emscripten-fastcomp - LLVM with the emscripten fastcomp javascript backend

Age	Commit message (Expand)	Author
2007-12-06	Fix for PR1831: if all defs of an interval are re-materializable, then it's a...	Evan Cheng
2007-12-05	MachineInstr can change. Store indexes instead.	Evan Cheng
2007-12-05	If a split live interval is spilled again, remove the kill marker on its last...	Evan Cheng
2007-12-05	Clobber more bugs.	Evan Cheng
2007-12-05	Fix kill info for split intervals.	Evan Cheng
2007-12-05	- Mark last use of a split interval as kill instead of letting spiller track it.	Evan Cheng
2007-12-04	Discard split intervals made empty due to folding.	Evan Cheng
2007-12-03	Typo	Evan Cheng
2007-12-03	Update kill info for uses of split intervals.	Evan Cheng
2007-12-02	Remove redundant foldMemoryOperand variants and other code clean up.	Evan Cheng
2007-12-01	Fix a bug where splitting cause some unnecessary spilling.	Evan Cheng
2007-12-01	Allow some reloads to be folded in multi-use cases. Specifically testl r, r -...	Evan Cheng
2007-11-30	Do not fold reload into an instruction with multiple uses. It issues one extr...	Evan Cheng
2007-11-29	Do not lose rematerialization info when spilling already split live intervals.	Evan Cheng
2007-11-29	Fix a major performance issue with splitting. If there is a def (not def/use)	Evan Cheng
2007-11-29	Replace the odd kill# hack with something less fragile.	Evan Cheng
2007-11-29	Fixed various live interval splitting bugs / compile time issues.	Evan Cheng
2007-11-28	Recover compile time regression.	Evan Cheng
2007-11-17	Live interval splitting:	Evan Cheng
2007-11-15	Fix a thinko in post-allocation coalescer.	Evan Cheng
2007-11-14	Clean up sub-register implementation by moving subReg information back to	Evan Cheng
2007-11-12	Refactor some code.	Evan Cheng
2007-11-07	Simplify my (il)logic.	Evan Cheng
2007-11-06	When the allocator rewrite a spill register with new virtual register, it rep...	Evan Cheng
2007-11-06	Fix a bug where a def use operand isn't being detected as a sub-register use.	Evan Cheng
2007-11-05	Fix PR1187.	Evan Cheng
2007-11-03	There are times when the coalescer would not coalesce away a copy but the copy	Evan Cheng
2007-10-17	Apply Chris' suggestions.	Evan Cheng
2007-10-17	Clean up code that calculate MBB live-in's.	Evan Cheng
2007-10-12	Did mean to leave this in. INSERT_SUBREG isn't being coalesced yet.	Evan Cheng
2007-10-12	EXTRACT_SUBREG coalescing support. The coalescer now treats EXTRACT_SUBREG like	Evan Cheng
2007-10-08	Kill cycle of an live range is always the last use index + 1.	Evan Cheng
2007-10-03	Use empty() member functions when that's what's being tested for instead	Dan Gohman
2007-09-14	Remove isReg, isImm, and isMBB, and change all their users to use	Dan Gohman
2007-09-06	Fix a memory leak.	Evan Cheng
2007-09-05	Use pool allocator for all the VNInfo's to improve memory access locality. Th...	Evan Cheng
2007-08-30	Try fold re-materialized load instructions into its uses.	Evan Cheng
2007-08-29	Change LiveRange so it keeps a pointer to the VNInfo rather than an index.	Evan Cheng
2007-08-16	Fix some kill info update bugs; add hidden option -disable-rematerialization ...	Evan Cheng
2007-08-13	Re-implement trivial rematerialization. This allows def MIs whose live interv...	Evan Cheng
2007-08-11	Code to maintain kill information during register coalescing.	Evan Cheng
2007-08-08	Adding kill info to val#.	Evan Cheng
2007-08-08	- Each val# can have multiple kills.	Evan Cheng
2007-08-07	- LiveInterval value#'s now have 3 components: def instruction #,	Evan Cheng
2007-06-27	If a livein is not used in the block. It's live through.	Evan Cheng
2007-06-27	Fix an obvious bug. Old code only worked for the entry block.	Evan Cheng
2007-06-19	Replace M_REMATERIALIZIBLE and the newly-added isOtherReMaterializableLoad	Dan Gohman
2007-06-14	Add a target hook to allow loads from constant pools to be rematerialized, an...	Dan Gohman
2007-06-08	Factor live variable analysis so it does not do register coalescing	David Greene
2007-05-14	Only worry about intervening kill if there are more than one live ranges in t...	Evan Cheng