Re: [patch] pthread(3) + malloc(3) init model

[Christos Zoulas <christos%zoulas.com@localhost>]

Looks clean to me. If it works, let's try it. We can always put it back.

christos

> On Feb 15, 2020, at 8:00 AM, Kamil Rytarowski <n54%gmx.com@localhost> wrote:
> 
> Signed PGP part
> I propose to separate the pthread_atfork(3) call from pthread_tsd_init()
> and move it into a distinct function.
> 
> I propose to call late TSD initialization after "pthread_atfork(NULL,
> NULL, pthread__fork_callback);" from pthread__init().
> 
> This change:
> 
> 1. Stops initializing jemalloc prematurely and unintentionally.
> 2. Eliminates '#if 0' hacks in pthread_mutex.c.
> 3. Restores control when to initialize a malloc implementation.
> 
> http://netbsd.org/~kamil/patch-00229-pthread_malloc_init_model.txt
> 
> No regressions are observed.
> 
> 
> <sanitizer.log>

Attachment: signature.asc
Description: Message signed with OpenPGP