<feed xmlns='http://www.w3.org/2005/Atom'>
<title>bcm5719-llvm/mlir/test/Dialect/FxpMathOps, branch meklort-10.0.1</title>
<subtitle>Project Ortega BCM5719 LLVM</subtitle>
<id>https://git.raptorcs.com/git/bcm5719-llvm/atom?h=meklort-10.0.1</id>
<link rel='self' href='https://git.raptorcs.com/git/bcm5719-llvm/atom?h=meklort-10.0.1'/>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/'/>
<updated>2019-10-24T22:01:09+00:00</updated>
<entry>
<title>Convert the Canonicalize and CSE passes to generic Operation Passes.</title>
<updated>2019-10-24T22:01:09+00:00</updated>
<author>
<name>River Riddle</name>
<email>riverriddle@google.com</email>
</author>
<published>2019-10-24T22:00:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=2b61b7979eb59df579a3a4bf6fe768ddf2a556f4'/>
<id>urn:sha1:2b61b7979eb59df579a3a4bf6fe768ddf2a556f4</id>
<content type='text'>
This allows for them to be used on other non-function, or even other function-like, operations. The algorithms are already generic, so this is simply changing the derived pass type. The majority of this change is just ensuring that the nesting of these passes remains the same, as the pass manager won't auto-nest them anymore.

PiperOrigin-RevId: 276573038
</content>
</entry>
<entry>
<title>Fix minor spelling tweaks (NFC)</title>
<updated>2019-10-20T07:11:34+00:00</updated>
<author>
<name>Kazuaki Ishizaki</name>
<email>kiszk@users.noreply.github.com</email>
</author>
<published>2019-10-20T07:11:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=8bfedb3ca599ccf3c507e721f4bf5e6a3b026f8c'/>
<id>urn:sha1:8bfedb3ca599ccf3c507e721f4bf5e6a3b026f8c</id>
<content type='text'>
Closes tensorflow/mlir#177

PiperOrigin-RevId: 275692653
</content>
</entry>
<entry>
<title>Change the attribute dictionary syntax to separate name and value with '='.</title>
<updated>2019-06-26T02:06:34+00:00</updated>
<author>
<name>River Riddle</name>
<email>riverriddle@google.com</email>
</author>
<published>2019-06-26T02:06:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=679a3b41911457ef0f4a79a3135bb7ecca6d2f97'/>
<id>urn:sha1:679a3b41911457ef0f4a79a3135bb7ecca6d2f97</id>
<content type='text'>
The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations:

Old:
  "foo: 10 : i32"

New:
  "foo = 10 : i32"
PiperOrigin-RevId: 255097928
</content>
</entry>
<entry>
<title>Modify the syntax of the the ElementsAttrs to print the type as a colon type.</title>
<updated>2019-06-25T23:06:58+00:00</updated>
<author>
<name>River Riddle</name>
<email>riverriddle@google.com</email>
</author>
<published>2019-06-25T23:06:13+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=4842b2d42e95ff4b5a8425ec7108c9e4e90e8b6c'/>
<id>urn:sha1:4842b2d42e95ff4b5a8425ec7108c9e4e90e8b6c</id>
<content type='text'>
This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr.

Example:
  dense&lt;5&gt; : tensor&lt;i32&gt;
  dense&lt;[3]&gt; : tensor&lt;1xi32&gt;
PiperOrigin-RevId: 255069157
</content>
</entry>
<entry>
<title>Rename -verify mlir-opt flag to -verify-expected-diagnostics</title>
<updated>2019-06-20T06:08:03+00:00</updated>
<author>
<name>Geoffrey Martin-Noble</name>
<email>gcmn@google.com</email>
</author>
<published>2019-06-19T18:21:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=d7d69569e72725686f1293aa042aa2cba9a1b11a'/>
<id>urn:sha1:d7d69569e72725686f1293aa042aa2cba9a1b11a</id>
<content type='text'>
This name has caused some confusion because it suggests that it's running op verification (and that this verification isn't getting run by default).

PiperOrigin-RevId: 254035268
</content>
</entry>
<entry>
<title>Refactor SplatElementsAttr to inherit from DenseElementsAttr as opposed to being a separate Attribute type. DenseElementsAttr provides a better internal representation for splat values as well as better API for accessing elements.</title>
<updated>2019-06-20T06:01:52+00:00</updated>
<author>
<name>River Riddle</name>
<email>riverriddle@google.com</email>
</author>
<published>2019-06-14T00:24:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=6a0555a87545bdd50c783271c08a9e50f871ca8f'/>
<id>urn:sha1:6a0555a87545bdd50c783271c08a9e50f871ca8f</id>
<content type='text'>
PiperOrigin-RevId: 253138287
</content>
</entry>
<entry>
<title>    Move Quantization -&gt; Dialect/QuantOps, FxpMathOps -&gt; Dialect/FxpMathOps.</title>
<updated>2019-05-20T20:41:55+00:00</updated>
<author>
<name>Stella Laurenzo</name>
<email>laurenzo@google.com</email>
</author>
<published>2019-05-14T18:03:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/bcm5719-llvm/commit/?id=d4dcf7de9e6f5f00177c534d765c5b24d9db8ed8'/>
<id>urn:sha1:d4dcf7de9e6f5f00177c534d765c5b24d9db8ed8</id>
<content type='text'>
    Adding the additional layer of directory was discussed offline and matches the Target/ tree. The names match the defacto convention we seem to be following where the C++ namespace is ^(.+)Ops/$ matched against the directory name.

    This is in preparation for patching the Quantizer into this tree, which would have been confusing without moving the Quantization dialect to its more proper home. It is left to others to move other dialects if desired.

    Tested:
      ninja check-mlir

--

PiperOrigin-RevId: 248171982
</content>
</entry>
</feed>
