processSEH(l, arch, ls.PData, ls.XData)
	}

	return &ls, nil
}

// PostProcessImports works to resolve inconsistencies with DLL import
// symbols; it is needed when building with more "modern" C compilers
// with internal linkage.
//
// Background: DLL import symbols are data (SNOPTRDATA) symbols whose
// name is of the form "__imp_XXX", which contain a pointer/reference

Other language plugins follow the same pattern.
The last portion of the directory path usually indicates the expected language of the source files.

Some compilers are capable of cross-compiling multiple languages in the same source directory.

</p>

<p>
Note that all these optimizations are permitted in C/C++ compilers:
a Go compiler sharing a back end with a C/C++ compiler must take care
to disable optimizations that are invalid for Go.
</p>

<p>
Note that the prohibition on introducing data races
does not apply if the compiler can prove that the races
do not affect correct execution on the target platform.

Some of the details map precisely to the machine, but some do not.
This is because the compiler suite (see
<a href="https://9p.io/sys/doc/compiler.html">this description</a>)
needs no assembler pass in the usual pipeline.
Instead, the compiler operates on a kind of semi-abstract instruction set,
and instruction selection occurs partly after code generation.

This is similar to a comparison between a high level programming language compared to assembly;
while its always possible to write better code in assembly, smart compilers can make optimizations humans are unlikely to,
such as loop unrolling.
Similarly, `krt` can make complex optimizations in one place, so each controller implementation doesn't, which is likely to increase

		defs = append(defs, `-D`, `GOPKGPATH="`+pkgpath+`"`)
	}
	compiler := envList("CC", cfg.DefaultCC(cfg.Goos, cfg.Goarch))
	if b.gccSupportsFlag(compiler, "-fsplit-stack") {
		defs = append(defs, "-fsplit-stack")
	}
	defs = tools.maybePIC(defs)
	if b.gccSupportsFlag(compiler, "-ffile-prefix-map=a=b") {
		defs = append(defs, "-ffile-prefix-map="+base.Cwd()+"=.")

#include "stablehlo/dialect/StablehloOps.h"  // from @stablehlo
#include "tensorflow/compiler/mlir/quantization/common/func.h"
#include "tensorflow/compiler/mlir/quantization/common/lift_as_function_call.h"
#include "tensorflow/compiler/mlir/quantization/stablehlo/utils/stablehlo_type_utils.h"
#include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.h"
#include "tensorflow/compiler/mlir/tensorflow/utils/xla_call_module_attrs.h"

exists in PPC64 assembler and is frequently used by PPC64 assembler writers.

PCALIGN $16
PCALIGN $8

By default, functions in Go are aligned to 16 bytes, as is the case in all
other compilers for PPC64. If there is a PCALIGN directive requesting alignment
greater than 16, then the alignment of the containing function must be
promoted to that same alignment or greater.

// or the empty string if there is none.
func (s *Sym) PackageName() string {
	name := s.nameWithoutInst()

	// Since go1.20, a prefix of "type:" and "go:" is a compiler-generated symbol,
	// they do not belong to any package.
	//
	// See cmd/compile/internal/base/link.go:ReservedImports variable.
	if s.goVersion >= ver120 && (strings.HasPrefix(name, "go:") || strings.HasPrefix(name, "type:")) {
		return ""
	}

 * Gradle will no longer automatically apply annotation processors that are on the compile classpath — use link:{groovyDslPath}/org.gradle.api.tasks.compile.CompileOptions.html#org.gradle.api.tasks.compile.CompileOptions:annotationProcessorPath[CompileOp...