<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/drivers/char, branch v2.6.34.8</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/drivers/char?h=v2.6.34.8</id>
<link rel='self' href='https://git.amat.us/linux/atom/drivers/char?h=v2.6.34.8'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2011-01-06T23:08:19Z</updated>
<entry>
<title>pcmcia: synclink_cs: fix information leak to userland</title>
<updated>2011-01-06T23:08:19Z</updated>
<author>
<name>Vasiliy Kulikov</name>
<email>segooon@gmail.com</email>
</author>
<published>2010-10-17T14:41:24Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=c32b9b64959bdfa3fcf7ea693869e25de1201c7e'/>
<id>urn:sha1:c32b9b64959bdfa3fcf7ea693869e25de1201c7e</id>
<content type='text'>
commit 5b917a1420d3d1a9c8da49fb0090692dc9aaee86 upstream.

Structure new_line is copied to userland with some padding fields unitialized.
It leads to leaking of stack memory.

Signed-off-by: Vasiliy Kulikov &lt;segooon@gmail.com&gt;
Signed-off-by: Dominik Brodowski &lt;linux@dominikbrodowski.net&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
</content>
</entry>
<entry>
<title>sis-agp: Remove SIS 760, handled by amd64-agp</title>
<updated>2011-01-06T23:08:05Z</updated>
<author>
<name>Ben Hutchings</name>
<email>ben@decadent.org.uk</email>
</author>
<published>2010-03-24T03:33:48Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d62dadba708a48105dac8d5bffe0897c6386b5f6'/>
<id>urn:sha1:d62dadba708a48105dac8d5bffe0897c6386b5f6</id>
<content type='text'>
commit d831692a1a8e9ceaaa9bb16bb3fc503b7e372558 upstream.

SIS 760 is listed in the device tables for both amd64-agp and sis-agp.
amd64-agp is apparently preferable since it has workarounds for some
BIOS misconfigurations that sis-agp doesn't handle.

Signed-off-by: Ben Hutchings &lt;ben@decadent.org.uk&gt;
Signed-off-by: Dave Airlie &lt;airlied@redhat.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
</content>
</entry>
<entry>
<title>char: Mark /dev/zero and /dev/kmem as not capable of writeback</title>
<updated>2011-01-06T23:08:00Z</updated>
<author>
<name>Jan Kara</name>
<email>jack@suse.cz</email>
</author>
<published>2010-09-21T09:49:01Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=c97442183f1dc5f94dd714c28a2ec64ea548f0ad'/>
<id>urn:sha1:c97442183f1dc5f94dd714c28a2ec64ea548f0ad</id>
<content type='text'>
commit 371d217ee1ff8b418b8f73fb2a34990f951ec2d4 upstream.

These devices don't do any writeback but their device inodes still can get
dirty so mark bdi appropriately so that bdi code does the right thing and files
inodes to lists of bdi carrying the device inodes.

Signed-off-by: Jan Kara &lt;jack@suse.cz&gt;
Signed-off-by: Jens Axboe &lt;jaxboe@fusionio.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
</content>
</entry>
<entry>
<title>Char: nozomi, set tty-&gt;driver_data appropriately</title>
<updated>2010-08-13T20:27:17Z</updated>
<author>
<name>Jiri Slaby</name>
<email>jslaby@suse.cz</email>
</author>
<published>2010-07-29T09:46:32Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=8eab907a34e3f26a303d114be8f2cbf2daf52365'/>
<id>urn:sha1:8eab907a34e3f26a303d114be8f2cbf2daf52365</id>
<content type='text'>
commit bf9c1fca9ae9a79ed209e7ab2c10b3862f3f6f72 upstream.

Sorry, one more fix, this one depends on the other, so this is rather 2/2.
--

tty-&gt;driver_data is used all over the code, but never set. This
results in oopses like:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000130
IP: [&lt;ffffffff814a0040&gt;] mutex_lock+0x10/0x40
...
Pid: 2157, comm: modem-manager Not tainted 2.6.34.1-0.1-desktop #1 2768DR7/2768DR7
RIP: 0010:[&lt;ffffffff814a0040&gt;]  [&lt;ffffffff814a0040&gt;] mutex_lock+0x10/0x40
RSP: 0018:ffff88007b16fa50  EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000130 RCX: 0000000000000003
RDX: 0000000000000003 RSI: 0000000000000286 RDI: 0000000000000130
RBP: 0000000000001000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000130
R13: 0000000000000001 R14: 0000000000000000 R15: ffff88007b16feb4
...
Call Trace:
 [&lt;ffffffffa077690d&gt;] ntty_write_room+0x4d/0x90 [nozomi]
...

Set tty-&gt;driver_data to the computed port in .install to not recompute it in
every place where needed. Switch .open to use driver_data too.

Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;
Cc: Alan Cox &lt;alan@linux.intel.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>Char: nozomi, fix tty-&gt;count counting</title>
<updated>2010-08-13T20:27:16Z</updated>
<author>
<name>Jiri Slaby</name>
<email>jslaby@suse.cz</email>
</author>
<published>2010-07-29T09:09:47Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=bf7ce0291fb87742e7093c34bf858d6ae1cb41da'/>
<id>urn:sha1:bf7ce0291fb87742e7093c34bf858d6ae1cb41da</id>
<content type='text'>
commit ee78bb95b7bea08b7774a02073ea2bb45611a9e1 upstream.

