Re: lib/54017: jemalloc deadlock?

[Jason Thorpe <thorpej%me.com@localhost>]

> On Mar 9, 2019, at 4:10 PM, matthew green <mrg%eterna.com.au@localhost> wrote:
> 
> the multiple definition problem seems to indicate that joerg's
> idea of hiding the stats functions by default is going to be
> the right solution.

The only symbols exposed by the malloc library should be the ones described by the relevant standards we conform to.  Any of the optional stuff specific to a particular malloc implementation should be relegated to underscore-oblivion.

> all these crashes makes me think that new jemalloc should be
> reverted in -current and delayed until netbsd-9 happens, so
> that we have heaps of time to find all the issues, rather
> than shipping a damaged or delayed netbsd-9.

+1

-- thorpej