Tech Agenda Track: Meeting Minutes 3/30

[emilypi]

HF Technical Agenda Task Force - Meeting Minutes
March 30, 2021

Attendees

Ben Gamari
Emily Pillmore
Edward Kmett
Michael Snoyman
Andrew Boardman

Previous Meeting Minutes:

Tech Agenda Track: Meeting Minutes 3/26

Not sure where the right place to record this is, but I asked a friend who had encountered issues diagnosing pinned-memory induced fragmentation what tools might have helped with the job, and this was the reply: “default output of any of the memory inspecting tools to know what fraction of the heap is pinned memory (it took a little while to realize that’s what was going on), and some kind of way to work backwards through heap references to the thing that’s retaining pinned memory (or anything, …

General

This is the first of the new series of weekly reporting meetings for individual proposal tracks. Previously, we had split into two tracks to parallelize the brainstorming needed to generate project ideas and proposals, now we are in a phase where we need to project plan and report like a stand up.

The formal proposal process will be delivered Wednesday, viewable on github. Emily will backport existing proposals and open PR’s for them in the repo.

Vector Types Proposal

Michael: reached out to winterland + Alexey. It was painful enough in the past on Random for Alexey that he doesn’t want to work on this stuff. Changes he’d like to see on primitive: unifying monadIO and monadPrim, stRef + IORef etc. Unifying prim + storable? New primops have landed in GHC to write to bytearrays. This points to a nicer unification for “writing bytes somewhere”.

Michael: Winterland has not responded yet. Let’s put bytearray in some kind of lower level package

Ben: There may be a way forward on bytestring which affects text too, and perhaps it’s just the bytearray-backed Bytestring work. Andrew is concerned about compatibility.

Ed: Worried about changing internals of Bytestring since it affects my prod code, but it’s serviceable.

Michael: we could just make primitive the core package

Ben: We can make ByteString a newtype around a PrimArray from the primitive package

Michael/Ed: can we make it easier to derive (either newtype or stock) the Prim typeclass?

Michael: Primitive has no bytearray slicing.

Ed: This would have to sit on top of prim vector/prim array

See:

Base proposal around vector-like types

Following up on the previous discussions around alternative standard libraries and preludes, I wrote up a proposal for what I’d like to see to address IMO a major pain point today: the proliferation of different packed memory representations. I’m going to add a link to this Discourse thread in the blog post as the official discussion point.

Towards a prevalent alternative prelude?

Three cheers! Let’s make it happen! I’d agree it’s mostly a coordination issue, but we also need to navigate cultural hurdles (why hasn’t this cleanup already happened?). We also have some strong egos to work with.

Haskell base proposal: unifying vector-like types

Action Items:

• Involve Andrewthad on moving primitive into base and bless it.
• Later: move text + bytestring onto byte arrays

GHC Performance Dashboard Proposal

Ben: Matt pickering and I had a productive morning on this. We have a dashboard now. It’s already looking pretty good. See: Grafana. I’ll update the proposal and we’ll figure out how to work on other projects. This would be a good GSOC project.

Emily: Is the GSOC proposal window still open?

Ben: Yeah, not sure. I have enough people I’m advising, and this would be a good way to find a mentor.

See:

Are we fast yet? Performance monitoring for the Haskell ecosystem - HedgeDoc

Action Items:

• Check GSOC proposal potential
• Continue working with Pickering

Text-Utf8 Proposal

Emily: Meeting next week to discuss project planning.
Ed: Don’t break text-icu. If you’re moving text to text-utf16, make text-utf16-icu and make text-icu work with text.
Emily: Will do.

See:
TODO

Action Items:

• Meeting among the maintainers next week.

Haskell Foundation Book Proposal

See:

2021-03-11 Proposal: Haskell Foundation book

Action Items:

• Nothing todo yet. Consider it a thunk.

GHC CI Proposal

Ben: Haven’t heard back from MSR. We’ve pared down the CI stuff alot, which hasn’t even affected test coverage, so things are good right now. But we are dependent on free resources from a particular provider still.

Ed: Could you give me an idea of how big the resourcing issue is?

Ben: 5 boxes on a hetzner are $70/m for a good box. It’s a pretty small cost.

Ed: So this is a $4k. We can just take the budget hit and just pay in the meantime.

Emily/Andrew: We don’t have to consider the total budget at this time. Amortize the costs over the year - we have funds!

Ben: Okay, i will write up a new proposal for this.

See:
TODO

Action Items:

• Write a “how will this help GHC?” document.
• Ben will write a proposal for the HF paying for this.

Windows Platform Proposal

Action Items:

• Set up a meeting with Emily, Michael, Ben, and Tamar.

Operation Project Matchmaker

Hecate: I am starting a new job, so this will live as a separate project for now.

See:
Project description

Action Items:

None.