Currently ntty_install omits to increment tty count and we get the
following warnings:
Warning: dev (noz2) tty-&gt;count(0) != #fd's(1) in tty_open

So to fix that, add one tty-&gt;count++ there.

Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;
Cc: Alan Cox &lt;alan@linux.intel.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>ipmi: set schedule_timeout_wait() value back to one</title>
<updated>2010-08-02T17:29:45Z</updated>
<author>
<name>Martin Wilck</name>
<email>martin.wilck@ts.fujitsu.com</email>
</author>
<published>2010-06-29T22:05:31Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=0dd6ec3a33cab9ee890c9ff671edea9202246c27'/>
<id>urn:sha1:0dd6ec3a33cab9ee890c9ff671edea9202246c27</id>
<content type='text'>
commit 8d1f66dc9b4f80a1441bc1c33efa98aca99e8813 upstream.

Fix a regression introduced by ae74e823cb7d ("ipmi: add parameter to limit
CPU usage in kipmid").

Some systems were seeing CPU usage go up dramatically with the recent
changes to try to reduce timer usage in the IPMI driver.  This was traced
down to schedule_timeout_interruptible(1) being changed to
schedule_timeout_interruptbile(0).  Revert that part of the change.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=16147

Reported-by: Thomas Jarosch &lt;thomas.jarosch@intra2net.com&gt;
Signed-off-by: Corey Minyard &lt;cminyard@mvista.com&gt;
Tested-by: Thomas Jarosch &lt;thomas.jarosch@intra2net.com&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>amd64-agp: Probe unknown AGP devices the right way</title>
<updated>2010-08-02T17:29:42Z</updated>
<author>
<name>Ben Hutchings</name>
<email>ben@decadent.org.uk</email>
</author>
<published>2010-03-24T03:36:31Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=104ce01cb95eb7eef75181b73f7c5a68a5d4b2a0'/>
<id>urn:sha1:104ce01cb95eb7eef75181b73f7c5a68a5d4b2a0</id>
<content type='text'>
commit 6fd024893911dcb51b4a0aa71971db5ba38f7071 upstream.

The current initialisation code probes 'unsupported' AGP devices
simply by calling its own probe function.  It does not lock these
devices or even check whether another driver is already bound to
them.

We must use the device core to manage this.  So if the specific
device id table didn't match anything and agp_try_unsupported=1,
switch the device id table and call driver_attach() again.

Signed-off-by: Ben Hutchings &lt;ben@decadent.org.uk&gt;
Signed-off-by: Dave Airlie &lt;airlied@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>tpm_tis: fix subsequent suspend failures</title>
<updated>2010-08-02T17:29:40Z</updated>
<author>
<name>Rajiv Andrade</name>
<email>srajiv@linux.vnet.ibm.com</email>
</author>
<published>2010-06-23T19:18:56Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d47f1d0c947a955b6d78befe9c6fa0928e4eff53'/>
<id>urn:sha1:d47f1d0c947a955b6d78befe9c6fa0928e4eff53</id>
<content type='text'>
commit 59f6fbe4291fcc078ba26ce4edf8373a7620a13a upstream.

Fix subsequent suspends by issuing tpm_continue_selftest during resume.
Otherwise, the tpm chip seems to be not fully initialized and will reject
the save state command during suspend, thus preventing the whole system
to suspend.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=16256

Signed-off-by: Helmut Schaa &lt;helmut.schaa@googlemail.com&gt;
Signed-off-by: Rajiv Andrade &lt;srajiv@linux.vnet.ibm.com&gt;
Cc: James Morris &lt;jmorris@namei.org&gt;
Cc: Debora Velarde &lt;debora@linux.vnet.ibm.com&gt;
Cc: David Safford &lt;safford@watson.ibm.com&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: James Morris &lt;jmorris@namei.org&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>TPM: ReadPubEK output struct fix</title>
<updated>2010-08-02T17:29:30Z</updated>
<author>
<name>Rajiv Andrade</name>
<email>srajiv@linux.vnet.ibm.com</email>
</author>
<published>2010-06-14T16:58:22Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=20bc0849443377b314abd94895ee6f79df821c78'/>
<id>urn:sha1:20bc0849443377b314abd94895ee6f79df821c78</id>
<content type='text'>
commit 02a077c52ef7631275a79862ffd9f3dbe9d38bc2 upstream.

This patch adds a missing element of the ReadPubEK command output,
that prevents future overflow of this buffer when copying the
TPM output result into it.

Prevents a kernel panic in case the user tries to read the
pubek from sysfs.

Signed-off-by: Rajiv Andrade &lt;srajiv@linux.vnet.ibm.com&gt;
Signed-off-by: James Morris &lt;jmorris@namei.org&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>ipmi: handle run_to_completion properly in deliver_recv_msg()</title>
<updated>2010-07-05T18:22:39Z</updated>
<author>
<name>Jiri Kosina</name>
<email>jkosina@suse.cz</email>
</author>
<published>2010-05-26T21:43:53Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=4b85622825dc6840081f5312450b16c903966d74'/>
<id>urn:sha1:4b85622825dc6840081f5312450b16c903966d74</id>
<content type='text'>
commit a747c5abc329611220f16df0bb4cf0ca4a7fdf0c upstream.

If run_to_completion flag is set, it means that we are running in a
single-threaded mode, and thus no locks are held.

This fixes a deadlock when IPMI notifier is being called during panic.

Signed-off-by: Jiri Kosina &lt;jkosina@suse.cz&gt;
Acked-by: Corey Minyard &lt;minyard@acm.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
</feed>
