Actually there is not enough parallelism in the computation of each cell. You can see “ Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks” for more details.
--- [Visit Topic](https://discuss.tvm.apache.org/t/comparison-between-tiramisu-and-tvm-and-halide/933/9) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/453d531a63225b7bbd7063149c2f41a27ba7b4f545e4031962e513d04cd327c8).