Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Likely incorrect bisection result produced #123

Closed
camelid opened this issue Dec 17, 2020 · 3 comments
Closed

Likely incorrect bisection result produced #123

camelid opened this issue Dec 17, 2020 · 3 comments

Comments

@camelid
Copy link
Member

camelid commented Dec 17, 2020

As seen in rust-lang/rust#80101 (comment), cargo-bisect-rustc seems to have produced an incorrect result. Additionally, it seemed to be installing a commit then skipped to a different one and then another without printing any information. And it checked a commit twice (cc #85).

Here's the full log from my usage of cargo-bisect-rustc in case it helps in diagnosing the problem:

(pnkfelix note: Of course, everywhere you see ==> No is a spot where the author chose "mark baseline", and ==> Yes means the choice was "mark regressed")

cargo bisect-rustc --preserve --prompt --start=2020-11-17 --end=2020-11-20 -- doc --document-private-items
installing nightly-2020-11-17
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 8.72s


nightly-2020-11-17 finished with exit code Some(0).
please select an action to take:
RESULT: nightly-2020-11-17, ===> No

installing nightly-2020-11-20
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 8.04s


nightly-2020-11-20 finished with exit code Some(0).
please select an action to take:
RESULT: nightly-2020-11-20, ===> Yes

installing nightly-2020-11-18
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 10.66s


nightly-2020-11-18 finished with exit code Some(0).
please select an action to take:
RESULT: nightly-2020-11-18, ===> No

installing nightly-2020-11-19
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 9.87s


nightly-2020-11-19 finished with exit code Some(0).
please select an action to take:
RESULT: nightly-2020-11-19, ===> Yes

searched toolchains nightly-2020-11-17 through nightly-2020-11-20


********************************************************************************
Regression in nightly-2020-11-19
********************************************************************************

fetching https://static.rust-lang.org/dist/2020-11-18/channel-rust-nightly-git-commit-hash.txt
nightly manifest 2020-11-18: 40 B / 40 B [================] 100.00 % 43.49 KB/s converted 2020-11-18 to c919f490bbcd2b29b74016101f7ec71aaa24bdbb
fetching https://static.rust-lang.org/dist/2020-11-19/channel-rust-nightly-git-commit-hash.txt
nightly manifest 2020-11-19: 40 B / 40 B [=================] 100.00 % 1.14 MB/s converted 2020-11-19 to 8256379832b5ecb7f71e8c5e2018446482223c12
looking for regression commit between 2020-11-18 and 2020-11-19
opening existing repository at "/Users/user/rust"
refreshing repository
fetching (via local git) commits from c919f490bbcd2b29b74016101f7ec71aaa24bdbb to 8256379832b5ecb7f71e8c5e2018446482223c12
opening existing repository at "/Users/user/rust"
refreshing repository
looking up first commit
looking up second commit
checking that commits are by bors and thus have ci artifacts...
finding bors merge commits
found 6 bors merge commits in the specified range
  commit[0] 2020-11-17UTC: Auto merge of #79138 - m-ou-se:rollup-owel5ld, r=m-ou-se
  commit[1] 2020-11-18UTC: Auto merge of #79132 - pietroalbini:beta-next, r=Mark-Simulacrum
  commit[2] 2020-11-18UTC: Auto merge of #77820 - jyn514:from-inner, r=petrochenkov
  commit[3] 2020-11-18UTC: Auto merge of #79159 - pietroalbini:woops, r=pietroalbini
  commit[4] 2020-11-18UTC: Auto merge of #79167 - m-ou-se:rollup-4g15apk, r=m-ou-se
  commit[5] 2020-11-18UTC: Auto merge of #78995 - Nadrieril:clean-empty-match, r=varkor
validated commits found, specifying toolchains

installing c919f490bbcd2b29b74016101f7ec71aaa24bdbb
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 10.60s


c919f490bbcd2b29b74016101f7ec71aaa24bdbb finished with exit code Some(0).
please select an action to take:
RESULT: c919f490bbcd2b29b74016101f7ec71aaa24bdbb, ===> No

installing 8256379832b5ecb7f71e8c5e2018446482223c12
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 13.33s


8256379832b5ecb7f71e8c5e2018446482223c12 finished with exit code Some(0).
please select an action to take:
RESULT: 8256379832b5ecb7f71e8c5e2018446482223c12, ===> Yes

installing c4f836ad1aceb83507810d9499f56988fd24578d
installing 87776d7d5322422e5239e153e793b687f7f9c379
installing 7d747db0d5dd8f08f2efb073e2e77a34553465a7
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 8.63s


7d747db0d5dd8f08f2efb073e2e77a34553465a7 finished with exit code Some(0).
please select an action to take:
RESULT: 7d747db0d5dd8f08f2efb073e2e77a34553465a7, ===> Yes

installing c919f490bbcd2b29b74016101f7ec71aaa24bdbb
testing...
 Documenting foo v0.1.0 (/Users/user/rust-issues/rust-issue-80101)
    Finished dev [unoptimized + debuginfo] target(s) in 8.58s


c919f490bbcd2b29b74016101f7ec71aaa24bdbb finished with exit code Some(0).
please select an action to take:
RESULT: c919f490bbcd2b29b74016101f7ec71aaa24bdbb, ===> No

searched toolchains c919f490bbcd2b29b74016101f7ec71aaa24bdbb through 8256379832b5ecb7f71e8c5e2018446482223c12


********************************************************************************
Regression in 7d747db0d5dd8f08f2efb073e2e77a34553465a7
********************************************************************************

==================================================================================
= Please file this regression report on the rust-lang/rust GitHub repository     =
=        New issue: https://github.com/rust-lang/rust/issues/new                 =
=     Known issues: https://github.com/rust-lang/rust/issues                     =
= Copy and paste the text below into the issue report thread.  Thanks!           =
==================================================================================

searched nightlies: from nightly-2020-11-17 to nightly-2020-11-20
regressed nightly: nightly-2020-11-19
searched commits: from https://github.com/rust-lang/rust/commit/c919f490bbcd2b29b74016101f7ec71aaa24bdbb to https://github.com/rust-lang/rust/commit/8256379832b5ecb7f71e8c5e2018446482223c12
regressed commit: https://github.com/rust-lang/rust/commit/7d747db0d5dd8f08f2efb073e2e77a34553465a7

<details>
<summary>bisected with <a href='https://github.com/rust-lang/cargo-bisect-rustc'>cargo-bisect-rustc</a> v0.6.0</summary>


Host triple: x86_64-apple-darwin
Reproduce with:
```bash
cargo bisect-rustc --preserve --prompt --start=2020-11-17 --end=2020-11-20 -- doc --document-private-items 
```
</details>

Note that the issue seems to be when bisecting CI artifacts; I think it's right about which nightly the issue is in. Also note that I was manually testing with --prompt because I was bisecting a visual rustdoc regression. However, I bisected three times and got the same result, so I think it's unlikely that it's human error.

@camelid
Copy link
Member Author

camelid commented Dec 17, 2020

Seems like it's potentially an off-by-one error since the commit that actually regressed was one later than the one cargo-bisect-rustc thought – the correct one was rust-lang/rust@c4f836a.

@dtolnay
Copy link
Member

dtolnay commented Aug 15, 2021

I agree something is off-by-one. This came up as well in #134 (comment).

@ehuss
Copy link
Collaborator

ehuss commented Oct 8, 2021

I'm going to close, as I suspect this is likely caused by #113. I published version 0.6.1, feel free to reopen if it continues with the newer version.

@ehuss ehuss closed this as completed Oct 8, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants