Nitin Gupta [Tue, 1 Jun 2010 08:01:23 +0000 (13:31 +0530)]
Staging: ramzswap: Support generic I/O requests
Currently, ramzwap devices (/dev/ramzswapX) can only
be used as swap disks since it was hard-coded to consider
only the first request in bio vector.
Now, we iterate over all the segments in an incoming
bio which allows us to handle all kinds of I/O requests.
ramzswap devices can still handle PAGE_SIZE aligned and
multiple of PAGE_SIZE sized I/O requests only. To ensure
that we get always get such requests only, we set following
request_queue attributes to PAGE_SIZE:
- physical_block_size
- logical_block_size
- io_min
- io_opt
Note: physical and logical block sizes were already set
equal to PAGE_SIZE and that seems to be sufficient to get
PAGE_SIZE aligned I/O.
Since we are no longer limited to handling swap requests
only, the next few patches rename ramzswap to zram. So,
the devices will then be called /dev/zram{0, 1, 2, ...}
Usage/Examples:
1) Use as /tmp storage
- mkfs.ext4 /dev/zram0
- mount /dev/zram0 /tmp
2) Use as swap:
- mkswap /dev/zram0
- swapon /dev/zram0 -p 10 # give highest priority to zram0
Performance:
- I/O benchamark done with 'dd' command. Details can be
found here:
http://code.google.com/p/compcache/wiki/zramperf
Summary:
- Maximum read speed (approx):
- ram disk: 1200 MB/sec
- zram disk: 600 MB/sec
- Maximum write speed (approx):
- ram disk: 500 MB/sec
- zram disk: 160 MB/sec
Issues:
- Double caching: We can potentially waste memory by having
two copies of a page -- one in page cache (uncompress) and
second in the device memory (compressed). However, during
reclaim, clean page cache pages are quickly freed, so this
does not seem to be a big problem.
- Stale data: Not all filesystems support issuing 'discard'
requests to underlying block devices. So, if such filesystems
are used over zram devices, we can accumulate lot of stale
data in memory. Even for filesystems to do support discard
(example, ext4), we need to see how effective it is.
- Scalability: There is only one (per-device) de/compression
buffer stats. This can lead to significant contention, especially
when used for generic (non-swap) purposes.
Naveen Jain [Wed, 9 Jun 2010 12:20:40 +0000 (07:20 -0500)]
Staging: ti-st: fix protocol counting
Older mechanism of counting the number of protocols
registered with ST was slow, in-efficient.
It used to check the protocol data for NULL for each
registration/unregistration.
With this change, counting protocols in maintained by
a single counter protos_registered.
Naveen Jain [Wed, 9 Jun 2010 08:45:32 +0000 (03:45 -0500)]
Staging: ti-st: cleanup sysfs code
Remove the older way of communicating with user-space
daemon UIM.
The older way involved creating of a new kobj and pid
sysfs file under it, and sending signal using that pid.
Now we communicate via rfkill to user-space UIM.
Background: UIM is the user-space daemon which upon
notification from ldisc driver, opens the tty, sets
default baud and then installs (tiocsetd) the ldisc.
Pavan Savoy [Mon, 24 May 2010 14:14:09 +0000 (19:44 +0530)]
Staging: Update TI-ST's TODO
Update the TODO of TI-ST to make sure it reflects current list
of activities that need to be done.
Also point it out to the user-space app code relevant to it.
Otavio Salvador [Fri, 21 May 2010 22:08:14 +0000 (19:08 -0300)]
Staging: vt6656: fix up built-in compilation together with rt2870
Fix this build error:
drivers/staging/vt6656/built-in.o: In function `rotr1':
(.text+0x1a878): multiple definition of `rotr1'
drivers/staging/rt2870/built-in.o:(.text+0x106c2): first defined here
drivers/staging/vt6656/built-in.o: In function `tkip_sbox':
(.text+0x1a848): multiple definition of `tkip_sbox'
drivers/staging/rt2870/built-in.o:(.text+0x10697): first defined here
drivers/staging/vt6656/built-in.o: In function `xor_32':
(.text+0x1ec24): multiple definition of `xor_32'
drivers/staging/rt2870/built-in.o:(.text+0x111c4): first defined here
drivers/staging/vt6656/built-in.o: In function `xor_128':
(.text+0x1ec00): multiple definition of `xor_128'
drivers/staging/rt2870/built-in.o:(.text+0x111dd): first defined here
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Commit 902d2411298a6e0b8a473a0163aa4bf6d5c8a073 converted U32 to u32
which you would think would be just fine. Unfortunatly, it seems that
VIA only builds their code on a 32bit processor (which makes sense if
you think about it), but this doesn't work on x86-64. So fix up the few
places where this really wanted to be an unsigned long width.
Cc: Charles Clément <caratorn@gmail.com> Cc: Forest Bond <forest@alittletooquiet.net>, Cc: Andres More <more.andres@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Charles Clément [Wed, 19 May 2010 18:30:51 +0000 (11:30 -0700)]
Staging: vt6655: use is_multicast_ether_addr instead of custom macro
Replace custom macro IS_MULTICAST_ADDRESS by is_multicast_ether_addr
from <linux/etherdevice.h>.
Remove linux/if_ether.h include as it is included in
linux/etherdevice.h already.
Signed-off-by: Charles Clément <caratorn@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Timofey Trofimov [Tue, 25 May 2010 14:19:54 +0000 (18:19 +0400)]
Staging: winbond: fix some checkpatch.pl issues in phy_calibration.c
This is a patch to the phy_calibration.c that fixes up almost all
warnings and errors (except 80 characters limit and lack of tabs errors)
found by the checkpatch.pl tool
Gustavo Silva [Fri, 18 Jun 2010 05:01:34 +0000 (00:01 -0500)]
Staging: comedi: drivers: fix coding style issues in pcl812.c
This is a patch to the pcl812.c file that fixes up the following
issues:
ERROR: code indent should use tabs where possible x 27
WARNING: line over 80 characters x 37
WARNING: please, no space before tabs x 13
WARNING: braces {} are not necessary for single statement blocks x 2
WARNING: printk() should include KERN_ facility level x 22
WARNING: braces {} are not necessary for any arm of this statement x 5
Ian Abbott [Mon, 14 Jun 2010 11:49:27 +0000 (12:49 +0100)]
Staging: Comedi: drivers: Remove comment copied from skel.c
The dmm32at.c and comedi_bond.c comedi driver files contain an
instructional comment block copied over from skel.c about how to format
a driver comment block. This comment was modified in skel.c by a
previous patch in this series to stop Comedi's 'dump_doc' script
treating it as an actual driver comment block. There isn't any need to
repeat this comment block in the other source files, so rather than
modify it, this patch just removes it from those files.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Mon, 14 Jun 2010 11:49:26 +0000 (12:49 +0100)]
Staging: comedi: drivers: skel: Add whitespace to example comment block
The Comedi team have a script 'dump_doc' to extract text from the driver
comment block (starting with line 'Driver:') in (most of) the comedi
driver source files. This was recently updated to allow and strip off a
" * " prefix at the start of each line (well actually, it uses the perl
substitution 's/^ ?\* ?//').
The skel.c file contains an instructional comment block about how to
format this driver comment block, but the updated 'dump_doc' script
mistakenly treats this as a valid driver comment block. This patch adds
some extra whitespace to stop the instructional comment block being
treated as a valid driver comment block by Comedi's 'dump_doc' script.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Mon, 14 Jun 2010 11:25:05 +0000 (12:25 +0100)]
Staging: comedi: drivers: das08: Add whitespace to driver comment block
Previous whitespace changes to das08.c removed some whitespace from a
"driver comment block" that the Comedi team would like to keep. "Header
continuation lines" in this comment block should be indented with
whitespace. (This is after the " * " at the start of each line.)
