Source code of Windows XP (NT5)
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

172 lines
4.1 KiB

// string.s: function to concatenate 2 strings
// Copyright (c) 2000, Intel Corporation
// All rights reserved.
//
// WARRANTY DISCLAIMER
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL INTEL OR ITS
// CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
// EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
// OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY OR TORT (INCLUDING
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Intel Corporation is the author of this code, and requests that all
// problem reports or change requests be submitted to it directly at
// http://developer.intel.com/opensource.
//
.file "string.s"
.section .text
// -- Begin strcat
.proc strcat#
.global strcat#
.align 32
strcat:
{ .mib
alloc r14=ar.pfs,2,6,0,8 //8 rotating registers, 7 locals
mov r11=pr //Save predicate register file
brp.loop.imp .bs1len, .bws1 // Put loop backedge target in TAR
}
// Setup for doing software pipelined loops
{ .mib
mov r9=r33
mov pr.rot=0x30000 // p16=p17=1
nop.b 0
};;
{ .mib
mov r14=r32
mov ar.ec=0
nop.b 0
} { .mib // Extra bundle to align bs1len.
mov r8=r32
nop.i 0
brp.loop.imp .bcat, .bwcat ;; // Put loop backedge target in TAR
}
.bs1len:
{ .mii
ld1.s r37=[r14],1 // *s (r37,r38,r39)
nop.i 0
(p19) chk.s r39,.natfault1_0 //
}
.bws1:
{ .mfb
(p19) cmp4.ne p17,p0=r39,r0 // *s==0 (p16,p17,p18)
nop.f 0
(p17) br.wtop.dptk .bs1len ;; //
}
//
// Now concatenate s2 into the end of s1
//
{ .mib
add r14=-3,r14 // Since ld1.s is 2 stages ahead
dep r15=1,r0,32,32 // rb = 0xffffffff00000000
clrrrb ;;
} { .mii
// Setup for doing software pipelined loops
or r32=r14,r9
mov pr.rot=0x30000 ;; // p16=p17=1
and r32=3,r32 ;;
} { .mib
cmp4.ne p10,p0=r32,r0
mov ar.ec=0
(p10) br.spnt .b_notaligned ;;
}
.bcat:
{ .mii
ld4.s r32=[r9],4 // *s1 (r32,r33,r34)
(p18) chk.s r33,.natfault2_0 //
(p18) pcmp1.eq r16=r33,r15 ;; // r16 !=0 only if a zero byte is found
}
.bwcat:
{ .mib
(p19) st4 [r14]=r34,4 // *s2=*s1
(p18) cmp4.eq p17,p0=r16,r0 // zero byte found?
(p17) br.wtop.dptk .bcat ;; //
}
{ .mfi
nop.m 0
nop.f 0
czx1.r r16 = r33
} ;;
{ .mfi
cmp.leu p2, p0 = 2, r16
nop.f 0
shr.u r35 = r33, 8
}
{ .mfi
cmp.eq p4, p0 = 3, r16
nop.f 0
cmp.ne p5, p0 = r0, r16
} ;;
{ .mfi
(p5)st1 [r14] = r33, 1
nop.f 0
shr.u r36 = r33, 16
};;
{ .mfi
(p2)st1 [r14] = r35,1
nop.f 0
nop.i 0
} ;;
{ .mfi
(p4)st1 [r14] = r36,1
nop.f 0
nop.i 0
};;
{ .mib
(p0) st1 [r14] = r0
nop.i 0
clrrrb
} ;;
{ .mib
nop.m 0
mov pr=r11,0x1003e
br.ret.sptk.many b0 ;;
}
.b_notaligned:
{ .mmi
ld1 r32=[r9],1 ;; // 2 cycle load causes 1 cycle stall
st1 [r14]=r32,1 // 3 cycles between st1 to avoid flush
cmp4.ne.unc p7,p0=r32,r0 ;; // Extra stop bit to force 3 cycles
} { .mib
nop.m 0
nop.i 0
(p7) br.cond.dptk .b_notaligned ;;
} { .mib
nop.m 0
mov pr=r11,0x1003e
br.ret.sptk.many b0 ;;
}
.natfault1_0:
{ .mmi
add r39=-3,r14 ;;
ld1 r39=[r39] // Redo the load
nop.i 0
} { .mib
nop.m 0
nop.i 0
br.sptk .bws1 ;;
}
.natfault2_0:
{ .mmi
add r33=-8,r9 ;;
ld4 r33=[r33] // *s1 (r32,r33,r34)
nop.i 0;;
} { .mib
nop.m 0
(p18) pcmp1.eq r16=r33,r15 // r16 !=0 only if a zero byte is found
br.sptk .bwcat ;;
}
_2_1_2auto_size == 0x0
// -- End strcat
.endp strcat#
// mark_proc_addr_taken strcat;
// End