On Wed, Nov 19, 2003 at 12:48:28PM +0000, Christoph Hellwig wrote:
> On Tue, Nov 18, 2003 at 01:29:22PM -0800, Mark Haverkamp wrote:
> > > +	pci_set_master(pdev);
> > > +	pci_set_dma_mask(pdev, 0xFFFFFFFFULL);
> >
> > I've been told that the return value of this should be checked as it is
> > possible for it to fail.
>
> Indeed.  This patches objective was to convert aacraid to the new-style
> probing, not to fix bugs, but I'll add the fix to the next revision of
> the patch anyway.

Ok here's a new patch.  Updates:

  - check pci_set_dma_mask return value
  - fix leak in the HBA remove path
  - fix leak in probe_one failure case
  - remove unused list of hosts
  - avoid scsi.h usage all over driver
  - mention the updates in the README file

Hi Doug,

you've been the last who touched inia100.c, so I may assume you
actually have the hardware?  I've updated the driver to the new
pci probing and scsi host registration code and it would be cool
if someone could test it so we could merge it into early 2.6.

Need to remove access_count from new seq_file code

From: 	Douglas Gilbert <dougg@torque.net>

This is an updated patch for the sg driver that takes into
account Patrick LaVarre's fix for negative reserved buffer
sizes found in lk 2.6.0-test9.

So it has the same changelog to the patch I sent on 2003/10/11:
     - add "struct cdev" [char_devs] objects to increase
       maximum number of sg devices from 256 to 8192
     - use seq_file interface for /proc/scsi/sg/*
       pseudo files
     - sysfs symlinks between the sysfs scsi device and the
       corresponding sg cdev node (and vice versa)

An edited "tree" output showing an example of these symlinks
was included in my previous post.

As noted in another thread, st (and osst) may need "cdevs"
and sysfs symlinks so SCSI tape devices have sysfs visibility
in lk 2.6 .

Also if both st and sg had sysfs visibility then Patrick
Mansfield's scsi_id program could be made to work for tape
drives (enclosures, tape robots, etc) by following these
symlinks.

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

2.05.00.05 changes
* error handling fixes, e.g. use of host_lock 

2.05.00.04 changes
* removed __init from mptscsih_setup
* removed __init from get_setup_token
* changed copyright from 2002 to 2003
* added new mailto, and removed Pam.Delaney
* added some fix for 32bit emulation when unloading mptctl module

Hi.

The patch below (against 2.6.0-test8) makes it possible to share the
interrupt when the aha1740 is configured for a level triggered interrupt.
It appears to work fine on my i486 EISA box with an AHA-1742A and an NE3200
Ethernet board sharing an irq. Comments, please.

From: 	Matthew Wilcox <willy@debian.org>

2.1.18f:
 - Rewrite the Kconfig help
 - Always honour CONFIG_SCSI_SYM53C8XX_IOMAPPED.  Alpha people used to
   have it forced off, Sparc people used to have it forced on.  (Thanks
   to Dann Frazier for testing on Alpha)
 - Simplify the NVRAM handling a bit.
 - SYM_OPT_NO_BUS_MEMORY_MAPPING is never set.
 - Remove PCI DMA abstraction.  (Christoph Hellwig)
 - Redo SCSI midlayer registration and unregistration to allow module
   load/unload to work.  Now copes with scsi_add_host() failing.  (Thanks
   to Brian King for testing)
 - Replace bcmp() with memcmp().
 - Change the MAINTAINER entry to myself.

Ninjas don't seem to like 64bit. The driver spew out so many
integer/pointer mismatch warnings that I gave up.

Mark it as !64BIT

-Andi

On Tue, Oct 07, 2003 at 07:33:23PM +0200, Andi Kleen wrote:
>
> aha152x seems to be not 64bit safe and spews out warnings on x86-64.
> As I think it's a ISA only driver anyways I just marked it as
> ISA only and !64BIT for Alpha's sake.


Matthew Wilcox pointed out that it was already marked ISA only.
I actually ment to change another driver, but looking at the source of one
it seems to be 64bit unclean too.

As there are 64bit architectures that have ISA slots (like old Alphas)
I think this patch is still appropiate.

