1//===---------------------------------------------------------------------===// 2// Testcases that crash the X86 backend because they aren't implemented 3//===---------------------------------------------------------------------===// 4 5These are cases we know the X86 backend doesn't handle. Patches are welcome 6and appreciated, because no one has signed up to implemented these yet. 7Implementing these would allow elimination of the corresponding intrinsics, 8which would be great. 9 101) vector shifts 112) vector comparisons 123) vector fp<->int conversions: PR2683, PR2684, PR2685, PR2686, PR2688 134) bitcasts from vectors to scalars: PR2804 145) llvm.atomic.cmp.swap.i128.p0i128: PR3462 15