tmp := make([]byte, 5)
	for i := 0; i <= 5; i++ {
		for j := i; j <= 5; j++ {
			for k := 0; k <= 6; k++ {
				// 0 <= i <= j <= 5; 0 <= k <= 6
				// Check that if we start with a buffer
				// of length j at offset i and ask for
				// Next(k), we get the right bytes.
				buf := NewBuffer(b[0:j])
				n, _ := buf.Read(tmp[0:i])
				if n != i {
					t.Fatalf("Read %d returned %d", i, n)
				}
				bb := buf.Next(k)

and we want to preserve the ability to compile pure Go code that
imports net without requiring gcc to be present at link time. (In this
case, the dynamic library requirement is less significant, because the
only library involved is libc.so, which can usually be assumed
present.)

This conflict between functionality and the gcc requirement means we

	// is too large we are basically just thrashing the CPU D-cache.
	// So if the result length is larger than an empirically-found
	// limit (8KB), we stop growing the source string once the limit
	// is reached and keep reusing the same source string - that
	// should therefore be always resident in the L1 cache - until we
	// have completed the construction of the result.

next:
	// Skip newlines.
	var tok lex.ScanToken
	for {
		tok = p.nextToken()
		// We save the line number here so error messages from this instruction
		// are labeled with this line. Otherwise we complain after we've absorbed
		// the terminating newline and the line numbers are off by one in errors.
		p.lineNum = p.lex.Line()
		switch tok {
		case '\n', ';':

(discussed below).

When preparing any change that is permitted by Go 1 compatibility
but may nonetheless break some existing programs,
we first engineer the change to keep as many existing programs working as possible.
For the remaining programs,
we define a new GODEBUG setting that
allows individual programs to opt back in to the old behavior.
A GODEBUG setting may not be added if doing so is infeasible,

	// That's the magic marker for a 64-bit file, so even though
	// it fits in a 32-bit field we must use the 64-bit field.
	// Go 1.5 and earlier got this wrong,
	// writing an invalid zip file.
	const size = 1<<32 - 1 - int64(len("END\n")) // before the "END\n" part
	buf := testZip64(t, size)
	testZip64DirectoryRecordLength(buf, t)
}

// Tests that we generate a zip64 file if the directory at offset
// 0xFFFFFFFF, but not before.

	paxGNUSparseMinor     = "GNU.sparse.minor"
	paxGNUSparseSize      = "GNU.sparse.size"
	paxGNUSparseRealSize  = "GNU.sparse.realsize"
)

// basicKeys is a set of the PAX keys for which we have built-in support.
// This does not contain "charset" or "comment", which are both PAX-specific,
// so adding them as first-class features of Header is unlikely.
// Users can use the PAXRecords field to set it themselves.

In general, machine-specific operations tend to appear as themselves, while more general concepts like
memory move and subroutine call and return are more abstract.
The details vary with architecture, and we apologize for the imprecision; the situation is not well-defined.
</p>

<p>
The assembler program is a way to parse a description of that
semi-abstract instruction set and turn it into instructions to be

	//	name and comment fields as whatever the local system encoding is.
	//
	// In order to avoid breaking readers without UTF-8 support,
	// we avoid setting the UTF-8 flag if the strings are CP-437 compatible.
	// However, if the strings require multibyte UTF-8 encoding and is a
	// valid UTF-8 string, then we set the UTF-8 bit.
	//
	// For the case, where the user explicitly wants to specify the encoding

// to avoid misleading negative results.
// This makes all the references to os.FileInfo in go1.txt
// be read as if they said fs.FileInfo, since os.FileInfo is now an alias.
// If there are many of these, we could do a more general solution,
// but for now the replacer is fine.
var aliasReplacer = strings.NewReplacer(
	"os.FileInfo", "fs.FileInfo",
	"os.FileMode", "fs.FileMode",
	"os.PathError", "fs.PathError",
)