Nicolas Vasilache

Autotuning improvements

1

This PR isolates a small subset from @ttheodor's #453. I took the commits that look independent, fix clear bugs or implement simple new behavior and do not touch autotuner-inl.h (which...

Support -mpcu flag and autodetect to trigger vectorization on proper vector lengths

1

This PR passes proper llvm::TargetMachine information in llvm_jit and codegen_llvm by introducing a proper TargetMachine at the LLVMJit level and avoids introducing adhoc objects. The TargetMachine is constructed either from...

[WIP] Exp/rnn

2

WIP cc @apaszke

[WIP][DO NOT MERGE] Experimental vector types

1

This is a WIP experiment, please do not review. I am looking on some feedback on how to best propagate vector types through Halide following up on the discussion from...

[DO NOT MERGE] Repro maybe uninitialized variable warning triggerring error appearing with TC_CHECK

1

For repro purposes only (@ftynse @skimo-openhub ) This reverts commit 02b7b057c986430b35c7a214a7e665f944ae535b.

Atm we only pass TC, input tensors and options.

The current [implementation](https://github.com/facebookresearch/TensorComprehensions/blob/master/tc/autotuner/parameters.cc#L333) is not portable, let's use device queried information to filter. Also make it backends-specific and split ```parameters.cc```, not in #307 though. Related to #354

#354 exhibits potential issues with size selections Revisit after #307 and tinker with the tuner.

Less strict TC + JIT requirements

2

Currently TC requires to compile a new fully-specialized version for each new tensor size. #327 has some context about usage in the C2 case: > What if dimensions changed? You...