instructions. It attempts to create high-level multi-operand GEPs,
though in cases where this isn't possible it falls back to casting
the pointer to i8* and emitting a GEP with that. Using GEP instructions
instead of ptrtoint+arithmetic+inttoptr helps pointer analyses that
don't use ScalarEvolution, such as BasicAliasAnalysis.
Also, make the AddrModeMatcher more aggressive in handling GEPs.
Previously it assumed that operand 0 of a GEP would require a register
in almost all cases. It now does extra checking and can do more
matching if operand 0 of the GEP is foldable. This fixes a problem
that was exposed by SCEVExpander using GEPs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72093 91177308-0d34-0410-b5e6-96231b3b80d8
getNoopOrSignExtend, and getTruncateOrNoop. These are similar
to getTruncateOrZeroExtend etc., except that they assert that
the conversion is either not widening or narrowing, as
appropriate. These will be used in some upcoming fixes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71632 91177308-0d34-0410-b5e6-96231b3b80d8
These values aren't analyzable, so they don't care if more information
about the loop trip count can be had. Also, SCEVUnknown is used for
a PHI while the PHI itself is being analyzed, so it needs to be left
in the Scalars map. This fixes a variety of subtle issues.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71533 91177308-0d34-0410-b5e6-96231b3b80d8
return the correct value when the cast operand is all zeros. This ought
to be pretty rare, because it would mean that the regular SCEV folding
routines missed a case, though there are cases they might legitimately
miss. Also, it's unlikely anything currently using GetMinTrailingZeros
cares about this case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71532 91177308-0d34-0410-b5e6-96231b3b80d8
add-recurrence to be exposed. Add a new SCEV folding rule to
help simplify expressions in the presence of these extra truncs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71264 91177308-0d34-0410-b5e6-96231b3b80d8
which are not analyzed with SCEV techniques, which can require
brute-forcing through a large number of instructions. This
fixes a massive compile-time issue on 400.perlbench (in
particular, the loop in MD5Transform).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71259 91177308-0d34-0410-b5e6-96231b3b80d8
to sorting SCEVs by their kind, sort SCEVs of the same kind according
to their operands. This helps avoid things like (a+b) being a distinct
expression from (b+a).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@71160 91177308-0d34-0410-b5e6-96231b3b80d8
CallbackVH, with fixes. allUsesReplacedWith need to
walk the def-use chains and invalidate all users of a
value that is replaced. SCEVs of users need to be
recalcualted even if the new value is equivalent. Also,
make forgetLoopPHIs walk def-use chains, since any
SCEV that depends on a PHI should be recalculated when
more information about that PHI becomes available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70927 91177308-0d34-0410-b5e6-96231b3b80d8
makes ScalarEvolution::deleteValueFromRecords, and it's code that
subtly needed to be called before ReplaceAllUsesWith, unnecessary.
It also makes ValueDeletionListener unnecessary.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70645 91177308-0d34-0410-b5e6-96231b3b80d8
it also forget any SCEVs associated with loop-header PHIs in the loop,
as they may be dependent on trip count information.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70633 91177308-0d34-0410-b5e6-96231b3b80d8
artificial "ptrtoint", as it tends to clutter up complicated
expressions. The cast operators now print both source and
destination types, which is usually sufficient.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70554 91177308-0d34-0410-b5e6-96231b3b80d8
compute an upper-bound value for the trip count, in addition to
the actual trip count. Use this to allow getZeroExtendExpr and
getSignExtendExpr to fold casts in more cases.
This may eventually morph into a more general value-range
analysis capability; there are certainly plenty of places where
more complete value-range information would allow more folding.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70509 91177308-0d34-0410-b5e6-96231b3b80d8
(sext i8 {-128,+,1} to i64) to i64 {-128,+,1}, where the iteration
crosses from negative to positive, but is still safe if the trip
count is within range.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70421 91177308-0d34-0410-b5e6-96231b3b80d8
print sext, zext, and trunc, instead of signextend, zeroextend,
and truncate, respectively, for consistency with the main IR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70405 91177308-0d34-0410-b5e6-96231b3b80d8
information to simplify [sz]ext({a,+,b}) to {zext(a),+,[zs]ext(b)},
as appropriate.
These functions and the trip count code each call into the other, so
this requires careful handling to avoid infinite recursion. During
the initial trip count computation, conservative SCEVs are used,
which are subsequently discarded once the trip count is actually
known.
Among other benefits, this change lets LSR automatically eliminate
some unnecessary zext-inreg and sext-inreg operation where the
operand is an induction variable.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70241 91177308-0d34-0410-b5e6-96231b3b80d8
type to truncate to should be the number of bits of the value that are
preserved, not the number that are clobbered with sign-extension.
