Elina Pasheva [Thu, 11 Jun 2009 13:32:01 +0000 (14:32 +0100)]
sierra: driver interface blacklisting
Interface blacklisting is necessary for non-serial interfaces that are handled
by a different driver. The interface blacklisting is implemented in sierra
driver per device. Each device in need of a blacklist has a static information
array kept in the driver. This array contains the interface numbers that are
blacklisted. The pointer for each blacklist array and the length
of that blacklist are 'bundled' in data structure sierra_iface_info. A pointer
to this information is set in id_table when the device is added to the id_table.
The following is summary of changes we have made to sierra.c driver in
this patch dealing with interface blacklisting support:
- Added data structure sierra_iface_info and function is_blacklisted()
to support blacklisting
- Modified sierra_probe() to handle blacklisted interfaces accordingly
- Improved comments in id_table
- Added new device in id_table with blacklist interface support
Signed-off-by: Elina Pasheva <epasheva@sierrawireless.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Elina Pasheva [Thu, 11 Jun 2009 13:30:21 +0000 (14:30 +0100)]
sierra: driver urb handling improvements
[Folded from eight patches into one as the original set according to the
author "All of the patches need to be applied to obtain a working product"
so keeping them split seems unhelpful
Merge fixes done versus other conflicting changes and moved the
spin_lock_init from open to setup time -- Alan]
Summary of the changes and code re-organization in this patch:
- The memory for urbs is allocated and urbs are submitted only for the active
interfaces (instead of pre-allocating these for all interfaces). This will
save memory especially in the case of using composite devices.
- The code has been re-organized and functionality has been extracted from
sierra_startup(), sierra_shutdown(), sierra_open(), sierra_close() and added
in helper functions sierra_release_urb(), sierra_stop_rx_urbs(),
sierra_submit_rx_urbs() and sierra_setup_urb()
- Added function sierra_release_urb() to free an urb and its transfer
buffer.
- Removed unecessary include file reference and comment.
- Added function sierra_stop_rx_urbs() that takes care of the release of
receive and interrupt urbs. This function is to be called by sierra_close()
whenever an interface is de-activated.
- Added new function sierra_submit_rx_urbs() that handles the submission of
receive urbs and interrupt urbs (if any) during the interface activation.
This function is to be called by sierra_open(). Added a second parameter to
pass the memory allocation (as suggested by Oliver Neukum) so that this
function can be used in post_reset() and resume().
- Added new function sierra_setup_urb() that contains the functionality to
allocate an urb, fill bulk urb using the supplied memory allocation flag
and release urb upon error. Added parameter so that the caller pass the
memory allocation flag for flexibility.
- Moved sierra_close() before sierra_open() to resolve dependencies
introduced by the code reorganization.
- Modified sierra_close() to call sierra_stop_rx_urbs() and
sierra_release_urb() functions added in previous patch.
- Modified sierra_open() to call sierra_setup_urb() and sierra_submit_rx_urbs()
functions; note urbs are allocated and submitted for each activated interface.
- Modified sierra_startup() so that allocation of urbs happens whenever an
interface is activated (urb allocation is moved to sierra_open()).
- Modified sierra_shutdown() so that urbs are freed whenever an interface is
de-activated (urb freeing moved to sierra_close() as shown in previous patch
from the series)
- Removed unecessary data structure from sierra_port_private_data
- Suppress an entry in logs by not re-submitting an urb when usb_submit_urb()
returns -EPERM, as this shows that usb_kill_urb() is running (as suggested by
Oliver Neukum)
Signed-off-by: Elina Pasheva <epasheva@sierrawireless.com> Signed-off-by: Alan Cox <alan.cox@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 13:29:29 +0000 (14:29 +0100)]
tty: resolve some sierra breakage
The various merges into the sierra driver inadvertently undid
commit 212b8f0c3f5a2280bfa1d6ab13a6fe98552becaa by Elina Pasheva
<epasheva@sierrawireless.com>. Put it back so the OBEX port works again.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 13:03:13 +0000 (14:03 +0100)]
tty: Untangle termios and mm mutex dependencies
Although this doesn't cause any problems it could potentially do so for
future mmap using devices. No real work is needed to sort it out so untangle
it before it causes problems
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 12:56:16 +0000 (13:56 +0100)]
tty: fix bluetooth scribbling on low latency flags
Bluetooth shouldn't be doing this as most drivers don't support the flag,
furthermore it shouldn't be needed with newer buffering. This becomes rather
more visible as the locking fixes make the abuse of low_latency visible as
spew on the users console/dmesg.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 12:55:34 +0000 (13:55 +0100)]
ft232: support the ASYNC_LOW_LATENCY flag
This allows users to use the standard setserial command with this FT232
feature as well as obscure chip specific interfaces we have now. We keep
track of and respect the sysfs value for non-low-latency cases. In theory we
could do smart stuff with VTIME and the like but this seems of questionable
worth.
