valgrind error with num_cpus

[andrieshiemstra]

my error:

running 6 tests
test auto_id_map::tests::test_aim ... ok
test eventloop::tests::test ... ok
test eventloop::tests::test_sync ... ok
test js_utils::facades::values::tests::test_jsvf ... ok
==5283== Thread 2 task_manager::te:
==5283== Syscall param statx(file_name) points to unaddressable byte(s)
==5283==    at 0x4AFD9FE: statx (statx.c:29)
==5283==    by 0x2DA930: statx (weak.rs:136)
==5283==    by 0x2DA930: std::sys::unix::fs::try_statx (fs.rs:139)
==5283==    by 0x2CC035: file_attr (fs.rs:776)
==5283==    by 0x2CC035: metadata (fs.rs:493)
==5283==    by 0x2CC035: buffer_capacity_required (fs.rs:609)
==5283==    by 0x2CC035: <std::fs::File as std::io::Read>::read_to_string (fs.rs:645)
==5283==    by 0x27536E: num_cpus::linux::Cgroup::param (linux.rs:214)
==5283==    by 0x275219: num_cpus::linux::Cgroup::quota_us (linux.rs:203)
==5283==    by 0x2750A2: num_cpus::linux::Cgroup::cpu_quota (linux.rs:188)
==5283==    by 0x274CB1: num_cpus::linux::load_cgroups (linux.rs:149)
==5283==    by 0x27494D: num_cpus::linux::init_cgroups (linux.rs:129)
==5283==    by 0x27D3A8: core::ops::function::FnOnce::call_once (function.rs:227)
==5283==    by 0x27ED7A: std::sync::once::Once::call_once::{{closure}} (once.rs:262)
==5283==    by 0x136C27: std::sync::once::Once::call_inner (once.rs:419)
==5283==    by 0x27ED05: std::sync::once::Once::call_once (once.rs:262)
==5283==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==5283== 
==5283== Syscall param statx(buf) points to unaddressable byte(s)
==5283==    at 0x4AFD9FE: statx (statx.c:29)
==5283==    by 0x2DA930: statx (weak.rs:136)
==5283==    by 0x2DA930: std::sys::unix::fs::try_statx (fs.rs:139)
==5283==    by 0x2CC035: file_attr (fs.rs:776)
==5283==    by 0x2CC035: metadata (fs.rs:493)
==5283==    by 0x2CC035: buffer_capacity_required (fs.rs:609)
==5283==    by 0x2CC035: <std::fs::File as std::io::Read>::read_to_string (fs.rs:645)
==5283==    by 0x27536E: num_cpus::linux::Cgroup::param (linux.rs:214)
==5283==    by 0x275219: num_cpus::linux::Cgroup::quota_us (linux.rs:203)
==5283==    by 0x2750A2: num_cpus::linux::Cgroup::cpu_quota (linux.rs:188)
==5283==    by 0x274CB1: num_cpus::linux::load_cgroups (linux.rs:149)
==5283==    by 0x27494D: num_cpus::linux::init_cgroups (linux.rs:129)
==5283==    by 0x27D3A8: core::ops::function::FnOnce::call_once (function.rs:227)
==5283==    by 0x27ED7A: std::sync::once::Once::call_once::{{closure}} (once.rs:262)
==5283==    by 0x136C27: std::sync::once::Once::call_inner (once.rs:419)
==5283==    by 0x27ED05: std::sync::once::Once::call_once (once.rs:262)
==5283==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==5283== 
test cache::tests::test_cache ... ok
test task_manager::tests::test ... ok

test result: ok. 6 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 9.45s

==5283== 
==5283== HEAP SUMMARY:
==5283==     in use at exit: 1,056 bytes in 2 blocks
==5283==   total heap usage: 577 allocs, 575 frees, 97,207 bytes allocated
==5283== 
==5283== LEAK SUMMARY:
==5283==    definitely lost: 0 bytes in 0 blocks
==5283==    indirectly lost: 0 bytes in 0 blocks
==5283==      possibly lost: 0 bytes in 0 blocks
==5283==    still reachable: 1,056 bytes in 2 blocks
==5283==         suppressed: 0 bytes in 0 blocks
==5283== Reachable blocks (those to which a pointer was found) are not shown.
==5283== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==5283== 
==5283== For lists of detected and suppressed errors, rerun with: -s
==5283== ERROR SUMMARY: 2 errors from 2 contexts (suppressed: 0 from 0)
Error: Process completed with exit code 123.

look sthe same as this issue, but i'm not going to build valgrind from source in my CI, todo check in 3 or 4 months if valgrinbd 3.19 is released and fixes this...

https://users.rust-lang.org/t/simple-fs-read-to-string-program-causes-valgrind-memcheck-errors/68722