Commit Graph

716 Commits

Author SHA1 Message Date
Benjamin Kramer 6094f30da2 SCEV: Make the final add of an inbounds GEP nuw if we know that the index is positive.
We can't do this for the general case as saying a GEP with a negative index
doesn't have unsigned wrap isn't valid for negative indices.
  %gep = getelementptr inbounds i32* %p, i64 -1

But an inbounds GEP cannot run past the end of address space. So we check for
the very common case of a positive index and make GEPs derived from that NUW.
Together with Andy's recent non-unit stride work this lets us analyze loops
like

  void foo3(int *a, int *b) {
    for (; a < b; a++) {}
  }

PR12375, PR12376.

Differential Revision: http://llvm-reviews.chandlerc.com/D2033

llvm-svn: 193514
2013-10-28 07:30:06 +00:00
Andrew Trick d5990ad9b6 Clarify SCEV comments.
We handle for(i=n; i>0; i -= s) by canonicalizing within SCEV to for(i=-n; i<0; i += s).

llvm-svn: 193147
2013-10-22 05:09:40 +00:00
Matt Arsenault 404c60a7c3 Use more type helper functions
llvm-svn: 193109
2013-10-21 19:43:56 +00:00
Matt Arsenault be18b8a3ca Fix creating bitcasts between address spaces in SCEV.
The test before wasn't successfully testing this
since it was missing the datalayout piece to change
the size of the second address space.

llvm-svn: 193102
2013-10-21 18:41:10 +00:00
Matt Arsenault 4ed49b5301 Remove unused SCEV functions
llvm-svn: 193097
2013-10-21 18:08:09 +00:00
Andrew Trick 768b917dc8 SCEV should use NSW to get trip count for positive nonunit stride loops.
SCEV currently fails to compute loop counts for nonunit stride
loops. This comes up frequently. It prevents loop optimization and
forces vectorization to insert extra loop checks.

For example:
void foo(int n, int *x) {
 for (int i = 0; i < n; i += 3) {
   x[i] = i;
   x[i+1] = i+1;
   x[i+2] = i+2;
 }
}

We need to properly handle the case in which limit > INT_MAX-stride. In
the above case: n > INT_MAX-3. In this case the loop counter will step
beyond the limit and overflow at the same time. However, knowing that
signed integer overlow in undefined, we can assume the loop test
behavior is arbitrary after overflow. This obeys both C undefined
behavior rules, and the more strict LLVM poison value rules.

I'm finally fixing this in response to Hal Finkel's persistence.
The most probable reason that we never optimized this before is that
we were being careful to handle case where the developer expected a
side-effect free infinite loop relying on overflow:

for (int i = 0; i < n; i += s) {
  ++j;
}
return j;

If INT_MAX+1 is a multiple of s and n > INT_MAX-s, then we might
expect an infinite loop. However there are plenty of ways to achieve
this effect without relying on undefined behavior of signed overflow.

llvm-svn: 193015
2013-10-18 23:43:53 +00:00
Matt Arsenault 4c265906cc Minor code simplification
llvm-svn: 191579
2013-09-27 22:38:23 +00:00
Matt Arsenault a90a18e0ea Teach ScalarEvolution about pointer address spaces
llvm-svn: 190425
2013-09-10 19:55:24 +00:00
Andrew Trick c3bc8b8de6 Fix a severe compile time problem when forming large SCEV expressions.
This fix is very lightweight. The same fix already existed for AddRec
but was missing for NAry expressions.

This is obviously an improvement and I'm unsure how to test compile
time problems.

Patch by Xiaoyi Guo!

llvm-svn: 187475
2013-07-31 02:43:40 +00:00
Shuxin Yang 23773b34c6 Stylistic change.
Thank Nick for figuring out these problems.

llvm-svn: 186146
2013-07-12 07:25:38 +00:00
Craig Topper 2cd5ff8003 Use SmallVectorImpl& instead of SmallVector to avoid repeating small vector size.
llvm-svn: 186098
2013-07-11 16:22:38 +00:00
Benjamin Kramer fc3ea6f4bc Don't use a potentially expensive shift if all we want is one set bit.
No functionality change.

llvm-svn: 186095
2013-07-11 16:05:50 +00:00
Hal Finkel ff666bd962 Don't crash in SE dealing with ashr x, -1
ScalarEvolution::getSignedRange uses ComputeNumSignBits from ValueTracking on
ashr instructions. ComputeNumSignBits can return zero, but this case was not
handled correctly by the code in getSignedRange which was calling:
  APInt::getSignedMinValue(BitWidth).ashr(NS - 1)
with NS = 0, resulting in an assertion failure in APInt::ashr.

Now, we just return the conservative result (as with NS == 1).

