 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
fixed one issue (diff)
1:43 PM Changeset in ghc [cb5260d4]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Edward Z. Yang <ezyang@…>
Immediately tag initialization code to prevent untagged spills. When allocating new objects on the heap, we previously returned a CmmExpr containing the heap pointer as well as the tag expression, which would be added to the code graph upon first usage. Unfortunately, this meant that untagged heap pointers living in registers might be spilled to the stack, where they interacted poorly with garbage collection (we saw this bug specifically with the compacting garbage collector.) This fix immediately tags the register containing the heap pointer, so that unless we have extremely unfriendly spill code, the new pointer will never be spilled to the stack untagged. An alternate solution might have been to modify allocDynClosure to tag the pointer upon the initial register allocation, but not all invocations of allocDynClosure tag the resulting pointer, and threading the consequent CgIdInfo for the cases that did would have been annoying.
1:16 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
extra note about allocDynClosure (diff)
1:11 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
more things wrong (diff)
12:46 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
Note about -ddump-opt-cmm versus -ddump-cmmz/-ddump-raw-cmm (diff)
1:37 AM Ticket #5041 (Incorrect Read deriving for MagicHash constructors) created by dolio
Greetings, A fellow in #haskell on freenode just discovered the following …

Mar 22, 2011:

11:16 PM Ticket #4984 (OS X: ld: warning: -read_only_relocs cannot be used with x86_64) closed by igloo
fixed: Fixed in HEAD and 7.0.3: […]
11:09 PM Building/Architecture/Idiom/PhaseOrdering edited by igloo
(diff)
6:21 PM Changeset in ghc [90d7a88f]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Ian Lynagh <igloo@…>
bindist checker improvements * Some refactoring * Support for Windows filenames * Some support for installed trees (as Windows "bindists" are really install trees)
3:31 PM Ticket #5038 (internal error with rebindable if-then-else) created by Philonous
Defining ifThenElse in a module with RebindableSyntax? enabled and using …
3:25 PM Commentary/Compiler/NewCodeGenStupidity edited by ezyang
Up and Down, Up and Down, I will lead you Up and Down (diff)
3:16 PM Changeset in ghc [419821c7]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Edward Z. Yang <ezyang@…>
Split out "Raw Cmm" dump to its own flag -ddump-raw-cmm Previously, -ddump-cmmz was used to dump out the "Raw Cmm" right before it gets passed to the backends, as well as all of the intermediate phases in the new code generator. Unfortunately, there are a lot of intermediate phases, which means -ddump-cmmz takes a very long time with -fnew-codegen for large programs, even if you're only interested in the final result. Here we split up -ddump-cmmz into -ddump-cmmz (now exclusively for the new code generator) and -ddump-raw-cmm (for both code generation paths.) The old flag was not documented in the manual, so hopefully not too many people will be surprised by this change.
2:17 PM Changeset in ghc [3e3498a]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Edward Z. Yang <ezyang@…>
RednCounts can contain CAFs, so support them in cvtToClosureLbl.
6:43 AM Commentary/Compiler/Backends/LLVM/WIP edited by dterei
(diff)
1:12 AM Ticket #5037 (TH mkName bug) created by igloo
This is accepted: […] but this fails: […] with: […] I expect it …

Mar 21, 2011:

11:43 PM Changeset in ghc [cb28985]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Ian Lynagh <igloo@…>
Rename System.Event to GHC.Event It's just an internal GHC library, for now at least
10:31 PM Changeset in ghc [d8dea07]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Ian Lynagh <igloo@…>
Hack to get the compare tool to work on Windows "bindists"
5:12 PM Commentary/Compiler/NewCodeGen edited by ezyang
Link to Commentary/Compiler/NewCodeGenStupidity (diff)
5:11 PM Commentary/Compiler/NewCodeGenStupidity created by ezyang
Initial revision
4:52 PM Commentary/Compiler/NewCodeGen edited by ezyang
Updates to new codegen play (diff)
4:45 PM Debugging/CompiledCode edited by ezyang
Note about entry and info equivalence (diff)
11:00 AM Changeset in ghc [6c979675]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by simonpj@…>
Fix Trac #5028: zap occ info when doing the binder swap This fixes the Lint error, but still risks leaving stupid let { x=y } bindings in the code. But no time to fix that today. (Leave the ticket open for that reason.)

Mar 20, 2011:

10:18 PM Ticket #5036 (ghc bug: urk! lookup local fingerprint) created by guest
[…]
8:37 PM Ticket #4998 (performance regressions) closed by igloo
worksforme: I've had a look at fannkuchredux and I can't see any slow down, either …
12:51 PM BugSweep edited by michalt
Remove closed tickets. (diff)
12:42 PM WikiStart edited by michalt
Update "Forthcoming plannd releases". (diff)

Mar 19, 2011:

6:50 PM Ticket #5035 (Typo in documentation section 7.8.7.2) created by sankeld
The Phrase: […] Should be […]

Mar 18, 2011:

8:28 PM Ticket #5034 (Performance of Data.Graph.{preorderF, postorderF}) created by michalt
The functions can be quite slow in some cases, but it's easy to improve …
2:21 PM Changeset in ghc [17bfe098]at-defaultsatomicsbetter-ho-cardinalitycardinalitycoerciblecoloured-corecpr-sum-typescrosscross-compiler-alienlessdata-kind-syntaxdecision-procedureencodingghc-7.2ghc-7.4ghc-7.6ghc-7.8ghc-axiomsghc-constraint-solverghc-deferghc-lwc2ghc-new-coghc-new-flavorghc-parmake-gsocghc-spjimp-param-classknown-key-serializationlate-dmdlate-lam-liftlocal-gcmonad-compnew-demand-to-mergenewcgno-pred-tyoverlapping-tyfamsprofilingreal-src-loc-spansdocsilent-sc-argssimdsrclocsupercompilertc-arrowstc-untouchablesth-newth-new-7.6ticky-for-all-letstype-holes-branchtype-natstype-nats-simpleunboxed-tuple-argumentsunboxed-tuple-arguments2wip/Cabal-1.20wip/T4404wip/T5084wip/T7704wip/T8545-ghc-7.8wip/T8592wip/T8776wip/T8995-level-generalisationwip/cbv-conv-thunkwip/common-contextwip/cpr-vs-jpwip/exprAritywip/nested-cprwip/pattern-synonymswip/recurs-compatwip/simdwip/th-new by Simon Marlow <marlowsd@…>
pruneSparkQueue: handle CAFs
Note: See TracTimeline for information about the timeline view.