- 63602fb Do not CopyFromReg physregs for live-in values. Instead, create a vreg for by Chris Lattner · 20 years ago
- 16cd04d rename the ADJCALLSTACKDOWN/ADJCALLSTACKUP nodes to be CALLSEQ_START/BEGIN. by Chris Lattner · 20 years ago
- c6f4181 Add a new -enable-x86-fastcc option that enables passing the first by Chris Lattner · 20 years ago
- c57f682 Pass in Calling Convention to use into LowerCallTo by Chris Lattner · 20 years ago
- c610d42 X86 has more than just 32-bit registers by Chris Lattner · 20 years ago
- 6415bb4 Convert feature of the simple isel over for the pattern isel to use. by Chris Lattner · 20 years ago
- 19bb228 Silence some VC++ warnings by Jeff Cohen · 20 years ago
- 966cdfb Implement READPORT/WRITEPORT, implementing the last X86 regression tests by Chris Lattner · 20 years ago
- 4e6ce5f legalize readio/writeio into load/stores, fixing CodeGen/X86/io.llx with by Chris Lattner · 20 years ago
- a0dbf18 restore some non-dead code I removed last night breaking double casts to uint by Chris Lattner · 20 years ago
- a80d2bd Wrap long lines, remove dead code that is now handled by legalize by Chris Lattner · 20 years ago
- 2afa191 Fix FP -> bool casts by Chris Lattner · 20 years ago
- e3e0f27 Fix X86/2005-05-08-FPStackifierPHI.ll: ugly gross hack. by Chris Lattner · 20 years ago
- b5884d3 fix typo by Andrew Lenharth · 20 years ago
- 691ef2b Implement count leading zeros (ctlz), count trailing zeros (cttz), and count by Andrew Lenharth · 20 years ago
- c5dcb53 Add support for FSIN/FCOS when unsafe math ops are enabled. Patch contributed by by Chris Lattner · 20 years ago
- 2c56e8a Add support for FSQRT node, patch contributed by Morten Ofstad by Chris Lattner · 20 years ago
- 2d86ea2 Implement Value* tracking for loads and stores in the selection DAG. This enables one to use alias analysis in the backends. by Andrew Lenharth · 20 years ago
- 0e0a7a45 * Remove trailing whitespace * Convert tabs to spaces by Misha Brukman · 20 years ago
- 75f354b Handle (store &GV -> mem) as a store immediate. This often occurs for by Chris Lattner · 20 years ago
- 5bf2686 Fix some mysteriously missing {}'s which cause the miscompilation of by Chris Lattner · 20 years ago
- 91302a1 Z_E_I is gone by Chris Lattner · 20 years ago
- 4c52f0e Use live out sets for return values instead of imp_defs, which is cleaner and faster. by Chris Lattner · 20 years ago
- da4d469 This target does not support/want ISD::BRCONDTWOWAY by Chris Lattner · 20 years ago
- 6659bd7 X86 zero extends setcc results by Chris Lattner · 20 years ago
- 8db0af1 Add (untested) support for MULHS and MULHU. by Chris Lattner · 20 years ago
- b7edaa1 add support for FABS and FNEG by Chris Lattner · 20 years ago
- 43fdea0 This target doesn't support fabs/fneg yet. by Chris Lattner · 20 years ago
- b38a749 Add support for 64-bit shifts. by Chris Lattner · 20 years ago
- 7ce7eff Add support for ISD::UNDEF to the X86 be by Chris Lattner · 20 years ago
- 71df3f8 don't depend on the cfg being set up yet by Chris Lattner · 20 years ago
- 8e21e71 Change interface to LowerCallTo to take a boolean isVarArg argument. by Nate Begeman · 20 years ago
- 85fdeb2 Remove comments that are now meaningless from the pattern ISels, at Chris's by Nate Begeman · 20 years ago
- 7805fa4 Don't emit two comparisons when comparing a FP value against zero! by Chris Lattner · 20 years ago
- e4d5c44 This mega patch converts us from using Function::a{iterator|begin|end} to by Chris Lattner · 20 years ago
- 572dd08 Silence some uninit variable warnings. by Chris Lattner · 20 years ago
- 3576c84 We can fold promoted and non-promoted loads into divs also! by Chris Lattner · 21 years ago
- 44129b5 Fold promoted loads into binary ops for FP, allowing us to generate m32 forms by Chris Lattner · 21 years ago
- 0a07883 Silence a warning. by Chris Lattner · 21 years ago
- fb0f53f Speed up folding operations into loads. by Chris Lattner · 21 years ago
- 67b1c3c The ever-important vanity pass name :) by Chris Lattner · 21 years ago
- b62e1e2 Fix a FIXME: realize that argument stores are all independent (don't alias) by Chris Lattner · 21 years ago
- 19ad062 Implement ADD_PARTS/SUB_PARTS so that 64-bit integer add/sub work. This by Chris Lattner · 21 years ago
- bf52d49 Fix a crash compiling 134.perl. by Chris Lattner · 21 years ago
- c3c021b Fix a problem where were were literally selecting for INCREASED register by Chris Lattner · 21 years ago
- 4053b1e Implement Regression/CodeGen/X86/rotate.ll: emit rotate instructions (which by Chris Lattner · 21 years ago
- 30ea1e9 Match 16-bit shld/shrd instructions as well, implementing shift-double.llx:test5 by Chris Lattner · 21 years ago
- 8571637 Codegen long >> 2 to this: by Chris Lattner · 21 years ago
- 009b55b X86 shifts mask the amount. by Chris Lattner · 21 years ago
- 477c931 Code to handle FP_EXTEND is dead now. X86 doesn't support any data types to by Chris Lattner · 21 years ago
