[inductor] optimize isa dry compile time.

[xuhancn]

Fixes #100378
Original issue caused by startup dry compile need cost almost 1 second.

This PR add compiler version info, isa build options and pytorch version info to the test binary path hash.
So same compile, same isa and same pytorch can skip the dry compile.

Local test:
First time:

We need to compile all c++ modules and it cost 16.5s.

Second time:

We skipped dry compile due to the same isa fingerprint. It is only cost 0.36s.

cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/124602

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit fe29f70 with merge base 87079f5 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[xuhancn]

@pytorchbot rebase

[pytorchmergebot]

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

[pytorchmergebot]

Successfully rebased xu_optimize_isa_dry_compile onto refs/remotes/origin/viable/strict, please pull locally before adding more changes (for example, via git checkout xu_optimize_isa_dry_compile && git pull --rebase)

[ezyang]

I guess we can land this as is, but you're still doing multiple(!) subprocess calls, and I would hope that there was some way to architect this where those were not needed either.

[xuhancn]

I guess we can land this as is, but you're still doing multiple(!) subprocess calls, and I would hope that there was some way to architect this where those were not needed either.

Here has some lru_cache to cache result. What optimization you still want to do?

[xuhancn]

3e09156 @ezyang I add a cache to pick_vec_isa and it would reduce the callee function run.

[xuhancn]

@malfet Any comment to me for next step?

[ezyang]

lru_cache doesn't help, you need a cache that persists across processes.

[xuhancn]

lru_cache doesn't help, you need a cache that persists across processes.

I guess you means mutiple process call pick_vec_isa to get isa for cpp build.
I think what I can do is that write the isa info to the temp file. But still a problem, we always need to get compiler version information to ensure the compiler's capability can cover the isa level. At the same time, get compile version still need to call subprocess. So I think we maybe not have potential to reduce call subprocess.

[xuhancn]

@ezyang we can land this PR firstly.
If any other optimize idea let create new PR to land.

[ezyang]

ci fails look real

[xuhancn]

ci fails look real

Let me debug for it.

[xuhancn]

ci fails look real

@ezyang the CI was failed as the reason that lru_cache for pick_vec_isa cache the value not match the test case expected result. After revert the related code, it works well.

[xuhancn]

@pytorchmergebot merge

[pytorchmergebot]

Merge failed

Reason: This PR needs a release notes: label
If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

Details for Dev Infra team

Raised by workflow job

[xuhancn]

@pytorchbot label "topic: not user facing"

[xuhancn]

@pytorchmergebot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here