30e30c76ae
Initial checkin of regression test for generating the PowerPC rlwimi
...
instruction, which performs bitfield inserts. All tests should compile
to one rlwimi instruction.
llvm-svn: 17206
2004-10-24 10:30:22 +00:00
8e97254431
New testcase
...
llvm-svn: 17176
2004-10-23 03:10:01 +00:00
96502da2c0
New testcase. Check for the ability to promote a malloc whose initialzed state is checked by the program
...
llvm-svn: 17160
2004-10-22 06:43:07 +00:00
033e156b54
New testcase
...
llvm-svn: 17115
2004-10-18 04:06:41 +00:00
61011accfd
New testcase that ADCE shouldn't crash on
...
llvm-svn: 17105
2004-10-17 23:44:02 +00:00
5f8becb224
New testcase that crashes the inliner
...
llvm-svn: 17103
2004-10-17 23:20:29 +00:00
d047673192
New testcase, no PHI should be inserted.
...
llvm-svn: 17097
2004-10-17 21:25:32 +00:00
d2c8ed1170
New testcase, rework testcases to fail if there are any gep's other than those
...
involving %B instead of allowing any geps except %A's.
llvm-svn: 17065
2004-10-16 19:44:23 +00:00
17ce82f71d
New testcase
...
llvm-svn: 17061
2004-10-16 18:28:01 +00:00
6299141c4e
When converting phi nodes into select instructions, we shouldn't promote PHI
...
nodes unless we KNOW that we are able to promote all of them. In this case
promoting the phi to a select is silly because we will always have to do the
call conditionally. As such, select promotion is actually a pessimization.
llvm-svn: 16972
2004-10-14 05:12:50 +00:00
c14504c92c
New testcase for memmove -> memcpy transform
...
llvm-svn: 16931
2004-10-12 04:51:55 +00:00
615bd303d1
This testcase ensures that we can SRA a global even if part of the global
...
cannot be SRA'd
llvm-svn: 16915
2004-10-11 05:00:30 +00:00
214956dd9d
These testcases check to see if we can promote malloc's to globals in some
...
cases.
llvm-svn: 16914
2004-10-11 05:00:12 +00:00
cd1e756469
Another testcase. We should be able to devirtualize the call even though
...
we cannot completely eliminate the global.
llvm-svn: 16900
2004-10-10 23:12:33 +00:00
89e0b39fe5
New testcase that crashes -globalopt. I found this through inspection, not
...
actually in the wild :)
llvm-svn: 16877
2004-10-10 17:06:09 +00:00
924ce79136
New testcase we were not handling before.
...
llvm-svn: 16875
2004-10-10 16:46:48 +00:00
ed8d767883
A testcase that globalopt now handles
...
llvm-svn: 16872
2004-10-09 21:50:05 +00:00
cdc19d6a0e
New testcase
...
llvm-svn: 16862
2004-10-09 02:50:01 +00:00
8f9a697431
New testcase
...
llvm-svn: 16856
2004-10-08 20:57:46 +00:00
6935f560a5
New testcase
...
llvm-svn: 16841
2004-10-08 17:29:47 +00:00
fc95027b7a
New testcase. The setcc is only used by a select, but not as a condition:
...
it cannot be folded in.
llvm-svn: 16839
2004-10-08 16:33:40 +00:00
9467062bcf
New testcase
...
llvm-svn: 16835
2004-10-08 05:03:25 +00:00
7bfe4032fc
New testcase
...
llvm-svn: 16833
2004-10-08 03:41:59 +00:00
31d9e6f922
These files now live in Transforms/GlobalOpt
...
llvm-svn: 16809
2004-10-07 19:16:43 +00:00
5860106954
Move these files from Transforms/GlobalConstifier
...
llvm-svn: 16808
2004-10-07 19:16:26 +00:00
381fbf1616
Add new testcase, rename pass
...
llvm-svn: 16799
2004-10-07 04:07:08 +00:00
645bcf6c5d
New testcase
...
llvm-svn: 16791
2004-10-06 20:42:51 +00:00
52783ab1d8
New testcase
...
llvm-svn: 16768
2004-10-06 15:07:56 +00:00
7d83efbc0b
When tail duplicating these functions, the add instruction should not be
...
duplicated, even though the block it is in is duplicated.
llvm-svn: 16757
2004-10-06 03:26:38 +00:00
07d1d7ede5
Another testcase for constness linkage
...
llvm-svn: 16691
2004-10-05 02:16:01 +00:00
e0d464bda2
Testcase to ensure that the 'constant' flag follows the definition when there
...
is a question.
llvm-svn: 16690
2004-10-05 02:12:20 +00:00
db76a3db91
New testcase for PR450
...
llvm-svn: 16670
2004-10-05 00:18:21 +00:00
f27b48f187
The verifier takes 50s on this testcase, that is not good n stuff
...
llvm-svn: 16592
2004-09-29 20:07:25 +00:00
d9d8538995
New testcasses for divides that can be eliminated.
...
llvm-svn: 16587
2004-09-29 17:37:07 +00:00
5e0ab5c5d8
New testcases
...
llvm-svn: 16581
2004-09-29 05:06:12 +00:00
5933e997d8
More testcases for recent instcombine patches
...
llvm-svn: 16569
2004-09-29 03:18:20 +00:00
efc0614cd6
New testcase
...
llvm-svn: 16560
2004-09-28 22:31:49 +00:00
3cbcd25596
New testcases
...
llvm-svn: 16556
2004-09-28 21:39:35 +00:00
cbd7a98cb8
New testcase
...
llvm-svn: 16548
2004-09-28 18:21:01 +00:00
3dd0ae64a9
New testcases
...
llvm-svn: 16547
2004-09-28 18:18:35 +00:00
64ee11511f
New testcase for a latent bug in instcombine that I have been chasing for
...
1.5 days :(
llvm-svn: 16545
2004-09-28 17:50:05 +00:00
7b30f27695
Testcase for LLVM PR445
...
llvm-svn: 16537
2004-09-28 03:36:39 +00:00
02060ae68e
New testcase that crashes the C++ FE, encountered while working on PR445
...
llvm-svn: 16532
2004-09-28 02:29:40 +00:00
5f9319a789
New testcase
...
llvm-svn: 16527
2004-09-27 19:25:20 +00:00
fbfb4d43c4
Testcases for rev 250 of InstructionCombining.cpp
...
llvm-svn: 16522
2004-09-27 16:21:26 +00:00
8c1c691755
Add some tests for shr-and folding
...
llvm-svn: 16507
2004-09-24 15:18:43 +00:00
45a5ace32e
New testcase for a pattern that occurs 20 times in perlbmk
...
llvm-svn: 16503
2004-09-23 21:42:49 +00:00
fb4735980b
load null is undefined behavior, this should fold
...
llvm-svn: 16498
2004-09-23 15:45:20 +00:00
803fb3c2c3
Two testcases for invalid transformations that instcombine is doing
...
llvm-svn: 16446
2004-09-20 10:14:27 +00:00
75cf9f08cc
New testcase for PR362: Icky code generated for std::min/std::max
...
llvm-svn: 16408
2004-09-19 19:17:10 +00:00
Nate Begeman