Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Eliminate crash for nested guards in record initialization #8196

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.

Sign up for GitHub

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

Jump to bottom
Merged
bjorng merged 1 commit into from
Mar 5, 2024
Merged

Eliminate crash for nested guards in record initialization #8196

bjorng merged 1 commit into from
Mar 5, 2024

Conversation

@bjorng
Copy link
Contributor

@bjorng bjorng commented Feb 27, 2024

Closes #8190

@bjorng bjorng added team:VM Assigned to OTP team VM fix labels Feb 27, 2024
@bjorng bjorng self-assigned this Feb 27, 2024
@bjorng bjorng added the testing currently being tested, tag is used by OTP internal CI label Feb 27, 2024
@github-actions
Copy link
Contributor

github-actions bot commented Feb 27, 2024 •
edited
Loading

CT Test Results

    2 files     93 suites   34m 34s ⏱️
2 018 tests 1 970 ✅ 48 💤 0 ❌
2 327 runs  2 277 ✅ 50 💤 0 ❌

Results for commit 1b23ddb.

♻️ This comment has been updated with latest results.

To speed up review, make sure that you have read Contributing to Erlang/OTP and that all checks pass.

See the TESTING and DEVELOPMENT HowTo guides for details about how to run test locally.

Artifacts

// Erlang/OTP Github Action Bot

@RobinMorisset
Copy link
Contributor

RobinMorisset commented Feb 27, 2024

Thanks for the fix.
It reduced by 100x the number of failing testcases, but there is one that remains and it changed its error message:

-record(r0, {rf0}).
-record(r1, {
    rf1 = fun
        (0.0) ->
            (list_to_bitstring(ok))#r0.rf0;
        (_) when ok ->
            ok
    end
}).

f() ->
    <<0 || (#r1{})>>.

still fails, and now its error message is

Function: f/0
/home/rmorisset/minimized/test7949.erl: internal error in pass core_to_ssa:
exception error: no match of right hand side value {b_remote,{b_literal,erlang},{b_literal,list_to_bitstring},1}
  in function  beam_core_to_ssa:call_cg/5 (beam_core_to_ssa.erl, line 2788)
  in call from beam_core_to_ssa:guard_cg/3 (beam_core_to_ssa.erl, line 2726)
  in call from beam_core_to_ssa:guard_cg/3 (beam_core_to_ssa.erl, line 2720)
  in call from beam_core_to_ssa:guard_clause_cg/3 (beam_core_to_ssa.erl, line 2701)
  in call from beam_core_to_ssa:match_cg/3 (beam_core_to_ssa.erl, line 2403)
  in call from beam_core_to_ssa:match_cg/3 (beam_core_to_ssa.erl, line 2404)
  in call from beam_core_to_ssa:cg/2 (beam_core_to_ssa.erl, line 2357)
  in call from beam_core_to_ssa:cg_fun/2 (beam_core_to_ssa.erl, line 2319)

I don't know whether it is a related bug, or a different one that was masked by the issue in erl_expand_records.

@bjorng bjorng force-pushed the bjorn/fix-erl_expand_records/GH-8190 branch from fa50647 to c8dd54f Compare February 28, 2024 06:04
@bjorng
Copy link
Contributor Author

bjorng commented Feb 28, 2024

I've updated the pull request to fix the second bug too.

@RobinMorisset
Copy link
Contributor

RobinMorisset commented Feb 28, 2024

Two other related crashes were found overnight:

-record(r0, {rf0}).
-record(r1, {
    rf0 = {
        ((ok)#r0{})#r0.rf0,
        case ok of
            _ when ok ->
                ok
        end
    }
}).

f() ->
    #{ok => ok || (#r1{})}.

and

-record(r0, {rf0}).
-record(r1, {
    rf0 =
        (begin
            ((_ = ok)#r0.rf0),
            (try
                ok
            catch
                _ when ok ->
                    ok
            end)
        end)
}).

f() ->
    #{ok => ok || (#r1{})}.

crash with this PR with

Function: f/0
/home/rmorisset/minimized/test185701.erl: internal error in pass core:
exception error: no function clause matching v3_core:unforce_tree(
... [very big expression I'm omitting] ...
)
  in function  v3_core:unforce/3 (v3_core.erl, line 593)
  in call from v3_core:force_booleans/4 (v3_core.erl, line 545)
  in call from v3_core:gexpr/3 (v3_core.erl, line 370)
  in call from v3_core:gexpr_top/2 (v3_core.erl, line 357)
  in call from v3_core:guard/2 (v3_core.erl, line 344)
  in call from v3_core:clause/2 (v3_core.erl, line 306)
  in call from v3_core:clauses/2 (v3_core.erl, line 298)

Without the PR, they crash with the same error message as the original testcase.

This was referenced Feb 28, 2024
Closed
Closed
@bjorng
Eliminate crash for nested guards in record initialization
1b23ddb 
Also fix other bugs in the `erl_expand_records` pass hidden behind the
original bug.

Closes erlang#8190
@bjorng bjorng force-pushed the bjorn/fix-erl_expand_records/GH-8190 branch from c8dd54f to 1b23ddb Compare March 4, 2024 05:23
@bjorng
Copy link
Contributor Author

bjorng commented Mar 4, 2024

I've force-pushed another update to fix the last two crashes.

@bjorng bjorng merged commit b18a791 into erlang:master Mar 5, 2024
@bjorng bjorng deleted the bjorn/fix-erl_expand_records/GH-8190 branch March 5, 2024 11:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

@bjorng bjorng

Labels

fix team:VM Assigned to OTP team VM testing currently being tested, tag is used by OTP internal CI

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

[erlc] internal error in erl_expand_records:in_guard/1

2 participants

@bjorng @RobinMorisset