Add a JIT interface for row-wise default-nullable functions.
Not actually implemented, though. It does print out some jit-compiled stuff,
but that's about it. For example, this query:
select number from system.numbers where something(cast(number as Float64)) == 4
results in this on server's stderr:
define double @"something(CAST(number, 'Float64'))"(void**, i8*, void*) {
"something(CAST(number, 'Float64'))":
ret double 1.234500e+04
}
(and an exception, because that's what the non-jitted method does.)
As one may notice, this function neither reads the input (first argument;
tuple of arrays) nor writes the output (third argument; array), instead
returning some general nonsense.
In addition, `#if USE_EMBEDDED_COMPILER` doesn't work for some reason,
including LLVM headers requires -Wno-unused-parameter, this probably only
works on LLVM 5.0 due to rampant API instability, and I'm definitely
no expert on CMake. In short, there's still a long way to go.
2018-04-23 22:29:39 +00:00
|
|
|
#pragma once
|
|
|
|
|
|
2018-04-25 15:19:22 +00:00
|
|
|
#include <Common/config.h>
|
|
|
|
|
|
|
|
|
|
#if USE_EMBEDDED_COMPILER
|
|
|
|
|
|
2018-05-06 09:32:36 +00:00
|
|
|
#include <Interpreters/ExpressionActions.h>
|
|
|
|
|
|
Add a JIT interface for row-wise default-nullable functions.
Not actually implemented, though. It does print out some jit-compiled stuff,
but that's about it. For example, this query:
select number from system.numbers where something(cast(number as Float64)) == 4
results in this on server's stderr:
define double @"something(CAST(number, 'Float64'))"(void**, i8*, void*) {
"something(CAST(number, 'Float64'))":
ret double 1.234500e+04
}
(and an exception, because that's what the non-jitted method does.)
As one may notice, this function neither reads the input (first argument;
tuple of arrays) nor writes the output (third argument; array), instead
returning some general nonsense.
In addition, `#if USE_EMBEDDED_COMPILER` doesn't work for some reason,
including LLVM headers requires -Wno-unused-parameter, this probably only
works on LLVM 5.0 due to rampant API instability, and I'm definitely
no expert on CMake. In short, there's still a long way to go.
2018-04-23 22:29:39 +00:00
|
|
|
namespace DB
|
|
|
|
|
{
|
|
|
|
|
|
2018-04-29 01:00:26 +00:00
|
|
|
/// For each APPLY_FUNCTION action, try to compile the function to native code; if the only uses of a compilable
|
|
|
|
|
/// function's result are as arguments to other compilable functions, inline it and leave the now-redundant action as-is.
|
|
|
|
|
void compileFunctions(ExpressionActions::Actions & actions, const Names & output_columns, const Block & sample_block);
|
Add a JIT interface for row-wise default-nullable functions.
Not actually implemented, though. It does print out some jit-compiled stuff,
but that's about it. For example, this query:
select number from system.numbers where something(cast(number as Float64)) == 4
results in this on server's stderr:
define double @"something(CAST(number, 'Float64'))"(void**, i8*, void*) {
"something(CAST(number, 'Float64'))":
ret double 1.234500e+04
}
(and an exception, because that's what the non-jitted method does.)
As one may notice, this function neither reads the input (first argument;
tuple of arrays) nor writes the output (third argument; array), instead
returning some general nonsense.
In addition, `#if USE_EMBEDDED_COMPILER` doesn't work for some reason,
including LLVM headers requires -Wno-unused-parameter, this probably only
works on LLVM 5.0 due to rampant API instability, and I'm definitely
no expert on CMake. In short, there's still a long way to go.
2018-04-23 22:29:39 +00:00
|
|
|
|
|
|
|
|
}
|
2018-04-25 15:19:22 +00:00
|
|
|
|
|
|
|
|
#endif
|
