way or the other. Rewriting the code itself prevents subsequent analysis
passes from making contradictory conclusions about the code that could
cause an infeasible path to be made feasible.
llvm-svn: 46427
void bork() {
int *address = 0;
*address = 0;
}
It's compiled into LLVM code that looks like this:
define void @bork() noreturn nounwind {
entry:
unreachable
}
This is bad on some platforms (like PPC) because it will generate the label for
the function but no body. The label could end up being associated with some
non-code related stuff, like a section. This places a "trap" instruction if the
SimplifyCFG pass removed all code from the function leaving only one
"unreachable" instruction.
llvm-svn: 46387
drop attributes on varargs call arguments. Also, it could generate
invalid IR if the transformed call already had the 'nest' attribute
somewhere (this can never happen for code coming from llvm-gcc,
but it's a theoretical possibility). Fix both problems.
llvm-svn: 45973
a load/store of i64. The later prevents promotion/scalarrepl of the
source and dest in many cases.
This fixes the 300% performance regression of the byval stuff on
stepanov_v1p2.
llvm-svn: 45945
direct calls bails out unless caller and callee have essentially
equivalent parameter attributes. This is illogical - the callee's
attributes should be of no relevance here. Rework the logic, which
incidentally fixes a crash when removed arguments have attributes.
llvm-svn: 45658
a direct call with cast parameters and cast return
value (if any), instcombine was prepared to cast any
non-void return value into any other, whether castable
or not. Add a new predicate for testing whether casting
is valid, and check it both for the return value and
(as a cleanup) for the parameters.
llvm-svn: 45657
it is only a partial fix. This change is noise for most programs, but
speeds up Shootout-C++/matrix by 20%, Ptrdist/ks by 24%, smg2000 by 8%,
hexxagon by 9%, bzip2 by 9% (not sure I trust this), ackerman by 13%, etc.
OTOH, it slows down Shootout/fib2 by 40% (I'll update PR1877 with this info).
llvm-svn: 45354
calls 'nounwind'. It is important for correct C++
exception handling that nounwind markings do not get
lost, so this transformation is actually needed for
correctness.
llvm-svn: 45218
calls. Remove special casing of inline asm from the
inliner. There is a potential problem: the verifier
rejects invokes of inline asm (not sure why). If an
asm call is not marked "nounwind" in some .ll, and
instcombine is not run, but the inliner is run, then
an illegal module will be created. This is bad but
I'm not sure what the best approach is. I'm tempted
to remove the check in the verifier...
llvm-svn: 45073
2. Using zero-extended value of Scale and unsigned division is safe provided
that Scale doesn't have the sign bit set.
Previously these 2 instructions:
%p = bitcast [100 x {i8,i8,i8}]* %x to i8*
%q = getelementptr i8* %p, i32 -4
were combined into:
%q = getelementptr [100 x { i8, i8, i8 }]* %x, i32 0,
i32 1431655764, i32 0
what was incorrect.
llvm-svn: 44936
throw exceptions", just mark intrinsics with the nounwind
attribute. Likewise, mark intrinsics as readnone/readonly
and get rid of special aliasing logic (which didn't use
anything more than this anyway).
llvm-svn: 44544
into alias analysis. This meant updating the API
which now has versions of the getModRefBehavior,
doesNotAccessMemory and onlyReadsMemory methods
which take a callsite parameter. These should be
used unless the callsite is not known, since in
general they can do a better job than the versions
that take a function. Also, users should no longer
call the version of getModRefBehavior that takes
both a function and a callsite. To reduce the
chance of misuse it is now protected.
llvm-svn: 44487
the function type, instead they belong to functions
and function calls. This is an updated and slightly
corrected version of Reid Spencer's original patch.
The only known problem is that auto-upgrading of
bitcode files doesn't seem to work properly (see
test/Bitcode/AutoUpgradeIntrinsics.ll). Hopefully
a bitcode guru (who might that be? :) ) will fix it.
llvm-svn: 44359
trivial difference in function attributes, allow calls to it to
be converted to direct calls. Based on a patch by Török Edwin.
While there, move the various lists of mutually incompatible
parameters etc out of the verifier and into ParameterAttributes.h.
llvm-svn: 44315
or getTypeSizeInBits as appropriate in ScalarReplAggregates.
The right change to make was not always obvious, so it would
be good to have an sroa guru review this. While there I noticed
some bugs, and fixed them: (1) arrays of x86 long double have
holes due to alignment padding, but this wasn't being spotted
by HasStructPadding (renamed to HasPadding). The same goes
for arrays of oddly sized ints. Vectors also suffer from this,
in fact the problem for vectors is much worse because basic
vector assumptions seem to be broken by vectors of type with
alignment padding. I didn't try to fix any of these vector
problems. (2) The code for extracting smaller integers from
larger ones (in the "int union" case) was wrong on big-endian
machines for integers with size not a multiple of 8, like i1.
