aaron/ispc - ispc - git.frat.tech

aaron/ispc

Fork 0

Commit Graph

Author	SHA1	Message	Date
Matt Pharr	974b40c8af	Add type suffix to comparison ops in C++ output. e.g. "__equal()" -> "__equal_float()", etc. No functional change; this is necessary groundwork for a forthcoming peephole optimization that eliminates ANDs of masks in some cases.	2012-07-07 07:50:59 -07:00
Jean-Luc Duprat	ac421f68e2	Ongoing support for int64 for KNC: Fixes to __load and __store. Added __add, __mul, __equal, __not_equal, __extract_elements, __smear_i64, __cast_sext, __cast_zext, and __scatter_base_offsets32_float. __rcp_varying_float now has a fast-math and full-precision implementation.	2012-07-05 17:05:42 -07:00
Jean-Luc Duprat	95d8f76ec3	Added prelimary support for Intel's Xeon Phi KNC processor. float, int32 and double support is included; int8, int16 and int64 not supported yet. This is work in progress and not considered stable yet.	2012-06-28 12:00:55 -07:00

Author

SHA1

Message

Date

Matt Pharr

974b40c8af

Add type suffix to comparison ops in C++ output.

e.g. "__equal()" -> "__equal_float()", etc.

No functional change; this is necessary groundwork for a forthcoming
peephole optimization that eliminates ANDs of masks in some cases.

2012-07-07 07:50:59 -07:00

Jean-Luc Duprat

ac421f68e2

Ongoing support for int64 for KNC:

Fixes to __load and __store.
Added __add, __mul, __equal, __not_equal, __extract_elements, __smear_i64, __cast_sext, __cast_zext,
and __scatter_base_offsets32_float.

__rcp_varying_float now has a fast-math and full-precision implementation.

2012-07-05 17:05:42 -07:00

Jean-Luc Duprat

95d8f76ec3

Added prelimary support for Intel's Xeon Phi KNC processor.

float, int32 and double support is included; int8, int16 and int64
not supported yet.

This is work in progress and not considered stable yet.

2012-06-28 12:00:55 -07:00

3 Commits