netdev - Re: [bug?] r8169: hangs under heavy load

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 for Android: free password hash cracker in your pocket

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

Date:	Sat, 26 Nov 2011 05:07:03 +0100
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Francois Romieu <romieu@...zoreil.com>
Cc:	Jonathan Nieder <jrnieder@...il.com>, netdev@...r.kernel.org,
	nic_swsd@...ltek.com, linux-kernel@...r.kernel.org,
	Armin Kazmi <armin.kazmi@...dortmund.de>,
	Gerd <booster@...ke7.net>
Subject: Re: [bug?] r8169: hangs under heavy load

Le samedi 26 novembre 2011 à 01:44 +0100, Francois Romieu a écrit :
> Eric Dumazet <eric.dumazet@...il.com> :
> > Le vendredi 25 novembre 2011 ?? 23:22 +0100, Francois Romieu a ??crit :
> [...]
> > > This path is not used.
> > 
> > OK, then we receive a RxFIFOOver indication while napi handler is
> > running (quite possible if machine under network load)
> 
> Welcome to the post 92fc43b4159b518f5baae57301f26d770b0834c9 world. :o/
> 
> The hack below could help (the XID matches RTL_GIGA_MAC_VER_22):
> 
> diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
> index 6f06aa1..4fcfb3f 100644
> --- a/drivers/net/ethernet/realtek/r8169.c
> +++ b/drivers/net/ethernet/realtek/r8169.c
> @@ -5818,7 +5818,6 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
>  			switch (tp->mac_version) {
>  			/* Work around for rx fifo overflow */
>  			case RTL_GIGA_MAC_VER_11:
> -			case RTL_GIGA_MAC_VER_22:
>  			case RTL_GIGA_MAC_VER_26:
>  				netif_stop_queue(dev);
>  				rtl8169_tx_timeout(dev);
> @@ -5835,6 +5834,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
>  			case RTL_GIGA_MAC_VER_31:
>  			/* Experimental science. Pktgen proof. */
>  			case RTL_GIGA_MAC_VER_12:
> +			case RTL_GIGA_MAC_VER_22:
>  			case RTL_GIGA_MAC_VER_25:
>  				if (status == RxFIFOOver)
>  					goto done;
> 
> 
> Traffic may stop flowing though. The patch below should be better:
> 

Yes, this is a workaround, and following _is_ the way to go...




> diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
> index 6f06aa1..54be9e1 100644
> --- a/drivers/net/ethernet/realtek/r8169.c
> +++ b/drivers/net/ethernet/realtek/r8169.c
> @@ -5820,8 +5820,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
>  			case RTL_GIGA_MAC_VER_11:
>  			case RTL_GIGA_MAC_VER_22:
>  			case RTL_GIGA_MAC_VER_26:
> -				netif_stop_queue(dev);
> -				rtl8169_tx_timeout(dev);
> +				rtl8169_schedule_work(dev, rtl8169_reset_task);
>  				goto done;
>  			/* Testers needed. */
>  			case RTL_GIGA_MAC_VER_17:
> 

Please Jonathan, Gerd, try this fix !


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html