^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) /* SPDX-License-Identifier: GPL-2.0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * arch/alpha/lib/ev67-strcat.S
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) * 21264 version contributed by Rick Gorton <rick.gorton@alpha-processor.com>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * Append a null-terminated string from SRC to DST.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) * Much of the information about 21264 scheduling/coding comes from:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) * Compiler Writer's Guide for the Alpha 21264
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * abbreviated as 'CWG' in other comments here
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) * ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) * Scheduling notation:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) * E - either cluster
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) * U - upper subcluster; U0 - subcluster U0; U1 - subcluster U1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) * L - lower subcluster; L0 - subcluster L0; L1 - subcluster L1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) * Try not to change the actual algorithm if possible for consistency.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) * Commentary: It seems bogus to walk the input string twice - once
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) * to determine the length, and then again while doing the copy.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) * A significant (future) enhancement would be to only read the input
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) * string once.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <asm/export.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) .text
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) .align 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) .globl strcat
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) .ent strcat
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) strcat:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) .frame $30, 0, $26
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) .prologue 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) mov $16, $0 # E : set up return value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) /* Find the end of the string. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) ldq_u $1, 0($16) # L : load first quadword (a0 may be misaligned)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) lda $2, -1 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) insqh $2, $16, $2 # U :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) andnot $16, 7, $16 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) or $2, $1, $1 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) cmpbge $31, $1, $2 # E : bits set iff byte == 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) bne $2, $found # U :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) $loop: ldq $1, 8($16) # L :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) addq $16, 8, $16 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) cmpbge $31, $1, $2 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) beq $2, $loop # U :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) $found: cttz $2, $3 # U0 :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) addq $16, $3, $16 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) /* Now do the append. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) mov $26, $23 # E :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) br __stxcpy # L0 :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) .end strcat
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) EXPORT_SYMBOL(strcat)