Some instructions from the original loop, when vectorized, can become trivially
dead. This happens because of the way we structure the new loop. For example,
we create new induction variables and induction variable "steps" in the new
loop. Thus, when we go to vectorize the original induction variable update, it
may no longer be needed due to the instructions we've already created. This
patch prevents us from creating these redundant instructions. This reduces code
size before simplification and allows greater flexibility in code generation
since we have fewer unnecessary instruction uses.
Differential Revision: https://reviews.llvm.org/D25631
llvm-svn: 284631
This change is motivated by the case when IndVarSimplify doesn't widen a comparison of IV increment because it can't prove IV increment being non-negative. We end up with a redundant trunc of the widened increment on this example.
for.body:
%i = phi i32 [ %start, %for.body.lr.ph ], [ %i.inc, %for.inc ]
%within_limits = icmp ult i32 %i, 64
br i1 %within_limits, label %continue, label %for.end
continue:
%i.i64 = zext i32 %i to i64
%arrayidx = getelementptr inbounds i32, i32* %base, i64 %i.i64
%val = load i32, i32* %arrayidx, align 4
br label %for.inc
for.inc:
%i.inc = add nsw nuw i32 %i, 1
%cmp = icmp slt i32 %i.inc, %limit
br i1 %cmp, label %for.body, label %for.end
There is a range check inside of the loop which guarantees the IV to be non-negative. NSW on the increment guarantees that the increment is also non-negative. Teach IndVarSimplify to use the range check to prove non-negativity of loop increments.
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D25738
llvm-svn: 284629
This augments the STLExtras toolset with a zip iterator and range
adapter. Zip comes in two varieties: `zip`, which will zip to the
shortest of the input ranges, and `zip_first`, which limits its
`begin() == end()` checks to just the first range.
Recommit r284035 after MSVC2013 support has been dropped.
Patch by: Bryant Wong <github.com/bryant>
Differential Revision: https://reviews.llvm.org/D23252
llvm-svn: 284623
Initializing a ThreadPool with ThreadCount = 1 spawns a thread even
though we don't need to. This is at least slower than it needs to be,
and at worst may somehow be exacerbating PR30735 (llvm-cov times out
on ARM bots).
As a follow-up, I'll try to add logic to llvm::ThreadPool to avoid
spawning a thread when ThreadCount = 1.
llvm-svn: 284621
Summary:
Changes default backend parallelism from thread::hardware_concurrency to
the new llvm::heavyweight_hardware_concurrency, which for X86 Linux
defaults to the number of physical cores (and will fall back to
thread::hardware_concurrency otherwise). This avoid oversubscribing
the physical cores using hyperthreading.
Reviewers: mehdi_amini, pcc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25775
llvm-svn: 284618
This reverts commit r284590 as it fails on the mingw buildbot. I think I know the
fix, but I cannot test it right now. Will reapply when I verify it works ok.
This reverts r284590.
llvm-svn: 284615
Use mask and negate for legalization of i1 source type with SIGN_EXTEND_INREG.
With the mask, this should be no worse than 2 shifts. The mask can be eliminated
in some cases, so that should be better than 2 shifts.
This change exposed some missing folds related to negation:
https://reviews.llvm.org/rL284239https://reviews.llvm.org/rL284395
There may be others, so please let me know if you see any regressions.
Differential Revision: https://reviews.llvm.org/D25485
llvm-svn: 284611
This required reengineering of some of the part of liveness calculation,
including fixing some issues caused by the limitations of the previous
approach. The current code is not necessarily the fastest, but it should
be functionally correct (at least more so than before). The compile-time
performance will be addressed in the future.
llvm-svn: 284609
Summary:
std::chrono mostly covers the functionality of llvm::sys::TimeValue and
lldb_private::TimeValue. This header adds a bit of utility functions and
typedefs, which make the usage of the library and porting code from TimeValues
easier.
Rationale:
- TimePoint typedef - precision of system_clock is implementation defined -
using a well-defined precision helps maintain consistency between platforms,
makes it interact better with existing TimeValue classes, and avoids cases
there a time point is implicitly convertible to a specific precision on some
platforms but not on others.
- system_clock::to_time_t only accepts time_points with the default system
precision (even though time_t has only second precision on all platforms we
support). To avoid the need for explicit casts, I have added a toTimeT()
wrapper function. toTimePoint(time_t) was not strictly necessary, but I have
added it for symmetry.
Reviewers: zturner, mehdi_amini
Subscribers: beanz, mgorny, llvm-commits, modocache
Differential Revision: https://reviews.llvm.org/D25416
llvm-svn: 284590
Most z13 vector instructions have a base form where the data type of
the operation (whether to consider the vector to be 16 bytes, 8
halfwords, 4 words, or 2 doublewords) is encoded into a mask field,
and then a set of extended mnemonics where the mask field is not
present but the data type is encoded into the mnemonic name.
Currently, LLVM only supports the type-specific forms (since those
are really the ones needed for code generation), but not the base
type-generic forms.
To complete the assembler support and make it fully compatible with
the GNU assembler, this commit adds assembler aliases for all the
base forms of the various vector instructions.
It also adds two more alias forms that are documented in the PoP:
VFPSO/VFPSODB/WFPSODB -- generic form of VFLCDB etc.
VNOT -- special variant of VNO
llvm-svn: 284586
The vfee[bhf], vfene[bhf], and vistr[bhf] assembler mnemonics are
documented in the Principles of Operation to have an optional last
operand to encode arbitrary values in a mask field.
This commit adds support for those optional operands, and cleans up
the patterns to generate vector string instruction as bit. No change
to code generation intended.
llvm-svn: 284585
Declare the LLVM_CMAKE_PATH to the source directory location of CMake
files, in order to make it possible to easily use them in subprojects.
Such a variable is already declared in most of LLVM projects
(and inconsistently mixed with direct source tree references), including
Clang, LLDB, compiler-rt, libcxx... Declaring it inside main LLVM tree
makes it possible to avoid having to declare fallback values or use
conditionals in those projects.
It should be noted that in some of the subprojects LLVM_CMAKE_PATH is
used to reference generated LLVMConfig.cmake file. However, these
references are conditional to stand-alone builds and explicitly
including this file is unnecessary in combined builds.
Differential Revision: https://reviews.llvm.org/D25724
llvm-svn: 284581
The TBB and TBH instructions in Thumb-2 allow jump tables to be compressed into sequences of bytes or shorts respectively. These instructions do not exist in Thumb-1, however it is possible to synthesize them out of a sequence of other instructions.
It turns out this sequence is so short that it's almost never a lose for performance and is ALWAYS a significant win for code size.
TBB example:
Before: lsls r0, r0, #2 After: add r0, pc
adr r1, .LJTI0_0 ldrb r0, [r0, #6]
ldr r0, [r0, r1] lsls r0, r0, #1
mov pc, r0 add pc, r0
=> No change in prologue code size or dynamic instruction count. Jump table shrunk by a factor of 4.
The only case that can increase dynamic instruction count is the TBH case:
Before: lsls r0, r4, #2 After: lsls r4, r4, #1
adr r1, .LJTI0_0 add r4, pc
ldr r0, [r0, r1] ldrh r4, [r4, #6]
mov pc, r0 lsls r4, r4, #1
add pc, r4
=> 1 more instruction in prologue. Jump table shrunk by a factor of 2.
So there is an argument that this should be disabled when optimizing for performance (and a TBH needs to be generated). I'm not so sure about that in practice, because on small cores with Thumb-1 performance is often tied to code size. But I'm willing to turn it off when optimizing for performance if people want (also note that TBHs are fairly rare in practice!)
llvm-svn: 284580
This will get the same ConstantSDNode scalar or vector splat value as the current separate dyn_cast<ConstantSDNode> / isVector() approach.
llvm-svn: 284578
This renames the function for checking FP function attribute values and also
adds more build attribute tests (which are in separate files because build
attributes are set per file).
