Retry tests #3229
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,102 @@ | ||
| #!/bin/bash | ||
|
|
||
| # Will catch exit code 5 when tests are deselected from previous passing run | ||
| # (relevent for --last-failed-no-failures none) | ||
| last_failed_no_failures_code=5 | ||
|
|
||
| # functions shared across test files | ||
| run_tests() { | ||
| # Set defaults | ||
| local core_args="-vvv tests/ignite" | ||
| local cache_dir=".unknown-cache" | ||
| local skip_distrib_tests=1 | ||
| local match_tests_expression="" | ||
| local trap_deselected_exit_code=1 | ||
| local use_last_failed=0 | ||
| local use_coverage=0 | ||
| local world_size=0 | ||
| # Always clean up pytest.ini | ||
| trap 'rm -f pytest.ini' RETURN | ||
| # Parse arguments | ||
| while [[ $# -gt 0 ]] | ||
| do | ||
| key="$1" | ||
| case $key in | ||
| --core_args) | ||
| core_args="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --cache_dir) | ||
| cache_dir="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --skip_distrib_tests) | ||
| skip_distrib_tests="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --match_tests_expression) | ||
| match_tests_expression="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --trap_deselected_exit_code) | ||
| trap_deselected_exit_code="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --use_last_failed) | ||
| use_last_failed="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --use_coverage) | ||
| use_coverage="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| --world_size) | ||
| world_size="$2" | ||
| shift | ||
| shift | ||
| ;; | ||
| *) | ||
| echo "Error: Unknown argument $key" | ||
| exit 1 | ||
| shift | ||
| ;; | ||
| esac | ||
| done | ||
|
|
||
| if [ "${skip_distrib_tests}" -eq "1" ]; then | ||
| # can be overwritten by core_args | ||
| skip_distrib_opt="-m 'not distributed and not tpu and not multinode_distributed'" | ||
| else | ||
| skip_distrib_opt="" | ||
| fi | ||
|
|
||
|
|
||
| echo [pytest] > pytest.ini ; echo "cache_dir=${cache_dir}" >> pytest.ini | ||
|
|
||
| # Assemble options for the pytest command | ||
| pytest_args="${skip_distrib_opt} ${core_args} --treat-unrun-as-failed -k '${match_tests_expression}'" | ||
| if [ "${use_last_failed:-0}" -eq "1" ] && [ -d "${cache_dir}" ]; then | ||
| pytest_args="--last-failed --last-failed-no-failures none ${pytest_args}" | ||
| fi | ||
| if [ "${use_coverage}" -eq "1" ]; then | ||
| pytest_args="--cov ignite --cov-append --cov-report term-missing --cov-report xml ${pytest_args}" | ||
| fi | ||
| if [ ! "${world_size}" -eq "0" ]; then | ||
| export WORLD_SIZE="${world_size}" | ||
| pytest_args="--dist=each --tx ${WORLD_SIZE}*popen//python=python ${pytest_args}" | ||
| fi | ||
|
|
||
| # Run the command | ||
| if [ "$trap_deselected_exit_code" -eq "1" ]; then | ||
| CUDA_VISIBLE_DEVICES="" eval "pytest ${pytest_args}" || { exit_code=$?; if [ "$exit_code" -eq ${last_failed_no_failures_code} ]; then echo "All tests deselected"; else exit $exit_code; fi; } | ||
|
There was a problem hiding this comment. why do we need There was a problem hiding this comment. I added an eval here because of some bugs I was running into where things like "-k ''" ends up as "-k" in the final command. The horrors of using eval are somewhat mitigated by the "-x" bash flag so that bugs in the command can be spotted more quickly. I think consistently using arrays for assembling commands in bash is a better alternative but I think using python is the best long term solution. |
||
| else | ||
| CUDA_VISIBLE_DEVICES="" eval "pytest ${pytest_args}" | ||
| fi | ||
| } | ||
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I wonder whether it would be more simple to write this script in python for later maintenance instead of a bash script and how much effort it would take?
If you think this is feasible we can do that in a follow-up PR
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it would be easy enough: largely just providing a simple cli, setting environment variables, and assembling commands to run as a subprocess. A few hours probably. Perhaps a few more to iron out issues and add some tests for it.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, sounds good, let's make a python script instead of this bash script in a follow-up PR