General protection fault with aborted ircomm FIR connection
Hi,
first of all, no proprietary modules have ever been loaded. The "Tainted" refers to "SUSE unsupported" modules. Machine is a Samsung P35 laptop (x86). Kernel is 2.6.16.21 with SUSE patches (which don't touch IRDA afaics). FIR chipset is served by nsc-ircc dongle_id=0x08.
The crash happened when I used gammu to connect to my nokia mobile phone over /dev/ircomm0. I moved the phone out of the IR beam by accident and then killed gammu with Ctrl-C while it still had the connection open. At that moment, the kernel spewed a general protection fault on me.
general protection fault: 0000 [#1] last sysfs file: /class/net/irda0/ifindex Modules linked in: bluetooth af_packet nsc_ircc xt_pkttype ipt_LOG xt_limit
cpufreq_ondemand cpufreq_userspace cpufreq_powersave speedstep_centrino freq_table ircomm_tty ircomm irda snd_pcm_oss crc_ccitt snd_mixer_oss snd_seq snd_seq_device edd asus_acpi button battery ac ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables ip6table_filter ip6_tables x_tables ipv6 apparmor aamatch_pcre nls_utf8 ntfs loop dm_mod ipw2200 snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc pcmcia ieee80211 ieee80211_crypt ide_cd cdrom firmware_class intel_agp agpgart i8xx_tco shpchp ehci_hcd uhci_hcd pci_hotplug usbcore yenta_socket rsrc_nonstatic pcmcia_core ohci1394 ieee1394 8139too mii parport_pc lp parport reiserfs fan thermal processor piix radeonfb i2c_algo_bit i2c_core ide_disk ide_core
CPU: 0 EIP: 0060:[<d530722f>] Tainted: G U VLI EFLAGS: 00010082 (2.6.16.21-0.13-default #1)
EIP is at 0xd530722f
eax: ccf59e40 ebx: ccf59e40 ecx: 00000000 edx: 00000001
esi: cab86240 edi: 00000000 ebp: ccf59e18 esp: ccf59df8
ds: 007b es: 007b ss: 0068
Process gammu (pid: 4100, threadinfo=ccf58000 task=ddd8fab0)
Stack: <0>c01140c7 00000000 00000001 d5307218 d5307200 00000246 d5307200 cbac4414
ccf59e2c c01140f7 00000000 00000000 ccb2d400 d5307224 c0221081 00000000
e1f1b313 e1f2a500 d5307200 c021ea58 d5307224 cab86240 c021f1d8 00000008
Call Trace:
[<c01140c7>] __wake_up_common+0x2e/0x4d
[<c01140f7>] __wake_up+0x11/0x1a
[<c0221081>] sock_def_wakeup+0x19/0x1b
[<e1f1b313>] irda_release+0x2f/0x109 [irda]
[<c021ea58>] sock_release+0x11/0x63
[<c021f1d8>] sock_close+0x26/0x2a
[<c014b9f9>] __fput+0x9e/0x14c
[<c0149492>] filp_close+0x4e/0x54
[<c0118b00>] put_files_struct+0x63/0xa5
[<c0119883>] do_exit+0x197/0x617
[<c0119d55>] sys_exit_group+0x0/0xd
[<c01201f5>] get_signal_to_deliver+0x34b/0x35b
[<c010220f>] do_notify_resume+0x89/0x5a0
[<c01259bc>] autoremove_wake_function+0x0/0x2d
[<c014b362>] vfs_read+0xb9/0x14d
[<c014b6bd>] sys_read+0x3c/0x63
[<c0102a8e>] work_notifysig+0x13/0x25
Code: 00 00 00 00 80 e1 f1 e1 00 00 00 00 40 62 b8 ca 00 d4 b2 cc 4c 9e f5 cc 4c 9e f5 cc 01 00 00 00 00 00 00 00 00 00 00 00 84 7e b8 <ca> ac 60 d0 df 34 75 30 d5 6c a2 4dc1 50 44 ac cb 50 44 ac cb
<1>Fixing recursive fault but reboot is needed!
If you need any further info to debug this, please tell me.
Regards,
Carl-Daniel
Hi,
Quoted 11 lines · Carl-Daniel Hailfinger
Hi,
first of all, no proprietary modules have ever been loaded. The "Tainted" refers to "SUSE unsupported" modules. Machine is a Samsung P35 laptop (x86). Kernel is 2.6.16.21 with SUSE patches (which don't touch IRDA afaics). FIR chipset is served by nsc-ircc dongle_id=0x08.
The crash happened when I used gammu to connect to my nokia mobile phone over
/dev/ircomm0. I moved the phone out of the IR beam by accident and then killed gammu with Ctrl-C while it still had the connection open. At that moment, the kernel spewed a general protection fault on me.
I think I managed to reproduce (and maybe fix) this bug. Could you please check if the following patch fixes it for you as it does for me:
net/irda/af_irda.c +7 -2
@@ -132,13 +132,14 @@ static void irda_disconnect_indication(v/* Prevent race conditions with irda_release() and irda_shutdown() */ if (!sock_flag (sk,SOCK_DEAD ) && sk->sk_state !=TCP_CLOSE ) { +lock_sock (sk); sk->sk_state =TCP_CLOSE ; sk->sk_err =ECONNRESET ; sk->sk_shutdown |=SEND_SHUTDOWN ; sk->sk_state_change (sk); -/* Uh-oh... Should use sock_orphan ? */ -sock_set_flag (sk,SOCK_DEAD ); +sock_orphan (sk); +release_sock (sk);/* Close our TSAP. * If we leave it open, IrLMP put it back into the list of
@@ -1212,6 +1213,7 @@ static int irda_release(struct socket *sif (sk ==NULL )return 0 ; +lock_sock (sk); sk->sk_state =TCP_CLOSE ; sk->sk_shutdown |=SEND_SHUTDOWN ; sk->sk_state_change (sk);
@@ -1221,6 +1223,7 @@ static int irda_release(struct socket *ssock_orphan (sk); sock->sk =NULL ; +release_sock (sk);/* Purge queues (see sock_init_data()) */ skb_queue_purge (&sk->sk_receive_queue);
@@ -1353,6 +1356,7 @@ static int irda_recvmsg_dgram(struct kioIRDA_DEBUG (4 ,"%s()\n" , __FUNCTION__); IRDA_ASSERT(self != NULL, return -1;); + IRDA_ASSERT(!sock_error(sk), return -1;); skb =skb_recv_datagram (sk, flags & ~MSG_DONTWAIT , flags &MSG_DONTWAIT , &err);
@@ -1405,6 +1409,7 @@ static int irda_recvmsg_stream(struct kiIRDA_DEBUG (3 ,"%s()\n" , __FUNCTION__); IRDA_ASSERT(self != NULL, return -1;); + IRDA_ASSERT(!sock_error(sk), return -1;);if (sock->flags & __SO_ACCEPTCON)return (-EINVAL );
Signature · 1 line
VGER BF report: U 0.499581