<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/arch, branch v2.6.32.4</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/arch?h=v2.6.32.4</id>
<link rel='self' href='https://git.amat.us/linux/atom/arch?h=v2.6.32.4'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2010-01-18T18:19:51Z</updated>
<entry>
<title>module: handle ppc64 relocating kcrctabs when CONFIG_RELOCATABLE=y</title>
<updated>2010-01-18T18:19:51Z</updated>
<author>
<name>Rusty Russell</name>
<email>rusty@rustcorp.com.au</email>
</author>
<published>2009-12-15T22:28:32Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=54f1b39ce06aaf023db558ce4cc73f1d550d0d53'/>
<id>urn:sha1:54f1b39ce06aaf023db558ce4cc73f1d550d0d53</id>
<content type='text'>
commit d4703aefdbc8f9f347f6dcefcddd791294314eb7 upstream.

powerpc applies relocations to the kcrctab.  They're absolute symbols,
but it's not completely unreasonable: other archs may too, but the
relocation is often 0.

http://lists.ozlabs.org/pipermail/linuxppc-dev/2009-November/077972.html

Inspired-by: Neil Horman &lt;nhorman@tuxdriver.com&gt;
Signed-off-by: Rusty Russell &lt;rusty@rustcorp.com.au&gt;
Tested-by: Neil Horman &lt;nhorman@tuxdriver.com&gt;
Acked-by: Paul Mackerras &lt;paulus@samba.org&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>Revert "x86: Side-step lguest problem by only building cmpxchg8b_emu for pre-Pentium"</title>
<updated>2010-01-18T18:19:16Z</updated>
<author>
<name>Rusty Russell</name>
<email>rusty@rustcorp.com.au</email>
</author>
<published>2010-01-05T02:18:49Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=b7e5f77916291b3c80fdf34dd2a649d8d10138e8'/>
<id>urn:sha1:b7e5f77916291b3c80fdf34dd2a649d8d10138e8</id>
<content type='text'>
commit db677ffa5f5a4f15b9dad4d132b3477b80766d82 upstream.

This reverts commit ae1b22f6e46c03cede7cea234d0bf2253b4261cf.

As Linus said in 982d007a6ee: "There was something really messy about
cmpxchg8b and clone CPU's, so if you enable it on other CPUs later, do it
carefully."

This breaks lguest for those configs, but we can fix that by emulating
if we have to.

Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=14884
Signed-off-by: Rusty Russell &lt;rusty@rustcorp.com.au&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>untangle the do_mremap() mess</title>
<updated>2010-01-18T18:19:11Z</updated>
<author>
<name>Al Viro</name>
<email>viro@zeniv.linux.org.uk</email>
</author>
<published>2010-01-14T19:39:39Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=1f51eb3a881359e97dc2c228e55c83fba598e349'/>
<id>urn:sha1:1f51eb3a881359e97dc2c228e55c83fba598e349</id>
<content type='text'>
This backports the following upstream commits all as one patch:
	54f5de709984bae0d31d823ff03de755f9dcac54
	ecc1a8993751de4e82eb18640d631dae1f626bd6
	1a0ef85f84feb13f07b604fcf5b90ef7c2b5c82f
	f106af4e90eadd76cfc0b5325f659619e08fb762
	097eed103862f9c6a97f2e415e21d1134017b135
	935874141df839c706cd6cdc438e85eb69d1525e
	0ec62d290912bb4b989be7563851bc364ec73b56
	c4caa778157dbbf04116f0ac2111e389b5cd7a29
	2ea1d13f64efdf49319e86c87d9ba38c30902782
	570dcf2c15463842e384eb597a87c1e39bead99b
	564b3bffc619dcbdd160de597b0547a7017ea010
	0067bd8a55862ac9dd212bd1c4f6f5bff1ca1301
	f8b7256096a20436f6d0926747e3ac3d64c81d24
	8c7b49b3ecd48923eb64ff57e07a1cdb74782970
	9206de95b1ea68357996ec02be5db0638a0de2c1
	2c6a10161d0b5fc047b5bd81b03693b9af99fab5
	05d72faa6d13c9d857478a5d35c85db9adada685
	bb52d6694002b9d632bb355f64daa045c6293a4e
	e77414e0aad6a1b063ba5e5750c582c75327ea6a
	aa65607373a4daf2010e8c3867b6317619f3c1a3

Backport done by Greg Kroah-Hartman.  Only minor tweaks were needed.

