Skip to content

Releases: ROCm/rocPRIM

rocPRIM 2.10.0

23 Nov 00:51
22d413d
Compare
Choose a tag to compare

point release for ROCm 2.10

rocPRIM 2.9.0

15 Oct 19:06
cebf382
Compare
Choose a tag to compare
Updating rocPRIM version for rocm2.9 (#102)

rocPRIM 2.7.0

16 Sep 15:54
cebf382
Compare
Choose a tag to compare

Release for rocm 2.7

  • performance optimizations for gfx906 architecture GPUs.

rocPRIM 2.7.2

16 Sep 15:54
Compare
Choose a tag to compare

Bug Fix release

rocPRIM 2.8.0

24 Sep 19:48
a220074
Compare
Choose a tag to compare

Minor bug fixes

rocPRIM 2.6.0

08 Jul 19:58
c5c8b35
Compare
Choose a tag to compare

Point release for rocm 2.6 with minor bugfixes

rocPRIM 2.5.0

03 Jun 22:40
aec473f
Compare
Choose a tag to compare

Enables rocThrust and hipCUB features.

Packaged as part of ROCm 2.5 release. It is available via repo.radeon.com for Ubuntu 16 and Centos 7.

Fixes for Fiji GPU failures are included.

rocPRIM 2.3

12 Apr 21:58
41ff492
Compare
Choose a tag to compare

Point Release for rocm 2.3

  • Minor bug fixes

rocPRIM v1.0.0

15 Jun 16:04
Compare
Choose a tag to compare
rocPRIM v1.0.0

rocPRIM v0.3.0

16 Apr 09:16
Compare
Choose a tag to compare
rocPRIM v0.3.0 Pre-release
Pre-release

Milestone 3

All functions needed for Caffe2 and Tensorflow 1.3 are now finished. Optimizations are only selectively done, where the rest should arrive with milestone 4.

Done in milestones 1 and 2:

  • Scan, reduce and sort algorithms (warp, block, device)
  • Block and thread I/O primitives
  • Block data exchange primitives
  • Reduce-by-key, transform (device)
  • Discontinuity algorithm (block)

Added in this milestone:

  • Fancy iterators
  • Segmented reduction, scan and sort (device)
  • Select (copy if) and unique operations (device)
  • Histogram algorithm (block, device)
  • Run length encode algorithm (device)

Not yet finished:

  • Partition algorithm (device)
  • Comparison sort (warp, block, device), merge (device)