Differential Revision: https://reviews.llvm.org/D25625
llvm-svn: 284571
msc18 doesn't recognize "using BaseT::BaseT;"
llvm\include\llvm/ADT/DenseSet.h(213) : error C2875: using-declaration causes a multiple declaration of 'BaseT'
llvm\include\llvm/ADT/DenseSet.h(214) : see reference to class template instantiation 'llvm::DenseSet<ValueT,ValueInfoT>' being compiled
llvm\include\llvm/ADT/DenseSet.h(231) : error C2875: using-declaration causes a multiple declaration of 'BaseT'
llvm\include\llvm/ADT/DenseSet.h(232) : see reference to class template instantiation 'llvm::SmallDenseSet<ValueT,InlineBuckets,ValueInfoT>' being compiled
llvm-svn: 284570
Summary:
This allows us to create broadcasts of 128-bit vector loads into 512-bit vectors.
New patterns added to support 8-bit and 16-bit vector types and v2f64/v2i64->v8f64/v8i64 without DQI instructions.
There also fallback patterns when the load can't be folded. These patterns are a little complex as we first need to insert the lower 128-bits into the second 128-bits using a zmm subvector insert instruction. We need to use a zmm insert in case VLX isn't available. Then use another zmm sub vector insert to take those 256-bits and insert them into the upper bits. Since we used a zmm insert to create the 256-bits we also need to do a extract_subreg to get just the lower 256-bits to pass to the second insert.
The outer insert for the fallback patterns should have its type correct because eventually we should also supported masked operations here too. So we need a DQI and a NoDQI version of the v16f32/v16i32 patterns.
Reviewers: RKSimon, delena, igorb
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25651
llvm-svn: 284567
Example of output:
COVERAGE:
COVERED: in DSO2(int) /pathto/DSO2.cpp:6
COVERED: in DSO2(int) /pathto/DSO2.cpp:8
COVERED: in DSO1(int) /pathto/DSO1.cpp:6
COVERED: in DSO1(int) /pathto/DSO1.cpp:8
COVERED: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:16
COVERED: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:19
COVERED: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:25
COVERED: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:26
MODULE_WITH_COVERAGE: /pathto/libLLVMFuzzer-DSO1.so
UNCOVERED_LINE: in DSO1(int) /pathto/DSO1.cpp:9
UNCOVERED_FUNC: in Uncovered1()
MODULE_WITH_COVERAGE: /pathto/libLLVMFuzzer-DSO2.so
UNCOVERED_LINE: in DSO2(int) /pathto/DSO2.cpp:9
UNCOVERED_FUNC: in Uncovered2()
MODULE_WITH_COVERAGE: /pathto/LLVMFuzzer-DSOTest
UNCOVERED_LINE: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:21
UNCOVERED_LINE: in LLVMFuzzerTestOneInput /pathto/DSOTestMain.cpp:27
UNCOVERED_FILE: /pathto/DSOTestExtra.cpp
Several things are not perfect here:
* we are using objdump+awk instead of sancov because sancov does not support DSOs yet.
* this breaks in the presence of ASAN_OPTIONS=strip_path_prefix=...
(need to implement another API to get the module name by PC)
llvm-svn: 284554
Summary:
The original heuristic to break critical edge during machine sink is relatively conservertive: when there is only one instruction sinkable to the critical edge, it is likely that the machine sink pass will not break the critical edge. This leads to many speculative instructions executed at runtime. However, with profile info, we could model the splitting benefits: if the critical edge has 50% taken rate, it would always be beneficial to split the critical edge to avoid the speculated runtime instructions. This patch uses profile to guide critical edge splitting in machine sink pass.
