The only documentation about the MLQUEUE returns is in the code, it
should also be in here.

I found a minor bug in that patch.  Its in mptbase_remove(). I saw
it this morning when compiling a driver for x86_64 platform. Funny
how it didn't complain in x86 system.

It's vestigial (it doesn't have a build entry in the Makefile)

It hasn't compiled for ages and no-one seems to care ...

The qla2xxx drivers take its place

From: Bob Doyle <doyle@primenet.com>

Deleted dead code:
- BusLogic_ReportTargetDeviceInfo()
- BusLogic_SelectQueueDepths()
- BusLogic_SendBusDeviceReset()
- BusLogic_ResetCommand()

Fixed:
- removed dependency on "scsi_obsolete.h"
  o Changes to BusLogic_ResetHostAdapter()
  o Changes to BusLogic_AbortCommand()
- removed the gratuitous typedefs per Christoph's comments
- several functions with incorrect __init and __exit sections.
- Removed vestiges of the 'Error Recovery' command line function (which
  is fundamentally incompatible with the new EH design).

Left to do:
- Test eh_abort_handler
- Test eh_host_reset_handler
- Test as module
- Straighten out FlashPoint conditional compile so it doesn't
  require #include "FlashPoint.c". 
- Please ignore the long line length and the very lengthy
  identifiers (for now).

Add SG_IO support for st, so we can send scsi commands directly to an st
device.

Though st (still?) needs to move to move away from cdev for udev to
function with it.

This patch against a recent bk 2.6 changes scsi_cmd_ioctl to take a
gendisk as an argument instead of a request_queue_t. This allows scsi char
devices to use the scsi_cmd_ioctl interface.

In turn, change bio_map_user to also pass a request_queue_t, and add a
__bio_add_page helper that takes a request_queue_t.

Tested ide cd burning with no problems.

If the scsi upper level scsi_cmd_ioctl usage were consolidated in
scsi_prep_fn, we could pass a request_queue_t instead of a gendisk to
scsi_cmd_ioctl.

Make sg and st respect any block alignment
constraints for the data.  Usb storage already
sets these to what it needs.

Also reset the default to 7 for other SCSI devices.

If a request is sent through scsi_wait_req the function may be woken up
from the completion by a function other than scsi_wait_done. This can
happen as a result of cases that return BLKPREP_KILL in the scsi_prep_fn
function.

author: Mike Anderson <andmike@us.ibm.com>
patch_version: Thu Jan 29 09:03:44 UTC 2004

 patched-2.6-andmike/drivers/scsi/scsi_lib.c |    2 ++
 1 files changed, 2 insertions(+)

This patch against recent mainline bk removes the bogus WARN_ON for
single_lun devices, and a meaningless comment.

We clear the starget_sdev_user, and immediately blk_run_queue for the LUN
that just issued IO. Another LUN could race in scsi_request_fn, but it is
most likely that the last user will get there first, and reset
starget_sdev_user. If it does not, it will have to wait for the other LUN
to finish all of its IO.

From: 	Andrew Vasquez <andrew.vasquez@qlogic.com>

-  Bump version number -> 8.00.00b10.

-  Use down() not down_interruptible() while waiting for mailbox
   completions.
  
        o down_interruptible() is the wrong thing to do here 
          (in case a signal hits you sooo do the wrong thing
          and corrupt the semaphore)
          (Arjan van de Ven arjanv_redhat.com)

-  Misc. ISP6312/ISP6322 changes:
  
        o Asynchronous IOCB handling is not available on 
          ISP6312 and ISP6322 type boards.
  
        o Firmware type (FLX) desgnation fix. 

-  Clean up some extraneous code:
  
        o Remove duplicate call to qla2x00_config_os().
        o Remove unused structure member rsvd in struct dev_id.

-  Create single module firmware loaders for each firmware image
   shipped with the driver (2/2).

-  Create single module firmware loaders for each firmware image
   shipped with the driver.

-  Resync with latest released firmware 3.02.21.

From: 	Andrew Vasquez <andrew.vasquez@qlogic.com>

RIO mode 3 (originally used in the qla2xxx driver) will not work on
big-endian platforms due to the firmware returning only the first-word
of the dword command handle returned after SCSI completion.  Use RIO
mode 4 since full 32bit handle is returned during status.

I've recently added an IBM p630 machine (ppc64) to my test ring with a
mix of 2200 and 2300 cards -- I feel fairly confident that most endian
type issues are addressed.

From: 	Andrew Vasquez <andrew.vasquez@qlogic.com>

The driver does not issue GFT_IDs for each of the ports discovered
during its SNS scan.  Remove unused code.

From: 	Andrew Vasquez <andrew.vasquez@qlogic.com>

The 2200 firmware doesn't support the MS IOCB interface the driver 
uses to issue CT commands to the SNS.  Basically, I had to 
forward-port some code from the 6.x driver to get the SNS calls 
to go through.  I've had several machines (mixture of 2200s/
2300s/others with loop and fabric connections) running with this
patch for the past week.

