The kvm mmio read path looks like:

 1. guest read faults
 2. kvm emulates read, calls emulator_read_emulated()
 3. fails as a read requires userspace help
 4. exit to userspace
 5. userspace emulates read, kvm sets vcpu->mmio_read_completed
 6. re-enter guest, fault again
 7. kvm emulates read, calls emulator_read_emulated()
 8. succeeds as vcpu->mmio_read_emulated is set
 9. instruction completes and guest is resumed

A problem surfaces if the userspace exit (step 5) also requests an interrupt
injection.  In that case, the guest does not re-execute the original
instruction, but the interrupt handler.  The next time an mmio read is
exectued (likely for a different address), step 3 will find
vcpu->mmio_read_completed set and return the value read for the original
instruction.

The problem manifested itself in a few annoying ways:
- little squares appear randomly on console when switching virtual terminals
- ne2000 fails under nfs read load
- rtl8139 complains about "pci errors" even though the device model is
  incapable of issuing them.

Fix by skipping interrupt injection if an mmio read is pending.

A better fix is to avoid re-entry into the guest, and re-emulating immediately
instead.  However that's a bit more complex.
Signed-off-by: NAvi Kivity <avi@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This makes the vmwrite errors on vm shutdown go away.
Signed-off-by: NAvi Kivity <avi@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Both "=r" and "=g" breaks my build on i386:

  $ make
    CC [M]  drivers/kvm/vmx.o
  {standard input}: Assembler messages:
  {standard input}:3318: Error: bad register name `%sil'
  make[1]: *** [drivers/kvm/vmx.o] Error 1
  make: *** [_module_drivers/kvm] Error 2

The reason is that setbe requires an 8-bit register but "=r" does not
constrain the target register to be one that has an 8-bit version on
i386.

According to

	http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10153

the correct constraint is "=q".
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Function short description should be on only one line.
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Unhide the SMBus on the Asus P4P800-X (and probably some other
models of the family.) This gives access to the memory module SPD
EEPROMs.

Thanks to Winbond for supporting the lm-sensors project with the
donation of this motherboard.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

In appendix a patch for the nokia 6233 mobile phone is included.
The patch is against 2.6.20-rc5. It is my first patch. Hopefully it has
the right format. The code makes my nokia 6233 on my computer work.


From: Manuel Osdoba <manuel.osdoba@tu-ilmenau.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Attached patch fixes typo in USB driver reported by Chase Douglas on linux-cirrus mailing
list. http://www.freelists.org/archives/linux-cirrus/12-2006/msg00003.htmlSigned-off-by: NPetr Stetiar <ynezz@true.cz>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Disable the USB_MULTITHREAD_PROBE option because it causes crashes on
people's machines and they never remember to actually read the config
help files.

No one likes this, everyone hates it, I'm going to go eat worms...

The full logic will be ripped out later.
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This fixes http://bugzilla.kernel.org/show_bug.cgi?id=7814Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

American Megatrends did something wrong in their floppy emulator. It breaks
with both kinds of MODE SENSE which our stack sends. Alan and I tried a few
tweaks, and got LUNs sensed right, but US_FL_NO_WP_DETECT is still needed.

I set the firmware bracket to 1.00 exactly, in case AMI or Sun fix it with a
firmware update. Hey, you never know.
Signed-off-by: NPete Zaitcev <zaitcev@redhat.com>
Signed-off-by: NPhil Dibowitz <phil@ipom.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

on request of the sourceforge project for this device, a kind of
robotized CD storage, it should be ignored by the generic driver.
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Bug fix for driver rndis_host which fixes rndis_host probing certain
Nokia S60 (Series 60) mobiles. While the rndis_host get probed by usbnet
and tries to bind the Nokia mobile the bind is going to fail. The
rndis_host module tries to release the device, in a wrong way, which
cause the oops.

Fixes Bugzilla #7201
Signed-off-by: NDaniel Gollub <dgollub@suse.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Different AX88772 dongles use different PHYs; the chip is capable of using
both a primary and secondary PHY, and supports an internal and external PHY.

It appears that some DUB-E100 devices use the internal PHY, so trying to use
an external one will not work (note that this is different across revisions,
as well; the "A" and "B" revs of the DUB-E100 use different PHYs!).  The data
sheet for the AX88772 chip specifies that the internal PHY id will be 0x10,
so if that's read from the EEPROM, we should use that rather than attempting
to use an external PHY.