Another bug found by llvm-stress.

llvm-svn: 185955
2013-07-09 18:16:16 +00:00
Shuxin Yang efc4c01ed3 Fix a SCEV update problem.
The symptom is seg-fault, and the root cause is that a SCEV contains a SCEVUnknown
which has null-pointer to a llvm::Value.

 This is how the problem take place:
 ===================================
  1). In the pristine input IR, there are two relevant instrutions Op1 and Op2, 
     Op1's corresponding SCEV (denoted as SCEV(op1)) is a SCEVUnknown, and
     SCEV(Op2) contains SCEV(Op1).  None of these instructions are dead.

     Op1 : V1 = ...
     ...
     Op2 : V2 = ... // directly or indirectly (data-flow) depends on Op1
    
  2) Optimizer (LSR in my case) generates an instruction holding the equivalent
     value of Op1, making Op1 dead. 
     Op1': V1' = ...
     Op1: V1 = ... ; now dead)
     Op2 : V2 = ... //Now deps on Op1', but the SCEV(Op2) still contains SCEV(Op1)

  3) Op1 is deleted, and call-back function is called to reset 
     SCEV(Op1) to indicate it is invalid. However, SCEV(Op2) is not 
     invalidated as well.

  4) Following pass get the cached, invalid SCEV(Op2), and try to manipulate it,
     and cause segfault. 

 The fix:
 ========
 It seems there is no clean yet inexpensive fix. I write to dev-list
soliciting good solution, unforunately no ack. So, I decide to fix this 
problem in a brute-force way:

  When ScalarEvolution::getSCEV is called, check if the cached SCEV 
contains a invalid SCEVUnknow, if yes, remove the cached SCEV, and
re-evaluate the SCEV from scratch.

  I compile buch of big *.c and *.cpp, fortunately, I don't see any increase
in compile time.

 Misc:
=====
 The reduced test-case has 2357 lines of code+other-stuff, too big to commit.

 rdar://14283433

llvm-svn: 185843
2013-07-08 17:33:13 +00:00
Craig Topper 31ee5866de Use SmallVectorImpl::iterator/const_iterator instead of SmallVector to avoid specifying the vector size.
llvm-svn: 185540
2013-07-03 15:07:05 +00:00
Aaron Ballman d07f55185c Silencing an MSVC warning about */ being found outside of a comment.
llvm-svn: 183175
2013-06-04 01:01:56 +00:00
Andrew Trick ee9143acf5 Prevent loop-unroll from making assumptions about undefined behavior.
Fixes rdar:14036816, PR16130.

There is an opportunity to compute precise trip counts for 'or'
expressions and multi-exit loops.
rdar:14038809: Optimize trip count computation for multi-exit loops.

To do this we need to record the fact that ExitLimit assumes NSW. When
it does not we can safely assume that the loop trip count is the
minimum ExitLimt across all subexpressions and loop exits.

llvm-svn: 183060
2013-05-31 23:34:46 +00:00
Andrew Trick 5b245a16fa Fix ScalarEvolution::ComputeExitLimitFromCond for 'or' conditions.
Fixes PR16130 - clang produces incorrect code with loop/expression at -O2.

This is a 2+ year old bug that's now holding up the release. It's a
case where we knowingly made aggressive assumptions about undefined
behavior. These assumptions are wrong when SCEV is computing a
subexpression that does not directly control the branch. With this
fix, we avoid making assumptions in those cases but still optimize the
common case. SCEV's trip count computation for exits controlled by
'or' expressions is now analagous to the trip count computation for
loops with multiple exits. I had already fixed the multiple exit case
to be conservative.

llvm-svn: 182989
2013-05-31 06:43:25 +00:00
Andrew Trick 9093e15066 Fix SCEV forgetMemoizedResults should search and destroy backedge exprs.
Fixes PR15570: SEGV: SCEV back-edge info invalid after dead code removal.

Indvars creates a SCEV expression for the loop's back edge taken
count, then determines that the comparison is always true and
removes it.

When loop-unroll asks for the expression, it contains a NULL
SCEVUnknkown (as a CallbackVH).

forgetMemoizedResults should invalidate the loop back edges expression.

llvm-svn: 177986
2013-03-26 03:14:53 +00:00
Chandler Carruth 9fb823bbd4 Move all of the header files which are involved in modelling the LLVM IR
into their new header subdirectory: include/llvm/IR. This matches the
directory structure of lib, and begins to correct a long standing point
of file layout clutter in LLVM.

There are still more header files to move here, but I wanted to handle
them in separate commits to make tracking what files make sense at each
layer easier.

The only really questionable files here are the target intrinsic
tablegen files. But that's a battle I'd rather not fight today.

I've updated both CMake and Makefile build systems (I think, and my
tests think, but I may have missed something).