This fixes regressions in ldecod.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69704 91177308-0d34-0410-b5e6-96231b3b80d8
as they appear in LLVM IR. This isn't particularly interesting
on its own; this is just setting up some infrastructure.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69655 91177308-0d34-0410-b5e6-96231b3b80d8
size from the integer, requiring zero extension or truncation. Don't
create ZExtInsts with pointer types. This fixes a regression in
consumer-jpeg.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69307 91177308-0d34-0410-b5e6-96231b3b80d8
have pointer types, though in contrast to C pointer types, SCEV
addition is never implicitly scaled. This not only eliminates the
need for special code like IndVars' EliminatePointerRecurrence
and LSR's own GEP expansion code, it also does a better job because
it lets the normal optimizations handle pointer expressions just
like integer expressions.
Also, since LLVM IR GEPs can't directly index into multi-dimensional
VLAs, moving the GEP analysis out of client code and into the SCEV
framework makes it easier for clients to handle multi-dimensional
VLAs the same way as other arrays.
Some existing regression tests show improved optimization.
test/CodeGen/ARM/2007-03-13-InstrSched.ll in particular improved to
the point where if-conversion started kicking in; I turned it off
for this test to preserve the intent of the test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69258 91177308-0d34-0410-b5e6-96231b3b80d8
to more accurately describe what it does. Expand its doxygen comment
to describe what the backedge-taken count is and how it differs
from the actual iteration count of the loop. Adjust names and
comments in associated code accordingly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65382 91177308-0d34-0410-b5e6-96231b3b80d8
trip count value when the original loop iteration condition is
signed and the canonical induction variable won't undergo signed
overflow. This isn't required for correctness; it just preserves
more information about original loop iteration values.
Add a getTruncateOrSignExtend method to ScalarEvolution,
following getTruncateOrZeroExtend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64918 91177308-0d34-0410-b5e6-96231b3b80d8
modified in a way that may effect the trip count calculation. Change
IndVars to use this method when it rewrites pointer or floating-point
induction variables instead of using a doInitialization method to
sneak these changes in before ScalarEvolution has a chance to see
the loop. This eliminates the need for LoopPass to depend on
ScalarEvolution.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64810 91177308-0d34-0410-b5e6-96231b3b80d8
loop induction on LP64 targets. When the induction variable is
used in addressing, IndVars now is usually able to inserst a
64-bit induction variable and eliminates the sign-extending cast.
This is also useful for code using C "short" types for
induction variables on targets with 32-bit addressing.
Inserting a wider induction variable is easy; the tricky part is
determining when trunc(sext(i)) expressions are no-ops. This
requires range analysis of the loop trip count. A common case is
when the original loop iteration starts at 0 and exits when the
induction variable is signed-less-than a fixed value; this case
is now handled.
This replaces IndVarSimplify's OptimizeCanonicalIVType. It was
doing the same optimization, but it was limited to loops with
constant trip counts, because it was running after the loop
rewrite, and the information about the original induction
variable is lost by that point.
Rename ScalarEvolution's executesAtLeastOnce to
isLoopGuardedByCond, generalize it to be able to test for
ICMP_NE conditions, and move it to be a public function so that
IndVars can use it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64407 91177308-0d34-0410-b5e6-96231b3b80d8
Use it to safely handle less-than-or-equals-to exit conditions in loops. These
also occur when the loop exit branch is exit on true because SCEV inverses the
icmp predicate.
Use it again to handle non-zero strides, but only with an unsigned comparison
in the exit condition.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@59528 91177308-0d34-0410-b5e6-96231b3b80d8
If this patch causes a performance regression for anyone, please let me know,
and it can be fixed in a different way with much more effort.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@59384 91177308-0d34-0410-b5e6-96231b3b80d8
returned by BinomialCoefficient and don't try to operate with them. This
replaces the previous fix for PR2857.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57431 91177308-0d34-0410-b5e6-96231b3b80d8
- Recognize expressions like "x > -1 ? x : 0" as min/max and turn them
into expressions like "x < 0 ? 0 : x", which is easily recognizable
as a min/max operation.
- Refrain from folding expression like "y/2 < 1" to "y < 2" when the
comparison is being used as part of a min or max idiom, like
"y/2 < 1 ? 1 : y/2". In that case, the division has another use, so
folding doesn't eliminate it, and obfuscates the min/max, making it
harder to recognize as a min/max operation.
These benefit ScalarEvolution, CodeGen, and anything else that wants to
recognize integer min and max.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@56246 91177308-0d34-0410-b5e6-96231b3b80d8
its callers to emit a space character before calling it when a
space is needed.
This fixes several spurious whitespace issues in
ScalarEvolution's debug dumps. See the test changes for
examples.
This also fixes odd space-after-tab indentation in the output
for switch statements, and changes calls from being printed like
this:
call void @foo( i32 %x )
to this:
call void @foo(i32 %x)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@56196 91177308-0d34-0410-b5e6-96231b3b80d8
continue past the first conditional branch when looking for a
relevant test. This helps it avoid using MAX expressions in
loop trip counts in more cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54697 91177308-0d34-0410-b5e6-96231b3b80d8
version uses a new algorithm for evaluating the binomial coefficients
which is significantly more efficient for AddRecs of more than 2 terms
(see the comments in the code for details on how the algorithm works).
