If a driver calls scsi_register, but then has a problem in there
detect where they need to call scsi_unregister the parent pointer of
the struct device may never be set.

 drivers/scsi/hosts.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletion(-)

This is the 2.5 version of a 2.4 patch posted to the list long
ago, the aacraid thread reminded me of it.

The problem is that certain highend arrays allow to mark a r/o
volume writeable on the fly so we have to call check_disk_change
for write-protected devices in sd_open, too.

No need to keep around the non-dma mapping code in 2.6

most callers really want GFP_KERNEL, not GFP_ATOMIC.

This is a followup to Doug's comments and older work.  It kills
sdev->tagged_queue which wasn't ever set in 2.5/2.6 except through
obscure and broken ioctls (!).  As a reason of that tagged queing
didn't work for a lot of drivers, so this does change behaviour.
Be careful..

James, can you review the code in 53c700.c?  Calling scsi_activate_tcq
in ->queuecommand rather than ->slave_configure looks rather strange to
me..

Synchronize all scanning activity, this fixes long-standing races
vs /proc/scsi/scsi and sysfs addition and deletion of devices.

Note that this does not serialize removing, the lists will get
their own locking soon.

scsi_scan_host_selected was the only prototype without 'extern'

proc_mkdir can also take absolute pathes, so we can avoid the export.

There's no more essential functionality in it so allow the
embedded folks to configure it out.

The logging level is now controlled by a
/proc/sys/dev/scsi/logging_level sysctl instead of /proc/scsi/scsi.
The format is the same as the logging_level module parameter.

Instead of hard coding the number of io ports (to the wrong value no
less) use the pci_resource_length to determine the number.

After searching for a device to free, only free it if it was found.

Clean up the initialization sequence for the adapter. Makes it easier
to follow.

Cleanup of the adapter initialization sequence. Now it's clear what
is going on and what has been done at any point. It also keeps the
initialization of various things together and not spread out over a
bunch of different functions. This then made it possible to ensure
that appropriate the resources were correctly released in the event
of failure.

This cleans up the device management. It makes the init and cleanup
seperate functions that basically do the opposite type things in
reverse order. Makes it clear which functions remove a device and
which ones free a device (or devices.)

I forgot to make some of the new functions added during the list
cleanups static. This patch just declares those new functions static.

When testing sym2 hotplug I found a few places where we need to use
__devinit, not __init.

This patch enables the advansys driver to build when
ADVANSYS_DEBUG is #defined.

patch_name:	scsi_advan_260t3.patch
patch_version:	2003-08-10.22:37:15
author:		Randy.Dunlap <rddunlap@osdl.org>
description:	enable compile with ADVANSYS_DEBUG #defined
product:	Linux
product_versions: 260-test3
URL:		_
maintainer:	unknown
diffstat:	=
 drivers/scsi/advansys.c |   13 +++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

Same as the ppa driver, keeping them in sync.

patch_name:	scsi_imm_unreg.patch
patch_version:	2003-08-19.21:14:54
author:		Randy.Dunlap <rddunlap@osdl.org>
description:	scsi imm driver needs to call scsi_unregister();
product:	Linux
product_versions: 260-test3
diffstat:	=
 drivers/scsi/imm.c |    1 +
 1 files changed, 1 insertion(+)

[Switch GFDL to GPL]

Currently mpt fusion does not handle IOMMU overflow (pci_map_sg returning 0)
very gracefully. It gets not reported to the upper layers. This patch fixes this.

As a related note the fusion driver tends to trigger the NMI watchdog as soon
as it goes into any error recovery, because it busy waits for seconds with
interrupts disabled (seems to be still true with the new error handling in 2.6).
This is a big inconvenient because it leads to a forced oops.

I tried to work around it by exporting touch_nmi_watchdog and using it in
the delays, but Linus was opposed to this approach. It would be nice
if someone could fix this. Afaik in 2.6 the error recovery should mostly
run in process context, so it should be possible to use schedule_timeout()
with interrupts on for the delays.

At least on x86-64 the NMI watchdog runs by default and even on i386 it is a
very useful debugging tool.

sound/pcmcia/vx/vx_entry.c:

This patch removes the PCMCIA timer release functionality which is no
longer required. Without this the module does not compile.

Too much cut-and-paste, noticed by Dmitry Torokhov <dtor_core@ameritech.net>

into samba.org:/home/paulus/kernel/for-linus-ppc

On PPC32 we reorder the arguments so they fit into 6 registers.  Glibc will
need a two-line stub to change them from the standard order to the ordering
used by the system call: (fd, advice, offset, len).

into samba.org:/home/paulus/kernel/for-linus-ppc

into samba.org:/stuff/paulus/kernel/for-linus-ppc

into samba.org:/stuff/paulus/kernel/for-linus-ppc

into samba.org:/stuff/paulus/kernel/for-linus-ppc

into samba.org:/stuff/paulus/kernel/for-linus-ppc

In both 2.4 and in 2.6, error handling for bad cdrom media is
wrong.  And it is my fault I'm afraid, since I botched an earlier
fix for the problem by putting the fix in the wrong spot.

My kids have a "Jumpstart Toddlers" cd they have long since
completely killed, which makes a great test disc.  Without this
fix, the best time projection I can get for completing a dd type
sector copy is about 2 years...  Most of that is spent thrashing
about in kernel space trying to re-read sectors we already know
are not correctable....  After the fix, I was able to rip a copy
the CD (or rather muddle through it getting lots of EIO errors)
in about 15 minutes.

Attached is the fix for 2.6.x,

into samba.org:/stuff/paulus/kernel/for-linus-ppc

of requiring users to select SCSI in order to see it.

into fys.uio.no:/home/linux/bitkeeper/nfsclient-2.5