(This replaces #121740. Sorry for wasting your time.)
This is a follow-up to a previous commit (ee7ca0d) which eliminated
several "TODO: make CombinerHelper methods const" remarks. As promised
in that ealier commit, this change completes the set by also making the
methods of AMDGPUCombinerHelper const so that the Helper member of
AMDGPUPreLegalizerCombinerImpl can be const rather than explicitly
mutable.
This combine pattern perform the below transformation.
fmul x, select(y, A, B) -> fldexp (x, select i32 (y, a, b))
fmul x, select(y, -A, -B) -> fldexp ((fneg x), select i32 (y, a, b))
where, A=2^a & B=2^b ; a and b are integers.
It is a follow-up PR to implement the above combine for globalIsel, as
the corresponding DAG combine has been done for SelectionDAG Isel
(#111109)
Add missing include guards to LLVM header files that did not previously
have them and update existing include guards to ensure that they enclose
all non-whitespace, non-comment text to enable these headers for the
multiple-include optimization.
Differential Revision: https://reviews.llvm.org/D150511
If we have legal f16 instructions but no f16 med3, we can save
one instruction by expanding out the min/max sequence compared
to casting to f32 and casting back.
If possible fold fneg into instruction above if users cannot fold mods and we
know it will decrease instruction count.
Follows same logic as SDAG combiner in choosing opportunities to combine.
Differential Revision: https://reviews.llvm.org/D112827
