linux-kernel - Re: [PATCH v2] drivers: s390: net: ctcm: migrate variables to handle y2038 problem

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Fri, 31 Oct 2014 14:38:35 +0100
From:	Ursula Braun <ubraun@...ux.vnet.ibm.com>
To:	Aya Mahfouz <mahfouz.saif.elyazal@...il.com>
Cc:	ursula.braun@...ibm.com, blaschka@...ux.vnet.ibm.com,
	schwidefsky@...ibm.com, heiko.carstens@...ibm.com,
	linux390@...ibm.com, linux-s390@...r.kernel.org,
	linux-kernel@...r.kernel.org, arnd@...db.de,
	opw-kernel@...glegroups.com
Subject: Re: [PATCH v2] drivers: s390: net: ctcm: migrate variables to
 handle y2038 problem

thanks, Aya. I miss drivers/s390/net/ctcm_main.c in your patch. And
"time_before()" is not the right function to use here, because it
returns a Bool value. Purpose of the ctcm timespec stuff is to determine
the time interval between triggering the send operation and receiving
the confirmation of the send operation. The maximum value is saved and
can be displayed in the statistics.

Regards, Ursula Braun

On Fri, 2014-10-31 at 11:18 +0200, Aya Mahfouz wrote:
> This patch is concerned with migrating the time variables for the s390
> network driver. The changes handle the y2038 problem where timespec will
> overflow in the year 2038. timespec was replaced by unsigned long and
> all time variables get their values from the jiffies global variable.
> This was done for the sake of speed and efficiency.
> 
> Signed-off-by: Aya Mahfouz <mahfouz.saif.elyazal@...il.com>
> ---
> v1: Arnd has advised me to provide you with options for time
>     calculation. The first option: "accuracy" is used in this
>     patch. The second option: "speed" can be done through
>     jiffies.
> 
> v2: Moved on to the speed option. Let me know if I explicitly
>     need to include the jiffies header. The module compiles with
>     no problems on my side.
> 
>  drivers/s390/net/ctcm_fsms.c | 18 +++++++-----------
>  drivers/s390/net/ctcm_main.h |  2 +-
>  2 files changed, 8 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/s390/net/ctcm_fsms.c b/drivers/s390/net/ctcm_fsms.c
> index fb92524..34b9820 100644
> --- a/drivers/s390/net/ctcm_fsms.c
> +++ b/drivers/s390/net/ctcm_fsms.c
> @@ -251,13 +251,11 @@ static void chx_txdone(fsm_instance *fi, int event, void *arg)
>  	int first = 1;
>  	int i;
>  	unsigned long duration;
> -	struct timespec done_stamp = current_kernel_time(); /* xtime */
> +	unsigned long done_stamp = jiffies;
> 
>  	CTCM_PR_DEBUG("%s(%s): %s\n", __func__, ch->id, dev->name);
> 
> -	duration =
> -	    (done_stamp.tv_sec - ch->prof.send_stamp.tv_sec) * 1000000 +
> -	    (done_stamp.tv_nsec - ch->prof.send_stamp.tv_nsec) / 1000;
> +	duration = time_before(done_stamp, ch->prof.send_stamp);
>  	if (duration > ch->prof.tx_time)
>  		ch->prof.tx_time = duration;
> 
> @@ -307,7 +305,7 @@ static void chx_txdone(fsm_instance *fi, int event, void *arg)
>  		spin_unlock(&ch->collect_lock);
>  		ch->ccw[1].count = ch->trans_skb->len;
>  		fsm_addtimer(&ch->timer, CTCM_TIME_5_SEC, CTC_EVENT_TIMER, ch);
> -		ch->prof.send_stamp = current_kernel_time(); /* xtime */
> +		ch->prof.send_stamp = jiffies;
>  		rc = ccw_device_start(ch->cdev, &ch->ccw[0],
>  						(unsigned long)ch, 0xff, 0);
>  		ch->prof.doios_multi++;
> @@ -1229,14 +1227,12 @@ static void ctcmpc_chx_txdone(fsm_instance *fi, int event, void *arg)
>  	int		rc;
>  	struct th_header *header;
>  	struct pdu	*p_header;
> -	struct timespec done_stamp = current_kernel_time(); /* xtime */
> +	unsigned long done_stamp = jiffies;
> 
>  	CTCM_PR_DEBUG("Enter %s: %s cp:%i\n",
>  			__func__, dev->name, smp_processor_id());
> 
> -	duration =
> -		(done_stamp.tv_sec - ch->prof.send_stamp.tv_sec) * 1000000 +
> -		(done_stamp.tv_nsec - ch->prof.send_stamp.tv_nsec) / 1000;
> +	duration = time_before(done_stamp, ch->prof.send_stamp);
>  	if (duration > ch->prof.tx_time)
>  		ch->prof.tx_time = duration;
> 
> @@ -1361,7 +1357,7 @@ static void ctcmpc_chx_txdone(fsm_instance *fi, int event, void *arg)
> 
>  	ch->ccw[1].count = ch->trans_skb->len;
>  	fsm_addtimer(&ch->timer, CTCM_TIME_5_SEC, CTC_EVENT_TIMER, ch);
> -	ch->prof.send_stamp = current_kernel_time(); /* xtime */
> +	ch->prof.send_stamp = jiffies;
>  	if (do_debug_ccw)
>  		ctcmpc_dumpit((char *)&ch->ccw[0], sizeof(struct ccw1) * 3);
>  	rc = ccw_device_start(ch->cdev, &ch->ccw[0],
> @@ -1827,7 +1823,7 @@ static void ctcmpc_chx_send_sweep(fsm_instance *fsm, int event, void *arg)
>  	fsm_newstate(wch->fsm, CTC_STATE_TX);
> 
>  	spin_lock_irqsave(get_ccwdev_lock(wch->cdev), saveflags);
> -	wch->prof.send_stamp = current_kernel_time(); /* xtime */
> +	wch->prof.send_stamp = jiffies;
>  	rc = ccw_device_start(wch->cdev, &wch->ccw[3],
>  					(unsigned long) wch, 0xff, 0);
>  	spin_unlock_irqrestore(get_ccwdev_lock(wch->cdev), saveflags);
> diff --git a/drivers/s390/net/ctcm_main.h b/drivers/s390/net/ctcm_main.h
> index 477c933..6f4417c 100644
> --- a/drivers/s390/net/ctcm_main.h
> +++ b/drivers/s390/net/ctcm_main.h
> @@ -121,7 +121,7 @@ struct ctcm_profile {
>  	unsigned long doios_multi;
>  	unsigned long txlen;
>  	unsigned long tx_time;
> -	struct timespec send_stamp;
> +	unsigned long send_stamp;
>  };
> 
>  /*


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/