<feed xmlns='http://www.w3.org/2005/Atom'>
<title>llvm/test/CodeGen/AArch64, branch testing</title>
<subtitle>http://llvm.org</subtitle>
<id>https://git.amat.us/llvm/atom/test/CodeGen/AArch64?h=testing</id>
<link rel='self' href='https://git.amat.us/llvm/atom/test/CodeGen/AArch64?h=testing'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/'/>
<updated>2013-03-09T18:25:40Z</updated>
<entry>
<title>Test case hygiene.</title>
<updated>2013-03-09T18:25:40Z</updated>
<author>
<name>Benjamin Kramer</name>
<email>benny.kra@googlemail.com</email>
</author>
<published>2013-03-09T18:25:40Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=1cb47b9afe77231d6b87b8445dfec475fb5e59cd'/>
<id>urn:sha1:1cb47b9afe77231d6b87b8445dfec475fb5e59cd</id>
<content type='text'>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176772 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: specify full triple in test as only Linux works for now.</title>
<updated>2013-03-08T15:27:30Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-03-08T15:27:30Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=fa3dc9ffbb5d8b9bb742940388dd14fc5f282362'/>
<id>urn:sha1:fa3dc9ffbb5d8b9bb742940388dd14fc5f282362</id>
<content type='text'>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176692 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: expand sincos operations, we don't support them.</title>
<updated>2013-03-08T13:55:07Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-03-08T13:55:07Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=69fe178f7781fa3c01d013ac7b7858926064f6ca'/>
<id>urn:sha1:69fe178f7781fa3c01d013ac7b7858926064f6ca</id>
<content type='text'>
Patch based on Mans Rullgard's.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176688 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: be more careful resorting to inefficient addressing for weak vars.</title>
<updated>2013-02-28T14:36:31Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-28T14:36:31Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=6ff20f205b2aa126b268bcada9920f56715161be'/>
<id>urn:sha1:6ff20f205b2aa126b268bcada9920f56715161be</id>
<content type='text'>
If an otherwise weak var is actually defined in this unit, it can't be
undefined at runtime so we can use normal global variable sequences (ADRP/ADD)
to access it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176259 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: don't drop GlobalAddress offset when handling extern_weak decls.</title>
<updated>2013-02-28T14:36:24Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-28T14:36:24Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=5366ab21f4595d0e3888b2d23f38469da2465b8d'/>
<id>urn:sha1:5366ab21f4595d0e3888b2d23f38469da2465b8d</id>
<content type='text'>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176258 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: Use cbnz instead of cmp/b.ne pair for atomic operations.</title>
<updated>2013-02-28T13:52:07Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-28T13:52:07Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=279b9184c2ff4fea93b198a3519b8cb3a1d8d195'/>
<id>urn:sha1:279b9184c2ff4fea93b198a3519b8cb3a1d8d195</id>
<content type='text'>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176253 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: remove ConstantIsland pass &amp; put literals in separate section.</title>
<updated>2013-02-15T09:33:43Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-15T09:33:43Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=1e8839302b70d77de63844332bdee9ce7d06f2c9'/>
<id>urn:sha1:1e8839302b70d77de63844332bdee9ce7d06f2c9</id>
<content type='text'>
This implements the review suggestion to simplify the AArch64 backend. If we
later discover that we *really* need the extra complexity of the
ConstantIslands pass for performance reasons it can be resurrected.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175258 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>AArch64: refactor frame handling to use movz/movk for overlarge offsets.</title>
<updated>2013-02-15T09:33:26Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-15T09:33:26Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=148ac534fc5592ed7031efde9a577890f078068b'/>
<id>urn:sha1:148ac534fc5592ed7031efde9a577890f078068b</id>
<content type='text'>
In the near future litpools will be in a different section, which means that
any access to them is at least two instructions. This makes the case for a
movz/movk pair (if total offset &lt;= 32-bits) even more compelling.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175257 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>Implement external weak (ELF) symbols on AArch64</title>
<updated>2013-02-06T16:43:33Z</updated>
<author>
<name>Tim Northover</name>
<email>Tim.Northover@arm.com</email>
</author>
<published>2013-02-06T16:43:33Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=8a06229c89f848bf742e2b88423d02558b7ca638'/>
<id>urn:sha1:8a06229c89f848bf742e2b88423d02558b7ca638</id>
<content type='text'>
Weakly defined symbols should evaluate to 0 if they're undefined at
link-time. This is impossible to do with the usual address generation
patterns, so we should use a literal pool entry to materlialise the
address.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174518 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
<entry>
<title>Reapply r174343, with a fix for a scary DAG combine bug where it failed to differentiate between the alignment of the</title>
<updated>2013-02-05T19:24:39Z</updated>
<author>
<name>Owen Anderson</name>
<email>resistor@mac.com</email>
</author>
<published>2013-02-05T19:24:39Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/llvm/commit/?id=b48783b09127136c98cc6b0399be757728504e56'/>
<id>urn:sha1:b48783b09127136c98cc6b0399be757728504e56</id>
<content type='text'>
base point of a load, and the overall alignment of the load.  This caused infinite loops in DAG combine with the
original application of this patch.

ORIGINAL COMMIT LOG:
When the target-independent DAGCombiner inferred a higher alignment for a load,
it would replace the load with one with the higher alignment.  However, it did
not place the new load in the worklist, which prevented later DAG combines in
the same phase (for example, target-specific combines) from ever seeing it.

This patch corrects that oversight, and updates some tests whose output changed
due to slightly different DAGCombine outputs.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174431 91177308-0d34-0410-b5e6-96231b3b80d8
</content>
</entry>
</feed>
