TIMEOUT: igammac_float64_20_20_float64_20_20_chlo.mlir.test

[apivovarov]

What happened?

igammac_float64_20_20_float64_20_20_chlo.mlir.test timed out if executed along with all other tests

bazel test //...
TIMEOUT: //stablehlo/testdata:igammac_float64_20_20_float64_20_20_chlo.mlir.test (Summary)

However, when run individually, it passes. Total time: 88 sec. Testing time: 53 sec

bazel test //stablehlo/testdata:igammac_float64_20_20_float64_20_20_chlo.mlir.test
INFO: Elapsed time: 88.440s, Critical Path: 78.23s
INFO: 2 processes: 2 processwrapper-sandbox.
INFO: Build completed successfully, 2 total actions
//stablehlo/testdata:igammac_float64_20_20_float64_20_20_chlo.mlir.test  PASSED in 53.6s

Steps to reproduce your issue

bazel test //...

TIMEOUT: //stablehlo/testdata:igammac_float64_20_20_float64_20_20_chlo.mlir.test (Summary)

Version information

e44958c

[GleasonK]

Feel free to filter this testpoint using:

// RUN-DISABLED(#2590, timeout): stablehlo-opt --chlo-pre-serialization-pipeline -inline %s | stablehlo-translate --interpret

Also if you get to this (i'll try to get this done today or tomorrow), could you update igamma_float64_20_20_float64_20_20_chlo.mlir to use the same DISABLED tag, will investigate both as a part of this ticket.

I'm not sure what's going on, but I think something with Bazel is extremely slow with the interpreter. The igamma test on a cmake took something like 1s, but on bazel 80s.. I need to look into what's going on, but don't want this failure blocking you in the meantime.

[GleasonK]

Update: Filtered tests to this bug ticket in #2586