With some memory model other than FLATMEM, the single node can
contains some holes so there might be many invalid pages. For
example, with two 256M memory and one 256M hole, some variables
(num_physpage, totalpages, nr_kernel_pages, nr_all_pages, etc.) will
indicate that there are 768MB on this system. This is not desired
because, for example, alloc_large_system_hash() allocates too many
entries.
Use free_area_init_node() with counted zholes_size[] instead of
free_area_init().
For num_physpages, use number of ram pages instead of max_low_pfn.
1. MIPS should select SPARSEMEM_STATIC since allocating bootmem in
memory_present() will corrupt bootmap area.
2. pfn_valid() for SPARSEMEM is defined in linux/mmzone.h
Ralf Baechle [Thu, 22 Jun 2006 21:42:32 +0000 (22:42 +0100)]
[MIPS] Avoid interprocessor function calls.
On the 34K where multiple virtual processors are implemented in a single
core and share a single TLB, interprocessor function calls are not needed
to flush a cache, so avoid them.
Arjan van de Ven [Thu, 13 Jul 2006 12:46:03 +0000 (14:46 +0200)]
[PATCH] lockdep: annotate mm/slab.c
mm/slab.c uses nested locking when dealing with 'off-slab'
caches, in that case it allocates the slab header from the
(on-slab) kmalloc caches. Teach the lock validator about
this by putting all on-slab caches into a separate class.
this patch has no effect on non-lockdep kernels.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] m68knommu: fix result type in get_user() macro
Keep the result holder variable the same type as the quantity we are
retreiving in the get_user() macro - don't go through a pointer version
of the user space address type.
Using the address type causes problems if the address type was const
(newer versions of gcc quite rightly error out for that condition).
[PATCH] i386: system.h: remove extra semicolons and fix order
include/asm-i386/system.h has trailing semicolons in some of the
macros that cause legitimate code to fail compilation, so remove
them. Also remove extra blank lines within one group of macros.
And put stts() and clts() back together; they got separated somehow.
David S. Miller [Thu, 13 Jul 2006 06:19:31 +0000 (23:19 -0700)]
[SPARC]: Fix OF register translations under sub-PCI busses.
There is an implicit assumption in the code that ranges will translate
to something that can fit in 2 32-bit cells, or a 64-bit value. For
certain kinds of things below PCI this isn't necessarily true.
Here is what the relevant OF device hierarchy looks like for one of
the serial controllers on an Ultra5:
2) Pass-through "SUNW,simba" since it lacks ranges property
3) Match "82010814.00000000.f13083f8" to third ranges property of PCI
controller node 'SUNW,sabre', and we arrive at the final physical
MMIO address of "0x1fff13083f8".
Due to the 2-cell assumption, we couldn't translate to a PCI 3-cell
value, and we couldn't perform a pass-thru on it either.
It was easiest to just stop splitting the ranges application operation
between two methods, ->map and ->translate, and just let ->map do all
the work. That way it would work purely on 32-bit cell arrays instead
of having to "return" some value like a u64.
It's still not %100 correct because the out-of-range check is still
done using the 64 least significant bits of the range and address.
But it does work for all the cases I've thrown at it so far.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Wed, 12 Jul 2006 22:59:53 +0000 (15:59 -0700)]
[SPARC64]: Fix 2 bugs in sabre_irq_build()
When installing the IRQ pre-handler, we were not setting up the second
argument correctly. It should be a pointer to the sabre_irq_data, not
the config space PIO address.
Furthermore, we only need this pre-handler installed if the device
sits behind a PCI bridge that is not Sabre or Simba/APB.
Signed-off-by: David S. Miller <davem@davemloft.net>
David Woodhouse [Wed, 12 Jul 2006 20:56:53 +0000 (13:56 -0700)]
[SPARC64]: Fix make headers_install
A minor typo in the include/asm-sparc64/Kbuild file prevents the make
headers_install from building a useful tree of kernel headers for
sparc64.
