JackCaoG comments

Results 401 comments of


                                            JackCaoG

PyTorch/XLA `gtest` needs to be updated

Thanks @ngam for the heads up, I will try to test after we update our tf pins.

[SPMD][PoC] compile & execute with PjRt

let's rebase again, I will try to review it today or tmr.

[SPMD][PoC] compile & execute with PjRt

you need to update this pr a bit to account for new tf update, should not be too much code changes through

[SPMD][PoC] compile & execute with PjRt

> > you need to update this pr a bit to account for new tf update, should not be too much code changes through > > Do we have a...

[SPMD][PoC] compile & execute with PjRt

Thanks, I will take a look later today

[SPMD][PoC] compile & execute with PjRt

Seems like test strict out crashed. @yeounoh were you able to run a simple pytorch/xla program locally?

[SPMD][PoC] compile & execute with PjRt

Seesm irrelevant, let me just restart the gpu ci

[SPMD][PoC] compile & execute with PjRt

I will take another pass and try to merge it.

Codegen addcdiv and addcmul

https://github.com/pytorch/pytorch/pull/82970#pullrequestreview-1065922141 fixed the device issue.

Codegen addcdiv and addcmul

Issue coming from trying to add f64 and f32 ``` 2 root error(s) found. (0) INTERNAL: during context [Unknown]: Seen floating point types of different precisions in %add.122 = f64[1,16]{1,0}...