summaryrefslogtreecommitdiff
path: root/compiler/cmm
Commit message (Expand)AuthorAgeFilesLines
* Require a bang pattern when unlifted types are where/let bound; #3182Ian Lynagh2009-04-242-2/+2
* eliminate warningsdias@eecs.tufts.edu2009-04-032-2/+1
* Buggy optimizations caused function-call return to share the function's entry...dias@eecs.tufts.edu2009-03-311-25/+50
* Better handling of node parameter in calling conventionsdias@eecs.tufts.edu2009-03-253-9/+13
* Code simplifications due to call/return separation; some improvements to how ...dias@eecs.tufts.edu2009-03-233-46/+13
* Code simplification due to separate call/return conventionsdias@eecs.tufts.edu2009-03-233-30/+41
* Calls with and without passing node arguments more clearly separateddias@eecs.tufts.edu2009-03-234-16/+30
* Another small step: call and return conventions specified separately when mak...dias@eecs.tufts.edu2009-03-234-13/+13
* Small step toward call-conv improvement: separate out calls and returnsdias@eecs.tufts.edu2009-03-234-21/+29
* use a UniqSet for is MathFun, this list is getting quite largeSimon Marlow2009-03-301-4/+5
* update list of C math functionsBertram Felgenhauer2009-03-231-17/+77
* Comment explaining use of seq in DFMonaddias@eecs.tufts.edu2009-03-181-0/+4
* Calling convention bug and cleanupdias@eecs.tufts.edu2009-03-172-39/+60
* stack overflows and out of memory'sdias@eecs.tufts.edu2009-03-167-82/+84
* A few bug fixes; some improvements spurred by paper writingdias@eecs.harvard.edu2009-03-0324-646/+602
* Correct an IsFunction that should be IsDataIan Lynagh2009-02-071-1/+1
* Fix calling maths functions when compiling via CIan Lynagh2009-02-061-1/+2
* When generating C, don't pretend functions are dataIan Lynagh2009-02-065-34/+47
* Fix warnings in CmmInfoIan Lynagh2008-12-291-11/+3
* Fix warnings in CmmCPSGenIan Lynagh2008-12-291-13/+23
* Fix warnings in CmmProcPointIan Lynagh2008-12-291-10/+1
* Fix warnings in CmmCallConvIan Lynagh2008-12-291-13/+18
* Fix warnings in CmmLiveIan Lynagh2008-12-291-8/+1
* Fix warnings in CmmCPSIan Lynagh2008-12-291-14/+8
* Fix warnings in CmmUtilsIan Lynagh2008-12-291-11/+5
* Comment out dead function breakProcIan Lynagh2008-12-291-0/+4
* Fix warnings in CmmBrokenBlockIan Lynagh2008-12-291-11/+5
* Fix #2838: we should narrow a CmmInt before converting to ImmIntegerSimon Marlow2008-12-092-17/+19
* Remove INLINE pragmas on recursive functionssimonpj@microsoft.com2008-12-051-5/+0
* add missing case to Ord GlobalReg (EagerBlackhole == EagerBlackhole)Simon Marlow2008-11-281-0/+1
* don't emit CmmComments for nowSimon Marlow2008-11-271-1/+5
* drop some debugging traces and use only one flag for new codegendias@eecs.harvard.edu2008-11-261-2/+2
* one more missing patch from new codegen pathdias@eecs.harvard.edu2008-11-261-0/+2
* Removed warnings, made Haddock happy, added examples in documentationdias@eecs.harvard.edu2008-10-1712-56/+79
* Fixed linear regalloc bug, dropped some tracing codedias@eecs.harvard.edu2008-10-166-32/+31
* Keep update frames live even in functions that never returndias@eecs.harvard.edu2008-10-142-3/+12
* Removed space and time inefficiency in procpoint splittingdias@eecs.harvard.edu2008-10-142-5/+21
* Clarify the SRT building processdias@eecs.harvard.edu2008-10-142-23/+55
* Floating infotables were reversed in C back enddias@eecs.harvard.edu2008-10-132-4/+6
* forgot a few filesdias@eecs.harvard.edu2008-10-132-0/+990
* Big collection of patches for the new codegen branch.dias@eecs.harvard.edu2008-10-1333-1219/+1047
* Merging in the new codegen branchdias@eecs.harvard.edu2008-08-1437-2368/+3248
* Add optional eager black-holing, with new flag -feager-blackholingSimon Marlow2008-11-184-5/+3
* Return parser errors and warnings instead of dying.Thomas Schilling2008-09-141-8/+12
* haddock 2 doc fixesIan Lynagh2008-08-281-1/+1
* Document CmmZipUtilMax Bolingbroke2008-07-311-1/+1
* Document ZipCfgMax Bolingbroke2008-07-311-1/+1
* Document MachOpMax Bolingbroke2008-07-311-5/+5
* Document DataflowMax Bolingbroke2008-07-311-1/+1
* Document DFMonadMax Bolingbroke2008-07-311-1/+1
View Lorry Controller Status