Signed-off-by: David Woodhouse <dwmw2@infradead.org> Signed-off-by: Tom "spot" Callaway <tcallawa@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Merge commit master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 of HEAD
* HEAD:
[NET]: fix __sk_stream_mem_reclaim
[Bluetooth] Fix deadlock in the L2CAP layer
[Bluetooth] Let BT_HIDP depend on INPUT
[Bluetooth] Avoid NULL pointer dereference with tty->driver
[Bluetooth] Remaining transitions to use kzalloc()
[WAN]: converting generic HDLC to use netif_dormant*()
[IPV4]: Fix error handling for fib_insert_node call
[NETROM] lockdep: fix false positive
[ROSE] lockdep: fix false positive
[AX.25]: Optimize AX.25 socket list lock
[IPCOMP]: Fix truesize after decompression
[IPV6]: Use ipv6_addr_src_scope for link address sorting.
[TCP] tcp_highspeed: Fix AI updates.
[MAINTAINERS]: Add proper entry for TC classifier
[NETROM]: Drop lock before calling nr_destroy_socket
[NETROM]: Fix locking order when establishing a NETROM circuit.
[AX.25]: Fix locking of ax25 protocol function list.
[IPV6]: order addresses by scope
Merge commit master.kernel.org:/pub/scm/linux/kernel/git/gregkh/i2c-2.6 of HEAD
* HEAD:
[PATCH] hwmon: Documentation update for abituguru
[PATCH] hwmon: Fix for first generation Abit uGuru chips
[PATCH] hwmon: New maintainer for w83791d
[PATCH] pca9539: Honor the force parameter
[PATCH] i2c-algo-bit: Wipe out dead code
[PATCH] i2c: Handle i2c_add_adapter failure in i2c algorithm drivers
[PATCH] i2c: New mailing list
[PATCH] i2c-ite: Plan for removal
[PATCH] i2c-powermac: Fix master_xfer return value
[PATCH] scx200_acb: Fix the block transactions
[PATCH] scx200_acb: Fix the state machine
[PATCH] i2c-iop3xx: Avoid addressing self
[PATCH] i2c: Fix 'ignore' module parameter handling in i2c-core
Len Brown [Thu, 13 Jul 2006 02:46:42 +0000 (22:46 -0400)]
[PATCH] Revert "ACPI: execute Notify() handlers on new thread"
This effectively reverts commit b8d35192c55fb055792ff0641408eaaec7c88988
by reverts acpi_os_queue_for_execution() to what it was before that,
except it changes the name to acpi_os_execute() to match ACPICA 20060512.
Signed-off-by: Len Brown <len.brown@intel.com>
[ The thread execution doesn't actually solve the bug it set out to
solve (see
http://bugzilla.kernel.org/show_bug.cgi?id=5534
for more details) because the new events can get caught behind the AML
semaphore or other serialization. And when that happens, the notify
threads keep on piling up until the system dies. ]
Ian McDonald [Thu, 13 Jul 2006 00:58:53 +0000 (17:58 -0700)]
[NET]: fix __sk_stream_mem_reclaim
__sk_stream_mem_reclaim is only called by sk_stream_mem_reclaim.
As such the check on sk->sk_forward_alloc is not needed and can be
removed.
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
Randy Dunlap [Thu, 22 Jun 2006 22:14:07 +0000 (15:14 -0700)]
[PATCH] Driver core: fix driver-core kernel-doc
Warning(/var/linsrc/linux-2617-g4//drivers/base/core.c:574): No description found for parameter 'class'
Warning(/var/linsrc/linux-2617-g4//drivers/base/core.c:574): No description found for parameter 'devt'
Warning(/var/linsrc/linux-2617-g4//drivers/base/core.c:626): No description found for parameter 'devt'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When changing power states from D0->DX and then from DX->D0, some
Intel PCIE chipsets will cause a device reset to occur. This will
cause problems for any D State other than D3, since any state
information that the driver will expect to be present coming from
a D1 or D2 state will have been cleared. This patch addes a
flag to the pci_dev structure to indicate that devices should
not use states D1 or D2, and will set that flag for the affected
chipsets. This patch also modifies pci_set_power_state() so that
when a device driver tries to set the power state on
a device that is downstream from an affected chipset, or on one
of the affected devices it only allows state changes to or
from D0 & D3. In addition, this patch allows the delay time
between D3->D0 to be changed via a quirk. These chipsets also
need additional time to change states beyond the normal 10ms.
