Fix misaligned addresses with batched loggers

ginkgo-project / ginkgo

Numerical linear algebra software package

https://ginkgo-project.github.io/

BSD 3-Clause "New" or "Revised" License

401 stars 88 forks source link

Fix misaligned addresses with batched loggers #1578

Closed pratikvn closed 4 months ago

pratikvn commented 6 months ago

This PR fixes errors with cuda misaligned addresses in the batched loggers when mixing 32 bit and 64 bit types reported in #1576. It adds a setup to allow aliasing workspace pointers with arrays of different types, which is necessary to prevent repeated allocations while preventing misaligned issues.

Fixes #1576

pratikvn commented 6 months ago

In this case the maximum number of elements possible is equal to the number of batch items. So, I think it is definitely possible to have more than max(int32) for that. Additionally, this is not a shared memory issue, but an issue of using a workspace and storing pointers to different types within the workspace that was causing the misaligned accesses.

Nevertheless, your suggestion on the CUB approach is a nice way to do it, and I will look into that.

upsj commented 5 months ago

This change only deals with final iteration counts, which is where IMO 64 bit integers make no sense.

pratikvn commented 5 months ago

Yes, only the final iteration counts are being logged, but the number of entries scales with the number of batch items that are being solved, which can possibly more than max(int32)

upsj commented 5 months ago

The type you use to index an array is independent of the value type of the array, I am talking about the value type

pratikvn commented 5 months ago

Ah, yes. I misunderstood your comment. I am looking into the CUB blueprint. That should be usable for correct aligntment of both the workspace vectors and for the shared memory objects.