Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
P
POCL
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
TRACES-public
Vortex
POCL
Commits
7f535ae6
Commit
7f535ae6
authored
2 years ago
by
Michal Babej
Browse files
Options
Downloads
Patches
Plain Diff
Update documentation
parent
6cf69be6
No related branches found
No related tags found
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
CHANGES
+5
-3
5 additions, 3 deletions
CHANGES
doc/notes-4.0.txt
+60
-0
60 additions, 0 deletions
doc/notes-4.0.txt
with
65 additions
and
3 deletions
CHANGES
+
5
−
3
View file @
7f535ae6
4.0
unreleased
4.0
June 2023
=================
=================
Notable User Facing Changes
Notable User Facing Changes
...
@@ -36,8 +36,8 @@ Notable User Facing Changes
...
@@ -36,8 +36,8 @@ Notable User Facing Changes
local memory size for the CPU devices.
local memory size for the CPU devices.
- CUDA: Added Direct Peer-to-Peer buffer migrations for much better performance
- CUDA: Added Direct Peer-to-Peer buffer migrations for much better performance
scaling in multi-GPU scenarios
scaling in multi-GPU scenarios
-
P
artial support for cl_khr_fp16 when compiled with LLVM 16+.
The
-
CPU drivers gained p
artial support for cl_khr_fp16 when compiled with LLVM 16+.
implementation relies on Clang, and may result in emulation if
The
implementation relies on Clang, and may result in emulation if
the CPU does not support the required instruction set.
the CPU does not support the required instruction set.
- Linux/x86-64 only: SIGFPE handler has been changed to skip instructions
- Linux/x86-64 only: SIGFPE handler has been changed to skip instructions
causing division-by-zero, only if it occured in one of the CPU driver
causing division-by-zero, only if it occured in one of the CPU driver
...
@@ -49,6 +49,8 @@ Notable User Facing Changes
...
@@ -49,6 +49,8 @@ Notable User Facing Changes
to reflect the hardware they're driving instead of implementation details.
to reflect the hardware they're driving instead of implementation details.
- POCL_MAX_PTHREAD_COUNT renamed to POCL_CPU_MAX_CU_COUNT; the old env. variable is deprecated
- POCL_MAX_PTHREAD_COUNT renamed to POCL_CPU_MAX_CU_COUNT; the old env. variable is deprecated
but still works
but still works
- it's now possible to disable libhwloc support even when it's present, using -DENABLE_HWLOC=0
CMake option
Notable Fixes
Notable Fixes
-------------
-------------
...
...
This diff is collapsed.
Click to expand it.
doc/notes-4.0.txt
0 → 100644
+
60
−
0
View file @
7f535ae6
Portable Computing Language (PoCL) v4.0 released
------------------------------------------------
PoCL is a portable open source (MIT-licensed) implementation of the
OpenCL standard. It aims to support the minimal v3.0 feature set (official
conformance stamp not yet applied for).
In addition to being an easily portable multi-device (truly
heterogeneous) open-source OpenCL implementation, a major goal of the
project is improving interoperability for diversity of OpenCL-capable
devices by integrating them to a single centrally orchestrated platform.
Another key goal is to enhance performance portability of OpenCL
programs across device types utilizing runtime and compiler techniques.
Upstream PoCL currently supports various CPUs, NVIDIA GPUs via libcuda,
Vulkan-compatible GPUs (experimental) and ASIPs (experimental,
see: http://openasip.org). It is also known to have multiple (private)
adaptations in active production use.
Release Highlights
------------------
* Support for Clang/LLVM 16.0
* Support for Clang/LLVM 6-9 (inclusive) dropped
* More improvements to SPIR-V support for CPU and CUDA drivers
* CPU drivers: gained support for program-scope variables, generic AS,
cl_khr_fp16 (LLVM 16+), cl_khr_subgroups, cl_intel_unified_shared_memory.
* New driver for devices using the Level Zero API.
Misc. Notes
-----------
* Please note that there's an official PoCL "maintenance policy" in
place. This text describes the policy and how you can get your favourite
project that uses OpenCL to remain regression free in the future PoCL
releases: http://portablecl.org/docs/html/maintainer-policy.html
* We are looking for ARM CPU and RISC-V CPU maintainers (and any other
target of your interest). If you are interested in ensuring PoCL stays
stable for these processor architectures in the future, please contact us.
* The dedicated TTASim driver will be deprecated once the AlmaIF driver
reaches feature parity with it.
Acknowledgements
----------------
Customized Parallel Computing (CPC) research group of Tampere University,
Finland leads the development of PoCL on the side and for the needs of
their research projects. This project has received funding from European
Union's Horizon 2020 research and innovation programme under Grant
Agreement No 871738 (CPSoSaware), Academy of Finland (decision #331344) and
Business Finland's AISA project.
The financial support is very much appreciated -- it keeps this open
source project going!
Links
-----
Release web page: http://portablecl.org/pocl-4.0.html
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment