(switched to email. Please respond via emailed reply-to-all, not via the
bugzilla web interface).
I assume this is an atheros driver problem, not a core networking
problem..
On Tue, 2 Sep 2008 12:07:51 -0700 (PDT)
[email protected] wrote:
> http://bugzilla.kernel.org/show_bug.cgi?id=11483
>
> Summary: Raw socket tx/rx program for "ath0" shows "sending: : No
> buffer space available"
> Product: Networking
> Version: 2.5
> KernelVersion: Linux 2.6.26.3 armv5teb
> Platform: All
> OS/Version: Linux
> Tree: Mainline
> Status: NEW
> Severity: high
> Priority: P1
> Component: IPV4
> AssignedTo: [email protected]
> ReportedBy: [email protected]
>
>
> Latest working kernel version:Linux 2.6.26.3 armv5teb
> Earliest failing kernel version:Linux 2.6.26.3 armv5teb
That is contradictory. What we're asking is "is this a regression? If
so, when did we break it?" I don't think the atheros driver is old
enough to have regressed.
> Distribution:OpenWrt Kamikaze 7.09, madwifi
> Hardware Environment: GW2348-4 board with ARM (IXP425 processor) + Ubiquiti SR5
> radios at miniPCI slots
> Software Environment:Socket programming with Iperf
> Problem Description:
> Message, "sending: : No buffer space available", is shown when bulk IP traffic
> is transmitted or received, using raw socket transmission program
>
> Also, some kernel network module crash message is displayed. Please find the
> attached crash message encountered
>
> Steps to reproduce:
> 1. start the User level socket program for routing custom IP packets, (by using
> raw transmission and reception socket) in "ath0" interface.
> 2. Now start the testing using Iperf for testing routing of the user program
>
>
> --
> Also refer to the attachment
>
: in mcmp_snmp_smtx function
:
: SNMP Packets sending..........
: SNMP Packets sending. over.........NETDEV WATCHDOG: wifi1: transmit timed out
: ------------[ cut here ]------------
: WARNING: at net/sched/sch_generic.c:223 ()
: Modules linked in: prism54 orinoco_plx orinoco_pci ipw2200 ipw2100 orinoco hermes ath_pci wlan_xauth wlan_wep wlan_tkip wlan_ccmp wlan_acl ath_rate_minstrel ath_hal(P) wlan_scan_ap wlan nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp ipt_TTL xt_MARK ipt_ECN xt_CLASSIFY ipt_ttl xt_time ipt_time xt_tcpmss xt_statistic xt_mark xt_mac xt_length ipt_ecn xt_DSCP xt_dscp ipt_LOG xt_CHAOS xt_DELUDE xt_TARPIT xt_quota xt_portscan xt_pkttype xt_physdev iptable_raw ip6t_REJECT ip6t_mh ip6t_ah ip6table_raw ip6_queue ip6table_filter ip6_tables nf_conntrack_ipv6 ppp_async ppp_generic slhc crc_ccitt ath9k usbcore mac80211 cfg80211 ipv6 ieee80211_crypt_ccmp ieee80211_crypt_tkip ieee80211_crypt_wep ieee80211 ieee80211_crypt michael_mic arc4 aes_generic deflate ecb cbc crypto_blkcipher crypto_hash cryptomgr crypto_algapi [last unloaded: wlan_scan_sta]
: Function entered at [<c0025890>] from [<c0032128>]
: Function entered at [<c00320e4>] from [<c0152c00>]
: r6:c0221aa0 r5:c3090000 r4:c022c45c
: Function entered at [<c0152b54>] from [<c003b13c>]
: r5:c0152b54 r4:00000100
: Function entered at [<c003aff4>] from [<c0036ff4>]
: r8:0001cea8 r7:c0206058 r6:0000000a r5:c02218ac r4:00000001
: Function entered at [<c0036f98>] from [<c0037380>]
: r6:00000000 r5:c020fc8c r4:00000005
: Function entered at [<c003733c>] from [<c0021048>]
: Function entered at [<c0021000>] from [<c0021664>]
: Exception stack(0xc0203f64 to 0xc0203fac)
: 3f60: c3d48000 c3c1a8a0 a0000013 60000013 c0022ae4 c0202000 c001eef8
: 3f80: c0206058 0001cea8 690541c2 0001cd3c c0203fc0 c0203fac c0203fac c00229a8
: 3fa0: c00229a8 60000013 ffffffff
: r6:00000020 r5:0000001f r4:ffffffff
: Function entered at [<c0022970>] from [<c01bb3a4>]
: r5:c021c230 r4:c0224310
: Function entered at [<c01bb350>] from [<c0008b80>]
: Function entered at [<c00088fc>] from [<00008034>]
: ---[ end trace 9e368de2b60a35e4 ]---
:
I suspect that testing 2.6.27-rc5 would be useful - quite a few fixes
are flowing into that driver.