Closes-bug: http://bugzilla.kernel.org/show_bug.cgi?id=9120 Signed-off-by: Alan Cox <alan@linux.intel.com) Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Craig Shelley [Thu, 11 Jun 2009 12:54:40 +0000 (13:54 +0100)]
USB: CP210X Replace CP2101 with CP210x
This patch replaces the string "CP2101" with "CP210x" within cp210x.c
This is to reduce confusion about the fact that the driver is actually
compatible with CP2101, CP2102 and CP2103 devices.
Signed-off-by: Craig Shelley <craig@microtron.org.uk>
(Fixed some collisions merging)
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Craig Shelley [Thu, 11 Jun 2009 12:53:30 +0000 (13:53 +0100)]
USB: CP210X Use official request code definitions
The CP210X driver was developed without official device specifications.
This has lead to an incorrect assumption that all GET request codes are
equal to the corresponding SET request code +1.
This patch removes this incorrect assumption, and uses request code
definitions based on the updated GPL driver from SiLabs.
This modification is needed before extended functionality such as GPIO
on CP2103 can be supported.
Sonic Zhang [Thu, 11 Jun 2009 12:51:33 +0000 (13:51 +0100)]
Blackfin Serial Driver: disable dma rx interrupt only rather than all irqs
The UART RX handling code isn't exactly speeding, so don't go disabling
all interrupts when processing the buffer. Just disable the relevant DMA
interrupt. This greatly improves latency of the system when utilizing the
UART.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
05000231 - UART STB Bit Incorrectly Affects Receiver Setting
For processors affected by this, we cannot safely allow CSTOPB to be set
as the UART will then be unable to properly clock in bytes.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sonic Zhang [Thu, 11 Jun 2009 12:45:07 +0000 (13:45 +0100)]
Blackfin Serial Driver: annotate anomalies 05000215 and 05000099
Add some comments for how these anomalies are addressed:
05000215 - UART TX Interrupt Masked Erroneously
We always clear ETBEI within last UART TX interrupt to end
a string. It is always set when starting a new tx transfer.
05000099 - UART Line Status Register (UART_LSR) Bits Are Not Updated at
the Same Time
This anomaly affects driver only in POLL code where multi bits of
UART_LSR are checked. It doesn't affect current bfin_5xx.c driver.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sonic Zhang [Thu, 11 Jun 2009 12:42:57 +0000 (13:42 +0100)]
Blackfin Serial Driver: handle irregular DMA register status in auto start mode
This bug is caused by irregular behavior of DMA register CURR_X_COUNT
and CURR_Y_COUNT when an auto restart uart rx DMA run to last byte in
DMA buffer, trigger the interrupt and stay at this possiton. The status
of current x and y is 0:7 instead of 512:8 or 0:8. The driver doesn't
take care of this case when calculating the position.
URL: http://blackfin.uclinux.org/gf/tracker/5063 Reported-by: Tomasz Motylewski <t.motylewski@bfad.de> Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Graf Yang [Thu, 11 Jun 2009 12:42:17 +0000 (13:42 +0100)]
Blackfin Serial Driver: fix missing new lines when under load
Add a SSYNC() into bfin_serial_dma_tx_chars() to ensure DMA registers are
written with new data otherwise we might miss a byte or two when the
system is under load. PIO mode is OK though.
Signed-off-by: Graf Yang <graf.yang@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Robin Getz [Thu, 11 Jun 2009 12:38:57 +0000 (13:38 +0100)]
Blackfin Serial Driver: fix baudrate for early_printk
Since we already setup the early console UART in
arch/blackfin/kernel/early_printk.c, and common functions which are
enabled from the .setup will override the proper settings later, don't
fill in these structures. Otherwise we get mangled baudrate settings when
using early_printk.
Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mike Frysinger [Thu, 11 Jun 2009 12:37:11 +0000 (13:37 +0100)]
Blackfin SPORT UART: rewrite inline assembly
Hopefuly the new version is easier to read, but in the process it declares
proper clobber lists and better constraints so that GCC can do a better
job at allocating free registers.
