tree 9a946397974d40249a3d5e7e5cb52911dccba607
parent 245def9def7b025644a8cf991ba24c53a50822c6
author Alex Bradbury <asb@igalia.com> 1746703994 +0100
committer GitHub <noreply@github.com> 1746703994 +0100
gpgsig -----BEGIN PGP SIGNATURE-----
 
 wsFcBAABCAAQBQJoHJZ6CRC1aQ7uu5UhlAAAUOsQABnVfs3qEReSo74wFn4O14Ku
 DqEq+u4KJ5hwQlSv5KRGwpB0WRlpwAUp4cGx4dtYPYjFqSsZR6IzURBLsSpbmy6c
 /KVwDyco/JXz4yUSmxDCmOA9YZDRWLJpO8eX1tszxHr4FIBJA5Wpqq87f9rRqQYj
 8v5kmheh0gGzh5d7ODMlFnh+Emaf3q4oHIjPiXBo4QLfLHwZh8EINW7viDc+ZYqF
 jEnaSKTnKbhvWxSeh3Xck//H12fBGUZCM+BNhs54DGKHHJRJJy+dgWsMBq/eJjnn
 ky/TdvKurfGAp8bj7bgv1tty99LiezJzSHPgDMyB332XFw6mANvnVdKZAtmhOelG
 wm5p9HFxGdLpilWvZMptj5w1kWDlTH7piOa3DmgaiG06q29MPvojqlpURRkxE4O7
 FQl5SPCt+1KCqOMV+azWv4G28K+uVg8f/KrSzaUOESiaBL+4vkpVu5tOMXW8ub+5
 UpnuYK33lZc3KPpHzUF28drq2BVTKvpDM+DVtYj3sxrbJtJ/leeh+ufXz2FAeZeZ
 ln4DyZzvyQTfawFI7tXMxYEN6Lr7CY9jIryb+22VgxOfHdDnKpopZ29h3xe5D2iz
 38MPAyPV9+2DcFoRwXXr6L3nc1LSyccCqya9Jbpzbup7K2M1hZEzAmwI+xErsxH/
 RCqAAr+C6OpwhccxRQ1v
 =bXro
 -----END PGP SIGNATURE-----
 

[RISCV][TII] Add and use new hook to simplify/canonicalize instructions after MachineCopyPropagation (#137973)

PR #136875 was posted as a draft PR that handled a subset of these
cases, using the CompressPat mechanism. The consensus from that
discussion (and a conclusion I agree with) is that it would be
beneficial doing this optimisation earlier on, and in a way that isn't
limited just to cases that can be handled by instruction compression.

The most common source for instructions that can be
optimized/canonicalized in this way is through tail duplication in
MachineBlockPlacement followed by machine copy propagation. For RISC-V,
choosing a more canonical instruction allows it to be compressed when it
couldn't be before. There is the potential that it would make other
MI-level optimisations easier.

This modifies ~910 instructions across an llvm-test-suite build
including SPEC2017, targeting rva22u64. Looking at the diff, it seems
there's room for eliminating instructions or further propagating after
this.

Coverage of instructions is based on observations from a script written
to find redundant or improperly canonicalized instructions (though I aim
to support all instructions in a 'group' at once, e.g. MUL* even if I
only saw some variants of MUL in practice).