Commit Graph

17 Commits

Author SHA1 Message Date
Rasmus Munk Larsen
b6fcddccfc Get rid of pblend packet op.
There was only a single code path left in TensorEvaluator using pblend. We can replace that with a call to the more general TernarySelectOp and get rid of pblend entirely from Core.

Closes #2998

See merge request libeigen/eigen!2056

Co-authored-by: Rasmus Munk Larsen <rmlarsen@google.com>
2025-11-03 23:27:50 +00:00
Rasmus Munk Larsen
ece9a4c0b6 Always vectorize abs2() for non-complex types.
For several packet types, `abs2` was not vectorized even if it only requires `pmul`. Get rid of the confusing and redundant `HasAbs2` enum and instead check `HasMul` in addition to making sure the scalar type is not complex.

See merge request libeigen/eigen!2048
2025-10-30 02:42:56 +00:00
Rasmus Munk Larsen
97c7cc6200 Explicitly use the packet trait HasPow to control whether Pow is vectorized. 2025-07-18 21:51:42 +00:00
Rasmus Munk Larsen
0d366f6532 Vectorize erfc(x) for double and improve erfc(x) for float. 2024-11-08 17:21:11 +00:00
qile lin
072ec9d954 Fix a bug for pcmp_lt_or_nan and Add sqrt support for SVE 2024-09-04 21:45:39 +00:00
qile lin
3b5a1b4157 sve instrinsics with "_x" suffix will be faster than "_z" suffix 2024-08-23 12:52:22 +00:00
Charles Schlosser
fb95e90f7f Add truncation op 2024-04-29 23:45:49 +00:00
Rasmus Munk Larsen
4d419e2209 Rename generic_fast_tanh_float to ptanh_float and move it to... 2024-02-16 21:27:22 +00:00
Tobias Wood
f38e16c193 Apply clang-format 2023-11-29 11:12:48 +00:00
Antonio Sánchez
6e4d5d4832 Add IWYU private pragmas to internal headers. 2023-08-21 16:25:22 +00:00
Rasmus Munk Larsen
df1049ddf4 Small packet math cleanup. 2023-04-04 16:14:32 +00:00
Guoqiang QI
32a3f9ac33 Improve plogical_shift_* implementations and fix typo in SVE/PacketMath.h 2022-05-23 09:33:49 +00:00
Sean McBride
f1b9692d63 Removed EIGEN_UNUSED decorations from many functions that are in fact used 2022-03-03 20:19:33 +00:00
Rasmus Munk Larsen
d7d0bf832d Issue an error in case of direct inclusion of internal headers. 2021-09-10 19:12:26 +00:00
Rasmus Munk Larsen
bffd267d17 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
Antonio Sanchez
7ff0b7a980 Updated pfrexp implementation.
The original implementation fails for 0, denormals, inf, and NaN.

See #2150
2021-02-17 02:23:24 +00:00
David Tellenbach
65e2169c45 Add support for Arm SVE
This patch adds support for Arm's new vector extension SVE (Scalable Vector Extension). In contrast to other vector extensions that are supported by Eigen, SVE types are inherently *sizeless*. For the use in Eigen we fix their size at compile-time (note that this is not necessary in general, SVE is *length agnostic*).

During compilation the flag `-msve-vector-bits=N` has to be set where `N` is a power of two in the range of `128`to `2048`, indicating the length of an SVE vector.

Since SVE is rather young, we decided to disable it by default even if it would be available. A user has to enable it explicitly by defining `EIGEN_ARM64_USE_SVE`.

This patch introduces the packet types `PacketXf` and `PacketXi` for packets of `float` and `int32_t` respectively. The size of these packets depends on the SVE vector length. E.g. if `-msve-vector-bits=512` is set, `PacketXf` will contain `512/32 = 16` elements.

This MR is joint work with Miguel Tairum <miguel.tairum@arm.com>.
2021-01-21 21:11:57 +00:00