path: root/lib/Target
AgeCommit message (Expand)AuthorFilesLines
2005-04-20Remove trailing whitespace, patch by Markus Oberhumer.Misha Brukman3-115/+115
2005-04-19Add completely untested support for mtcrf/mfcrf encodingChris Lattner1-0/+7
2005-04-19switch over the rest of the formats that use RC to use isDOTChris Lattner2-71/+73
2005-04-19Convert the XForm instrs and XSForm instruction over to use isDOTChris Lattner2-52/+60
2005-04-19Now that the ppc64 and vmx operands of I are always 0, forward substituteChris Lattner1-36/+30
2005-04-19convert over bform and iform instructionsChris Lattner2-17/+17
2005-04-19Convert over DForm and DSForm instructionsChris Lattner2-90/+80
2005-04-19Convert XLForm and XForm instructions over to use PPC64 when appropriate.Chris Lattner2-129/+112
2005-04-19Convert XO XS and XFX forms to use isPPC64Chris Lattner2-53/+52
2005-04-19Turn PPC64 and VMX into classes that can be added to instructions instead ofChris Lattner2-39/+40
2005-04-18Next round of PPC CR optimizations. For the following code:Nate Begeman1-62/+95
2005-04-18Change codegen for setcc to read the bit directly out of the conditionNate Begeman3-45/+36
2005-04-18Handle ExternalSymbol operands in the PPC JITChris Lattner1-9/+13
2005-04-15Make pattern isel default for ppcNate Begeman5-19/+36
2005-04-14fix callsAndrew Lenharth2-1/+2
2005-04-14a 21264 fix, and fix the operator precidence on an and -> zap check (should f...Andrew Lenharth2-3/+11
2005-04-14print negative 64 bit immediates as negative numbers, makes things a littleDuraid Madina2-2/+9
2005-04-14oops, this stopped us turning movl r4=0xFFFFFFFF;; and rX, r4 into zxt4Duraid Madina1-1/+1
2005-04-14Implement multi-way branches through logical ops on condition registers.Nate Begeman2-3/+69
2005-04-14we have zextloads, not sextloads!Duraid Madina1-1/+1
2005-04-14Add the necessary support to codegen condition register logical ops withNate Begeman4-17/+72
2005-04-13Start allocating condition registers. Almost all explicit uses of CR0 areNate Begeman1-23/+26
2005-04-13Implement the fold shift X, zext(Y) -> shift X, Y at the target level,Nate Begeman1-6/+22
2005-04-13Disbale the broken fold of shift + sz[ext] for nowNate Begeman1-19/+0
2005-04-13WOW, function calls still seem to work after this.Andrew Lenharth3-19/+30
2005-04-13prepare for func call optimizationAndrew Lenharth1-1/+1
2005-04-13* add the shladd instructionDuraid Madina2-0/+23
2005-04-13add matches for SxADDL and company, as well as simplify the SxADDQ codeAndrew Lenharth1-78/+48
2005-04-13* if ANDing with a constant of the form:Duraid Madina1-2/+45
2005-04-13added all flavors of zap for andingAndrew Lenharth1-0/+20
2005-04-13Fix some mysteriously missing {}'s which cause the miscompilation ofChris Lattner1-2/+2
2005-04-13remove one more occurance of this that snuck inChris Lattner1-1/+1
2005-04-13Remove support for ZERO_EXTEND_INREG. This pessimizes code, genering stuffChris Lattner1-21/+0
2005-04-13Remove special handling of ZERO_EXTEND_INREG. This pessimizes code, causingChris Lattner1-17/+0
2005-04-13Elimate handling of ZERO_EXTEND_INREG. This causes the PPC backend to emitChris Lattner2-25/+0
2005-04-13Z_E_I is goneChris Lattner1-2/+0
2005-04-12Fold shift by size larger than type size to undefNate Begeman1-1/+0
2005-04-12Implement setcc op, -1 sequencesNate Begeman1-22/+41
2005-04-12* OK, after changing to use liveIn/liveOut instead of IDEFs,Duraid Madina2-5/+18
2005-04-12Get rid of idefs for arguments (oops)Andrew Lenharth1-5/+5
2005-04-12Get rid of idefs for argumentsAndrew Lenharth1-5/+28
2005-04-12Put out* into the allocation order, allowing the register allocator toChris Lattner1-4/+3
2005-04-12Make sure to realize that calls use their argument regsChris Lattner1-0/+1
2005-04-12stop emitting IDEFs for args - change to using liveIn/liveOutDuraid Madina2-4/+31
2005-04-12Initial support for allocation condition registersNate Begeman5-13/+67
2005-04-12Implement bitfield clearsNate Begeman2-11/+35
2005-04-11Update PPC readme. Remove things that are done or aren't ppc specificNate Begeman1-6/+2
2005-04-11IA64 supports this operation.Chris Lattner1-1/+0
2005-04-11ORo sets CR0Chris Lattner1-0/+1
2005-04-11Revert the previous patch, which I didn't mean to check in.Chris Lattner2-37/+26