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Matthew Garrett [Fri, 30 Jun 2006 09:31:25 +0000 (02:31 -0700)]
[PATCH] PCI: Clear abnormal poweroff flag on VIA southbridges, fix resume
Some VIA southbridges contain a flag in the ACPI register space that
indicates whether an abnormal poweroff has occured, presumably with the
intention that it can be cleared on clean shutdown. Some BIOSes check this
flag at resume time, and will re-POST the system rather than jump back to
the OS if it's set. Clearing it at boot time appears to be sufficient.
I'm not sure if drivers/pci/quirks.c is the right place to do it, but I'm
not sure where would be cleaner.
[PATCH] USB: move usb-serial.h to include/linux/usb/
USB serial outside of the kernel tree can not build properly due to
usb-serial.h being buried down in the source tree. This patch moves the
location of the file to include/linux/usb and fixes up all of the usb
serial drivers to handle the move properly.
Cc: Sergei Organov <osv@javad.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Anydata is using usb_serial_generic_write_bulk_callback() for its
read URB, but it should use usb_serial_generic_read_bulk_callback()
instead (it's a read URB, isn't it?).
Reported by Jon K Hellan <hellan@acm.org>.
Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Thu, 6 Jul 2006 22:48:53 +0000 (15:48 -0700)]
[PATCH] USB: gadget section fixups
Recent section changes broke gadget builds on some platforms. This patch
is the best fix that's available until better section markings exist:
- There's a lot of cleanup code that gets used in both init and exit paths;
stop marking it as "__exit".
(Best fix for this would be an "__init_or_exit" section marking, putting
the cleanup in __init when __exit sections get discarded else in __exit.)
- Stop marking the use-once probe routines as "__init" since references
to those routines are not allowed from driver structures. They're now
marked "__devinit", which in practice is a net lose.
(Best fix for this is likely to separate such use-once probe routines
from the driver structure ... but in general, all busses that aren't
hotpluggable will be forced to waste memory for all probe-only code.)
In general these broken section rules waste an average of two to four kBytes
per driver of code bloat ... because none of the relevant code can ever be
reused after module initialization.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Colin Leroy [Tue, 11 Jul 2006 09:36:43 +0000 (11:36 +0200)]
[PATCH] USB: Add one VID/PID to ftdi_sio
This patch adds the Testo USB interface to the list of devices
recognized by the ftdi_sio module. This device is based on a FT232BL
chip, and is used as an interface to get data from digital sensors
(thermometer, etc). See http://www.testo.com/
Signed-off-by: Colin Leroy <colin@colino.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Fri, 7 Jul 2006 17:45:13 +0000 (13:45 -0400)]
[PATCH] USB: unusual_devs entry for Nokia N91
This patch (as745) adds an unusual_devs entry for the Nokia N91, just like
the entry for the N80 added a couple of weeks ago. Apparently Nokia isn't
using very good firmware these days...
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Dan Streetman [Wed, 5 Jul 2006 23:17:27 +0000 (19:17 -0400)]
[PATCH] USB: add ZyXEL vendor/product ID to rtl8150 driver
I just got a "ZyXEL Prestige USB Adapter" that is actually RTL8150
adapter. Here is the relevant /proc/bus/usb/devices output (after
adding the vendor/product IDs to the driver):
This patch adds the ZyXEL vendor ID to the rtl8150.c driver. The
device has absolutely no identifying marks on the outside for model
type, just a serial number, and I can't find anything on ZyXEL's
website, so I called the product ID PRODUCT_ID_PRESTIGE to match the
product string.
Signed-off-by: Dan Streetman <ddstreet@ieee.org> Acked-by: <petkan@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Navaho Gunleg [Fri, 30 Jun 2006 07:44:03 +0000 (09:44 +0200)]
[PATCH] USB: add support for WiseGroup., Ltd SmartJoy Dual PLUS Adapter
This patch is to get the WiseGroup.,Ltd SmartJoy Dual Plus PS2-to-USB
Adapter [0x6677:0x8802] correctly detected. It sets the NOGET and
MULTI_INPUT quirks to make 2 joystick nodes appear in stead of only
one.
The below patch fixes the ipw module in kernel 2.6.17 for me; without
this change it simply does not work at all (all but the first writes are
refused because write_urb_busy is always 1).
