RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-21 03:53:04 +02:00
Code Issues Projects Releases Wiki Activity
80,650 Commits 4 Branches 5 Tags 704 MiB
64efa0a7a8
Commit Graph

4181 Commits

Author SHA1 Message Date
Benjamin Kramer
2a3719125f LVI: Recognize the form instcombine canonicalizes range checks into when forming constant ranges. 
This could probably be made a lot smarter, but this is a common case and doesn't require LVI to scan a lot
of code. With this change CVP can optimize away the "shift == 0" case in Hashing.h that only gets hit when
"shift" is in a range not containing 0.

llvm-svn: 151919
 2012-03-02 15:34:43 +00:00
Eli Friedman
1ff1d1f1bc Duncan pointed out that if the alignment isn't explicitly specified, it defaults to the ABI alignment. Given that, make this code a bit more aggressive in such cases. 
llvm-svn: 151584
 2012-02-27 23:16:46 +00:00
Eli Friedman
15f56db6c0 Teach BasicAA about the LLVM IR rules that allow reading past the end of an object given sufficient alignment. Fixes PR12098. 
llvm-svn: 151553
 2012-02-27 20:46:07 +00:00
Rafael Espindola
2d9b864afe Fix this assert. IP can point to an instruction with strange dominance 
properties (invoke). Just assert that the instruction we return dominates
the insertion point.

llvm-svn: 151511
 2012-02-27 02:13:03 +00:00
Rafael Espindola
c333947449 Don't call dominates on unreachable instructions. Should fix the dragonegg 
build. Testcase is still reducing.

llvm-svn: 151474
 2012-02-26 05:30:08 +00:00
Rafael Espindola
137548bb3e And update the comment... 
llvm-svn: 151472
 2012-02-26 02:36:56 +00:00
Rafael Espindola
e63adef174 Enable the assert that got all this dominator work started. 
llvm-svn: 151471
 2012-02-26 02:29:18 +00:00
Rafael Espindola
34b7c064cb Change the implementation of dominates(inst, inst) to one based on what the 
verifier does. This correctly handles invoke.
Thanks to Duncan, Andrew and Chris for the comments.
Thanks to Joerg for the early testing.

llvm-svn: 151469
 2012-02-26 02:19:19 +00:00
Nick Lewycky
a93c874757 Reinstate the optimization from r151449 with a fix to not turn 'gep %x' into 
'gep null' when the icmp predicate is unsigned (or is signed without inbounds).

llvm-svn: 151467
 2012-02-26 02:09:49 +00:00
Rafael Espindola
d9e654ca63 Don't call dominates on unreachable instructions. 
llvm-svn: 151466
 2012-02-26 01:50:14 +00:00
Nick Lewycky
849715d31f Roll these back to r151448 until I figure out how they're breaking 
MultiSource/Applications/lua.

llvm-svn: 151463
 2012-02-25 23:01:19 +00:00
Nick Lewycky
1636c6eaef An argument and a local identified object (eg. a noalias call) could turn out 
equal if both are null. In the test, scope type %t and global @y by adding a
'gep' prefix to them.

llvm-svn: 151452
 2012-02-25 20:19:07 +00:00
Nick Lewycky
808d0e3fa7 Fix five-letter typo in comment. 
llvm-svn: 151450
 2012-02-25 19:12:58 +00:00
Nick Lewycky
94be1c7d95 Teach instsimplify to be more aggressive when analyzing comparisons of pointers 
by using llvm::isIdentifiedObject. Also teach it to handle GEPs that have
the same base pointer and constant operands. Fixes PR11238!

llvm-svn: 151449
 2012-02-25 19:07:42 +00:00
Nick Lewycky
dce2cfc335 Move isKnownNonNull from private implementation detail of BasicAA to a public 
function that others can use, next to llvm::isIdentifiedObject.

llvm-svn: 151446
 2012-02-25 10:56:28 +00:00
Chris Lattner
b01936f21a fix PR12075, a regression in a recent transform I added. In unreachable code, gep chains can be infinite. Just like "stripPointerCasts", use a set to keep track of visited instructions so we don't recurse infinitely. 
llvm-svn: 151383
 2012-02-24 19:01:58 +00:00
Rafael Espindola
c378c75761 Fix typo. 
llvm-svn: 151238
 2012-02-23 05:38:51 +00:00
Chad Rosier
3703a1917a Remove extra semi-colons. 
llvm-svn: 151169
 2012-02-22 17:25:00 +00:00
Rafael Espindola
4f3e648dd0 Improve comment. Thanks for Andrew for the suggestion. 
llvm-svn: 151127
 2012-02-22 03:44:46 +00:00
Rafael Espindola
23cd372dbf Semantically revert 151015. Add a comment on why we should be able to assert 
the dominance once the dominates method is fixed and why we can use the builder's
insertion point.
Fixes pr12048.

