[Bug 2141514] Re: [needs-packaging] rocwmma - in ubuntu
Talha Can Havadar
2141514 at bugs.launchpad.net
Thu Feb 19 16:02:06 UTC 2026
Hello Frank,
- updated the patches with gbp so they should have an header including Forwarded field and Last-Update
- rocwmma is header only library so no so file is produced to consume
- updated the control file with descriptions
- build is taking long time one causing problems in launchpad infra nproc workaround which is inherited from llvm-toolchain package helps me to workaround this issue. Do you want me to put this in lintian-overrides?
- about the license I think I am following what is suggested for packaging which needs to follow how the software licensed. (but maybe I am interpreting it wrongly, https://library.canonical.com/legal-processes/licensing-policy#:~:text=places%2E-,Packaging )
- updated maintainer field
- about size of -tests package , yeah I hope it wont be a problem but need to hear from AA indeed.
Can you review again? and what is your idea on lintian overrides for
nproc?
If source is still not available in ppa, changes are already pushed to
the git repo in description
--
You received this bug notification because you are a member of Ubuntu
Sponsors, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/2141514
Title:
[needs-packaging] rocwmma - in ubuntu
Status in Ubuntu:
In Progress
Bug description:
Aiming to package v7.1.0 of the upstream to have better compatibility
with the ROCm stack (v7.1.0) in ubuntu resolute. From 7.2.0 onwards
upstream moved to https://github.com/ROCm/rocm-libraries so thats why
you see deprecation warning in the upstream repo but that is the
source for v7.1.0
The rocWMMA is a C++ library for accelerating mixed-precision matrix
multiply-accumulate (MMA) operations leveraging AMD GPU hardware.
rocWMMA makes it easier to break down MMA problems into fragments and
distribute block-wise MMA operations in parallel across GPU
wavefronts. The API consists of a header library, that can be used to
compile MMA acceleration directly into GPU kernel device code. This
can benefit from compiler optimization in the generation of kernel
assembly, and doesn't incur additional overhead costs of linking to
external runtime libraries or having to launch separate kernels.
rocWMMA includes sample projects to validate and demonstrate API
usage. These include simple GEMMs, performant GEMMs, DLRM, GEMV and
hipRTC integration.
[needs-packaging]:
Upstream: https://github.com/ROCm/rocWMMA
URL: https://code.launchpad.net/~bullwinkle-team/ubuntu/+source/rocwmma/+git/rocwmma/+ref/bullwinkle/llvm-21/ubuntu/devel
License: https://git.launchpad.net/~bullwinkle-team/ubuntu/+source/rocwmma/tree/debian/copyright?h=bullwinkle/llvm-21/ubuntu/devel
PPAs:
rocm-devel-21: https://launchpad.net/~bullwinkle-team/+archive/ubuntu/rocm-devel-21
just-for-this-bug: https://launchpad.net/~tchavadar/+archive/ubuntu/lp2141514
- all archs and -proposed enabled
autopkgtest: https://autopkgtest.ubuntu.com/user/tchavadar/ppa/lp2141514
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+bug/2141514/+subscriptions
More information about the Ubuntu-sponsors
mailing list