This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Should glibc be fully reentrant? -- No. (Roland was right).
- From: OndÅej BÃlka <neleai at seznam dot cz>
- To: Florian Weimer <fweimer at redhat dot com>
- Cc: Carlos O'Donell <carlos at redhat dot com>, GNU C Library <libc-alpha at sourceware dot org>, Roland McGrath <roland at hack dot frob dot com>, Siddhesh Poyarekar <siddhesh at redhat dot com>
- Date: Thu, 11 Dec 2014 14:33:53 +0100
- Subject: Re: Should glibc be fully reentrant? -- No. (Roland was right).
- Authentication-results: sourceware.org; auth=none
- References: <5488F9C6 dot 9080605 at redhat dot com> <54898FE0 dot 8060701 at redhat dot com>
On Thu, Dec 11, 2014 at 01:36:48PM +0100, Florian Weimer wrote:
> On 12/11/2014 02:56 AM, Carlos O'Donell wrote:
> >After serious review it seems like we will need to start by
> >saying that an interposed malloc must operate as-if it were
> >in async-signal context and use only those functions which
> >are marked as async-signal-safe.
>
> Practically speaking, this seems rather restrictive. Most mallocs
> need locking, and implementations may want to use pthread mutexes.
> Those are not async-signal-safe, and I doubt they can be made
> reentrant, either.
>
No, making them reentrant is relatively simple as I wrote before, just
wrap calls in something
static __thread int recursed
if (recursed)
return allocate_by_mmap();
else
{
recursed=1;
normal;
recursed=0;
}
When freeing/reallocing simplest way is leak memory or put it into
atomic list. Also for performance you need to place recursed field into
already existing allocator per-thread structure.