I've also re-sorted the includes throughout the project. I'll be
committing updates to Clang, DragonEgg, and Polly momentarily.

llvm-svn: 171366
2013-01-02 11:36:10 +00:00
Chandler Carruth ed0881b2a6 Use the new script to sort the includes of every file under lib.
Sooooo many of these had incorrect or strange main module includes.
I have manually inspected all of these, and fixed the main module
include to be the nearest plausible thing I could find. If you own or
care about any of these source files, I encourage you to take some time
and check that these edits were sensible. I can't have broken anything
(I strictly added headers, and reordered them, never removed), but they
may not be the headers you'd really like to identify as containing the
API being implemented.

Many forward declarations and missing includes were added to a header
files to allow them to parse cleanly when included first. The main
module rule does in fact have its merits. =]

llvm-svn: 169131
2012-12-03 16:50:05 +00:00
Benjamin Kramer ba11a9892c Follow up to 168711: It's safe to base this analysis on the found compare, just return the value for the right predicate.
Thanks to Andy for catching this.

llvm-svn: 168921
2012-11-29 19:07:57 +00:00
Andrew Trick fa59403bfd Improve isImpliedCond comment a bit.
llvm-svn: 168914
2012-11-29 18:35:13 +00:00
Benjamin Kramer e20e124280 SCEV: Even if the latch terminator is foldable we can't deduce the result of an unrelated condition with it.
Fixes PR14432.

llvm-svn: 168711
2012-11-27 18:16:32 +00:00
Chandler Carruth 7ec5085e01 Revert the series of commits starting with r166578 which introduced the
getIntPtrType support for multiple address spaces via a pointer type,
and also introduced a crasher bug in the constant folder reported in
PR14233.

These commits also contained several problems that should really be
addressed before they are re-committed. I have avoided reverting various
cleanups to the DataLayout APIs that are reasonable to have moving
forward in order to reduce the amount of churn, and minimize the number
of commits that were reverted. I've also manually updated merge
conflicts and manually arranged for the getIntPtrType function to stay
in DataLayout and to be defined in a plausible way after this revert.

Thanks to Duncan for working through this exact strategy with me, and
Nick Lewycky for tracking down the really annoying crasher this
triggered. (Test case to follow in its own commit.)

After discussing with Duncan extensively, and based on a note from
Micah, I'm going to continue to back out some more of the more
problematic patches in this series in order to ensure we go into the
LLVM 3.2 branch with a reasonable story here. I'll send a note to
llvmdev explaining what's going on and why.

Summary of reverted revisions:

r166634: Fix a compiler warning with an unused variable.
r166607: Add some cleanup to the DataLayout changes requested by
         Chandler.
r166596: Revert "Back out r166591, not sure why this made it through
         since I cancelled the command. Bleh, sorry about this!
r166591: Delete a directory that wasn't supposed to be checked in yet.
r166578: Add in support for getIntPtrType to get the pointer type based
         on the address space.
llvm-svn: 167221
2012-11-01 08:07:29 +00:00
Benjamin Kramer 5bc077aa88 SCEV validator: Ignore CouldNotCompute/undef on both sides. This is mostly noise and blocks finding more severe bugs.
llvm-svn: 166873
2012-10-27 11:36:07 +00:00
Benjamin Kramer 24d270db57 SCEV validator: Add workarounds for some common false positives due to the way it handles strings.
llvm-svn: 166872
2012-10-27 10:45:01 +00:00
Benjamin Kramer 214935ee70 Add a basic verifier for SCEV's backedge taken counts.
Enabled with -verify-scev. This could be extended significantly but hopefully
catches the common cases now. Note that it's not enabled by default in any
configuration because the way it tries to distinguish SCEVs is still fragile and
may produce false positives. Also the test-suite isn't clean yet, one example
is that it fails if a pass drops an NSW bit but it's still present in SCEV's
cached. Cleaning up all those cases will take some time.

llvm-svn: 166786
2012-10-26 17:31:32 +00:00
Hal Finkel 30bd9346a0 getSmallConstantTripMultiple should never return zero.
When the trip count is -1, getSmallConstantTripMultiple could return zero,
and this would cause runtime loop unrolling to assert. Instead of returning
zero, one is now returned (consistent with the existing overflow cases).
Fixes PR14167.

llvm-svn: 166612
2012-10-24 19:46:44 +00:00
Micah Villmow 12d9127833 Add in support for getIntPtrType to get the pointer type based on the address space.
This checkin also adds in some tests that utilize these paths and updates some of the
clients.

