diff options
| author | Jakob Stoklund Olesen <stoklund@2pi.dk> | 2013-07-01 23:19:39 +0000 |
|---|---|---|
| committer | Jakob Stoklund Olesen <stoklund@2pi.dk> | 2013-07-01 23:19:39 +0000 |
| commit | 55daed27b7aec4a77f64e372b7f0d993ab4caaba (patch) | |
| tree | ceed488413ae69724aff0d6ff0eb9d8681c47939 /llvm/lib/Support/APFloat.cpp | |
| parent | fab01e5cea36143eb7d522b05fc50292179399b9 (diff) | |
| download | bcm5719-llvm-55daed27b7aec4a77f64e372b7f0d993ab4caaba.tar.gz bcm5719-llvm-55daed27b7aec4a77f64e372b7f0d993ab4caaba.zip | |
Remove floating point computations form SpillPlacement.cpp.
Patch by Benjamin Kramer!
Use the BlockFrequency class instead of floats in the Hopfield network
computations. This rescales the node Bias field from a [-2;2] float
range to two block frequencies BiasN and BiasP pulling in opposite
directions. This construct has a more predictable behavior when block
frequencies saturate.
The per-node scaling factors are no longer necessary, assuming the block
frequencies around a bundle are consistent.
This patch can cause the register allocator to make different spilling
decisions. The differences should be small.
llvm-svn: 185393
Diffstat (limited to 'llvm/lib/Support/APFloat.cpp')
0 files changed, 0 insertions, 0 deletions