Cc: David S. Miller &lt;davem@davemloft.net&gt;
Cc: Hugh Dickins &lt;hugh.dickins@tiscali.co.uk&gt;
Cc: Paul Mundt &lt;lethal@linux-sh.org&gt;
Cc: Russell King &lt;rmk+kernel@arm.linux.org.uk&gt;
Cc: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
Signed-off-by: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>x86, msr: msrs_alloc/free for CONFIG_SMP=n</title>
<updated>2010-01-06T23:05:12Z</updated>
<author>
<name>Borislav Petkov</name>
<email>petkovbb@googlemail.com</email>
</author>
<published>2009-12-16T23:16:25Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=2d9e1f02c8fec8e339eddf6c92e836c18525d328'/>
<id>urn:sha1:2d9e1f02c8fec8e339eddf6c92e836c18525d328</id>
<content type='text'>
commit 6ede31e03084ee084bcee073ef3d1136f68d0906 upstream.

Randy Dunlap reported the following build error:

"When CONFIG_SMP=n, CONFIG_X86_MSR=m:

ERROR: "msrs_free" [drivers/edac/amd64_edac_mod.ko] undefined!
ERROR: "msrs_alloc" [drivers/edac/amd64_edac_mod.ko] undefined!"

This is due to the fact that &lt;arch/x86/lib/msr.c&gt; is conditioned on
CONFIG_SMP and in the UP case we have only the stubs in the header.
Fork off SMP functionality into a new file (msr-smp.c) and build
msrs_{alloc,free} unconditionally.

Reported-by: Randy Dunlap &lt;randy.dunlap@oracle.com&gt;
Cc: H. Peter Anvin &lt;hpa@zytor.com&gt;
Signed-off-by: Borislav Petkov &lt;petkovbb@gmail.com&gt;
LKML-Reference: &lt;20091216231625.GD27228@liondog.tnic&gt;
Signed-off-by: H. Peter Anvin &lt;hpa@zytor.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>x86, msr: Add support for non-contiguous cpumasks</title>
<updated>2010-01-06T23:05:11Z</updated>
<author>
<name>Borislav Petkov</name>
<email>borislav.petkov@amd.com</email>
</author>
<published>2009-12-11T17:14:40Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=eb21839600f406f76a63e6da241db7f361f7ff52'/>
<id>urn:sha1:eb21839600f406f76a63e6da241db7f361f7ff52</id>
<content type='text'>
commit 505422517d3f126bb939439e9d15dece94e11d2c upstream.

The current rd/wrmsr_on_cpus helpers assume that the supplied
cpumasks are contiguous. However, there are machines out there
like some K8 multinode Opterons which have a non-contiguous core
enumeration on each node (e.g. cores 0,2 on node 0 instead of 0,1), see
http://www.gossamer-threads.com/lists/linux/kernel/1160268.

This patch fixes out-of-bounds writes (see URL above) by adding per-CPU
msr structs which are used on the respective cores.

Additionally, two helpers, msrs_{alloc,free}, are provided for use by
the callers of the MSR accessors.

Cc: H. Peter Anvin &lt;hpa@zytor.com&gt;
Cc: Mauro Carvalho Chehab &lt;mchehab@redhat.com&gt;
Cc: Aristeu Rozanski &lt;aris@redhat.com&gt;
Cc: Randy Dunlap &lt;randy.dunlap@oracle.com&gt;
Cc: Doug Thompson &lt;dougthompson@xmission.com&gt;
Signed-off-by: Borislav Petkov &lt;borislav.petkov@amd.com&gt;
LKML-Reference: &lt;20091211171440.GD31998@aftab&gt;
Signed-off-by: H. Peter Anvin &lt;hpa@zytor.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>x86, msr: Unify rdmsr_on_cpus/wrmsr_on_cpus</title>
<updated>2010-01-06T23:05:07Z</updated>
<author>
<name>Borislav Petkov</name>
<email>borislav.petkov@amd.com</email>
</author>
<published>2009-07-30T09:10:02Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=a89a9e1d729d7f1bc0d18bea800087bdfcfe4957'/>
<id>urn:sha1:a89a9e1d729d7f1bc0d18bea800087bdfcfe4957</id>
<content type='text'>
commit b8a4754147d61f5359a765a3afd3eb03012aa052 upstream.