Thanks.
On Tue, Sep 02, 2008 at 12:29:38PM -0700, Andrew Morton wrote:
>
> (switched to email. Please respond via emailed reply-to-all, not via the
> bugzilla web interface).
>
> I assume this is an atheros driver problem, not a core networking
> problem..
ath_pci -- looks like madwifi to me...
--
John W. Linville
[email protected]
2008/9/2 John W. Linville <[email protected]>:
> On Tue, Sep 02, 2008 at 12:29:38PM -0700, Andrew Morton wrote:
>>
>> (switched to email. Please respond via emailed reply-to-all, not via the
>> bugzilla web interface).
>>
>> I assume this is an atheros driver problem, not a core networking
>> problem..
>
> ath_pci -- looks like madwifi to me...
>
> wifi1: transmit timed out
MadWiFi indeed, normaly ath5k would handle this card (AR5213 + RF5112A
without 2Ghz part) but ath5k isn't loaded.
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
On Tue, Sep 2, 2008 at 12:41 PM, John W. Linville
<[email protected]> wrote:
> On Tue, Sep 02, 2008 at 12:29:38PM -0700, Andrew Morton wrote:
>>
>> (switched to email. Please respond via emailed reply-to-all, not via the
>> bugzilla web interface).
>>
>> I assume this is an atheros driver problem, not a core networking
>> problem..
>
> ath_pci -- looks like madwifi to me...
I see ath_pci but I also see ath9k, and a whole bunch of other
wireless drivers loaded (strange). Anyway though the user reports the
card used is an Ubiquity SR5. These are based on Atheros AR5004
(802.11a) so ath9k cannot be used anyway, this makes it a MadWifi
issue.
Luis
On Tue, Sep 02, 2008 at 12:29:38PM -0700, Andrew Morton wrote:
>
> (switched to email. Please respond via emailed reply-to-all, not via the
> bugzilla web interface).
>
> I assume this is an atheros driver problem, not a core networking
> problem..
>
> On Tue, 2 Sep 2008 12:07:51 -0700 (PDT)
> [email protected] wrote:
>
> > http://bugzilla.kernel.org/show_bug.cgi?id=11483
> >
> > Summary: Raw socket tx/rx program for "ath0" shows "sending: : No
> > buffer space available"
> > Product: Networking
> > Version: 2.5
> > KernelVersion: Linux 2.6.26.3 armv5teb
> > Platform: All
> > OS/Version: Linux
> > Tree: Mainline
> > Status: NEW
> > Severity: high
> > Priority: P1
> > Component: IPV4
> > AssignedTo: [email protected]
> > ReportedBy: [email protected]
> >
> >
> > Latest working kernel version:Linux 2.6.26.3 armv5teb
> > Earliest failing kernel version:Linux 2.6.26.3 armv5teb
>
> That is contradictory. What we're asking is "is this a regression? If
> so, when did we break it?" I don't think the atheros driver is old
> enough to have regressed.
