Commits · 794f15868a11fbc891ad854e706f30eb91453ad8 · Raul Torres / llvm-target-spread · GitLab

Oct 20, 2005
- forgot this one · 794f1586
  Andrew Lenharth authored Oct 20, 2005
  794f1586
- ret 0; works, not much else · 7b698670
  Andrew Lenharth authored Oct 20, 2005
  7b698670
Oct 19, 2005
- This fixes PR638: · 196e8c1f
  John Criswell authored Oct 19, 2005
  196e8c1f
- Added InstrSchedClass to each of the PowerPC Instructions. · 74ab9960
  Jim Laskey authored Oct 19, 2005
  74ab9960
- Write patterns for the various shl and srl patterns that don't involve · 9f3c26c4
  Nate Begeman authored Oct 19, 2005
  9f3c26c4
- Push processor descriptions to the top of target and add command line info. · 97611000
  Jim Laskey authored Oct 19, 2005
  97611000
- now that tblgen is smarter, use integers directly. This should help Andrew too · c16b0c38
  Chris Lattner authored Oct 19, 2005
  c16b0c38
- teach ppc backend these are copies · 5f376232
  Chris Lattner authored Oct 19, 2005
  5f376232
- Convert these cases to patterns · 5b6f4dc6
  Chris Lattner authored Oct 19, 2005
  5b6f4dc6
- Woo, it kinda works. We now generate this atrociously bad, but correct, · 9eaa6bac
  Nate Begeman authored Oct 19, 2005
  9eaa6bac
- apply some tblgen majik to simplify the X register definitions · ecdf8423
  Chris Lattner authored Oct 19, 2005
  ecdf8423
- Teach Legalize how to do something with EXTRACT_ELEMENT when the type of · 5172ce64
  Nate Begeman authored Oct 19, 2005
  5172ce64
- Make a new reg class for 64 bit regs that aliases the 32 bit regs. This · 92e77502
  Nate Begeman authored Oct 19, 2005
  92e77502
- Add the ability to lower return instructions to TargetLowering. This · 78afac2d
  Nate Begeman authored Oct 18, 2005
  78afac2d
- Fix Generic/2005-10-18-ZeroSizeStackObject.ll by not requesting a zero · 0a71a9ac
  Chris Lattner authored Oct 18, 2005
  0a71a9ac
- remove hack · 8396a308
  Chris Lattner authored Oct 18, 2005
  8396a308
Oct 18, 2005
- Simple edits; remove unimplimented cases and clarify long haul SLU cases. · d812a2e4
  Jim Laskey authored Oct 18, 2005
  d812a2e4
- Fix the JIT encoding of LWA, LD, STD, and STDU. · 5a2fb978
  Chris Lattner authored Oct 18, 2005
  5a2fb978
- Checking in first round of scheduling tablegen files. Not tied in as yet. · c6533006
  Jim Laskey authored Oct 18, 2005
  c6533006
- add a case · 53b9c3ad
  Chris Lattner authored Oct 18, 2005
  53b9c3ad
- Add an option to this pass. If it is set, we are allowed to internalize · 45517baf
  Chris Lattner authored Oct 18, 2005
  45517baf
- Fold (select C, load A, load B) -> load (select C, A, B). This happens quite · 6c14c35b
  Chris Lattner authored Oct 18, 2005
  6c14c35b
- Do the right thing and enable 64 bit regs under the control of a subtarget · e74dfbb9
  Nate Begeman authored Oct 18, 2005
  e74dfbb9
- First bits of 64 bit PowerPC stuff, currently disabled. A lot of this is · 0b71e007
  Nate Begeman authored Oct 18, 2005
  0b71e007
- Implement some feedback from Chris re: constant canonicalization · 418c6e40
  Nate Begeman authored Oct 18, 2005
  418c6e40
- Legalize BUILD_PAIR appropriately for upcoming 64 bit PowerPC work. · bd5f41a6
  Nate Begeman authored Oct 18, 2005
  bd5f41a6
Oct 17, 2005
- fold fmul X, +2.0 -> fadd X, X; · ec48a1bf
  Nate Begeman authored Oct 17, 2005
  ec48a1bf
- Make this work for FP constantexprs · da1b152c
  Chris Lattner authored Oct 17, 2005
  da1b152c
- Oops, X+0.0 isn't foldable, but X+-0.0 is. · 7fde91e3
  Chris Lattner authored Oct 17, 2005
  7fde91e3
- relax this a bit, as we only support the default rounding mode · 32979336
  Chris Lattner authored Oct 17, 2005
  32979336
- add a trivial fold · eeb2bda2
  Chris Lattner authored Oct 17, 2005
  eeb2bda2
Oct 16, 2005
- More PPC32 -> PPC changes, as well as merging some classes that were · 6cca84e4
  Nate Begeman authored Oct 16, 2005
  6cca84e4
- Fix this logic. · e540800d
  Chris Lattner authored Oct 15, 2005
  e540800d
- Add a case we were missing that was causing us to fail CodeGen/PowerPC/rlwinm.ll:test3 · 17cc9edd
  Chris Lattner authored Oct 15, 2005
  17cc9edd
- Remove some dead code now that the dag combiner exists. · c0896117
  Nate Begeman authored Oct 15, 2005
  c0896117
- Remove some dead code: the ORI/ORIS cases are autogen'd. This makes · d869bec4
  Chris Lattner authored Oct 15, 2005
  d869bec4
Oct 15, 2005
- prune #includes · 03354280
  Chris Lattner authored Oct 15, 2005
  03354280
- These instructions are now autogenerated · a52969c8
  Chris Lattner authored Oct 15, 2005
  a52969c8
- Add a pattern for FSQRTS · 286c1d7c
  Chris Lattner authored Oct 15, 2005
  286c1d7c
- remove dead code · efa38261
  Chris Lattner authored Oct 15, 2005
  efa38261