Project

General

Profile

Actions

Bug #1815

closed

assertion in lwkt_reltoken

Added by ftigeot about 14 years ago. Updated about 14 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
Start date:
Due date:
% Done:

0%

Estimated time:

Description

With a recent -2.7, I get this panic no more than 2 minutes after boot:

assertion: ref >= &td->td_toks_base && ref->tr_tok == tok in lwkt_reltoken

A core dump is available at this URL:
http://www.wolfpond.org/crash.dfly/

Actions #1

Updated by dillon about 14 years ago

:With a recent 2.7, I get this panic no more than 2 minutes after boot:
:
:assertion: ref >= &td
>td_toks_base && ref->tr_tok == tok in lwkt_reltoken
:
:A core dump is available at this URL:
:http://www.wolfpond.org/crash.dfly/
:
:--
:Francois Tigeot

If this was related to a ^S/^Q on a tty or pty it has just now
been fixed.
-Matt
Matthew Dillon
<>
Actions #2

Updated by ftigeot about 14 years ago

On Wed, Sep 01, 2010 at 01:03:41AM -0700, Matthew Dillon wrote:

:With a recent 2.7, I get this panic no more than 2 minutes after boot:
:
:assertion: ref >= &td
>td_toks_base && ref->tr_tok == tok in lwkt_reltoken
:
:A core dump is available at this URL:
:http://www.wolfpond.org/crash.dfly/

If this was related to a ^S/^Q on a tty or pty it has just now
been fixed.

I didn't have a chance to type these key sequences.

Howewer, most of the crashes occurred just after launching X, so it may
indeed have been related to a tty issue.

Actions #3

Updated by elekktretterr about 14 years ago

On Wed, Sep 01, 2010 at 01:03:41AM -0700, Matthew Dillon wrote:

I didn't have a chance to type these key sequences.

Howewer, most of the crashes occurred just after launching X, so it may
indeed have been related to a tty issue.

This is till happening.
System panics the very moment X tries to start on tty8.

Petr

Actions #4

Updated by elekktretterr about 14 years ago

On Wed, Sep 01, 2010 at 01:03:41AM -0700, Matthew Dillon wrote:

I didn't have a chance to type these key sequences.

Howewer, most of the crashes occurred just after launching X, so it may
indeed have been related to a tty issue.

This is till happening.
System panics the very moment X tries to start on tty8.

Petr

A small update on this panic. It also happens when just switching between
virtual terminals with ALT+F(2,3,4,5 etc).

Petr

Actions #5

Updated by dillon about 14 years ago

:
:> I didn't have a chance to type these key sequences.
:>
:> Howewer, most of the crashes occurred just after launching X, so it may
:> indeed have been related to a tty issue.
:
:This is till happening.
:System panics the very moment X tries to start on tty8.
:
:Petr

Ok, I'll try to reproduce w/ my workstation.  I may be offline for
a bit :-)
-Matt
Matthew Dillon
<>
Actions #6

Updated by dillon about 14 years ago

These particular issues should now be resolved though I don't
guarantee that a few more cases are waiting in the wings.

-Matt
Actions #7

Updated by ftigeot about 14 years ago

On Wed, Sep 01, 2010 at 11:00:18AM -0700, Matthew Dillon wrote:

These particular issues should now be resolved though I don't
guarantee that a few more cases are waiting in the wings.

No more troubles for me.

I must add that the GUI feels very fast now, much faster than 2.6 on the
same hardware. Latency must really have decreased with the latest mpsafe
commits.

Kudos !

Actions #8

Updated by dillon about 14 years ago

:No more troubles for me.
:
:I must add that the GUI feels very fast now, much faster than 2.6 on the
:same hardware. Latency must really have decreased with the latest mpsafe
:commits.
:
:Kudos !
:
:--
:Francois Tigeot

It feels faster to me too on my workstation.  I think a good chunk
of the reason is that the system isn't allowing held tokens to leak
across transitions to userland any more.
-Matt
Matthew Dillon
<>
Actions

Also available in: Atom PDF