2.4.22-pre3-ac1 write hang

	Hi,

	I just got a new Toshiba SPro M15-S405, I have been having

problems with write to the disk. For instance I have been trying to just untar a fresh 2.4.21 to apply the pre4 to try and after I untar the archive as soon as it begin to write out to the disk our i run sync it hangs. After this when I reboot it hangs when loading the kernel right after these lines

EXT3-fs: mounted filesystem with ordered data mode.
EXT3-fs: write access will be enabled during recovery.

Now to get the os to boot again I must boot off a rescue cdrom running a bare 2.4.18 kernel, then
run e2fsck, recover the journal reboot and it comes right back up. There is nothing in the log files,
I have tried with devfs without devfs, if i am watching a movie when it hangs if the sounds is playing it gets a little static everytime before
it hangs for a half second. I thought maybe it
was the sound drivers so i switch to alsa to test with with the same result. I have run memtest
for 31 hours one and half passes with full tests
with no problems, it hangs with vesafb and without.
I can't very well use this computer for work so
I am up to run any patchs and our test you would
like to do. When it hangs outside of x tho,
i can get the sysreq help and do the reboot and off but no hard-drive actions will happen can't sync
our remount read-only. when it hangs in x you can't do anything just hit the button. the bios is fully acpi, it has apm. i switched over and turned off acpi with the same results. it has no options for resource assignment on anything but serial and printer port. There is no serial port to huck up to aswell, it has resource for it i guess for the docking station. It seems big writes and then syncing trips the event now almost everytime in the pre3 it would stay up for a day our two then lock up on a big write. Another thing that is broken is rebooting,
if i do shutdown -r now it will reboot but hang before the lilo prompt appears but if i shutdown completely then turn it on it has no problems, like it doesn't really reset maybe. At any rate it is of no use to me in this shape, so
i have no problem running patchs our tests for the cause. it has been much enjoyment messing with the last few days. please forgive my spelling and grammer.

Attached is the normal includes

Dee

I got it stable by switching from ordered data mode to journaled mode, it runs very stable now. I have Journal Block Device support turned on, maybe something about not using journal data mode with that enabled. Do all new ext3 version partitions default to ordered mode ? Happy it works now anyway :)

Dee