is only used by a cast, and the casted type is the same size as the original
allocation, it would eliminate the cast by folding it into the allocation.
Unfortunately, it was placing the new allocation instruction right before
the cast, which could pull (for example) alloca instructions into the body
of a function. This turns statically allocatable allocas into expensive
dynamically allocated allocas, which is bad bad bad.
This fixes the problem by placing the new allocation instruction at the same
place the old one was, duh. :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13289 91177308-0d34-0410-b5e6-96231b3b80d8
the Module. The default behavior keeps functionality as before: the chosen
function is the one that remains.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13111 91177308-0d34-0410-b5e6-96231b3b80d8
loop. This eliminates the extra add from the previous case, but it's
not clear that this will be a performance win overall. Tommorows test
results will tell. :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13103 91177308-0d34-0410-b5e6-96231b3b80d8
over its USES. If it's dead it doesn't have any uses! :)
Thanks to the fabulous and mysterious Bill Wendling for pointing this out. :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13102 91177308-0d34-0410-b5e6-96231b3b80d8
Eventually it would be nice if CallGraph maintained an ilist of CallGraphNode's instead
of a vector of pointers to them, but today is not that day.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13100 91177308-0d34-0410-b5e6-96231b3b80d8
structure to being dynamically computed on demand. This makes updating
loop information MUCH easier.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13045 91177308-0d34-0410-b5e6-96231b3b80d8
that the exit block of the loop becomes the new entry block of the function.
This was causing a verifier assertion on 252.eon.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13039 91177308-0d34-0410-b5e6-96231b3b80d8
block. The primary motivation for doing this is that we can now unroll nested loops.
This makes a pretty big difference in some cases. For example, in 183.equake,
we are now beating the native compiler with the CBE, and we are a lot closer
with LLC.
I'm now going to play around a bit with the unroll factor and see what effect
it really has.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13034 91177308-0d34-0410-b5e6-96231b3b80d8
limited. Even in it's extremely simple state (it can only *fully* unroll single
basic block loops that execute a constant number of times), it already helps improve
performance a LOT on some benchmarks, particularly with the native code generators.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13028 91177308-0d34-0410-b5e6-96231b3b80d8
Basically we were using SimplifyCFG as a huge sledgehammer for a simple
optimization. Because simplifycfg does so many things, we can't use it
for this purpose.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12977 91177308-0d34-0410-b5e6-96231b3b80d8
Instead of producing code like this:
Loop:
X = phi 0, X2
...
X2 = X + 1
if (X != N-1) goto Loop
We now generate code that looks like this:
Loop:
X = phi 0, X2
...
X2 = X + 1
if (X2 != N) goto Loop
This has two big advantages:
1. The trip count of the loop is now explicit in the code, allowing
the direct implementation of Loop::getTripCount()
2. This reduces register pressure in the loop, and allows X and X2 to be
put into the same register.
As a consequence of the second point, the code we generate for loops went
from:
.LBB2: # no_exit.1
...
mov %EDI, %ESI
inc %EDI
cmp %ESI, 2
mov %ESI, %EDI
jne .LBB2 # PC rel: no_exit.1
To:
.LBB2: # no_exit.1
...
inc %ESI
cmp %ESI, 3
jne .LBB2 # PC rel: no_exit.1
... which has two fewer moves, and uses one less register.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12961 91177308-0d34-0410-b5e6-96231b3b80d8
This transforms code like this:
%C = or %A, %B
%D = select %cond, %C, %A
into:
%C = select %cond, %B, 0
%D = or %A, %C
Since B is often a constant, the select can often be eliminated. In any case,
this reduces the usage count of A, allowing subsequent optimizations to happen.
This xform applies when the operator is any of:
add, sub, mul, or, xor, and, shl, shr
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12800 91177308-0d34-0410-b5e6-96231b3b80d8
that have a constant operand. This implements
add.ll:test19, shift.ll:test15*, and others that are not tested
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12794 91177308-0d34-0410-b5e6-96231b3b80d8
Now we collect all of the call sites we are interested in inlining, then inline
them. This entirely avoids issues with trying to inline a call site we got by
inlining another call site. This also eliminates iterator invalidation issues.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12770 91177308-0d34-0410-b5e6-96231b3b80d8
This also implements some new features for the indvars pass, including
linear function test replacement, exit value substitution, and it works with
a much more general class of induction variables and loops.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12620 91177308-0d34-0410-b5e6-96231b3b80d8
#1 is to unconditionally strip constantpointerrefs out of
instruction operands where they are absolutely pointless and inhibit
optimization. GRRR!
#2 is to implement InstCombine/getelementptr_const.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12519 91177308-0d34-0410-b5e6-96231b3b80d8
With this fix we now successfully extract all 149 loops from 256.bzip2 without
crashing or miscompiling the program!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12493 91177308-0d34-0410-b5e6-96231b3b80d8
1. Names were not put on the new arguments created (ok, this just helps sanity :)
2. Fix outgoing pointer values
3. Do not insert stores for values that had not been computed
4. Fix some wierd problems with the outset calculation
This fixes CodeExtractor/2004-03-14-DominanceProblem.ll, making the extractor
work on at least one simple case!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12484 91177308-0d34-0410-b5e6-96231b3b80d8
as it is making effectively arbitrary modifications to the CFG and we don't
have a domset/domfrontier implementations that can handle the dynamic updates.
Instead of having a bunch of code that doesn't actually work in practice,
just demote any potentially tricky values to the stack (causing the problem
to go away entirely). Later invocations of mem2reg will rebuild SSA for us.