-Andi

As Matthew Wilcox pointed out - the ISA aha152x driver was already marked
as ISA only, so couldn't have been enabled on x86-64.

The warning I saw was actually for the PCMCIA aha152x driver.

Mark that one as !64BIT

During a make allyesconfig on x86-64 I noticed several integer/pointer
mismatch warnings in the bus logic driver.

patch_name:	buslogic_ehupdate_v3.patch
patch_version:	2003-10-02.14:10:32
author:		Randy.Dunlap <rddunlap@osdl.org>
description:	update BusLogic driver to use current SCSI
		  error handling model;
		remove duplicate doc comments -- use
		  Documentation/scsi/BusLogic.txt only;
product:	Linux
product_versions: 2.6.0-test6
diffstat:	=
 Documentation/scsi/BusLogic.txt |    2
 drivers/scsi/BusLogic.c         |  229 ++--------------------------------------
 2 files changed, 16 insertions(+), 215 deletions(-)

Consolidate and nicely log the scsi_device and scsi command before sending
and after completing a command to an adapter driver.

This small patch adds a "release" method to the "pseudo_0"
device to stop the noise when the scsi_debug module is
loaded.

Another annoyance that I was unable to get to the bottom
of was during "rmmod scsi_debug" **:
  Synchronizing SCSI cache for disk sda: <4>FAILED
    status = 0, message = 00, host = 1, driver = 00
That is a DID_NO_CONNECT error. So the LLD host is
being shut down before the sd driver gets a chance to
send through a SYNCHRONIZE CACHE command. If the user
instigates a rmmod (as distinct from the hardware
saying the host/device is gone), shouldn't a window
be left open for such a flushing type command. This
problem seems to have appeared recently.


** "echo -1 > add_host" in scsi_debug's driver directory
   (i.e. remove a host) also causes the same error so the
   problem is not perculiar to rmmod.

patch_name:	drivers_clean.patch
patch_version:	2003-09-09.17:01:58
author:		Randy.Dunlap <rddunlap@osdl.org>
description:	fix to remove these warnings:
  drivers/scsi/cpqfcTSinit.c:1583: warning: unused variable `timeout'
  drivers/scsi/cpqfcTSinit.c:1584: warning: unused variable `retries'
  drivers/scsi/cpqfcTSinit.c:1585: warning: unused variable `scsi_cdb'
  drivers/scsi/cpqfcTSinit.c:471: warning: `my_ioctl_done' defined but not used
product:	Linux
product_versions: 2.6.0-test6
changelog:	ifdef around my_ioctl_done();
		write a new, smaller version of cpqfcTS_TargetDeviceReset(),
		but keep the previous version for future updates;
maintainer:	Chase Maupin (support@compaq.com)
diffstat:	=
 drivers/scsi/cpqfcTSinit.c |   17 +++++++++++++----
 1 files changed, 13 insertions(+), 4 deletions(-)

From Jon Burgess:

  There is a problems with blank DVD media using the ide-cd driver.

  When we attempt to read the blank disk, the drive responds to the read
  request by returning a "blank media" error.  The kernel doesn't have
  any special case handling for this sense value and retries the request
  a couple of times, then gives up and does a bus reset and disables DMA
  to the device.

  Which obviously doesn't help the situation.

  The sense key value of 8 isn't listed in ide-cd.h, but it is listed in
  scsi.h as a "BLANK_CHECK" error.

  This trivial patch treats this error condition as a reason to abort
  the request.  This behaviour is the same as what we do with a blank CD-R.

  It looks like the same fix might be desired for 2.4 as well, although
  is perhaps not so important since scsi-ide is normally used instead.

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

 1/ make sure raid5 doesn't try to handle multiple overlaping
    requests at the same time as this would confuse things badly.
    Currently it justs BUGs if this is attempted.
 2/ Fix a possible data-loss-on-write problem.  If two or
    more bio's that write to the same page are processed at the
    same time, only the first was actually commited to storage.
 3/ Fix a use-after-free bug.  raid5 keeps the bio's it is given
    in linked lists when more than one bio touch a single page.
    In some cases the tail of this list can be freed, and
    the current test for 'are we at the end' isn't reliable.
    This patch strengths the test to make it reliable.

