Arnold Schwaighofer 11eb51e239 LoopVectorize: Vectorize all accesses in address space zero with unit stride ...

We can vectorize them because in the case where we wrap in the address space the
unvectorized code would have had to access a pointer value of zero which is
undefined behavior in address space zero according to the LLVM IR semantics.
(Thank you Duncan, for pointing this out to me).

Fixes PR16592.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186088 91177308-0d34-0410-b5e6-96231b3b80d8

2013-07-11 15:21:55 +00:00

..

ARM

…

X86

…

12-12-11-if-conv.ll

Reapply 184685 after the SetVector iteration order fix.

2013-06-24 12:09:15 +00:00

2012-10-20-infloop.ll

…

2012-10-22-isconsec.ll

…

bsd_regex.ll

…

bzip_reverse_loops.ll

…

calloc.ll

…

cast-induction.ll

…

cpp-new-array.ll

…

dbg.value.ll

…

debugloc.ll

LoopVectorize: Preserve debug location info

2013-06-28 00:38:54 +00:00

flags.ll

…

float-reduction.ll

…

funcall.ll

LoopVectorize: Math functions only read rounding mode

2013-07-01 00:54:44 +00:00

gcc-examples.ll

…

global_alias.ll

…

hoist-loads.ll

…

i8-induction.ll

…

if-conv-crash.ll

…

if-conversion-edgemasks.ll

LoopVectorize: Cache edge masks created during if-conversion

2013-06-27 20:31:06 +00:00

if-conversion-nest.ll

…

if-conversion-reduction.ll

…

if-conversion.ll

…

increment.ll

…

induction_plus.ll

…

induction.ll

…

infiniteloop.ll

…

intrinsic.ll

…

lcssa-crash.ll

…

lit.local.cfg

…

memdep.ll

Reapply 184685 after the SetVector iteration order fix.

2013-06-24 12:09:15 +00:00

metadata-unroll.ll

…

metadata-width.ll

…

minmax_reduction.ll

…

multiple-address-spaces.ll

…

no_idiv_reduction.ll

…

no_int_induction.ll

…

no_outside_user.ll

…

nofloat.ll

…

non-const-n.ll

…

nsw-crash.ll

…

phi-hang.ll

…

ptr_loops.ll

…

read-only.ll

…

reduction.ll

…

reverse_induction.ll

…

reverse_iter.ll

…

runtime-check-readonly.ll

…

runtime-check.ll

Reapply 184685 after the SetVector iteration order fix.

2013-06-24 12:09:15 +00:00

runtime-limit.ll

…

safegep.ll

LoopVectorize: Vectorize all accesses in address space zero with unit stride

2013-07-11 15:21:55 +00:00

same-base-access.ll

…

scalar-select.ll

…

simple-unroll.ll

…

small-loop.ll

…

start-non-zero.ll

…

store-shuffle-bug.ll

Fix spelling.

2013-06-27 01:01:11 +00:00

struct_access.ll

…

undef-inst-bug.ll

LoopVectorize: Use vectorized loop invariant gep index anchored in loop

2013-06-27 15:11:55 +00:00

value-ptr-bug.ll

…

vectorize-once.ll

…

write-only.ll

…