llvm-svn: 151125
 2012-02-22 03:21:39 +00:00
Rafael Espindola
8d727cc0cb s/the the/the/ 
llvm-svn: 151079
 2012-02-21 19:27:16 +00:00
Rafael Espindola
8ef9503d71 Use more idiomatic assert. 
llvm-svn: 151026
 2012-02-21 03:51:14 +00:00
Rafael Espindola
aec58a4177 Avoid warning on non assert builds. 
llvm-svn: 151025
 2012-02-21 03:48:30 +00:00
Rafael Espindola
232a730f45 It turns out that with the current scev organization ReuseOrCreateCast cannot 
know where users will be added. Because of this, it cannot use
Builder.GetInsertPoint at all.

This patch
* removes the FIXME about adding the assert.
* adds a comment explaining hy we don't have one.
* removes a broken logic that only works for some callers and is not needed
  since r150884.
* adds an assert to caller that would have caught the bug fixed by r150884.

llvm-svn: 151015
 2012-02-21 01:19:51 +00:00
Eric Christopher
742c530fa0 Make this a bit prettier and more obvious when a derived type isn't 
derived from anything.

llvm-svn: 150975
 2012-02-20 18:04:39 +00:00
Eric Christopher
69a0eb9258 If a derived type is also a composite type, print that information 
too.

llvm-svn: 150974
 2012-02-20 18:04:35 +00:00
Eric Christopher
ed41b1f1ef Add support for runtime languages on our forward declarations. 
llvm-svn: 150973
 2012-02-20 18:04:14 +00:00
Chris Lattner
50ad7c3f54 fold comparisons of gep'd alloca points with null to false, 
implementing PR12013.  We now compile the testcase to:

__Z4testv:                              ## @_Z4testv
## BB#0:                                ## %_ZN4llvm15SmallVectorImplIiE9push_backERKi.exit
	pushq	%rbx
	subq	$64, %rsp
	leaq	32(%rsp), %rbx
	movq	%rbx, (%rsp)
	leaq	64(%rsp), %rax
	movq	%rax, 16(%rsp)
	movl	$1, 32(%rsp)
	leaq	36(%rsp), %rax
	movq	%rax, 8(%rsp)
	leaq	(%rsp), %rdi
	callq	__Z1gRN4llvm11SmallVectorIiLj8EEE
	movq	(%rsp), %rdi
	cmpq	%rbx, %rdi
	je	LBB0_2
## BB#1:
	callq	_free
LBB0_2:                                 ## %_ZN4llvm11SmallVectorIiLj8EED1Ev.exit
	addq	$64, %rsp
	popq	%rbx
	ret

instead of:

__Z4testv:                              ## @_Z4testv
## BB#0:
	pushq	%rbx
	subq	$64, %rsp
	xorl	%eax, %eax
	leaq	(%rsp), %rbx
	addq	$32, %rbx
	movq	%rbx, (%rsp)
	movq	%rbx, 8(%rsp)
	leaq	64(%rsp), %rcx
	movq	%rcx, 16(%rsp)
	je	LBB0_2
## BB#1:
	movl	$1, 32(%rsp)
	movq	%rbx, %rax
LBB0_2:                                 ## %_ZN4llvm15SmallVectorImplIiE9push_backERKi.exit
	addq	$4, %rax
	movq	%rax, 8(%rsp)
	leaq	(%rsp), %rdi
	callq	__Z1gRN4llvm11SmallVectorIiLj8EEE
	movq	(%rsp), %rdi
	cmpq	%rbx, %rdi
	je	LBB0_4
## BB#3:
	callq	_free
LBB0_4:                                 ## %_ZN4llvm11SmallVectorIiLj8EED1Ev.exit
	addq	$64, %rsp
	popq	%rbx
	ret

This doesn't shrink clang noticably though.

llvm-svn: 150944
 2012-02-20 00:42:49 +00:00
Rafael Espindola
64fe8ada6e Temporarily disable this assert. Looks like it found a similar issue when 
building bullet.

llvm-svn: 150885
 2012-02-18 17:51:43 +00:00
Rafael Espindola
5154b9bedb Don't skip debug instructions when looking for the insertion point of 
the cast. If we do, we can end up with

   inst1
   ---------------  < Insertion point
   dbg inst
   new inst

instead of the desired

   inst1
   new inst
   ---------------  < Insertion point
   dbg inst

Another option would be for InsertNoopCastOfTo (or its callers) to move the
insertion point and we would end up with

   inst1
   dbg inst
   new inst
   ---------------  < Insertion point

but that complicates the callers. This fixes PR12018 (and firefox's build).

