When umounting a ppa device, I see (2.6.0-test3):

Releasing ppa0
Iomega VPI0 (ppa) interface did not call scsi_unregister
Call Trace:
 [<f89aeca6>] exit_this_scsi_driver+0xb6/0xfa [ppa]
 [<c013c08d>] sys_delete_module+0x18d/0x1e0
 [<c0155276>] do_munmap+0x136/0x1b0
 [<c0155333>] sys_munmap+0x43/0x70
 [<c0109657>] syscall_call+0x7/0xb

so ppa needs to call scsi_unregister().
Is this all that's needed?

--
~Randy


patch_name:	scsi_ppa_unreg.patch
patch_version:	2003-08-15.09:46:30
author:		Randy.Dunlap <rddunlap@osdl.org>
description:	ppa driver needs to call scsi_unregister(host);
product:	Linux
product_versions: 260-815
maintainer:	campbell@torque.net
diffstat:	=
 drivers/scsi/ppa.c |    1 +
 1 files changed, 1 insertion(+)

This patch has no code changes.  It is just a resync of all the version
numbers with the 6.10 driver and bumping up the driver version number.

From: David Jeffery <david_jeffery@adaptec.com>
(with minor reject fixes)

This patch removes several places where we were using kmalloc flag
tricks to always get memory <4GB instead of using the proper
dma_alloc_consistent() API.  It also no longer #errors on when compiled
for x86-64 kernels.

From: David Jeffery <david_jeffery@adaptec.com>
(with minor fixes for rejects)

This patch is a resync with the 6.10 driver for 2.4 and updates the 2.4
compatability code.  It adds a new wrapper function for differences
between the 2.4 and 2.6 scsi proc interface,  forces pci posting in 3
places, corrects using a meaningless constant of 0x80 to 0, and syncs
comments that had changed. 

From: 	David Jeffery <david_jeffery@adaptec.com>
(with fix ups for rejections on CONFIG_HIGHIO)

This patch has no code changes.  It is only the result of running
scripts/Lindent on drivers/scsi/ips.c in test3.  Rather than use this
patch, it may be advisable to just run 

sh scripts/Lindent drivers/scsi/ips.c 

The nasty 3-4 space indenting in ips.c has always made it a pain to
read.  The remaining patches in this patch set all assume ips.c has been
lindented. 

In sym_attach we call sym_hcb_attach. If it fails we call
sym_free_resources which will call sym_hcb_free. Unfortunately
sym_hcb_attach also calls sym_hcb_free on failure.

This results in a bunch of things being freed twice and it looks like
the sym2 memory allocator adds them to the freelist twice. Sometime later
on we allocate the memory twice, with weird consequences.

One problem we found when we hit a bad adapter, we need to use scsi_host_put.

This is the preliminary version with no error indications or
control over the types of failures.

From: Andries Brouwer <aebr@win.tue.nl>

I see an Oops in the SCSI code, caused by the fact that sdkp is NULL
in sd_shutdown. "How can that be?", you will ask - dev->driver_data was set
in sd_probe. But in my case sd_probe never finished. An insmod usb-storage
hangs forever, or at least for more than six hours, giving ample opportunity
to observe this race between sd_probe and sd_shutdown.
(Of course sd_probe hangs in sd_revalidate disk.)

Perhaps the obvious test is a good idea.
Locking seems meaningless - sd_probe will never finish.

This patch against scsi-misc-2.5 adds a sysfs attribute to allow scanning
(or rescanning) and deletion of scsi_devices.

It also allows scanning of entire hosts, channels, or targets.

It adds a per-host scan attribute, and a per scsi_device delete attribute.

This was in scsi_sysfs.c previously but given that it's not
sysfs-related and the whole scsi code is built around the driver
model now it's better to have it where it belongs.  Also allows
us to reduce the scsi_mod-wide globals.

DEV_STALE_* is used for some crude hacks in pcmcia netdrivers and it
seems people copied it blindly over to the scsi ones also it couldn't
ever be triggered (dev_link_t.open is never incremented e.g.)

The recent update made it check the scsi_remove_host retval which
is void now.

