Using LIS can be quite expensive, so caching of calculated region
live-ins and pressure is implemented. It does two things:
1. Caches the info for the second stage when we schedule with
decreased target occupancy.
2. Tracks the basic block from top to bottom thus eliminating the
need to scan whole register file liveness at every region split
in the middle of the block.
The scheduling is now done in 3 stages instead of two, with the first
one being really a no-op and only used to collect scheduling regions
as sent by the scheduler driver.
There is no functional change to the current behavior, only compilation
speed is affected. In general computeBlockPressure() could be simplified
if we switch to backward RP tracker, because scheduler sends regions
within a block starting from the last upward. We could use a natural
order of upward tracker to seamlessly change between regions of the same
block, since live reg set of a previous tracked region would become a
live-out of the next region. That however requires fixing upward tracker
to properly account defs and uses of the same instruction as both are
contributing to the current pressure. When we converge on the produced
pressure we should be able to switch between them back and forth. In
addition, backward tracker is less expensive as it uses LIS in recede
less often than forward uses it in advance.
At the moment the worst known case compilation time has improved from 26
minutes to 8.5.
Differential Revision: https://reviews.llvm.org/D33117
llvm-svn: 303184
According to Intel's Optimization Reference Manual for SNB+:
" For LEA instructions with three source operands and some specific situations, instruction latency has increased to 3 cycles, and must
dispatch via port 1:
- LEA that has all three source operands: base, index, and offset
- LEA that uses base and index registers where the base is EBP, RBP,or R13
- LEA that uses RIP relative addressing mode
- LEA that uses 16-bit addressing mode "
This patch currently handles the first 2 cases only.
Differential Revision: https://reviews.llvm.org/D32277
llvm-svn: 303183
This factors register pressure estimation mechanism from the
GCNSchedStrategy into the forward tracker to unify interface
with other strategies and expose it to other interested phases.
Differential Revision: https://reviews.llvm.org/D33105
llvm-svn: 303179
This function gives the wrong answer on some non-ELF platforms in some
cases. The function that does the right thing lives in Mangler.h. To try to
discourage people from using this function, give it a different name.
Differential Revision: https://reviews.llvm.org/D33162
llvm-svn: 303134
We don't use section-relative relocations on AArch64, so all symbols must be at
least visible to the linker (i.e. properly global or l_whatever, but not
L_whatever).
llvm-svn: 303118
ARM Neon has native support for half-sized vector registers (64 bits). This
is beneficial for example for 2D and 3D graphics. This patch adds the option
to lower MinVecRegSize from 128 via a TTI in the SLP Vectorizer.
*** Performance Analysis
This change was motivated by some internal benchmarks but it is also
beneficial on SPEC and the LLVM testsuite.
The results are with -O3 and PGO. A negative percentage is an improvement.
The testsuite was run with a sample size of 4.
** SPEC
* CFP2006/482.sphinx3 -3.34%
A pretty hot loop is SLP vectorized resulting in nice instruction reduction.
This used to be a +22% regression before rL299482.
* CFP2000/177.mesa -3.34%
* CINT2000/256.bzip2 +6.97%
My current plan is to extend the fix in rL299482 to i16 which brings the
regression down to +2.5%. There are also other problems with the codegen in
this loop so there is further room for improvement.
** LLVM testsuite
* SingleSource/Benchmarks/Misc/ReedSolomon -10.75%
There are multiple small SLP vectorizations outside the hot code. It's a bit
surprising that it adds up to 10%. Some of this may be code-layout noise.
* MultiSource/Benchmarks/VersaBench/beamformer/beamformer -8.40%
The opt-viewer screenshot can be seen at F3218284. We start at a colder store
but the tree leads us into the hottest loop.
* MultiSource/Applications/lambda-0.1.3/lambda -2.68%
* MultiSource/Benchmarks/Bullet/bullet -2.18%
This is using 3D vectors.
* SingleSource/Benchmarks/Shootout-C++/Shootout-C++-lists +6.67%
Noise, binary is unchanged.
* MultiSource/Benchmarks/Ptrdist/anagram/anagram +4.90%
There is an additional SLP in the cold code. The test runs for ~1sec and
prints out over 2000 lines. This is most likely noise.
* MultiSource/Applications/aha/aha +1.63%
* MultiSource/Applications/JM/lencod/lencod +1.41%
* SingleSource/Benchmarks/Misc/richards_benchmark +1.15%
Differential Revision: https://reviews.llvm.org/D31965
llvm-svn: 303116
This caused PR33053.
