When considering whether to split an instruction with a memory operand
into an explicit load and a register-based instruction, we currently
check that the resulting instruction has exactly 1 def. This prevents 2
important LICM optimizations: compares with memory operands, and double
indirect calls. All the tests and the test-suite pass without the check.
My guess as to original intent is to limit the additional register pressure
created by the new instruction, but given that we only split out a single
register, it is already limited.
The licm-dominance test now checks actual memory loads for hoisting instead of
undef, and it tests compares.
hoist-invariant-load.ll now checks for 2 hoists, the intended hoist, and a bonus
from calling a got-relative function in a loop.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273616 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
This instcombine rule folds away trunc operations that have value available from a prior load or store.
This kind of code can be generated as a result of GVN widening the load or from source code as well.
Reviewers: reames, majnemer, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21246
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273608 91177308-0d34-0410-b5e6-96231b3b80d8
Previously, we just unified any arguments that seemed to be related to
each other. With this patch, we now respect dereference levels, etc.
which should make us substantially more accurate. Proper handling of
StratifiedAttrs will be done in a later patch.
Patch by Jia Chen.
Differential Revision: http://reviews.llvm.org/D21536
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273596 91177308-0d34-0410-b5e6-96231b3b80d8
X86FrameLowering::adjustForHiPEPrologue() contains a hard-coded offset
into an Erlang Runtime System-internal data structure (the PCB). As the
layout of this data structure is prone to change, this poses problems
for maintaining compatibility.
To address this problem, the compiler can produce this information as
module-level named metadata. For example (where P_NSP_LIMIT is the
offending offset):
!hipe.literals = !{ !2, !3, !4 }
!2 = !{ !"P_NSP_LIMIT", i32 152 }
!3 = !{ !"X86_LEAF_WORDS", i32 24 }
!4 = !{ !"AMD64_LEAF_WORDS", i32 24 }
Patch by Magnus Lang
Differential Revision: http://reviews.llvm.org/D20363
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273593 91177308-0d34-0410-b5e6-96231b3b80d8
Recommiting after correcting over-eager Debug Value transfer fixing PR28270.
[DAG] Previously debug values would transfer debuginfo for the selected
start node for a replacement which allows for debug to be dropped.
Push debug value transfer to occur with node/value replacement in
SelectionDAG, remove now extraneous transfers of debug values.
This refixes PR9817 which was being incompletely checked in the
testsuite.
Reviewers: jyknight
Subscribers: dblaikie, llvm-commits
Differential Revision: http://reviews.llvm.org/D21037
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273585 91177308-0d34-0410-b5e6-96231b3b80d8
The previous code would always do 1 or 2 prefix compares;
explicitly only do one.
This speeds up debug -gen-asm-matcher by ~10% (e.g. X86: 40s -> 35s).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273583 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
SSAT saturates an integer, making sure that its value lies within
an interval [-k, k]. Since the constant is given to SSAT as the
number of bytes set to one, k + 1 must be a power of 2, otherwise
the optimization is not possible. Also, the select_cc must use <
and > respectively so that they define an interval.
Reviewers: mcrosier, jmolloy, rengolin
Subscribers: aemerson, rengolin, llvm-commits
Differential Revision: http://reviews.llvm.org/D21372
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273581 91177308-0d34-0410-b5e6-96231b3b80d8
This processor feature had been left out by mistake from the z13
ProcessorModel.
Reviewed by Ulrich Weigand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273567 91177308-0d34-0410-b5e6-96231b3b80d8
When simplifying a load we need to make sure that the type of the
simplified value matches the type of the instruction we're processing.
In theory, we can handle casts here as we deal with constant data, but
since it's not implemented at the moment, we at least need to bail out.
This fixes PR28262.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273562 91177308-0d34-0410-b5e6-96231b3b80d8
DeadStoreElimination can currently remove a small store rendered unnecessary by
a later larger one, but could not remove a larger store rendered unnecessary by
a series of later smaller ones. This adds that capability.
It works by keeping a map, which is used as an effective interval map, for each
store later overwritten only partially, and filling in that interval map as
more such stores are discovered. No additional walking or aliasing queries are
used. In the map forms an interval covering the the entire earlier store, then
it is dead and can be removed. The map is used as an interval map by storing a
mapping between the ending offset and the beginning offset of each interval.
