* odd gnome-terminal behavior in 2.5.67-mm3
@ 2003-04-24 19:26 David Mansfield
2003-04-25 13:24 ` James Strandboge
0 siblings, 1 reply; 3+ messages in thread
From: David Mansfield @ 2003-04-24 19:26 UTC (permalink / raw)
To: Andrew Morton, linux-kernel
Hi Andrew, list
I've been experiencing some odd behavior running 2.5.67-mm3 on my RH9
based desktop.
It's probably an application bug, but something strange is happening
anyway that doesn't happen in 'stable' kernels.
What happens is that gnome-terminal gets stuck in some sort of 'infinite
loop' when a lot of output is going to the screen and also keypresses are
going in (like paging through a large file - holding down pgup/pgdown).
Xterm doesn't seem to be affected.
If I kill off the shell running inside the terminal, it recovers (that is,
when gnome-terminal is running as a 'factory' the window that is hung
disappears but the process and all of the other windows are fine).
I captured some 'strace' of this while it was happening (sorry about the
wrapping, the lines are quite long):
ioctl(3, FIONREAD, [0]) = 0
poll([{fd=17, events=POLLIN}, {fd=3, events=POLLIN}, {fd=4,
events=POLLIN}, {fd=8, events=POLLIN}, {fd=10, events=POLLIN|POLLPRI},
{fd=11, events=POLLIN|POLLPRI}, {fd=12, ev
ents=POLLIN|POLLPRI}, {fd=13, events=POLLIN|POLLPRI}, {fd=15,
events=POLLIN|POLLPRI}, {fd=14, events=POLLIN|POLLPRI}], 10, 0) = 0
write(3, "+\20\1\0", 4) = 4
read(3, "\1\1-\205\0\0\0\0&\0\300\3\0\0\0\0\1\0\0\0\0\0\0\0\370"..., 32) =
32
ioctl(3, FIONREAD, [0]) = 0
write(3, "5\20\4\0.\356)\2q\266#\2\24\0\25\0007\0\5\0/\356)\2.\356"...,
328) = 328
read(3, "\1\1A\205\0\0\0\0&\0\300\3\0\0\0\0\0\0\0\0\0\0\0\0\370"..., 32) =
32
ioctl(3, FIONREAD, [0]) = 0
poll([{fd=17, events=POLLIN}, {fd=3, events=POLLIN}, {fd=4,
events=POLLIN}, {fd=8, events=POLLIN}, {fd=10, events=POLLIN|POLLPRI},
{fd=11, events=POLLIN|POLLPRI}, {fd=12, ev
ents=POLLIN|POLLPRI}, {fd=13, events=POLLIN|POLLPRI}, {fd=15,
events=POLLIN|POLLPRI}, {fd=14, events=POLLIN|POLLPRI}, {fd=22,
events=POLLIN}, {fd=23, events=POLLIN}, {fd=21,
events=POLLIN, revents=POLLIN}, {fd=24, events=POLLIN}, {fd=16,
events=POLLIN}, {fd=25, events=POLLIN}], 16, 0) = 1
read(21, 0xbfffe8f0, 4294965249) = -1 EFAULT (Bad address)
getpid() = 31085
write(2, "\n** (gnome-terminal:31085): WARN"..., 79) = 79
This exact sequence repeats ad infinitum.
The gnome-terminal is apparently doing an invalid read of -1 bytes after
poll is saying there is data available.
I don't know what this could mean.
David
--
/==============================\
| David Mansfield |
| lkml@dm.cobite.com |
\==============================/
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: odd gnome-terminal behavior in 2.5.67-mm3
2003-04-25 13:24 ` James Strandboge
@ 2003-04-25 13:06 ` Antonio Vargas
0 siblings, 0 replies; 3+ messages in thread
From: Antonio Vargas @ 2003-04-25 13:06 UTC (permalink / raw)
To: James Strandboge; +Cc: Linux Kernel Mailing List
On Fri, Apr 25, 2003 at 09:24:18AM -0400, James Strandboge wrote:
> On Thu, 2003-04-24 at 15:26, David Mansfield wrote:
> > Hi Andrew, list
> >
> > I've been experiencing some odd behavior running 2.5.67-mm3 on my RH9
> > based desktop.
> >
> > It's probably an application bug, but something strange is happening
> > anyway that doesn't happen in 'stable' kernels.
> >
> > What happens is that gnome-terminal gets stuck in some sort of 'infinite
> > loop' when a lot of output is going to the screen and also keypresses are
> > going in (like paging through a large file - holding down pgup/pgdown).
> >
> > Xterm doesn't seem to be affected.
>
> I'll chime in and mention that I've seen this too, and also doing a
> paste operation via highlight and middle click doesn't work in 2.5
> either. I assumed it was a libvte bug.
I've experienced this also when running "make menuconfig"
against 2.5.66 while running 2.5.66 on a redhat9 system.
I filled a bug-report on bugzilla.rehat.com and
they told me "bug is reproduced here. workaround is to
minimize the gnome-terminal while there is intense
terminal activity"... so I'm supposed to config my
kernel while it's terminal is not visible... great :)
I'd not mind trying to debug gnome-terminal if
I didn't need to install bazillons of libs
for his recompile...
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: odd gnome-terminal behavior in 2.5.67-mm3
2003-04-24 19:26 odd gnome-terminal behavior in 2.5.67-mm3 David Mansfield
@ 2003-04-25 13:24 ` James Strandboge
2003-04-25 13:06 ` Antonio Vargas
0 siblings, 1 reply; 3+ messages in thread
From: James Strandboge @ 2003-04-25 13:24 UTC (permalink / raw)
To: Linux Kernel Mailing List
On Thu, 2003-04-24 at 15:26, David Mansfield wrote:
> Hi Andrew, list
>
> I've been experiencing some odd behavior running 2.5.67-mm3 on my RH9
> based desktop.
>
> It's probably an application bug, but something strange is happening
> anyway that doesn't happen in 'stable' kernels.
>
> What happens is that gnome-terminal gets stuck in some sort of 'infinite
> loop' when a lot of output is going to the screen and also keypresses are
> going in (like paging through a large file - holding down pgup/pgdown).
>
> Xterm doesn't seem to be affected.
I'll chime in and mention that I've seen this too, and also doing a
paste operation via highlight and middle click doesn't work in 2.5
either. I assumed it was a libvte bug.
Jamie
--
James Strandboge
Targeted Performance Partners, LLC
Web: http://www.tpptraining.com
E-mail: jamie@tpptraining.com
Tel: (585) 271-8370
Fax: (585) 271-8373
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-04-25 21:49 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-04-24 19:26 odd gnome-terminal behavior in 2.5.67-mm3 David Mansfield
2003-04-25 13:24 ` James Strandboge
2003-04-25 13:06 ` Antonio Vargas
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).