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Gustavo Silva [Wed, 16 Jun 2010 06:01:07 +0000 (01:01 -0500)]
Staging: comedi: drivers: fix coding style issues in quatech_daqp_cs.c
This is a patch to the quatech_daqp_cs.c file that fixes up the following
issues:
ERROR: code indent should use tabs where possible x 1
WARNING: line over 80 characters x 1
WARNING: braces {} are not necessary for single statement blocks x 10
WARNING: printk() should include KERN_ facility level x 6
Gustavo Silva [Wed, 16 Jun 2010 05:17:52 +0000 (00:17 -0500)]
Staging: comedi: drivers: fix coding style issues in pcl816.c
This is a patch to the pcl816.c file that fixes up the following
issues:
ERROR: code indent should use tabs where possible x 2
WARNING: line over 80 characters x 34
WARNING: please, no space before tabs x 1
WARNING: braces {} are not necessary for single statement blocks x 6
WARNING: printk() should include KERN_ facility level x 15
WARNING: braces {} are not necessary for any arm of this statement x 1
Gustavo Silva [Sat, 12 Jun 2010 21:39:18 +0000 (16:39 -0500)]
Staging: comedi: drivers: fix coding style issues in me4000.c
This is a patch to the me4000.c file that fixes up the following
issues:
ERROR: space required after that close brace '}' x 13
ERROR: spaces required around that ':' (ctx:VxV) x 3
WARNING: line over 80 characters x 96
WARNING: braces {} are not necessary for any arm of this statement x 2
John Sheehan [Fri, 28 May 2010 15:33:25 +0000 (16:33 +0100)]
Staging: comedi: fix code warnings in s626.c
A patch for s626.c to fix some of the warnings
reported by the checkpatch.pl tool, namely,
printk() should include KERN_ facility level
unnecessary whitespace before a quoted newline
Signed-off-by: John Sheehan <john.d.sheehan@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
These checks are obviously pointless because kfree() can handle null
dereferences.
But really the main problem is that if the pointers were null that would
cause problems on the ealier lines. The dereferences would cause an
oops and the _release() functions use ->priv to determine which IRQ to
free. I looked into it and quite a few of the detach functions assume
link->priv is non-null. It seems like we can remove these checks.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Wed, 19 May 2010 13:10:03 +0000 (14:10 +0100)]
Staging: comedi: serial2002: Reduce stack usage on 'open'
Reduce stack usage in serial_2002_open() by allocating dig_in_config,
dig_out_config, chan_in_config, and chan_out_config temporary arrays
using kcalloc() and freeing them when done with.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Wed, 19 May 2010 13:10:01 +0000 (14:10 +0100)]
Staging: comedi: serial2002: handle allocation failures on 'open'
The comedi device 'open' method for the serial2002 driver frees any old
'maxdata_list' and 'range_table_list' arrays belonging to a subdevice
and allocates them again, but was missing checks for allocation failure.
If an allocation fails, free the 'maxdata_list' and 'range_table_list'
arrays for all subdevices and return an error.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ian Abbott [Wed, 19 May 2010 13:10:00 +0000 (14:10 +0100)]
Staging: comedi: Allow 'open' driver method to fail
Some comedi drivers should return an error from their 'open' method when
something goes wrong. Change the prototype of the 'open' method in
'struct comedi_device' to allow this, and change the drivers that use it.
Propagate any error to the 'open' file operation.
The corresponding 'close' method won't be called when the 'open' method
fails, so drivers failing the 'open' need to clean up any mess they
created.
The dt9812 and serial2002 drivers can now return an error on 'open'.
The jr3_pci driver also uses the 'open' method but doesn't fail it.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
wimax/i2400m: fix missing endian correction read in fw loader
net8139: fix a race at the end of NAPI
pktgen: Fix accuracy of inter-packet delay.
pkt_sched: gen_estimator: add a new lock
net: deliver skbs on inactive slaves to exact matches
ipv6: fix ICMP6_MIB_OUTERRORS
r8169: fix mdio_read and update mdio_write according to hw specs
gianfar: Revive the driver for eTSEC devices (disable timestamping)
caif: fix a couple range checks
phylib: Add support for the LXT973 phy.
