Differences
This shows you the differences between two versions of the page.
Next revision | Previous revision | ||
projects:nbitinteger [2016/10/30 18:13] – created sbs | projects:nbitinteger [2016/10/30 18:19] (current) – sbs | ||
---|---|---|---|
Line 1: | Line 1: | ||
- | **Lead Developers: | + | **Lead Developers: |
Many applications that deal with whole numbers do not require the full range provided by the default integer type. While this may be negligible in cases where only a few such numbers are needed, when dealing with large arrays of them the situation is radically different. Using " | Many applications that deal with whole numbers do not require the full range provided by the default integer type. While this may be negligible in cases where only a few such numbers are needed, when dealing with large arrays of them the situation is radically different. Using " | ||
Line 5: | Line 5: | ||
The development of a compilation strategy for m-dimensional arrays of n-bit integers poses a major challenge when attempting to target a wide range of architectures with vector instructions including mainstream architectures such as X86 and Sparc as well as novel multicores such as the Cell Broadband Engine or Intel' | The development of a compilation strategy for m-dimensional arrays of n-bit integers poses a major challenge when attempting to target a wide range of architectures with vector instructions including mainstream architectures such as X86 and Sparc as well as novel multicores such as the Cell Broadband Engine or Intel' | ||
- | Current Status: | + | **Current Status**: Artem looked into this and ended up doing layout transformations for vectorisation. This work should build on Artem' |
+ | |||
+ | **Needed Work:** | ||
+ | * Integration of new n-bit data types | ||
+ | * translation schemes for SIMD in a register | ||
+ | * implementation of SIMD in a register; including padding for n-dimensional arrays | ||
+ | * performance evaluation | ||