From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <v.shpilevoy@tarantool.org>
Received: from smtp43.i.mail.ru (smtp43.i.mail.ru [94.100.177.103])
 (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 26C3D445320
 for <tarantool-patches@dev.tarantool.org>;
 Wed, 15 Jul 2020 02:51:20 +0300 (MSK)
References: <1594221263-6228-1-git-send-email-alyapunov@tarantool.org>
 <1594221263-6228-12-git-send-email-alyapunov@tarantool.org>
From: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
Message-ID: <198d4c8b-0e6b-9cd0-1d80-8d40f6c68081@tarantool.org>
Date: Wed, 15 Jul 2020 01:51:18 +0200
MIME-Version: 1.0
In-Reply-To: <1594221263-6228-12-git-send-email-alyapunov@tarantool.org>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Subject: Re: [Tarantool-patches] [PATCH 11/16] tx: introduce conflict tracker
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: Aleksandr Lyapunov <alyapunov@tarantool.org>, tarantool-patches@dev.tarantool.org

Thanks for the patch!

See 1 comment below.

On 08.07.2020 17:14, Aleksandr Lyapunov wrote:
> ---
>  src/box/txn.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  src/box/txn.h |  5 +++
>  2 files changed, 103 insertions(+)
> 
> diff --git a/src/box/txn.c b/src/box/txn.c
> index 2bd44a3..b9f1737 100644
> --- a/src/box/txn.c
> +++ b/src/box/txn.c
> @@ -998,3 +1053,46 @@ void
>  tx_manager_free()
>  {
>  }
> +
> +int
> +txm_cause_conflict(struct txn *wreaker, struct txn *victim)
> +{
> +	struct tx_conflict_tracker *tracker = NULL;
> +	struct rlist *r1 = wreaker->conflict_list.next;
> +	struct rlist *r2 = wreaker->conflicted_by_list.next;
> +	while (r1 != &wreaker->conflict_list &&
> +	       r2 != &wreaker->conflicted_by_list) {
> +		tracker = rlist_entry(r1, struct tx_conflict_tracker,
> +				      in_conflict_list);
> +		if (tracker->wreaker == wreaker && tracker->victim == victim)
> +			break;
> +		tracker = rlist_entry(r2, struct tx_conflict_tracker,
> +				      in_conflicted_by_list);
> +		if (tracker->wreaker == wreaker && tracker->victim == victim)
> +			break;
> +		tracker = NULL;
> +		r1 = r1->next;
> +		r2 = r2->next;
> +	}
> +	if (tracker != NULL) {
> +		/* Move to the beginning of a list
> +		 * for a case of subsequent lookups */
> +		rlist_del(&tracker->in_conflict_list);
> +		rlist_del(&tracker->in_conflicted_by_list);
> +	} else {
> +		size_t size;
> +		tracker = region_alloc_object(&victim->region,
> +					      struct tx_conflict_tracker,
> +					      &size);
> +		if (tracker == NULL) {
> +			diag_set(OutOfMemory, size, "tx region",
> +				 "conflict_tracker");
> +			return -1;
> +		}
> +	}
> +	tracker->wreaker = wreaker;
> +	tracker->victim = victim;

Why don't you save anything into the victim? How will it be
removed from the lists if it will commit earlier than 'wreaker'?