Programming Languages Research Group: Git

2010-04-08	Evan Cheng	Coalescer should not delete copy instructions whose...	commit \| commitdiff \| tree
2010-04-08	Evan Cheng	Avoid using f64 to lower memcpy from constant string...	commit \| commitdiff \| tree
2010-04-08	Evan Cheng	Make post regalloc machine licm functional. It now...	commit \| commitdiff \| tree
2010-04-07	Evan Cheng	Fix typo.	commit \| commitdiff \| tree
2010-04-07	Evan Cheng	Add comments for missed opportunities.	commit \| commitdiff \| tree
2010-04-07	Evan Cheng	Fix typo.	commit \| commitdiff \| tree
2010-04-07	Evan Cheng	Post regalloc LICM. Work in progress.	commit \| commitdiff \| tree
2010-04-06	Evan Cheng	Code clean up. Move includes from VirtRegRewriter.h...	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Fix an obvious copy-n-paste bug. It's not known to...	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Add nounwind.	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Fix ADD32rr_alt instruction encoding bug. Patch by...	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Code clean up.	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Temporarily remove to disable building of ARM disassembler.	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Re-apply 100265 but instead disable building of ARM...	commit \| commitdiff \| tree
2010-04-05	Evan Cheng	Reverting 100265 to try to get buildbots green again...	commit \| commitdiff \| tree
2010-04-03	Evan Cheng	Code refactoring.	commit \| commitdiff \| tree
2010-04-02	Evan Cheng	Correctly lower memset / memcpy of undef. It should...	commit \| commitdiff \| tree
2010-04-02	Evan Cheng	Revert 100204. It broke a bunch of tests and apparently...	commit \| commitdiff \| tree
2010-04-02	Evan Cheng	After trivial coalescing, the MI being visited may...	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	In 64-bit mode, use i64 to lower memcpy / memset instead...	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	Skip checking preferred alignment of GVs defined in...	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	Add comments about DstAlign and SrcAlign.	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	- Avoid using floating point stores to implement memset...	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	Add -mcpu to memcpy / memset tests to ensure they behave...	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	Fix sdisel memcpy, memset, memmove lowering:	commit \| commitdiff \| tree
2010-04-01	Evan Cheng	Nehalem unaligned memory access is fast.	commit \| commitdiff \| tree
2010-03-30	Evan Cheng	Funky indentation.	commit \| commitdiff \| tree
2010-03-30	Evan Cheng	Avoid being influenced by the presence of dbg_value...	commit \| commitdiff \| tree
2010-03-29	Evan Cheng	Fix PR4975. Avoid referencing empty vector.	commit \| commitdiff \| tree
2010-03-29	Evan Cheng	Pool allocate SDDbgValue nodes.	commit \| commitdiff \| tree
2010-03-26	Evan Cheng	Do not sibcall if stack needs to be dynamically aligned.	commit \| commitdiff \| tree
2010-03-26	Evan Cheng	Allow trivial sibcall of vararg callee when no arguments...	commit \| commitdiff \| tree
2010-03-26	Evan Cheng	LiveVariables should clear kill / dead markers first...	commit \| commitdiff \| tree
2010-03-26	Evan Cheng	Try trivial remat before the coalescer gives up on...	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Add nounwind.	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Code clean up.	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Remove an unused option.	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Scheduler assumes SDDbgValue nodes are in source order...	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Remove a fixme that doesn't make sense any more.	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Make sure SDDbgValue.Invalid is initialized to false...	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Change how dbg_value sdnodes are converted into machine...	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Avoid being influenced by dbg_value instructions.	commit \| commitdiff \| tree
2010-03-25	Evan Cheng	Disable folding loads into tail call in 32-bit PIC...	commit \| commitdiff \| tree
2010-03-24	Evan Cheng	Move OptChkCall off LibCallOptimization into StrCpyOpt.	commit \| commitdiff \| tree
2010-03-24	Evan Cheng	dbg_value may end a block.	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Ignore dbg_value's.	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Teach isSafeToClobberEFLAGS to ignore dbg_value's....	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Code clean up.	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Add comment.	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Forgot this.	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Teach simplify libcall to transform __strcpy_chk to...	commit \| commitdiff \| tree
