llvm/utils/TableGen, branch release_27http://llvm.orghttps://git.amat.us/llvm/atom/utils/TableGen?h=release_272010-03-07T07:21:24Zturn off debug spew2010-03-07T07:21:24ZChris Lattnersabre@nondot.org2010-03-07T07:21:24Zurn:sha1:81d6d52592b928838d706e9fc5c9779de017d0d7
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97912 91177308-0d34-0410-b5e6-96231b3b80d8
more factoring.2010-03-07T07:20:49ZChris Lattnersabre@nondot.org2010-03-07T07:20:49Zurn:sha1:6fd326b7ff3f114f3b1eda05e1142e52222c6b54
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97911 91177308-0d34-0410-b5e6-96231b3b80d8
teach tblgen to be more aggressive when factoring CheckType nodes.2010-03-07T07:01:28ZChris Lattnersabre@nondot.org2010-03-07T07:01:28Zurn:sha1:9cdd9659c381001a200aa4667919297187fa5764
Now it will factor things like this:
CheckType i32
...
CheckOpcode ISD::AND
CheckType i64
...
into:
SwitchType:
i32: ...
i64:
CheckOpcode ISD::AND
...
This shrinks hte table by a few bytes, nothing spectacular.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97908 91177308-0d34-0410-b5e6-96231b3b80d8
add some helper functions and implement isContradictory 2010-03-07T06:29:26ZChris Lattnersabre@nondot.org2010-03-07T06:29:26Zurn:sha1:48aa5756a29a7b96850ac646d1edd806c9df4643
for CheckValueTypeMatcher. The isContradictory implementation
helps us factor better, shrinking x86 table from 79144 -> 78896
bytes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97905 91177308-0d34-0410-b5e6-96231b3b80d8
Make it not an error to specify -O* options several times.2010-03-05T04:46:39ZMikhail Glushenkovfoldr@codedgers.com2010-03-05T04:46:39Zurn:sha1:b5c4239606d3b34b5aac20b93d4c052ac4c39ce7
As in 'llvmc -O2 -O2 test.c'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97787 91177308-0d34-0410-b5e6-96231b3b80d8
so hey, it turns out that the histogram was completely wrong, because2010-03-04T01:34:29ZChris Lattnersabre@nondot.org2010-03-04T01:34:29Zurn:sha1:57fcd0361162ebe74d123d1d04a455c2c2b000a9
we sometimes emit nodes multiple times to string buffers to size them.
Compute the histogram correctly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97708 91177308-0d34-0410-b5e6-96231b3b80d8
zap fixme.2010-03-04T01:26:00ZChris Lattnersabre@nondot.org2010-03-04T01:26:00Zurn:sha1:6b4ae7172529503bd28333f488857c0d4fdc3e39
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97705 91177308-0d34-0410-b5e6-96231b3b80d8
now that complexpatterns are all emitted at the end of the match2010-03-04T01:25:36ZChris Lattnersabre@nondot.org2010-03-04T01:25:36Zurn:sha1:9752fb10a4ad9a8144740b092d2b4f513897835c
sequence, just emit instruction predicates right before them. This
exposes yet more factoring opportunitites, shrinking the X86 table
to 79144 bytes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97704 91177308-0d34-0410-b5e6-96231b3b80d8
change the new isel matcher to emit ComplexPattern matches2010-03-04T01:23:08ZChris Lattnersabre@nondot.org2010-03-04T01:23:08Zurn:sha1:57bf8a483edf97589c3e6085721e72fc187677a8
as the very last thing before node emission. This should
dramatically reduce the number of times we do 'MatchAddress'
on X86, speeding up compile time. This also improves comments
in the tables and shrinks the table a bit, now down to
80506 bytes for x86.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97703 91177308-0d34-0410-b5e6-96231b3b80d8
enhance comment output to specify what recorded slot2010-03-04T00:28:05ZChris Lattnersabre@nondot.org2010-03-04T00:28:05Zurn:sha1:d1aca7c8b4ea45fbee18fc18a21ed963175dd7bc
numbers a ComplexPat will match into.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97696 91177308-0d34-0410-b5e6-96231b3b80d8