net: Print num_rx_queues imbalance warning only when there are allocated queues
Linus Torvalds [Fri, 11 Jun 2010 21:18:47 +0000 (14:18 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: The file argument for fsync() is never null
Btrfs: handle ERR_PTR from posix_acl_from_xattr()
Btrfs: avoid BUG when dropping root and reference in same transaction
Btrfs: prohibit a operation of changing acl's mask when noacl mount option used
Btrfs: should add a permission check for setfacl
Btrfs: btrfs_lookup_dir_item() can return ERR_PTR
Btrfs: btrfs_read_fs_root_no_name() returns ERR_PTRs
Btrfs: unwind after btrfs_start_transaction() errors
Btrfs: btrfs_iget() returns ERR_PTR
Btrfs: handle kzalloc() failure in open_ctree()
Btrfs: handle error returns from btrfs_lookup_dir_item()
Btrfs: Fix BUG_ON for fs converted from extN
Btrfs: Fix null dereference in relocation.c
Btrfs: fix remap_file_pages error
Btrfs: uninitialized data is check_path_shared()
Btrfs: fix fallocate regression
Btrfs: fix loop device on top of btrfs
Linus Torvalds [Fri, 11 Jun 2010 21:15:44 +0000 (14:15 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
PCI: clear bridge resource range if BIOS assigned bad one
PCI: hotplug/cpqphp, fix NULL dereference
Revert "PCI: create function symlinks in /sys/bus/pci/slots/N/"
PCI: change resource collision messages from KERN_ERR to KERN_INFO
Yinghai Lu [Thu, 3 Jun 2010 20:43:03 +0000 (13:43 -0700)]
PCI: clear bridge resource range if BIOS assigned bad one
Yannick found that video does not work with 2.6.34. The cause of this
bug was that the BIOS had assigned the wrong range to the PCI bridge
above the video device. Before 2.6.34 the kernel would have shrunk
the size of the bridge window, but since d65245c PCI: don't shrink bridge resources
the kernel will avoid shrinking BIOS ranges.
So zero out the old range if we fail to claim it at boot time; this will
cause us to allocate a new range at startup, restoring the 2.6.34
behavior.
Jiri Slaby [Wed, 9 Jun 2010 20:31:13 +0000 (22:31 +0200)]
PCI: hotplug/cpqphp, fix NULL dereference
There are devices out there which are PCI Hot-plug controllers with
compaq PCI IDs, but are not bridges, hence have pdev->subordinate
NULL. But cpqphp expects the pointer to be non-NULL.
Add a check to the probe function to avoid oopses like:
BUG: unable to handle kernel NULL pointer dereference at 00000050
IP: [<f82e3c41>] cpqhpc_probe+0x951/0x1120 [cpqphp]
*pdpt = 0000000033779001 *pde = 0000000000000000
...
Bjorn Helgaas [Thu, 3 Jun 2010 19:47:18 +0000 (13:47 -0600)]
PCI: change resource collision messages from KERN_ERR to KERN_INFO
We can often deal with PCI resource issues by moving devices around. In
that case, there's no point in alarming the user with messages like these.
There are many bug reports where the message itself is the only problem,
e.g., https://bugs.launchpad.net/ubuntu/+source/linux/+bug/413419 .
Dan Carpenter [Sat, 29 May 2010 09:49:07 +0000 (09:49 +0000)]
Btrfs: The file argument for fsync() is never null
The "file" argument for fsync is never null so we can remove this check.
What drew my attention here is that 7ea8085910e: "drop unused dentry
argument to ->fsync" introduced an unconditional dereference at the
start of the function and that generated a smatch warning.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Sage Weil [Mon, 17 May 2010 17:15:27 +0000 (17:15 +0000)]
Btrfs: avoid BUG when dropping root and reference in same transaction
If btrfs_ioctl_snap_destroy() deletes a snapshot but finishes
with end_transaction(), the cleaner kthread may come in and
drop the root in the same transaction. If that's the case, the
root's refs still == 1 in the tree when btrfs_del_root() deletes
the item, because commit_fs_roots() hasn't updated it yet (that
happens during the commit).
