[THEME] Universal 64-bit index type support in Thrust/CUB algorithms

[jrhemstad]

Summary

Users expect Thrust/CUB algorithms to work with input sequences larger than INT_MAX. However, for historical reasons, many Thrust/CUB algorithms are hard-coded to use int as the index/offset type. As a result, those algorithms will fail (often silently) when provided inputs whose extent is larger than std::numeric_limits<int>::max().

In order to provide a delightful and consistent user experience, every Thrust/CUB should be able to work with inputs larger than INT_MAX. However, we likely want different solutions in Thrust vs CUB.

Intended Outcome

Thrust

• All Thrust algorithms should work with large inputs by default.
• Users should be able to opt-in to explicitly specify the offset type to avoid any compile/run-time overhead

CUB

• CUB algorithms need not work with large inputs by default, but should enable users to work with large inputs

Depends On

• [EPIC] Improve separation and elminate redundancy between Thrust and CUB #24

Tasks

• [EPIC] Thrust large input support #49
• [EPIC]: CUB large input support #50

Related Info

NVIDIA/cub#212 was the previous tracking issue in CUB that has a lot of good discussion about what we want a solution to look like in Thrust vs CUB.

[bhack]

Any news on this?