From:         Andrew Vasquez <praka@users.sourceforge.net>
http://sourceforge.net/projects/linux-qla2xxx/

A very simple patch against 2.6.0 follows, which fixes an improper file
reference in the scsi sym53c8xx_2 driver help text. The referenced
documentation file was moved in 2.5.48 but the help text was not updated
accordingly.

 linux-262-rc1-bk1-rddunlap/drivers/scsi/Kconfig |    2 +-
 1 files changed, 1 insertion(+), 1 deletion(-)

From: Timmy Yee <shoujun@masterofpi.org>

In drivers/scsi/aha1542.c, kmalloc() is called with no memtype (i.e.
without some flag like GFP_KERNEL). The following patch will fix that.

diff -puN drivers/scsi/aha1542.c~aha1542_kmalloc_type drivers/scsi/aha1542.c


 linux-262-rc1-bk1-rddunlap/drivers/scsi/aha1542.c |    2 +-
 1 files changed, 1 insertion(+), 1 deletion(-)

From: Timmy Yee <shoujun@masterofpi.org>

The aha1542 driver calls panic() if kmalloc() fails, which it shouldn't
do. This patch changes that by having the code return a nonzero value, so
it tells the SCSI mid-layer to retry the command, as suggested by Randy.


diffstat:=
 drivers/scsi/aha1542.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

From: Moore, Eric Dean <Emoore@lsil.com>

Here is list of fix's.
* added new PCI API support
* added ACPI support
* added CONFIG_LBA, READ16, WRITE16 support
* underun fix
* chain buffer free list not being init properly
* reduce task management 
        (abort=2sec,reset bus=5sec, timeout=10sec)
* Hot plug fix's requested from Christoph Hellwig, and several
others from the linux-scsi@ list.

> This patch is against the sg driver found in lk 2.6.1-bk6 .
> It is an update on the sg patch sent in the "devices with more
> than one node" thread:
> http://marc.theaimsgroup.com/?l=linux-scsi&m=107415435300783&w=2

Attached is an updated patch adding the cdev_unmap() call
which Kai Makisara reported as required.

Noticed by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Fixed By: Juergen E. Fischer <fischer@linux-buechse.de>

On Sat, Jan 24, 2004 at 15:02:12 +0100, Guennadi Liakhovetski wrote:
> Correct and forgive me if I am wrong, it looks like you are now poking at
> ports without requesting them - in *_porttest. Is it a good idea?

Ouch.  No, of course not, you are right.  Thanks for pointing.  I
thought I had moved the port testing to aha152x_probe_one...

Following patch cleans that up.

The port test during autoconfiguration was also done without requesting
the region first and there was a bug in auto configuration of a TC1550
controller (both even before the patch).

attached my patch for the aha152x driver.  Tested for pcmcia and not
pcmcia and various devices.

changes in the aha152x driver:
- gather code that is not used by PCMCIA at the end
- move request_region for !PCMCIA case to detection
- migration to new scsi host api (remove legacy code)
- free host scribble before scsi_done
- fix error handling
- one isapnp device added to id_table

and in the pcmcia stub module:
- default to synchronous transfers
- release_region hack removed
- let core module do the scsi host api calls.

The patch at the end of this message fixes the following problems:

- cdev allocation moved outside st_dev_arr_lock, prevents the debugging
error messages reported by Mike and Mike
- cdev_unmap() added before cdev_del(), prevents oops (and kernel data
corruption) in case someone tries to use a device after removing the
module
- better error handling for failures in cdev allocation

As far as I am concerned, the patch is tested but testing by others might
be useful :-)

into hopper.(none):/usr/src/kernel/linux-2.6

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

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

tty->driver_data or state->port may end up being NULL in uart_close.
Make sure that we correctly clean up in this case, rather than
oopsing.

There is no way that tty can be NULL in uart_put_char() and
uart_write().  Eliminate these redundant tests.

From: Jim Collette <jim@hamachi.net>

There's an exit path in i8042_interrupt() which forgets to rearm the timer.
It can make the mouse die when X is started.

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

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

Current driver doesn't do shared irq properly.  When testing on
a laptop here irq 3 get shared between pcmcia slot and tty/IRDA

Andreas Happe <andreashappe@gmx.net> writes:
> my notebook (hp/compaq nx7000) still crashes when using 8139cp (runs
> rock solid with 8139too driver). The computer just locks up, there is no
> dmesg output. This has happened since I've got this laptop (around
> november '03).

It seems 8139cp.c has the race condition of rx_poll and interrupt.

NOTE, since I don't have this device, patch is untested. Sorry.

into kernel.bkbits.net:/home/davem/irda-2.6