2010-03-23	Evan Cheng	Fix an incorrect logic causing instcombine to miss...	commit \| commitdiff \| tree
2010-03-22	Evan Cheng	Fix 80 col violation.	commit \| commitdiff \| tree
2010-03-22	Evan Cheng	Reduce indentation.	commit \| commitdiff \| tree
2010-03-22	Evan Cheng	80 col violation.	commit \| commitdiff \| tree
2010-03-20	Evan Cheng	Stupid svn. Add back to the lost sibcall tests.	commit \| commitdiff \| tree
2010-03-20	Evan Cheng	If call result is in ST0 and it is not being passed...	commit \| commitdiff \| tree
2010-03-18	Evan Cheng	Turning off post-ra scheduling for x86. It isn't a...	commit \| commitdiff \| tree
2010-03-17	Evan Cheng	X86 address mode matching code MatchAddressRecursively...	commit \| commitdiff \| tree
2010-03-16	Evan Cheng	Fix liveintervals handling of dbg_value instructions.	commit \| commitdiff \| tree
2010-03-15	Evan Cheng	Avoid sibcall optimization if either caller or callee...	commit \| commitdiff \| tree
2010-03-14	Evan Cheng	Rename SDDbgValue.h to SDNodeDbgValue.h for consistency.	commit \| commitdiff \| tree
2010-03-14	Evan Cheng	Fix jit encoding bugs.	commit \| commitdiff \| tree
2010-03-14	Evan Cheng	Check in tablegen changes to fix disassembler related...	commit \| commitdiff \| tree
2010-03-14	Evan Cheng	Do not force indirect tailcall through fixed registers...	commit \| commitdiff \| tree
2010-03-13	Evan Cheng	Fix a typo in ValueTracking that's causing instcombine...	commit \| commitdiff \| tree
2010-03-11	Evan Cheng	In case of tail call size of Ins and InVals may not...	commit \| commitdiff \| tree
2010-03-11	Evan Cheng	Bad bad bug. x86 force indirect tail call address into...	commit \| commitdiff \| tree
2010-03-11	Evan Cheng	The check for coalescing a virtual register to a physical...	commit \| commitdiff \| tree
2010-03-10	Evan Cheng	Fix typo.	commit \| commitdiff \| tree
2010-03-10	Evan Cheng	Unbreak test on Linux.	commit \| commitdiff \| tree
2010-03-10	Evan Cheng	Enable machine cse pass.	commit \| commitdiff \| tree
2010-03-10	Evan Cheng	Add a couple more heuristics to neuter machine cse...	commit \| commitdiff \| tree
2010-03-09	Evan Cheng	Allow more cross-rc coalescing.	commit \| commitdiff \| tree
2010-03-09	Evan Cheng	Revert accidental commit.	commit \| commitdiff \| tree
2010-03-09	Evan Cheng	- Make the machine cse dumb coalescer (as opposed to...	commit \| commitdiff \| tree
2010-03-08	Evan Cheng	Don't waste time trying to CSE labels, phis, inline...	commit \| commitdiff \| tree
2010-03-08	Evan Cheng	Restrict machine cse to really trivial coalescing....	commit \| commitdiff \| tree
2010-03-08	Evan Cheng	Re-commit 97860 with fix. getMallocAllocatedType may...	commit \| commitdiff \| tree
2010-03-08	Evan Cheng	Add documentation on sibling call optimization. Rename...	commit \| commitdiff \| tree
2010-03-06	Evan Cheng	Don't update physical register def.	commit \| commitdiff \| tree
2010-03-06	Evan Cheng	Transform @llvm.objectsize to integer if the argument...	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Safely turn memset_chk etc. to non-chk variant if the...	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Instcombine should turn llvm.objectsize of a alloca...	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Fix typo.	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Fix an oops in x86 sibcall optimization. If the ByVal...	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Rever 96389 and 96990. They are causing some miscompilation...	commit \| commitdiff \| tree
2010-03-05	Evan Cheng	Add missing break for Intrinsic::objectsize case. It...	commit \| commitdiff \| tree
2010-03-04	Evan Cheng	Run machine licm before machine cse to avoid messing...	commit \| commitdiff \| tree
2010-03-04	Evan Cheng	Avoid cse load instructions unless they are known to...	commit \| commitdiff \| tree
2010-03-04	Evan Cheng	Rename -machine-cse to -enable-machine-cse.	commit \| commitdiff \| tree
2010-03-04	Evan Cheng	Look ahead a bit to determine if a physical register...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Fix a logic error. An instruction that has a live physical...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Remove PHINodeTraits and use MachineInstrExpressionTrait...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Move MachineInstrExpressionTrait::getHashValue() out...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Re-apply r97667 but with a little bit of thought put...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Fix funky indentation and add comments.	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Move DenseMapInfo for MachineInstr* to MachineInstr.h	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Machine CSE work in progress. It's doing some CSE now...	commit \| commitdiff \| tree
2010-03-03	Evan Cheng	Add MachineRegisterInfo::hasOneUse and hasOneNonDBGUse.	commit \| commitdiff \| tree
next

C/C++ LLVM-based compilers that forbids OOTA behaviors

RSS Atom