Petr Vandrovec noticed a problem where the thread group leader
would not be properly reaped if the parent of the thread group
was ignoring SIGCHLD, and the thread group leader had exited
before the last sub-thread.

Fixed by Ingo Molnar.

This time we have a SMP memory ordering issue in prepare_to_wait(),
where we really need to make sure that subsequent tests for the
event we are waiting for can not migrate up to before the wait
queue has been set up.

In 2.5.x, the BKL was pushed from vfs_readdir() into the filesystem
specific functions.  But only the unlock_kernel() made it into the HPFS
code, lock_kernel() got lost on the way.  This rendered the filesystem
unusable.

This adds the missing lock_kernel().  It's been tested by Timo Maier who
also reported the problem earlier today.

The previous scsi_ioctl.c patch didn't cleanup the buffer/bio in the
error case. 

Fix it by copying the command data earlier.

corruption on SMP because of another CPU still accessing a waitqueue
even after it was de-allocated.

Use a careful version of the list emptiness check to make sure we
don't de-allocate the stack frame before the waitqueue is all done.

into kroah.com:/home/greg/linux/BK/usb-2.6

visor.c defines one empty slot in USB ids table that can be filled in at
runtime using module parameters. file2alias generates catch-all alias for it:

alias usb:v*p*dl*dh*dc*dsc*dp*ic*isc*ip* visor

patch adds the same sanity check as in depmod to scripts/file2alias.

This patch fixes the bug where running ppp over a ttyUSB device would fail.

This patch was integrated by you in 2.4 six months ago.  Unfortunately
it never got into 2.5.  Without it you can end up with crashes such
as http://bugs.debian.org/218670

apart from locking bugs, there are other races. This fixes one with
signal delivery. The signal should be delivered _before_ the reciever
is woken.

this fixes two instances of GFP_KERNEL from completion handlers.

On Mon, Dec 01, 2003 at 11:21:58AM -0800, Greg KH wrote:
> Can't you use xsane without the scanner kernel driver?  I thought the
> latest versions used libusb/usbfs to talk directly to the hardware.
> Because of this, the USB scanner driver is marked to be removed from the
> kernel sometime in the near future.

After a bit of mucking around (and possibly finding a bug with debian's
libusb/xsane/hotplug interaction, nothing seems to run
/etc/hotplug/usb/libusbscanner and thus only root can scan, anyone whose
got this working please let me know), the problem does not exist if I
only use  libusb xsane.

How about the following:

This patch fixes some obvious errors in the jumpshot and datafab drivers.

This should close out Bugzilla bug #1408

> Date: Mon, 1 Dec 2003 12:14:53 -0500 (EST)
> From: Alan Stern <stern@rowland.harvard.edu>
> Subject: Patch from Eduard Hasenleithner
> To: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
> cc: USB Storage List <usb-storage@one-eyed-alien.net>
>
> Matt:
>
> Did you see this patch?  It was posted to the usb-development mailing list
> about a week ago, before I started making all my changes.  It is clearly
> correct and necessary.
>
> Alan Stern

James reported a bug in scsi_ioctl.c where it mem copies a user pointer
instead of using copy_from_user(). I inadvertently introduced this one
when getting rid of CDROM_SEND_PACKET. Here's a trivial patch to fix it.

TCP IPV6 ->hash() method should not grab a socket reference.

If a device can't be configured, the current test9 code forgets
to clean it out of sysfs.  This resolves that issue, so the retry
in usb_new_device() stands a chance of working.

The enumeration code still doesn't handle such errors well, but
at least this way that hub port can be used for another device.

They should be bound to the interface the driver is attached to, not
the device.

The problem was that the cursor was in the list being walked, and when
the pointer pointed to the cursor the list_del/list_add_tail pair would
oops trying to find the entry pointed to by the prev pointer of the
deleted cursor element.

The solution I found was to move the list_del earlier, before the
beginning of the list walk. since it is not used during the list walk and
should not count in the list enumeration it can be deleted, then the
list pointer cannot point to it so it can be added safely with the
list_add_tail without oopsing, and everything works as expected.

I am unable to oops this version with any of my test programs. 

Patch acked by Al Viro.