This problem was there in 2.6.15 as well, but at that point I used the
(updated) ipw.c, version 0.4, from
http://www.neology.co.za/products/opensource/ipwireless/ which no longer
compiles with 2.6.17. It can be made to after a few changes but
obviously it's easier if the built-in ipw driver works instead of having
to download one from the neology site.
[PATCH] USB: remove empty destructor from drivers/usb/mon/mon_text.c
Remove destructor and call kmem_cache_create with NULL for the destructor.
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Cc: Pete Zaitcev <zaitcev@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Frank Gevaerts [Fri, 30 Jun 2006 09:34:45 +0000 (02:34 -0700)]
[PATCH] USB: ipaq.c timing parameters
Adds configurable waiting periods to the ipaq connection code. These are
not needed when the pocketpc device is running normally when plugged in,
but they need extra delays if they are physically connected while
rebooting.
There are two parameters :
* initial_wait : this is the delay before the driver attemts to start the
connection. This is needed because the pocktpc device takes much
longer to boot if the driver starts sending control packets too soon.
* connect_retries : this is the number of times the control urb is
retried before finally giving up. The patch also adds a 1 second delay
between retries.
I'm not sure if the cases where this patch is useful are general enough
to include this in the kernel.
Signed-off-by: Frank Gevaerts <frank.gevaerts@fks.be> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Frank Gevaerts [Fri, 30 Jun 2006 09:34:44 +0000 (02:34 -0700)]
[PATCH] USB: ipaq.c bugfixes
This patch fixes several problems in the ipaq.c driver with connecting
and disconnecting pocketpc devices:
* The read urb stayed active if the connect failed, causing nullpointer
dereferences later on.
* If a write failed, the driver continued as if nothing happened. Now it
handles that case the same way as other usb serial devices (fix by
Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>)
Signed-off-by: Frank Gevaerts <frank.gevaerts@fks.be> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Fri, 30 Jun 2006 09:34:42 +0000 (02:34 -0700)]
[PATCH] USB: ehci: fix bogus alteration of a local variable
In a rare and all-but-unused path, the EHCI driver could reuse a variable
in a way that'd make trouble. Specifically, if the first root hub port
gets an overcurrent event (rare) during a remote wakeup scenario (all but
unused in today's Linux, except for folk working with suspend-to-RAM and
similar sleep states), that would look like a fatal error which would shut
down the controller. Fix by not reusing that variable.
Spotted by Per Hallsmark <saxofon@musiker.nu>
Fixes http://bugzilla.kernel.org/show_bug.cgi?id=6661
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Kevin Lloyd [Fri, 30 Jun 2006 18:17:55 +0000 (11:17 -0700)]
[PATCH] USB: add driver for non-composite Sierra Wireless devices
This patch creates a new driver, sierra.c, that supports the new
non-composite Sierra Wireless WWAN devices. The older Sierra
Wireless and Airprime devices are supported in airprime.c.
Signed-off-by: Kevin Lloyd <linux@sierrawireless.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Mon, 26 Jun 2006 11:59:17 +0000 (12:59 +0100)]
[PATCH] USB serial ftdi_sio: Prevent userspace DoS
This patch limits the amount of outstanding 'write' data that can be
queued up for the ftdi_sio driver, to prevent userspace DoS attacks (or
simple accidents) that use up all the system memory by writing lots of
data to the serial port.
The original patch was by Guillaume Autran, who in turn based it on the
same mechanism implemented in the 'visor' driver. I (Ian Abbott)
re-targeted the patch to the latest sources, fixed a couple of errors,
renamed his new structure members, and updated the implementations of
the 'write_room' and 'chars_in_buffer' methods to take account of the
number of outstanding 'write' bytes. It seems to work fine, though at
low baud rates it is still possible to queue up an amount of data that
takes an age to shift (a job for another day!).
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
So priv->outstanding_urbs ends up as (unsigned long)(-1). Not good!
I haven't seen this happen with the visor driver as I don't have the
hardware, but I have seen it while testing a patch to implement the same
functionality in the ftdi_sio driver (patch not yet submitted).
The fix is pretty simple: don't reinitialize outstanding_urbs in
visor_open. (Again, I haven't tested the fix in visor, but I have
tested it in ftdi_sio.)
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / linux-beck.git / log