I discovered this problem when investigating a performance issue with code like
this on PowerPC:
#include <complex>
using namespace std;
complex<float> bar(complex<float> C);
complex<float> foo(complex<float> C) {
return bar(C)*C;
}
which produces this:
define void @_Z4testSt7complexIfE(%"struct.std::complex"* noalias nocapture sret %agg.result, i64 %c.coerce) {
entry:
%ref.tmp = alloca i64, align 8
%tmpcast = bitcast i64* %ref.tmp to %"struct.std::complex"*
%c.sroa.0.0.extract.shift = lshr i64 %c.coerce, 32
%c.sroa.0.0.extract.trunc = trunc i64 %c.sroa.0.0.extract.shift to i32
%0 = bitcast i32 %c.sroa.0.0.extract.trunc to float
%c.sroa.2.0.extract.trunc = trunc i64 %c.coerce to i32
%1 = bitcast i32 %c.sroa.2.0.extract.trunc to float
call void @_Z3barSt7complexIfE(%"struct.std::complex"* nonnull sret %tmpcast, i64 %c.coerce)
%2 = bitcast %"struct.std::complex"* %agg.result to i64*
%3 = load i64, i64* %ref.tmp, align 8
store i64 %3, i64* %2, align 4 ; <--- ***** THIS SHOULD NOT BE HERE ****
%_M_value.realp.i.i = getelementptr inbounds %"struct.std::complex", %"struct.std::complex"* %agg.result, i64 0, i32 0, i32 0
%4 = lshr i64 %3, 32
%5 = trunc i64 %4 to i32
%6 = bitcast i32 %5 to float
%_M_value.imagp.i.i = getelementptr inbounds %"struct.std::complex", %"struct.std::complex"* %agg.result, i64 0, i32 0, i32 1
%7 = trunc i64 %3 to i32
%8 = bitcast i32 %7 to float
%mul_ad.i.i = fmul fast float %6, %1
%mul_bc.i.i = fmul fast float %8, %0
%mul_i.i.i = fadd fast float %mul_ad.i.i, %mul_bc.i.i
%mul_ac.i.i = fmul fast float %6, %0
%mul_bd.i.i = fmul fast float %8, %1
%mul_r.i.i = fsub fast float %mul_ac.i.i, %mul_bd.i.i
store float %mul_r.i.i, float* %_M_value.realp.i.i, align 4
store float %mul_i.i.i, float* %_M_value.imagp.i.i, align 4
ret void
}
the problem here is not just that the i64 store is unnecessary, but also that
it blocks further backend optimizations of the other uses of that i64 value in
the backend.
In the future, we might want to add a special case for handling smaller
accesses (e.g. using a bit vector) if the map mechanism turns out to be
noticeably inefficient. A sorted vector is also a possible replacement for the
map for small numbers of tracked intervals.
Differential Revision: http://reviews.llvm.org/D18586
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273559 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
The backend has no reason to behave like a driver and should generally do
as it's told (and error out if it can't) instead of trying to figure out
what the API user meant. The default ABI is still derived from the arch
component as a concession to backwards compatibility.
API-users that previously passed an explicit CPU and a triple that was
inconsistent with the CPU (e.g. mips-linux-gnu and mips64r2) may get a
different ABI to what they got before. However, it's expected that there
are no such users on the basis that CodeGen has been asserting that the
triple is consistent with the selected ABI for several releases. API-users
that were consistent or passed '' or 'generic' as the CPU will see no
difference.
Reviewers: sdardis, rafael
Subscribers: rafael, dsanders, sdardis, llvm-commits
Differential Revision: http://reviews.llvm.org/D21466
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273557 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
When parseAnyRegister() encounters a symbol alias, it parses integers and adds
a corresponding expression to the operand list. This is clearly wrong since the
only operands that parseAnyRegister() should be accepting are registers.
It's not clear why this code was added and there are no test cases that cover
it. I think it might be leftover from when searchSymbolAlias() was more widely
used.
Reviewers: sdardis
Subscribers: dsanders, sdardis, llvm-commits
Differential Revision: http://reviews.llvm.org/D21377
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273555 91177308-0d34-0410-b5e6-96231b3b80d8
This flag was introduced in r269655 with the new diagnostic handler for llc. Its
purpose was to keep the old behavior for some of the tests that didn't recover
well after an error. Those tests have been fixed, so now it's safe to remove the
flag entirely.
Fixes PR27759.
Differential Revision: http://reviews.llvm.org/D21368
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273554 91177308-0d34-0410-b5e6-96231b3b80d8
Currently isComplete = 1 requires that every instruction must
be described, declared unsupported or marked as having no
scheduling information for a processor.
For some backends such as MIPS, this requirement entails
long regex lists of instructions that are unsupported.
This patch teaches Tablegen to skip over instructions that
are associated with unsupported feature when checking if the
scheduling model is complete.
Patch by: Daniel Sanders
Contributions by: Simon Dardis
Reviewers: MatzeB
Differential Reviewer: http://reviews.llvm.org/D20522
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273551 91177308-0d34-0410-b5e6-96231b3b80d8
The exit-on-error flag was necessary in order to avoid an assertion when
handling DYNAMIC_STACKALLOC nodes in SelectionDAGLegalize.
We can avoid the assertion by creating some dummy nodes. This enables us to
remove the exit-on-error flag on the first 2 run lines (SI), but on the third
run line (R600) we would run into another assertion when trying to reserve
indirect registers. This patch also replaces that assertion with an early exit
from the function.
Fixes PR27761.
Differential Revision: http://reviews.llvm.org/D20852
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273550 91177308-0d34-0410-b5e6-96231b3b80d8
IfConversion used to always add the undef flag when adding a use operand
on a newly predicated instruction. This would be an operand for the register
being conditionally redefined. Due to the undef flag, the liveness of this
register prior to the predicated instruction would get lost.
This patch changes this so that such use operands are added only when the
register is live, without the undef flag.
Reviewed by Quentin Colombet.
http://reviews.llvm.org/D209077
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273545 91177308-0d34-0410-b5e6-96231b3b80d8
This is a cleanup commit similar to r271555, but for ARM.
The end goal is to get rid of the isSwift / isCortexXY / isWhatever methods.
Since the ARM backend seems to have quite a lot of calls to these methods, I
intend to submit 5-6 subtarget features at a time, instead of one big lump.
Differential Revision: http://reviews.llvm.org/D21432
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@273544 91177308-0d34-0410-b5e6-96231b3b80d8