It also fixes some bugs: it removes the arbitrary length restriction for
AddRecs, it fixes the silent generation of incorrect code for AddRecs
which require a wide calculation width, and it fixes an issue where we
were incorrectly truncating the iteration count too far when evaluating
an AddRec expression narrower than the induction variable.
There are still a few related issues I know of: I think there's
still an issue with the SCEVExpander expansion of AddRec in terms of
the width of the induction variable used. The hack to avoid generating
too-wide integers shouldn't be necessary; instead, the callers should be
considering the cost of the expansion before expanding it (in addition
to not expanding too-wide integers, we might not want to expand
expressions that are really expensive, especially when optimizing for
size; calculating an length-17 32-bit AddRec currently generates about 250
instructions of straight-line code on X86). Also, for long 32-bit
AddRecs on X86, CodeGen really sucks at scheduling the code. I'm planning on
filing follow-up PRs for these issues.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54332 91177308-0d34-0410-b5e6-96231b3b80d8
time applying to the implicit comparison in smin expressions. The
correct way to transform an inequality into the opposite
inequality, either signed or unsigned, is with a not expression.
I looked through the SCEV code, and I don't think there are any more
occurrences of this issue.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54194 91177308-0d34-0410-b5e6-96231b3b80d8
SGT exit condition. Essentially, the correct way to flip an inequality
in 2's complement is the not operator, not the negation operator.
That said, the difference only affects cases involving INT_MIN.
Also, enhance the pre-test search logic to be a bit smarter about
inequalities flipped with a not operator, so it can eliminate the smax
from the iteration count for simple loops.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54184 91177308-0d34-0410-b5e6-96231b3b80d8
bail after 256-bits to avoid producing code that the backends can't handle.
Previously, we capped it at 64-bits, preferring to miscompile in those cases.
This change also reverts much of r52248 because the invariants the code was
expecting are now being met.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53812 91177308-0d34-0410-b5e6-96231b3b80d8
with code that was expecting different bit widths for different values.
Make getTruncateOrZeroExtend a method on ScalarEvolution, and use it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52248 91177308-0d34-0410-b5e6-96231b3b80d8
several things that were neither in an anonymous namespace nor static
but not intended to be global.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51017 91177308-0d34-0410-b5e6-96231b3b80d8
manually performing the comparison. This allows the special
case to work correctly even in the case where someone is
experimenting with a different comparison function :-).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49670 91177308-0d34-0410-b5e6-96231b3b80d8
Parse reversed smax and umax as smin and umin and express them with negative
or binary-not SCEVs (which are really just subtract under the hood).
Parse 'xor %x, -1' as (-1 - %x).
Remove dead code (ConstantInt::get always returns a ConstantInt).
Don't use getIntegerSCEV(-1, Ty). The first value is an int, then it gets
passed into a uint64_t. Instead, create the -1 directly from
ConstantInt::getAllOnesValue().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47360 91177308-0d34-0410-b5e6-96231b3b80d8
variable (with step 1) and m is its final value. Then, the correct trip
count is SMAX(m,n)-n. Previously, we used SMAX(0,m-n), but m-n may
overflow and can't in general be interpreted as signed.
Patch by Nick Lewycky.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47007 91177308-0d34-0410-b5e6-96231b3b80d8
to the RHS. This simple change allows to compute loop iteration count
for loops with condition similar to the one in the testcase (which seems
to be quite common).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46959 91177308-0d34-0410-b5e6-96231b3b80d8
arbitrary iteration.
The patch:
1) changes SCEVSDivExpr into SCEVUDivExpr,
2) replaces PartialFact() function with BinomialCoefficient(); the
computations (essentially, the division) in BinomialCoefficient() are
performed with the apprioprate bitwidth necessary to avoid overflow;
unsigned division is used instead of the signed one.
Computations in BinomialCoefficient() require support from the code
generator for APInts. Currently, we use a hack rounding up the
neccessary bitwidth to the nearest power of 2. The hack is easy to turn
off in future.
One remaining issue: we assume the divisor of the binomial coefficient
formula can be computed accurately using 16 bits. It means we can handle
AddRecs of length up to 9. In future, we should use APInts to evaluate
the divisor.
Thanks to Nicholas for cooperation!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46955 91177308-0d34-0410-b5e6-96231b3b80d8
Reimplement the xform in Analysis/ConstantFolding.cpp where we can use
targetdata to validate that it is safe. While I'm in there, fix some const
correctness issues and generalize the interface to the "operand folder".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44817 91177308-0d34-0410-b5e6-96231b3b80d8
is disabled in the sense that it will refuse to create one from a UDiv
instruction, until the code is better tested.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44163 91177308-0d34-0410-b5e6-96231b3b80d8
Use APFloat in UpgradeParser and AsmParser.
Change all references to ConstantFP to use the
APFloat interface rather than double. Remove
the ConstantFP double interfaces.
Use APFloat functions for constant folding arithmetic
and comparisons.
(There are still way too many places APFloat is
just a wrapper around host float/double, but we're
getting there.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41747 91177308-0d34-0410-b5e6-96231b3b80d8