llvm-svn: 150884
 2012-02-18 17:22:58 +00:00
Eli Friedman
be89455c98 Fix a rather nasty regression from r150690: LHS != RHS does not imply LHS->stripPointerCasts() != RHS->stripPointerCasts(). 
llvm-svn: 150863
 2012-02-18 03:29:25 +00:00
Dan Gohman
eb02ab2f74 Remove a comment about an alternative approach that wouldn't 
actually work, at least as described. LLVM Metadata is not
intended to suppress LLVM IR rules, as it can be stripped at
any time.

llvm-svn: 150821
 2012-02-17 18:33:38 +00:00
Eric Christopher
61e7e5c8c9 Typo in variable name. 
llvm-svn: 150796
 2012-02-17 07:08:46 +00:00
Benjamin Kramer
3bef0c77d1 Revert "InstSimplify: Strip pointer casts early." 
Turns out this isn't safe, because the code below depends on LHS and RHS having
the same type.

llvm-svn: 150695
 2012-02-16 15:19:59 +00:00
Benjamin Kramer
0911ace9f3 InstSimplify: Strip pointer casts early. 
llvm-svn: 150694
 2012-02-16 15:03:04 +00:00
Benjamin Kramer
8c809e592f InstSimplify: Ignore pointer casts when constant folding compares between pointers. 
llvm-svn: 150690
 2012-02-16 13:49:39 +00:00
Hal Finkel
485162e321 Have AliasSet::aliasesUnknownInst use pointer TBAA info when available 
llvm-svn: 150249
 2012-02-10 15:52:39 +00:00
Duncan Sands
931ce8ee15 Fix PR11948: the result type of an icmp may be a vector of boolean - 
don't assume it is a boolean.

llvm-svn: 150247
 2012-02-10 14:31:24 +00:00
Eric Christopher
4330499af4 Add support for a temporary forward decl type. We want this so we 
can rauw forward declarations if we decide to emit the full type.

Part of rdar://10809898

llvm-svn: 150024
 2012-02-08 00:22:26 +00:00
Devang Patel
5b37f2548b Remove tabs. 
llvm-svn: 150022
 2012-02-08 00:17:07 +00:00
Craig Topper
639b152ca5 Convert assert(0) to llvm_unreachable 
llvm-svn: 149967
 2012-02-07 05:05:23 +00:00
Kostya Serebryany
f4be131943 The patch resolves the conflict between AddressSanitizer and load widening (GVN). 
The problem initially reported by Mozilla folks (http://code.google.com/p/address-sanitizer/issues/detail?id=20),
but it also prevents us from enabling LLVM bootstrap with AddressSanitizer.

llvm-svn: 149925
 2012-02-06 22:48:56 +00:00
Chris Lattner
7a6bd0185e Remove some dead code and tidy things up now that vectors use ConstantDataVector 
instead of always using ConstantVector.

llvm-svn: 149912
 2012-02-06 21:56:39 +00:00
Bill Wendling
8c63e349bc [unwind removal] Remove all of the code for the dead 'unwind' instruction. There 
were no 'unwind' instructions being generated before this, so this is in effect
a no-op.

llvm-svn: 149906
 2012-02-06 21:44:22 +00:00
Bill Wendling
4e92f798ff [unwind removal] We no longer have 'unwind' instructions being generated, so 
remove the code that handles them.

llvm-svn: 149901
 2012-02-06 21:16:41 +00:00
Devang Patel
864da4b549 DebugInfo: Provide a new hook to encode relationship between a property and an ivar. 
llvm-svn: 149874
 2012-02-06 17:49:43 +00:00
Duncan Sands
cc472eec96 Persuade GCC that there is nothing worth warning about here (there isn't). 
llvm-svn: 149834
 2012-02-05 14:20:11 +00:00
Chris Lattner
9782adedd7 reapply the patches reverted in r149470 that reenable ConstantDataArray, 
but with a critical fix to the SelectionDAG code that optimizes copies
from strings into immediate stores: the previous code was stopping reading
string data at the first nul.  Address this by adding a new argument to
llvm::getConstantStringInfo, preserving the behavior before the patch.

llvm-svn: 149800
 2012-02-05 02:29:43 +00:00
Qirun Zhang
78cd86ecbf remove the blank line from previous ci. 
llvm-svn: 149758
 2012-02-04 03:18:47 +00:00
Qirun Zhang
ea06b61d0c test commit. 
add a blank line.

llvm-svn: 149757
 2012-02-04 03:15:26 +00:00