From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <v.shpilevoy@tarantool.org>
Received: from smtp52.i.mail.ru (smtp52.i.mail.ru [94.100.177.112])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
 (No client certificate requested)
 by dev.tarantool.org (Postfix) with ESMTPS id 17ADA452566
 for <tarantool-patches@dev.tarantool.org>;
 Sat,  2 Nov 2019 19:21:09 +0300 (MSK)
References: <71d3e28eadd9ca54d5f564ed25d4276d472d5e6d.1572432961.git.imeevma@gmail.com>
 <20191031152306.GD2636@atlas>
 <05dc9a5b-f3c8-37f9-3dd7-63cfc966bf11@tarantool.org>
From: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
Message-ID: <4be44137-f646-ed77-d3aa-494cb887dcf3@tarantool.org>
Date: Sat, 2 Nov 2019 17:26:49 +0100
MIME-Version: 1.0
In-Reply-To: <05dc9a5b-f3c8-37f9-3dd7-63cfc966bf11@tarantool.org>
Content-Type: text/plain; charset="utf-8"
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Subject: Re: [Tarantool-patches] [PATCH v1 1/1] netbox: fix memleak in
 lbox_tuple_format_new()
List-Id: Tarantool development patches <tarantool-patches.dev.tarantool.org>
List-Unsubscribe: <https://lists.tarantool.org/mailman/options/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=unsubscribe>
List-Archive: <https://lists.tarantool.org/pipermail/tarantool-patches/>
List-Post: <mailto:tarantool-patches@dev.tarantool.org>
List-Help: <mailto:tarantool-patches-request@dev.tarantool.org?subject=help>
List-Subscribe: <https://lists.tarantool.org/mailman/listinfo/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=subscribe>
To: Imeev Mergen <imeevma@tarantool.org>, Kirill Yukhin <kyukhin@tarantool.org>
Cc: tarantool-patches@freelists.org, tarantool-patches@dev.tarantool.org

Hi! Thanks for the fix!

LGTM.

On 31/10/2019 17:43, Imeev Mergen wrote:
> 
> On 10/31/19 6:23 PM, Konstantin Osipov wrote:
>> * imeevma@tarantool.org <imeevma@tarantool.org> [19/10/30 14:03]:
>>> Memleak appeared due to the fact that when creating a dictionary
>>> its reference count is 1 from the beginning. Later, when the
>>> dictionary is used to create tuple_format, its reference counter
>>> increased by 1 and it became equal to 2. After removing
>>> tuple_format, the reference counter for dict decreased by one, so
>>> it became equal to 1. Since ref counter is not equal to 0, dict
>>> not deleted, causing a memory leak.
>>>
>>> Closes #4588
>> I wonder if you could come up with a test case? Is memory used by
>> dict accounted in any stats?
>>
> I did not find anything in box, and I did not see anything about
> the statistics in the tuple_dictionary_new() function. Doesn't
> look like we accounting this. At the moment, I can’t come up with
> a test.
> 
> You think we should add this in stats?
> 

There is no accounting for the heap memory. We tried to add it via
symbol replacement, but it didn't work properly. I think we can
omit a test here.