This wasn't a problem before only because
btrfs_ioctl_snap_destroy() would commit the transaction before dropping
the dentry reference, so the dead root wouldn't get queued up until
after the fs root item was updated in the btree.
Since it is not an error to drop the root reference and the root in the
same transaction, just drop the BUG_ON() in btrfs_del_root().
Signed-off-by: Sage Weil <sage@newdream.net> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Shi Weihua [Tue, 18 May 2010 00:51:54 +0000 (00:51 +0000)]
Btrfs: prohibit a operation of changing acl's mask when noacl mount option used
when used Posix File System Test Suite(pjd-fstest) to test btrfs,
some cases about setfacl failed when noacl mount option used.
I simplified used commands in pjd-fstest, and the following steps
can reproduce it.
------------------------
# cd btrfs-part/
# mkdir aaa
# setfacl -m m::rw aaa <- successed, but not expected by pjd-fstest.
------------------------
I checked ext3, a warning message occured, like as:
setfacl: aaa/: Operation not supported
Certainly, it's expected by pjd-fstest.
So, i compared acl.c of btrfs and ext3. Based on that, a patch created.
Fortunately, it works.
Signed-off-by: Shi Weihua <shiwh@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Shi Weihua [Tue, 18 May 2010 00:50:32 +0000 (00:50 +0000)]
Btrfs: should add a permission check for setfacl
On btrfs, do the following
------------------
# su user1
# cd btrfs-part/
# touch aaa
# getfacl aaa
# file: aaa
# owner: user1
# group: user1
user::rw-
group::rw-
other::r--
# su user2
# cd btrfs-part/
# setfacl -m u::rwx aaa
# getfacl aaa
# file: aaa
# owner: user1
# group: user1
user::rwx <- successed to setfacl
group::rw-
other::r--
------------------
but we should prohibit it that user2 changing user1's acl.
In fact, on ext3 and other fs, a message occurs:
setfacl: aaa: Operation not permitted
This patch fixed it. Signed-off-by: Shi Weihua <shiwh@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_read_fs_root_no_name() returns ERR_PTRs on error so I added a
check for that. It's not clear to me if it can also return NULL
pointers or not so I left the original NULL pointer check as is.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Dan Carpenter [Sat, 29 May 2010 09:46:47 +0000 (09:46 +0000)]
Btrfs: unwind after btrfs_start_transaction() errors
This was added by a22285a6a3: "Btrfs: Integrate metadata reservation
with start_transaction". If we goto out here then we skip all the
unwinding and there are locks still held etc.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Linus Torvalds [Fri, 11 Jun 2010 16:55:50 +0000 (09:55 -0700)]
Merge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6
* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
kbuild: Create output directory in Makefile.modbuiltin
kbuild: Generate modules.builtin in make modules
Linus Torvalds [Fri, 11 Jun 2010 16:55:21 +0000 (09:55 -0700)]
Merge branch 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia-2.6
* 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia-2.6:
pcmcia: avoid validate_cis failure on CIS override
pcmcia: dev_node removal bugfix
pcmcia: yenta_socket.c Remove extra #ifdef CONFIG_YENTA_TI
pcmcia: only keep saved I365_CSCINT flag if there is no PCI irq
Linus Torvalds [Fri, 11 Jun 2010 16:52:23 +0000 (09:52 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
ceph: try to send partial cap release on cap message on missing inode
ceph: release cap on import if we don't have the inode
ceph: fix misleading/incorrect debug message
ceph: fix atomic64_t initialization on ia64
ceph: fix lease revocation when seq doesn't match
ceph: fix f_namelen reported by statfs
ceph: fix memory leak in statfs
ceph: fix d_subdirs ordering problem
Dan Carpenter [Tue, 1 Jun 2010 08:23:11 +0000 (08:23 +0000)]
Btrfs: uninitialized data is check_path_shared()
refs can be used with uninitialized data if btrfs_lookup_extent_info()
fails on the first pass through the loop. In the original code if that
happens then check_path_shared() probably returns 1, this patch
changes it to return 1 for safety.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>