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Blackfin SPORT UART: fix data misses while using transmit frame sync
SPORT transmit frame sync (TFS) isn't used as an electrical signal during
normal SPORT UART emulation. However, it is useful in EIA RS-485
emulation as RS-485 Transceiver Driver Enable DE strobe.
This patch configures:
TFS to be active high in order to drive an DE strobe of
an eventually connected RS-485 Transceiver.
Late frame sync mode (LATFS) gating the entire TX shift cycle.
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michael Buesch [Thu, 11 Jun 2009 12:06:31 +0000 (13:06 +0100)]
parport_pc: Fix subscription bugs
This patch fixes array subscription bugs in the parport_pc driver.
drivers/parport/parport_pc.c: In function ‘parport_irq_probe’:
drivers/parport/parport_pc.c:1589: warning: array subscript is above array bounds
drivers/parport/parport_pc.c: In function ‘parport_pc_probe_port’:
drivers/parport/parport_pc.c:1579: warning: array subscript is above array bounds
The patch also fixes a few other array bugs, which the compiler was
unable to find. Coding style violations are also fixed.
Signed-off-by: Michael Buesch <mb@bu3sch.de> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 12:04:27 +0000 (13:04 +0100)]
tty: remove buffer special casing
Long long ago a 4K kmalloc allocated two pages so the tty layer used the
page allocator, except on some machines where the page size was huge. This was
removed from the core tty layer with the tty buffer re-implementation but not
from tty_audit or the n_tty ldisc.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Olivier Bornet [Thu, 11 Jun 2009 11:54:20 +0000 (12:54 +0100)]
tty: iuu_phoenix: set termios.
set_termios can now be used for setting the parity and the stopbits. This is
needed to use with cards which use a different parity then the parity used at
start (even).
If the iuu_uart_baud function return an error, we will return the old_termios
instead of the new one.
Signed-off-by: Olivier Bornet <Olivier.Bornet@puck.ch>
This was then revamped to use the various helpers, not copy non-hardware
bits any to add mark/space parity and csize reporting
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:50:12 +0000 (12:50 +0100)]
tty: rewrite the ldisc locking
There are several pretty much unfixable races in the old ldisc code, especially
with respect to pty behaviour and also to hangup. It's easier to rewrite the
code than simply try and patch it up.
This patch
- splits the ldisc from the tty (so we will be able to refcount it more cleanly
later)
- introduces a mutex lock for ldisc changing on an active device
- fixes the complete mess that hangup caused
- implements hopefully correct setldisc/close/hangup locking
There are still some problems around pty pairs that have always been there but
at least it is now possible to understand the code and fix further problems.
This fixes the following known bugs
- hang up can leak ldisc references
- hang up may not call open/close on ldisc in a matched way
- pty/tty pairs can deadlock during an ldisc change
- reading the ldisc proc files can cause every ldisc to be loaded
and probably a few other of the mysterious ldisc race reports.
I'm sure it also adds the odd new one.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:46:41 +0000 (12:46 +0100)]
tty: Fix oops when scanning the polling list for kgdb
Costantino Leandro found a bug in tty_find_polling_driver and provided a
patch that fixed the crash but not the underlying bug. This fixes the
underlying bug where the list walk corrupts the values it is using on a
match but then reuses them if the open fails.
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:44:17 +0000 (12:44 +0100)]
tty: throttling race fix
The tty throttling code can race due to the lock drops. It takes very high
loads but this has been observed and verified by Rob Duncan.
The basic problem is that on an SMP box we can go
CPU #1 CPU #2
need to throttle ?
suppose we should buffer space cleared
are we throttled
yes ? - unthrottle
call throttle method
This changeet take the termios lock to protect against this. The termios
lock isn't the initial obvious candidate but many implementations of throttle
methods already need to poke around their own termios structures (and nobody
really locks them against a racing change of flow control).
This does mean that anyone who is setting tty->low_latency = 1 and then
calling tty_flip_buffer_push from their unthrottle method is going to end up
collapsing in a pile of locks. However we've removed all the known bogus
users of low_latency = 1 and such use isn't safe anyway for other reasons so
catching it would be an improvement.
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andre Przywara [Thu, 11 Jun 2009 11:41:57 +0000 (12:41 +0100)]
8250_pci: add the OXCB950 chip to the 8250 PCI driver.
