|  | /* | 
|  | * arch/alpha/lib/ev67-strlen.S | 
|  | * 21264 version by Rick Gorton <rick.gorton@alpha-processor.com> | 
|  | * | 
|  | * Finds length of a 0-terminated string.  Optimized for the | 
|  | * Alpha architecture: | 
|  | * | 
|  | *	- memory accessed as aligned quadwords only | 
|  | *	- uses bcmpge to compare 8 bytes in parallel | 
|  | * | 
|  | * Much of the information about 21264 scheduling/coding comes from: | 
|  | *	Compiler Writer's Guide for the Alpha 21264 | 
|  | *	abbreviated as 'CWG' in other comments here | 
|  | *	ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html | 
|  | * Scheduling notation: | 
|  | *	E	- either cluster | 
|  | *	U	- upper subcluster; U0 - subcluster U0; U1 - subcluster U1 | 
|  | *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1 | 
|  | */ | 
|  |  | 
|  | .set noreorder | 
|  | .set noat | 
|  |  | 
|  | .globl	strlen | 
|  | .ent	strlen | 
|  | .align 4 | 
|  | strlen: | 
|  | ldq_u	$1, 0($16)	# L : load first quadword ($16  may be misaligned) | 
|  | lda	$2, -1($31)	# E : | 
|  | insqh	$2, $16, $2	# U : | 
|  | andnot	$16, 7, $0	# E : | 
|  |  | 
|  | or	$2, $1, $1	# E : | 
|  | cmpbge	$31, $1, $2	# E : $2  <- bitmask: bit i == 1 <==> i-th byte == 0 | 
|  | nop			# E : | 
|  | bne	$2, $found	# U : | 
|  |  | 
|  | $loop:	ldq	$1, 8($0)	# L : | 
|  | addq	$0, 8, $0	# E : addr += 8 | 
|  | cmpbge	$31, $1, $2	# E : | 
|  | beq	$2, $loop	# U : | 
|  |  | 
|  | $found: | 
|  | cttz	$2, $3		# U0 : | 
|  | addq	$0, $3, $0	# E : | 
|  | subq	$0, $16, $0	# E : | 
|  | ret	$31, ($26)	# L0 : | 
|  |  | 
|  | .end	strlen |