[External] Re: [RFC PATCH] move memset out of hold lock when rte_free

[Fengnan Chang]

This problem had fix in this patch
http://patches.dpdk.org/project/dpdk/patch/20230912090415.48709-1-changfengnan@bytedance.com/
I'm doing long-term test, especially rte_zmalloc.

Fengnan Chang <changfengnan at bytedance.com> 于2023年9月12日周二 10:44写道：
>
> This patch still have problem, I'll fix next version.
>
> Stephen Hemminger <stephen at networkplumber.org> 于2023年9月6日周三 23:08写道：
> >
> > On Thu, 31 Aug 2023 19:19:37 +0800
> > Fengnan Chang <changfengnan at bytedance.com> wrote:
> >
> > > +#ifndef RTE_MALLOC_DEBUG
> > > +     if (internal_conf->legacy_mem) {
> > > +             /* If orig_elem is dirty, the joint element is clean, we need do memset now */
> > > +             if (elem->orig_elem->dirty && !elem->dirty)
> > > +                     memset(ptr, 0, data_len);
> > > +     } else if (!elem->dirty) {
> > > +             memset(ptr, 0, data_len);
> > > +     }
> > > +#else
> > > +     /* Always poison the memory in debug mode. */
> > > +     memset(ptr, MALLOC_POISON, data_len);
> > > +#endif
> >
> > The code reads better if positive clause was first.
>
> Got it,  I'll do as you suggest in next version.
> > I.e.
> >
> > #ifdef RTE_MALLOC_DEBUG
> >         /* Always poison the memory in debug mode. */
> >         memset(ptr, MALLOC_POISON, data_len);
> > #else
> > ...