tree 6d317de0a01721b5cbfb0b5d9c056661c2e2d932
parent 2bef72007dd975fa7469d2b1142378d037610981
author Scott Todd <scott.todd0@gmail.com> 1722352558 -0700
committer GitHub <noreply@github.com> 1722352558 -0700
gpgsig -----BEGIN PGP SIGNATURE-----
 
 wsFcBAABCAAQBQJmqQOuCRC1aQ7uu5UhlAAAWb8QACi1DV92cb5SfgQ7koa6jNdC
 PkGuQ6MaM4DcPTAWuKobWT0QrBudQJoI/FEJUSZg0j2DnYQIq3YQ6ZBCkCqlzKFw
 QR3MKEeGkwzwrdUFu2z45PNnEVOnMmM/RT4q8EB8SIj782u0jYR6XmmfW9AL7vLP
 xmX08tmW49gmbXzgbttEe0u5SPG1FiawBHL5s1J6MEt3le/kHTPsBMGP/bHDzyl/
 R55WaRyQ8sXVBkxzY7841x5KL264wZ2jpYXAjvrzabhu8zIBmL79/OQQhE2N8kq2
 E/ZnSkAX7VdGdcwF+cZrpfesIZYiARzOAHomN+Rg6XyHxsKtRFy8kPXfqKCXug49
 K0TXUtwNcYvO2q3e2KJe9FOOinkMZefApmL4PcGA4na6i2ldpY8STI3bLj+jXQ56
 6ckxHjwVWKAccpCown81JI1muG4KlcjIewsstPxeuDlPp2TivSFEQO1T9NS0jMVn
 7mo3cMraqF2cWWm152cfCdTn2JefFAsm4cPbAvw+uO8uO5IQkMSBXxtFuUfrcrxU
 GUfmEYqXk8whhwpY9SpweZcBK6HAk8Zec5Ym7YFpyrCKixcZA45DzqZiblRQsvxr
 xwLGJZ0WqevFwsV5TMERBPxpnpucSLbNxOSuhPo/1BgVb9bnS58EAGPXSEyPYQzU
 QcycPHfNQaIOn+i5Oeqa
 =GmJm
 -----END PGP SIGNATURE-----
 

Simplify tests/e2e/tensor_ops. (#17854)

Forked from https://github.com/iree-org/iree/pull/17766

* ~~Move `tests/e2e/tensor_ops/pack_i8.mlir` to
`tests/e2e/cpu_specific/pack_i8_vnni.mlir` and add a comment explaining
what is VNNI specific about it~~
* Enable `tensor_cast` and `unpack` tests on CUDA
* Add Metal and ROCm/HIP test suites to Bazel
* Share test srcs lists between:
  * CPU and VMVX
  * Vulkan and Metal (both using SPIR-V codegen)
  * CUDA and ROCm (both using LLVMGPU codegen)