The performance impact on speccpu2006 on Intel sandybridge machines:
spec/2006/fp/C++/444.namd 25.3 +0.26%
spec/2006/fp/C++/447.dealII 45.96 -0.10%
spec/2006/fp/C++/450.soplex 41.97 +1.49%
spec/2006/fp/C++/453.povray 36.83 -0.96%
spec/2006/fp/C/433.milc 23.81 +0.32%
spec/2006/fp/C/470.lbm 41.17 +0.34%
spec/2006/fp/C/482.sphinx3 48.13 +0.69%
spec/2006/int/C++/471.omnetpp 22.45 +3.25%
spec/2006/int/C++/473.astar 21.35 -2.06%
spec/2006/int/C++/483.xalancbmk 36.02 -2.39%
spec/2006/int/C/400.perlbench 33.7 -0.17%
spec/2006/int/C/401.bzip2 22.9 +0.52%
spec/2006/int/C/403.gcc 32.42 -0.54%
spec/2006/int/C/429.mcf 39.59 +0.19%
spec/2006/int/C/445.gobmk 26.98 -0.00%
spec/2006/int/C/456.hmmer 24.52 -0.18%
spec/2006/int/C/458.sjeng 28.26 +0.02%
spec/2006/int/C/462.libquantum 55.44 +3.74%
spec/2006/int/C/464.h264ref 46.67 -0.39%
geometric mean +0.20%
Manually checked 473 and 471 to verify the diff is in the noise range.
Reviewers: rengolin, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24818
llvm-svn: 284545
Summary:
This pass shrink-wraps a condition to some library calls where the call
result is not used. For example:
sqrt(val);
is transformed to
if (val < 0)
sqrt(val);
Even if the result of library call is not being used, the compiler cannot
safely delete the call because the function can set errno on error
conditions.
Note in many functions, the error condition solely depends on the incoming
parameter. In this optimization, we can generate the condition can lead to
the errno to shrink-wrap the call. Since the chances of hitting the error
condition is low, the runtime call is effectively eliminated.
These partially dead calls are usually results of C++ abstraction penalty
exposed by inlining. This optimization hits 108 times in 19 C/C++ programs
in SPEC2006.
Reviewers: hfinkel, mehdi_amini, davidxl
Subscribers: modocache, mgorny, mehdi_amini, xur, llvm-commits, beanz
Differential Revision: https://reviews.llvm.org/D24414
llvm-svn: 284542
Summary:
The original heuristic to break critical edge during machine sink is relatively conservertive: when there is only one instruction sinkable to the critical edge, it is likely that the machine sink pass will not break the critical edge. This leads to many speculative instructions executed at runtime. However, with profile info, we could model the splitting benefits: if the critical edge has 50% taken rate, it would always be beneficial to split the critical edge to avoid the speculated runtime instructions. This patch uses profile to guide critical edge splitting in machine sink pass.
The performance impact on speccpu2006 on Intel sandybridge machines:
spec/2006/fp/C++/444.namd 25.3 +0.26%
spec/2006/fp/C++/447.dealII 45.96 -0.10%
spec/2006/fp/C++/450.soplex 41.97 +1.49%
spec/2006/fp/C++/453.povray 36.83 -0.96%
spec/2006/fp/C/433.milc 23.81 +0.32%
spec/2006/fp/C/470.lbm 41.17 +0.34%
spec/2006/fp/C/482.sphinx3 48.13 +0.69%
spec/2006/int/C++/471.omnetpp 22.45 +3.25%
spec/2006/int/C++/473.astar 21.35 -2.06%
spec/2006/int/C++/483.xalancbmk 36.02 -2.39%
spec/2006/int/C/400.perlbench 33.7 -0.17%
spec/2006/int/C/401.bzip2 22.9 +0.52%
spec/2006/int/C/403.gcc 32.42 -0.54%
spec/2006/int/C/429.mcf 39.59 +0.19%
spec/2006/int/C/445.gobmk 26.98 -0.00%
spec/2006/int/C/456.hmmer 24.52 -0.18%
spec/2006/int/C/458.sjeng 28.26 +0.02%
spec/2006/int/C/462.libquantum 55.44 +3.74%
spec/2006/int/C/464.h264ref 46.67 -0.39%
geometric mean +0.20%
Manually checked 473 and 471 to verify the diff is in the noise range.
Reviewers: rengolin, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24818
llvm-svn: 284541