Probably this is impossible to hit via llvm-gcc, but I fixed
it anyway while there and added a testcase. I also got rid of
some trailing whitespace and changed a function name which
had an obvious typo in it.
llvm-svn: 43672
metric is way off for these in general, and this works around
buggy code like that in PR1764. we'll see if there is a big
performance impact of this. If so, I'll revert it tomorrow.
llvm-svn: 43668
The meaning of getTypeSize was not clear - clarifying it is important
now that we have x86 long double and arbitrary precision integers.
The issue with long double is that it requires 80 bits, and this is
not a multiple of its alignment. This gives a primitive type for
which getTypeSize differed from getABITypeSize. For arbitrary precision
integers it is even worse: there is the minimum number of bits needed to
hold the type (eg: 36 for an i36), the maximum number of bits that will
be overwriten when storing the type (40 bits for i36) and the ABI size
(i.e. the storage size rounded up to a multiple of the alignment; 64 bits
for i36).
This patch removes getTypeSize (not really - it is still there but
deprecated to allow for a gradual transition). Instead there is:
(1) getTypeSizeInBits - a number of bits that suffices to hold all
values of the type. For a primitive type, this is the minimum number
of bits. For an i36 this is 36 bits. For x86 long double it is 80.
This corresponds to gcc's TYPE_PRECISION.
(2) getTypeStoreSizeInBits - the maximum number of bits that is
written when storing the type (or read when reading it). For an
i36 this is 40 bits, for an x86 long double it is 80 bits. This
is the size alias analysis is interested in (getTypeStoreSize
returns the number of bytes). There doesn't seem to be anything
corresponding to this in gcc.
(3) getABITypeSizeInBits - this is getTypeStoreSizeInBits rounded
up to a multiple of the alignment. For an i36 this is 64, for an
x86 long double this is 96 or 128 depending on the OS. This is the
spacing between consecutive elements when you form an array out of
this type (getABITypeSize returns the number of bytes). This is
TYPE_SIZE in gcc.
Since successive elements in a SequentialType (arrays, pointers
and vectors) need to be aligned, the spacing between them will be
given by getABITypeSize. This means that the size of an array
is the length times the getABITypeSize. It also means that GEP
computations need to use getABITypeSize when computing offsets.
Furthermore, if an alloca allocates several elements at once then
these too need to be aligned, so the size of the alloca has to be
the number of elements multiplied by getABITypeSize. Logically
speaking this doesn't have to be the case when allocating just
one element, but it is simpler to also use getABITypeSize in this
case. So alloca's and mallocs should use getABITypeSize. Finally,
since gcc's only notion of size is that given by getABITypeSize, if
you want to output assembler etc the same as gcc then getABITypeSize
is the size you want.
Since a store will overwrite no more than getTypeStoreSize bytes,
and a read will read no more than that many bytes, this is the
notion of size appropriate for alias analysis calculations.
In this patch I have corrected all type size uses except some of
those in ScalarReplAggregates, lib/Codegen, lib/Target (the hard
cases). I will get around to auditing these too at some point,
but I could do with some help.
Finally, I made one change which I think wise but others might
consider pointless and suboptimal: in an unpacked struct the
amount of space allocated for a field is now given by the ABI
size rather than getTypeStoreSize. I did this because every
other place that reserves memory for a type (eg: alloca) now
uses getABITypeSize, and I didn't want to make an exception
for unpacked structs, i.e. I did it to make things more uniform.
This only effects structs containing long doubles and arbitrary
precision integers. If someone wants to pack these types more
tightly they can always use a packed struct.
llvm-svn: 43620
- ChangeCompareStride only reuse stride that is larger than current stride. It
will let the general reuse mechanism to try to reuse a smaller stride.
- Watch out for multiplication overflow in ChangeCompareStride.
- Replace std::set with SmallPtrSet.
llvm-svn: 43408
and the compaison is against a constant value, try eliminate the stride
by moving the compare instruction to another stride and change its
constant operand accordingly. e.g.
loop:
...
v1 = v1 + 3
v2 = v2 + 1
if (v2 < 10) goto loop
=>
loop:
...
v1 = v1 + 3
if (v1 < 30) goto loop
llvm-svn: 43336
- Avoid attempting stride-reuse in the case that there are users that
aren't addresses. In that case, there will be places where the
multiplications won't be folded away, so it's better to try to
strength-reduce them.
- Several SSE intrinsics have operands that strength-reduction can
treat as addresses. The previous item makes this more visible, as
any non-address use of an IV can inhibit stride-reuse.
- Make ValidStride aware of whether there's likely to be a base
register in the address computation. This prevents it from thinking
that things like stride 9 are valid on x86 when the base register is
already occupied.
Also, XFAIL the 2007-08-10-LEA16Use32.ll test; the new logic to avoid
stride-reuse elimintes the LEA in the loop, so the test is no longer
testing what it was intended to test.
llvm-svn: 43231
Fix DecomposeSimpleLinearExpr to handle simple constants better.
Don't nuke gep(bitcast(allocation)) if the bitcast(allocation) will
fold the allocation. This fixes PR1728 and Instcombine/malloc3.ll
llvm-svn: 42891
for (i=A; i<N; i++) {
if (i < X && i > Y)
do_something();
}
is transformed into
U=min(N,X); L=max(A,Y);
for (i=L;i<U;i++)
do_somethihg();
llvm-svn: 42299
