mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2025-01-07 11:33:44 +00:00
af57bdf7d6
Sort the operands of the other entries in the current vectorization root according to the first entry's operands opcodes. %conv0 = uitofp ... %load0 = load float ... = fmul %conv0, %load0 = fmul %load0, %conv1 = fmul %load0, %conv2 Make sure that we recursively vectorize <%conv0, %conv1, %conv2> and <%load0, %load0, %load0>. This makes it more likely to obtain vectorizable trees. We have to be careful when we sort that we don't destroy 'good' existing ordering implied by source order. radar://15080067 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191977 91177308-0d34-0410-b5e6-96231b3b80d8 |
||
|---|---|---|
| .. | ||
| barriercall.ll | ||
| cast.ll | ||
| cmp_sel.ll | ||
| compare-reduce.ll | ||
| crash_7zip.ll | ||
| crash_bullet3.ll | ||
| crash_bullet.ll | ||
| crash_dequeue.ll | ||
| crash_flop7.ll | ||
| crash_lencod.ll | ||
| crash_mandeltext.ll | ||
| crash_netbsd_decompress.ll | ||
| crash_sim4b1.ll | ||
| crash_smallpt.ll | ||
| cross_block_slp.ll | ||
| cse.ll | ||
| cycle_dup.ll | ||
| debug_info.ll | ||
| diamond.ll | ||
| external_user.ll | ||
| extract.ll | ||
| flag.ll | ||
| hoist.ll | ||
| horizontal.ll | ||
| implicitfloat.ll | ||
| in-tree-user.ll | ||
| insert-element-build-vector.ll | ||
| lit.local.cfg | ||
| long_chains.ll | ||
| loopinvariant.ll | ||
| multi_block.ll | ||
| multi_user.ll | ||
| odd_store.ll | ||
| operandorder.ll | ||
| ordering.ll | ||
| phi3.ll | ||
| phi_landingpad.ll | ||
| phi_overalignedtype.ll | ||
| phi.ll | ||
| pr16571.ll | ||
| pr16628.ll | ||
| pr16899.ll | ||
| reduction2.ll | ||
| reduction.ll | ||
| rgb_phi.ll | ||
| saxpy.ll | ||
| simple-loop.ll | ||
| simplebb.ll | ||
| tiny-tree.ll | ||
| vector.ll | ||