Re: [PATCH] Make xmalloc and xrealloc thread-safe

From: Fredrik Kuivinen <frekui@gmail.com>
To: Nicolas Pitre <nico@fluxnic.net>
Cc: Shawn Pearce <spearce@spearce.org>,
	Junio C Hamano <gitster@pobox.com>,
	git@vger.kernel.org, Johannes Sixt <j6t@kdbg.org>
Subject: Re: [PATCH] Make xmalloc and xrealloc thread-safe
Date: Wed, 31 Mar 2010 08:57:55 +0200	[thread overview]
Message-ID: <s2t4c8ef71003302357x5e9defa1l4dbde6391d533ca5@mail.gmail.com> (raw)
In-Reply-To: <alpine.LFD.2.00.1003271035360.694@xanadu.home>

On Sat, Mar 27, 2010 at 19:59, Nicolas Pitre <nico@fluxnic.net> wrote:
> On Sat, 27 Mar 2010, Fredrik Kuivinen wrote:
>
>> On Wed, Mar 24, 2010 at 21:22, Nicolas Pitre <nico@fluxnic.net> wrote:
>> > +static void try_to_free_from_threads(size_t size)
>> > +{
>> > +       read_lock();
>> > +       release_pack_memory(size, -1);
>> > +       read_unlock();
>> > +}
>> > +
>>
>> Will this really work in all cases? In the find_deltas -> try_delta ->
>> read_sha1_file -> ... -> xmalloc call path, the mutex is already
>> locked when we get to xmalloc.
>
> You're right.  Damn.

A simple fix is to make it a recursive mutex instead. This will work
with a minimal change in win32 as well as the CRITICAL_SECTION type is
recursive.

I guess the downside is that the locking potentially gets slightly slower.

- Fredrik