1 //===---------------------------------------------------------------------===// 2 // Testcases that crash the X86 backend because they aren't implemented 3 //===---------------------------------------------------------------------===// 4 5 These are cases we know the X86 backend doesn't handle. Patches are welcome 6 and appreciated, because no one has signed up to implemented these yet. 7 Implementing these would allow elimination of the corresponding intrinsics, 8 which would be great. 9 10 1) vector shifts 11 2) vector comparisons 12 3) vector fp<->int conversions: PR2683, PR2684, PR2685, PR2686, PR2688 13 4) bitcasts from vectors to scalars: PR2804 14 5) llvm.atomic.cmp.swap.i128.p0i128: PR3462 15