llvm-svn: 166578
2012-10-24 15:52:52 +00:00
Micah Villmow cdfe20b97f Move TargetData to DataLayout.
llvm-svn: 165402
2012-10-08 16:38:25 +00:00
Sylvestre Ledru 91ce36c986 Revert 'Fix a typo 'iff' => 'if''. iff is an abreviation of if and only if. See: http://en.wikipedia.org/wiki/If_and_only_if Commit 164767
llvm-svn: 164768
2012-09-27 10:14:43 +00:00
Sylvestre Ledru 721cffd53a Fix a typo 'iff' => 'if'
llvm-svn: 164767
2012-09-27 09:59:43 +00:00
Manman Ren 49d684e1e2 Release build: guard dump functions with
"#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)"

No functional change. Update r163344.

llvm-svn: 163679
2012-09-12 05:06:18 +00:00
Manman Ren c3366ccecb Release build: guard dump functions with "ifndef NDEBUG"
No functional change.

llvm-svn: 163344
2012-09-06 19:55:56 +00:00
Nick Lewycky fb78083b1c Stay rational; don't assert trying to take the square root of a negative value.
If it's negative, the loop is already proven to be infinite. Fixes PR13489!

llvm-svn: 161107
2012-08-01 09:14:36 +00:00
Andrew Trick 365e31c36c Factor SCEV traversal code so I can use it elsewhere. No functionality.
llvm-svn: 160203
2012-07-13 23:33:03 +00:00
Dan Gohman 3d1512384f Delete code for folding undefs in ScalarEvolution. It's invalid in
obscure ways, and it isn't actually important in the real world.

llvm-svn: 159969
2012-07-09 23:51:20 +00:00
Benjamin Kramer e2ef47c145 Reduce use list thrashing by using DenseMap's find_as for maps with ValueHandle keys.
No functionality change.

llvm-svn: 159497
2012-06-30 22:37:15 +00:00
Nick Lewycky 474112d82c If the step value is a constant zero, the loop isn't going to terminate. Fixes
the assert reported in PR13228!

llvm-svn: 159393
2012-06-28 23:44:57 +00:00
Benjamin Kramer bde9176663 Fix typos found by http://github.com/lyda/misspell-check
llvm-svn: 157885
2012-06-02 10:20:22 +00:00
Benjamin Kramer 406a2db1f6 Make sure that we're dealing with a binary SCEVExpr when simplifying.
llvm-svn: 157704
2012-05-30 18:42:43 +00:00
Benjamin Kramer 50b26ebb2b Teach SCEV's icmp simplification logic that a-b == 0 is equivalent to a == b.
This also required making recursive simplifications until
nothing changes or a hard limit (currently 3) is hit.

With the simplification in place indvars can canonicalize
loops of the form
for (unsigned i = 0; i < a-b; ++i)
into
for (unsigned i = 0; i != a-b; ++i)
which used to fail because SCEV created a weird umax expr
for the backedge taken count.

llvm-svn: 157701
2012-05-30 18:32:23 +00:00
Andrew Trick a3f9043196 SCEV: Handle a corner case reducing AddRecExpr * AddRecExpr
If integer overflow causes one of the terms to reach zero, that can
force the entire expression to zero.

Fixes PR12929: cast<Ty>() argument of incompatible type

llvm-svn: 157673
2012-05-30 03:35:20 +00:00
Andrew Trick 946f76bf33 Reformat the loop that does AddRecExpr * AddRecExpr reduction.
No functionality.

llvm-svn: 157672
2012-05-30 03:35:17 +00:00
Andrew Trick 7fa4e0fea6 SCEV: Add MarkPendingLoopPredicates to avoid recursive isImpliedCond.
getUDivExpr attempts to simplify by checking for overflow.
isLoopEntryGuardedByCond then evaluates the loop predicate which
may lead to the same getUDivExpr causing endless recursion.

Fixes PR12868: clang 3.2 segmentation fault.

llvm-svn: 157092
2012-05-19 00:48:25 +00:00
Nuno Lopes c2a170e26e reuse the result of some expensive computations in getSignExtendExpr() and getZeroExtendExpr()
this gives a speedup of > 80 in a debug build in the test case of PR12825 (php_sha512_crypt_r)

llvm-svn: 156849
2012-05-15 20:20:14 +00:00
Nuno Lopes ab5c924006 minor simplification to code: Ty is already a SCEV type; don't need to run getEffectiveSCEVType() twice
llvm-svn: 156823
2012-05-15 15:44:38 +00:00
Dan Gohman 0291246ce7 Rewrite ScalarEvolution::hasOperand to use an explicit worklist instead
of recursion, to avoid excessive stack usage on deep expressions.

llvm-svn: 156554
2012-05-10 17:21:30 +00:00
Benjamin Kramer e364d195e9 Revert "SCEV: When expanding a GEP the final addition to the base pointer has NUW but not NSW."
This isn't right either, reverting for now.

llvm-svn: 154910
2012-04-17 06:33:57 +00:00