From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ferenc =?utf-8?b?V8OhZ25lcg==?= Subject: Re: WARNING =?utf-8?b?KHNrYl9nc29fc2VnbWVudCk=?= with tg3+bridge in 2.6.27 Date: Thu, 7 May 2009 21:23:14 +0000 (UTC) Message-ID: References: <20081011172002.GA12659@pest> <20081013.004226.171259546.davem@davemloft.net> <20081022074502.GA25058@gondor.apana.org.au> <20081022175704.GA25286@xw6200.broadcom.net> <20081023022001.GB304@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: netdev@vger.kernel.org Return-path: Received: from main.gmane.org ([80.91.229.2]:46044 "EHLO ciao.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751626AbZEGXaE (ORCPT ); Thu, 7 May 2009 19:30:04 -0400 Received: from root by ciao.gmane.org with local (Exim 4.43) id 1M2D2R-0006Fw-F4 for netdev@vger.kernel.org; Thu, 07 May 2009 23:30:03 +0000 Received: from tac.ki.iif.hu ([193.6.222.43]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 07 May 2009 23:30:03 +0000 Received: from wferi by tac.ki.iif.hu with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 07 May 2009 23:30:03 +0000 Sender: netdev-owner@vger.kernel.org List-ID: Herbert Xu gondor.apana.org.au> writes: > On Wed, Oct 22, 2008 at 10:57:04AM -0700, Matt Carlson wrote: > >> Looking at the dmesg output, I see that the tg3 device is a 5714. >> NETIF_F_TSO6 should not be enabled for the 5714. Why is the driver >> getting an IPv6 TSO packet? > > This might be caused by the bridge enabling TSO6 due to GSO > being enabled on the underlying device which is a result of > it supporting SG. Hi, Was this bug finally found and fixed? I get the enclosed warnings with eth0: Tigon3 [partno(BCM95701A10) rev 0105 PHY(5701)] (PCIX:133MHz:64-bit) 10/100/1000Base-T Ethernet 00:06:5b:f3:43:66 eth0: RXcsums[1] LinkChgREG[1] MIirq[1] ASF[0] WireSpeed[1] TSOcap[0] eth0: dma_rwctrl[76db000f] dma_mask[64-bit] cards in active/backup bonding setup, without any bridging under 2.6.26-2-i686, the Debian Lenny stock kernel. If if it was fixed, I wonder if the fix could be perhaps backported to this kernel. If not, I'd like to help finding it. I could get rid of them by turning of GSO on both physical interfaces of the bond. -- Thanks, Feri. WARNING: at net/core/dev.c:1451 skb_gso_segment+0x6b/0x14e() Modules linked in: 8021q bonding nf_conntrack_tftp ipt_recent xt_multiport ip6t_REJECT ipt_REJECT nf_conntrack_ipv6 nf_conntrack_ipv4 xt_state nf_conntrack xt_tcpudp xt_comment ip6t_LOG ipt_LOG xt_limit ip6table_filter iptable_filter ip6table_raw ip6_tables ipv6 iptable_raw ip_tables x_tables ext3 jbd mbcache ipmi_devintf ipmi_si ipmi_msghandler loop psmouse evdev serio_raw snd_pcm snd_timer snd soundcore snd_page_alloc pcspkr shpchp pci_hotplug button i2c_piix4 i2c_core rocket sworks_agp agpgart dcdbas xfs dm_mirror dm_log dm_snapshot dm_mod raid1 md_mod ide_cd_mod cdrom sg ata_generic sd_mod libata dock serverworks aic7xxx scsi_transport_spi scsi_mod floppy ide_pci_generic tg3 ohci_hcd ide_core usbcore thermal processor fan thermal_sys [last unloaded: scsi_wait_scan] Pid: 3956, comm: exim4 Not tainted 2.6.26-2-686 #1 [] warn_on_slowpath+0x40/0x66 [] tcp_packet+0x916/0x980 [nf_conntrack] [] hrtick_set+0x8f/0xd8 [] atomic_notifier_call_chain+0x9/0xc [] nf_ct_deliver_cached_events+0x3b/0x96 [nf_conntrack] [] skb_gso_segment+0x6b/0x14e [] dev_hard_start_xmit+0x1bc/0x24f [] dev_queue_xmit+0x20b/0x2bc [] neigh_resolve_output+0x1f1/0x21b [] ip6_output_finish+0x72/0xc7 [ipv6] [] ip6_output+0xa57/0xa73 [ipv6] [] nf_iterate+0x30/0x61 [] nf_hook_slow+0x3a/0x90 [] dst_output+0x0/0x7 [ipv6] [] ip6_xmit+0x2c7/0x387 [ipv6] [] inet6_csk_xmit+0x1e1/0x1f7 [ipv6] [] tcp_transmit_skb+0x6a2/0x6d5 [] tso_fragment+0xd5/0x10b [] __tcp_push_pending_frames+0x60b/0x6b2 [] tcp_sendmsg+0x7f5/0x8e3 [] sock_sendmsg+0xc7/0xe1 [] autoremove_wake_function+0x0/0x2d [] mnt_drop_write+0x1a/0xb4 [] generic_file_aio_read+0x493/0x4da [] xfs_iunlock+0x28/0x62 [xfs] [] sys_sendto+0xfc/0x127 [] xfs_file_aio_read+0x50/0x58 [xfs] [] do_sync_read+0xbf/0xfe [] getnstimeofday+0x37/0xbc [] sys_send+0x19/0x1d [] sys_socketcall+0xed/0x19e [] sysenter_past_esp+0x78/0xb1 [] acpi_pci_root_add+0xe7/0x296 ======================= ---[ end trace 8437de2d68accdc8 ]---