search for: oldfmareg

...p1 = fmul float %tmp, 0x401F25E360000000 %tmp2 = fadd float %tmp1, 0x3FC1A7B960000000 %tmp3 = select i1 undef, float 0x401F25E360000000, float %tmp2 store float %tmp3, float* undef br label %loop_bb } The code above is triggering a assertion failure when adjusting the live intervals, since OldFMAReg (%vreg12 in debug info) is actually defined in two blocks. I wonder if we can fix this by making the transformation simpler, that is, instead of doing: %vreg12<def> = COPY %vreg7; VSSRC:%vreg12,%vreg7 %vreg12<def,tied1> = XSMADDASP %vreg12<tied0>, %vreg0, %vreg4; VSSRC:%v...

On Tue, Mar 22, 2016 at 5:13 PM Eric Christopher wrote: > I think we can probably go ahead and throw this up on Phabricator for > review. I'd probably bring in Matthias for review as well. > > Thanks! > > -eric > (Follow up on the discuss in IRC) I tried to bridge PPC backend and commuteInstruction, not sure if correctly, but here I got some non-optimal results: in 12

On Fri, Feb 19, 2016 at 5:10 PM Tim Shen <timshen at google.com> wrote: > I wonder if we can fix this by making the transformation simpler, that is, > instead of doing: > I wrote a prototype (see attach) for this idea, it actually improves some of the test cases (e.g. fma-assoc.ll: test_FMADD_ASSOC1), but pessimize several other cases (e.g. test_FMADD_ASSOC_EXT1). I'm not

Ping? On Mon, Feb 22, 2016 at 1:06 PM Tim Shen <timshen at google.com> wrote: > On Fri, Feb 19, 2016 at 5:10 PM Tim Shen <timshen at google.com> wrote: > >> I wonder if we can fix this by making the transformation simpler, that >> is, instead of doing: >> > > I wrote a prototype (see attach) for this idea, it actually improves some > of the test cases

I think we can probably go ahead and throw this up on Phabricator for review. I'd probably bring in Matthias for review as well. Thanks! -eric On Wed, Mar 16, 2016 at 10:53 AM Tim Shen <timshen at google.com> wrote: > I implemented a proof of concept of a new generic MachineFunction SSA > pass. The code is not readable and not efficient yet, but it shows > interesting

I implemented a proof of concept of a new generic MachineFunction SSA pass. The code is not readable and not efficient yet, but it shows interesting results: In fma.ll @test_FMSUB2 (return dummy(A * B + C, A * B - D)): before: fmr 0, 1 xsmaddadp 3, 0, 2 xsmsubmdp 0, 2, 4 fmr 1, 3 fmr 2, 0 bl dummy2 after: xsmsubadp 4, 1, 2 xsmaddmdp

I wonder if we can do this in a separate analysis MachineFunction SSA pass. 1) SelectionDAG will generate a pseudo instruction MutatingFMA. When it's generated it's allowed to have d = a * b + c form, where d doesn't have to be in {a, b, c}. 2) Later, the proposed pass uses an algorithm to decide for instruction MI: `%vreg0 = MutatingFMA %vreg1, %vreg2, %vreg3`, it should tie %vreg0