On Wed, 2011-10-12 at 14:05 +0300, Matrosov Dmitriy wrote: > Package: linux-2.6 > Version: 2.6.32-38 > Severity: normal > Tags: squeeze > > Steps to reproduce: > 1. Reboot computer. > 2. Boot into runlevel 2. > 3. Switch to console and go to single-user runlevel: > # telinit 1 > 4. Enter root password and switch to runlevel 2 again: > # telinit 2 > 5. And now, as i understand, second init copy (which run by telinit) will > crash. > > I can't reproduce this again without rebooting system, i.e. it occurs only > after reboot and only once. > Also, i can't reproduce this with previous kernel (=2.6.32-35squeeze2), and, > though, i don't sure that this is kernel bug, but it seems, like last kernel > upgrade (=2.6.32-38) triggers it.
This isn't a crash. [...] > [ 75.608119] ------------[ cut here ]------------ > [ 75.608136] WARNING: at > /build/buildd-linux-2.6_2.6.32-38-i386-G9QRBd/linux-2.6-2.6.32/debian/build/source_i386_none/drivers/char/tty_ldisc.c:738 > tty_ldisc_reinit+0x3e/0x6c() > [ 75.608144] Hardware name: 6460EDG > [ 75.608148] Modules linked in: acpi_cpufreq cpufreq_stats > cpufreq_powersave cpufreq_conservative cpufreq_userspace pppoe pppox > ppp_generic slhc binfmt_misc uinput fuse ipt_REJECT ipt_LOG xt_limit > xt_tcpudp xt_state iptable_filter ipt_MASQUERADE iptable_nat nf_nat > nf_conntrack_ipv4 nf_conntrack nf_defrag_ipv4 iptable_mangle ip_tables > x_tables ext3 jbd ext2 tp_smapi thinkpad_ec loop firewire_sbp2 > snd_hda_codec_analog btusb snd_hda_intel bluetooth arc4 snd_hda_codec sg > snd_hwdep ecb iwlagn snd_pcm nouveau iwlcore sr_mod sdhci_pci sdhci ttm > snd_seq mac80211 snd_timer firewire_ohci cdrom drm_kms_helper snd_seq_device > joydev firewire_core mmc_core cfg80211 drm pcmcia yenta_socket snd > thinkpad_acpi soundcore rsrc_nonstatic i2c_algo_bit i2c_i801 wmi uhci_hcd > pcmcia_core ricoh_mmc crc_itu_t snd_page_alloc ehci_hcd i2c_core ata_piix > ata_generic e1000e rfkill led_class nvram usbcore nls_base processor button > ac battery psmouse evdev pcspkr video output serio_raw ext4 mbcache jbd2 crc > 16 dm_mod sd_mod crc_t10dif ahci libata scsi_mod thermal thermal_sys [last > unloaded: kvm] > [ 75.608352] Pid: 2422, comm: init Not tainted 2.6.32-5-686-bigmem #1 > [ 75.608357] Call Trace: > [ 75.608369] [<c1036819>] ? warn_slowpath_common+0x5e/0x8a > [ 75.608378] [<c103684f>] ? warn_slowpath_null+0xa/0xc > [ 75.608386] [<c11a38b9>] ? tty_ldisc_reinit+0x3e/0x6c > [ 75.608394] [<c11a39d6>] ? tty_ldisc_hangup+0xef/0x172 > [ 75.608403] [<c10a5c14>] ? remove_vma+0x43/0x48 > [ 75.608413] [<c119e136>] ? do_tty_hangup+0xd0/0x2b1 > [ 75.608422] [<c119e60d>] ? disassociate_ctty+0x3b/0x1b3 > [ 75.608430] [<c103961a>] ? do_exit+0x212/0x5cd > [ 75.608439] [<c1039a34>] ? do_group_exit+0x5f/0x82 > [ 75.608449] [<c10449bf>] ? get_signal_to_deliver+0x30d/0x32e > [ 75.608459] [<c10075b4>] ? do_notify_resume+0x6f/0x713 > [ 75.608469] [<c110a800>] ? security_task_wait+0xc/0xd > [ 75.608478] [<c103902e>] ? wait_consider_task+0x983/0x990 > [ 75.608488] [<c104a804>] ? remove_wait_queue+0xb/0x2f > [ 75.608496] [<c10391ea>] ? do_wait+0x1af/0x1e3 > [ 75.608504] [<c10392b6>] ? sys_wait4+0x98/0xa2 > [ 75.608514] [<c1037c44>] ? child_wait_callback+0x0/0x5f > [ 75.608522] [<c10392d3>] ? sys_waitpid+0x13/0x17 > [ 75.608530] [<c10082f4>] ? work_notifysig+0x13/0x1b > [ 75.608536] ---[ end trace a550edb008c670a3 ]--- [...]. This warning message was added by the bug fix "TTY: ldisc, do not close until there are readers". I don't believe that the warning a regression; it's just reporting a problem that already existed. If you would like to help us understand and fix the problem, please can you test the attached patch, following the instructions at <http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-official>. It should replace the warnings with log lines of the form "tty_ldisc_reinit: refcount of <name> is still <number>" which you should send back. Ben. -- Ben Hutchings Quantity is no substitute for quality, but it's the only one we've got.
From: Jiri Slaby <jsl...@suse.cz>
Date: Mon, 8 Aug 2011 16:00:59 +0200
Subject: [PATCH] TTY: tty_ldisc_wait_idle debug
Signed-off-by: Jiri Slaby <jsl...@suse.cz>
[bwh: Change file path for 2.6.32]
---
drivers/tty/tty_ldisc.c | 9 ++++++++-
1 files changed, 8 insertions(+), 1 deletions(-)
diff --git a/drivers/tty/tty_ldisc.c b/drivers/tty/tty_ldisc.c
index 512c49f..0b90801 100644
--- a/drivers/char/tty_ldisc.c
+++ b/drivers/char/tty_ldisc.c
@@ -26,6 +26,7 @@
#include <linux/bitops.h>
#include <linux/delay.h>
#include <linux/seq_file.h>
+#include <linux/ratelimit.h>
#include <linux/uaccess.h>
#include <asm/system.h>
@@ -758,11 +759,17 @@ static void tty_reset_termios(struct tty_struct *tty)
static int tty_ldisc_reinit(struct tty_struct *tty, int ldisc)
{
struct tty_ldisc *ld = tty_ldisc_get(ldisc);
+ int ret;
if (IS_ERR(ld))
return -1;
- WARN_ON_ONCE(tty_ldisc_wait_idle(tty));
+ ret = tty_ldisc_wait_idle(tty);
+ if (ret)
+ printk_ratelimited(KERN_DEBUG "%s: refcount of %ps is still %d\n",
+ __func__,
+ tty->ldisc->ops,
+ atomic_read(&tty->ldisc->users));
tty_ldisc_close(tty, tty->ldisc);
tty_ldisc_put(tty->ldisc);
--
1.7.6
signature.asc
Description: This is a digitally signed message part