Original commit message:
> The new experimental reduction intrinsics can now be used, so I'm enabling this
> for AArch64. We will need this for SVE anyway, so it makes sense to do this for
> NEON reductions as well.
>
> The existing code to match shufflevector patterns are replaced with a direct
> lowering of the reductions to AArch64-specific nodes. Tests updated with the
> new, simpler, representation.
>
> Differential Revision: https://reviews.llvm.org/D32247
llvm-svn: 303115
We were silently ignoring any features we couldn't match up, which led to
errors in an inline asm block missing the conventional "\n\t".
llvm-svn: 303108
Follow up to D33147
NVPTXTargetLowering::LowerCall was trusting the default argument values.
Fixes another 17 of the NVPTX '-verify-machineinstrs with EXPENSIVE_CHECKS' errors in PR32146.
Differential Revision: https://reviews.llvm.org/D33189
llvm-svn: 303082
This patch enables fusing dependent AESE/AESMC and AESD/AESIMC
instruction pairs on Cortex-A72, as recommended in the Software
Optimization Guide, section 4.10.
llvm-svn: 303073
Doing this means that if an LEApcrel is used in two places we will rematerialize
instead of generating two MOVs. This is particularly useful for printfs using
the same format string, where we want to generate an address into a register
that's going to get corrupted by the call.
Differential Revision: https://reviews.llvm.org/D32858
llvm-svn: 303054
Doing this lets us hoist it out of loops, and I've also marked it as
rematerializable the same as the thumb1 and thumb2 counterparts.
It looks like it being marked as such was just a mistake, as the commit that
made that change only mentions LEApcrelJT and in thumb1 and thumb2 only the
LEApcrelJT instructions were marked as having side-effects, so it looks like
the intent was to only mark LEApcrelJT as having side-effects but LEApcrel was
accidentally marked as such also.
Differential Revision: https://reviews.llvm.org/D32857
llvm-svn: 303053
NFC followup to D33147, this explicitly sets all the arguments (instead of relying on the defaults) to SelectionDAG::getMemIntrinsicNode to help identify -verify-machineinstrs issues.
llvm-svn: 303047
Replace SelectionDAG::getNode(ISD::SELECT, ...)
and SelectionDAG::getNode(ISD::VSELECT, ...)
with SelectionDAG::getSelect(...)
Saves a few lines of code and in some cases saves the need to explicitly
check the type of the desired node.
llvm-svn: 303024
Summary:
We should not change volatile loads/stores in promoting alloca to vector.
Reviewers:
arsenm
Differential Revision:
http://reviews.llvm.org/D33107
llvm-svn: 302943
This fixes 47 of the 75 NVPTX '-verify-machineinstrs with EXPENSIVE_CHECKS' errors in PR32146.
Differential Revision: https://reviews.llvm.org/D33147
llvm-svn: 302942
Summary:
Eli pointed out that it's unsafe to combine the shifts to ISD::SHL etc.,
because those are not defined for b > sizeof(a) * 8, even after some of
the combiners run.
However, PPCISD::SHL defines that behavior (as the instructions themselves).
Move the combination to the backend.
The tests in shift_mask.ll still pass.
Reviewers: echristo, hfinkel, efriedma, iteratee
Subscribers: nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D33076
llvm-svn: 302937
This patch adds min/max population count, leading/trailing zero/one bit counting methods.
The min methods return answers based on bits that are known without considering unknown bits. The max methods give answers taking into account the largest count that unknown bits could give.
Differential Revision: https://reviews.llvm.org/D32931
llvm-svn: 302925
Avoid using report_fatal_error, because it will ask the user to file a
bug. If the user attempts to disable SSE on x86_64 and them use floating
point, that's a bug in their code, not a bug in the compiler.
This is just a start. There are other ways to crash the backend in this
configuration, but they should be updated to follow this pattern.
Differential Revision: https://reviews.llvm.org/D27522
llvm-svn: 302835
According to Power ISA V3.0 document, the first source operand of mtvsrdd is constant 0 if r0 is specified. So the corresponding register constraint should be g8rc_nox0.
This bug caused wrong output generated by 401.bzip2 when -mcpu=power9 and fdo are specified.
Differential Revision: https://reviews.llvm.org/D32880
llvm-svn: 302834
Reid Kleckner