summaryrefslogtreecommitdiff
path: root/compiler/codeGen
Commit message (Collapse)AuthorAgeFilesLines
...
* fix warningSimon Marlow2009-07-281-1/+0
|
* Remove old 'foreign import dotnet' codeSimon Marlow2009-07-273-7/+0
| | | | It still lives in darcs, if anyone wants to revive it sometime.
* Remove GHC's haskell98 dependencyIan Lynagh2009-07-245-5/+5
|
* Remove unused importsIan Lynagh2009-07-0713-15/+0
|
* Trim unused imports detected by new unused-import codesimonpj@microsoft.com2009-07-061-1/+0
|
* Support for -fwarn-unused-do-bind and -fwarn-wrong-do-bind, as per #3263Max Bolingbroke2009-07-018-16/+19
|
* Add missing StgPrimCallOp case to isSimpleOpIan Lynagh2009-06-111-0/+1
|
* Add PrimCall to the STG layer and update Core -> STG translationDuncan Coutts2009-06-093-2/+19
| | | | | | | | | | It adds a third case to StgOp which already hold StgPrimOp and StgFCallOp. The code generation for the new StgPrimCallOp case is almost exactly the same as for out-of-line primops. They now share the tailCallPrim function. In the Core -> STG translation we map foreign calls using the "prim" calling convention to the StgPrimCallOp case. This is because in Core we represent prim calls using the ForeignCall stuff. At the STG level however the prim calls are really much more like primops than foreign calls.
* Fix #3132: a case of bogus code generationSimon Marlow2009-06-181-2/+28
|
* Remove old GUM/GranSim codeSimon Marlow2009-06-021-5/+1
|
* Make some showSDoc's use OneLineMode rather than PageModeIan Lynagh2009-03-311-2/+2
|
* Debugging by Sesame Street:dias@eecs.tufts.edu2009-04-031-1/+1
| | | | | | | | | | | | | | | | One of these things is not like the others: stdPattern :: [LRep] -> Maybe StgHalfWord stdPattern reps = case reps of [] -> Just ARG_NONE -- just void args, probably [N] -> Just ARG_N [P] -> Just ARG_N [F] -> Just ARG_F [D] -> Just ARG_D [L] -> Just ARG_L Today's debugging session was brought to you by the letter P.
* Better handling of node parameter in calling conventionsdias@eecs.tufts.edu2009-03-251-3/+7
| | | | | | - Previously, the node was taken as a parameter, then ignored, for static closures. Goofy. Now, the vestigial node parameters are gone.
* When calling gc, avoid saving node in static closuresdias@eecs.tufts.edu2009-03-232-4/+5
|
* Code simplifications due to call/return separation; some improvements to how ↵dias@eecs.tufts.edu2009-03-232-22/+22
| | | | node argument is managed
* Calls with and without passing node arguments more clearly separateddias@eecs.tufts.edu2009-03-233-6/+6
|
* Another small step: call and return conventions specified separately when ↵dias@eecs.tufts.edu2009-03-235-13/+14
| | | | making calls
* Small step toward call-conv improvement: separate out calls and returnsdias@eecs.tufts.edu2009-03-235-6/+6
|
* Removed a tracedias@eecs.tufts.edu2009-03-181-1/+1
|
* Calling convention bug and cleanupdias@eecs.tufts.edu2009-03-174-5/+3
| | | | | - yet another wrong calling convention; this one was a special case for returning one value.
* Inconsistent type and arguments in safe foreign calls...dias@eecs.tufts.edu2009-03-161-15/+15
| | | | | - The function argument was stripped from the argument list but not from the type. Now they're both stripped.
* Fixes to "Retract Hp *before* checking for HpLim==0"Simon Marlow2009-03-181-2/+7
|
* FIX biographical profiling (#3039, probably #2297)Simon Marlow2009-03-173-4/+21
| | | | | | | | | Since we introduced pointer tagging, we no longer always enter a closure to evaluate it. However, the biographical profiler relies on closures being entered in order to mark them as "used", so we were getting spurious amounts of data attributed to VOID. It turns out there are various places that need to be fixed, and I think at least one of them was also wrong before pointer tagging (CgCon.cgReturnDataCon).
* A few bug fixes; some improvements spurred by paper writingdias@eecs.harvard.edu2009-03-036-35/+47
| | | | | | | | | | | | Among others: - Fixed Stg->C-- translation of let-no-escapes -- it's important to use the right continuation... - Fixed infinite recursion in X86 backend (shortcutJump mishandled infinite loops) - Fixed yet another wrong calling convention -- primops take args only in vanilla regs, but they may return results on the stack! - Removed StackInfo from LGraph and Block -- now in LastCall and CmmZ - Updated avail-variable and liveness code
* Fix the build on amd64/LinuxIan Lynagh2009-02-063-2/+2
|
* When generating C, don't pretend functions are dataIan Lynagh2009-02-066-6/+10
| | | | | | | | | | | | | | | We used to generated things like: extern StgWordArray (newCAF) __attribute__((aligned (8))); ((void (*)(void *))(W_)&newCAF)((void *)R1.w); (which is to say, pretend that newCAF is some data, then cast it to a function and call it). This goes wrong on at least IA64, where: A function pointer on the ia64 does not point to the first byte of code. Intsead, it points to a structure that describes the function. The first quadword in the structure is the address of the first byte of code so we end up dereferencing function pointers one time too many, and segfaulting.
* Remove a redundant importIan Lynagh2009-01-141-1/+0
|
* Make the ASSERT more informativesimonpj@microsoft.com2009-01-131-1/+1
|
* Fix warnings in StgCmmForeignIan Lynagh2008-12-291-10/+11
|
* Fix warnings in CgCallConvIan Lynagh2008-12-291-10/+11
|
* Fix warnings in SMRepIan Lynagh2008-12-291-17/+15
|
* Fix warnings in ClosureInfoIan Lynagh2008-12-291-34/+43
|
* Fix warnings in CgTickyIan Lynagh2008-12-291-15/+18
|
* Fix warnings in CgConIan Lynagh2008-12-291-11/+4
|
* Use DynFlags to work out if we are doing ticky ticky profilingIan Lynagh2008-12-188-44/+55
| | | | We used to use StaticFlags
* Fix warnings in CgExprIan Lynagh2008-12-171-16/+24
|
* Fix warnings in CgBinderyIan Lynagh2008-12-171-13/+28
|
* Fix warnings in CgStackeryIan Lynagh2008-12-171-9/+1
|
* Fix warnings in CgCaseIan Lynagh2008-12-171-24/+24
|
* Remove some dead code from CgCaseIan Lynagh2008-12-171-1/+0
|
* Fix warnings in StgCmmProfIan Lynagh2008-12-171-13/+21
|
* Fix warnings in CgProfIan Lynagh2008-12-171-15/+21
|
* Fix warnings in CgInfoTblsIan Lynagh2008-12-171-13/+3
|
* Remove dead function srtLabelAndLength from CgInfoTblsIan Lynagh2008-12-171-12/+0
|
* Fix warnings in CgHeaperyIan Lynagh2008-12-171-11/+7
|
* Fix warnings in CgTailCallIan Lynagh2008-12-171-13/+7
|
* Remove a little dead code from CgTailCallIan Lynagh2008-12-171-1/+0
|
* Fix warnings in CodeGenIan Lynagh2008-12-171-11/+2
|
* Fix warnings in StgCmmTickyIan Lynagh2008-12-171-11/+20
|
* Remove dead code from CgUtilsIan Lynagh2008-12-171-1/+0
|
View Lorry Controller Status