This adds support for the following serial controller chip:
Oxford Semiconductor OXCB950 for PCI Cardbus interface
http://www.transdimension.com/products/serial/OXCB950.html
on this card:
ExSys EX-1370 1 port high-speed serial card for ExpressCard/34 slot
Signed-off-by: Andre Przywara <andre.przywara@amd.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jiri Slaby [Thu, 11 Jun 2009 11:29:27 +0000 (12:29 +0100)]
tty: cyclades, plx9060 casts cleanup
Remove ugly all-over-the-code casts of ctl_addr to 9060 space.
Add an union to the cyclades_card structure, which contains
a pointer to both 9050 and 9060 spaces.
The 9050 space layout is unknown, so let it still as a void
__iomem pointer.
Paul Fulghum [Thu, 11 Jun 2009 11:28:37 +0000 (12:28 +0100)]
tty: synclink_gt add receive pio mode
Add receive programmed IO mode to reduce receive latency
when using low data rates. The receive FIFO trigger
level of 128 bytes used in DMA mode creates excessive latency
when operating at low data rates. PIO mode is selected when user
application requests data in blocks of less than 128 bytes.
Signed-off-by: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:26:29 +0000 (12:26 +0100)]
tty: Bring the usb tty port structure into more use
This allows us to clean stuff up, but is probably also going to cause
some app breakage with buggy apps as we now implement proper POSIX behaviour
for USB ports matching all the other ports. This does also mean other apps
that break on USB will now work properly.
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:25:25 +0000 (12:25 +0100)]
tty: Implement a drain delay in the tty port
We need this for devices that cannot flush and wait, but which do not order
data and modem events. Without it we will hang up before all the data
clears the hardware. Needed for the USB changes.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Thu, 11 Jun 2009 11:24:17 +0000 (12:24 +0100)]
tty: Add carrier processing on close to the tty_port core
Some drivers implement this internally, others miss it out. Push the
behaviour into the core code as that way everyone will do it consistently.
Update the dtr rts method to raise or lower depending upon flags. Having a
single method in this style fits most of the implementations more cleanly than
two funtions.
We need this in place before we tackle the USB side
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Peter Botha [Wed, 10 Jun 2009 00:16:32 +0000 (17:16 -0700)]
char: mxser, fix ISA board lookup
There's a bug in the mxser kernel module that still appears in the
2.6.29.4 kernel.
mxser_get_ISA_conf takes a ioaddress as its first argument, by passing the
not of the ioaddr, you're effectively passing 0 which means it won't be
able to talk to an ISA card. I have tested this, and removing the !
fixes the problem.
Jan Kara [Tue, 9 Jun 2009 23:26:26 +0000 (16:26 -0700)]
jbd: fix race in buffer processing in commit code
In commit code, we scan buffers attached to a transaction. During this
scan, we sometimes have to drop j_list_lock and then we recheck whether
the journal buffer head didn't get freed by journal_try_to_free_buffers().
But checking for buffer_jbd(bh) isn't enough because a new journal head
could get attached to our buffer head. So add a check whether the journal
head remained the same and whether it's still at the same transaction and
list.
This is a nasty bug and can cause problems like memory corruption (use after
free) or trigger various assertions in JBD code (observed).
Signed-off-by: Jan Kara <jack@suse.cz> Cc: <stable@kernel.org> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Tue, 9 Jun 2009 23:26:24 +0000 (16:26 -0700)]
autofs4: remove hashed check in validate_wait()
The recent ->lookup() deadlock correction required the directory inode
mutex to be dropped while waiting for expire completion. We were
concerned about side effects from this change and one has been identified.
I saw several error messages.
They cause autofs to become quite confused and don't really point to the
actual problem.
Things like:
handle_packet_missing_direct:1376: can't find map entry for (43,1827932)
which is usually totally fatal (although in this case it wouldn't be
except that I treat is as such because it normally is).
do_mount_direct: direct trigger not valid or already mounted
/test/nested/g3c/s1/ss1
which is recoverable, however if this problem is at play it can cause
autofs to become quite confused as to the dependencies in the mount tree
because mount triggers end up mounted multiple times. It's hard to
accurately check for this over mounting case and automount shouldn't need
to if the kernel module is doing its job.
There was one other message, similar in consequence of this last one but I
can't locate a log example just now.