Thanks to Mitch Bradley for pointing this out!
Signed-off-by: NAndres Salomon <dilinger@debian.org>
Cc: David Hollis <dhollis@davehollis.com>
Cc: Chris Ball <cjb@laptop.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

ITM screens send invalid x/y data when not touched. this was fixes a while ago
but the problem is if the screen is not touched anymore the driver never does
not report BTN_TOUCH as zero. fix it by sending the report with the last valid
coordinates when pressure is released.
Signed-off-by: NDaniel Ritz <daniel.ritz@gmx.ch>
Cc: J.P. Delport <jpdelport@csir.co.za>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This reverts commit 99a10a60.

As per Hugh Dickins:

  "Nadia Derbey has reported that mmap of /dev/kmem no longer works with
   the kernel virtual address as offset, and Franck has confirmed that
   his patch came from a misunderstanding of what an offset means to
   /dev/kmem - whereas his patch description seems to say that he was
   correcting the offset on a few plaforms, there was no such problem to
   correct, and his patch was in fact changing its API on all platforms."
Suggested-by: NHugh Dickins <hugh@veritas.com>
Cc: Franck Bui-Huu <fbuihuu@gmail.com>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Andi Kleen <ak@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Andrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Adds GTCO CalComp Interwrite IPanels to the hid-core.c blacklist.
Signed-off-by: NJeremy A. Roberson <jroberson@gtcocalcomp.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The commit 4916b3a5 introduced a
hid regression between 2.6.19 and 2.6.20-rc1. The device put in
input_dev->cdev is now of type usb_device instead of usb_interface.

Before:
> # readlink -f /sys/class/input/input6/event4/device
> /sys/devices/pci0000:00/0000:00:10.0/usb2/2-1/2-1:1.1
After:
> # readlink -f /sys/class/input/input3/event3/device
> /sys/devices/pci0000:00/0000:00:10.0/usb1/1-1

This causes breakage:
> # udevinfo -q all -n /dev/input/event3
> P: /class/input/input3/event3
> N: input/event3
> S: input/by-path/pci-1-1--event-
> E: ID_SERIAL=noserial
> E: ID_PATH=pci-1-1-

No ID_MODEL, ID_VENDOR, ID_REVISION, ID_TYPE etc etc.

Fix this by assigning the intf->dev into hid->dev, and fixing
all the users.
Signed-off-by: NAnssi Hannula <anssi.hannula@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

There is no reason for generic hid-input.c to include usb-specific
input.h.

As a sideeffect, this also fixes warning of redefinition of dbg()
macro, when hid-input.c is compiled with DEBUG (as there is a clash
between dbg() from hid.h and usb/input.h).
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

hid/hid-core.c references 'len' variable when DEBUG_DATA is defined,
but the actual name of the variable is 'size'. Fix it.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This change introduces a mapping for LED indicators between the HID
specification and the Linux input subsystem. The previous code properly
mapped the LEDs relevant for Keyboards, but garbeled the remaining ones.
With this change all LED enums from the input system get mapped to more
or less equivalent LED numbers from the HID specification.

This patch also ensures that the unused bits in a HID report to the
device are zeroed out. This makes the 3Dconnexion SpaceNavigator fully
usable with the linux input system.
Signed-off-by: NSimon Budig <simon@budig.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

I've got a newer MacBook with core2duo. Two keys on the keyboard are
swapped, "unswaping" works with the same trick as GEYSER3_ISO.
Signed-off-by: NAdrian Friedli <masteradi@gmx.ch>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

We have USB_HID _newly_ selected in configurations which didn't
have it before, which overrides CONFIG_HID and builds HID without
input support.

Nevertheless, here's a patch to solve more of the same that my original
patch attempted to solve.  The original patch is still required.  Seems
to solve the final instance of this problem here.
Signed-off-by: NRussell King <rmk+lkml@arm.linux.org.uk>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

With the following patch, my HighPoint 2310 with a Marvell 88SX7042 on
it seems to work OK.

The controller only has 4 ports, with MV_FLAG_DUAL_HC it seems to init 8
ports and fails miserably at probe time. There are no other devices mapped
to that chip, maybe it was just incorrectly specified in the first place?
Signed-off-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

libata EH ignores port-wide actions in per-dev action mask.  However,
device resume requests EH_SOFTRESET using per-dev action mask.  Under
certain circumstances, this results in not resetting frozen port after
resuming which causes failure of all commands.

