Miri: give the incremental session a chance to finish

[RalfJung]

View all comments

This helps with rust-lang/miri#5013. Reruns still aren't as fast as I'd like, but using nextest to run libcore tests shows a big difference:

before:
 Nextest run ID 56b7b355-6225-4083-96e3-aeeabbe80523 with nextest profile: default-miri
    Starting 10 tests across 2 binaries (2760 tests skipped)
        PASS [  31.264s] coretests::coretests any::any_downcast_mut
        PASS [  31.287s] coretests::coretests any::distinct_type_names
        PASS [  32.747s] coretests::coretests iter::traits::iterator::test_any
        PASS [  32.758s] coretests::coretests any::any_owning
        PASS [  32.864s] coretests::coretests any::dyn_type_name
        PASS [  32.973s] coretests::coretests any::any_downcast_ref
        PASS [  33.139s] coretests::coretests any::any_unsized
        PASS [  33.719s] coretests::coretests any::any_referenced
        PASS [  35.786s] coretests::coretests any::any_fixed_vec
        PASS [  38.951s] coretests::coretests num::dec2flt::parse::many_digits
────────────
     Summary [  38.955s] 10 tests run: 10 passed, 2760 skipped

after:
 Nextest run ID af603971-e41e-466d-8469-0425057fa325 with nextest profile: default-miri
    Starting 10 tests across 2 binaries (2761 tests skipped)
        PASS [  15.063s] coretests::coretests any::any_unsized
        PASS [  15.176s] coretests::coretests any::distinct_type_names
        PASS [  15.200s] coretests::coretests any::any_referenced
        PASS [  15.550s] coretests::coretests iter::traits::iterator::test_any
        PASS [  15.993s] coretests::coretests any::any_fixed_vec
        PASS [  17.506s] coretests::coretests num::dec2flt::parse::many_digits
        PASS [  17.902s] coretests::coretests any::any_downcast_ref
        PASS [  18.188s] coretests::coretests any::any_owning
        PASS [  19.270s] coretests::coretests any::dyn_type_name
        PASS [  19.702s] coretests::coretests any::any_downcast_mut
────────────
     Summary [  19.705s] 10 tests run: 10 passed, 2761 skipped

It still seems to spend at least 10s building the crate before Miri even begins running, no idea what it is doing in that time. But it's 15s less than before so that's good. :)

[RalfJung]

@bors try jobs=dist-x86_64-linux

[RalfJung]

@bors try jobs=dist-x86_64-linux

[rustbot]

These commits modify the Cargo.lock file. Unintentional changes to Cargo.lock can be introduced when switching branches and rebasing PRs.

If this was unintentional then you should revert the changes before this PR is merged.
Otherwise, you can ignore this comment.

miri is developed in its own repository. If the Miri part of this change can be broken out, consider making this change to rust-lang/miri instead. However, if Miri needs adjusting for rustc changes, just ignore this message.

cc @rust-lang/miri

[rustbot]

r? @nnethercote

rustbot has assigned @nnethercote.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 75 candidates
• Random selection from 19 candidates

[RalfJung]

r? @bjorn3

[rustbot]

Failed to set assignee to bjorn3: error sending request

Note: Only org members with at least the repository "read" role, users with write permissions, or people who have commented on the PR may be assigned.

[RalfJung]

It's kind of annoying that we get this output now... but warnings are rare in Miri so 🤷

View changes since the review

[RalfJung]

@bors try jobs=dist-x86_64-linux

[rust-bors]

⌛ Trying commit 19c9e78 with merge 5b05222…

(The previously running try build was automatically cancelled.)

To cancel the try build, run the command @bors try cancel.

Workflow: https://github.com/rust-lang/rust/actions/runs/31265178652

[bjorn3]

Failed to set assignee to bjorn3: error sending request

Note: Only org members with at least the repository "read" role, users with write permissions, or people who have commented on the PR may be assigned.

Github api hickup I guess.

r? bjorn3

[RalfJung]

@bors try cancel
I already tested this locally

[rust-bors]

Try build cancelled. Cancelled workflows:

• https://github.com/rust-lang/rust/actions/runs/31265178652

[RalfJung]

Okay this seems to work, thanks. :)
Miri now has its own codegen backend. That also let me remove this ad-hoc hook that you added to the dummy backend to make it work for Miri.

[RalfJung]

Looks to be finally ready :)