Since rdmsr_on_cpus and wrmsr_on_cpus are almost identical, unify them
into a common __rwmsr_on_cpus helper thus avoiding code duplication.

While at it, convert cpumask_t's to const struct cpumask *.

Signed-off-by: Borislav Petkov &lt;borislav.petkov@amd.com&gt;
Signed-off-by: H. Peter Anvin &lt;hpa@zytor.com&gt;
Signed-off-by: Ingo Molnar &lt;mingo@elte.hu&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>x86/ptrace: make genregs[32]_get/set more robust</title>
<updated>2010-01-06T23:04:35Z</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2009-12-17T15:04:56Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=b52d85562809f1e8b03d03f6ebc5aa28405d0576'/>
<id>urn:sha1:b52d85562809f1e8b03d03f6ebc5aa28405d0576</id>
<content type='text'>
commit 04a1e62c2cec820501f93526ad1e46073b802dc4 upstream.

The loop condition is fragile: we compare an unsigned value to zero, and
then decrement it by something larger than one in the loop.  All the
callers should be passing in appropriately aligned buffer lengths, but
it's better to just not rely on it, and have some appropriate defensive
loop limits.

Acked-by: Roland McGrath &lt;roland@redhat.com&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>KVM: LAPIC: make sure IRR bitmap is scanned after vm load</title>
<updated>2010-01-06T23:04:01Z</updated>
<author>
<name>Marcelo Tosatti</name>
<email>mtosatti@redhat.com</email>
</author>
<published>2009-12-14T19:37:35Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=3b96f9a68db8401e73955a6f6c33ec08ffed9e96'/>
<id>urn:sha1:3b96f9a68db8401e73955a6f6c33ec08ffed9e96</id>
<content type='text'>
commit 6e24a6eff4571002cd48b99a2b92dc829ce39cb9 upstream.

The vcpus are initialized with irr_pending set to false, but
loading the LAPIC registers with pending IRR fails to reset
the irr_pending variable.

Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>KVM: MMU: remove prefault from invlpg handler</title>
<updated>2010-01-06T23:04:00Z</updated>
<author>
<name>Marcelo Tosatti</name>
<email>mtosatti@redhat.com</email>
</author>
<published>2009-12-05T14:34:11Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=3a9f99234194d5f6c987098c288f29f200dabf5e'/>
<id>urn:sha1:3a9f99234194d5f6c987098c288f29f200dabf5e</id>
<content type='text'>
commit fb341f572d26e0786167cd96b90cc4febed830cf upstream.

The invlpg prefault optimization breaks Windows 2008 R2 occasionally.

The visible effect is that the invlpg handler instantiates a pte which
is, microseconds later, written with a different gfn by another vcpu.

The OS could have other mechanisms to prevent a present translation from
being used, which the hypervisor is unaware of.

While the documentation states that the cpu is at liberty to prefetch tlb
entries, it looks like this is not heeded, so remove tlb prefetch from
invlpg.

Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

</content>
</entry>
<entry>
<title>x86/amd-iommu: Fix initialization failure panic</title>
<updated>2010-01-06T23:03:57Z</updated>
<author>
<name>Joerg Roedel</name>
<email>joerg.roedel@amd.com</email>
</author>
<published>2009-12-21T14:51:23Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=e93166f10c741f247c7e172936811bad558b4135'/>
<id>urn:sha1:e93166f10c741f247c7e172936811bad558b4135</id>
<content type='text'>
commit 0f764806438d5576ac58898332e5dcf30bb8a679 upstream.

The assumption that acpi_table_parse passes the return value
of the hanlder function to the caller proved wrong
recently. The return value of the handler function is
totally ignored. This makes the initialization code for AMD
IOMMU buggy in a way that could cause a kernel panic on
initialization. This patch fixes the issue in the AMD IOMMU
driver.

Signed-off-by: Joerg Roedel &lt;joerg.roedel@amd.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@suse.de&gt;

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