This patch allows port-wide actions to be requested in per-dev action
mask.  Before EH recovery starts, port-wide actions will be collected.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Add PCI ID 0x5337 to supported PCI ID.  This is VT8237 in IDE mode.
Signed-off-by: NLuca Pedrielli <luca.pedrielli@barradev.it>
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Fix from http://bugzilla.kernel.org/show_bug.cgi?id=7747Signed-off-by: NAndrew Morton <akpm@osdl.org>
Cc: <sleepy@mike-neko.net>
Signed-off-by: NFrancois Romieu <romieu@fr.zoreil.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Ignore the VID readings when the motherboard has not designed
the function.  
Signed-off-by: NGong Jun <jgong@winbond.com>
Signed-off-by: NRudolf Marek <r.marek@assembler.cz>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

Catch the cases when alternative pins are used to route the
fan9-12 input.
Signed-off-by: NRudolf Marek <r.marek@assembler.cz>
Signed-off-by: NGong Jun <jgong@winbond.com>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

Ignore the temperature readings when its channel is disabled,
ignore AMDSI readings.
Signed-off-by: NGong Jun <jgong@winbond.com>
Signed-off-by: NRudolf Marek <r.marek@assembler.cz>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

I wonder how we came up with such a broken test in the first place.
Signed-off-by: NJean Delvare <khali@linux-fr.org>

Fix the driver to match the information in datasheet 1.0. AMD 
SI interface is marked as reserved, computing formula for 5VDD
and 5VSB is updated.
Signed-off-by: NGong Jun <jgong@winbond.com>
Signed-off-by: NRudolf Marek <r.marek@assembler.cz>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

The local versions of phy_read() and phy_write() in ucc_geth_phy.c conflict
with the prototypes in include/linux/phy.h, so this patch renames them,
moves them to the top of the file (while eliminating the redundant prototype),
and makes them static.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NKumar Gala <galak@kernel.crashing.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

The workqueue interface changed with David Howell's patch on 11/22/2006
(SHA 65f27f38).  Several drivers were
updated with that patch to handle the new interface, but ucc_geth.c
was not one of them.  This patch updates ucc_geth.c to support the new
model.

A compiler warning in set_mac_addr() was also fixed.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NKumar Gala <galak@kernel.crashing.org>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Driver is now at version 1.2.0.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Now that IRQ allocation is done in myri10ge_open(), we want to still
check when loading the driver that IRQ allocation could succeed later.

Additionaly, we fix the initialization and printing of netdev->irq.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Under some circumstances, using WC without the WC fifo is faster.
So we make it possible to tune wc_fifo with a module parameter.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Similar to this commit:

http://kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=d15e9c4d9a75702b30e00cdf95c71c88e3f3f51e

It's not safe in cp_start_xmit to blindly call spin_lock_irq and then
spin_unlock_irq, since it may very well be the case that cp_start_xmit
was called with interrupts already disabled (I came across this bug in
the context of netdump in RedHat kernels, but the same issue holds, for
example, in netconsole). Therefore, replace all instances of
spin_lock_irq and spin_unlock_irq with spin_lock_irqsave and
spin_unlock_irqrestore, respectively, in cp_start_xmit(). I tested this
against a fully-virtualized Xen guest using netdump, which happens to
use the 8139cp driver to talk to the emulated hardware. I don't have a
real piece of 8139cp hardware to test on, so someone else will have to
do that.
Signed-off-by: NChris Lalancette <clalance@redhat.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

The PCI-E modifications to bcm43xx do not set up the interrupt vector
correctly. Tested with BCM4311 (PCI-E) on x86_64 and BCM4306 (PCI) on i386.
Signed-off-by: NLarry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

There's a problem, pointed by Meelis Roos <mroos@linux.ee>, that, on ppc32 arch,
with some gcc versions (noticed with prerelease 4.1.2 20061115), compilation 
fails, due the lack of __ucmpdi2 to do the required 64-bit comparision.
This patch takes some sugestions made by Andrew Morton <akpm@osdl.org>,
Stelian Pop <stelian@popies.net> and Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>

autodetect LG TAPC G701D as tuner type 37.
Thanks to Adonis Papas, for pointing out the missing autodetection
for this tuner.
Signed-off-by: NMichael Krufky <mkrufky@linuxtv.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>