>
>
> > Distribution:OpenWrt Kamikaze 7.09, madwifi
> > Hardware Environment: GW2348-4 board with ARM (IXP425 processor) + Ubiquiti SR5
> > radios at miniPCI slots
> > Software Environment:Socket programming with Iperf
> > Problem Description:
> > Message, "sending: : No buffer space available", is shown when bulk IP traffic
> > is transmitted or received, using raw socket transmission program
> >
> > Also, some kernel network module crash message is displayed. Please find the
> > attached crash message encountered
> >
> > Steps to reproduce:
> > 1. start the User level socket program for routing custom IP packets, (by using
> > raw transmission and reception socket) in "ath0" interface.
> > 2. Now start the testing using Iperf for testing routing of the user program
> >
> >
> > --
> > Also refer to the attachment
> >
>
> : in mcmp_snmp_smtx function
> :
> : SNMP Packets sending..........
> : SNMP Packets sending. over.........NETDEV WATCHDOG: wifi1: transmit timed out
> : ------------[ cut here ]------------
> : WARNING: at net/sched/sch_generic.c:223 ()
> : Modules linked in: prism54 orinoco_plx orinoco_pci ipw2200 ipw2100 orinoco hermes ath_pci wlan_xauth wlan_wep wlan_tkip wlan_ccmp wlan_acl ath_rate_minstrel ath_hal(P) wlan_scan_ap wlan nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp ipt_TTL xt_MARK ipt_ECN xt_CLASSIFY ipt_ttl xt_time ipt_time xt_tcpmss xt_statistic xt_mark xt_mac xt_length ipt_ecn xt_DSCP xt_dscp ipt_LOG xt_CHAOS xt_DELUDE xt_TARPIT xt_quota xt_portscan xt_pkttype xt_physdev iptable_raw ip6t_REJECT ip6t_mh ip6t_ah ip6table_raw ip6_queue ip6table_filter ip6_tables nf_conntrack_ipv6 ppp_async ppp_generic slhc crc_ccitt ath9k usbcore mac80211 cfg80211 ipv6 ieee80211_crypt_ccmp ieee80211_crypt_tkip ieee80211_crypt_wep ieee80211 ieee80211_crypt michael_mic arc4 aes_generic deflate ecb cbc crypto_blkcipher crypto_hash cryptomgr crypto_algapi [last unloaded: wlan_scan_sta]
> : Function entered at [<c0025890>] from [<c0032128>]
> : Function entered at [<c00320e4>] from [<c0152c00>]
> : r6:c0221aa0 r5:c3090000 r4:c022c45c
> : Function entered at [<c0152b54>] from [<c003b13c>]
> : r5:c0152b54 r4:00000100
> : Function entered at [<c003aff4>] from [<c0036ff4>]
> : r8:0001cea8 r7:c0206058 r6:0000000a r5:c02218ac r4:00000001
> : Function entered at [<c0036f98>] from [<c0037380>]
> : r6:00000000 r5:c020fc8c r4:00000005
> : Function entered at [<c003733c>] from [<c0021048>]
> : Function entered at [<c0021000>] from [<c0021664>]
> : Exception stack(0xc0203f64 to 0xc0203fac)
> : 3f60: c3d48000 c3c1a8a0 a0000013 60000013 c0022ae4 c0202000 c001eef8
> : 3f80: c0206058 0001cea8 690541c2 0001cd3c c0203fc0 c0203fac c0203fac c00229a8
> : 3fa0: c00229a8 60000013 ffffffff
> : r6:00000020 r5:0000001f r4:ffffffff
> : Function entered at [<c0022970>] from [<c01bb3a4>]
> : r5:c021c230 r4:c0224310
> : Function entered at [<c01bb350>] from [<c0008b80>]
> : Function entered at [<c00088fc>] from [<00008034>]
> : ---[ end trace 9e368de2b60a35e4 ]---
> :
>
> I suspect that testing 2.6.27-rc5 would be useful - quite a few fixes
> are flowing into that driver.
Die MadWifi.
Luis