-
Ben and Aydin were discussing about a zero-copy interface for GraphBLAS matrices. This is something Tim D. wanted and he probably hacked it by giving CSR pointers, which is something Ben is also doing…
-
Currently, in our lowering passes, we add `graphblas.convert_layout` ops when we're given a CSC matrix when the "real" lowering expects a CSR matrix.
It might be cleaner to use [canonicalization p…
-
Where scalars and vectors are automatically promoted to vectors and matrices through replication of values to carry out existing GraphBLAS operations
[RankPromotionProposal.pdf](https://github.com/…
-
The ALP code here
https://github.com/Algebraic-Programming/ALP/blob/b50fe72e957ed4da10c1cd1c59d924260f051a8d/include/graphblas/bsp1d/exec.hpp#L184
immediately segfaults because of the `sizeof( s…
-
A suggestion/idea — I read the paper ["Vertical, Temporal, and Horizontal Scaling of Hierarchical Hypersparse GraphBLAS Matrices (2021)"](https://arxiv.org/pdf/2108.06650.pdf) and I was wondering whet…
-
I'm getting unexpectedly high memory usage with flox. Here's what I've been doing:
```python
import dask.distributed as dd
import dask.array as da
import numpy as np
import flox
cluster = dd…
-
`GrB_init` should be extended to pass in pointers to `malloc`, `calloc`, `realloc`, and `free`. Required for matrix import/export. See SuiteSparse:GraphBLAS `GxB_init()`.
-
We're making progress on the CUDA kernels for SuiteSparse:GraphBLAS.
The change to the C API will be very slight. See the new "cuda" branch on LAGraph, which is now synced with the latest LAGraph …
-
When trying to build SuiteSparse newer than 5.8.1 with either intel compilers version `2021.2.0` or `2021.4.0`, the build fails whilst building GraphBLAS with:
```
/dev/shm/jfg508/SuiteSparse/5.11.0…
-
[18Mar2020] Should LAGraph return a GrB_Matrix that is deferred?
Related question: are the algorithms is composable for which the result of one is passed to another.
Is GraphBLAS execution mode …