From: Jes Sorensen <jes@wildopensource.com>
with minor compile and rejection fixes

This patch might make the qla1280 driver build again in Linus' bk
tree. I don't have those sources here so I made the changes relative to
2.6.0-test2 and I am out of space on my laptop

Only compile tested, but if it works it works ;-)

CLONE_THREAD without CLONE_DETACHED will now return -EINVAL, and
for a while we will warn about anything that uses it (there are no
known users, but this will help pinpoint any problems if somebody
used to care about the invalid combination).

A part of the 64 bit kdev_t patch already got merged, and it changes the
format of /sys/block/<dev>/dev from %02x%02x to %u:%u.  The partition
could must also be changed.

e.g. cat /sys/block/hda/hda5/dev should return 3:5 instead of 0305

It was caused by improper IV calculation in loop.c

cryptoloop won't oops anymore if ECB mode is requested.

Just call them from the x86-64 entry code.

Also implement the x86-64 specific sys32_timer_create.

Add 32bit->64bit conversion functions for POSIX timers.

I kept timer_create architecture specific, because it does signal specific
stuff which is not portable enough for generic compat.

Add compat_sys_utimes for 32bit->64bit utimes conversion

Add compat_* functions for statfs64. The 32bit layout unfortunately
does not match x86-64.

This patch allows modules to work for ARM, and is the one thing which
prevents the standard tree from building for any ARM machine.

After reviewing the /proc/kcore and kclist issues, I've decided that I'm
no longer prepared to even _think_ about supporting /proc/kcore on ARM -
it just gets too ugly, and adds too much code to make it worth the effort,
the time or the energy to implement a solution to that problem.

/proc/kcore should probably go away, but in the meantime this just allows
ARM to ignore the issues.

makes 2.4->2.6 diff noticeable smaller

and add stripped down buffer_head variant (struct idetape_bh).

ide-tape is accessed by a char device (not a block one!),
it uses block layer only to queue requests.

also add missing Kconfig help entry from 2.4.x

- more locking fixes
- preserve gendev.parent of the old hwif in the new one

Noticed by Andries.Brouwer@cwi.nl.

Also:
- kill probe_lba_addressing() wrapper
- rename hwif->addressing to hwif->no_lba48

From Andries.Brouwer@cwi.nl.

- always use drive->capacity48 and kill drive->capacity

I've changed drive->capacity48 to drive->capacity64 to avoid confusion.

From Andries.Brouwer@cwi.nl.

- kill redundant, never executed code in lba_capacity_is_ok()
- add idedisk_supports_{hpa,lba48}() helpers
- don't recalculate drive->cyl for drives using LBA addressing,
  we never fall-back to CHS, so its useless and confusing
- remove wrong drive->head and drive->sect assignments for LBA-48
- don't overwrite id->lba_capacity and id->lba_capacity_2

HDIO_GETGEO_BIG_RAW is an ide specific hack introduced in 2.3.99-pre3.
There are no known programs using this ioctl.

Its aim was to provide current CHS translation to the user-space,
but very often it provides what driver thinks is a current translation
(drive with LBA have to support only one physical translation, also
 drive may not support chosen translation and there is no return value check).

hdparm -I can be used instead, it provides correct information
(and bogus data is still accessible through /proc/ide/hdX/geometry).

into home.osdl.org:/home/torvalds/v2.5/linux

into jen.americas.sgi.com:/src/lord/bitkeeper/xfs-2.5

by moving blk queue manipulation down into pagebuf. 
Fix some busted comments in page_buf.h, use a more
descriptive name for __pagebuf_iorequest.

SGI Modid: 2.5.x-xfs:slinx:155788a

The help for CONFIG_DEVPTS_FS claims that devfs "is a more general
facility".  But that apparently hasn't been true since 2.5.68.  This patch
removes that claim, and adds a warning to the DEVFS_FS help.

into home.osdl.org:/home/torvalds/v2.5/linux

into kernel.bkbits.net:/home/lord/xfs-2.5

From A1tmblwd@netscape.net

This updates my email address.

This corrects the following problem:

- release user buffer mapping early in read path (prevent oops in some
  HBA drivers)