<feed xmlns='http://www.w3.org/2005/Atom'>
<title>BMC/Intel-BMC/linux.git/drivers/md/persistent-data/dm-btree.c, branch dev-4.3</title>
<subtitle>Intel OpenBMC Linux kernel source tree (mirror)</subtitle>
<id>https://git.radix-linux.su/BMC/Intel-BMC/linux.git/atom?h=dev-4.3</id>
<link rel='self' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/atom?h=dev-4.3'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/'/>
<updated>2016-02-19T22:28:23+00:00</updated>
<entry>
<title>dm btree: fix leak of bufio-backed block in btree_split_sibling error path</title>
<updated>2016-02-19T22:28:23+00:00</updated>
<author>
<name>Mike Snitzer</name>
<email>snitzer@redhat.com</email>
</author>
<published>2015-11-23T21:24:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=7435c303d7e85e47738282becc3c8e15b4b02c06'/>
<id>urn:sha1:7435c303d7e85e47738282becc3c8e15b4b02c06</id>
<content type='text'>
commit 30ce6e1cc5a0f781d60227e9096c86e188d2c2bd upstream.

The block allocated at the start of btree_split_sibling() is never
released if later insert_at() fails.

Fix this by releasing the previously allocated bufio block using
unlock_block().

Reported-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>dm btree: fix leak of bufio-backed block in btree_split_beneath error path</title>
<updated>2015-10-23T18:02:55+00:00</updated>
<author>
<name>Mike Snitzer</name>
<email>snitzer@redhat.com</email>
</author>
<published>2015-10-22T14:56:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=4dcb8b57df3593dcb20481d9d6cf79d1dc1534be'/>
<id>urn:sha1:4dcb8b57df3593dcb20481d9d6cf79d1dc1534be</id>
<content type='text'>
btree_split_beneath()'s error path had an outstanding FIXME that speaks
directly to the potential for _not_ cleaning up a previously allocated
bufio-backed block.

Fix this by releasing the previously allocated bufio block using
unlock_block().

Reported-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Acked-by: Joe Thornber &lt;thornber@redhat.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>dm btree: remove unused "dm_block_t root" parameter in btree_split_sibling()</title>
<updated>2015-08-12T15:32:19+00:00</updated>
<author>
<name>Vivek Goyal</name>
<email>vgoyal@redhat.com</email>
</author>
<published>2015-07-06T15:55:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=0a8d4c3ef8f14cbd26d97767e3676722d4eebee5'/>
<id>urn:sha1:0a8d4c3ef8f14cbd26d97767e3676722d4eebee5</id>
<content type='text'>
Signed-off-by: Vivek Goyal &lt;vgoyal@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm btree: add ref counting ops for the leaves of top level btrees</title>
<updated>2015-08-12T14:50:37+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2015-08-12T14:12:09+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=b0dc3c8bc157c60b1d470163882be8c13e1950af'/>
<id>urn:sha1:b0dc3c8bc157c60b1d470163882be8c13e1950af</id>
<content type='text'>
When using nested btrees, the top leaves of the top levels contain
block addresses for the root of the next tree down.  If we shadow a
shared leaf node the leaf values (sub tree roots) should be incremented
accordingly.

This is only an issue if there is metadata sharing in the top levels.
Which only occurs if metadata snapshots are being used (as is possible
with dm-thinp).  And could result in a block from the thinp metadata
snap being reused early, thus corrupting the thinp metadata snap.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>dm btree: silence lockdep lock inversion in dm_btree_del()</title>
<updated>2015-07-06T14:45:02+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2015-07-03T13:51:32+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=1c7518794a3647eb345d59ee52844e8a40405198'/>
<id>urn:sha1:1c7518794a3647eb345d59ee52844e8a40405198</id>
<content type='text'>
Allocate memory using GFP_NOIO when deleting a btree.  dm_btree_del()
can be called via an ioctl and we don't want to recurse into the FS or
block layer.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>dm btree: fix a recursion depth bug in btree walking code</title>
<updated>2014-11-10T20:23:58+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2014-11-10T15:03:24+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=9b460d3699324d570a4d4161c3741431887f102f'/>
<id>urn:sha1:9b460d3699324d570a4d4161c3741431887f102f</id>
<content type='text'>
The walk code was using a 'ro_spine' to hold it's locked btree nodes.
But this data structure is designed for the rolling lock scheme, and
as such automatically unlocks blocks that are two steps up the call
chain.  This is not suitable for the simple recursive walk algorithm,
which retraces its steps.

This code is only used by the persistent array code, which in turn is
only used by dm-cache.  In order to trigger it you need to have a
mapping tree that is more than 2 levels deep; which equates to 8-16
million cache blocks.  For instance a 4T ssd with a very small block
size of 32k only just triggers this bug.

The fix just places the locked blocks on the stack, and stops using
the ro_spine altogether.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>dm btree: add dm_btree_find_lowest_key</title>
<updated>2014-01-09T21:29:17+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2013-12-20T15:41:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=f164e6900f2be2c29f5c11ca52af5bb824f40826'/>
<id>urn:sha1:f164e6900f2be2c29f5c11ca52af5bb824f40826</id>
<content type='text'>
dm_btree_find_lowest_key is the reciprocal of dm_btree_find_highest_key.
Factor out common code for dm_btree_find_{highest,lowest}_key.

dm_btree_find_lowest_key is needed for an upcoming DM target, as such it
is best to get this interface in place.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm btree: prefetch child nodes when walking tree for a dm_btree_del</title>
<updated>2013-08-23T13:02:14+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2013-08-09T11:59:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=04f17c802f447e76d13ade5bb78fbbf34baef0f8'/>
<id>urn:sha1:04f17c802f447e76d13ade5bb78fbbf34baef0f8</id>
<content type='text'>
dm-btree now takes advantage of dm-bufio's ability to prefetch data via
dm_bm_prefetch().  Prior to this change many btree node visits were
causing a synchronous read.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Signed-off-by: Alasdair G Kergon &lt;agk@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm btree: use pop_frame in dm_btree_del to cleanup code</title>
<updated>2013-08-23T13:02:14+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2013-08-09T11:48:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=cd5acf0b445030cd8b5bfd818f464997652485e5'/>
<id>urn:sha1:cd5acf0b445030cd8b5bfd818f464997652485e5</id>
<content type='text'>
Remove a visited leaf straight away from the stack, rather than
marking all it's children as visited and letting it get removed on the
next iteration.  May also offer a micro optimisation in dm_btree_del.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Signed-off-by: Alasdair G Kergon &lt;agk@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm persistent data: add btree_walk</title>
<updated>2013-03-01T22:45:50+00:00</updated>
<author>
<name>Joe Thornber</name>
<email>ejt@redhat.com</email>
</author>
<published>2013-03-01T22:45:50+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/BMC/Intel-BMC/linux.git/commit/?id=4e7f1f9089660aec3b5ab2645ce62777c6f4c6a2'/>
<id>urn:sha1:4e7f1f9089660aec3b5ab2645ce62777c6f4c6a2</id>
<content type='text'>
Add dm_btree_walk to iterate through the contents of a btree.
This will be used by the dm cache target.

Signed-off-by: Joe Thornber &lt;ejt@redhat.com&gt;
Signed-off-by: Alasdair G Kergon &lt;agk@redhat.com&gt;
</content>
</entry>
</feed>