When checking if a mount has already completed prior to adding a new mount
request to the wait queue we check if the dentry is hashed and, if so, if
it is a mount point. But, if a mount successfully completed while we
slept on the wait queue mutex the dentry must exist for the mount to have
completed so the test is not really needed.
Mounts can also be done on top of a global root dentry, so for the above
case, where a mount request completes and the wait queue entry has already
been removed, the hashed test returning false can cause an incorrect
callback to the daemon. Also, d_mountpoint() is not sufficient to check
if a mount has completed for the multi-mount case when we don't have a
real mount at the base of the tree.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mike Frysinger [Tue, 9 Jun 2009 23:26:23 +0000 (16:26 -0700)]
shm: fix unused warnings on nommu
The massive nommu update (8feae131) resulted in these warnings:
ipc/shm.c: In function `sys_shmdt':
ipc/shm.c:974: warning: unused variable `size'
ipc/shm.c:972: warning: unused variable `next'
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: David Howells <dhowells@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
cls_cgroup: Fix oops when user send improperly 'tc filter add' request
r8169: fix crash when large packets are received
Linus Torvalds [Tue, 9 Jun 2009 15:41:22 +0000 (08:41 -0700)]
Merge branch 'for-linus' of git://neil.brown.name/md
* 'for-linus' of git://neil.brown.name/md:
md/raid5: fix bug in reshape code when chunk_size decreases.
md/raid5 - avoid deadlocks in get_active_stripe during reshape
md/raid5: use conf->raid_disks in preference to mddev->raid_disk
it turns out that we need clear cpus_hardware_enabled in that case.
Reported-and-tested-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Minoru Usui [Tue, 9 Jun 2009 11:03:09 +0000 (04:03 -0700)]
cls_cgroup: Fix oops when user send improperly 'tc filter add' request
I found a bug in cls_cgroup_change() in cls_cgroup.c.
cls_cgroup_change() expected tca[TCA_OPTIONS] was set from user space properly,
but tc in iproute2-2.6.29-1 (which I used) didn't set it.
In the current source code of tc in git, it set tca[TCA_OPTIONS].
If we always use a newest iproute2 in git when we use cls_cgroup,
we don't face this oops probably.
But I think, kernel shouldn't panic regardless of use program's behaviour.
Signed-off-by: Minoru Usui <usui@mxm.nes.nec.co.jp> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Tue, 9 Jun 2009 11:01:02 +0000 (04:01 -0700)]
r8169: fix crash when large packets are received
Michael Tokarev reported receiving a large packet could crash
a machine with RTL8169 NIC.
( original thread at http://lkml.org/lkml/2009/6/8/192 )
Problem is this driver tells that NIC frames up to 16383 bytes
can be received but provides skb to rx ring allocated with
smaller sizes (1536 bytes in case standard 1500 bytes MTU is used)
When a frame larger than what was allocated by driver is received,
dma transfert can occurs past the end of buffer and corrupt
kernel memory.
Fix is to tell to NIC what is the maximum size a frame can be.
This bug is very old, (before git introduction, linux-2.6.10), and
should be backported to stable versions.
Reported-by: Michael Tokarev <mjt@tls.msk.ru> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Michael Tokarev <mjt@tls.msk.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
NeilBrown [Tue, 9 Jun 2009 06:32:22 +0000 (16:32 +1000)]
md/raid5: fix bug in reshape code when chunk_size decreases.
Now that we support changing the chunksize, we calculate
"reshape_sectors" to be the max of number of sectors in old
and new chunk size.
However there is one please where we still use 'chunksize'
rather than 'reshape_sectors'.
This causes a reshape that reduces the size of chunks to freeze.
NeilBrown [Tue, 9 Jun 2009 04:39:59 +0000 (14:39 +1000)]
md/raid5 - avoid deadlocks in get_active_stripe during reshape
md has functionality to 'quiesce' and array so that all pending
IO completed and no new IO starts. This is used to achieve a
stable state before making internal changes.
Currently this quiescing applies equally to normal IO, resync
IO, and reshape IO.
However there is a problem with applying it to reshape IO.
Reshape can have multiple 'stripe_heads' that must be active together.
If the quiesce come between allocating the first and the last of
such a collection, then we deadlock, as the last will not be allocated
until the quiesce is lifted, the quiesce will not be lifted until the
first (which has been allocated) gets used, and that first cannot be
used until the last is allocated.
