Rohit Santhanam 2d132d1736 Commit 52a5f982 broke conjhelper functionality for HIP GPUs. ...

This commit addresses this.

2021-06-25 19:28:00 +00:00

..

AltiVec

Small cleanup: Get rid of the macros EIGEN_HAS_SINGLE_INSTRUCTION_CJMADD and CJMADD, which were effectively unused, apart from on x86, where the change results in identically performing code.

2021-06-24 18:52:17 -07:00

AVX

Get rid of code duplication for conj_helper. For packets where LhsType=RhsType a single generic implementation suffices. For scalars, the generic implementation of pconj automatically forwards to numext::conj, so much of the existing specialization can be avoided. For mixed types we still need specializations.

2021-06-24 15:47:48 -07:00

AVX512

Get rid of code duplication for conj_helper. For packets where LhsType=RhsType a single generic implementation suffices. For scalars, the generic implementation of pconj automatically forwards to numext::conj, so much of the existing specialization can be avoided. For mixed types we still need specializations.

2021-06-24 15:47:48 -07:00

CUDA

Better CUDA complex division.

2021-04-29 17:39:58 +00:00

Default

Commit 52a5f982 broke conjhelper functionality for HIP GPUs.

2021-06-25 19:28:00 +00:00

GPU

Fix some CUDA warnings.

2021-02-24 00:16:31 +00:00

HIP/hcc

moving Half headers from CUDA dir to GPU dir, removing the HIP versions

2018-06-13 12:26:18 -04:00

MSA

Small cleanup: Get rid of the macros EIGEN_HAS_SINGLE_INSTRUCTION_CJMADD and CJMADD, which were effectively unused, apart from on x86, where the change results in identically performing code.

2021-06-24 18:52:17 -07:00

NEON

Small cleanup: Get rid of the macros EIGEN_HAS_SINGLE_INSTRUCTION_CJMADD and CJMADD, which were effectively unused, apart from on x86, where the change results in identically performing code.

2021-06-24 18:52:17 -07:00

SSE

Get rid of code duplication for conj_helper. For packets where LhsType=RhsType a single generic implementation suffices. For scalars, the generic implementation of pconj automatically forwards to numext::conj, so much of the existing specialization can be avoided. For mixed types we still need specializations.

2021-06-24 15:47:48 -07:00

SVE

Small cleanup: Get rid of the macros EIGEN_HAS_SINGLE_INSTRUCTION_CJMADD and CJMADD, which were effectively unused, apart from on x86, where the change results in identically performing code.

2021-06-24 18:52:17 -07:00

SYCL

[SYCL Function pointer Issue]: SYCL does not support function pointer inside the kernel, due to the portability issue of a function pointer and memory address space among host and accelerators. To fix the issue, function pointers have been replaced by function objects.

2020-11-12 01:50:28 +00:00

ZVector

Small cleanup: Get rid of the macros EIGEN_HAS_SINGLE_INSTRUCTION_CJMADD and CJMADD, which were effectively unused, apart from on x86, where the change results in identically performing code.

2021-06-24 18:52:17 -07:00