- 3dd6815 Remove more dead code. by Chris Lattner · 21 years ago
- a7a4c0f The selection dag code handles the promotions from F32 to F64 for us, so we by Chris Lattner · 21 years ago
- b422aea Fix 124.m88ksim. by Chris Lattner · 21 years ago
- 811482a Do not emit loads multiple times, potentially in the wrong places. by Chris Lattner · 21 years ago
- b71f8fc Eliminate bad assertions. by Chris Lattner · 21 years ago
- 4a10866 * Eliminate the TokenSet and just use the ExprMap for both tokens and values. by Chris Lattner · 21 years ago
- 883c86f Rely on the code in MatchAddress to do this work. Otherwise we fail to by Chris Lattner · 21 years ago
- 98a8ba0 Fix a problem where probing for addressing modes caused expressions to be by Chris Lattner · 21 years ago
- 869e043 Don't bother using max here. by Chris Lattner · 21 years ago
- 90d1be7 Do not give token factor nodes outrageous weights by Chris Lattner · 21 years ago
- 5c65981 Two changes: by Chris Lattner · 21 years ago
- e10269b Refactor load/op/store folding into it's own method, no functionality changes. by Chris Lattner · 21 years ago
- ae3ffba Fix a major regression last night that prevented us from producing [mem] op= reg by Chris Lattner · 21 years ago
- 2a4e508 Codegen this: by Chris Lattner · 21 years ago
- 4ff348b Fix test/Regression/CodeGen/X86/2005-01-17-CycleInDAG.ll and 132.ijpeg. by Chris Lattner · 21 years ago
- cb1aa8d Do not fold a load into a comparison that is used by more than one place. by Chris Lattner · 21 years ago
- c98279d Do not codegen 'xor bool, true' as 'not reg'. not reg inverts the upper bits by Chris Lattner · 21 years ago
- 4df0de9 Set up the shift and setcc types. by Chris Lattner · 21 years ago
- da2ce11 * Adjust to changes in TargetLowering interfaces. by Chris Lattner · 21 years ago
- e9ef81d Add support for truncstore and *extload. by Chris Lattner · 21 years ago
- 7f2afac Adjust to CopyFromREg changes. by Chris Lattner · 21 years ago
- 18c2f13 Add new ImplicitDef node, rename CopyRegSDNode class to RegSDNode. by Chris Lattner · 21 years ago
- 1d50b7f Codegen factor nodes more intelligently according to perceived register pressure. by Chris Lattner · 21 years ago
- c358071 Initial trivial (but stupid) codegen for this node. by Chris Lattner · 21 years ago
- 636e79a Add some really pedantic assertions to the load folding code. Fix a bunch by Chris Lattner · 21 years ago
- a0bb692 We can only fold a load into an op if there is exactly one use of the value. by Chris Lattner · 21 years ago
- 9bbd992 Try both ways to fold an add together. This allows us to generate this code by Chris Lattner · 21 years ago
- 2b93786 Fix a major miscompilation where we were overwriting the scale reg. by Chris Lattner · 21 years ago
- 48034fd Do not use the type of the RHS constant to determine the type of the operation. by Chris Lattner · 21 years ago
- 603fea9 Fix C++ more compilatiom errors by Jeff Cohen · 21 years ago
- a56cea4 Fix a compile error with VC++, which things that static const arrays need by Chris Lattner · 21 years ago
- 7dbcb75 Fix a bug that caused us to crash on povray. We weren't emitting an FP_REG_KILL into a block that had a successor with a FP PHI node. by Chris Lattner · 21 years ago
- 085c995 Fix a crash compiling povray on UINT_TO_FP from i16. by Chris Lattner · 21 years ago
- 4292830 There are no [mem] op= reg instructions for FP, so remove their entries. by Chris Lattner · 21 years ago
- db8c368 Fix a bug where we didn't insert FP_REG_KILL instructions into MBB's that by Chris Lattner · 21 years ago
- afce430 Fold TRUNCATE (LOAD P) into a smaller load from P. by Chris Lattner · 21 years ago
- ef6806c Be more careful about order of arg evalution for CopyToReg nodes. This shrinks by Chris Lattner · 21 years ago
- 7ea64f5 Fold some more [mem] op= val operators. This allows us to things like this by Chris Lattner · 21 years ago
- dbba22f Fold loads into sign/zero extends. instead of: by Chris Lattner · 21 years ago
- 837caa7 Comment out debug code :) by Chris Lattner · 21 years ago
- c4b6a78 Treat int constants as not requiring a register, since they are almost always by Chris Lattner · 21 years ago
- a5ade06 * Factor a bunch of binary operator cases into shared code. by Chris Lattner · 21 years ago
- 947d544 Fold multiplies by 3,5,9 into addressing modes when possible. by Chris Lattner · 21 years ago
- 51a2634 Instead of generating stuff like this: by Chris Lattner · 21 years ago
- 31805bf Implement MEMCPY natively in terms of rep movs* by Chris Lattner · 21 years ago
- 989de03 Implement memset -> rep stos* by Chris Lattner · 21 years ago
- 795069d Announce that we don't support mem ops yet. by Chris Lattner · 21 years ago
- a95589b Teach the address selector to make 'reg+reg' addressing modes. by Chris Lattner · 21 years ago
- d4dab92 Emit NOT instructions. by Chris Lattner · 21 years ago
- 6c07aee Fix a bug emitting branches that broke a lot of programs. by Chris Lattner · 21 years ago