Rohan Yadav

I see this error now: ``` [yadav2@lassen737:build]$ ./bin/taco "a(i) = B(i, j) * c(i)" -f=a:d -f=B:ds -f=c:d -g=c:d -time=5 -i=B:../../taco/petsc/test-mat.mtx B file read: 0.846304 ms B pack: 157.335 ms B...

Ah I think I didn't rebuild properly. It works for me now. Let me see what I can pull in from the commit to fix the issue on my branch.

This fails some tests where the assumption about lifetimes of TACO objects is somewhat backwards. The expr.storage and alloc.storage tests both assume that the created IndexExpr objects will outlive the...

Ah, I never understood why the CUDA code generator decided to emit `break` for a `Continue` operation. Do you think you could post the full generated code here? In my...

I see. In this case we definitely shouldn't be returning. > In my mind, if the for loop being continue'd out of is parallelized over GPU threads, then we would...

cc @stephenchouca this PR makes it easier to write benchmarks, so I'd like to merge it (and also into the `array_algebra` branch) for the ooplsa benchmarks

> Is this change really necessary probably not. I understand that you can benchmark it this way, but it just becomes sort of cumbersome and slower to run than being...

cc @fredrikbk

Conversions between formats are expressed in TACO by assignments between tensors of the desired source and destination formats. Unfortunately, there aren't utilities right now in TACO to directly express the...

DO NOT MERGE