It is not necessary to inhibit reshape IO when a quiesce is
requested. Those places in the code that require a full quiesce will
ensure the reshape thread is not running at all.
So allow reshape requests to get access to new stripe_heads without
being blocked by a 'quiesce'.
This only affects in-place reshapes (i.e. where the array does not
grow or shrink) and these are only newly supported. So this patch is
not needed in earlier kernels.
Linus Torvalds [Mon, 8 Jun 2009 19:31:53 +0000 (12:31 -0700)]
async: Fix lack of boot-time console due to insufficient synchronization
Our async work synchronization was broken by "async: make sure
independent async domains can't accidentally entangle" (commit d5a877e8dd409d8c702986d06485c374b705d340), because it would report
the wrong lowest active async ID when there was both running and
pending async work.
This caused things like no being able to read the root filesystem,
resulting in missing console devices and inability to run 'init',
causing a boot-time panic.
This fixes it by properly returning the lowest pending async ID: if
there is any running async work, that will have a lower ID than any
pending work, and we should _not_ look at the pending work list.
There were alternative patches from Jaswinder and James, but this one
also cleans up the code by removing the pointless 'ret' variable and
the unnecesary testing for an empty list around 'for_each_entry()' (if
the list is empty, the for_each_entry() thing just won't execute).
Fixes-bug: http://bugzilla.kernel.org/show_bug.cgi?id=13474 Reported-and-tested-by: Chris Clayton <chris2553@googlemail.com> Cc: Jaswinder Singh Rajput <jaswinder@kernel.org> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Mon, 8 Jun 2009 16:22:53 +0000 (09:22 -0700)]
Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
MIPS: Outline udelay and fix a few issues.
MIPS: ioctl.h: Fix headers_check warnings
MIPS: Cobalt: PCI bus is always required to obtain the board ID
MIPS: Kconfig: Remove "Support for" from Cavium system type
MIPS: Sibyte: Honor CONFIG_CMDLINE
SSB: BCM47xx: Export ssb_watchdog_timer_set
Ralf Baechle [Sat, 28 Feb 2009 09:44:28 +0000 (09:44 +0000)]
MIPS: Outline udelay and fix a few issues.
Outlining fixes the issue were on certain CPUs such as the R10000 family
the delay loop would need an extra cycle if it overlaps a cacheline
boundary.
The rewrite also fixes build errors with GCC 4.4 which was changed in
way incompatible with the kernel's inline assembly.
Relying on pure C for computation of the delay value removes the need for
explicit. The price we pay is a slight slowdown of the computation - to
be fixed on another day.
Linus Torvalds [Mon, 8 Jun 2009 15:29:31 +0000 (08:29 -0700)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm
* master.kernel.org:/home/rmk/linux-2.6-arm:
[ARM] 5543/1: arm: serial amba: add missing declaration in serial.h
[ARM] pxa: fix pxa27x_udc default pullup GPIO
[ARM] pxa/imote2: fix UCAM sensor board ADC model number
mx[23]: don't put clock lookups in __initdata
fix oops when using console=ttymxcN with N > 0
[ARM] ARMv7 errata: only apply fixes when running on applicable CPU
[ARM] 5534/1: kmalloc must return a cache line aligned buffer
Linus Torvalds [Mon, 8 Jun 2009 14:53:59 +0000 (07:53 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc:
sdhci-of: Fix the wrong accessor to HOSTVER register
mvsdio: fix config failure with some high speed SDHC cards
mvsdio: ignore high speed timing requests from the core
mmc/omap: Use disable_irq_nosync() from within irq handlers.
sdhci-of: Add fsl,esdhc as a valid compatible to bind against
mvsdio: allow automatic loading when modular
mxcmmc: Fix missing return value checking in DMA setup code.
mxcmmc : Reset the SDHC hardware if software timeout occurs.
omap_hsmmc: Trivial fix for a typo in comment
mxcmmc: decrease minimum frequency to make MMC cards work
Avi Kivity [Sat, 6 Jun 2009 09:34:39 +0000 (12:34 +0300)]
KVM: Explicity initialize cpus_hardware_enabled
Under CONFIG_MAXSMP, cpus_hardware_enabled is allocated from the heap and
not statically initialized. This causes a crash on reboot when kvm thinks
vmx is enabled on random nonexistent cpus and accesses nonexistent percpu
lists.
Fix by explicitly clearing the variable.
Cc: stable@kernel.org Reported-and-tested-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Avi Kivity <avi@redhat.com>