This fixes all of the major performance regressions with tail duplication
from LLVM 1.1. For example, this loop:
---
int popcount(int x) {
int result = 0;
while (x != 0) {
result = result + (x & 0x1);
x = x >> 1;
}
return result;
}
---
Used to be compiled into:
int %popcount(int %X) {
entry:
br label %loopentry
loopentry: ; preds = %entry, %no_exit
%x.0 = phi int [ %X, %entry ], [ %tmp.9, %no_exit ] ; <int> [#uses=3]
%result.1.0 = phi int [ 0, %entry ], [ %tmp.6, %no_exit ] ; <int> [#uses=2]
%tmp.1 = seteq int %x.0, 0 ; <bool> [#uses=1]
br bool %tmp.1, label %loopexit, label %no_exit
no_exit: ; preds = %loopentry
%tmp.4 = and int %x.0, 1 ; <int> [#uses=1]
%tmp.6 = add int %tmp.4, %result.1.0 ; <int> [#uses=1]
%tmp.9 = shr int %x.0, ubyte 1 ; <int> [#uses=1]
br label %loopentry
loopexit: ; preds = %loopentry
ret int %result.1.0
}
And is now compiled into:
int %popcount(int %X) {
entry:
br label %no_exit
no_exit: ; preds = %entry, %no_exit
%x.0.0 = phi int [ %X, %entry ], [ %tmp.9, %no_exit ] ; <int> [#uses=2]
%result.1.0.0 = phi int [ 0, %entry ], [ %tmp.6, %no_exit ] ; <int> [#uses=1]
%tmp.4 = and int %x.0.0, 1 ; <int> [#uses=1]
%tmp.6 = add int %tmp.4, %result.1.0.0 ; <int> [#uses=2]
%tmp.9 = shr int %x.0.0, ubyte 1 ; <int> [#uses=2]
%tmp.1 = seteq int %tmp.9, 0 ; <bool> [#uses=1]
br bool %tmp.1, label %loopexit, label %no_exit
loopexit: ; preds = %no_exit
ret int %tmp.6
}
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12457 91177308-0d34-0410-b5e6-96231b3b80d8
time from 615s to 1.49s on a large testcase that has a gigantic switch statement
that all of the blocks in the function go to (an intepreter).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12442 91177308-0d34-0410-b5e6-96231b3b80d8
Simplify the input/output finder. All elements of a basic block are
instructions. Any used arguments are also inputs. An instruction can only
be used by another instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12405 91177308-0d34-0410-b5e6-96231b3b80d8
extracted, and a function that contained a single top-level loop never had
the loop extracted, regardless of how much non-loop code there was.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12403 91177308-0d34-0410-b5e6-96231b3b80d8
* Don't insert a branch to the switch instruction after the call, just
make it a single block.
* Insert the new alloca instructions in the entry block of the original
function instead of having them execute dynamically
* Don't make the default edge of the switch instruction go back to the switch.
The loop extractor shouldn't create new loops!
* Give meaningful names to the alloca slots and the reload instructions
* Some minor code simplifications
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12402 91177308-0d34-0410-b5e6-96231b3b80d8
This also implements a two minor improvements:
* Don't insert live-out stores IN the region, insert them on the code path
that exits the region
* If the region is exited to the same block from multiple paths, share the
switch statement entry, live-out store code, and the basic block.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12401 91177308-0d34-0410-b5e6-96231b3b80d8
a member of the class. While we're at it, turn the collection into a set
instead of a vector to improve efficiency and make queries simpler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12400 91177308-0d34-0410-b5e6-96231b3b80d8
Require 'simplified' loops, not just raw natural loops. This fixes
CodeExtractor/2004-03-13-LoopExtractorCrash.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12381 91177308-0d34-0410-b5e6-96231b3b80d8
loop information won't see it, and we could have unreachable blocks pointing to
the non-header node of blocks in a natural loop. This isn't tidy, so have the
loopsimplify pass clean it up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12380 91177308-0d34-0410-b5e6-96231b3b80d8
Fix InstCombine/2004-03-13-InstCombineInfLoop.ll which caused an infinite
loop compiling (I think) povray.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12365 91177308-0d34-0410-b5e6-96231b3b80d8
* Be a lot more accurate about what the effects will be when inlining a call
to a function when an argument is an alloca.
* Dramatically reduce the penalty for inlining a call in a large function.
This heuristic made it almost impossible to inline a function into a large
function, no matter how small the callee is.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12363 91177308-0d34-0410-b5e6-96231b3b80d8
On the testcase from GCC PR12440, which has a LOT of loops (1392 of which require
preheaders to be inserted), this speeds up the loopsimplify pass from 1.931s to
0.1875s. The loop in question goes from 1.65s -> 0.0097s, which isn't bad. All of
these times are a debug build.
This adds a dependency on DominatorTree analysis that was not there before, but
we always had dominatortree available anyway, because LICM requires both loop
simplify and DT, so this doesn't add any extra analysis in practice.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12362 91177308-0d34-0410-b5e6-96231b3b80d8
This allows pointers to aggregate objects, whose elements are only read, to
be promoted and passed in by element instead of by reference. This can
enable a LOT of subsequent optimizations in the caller function.
It's worth pointing out that this stuff happens a LOT of C++ programs, because
objects in templates are generally passed around by reference. When these
templates are instantiated on small aggregate or scalar types, however, it is
more efficient to pass them in by value than by reference.
This transformation triggers most on C++ codes (e.g. 334 times on eon), but
does happen on C codes as well. For example, on mesa it triggers 72 times,
and on gcc it triggers 35 times. this is amazingly good considering that
we are using 'basicaa' so far.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12202 91177308-0d34-0410-b5e6-96231b3b80d8