Skip to content

Prepare changelog for v4.1 #745

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
abouteiller wants to merge 2 commits into
base: master
Choose a base branch
from
Draft

Prepare changelog for v4.1 #745

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7019c4b
Prepare changelog for v4.1
abouteiller Mar 14, 2025
eee6a5d
changelog: pr#733 don't bind by default
abouteiller Mar 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 29 additions & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,35 @@ changes to this project (since v1.1.0).
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

v4.1.2503
---------

### Added
- PaRSEC API 4.1
- New query to obtain the peer-access mask between GPUs: PR#716
- GPU Data copies can have no primary CPU copies, the CPU copy is allocated as needed if we run out of GPU memory: PR#711
- PaRSEC objects can have customized free callbacks: PR#731

### Changed
- PaRSEC will not bind threads by default, but will instead inherit binding from the external launcher (e.g., mpiexec, srun). In some cases the launcher will bind 1-process-per-node on a single core, be mindful about checking the binding with the external launcher (e.g., `mpiexec --report-binding` for Open MPI, `srun --cpu-bind=verbose` in Slurm, etc.). PR#730

### Deprecated

### Removed

### Fixed
- Usage of MPI hint allow_overtake would taint the input communicator: PR#708
- Non-POSIX calls prevented compilation on Windows: PR#719
- Eviction of memory under-transfer would cause assert: PR#733
- Tasks scheduled from tasks could deadlock: PR#736

### Known Bugs

- Enabling the `RECURSIVE` device will cause crashes (it is disabled by default in this release); see issues #548, #541.
- Running out of GPU memory when using the NEW keyword in PTG may cause deadlocks; see issue #527.

### Security


v4.0.2411
---------
Expand Down