123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203 |
- /*
- * Copyright (c) 2006 - 2008, Intel Corporation. All rights reserved.
- * This program and the accompanying materials are licensed and made
- available
- * under the terms and conditions of the BSD License which accompanies
- this
- * distribution. The full text of the license may be found at
- * http://opensource.org/licenses/bsd-license.php.
- *
- * THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS"
- BASIS,
- * WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR
- * IMPLIED.
- */
- .text
- .globl setjmp
- .type setjmp, @function
- setjmp:
- alloc loc0 = ar.pfs, 1, 2, 1, 0
- ;;
- mov r14 = ar.unat
- mov r15 = ar.bsp
- add r10 = 0x10*20, in0
- ;;
- stf.spill.nta [in0] = f2, 0x10
- st8.spill.nta [r10] = r4, 8
- mov r21 = b1
- ;;
- stf.spill.nta [in0] = f3, 0x10
- st8.spill.nta [r10] = r5, 8
- mov r22 = b2
- ;;
- stf.spill.nta [in0] = f4, 0x10
- st8.spill.nta [r10] = r6, 8
- mov r23 = b3
- ;;
- stf.spill.nta [in0] = f5, 0x10
- st8.spill.nta [r10] = r7, 8
- mov r24 = b4
- ;;
- stf.spill.nta [in0] = f16, 0x10
- st8.spill.nta [r10] = sp, 8
- mov r25 = b5
- ;;
- stf.spill.nta [in0] = f17, 0x10
- st8.nta [r10] = loc1, 8
- mov r16 = pr
- ;;
- stf.spill.nta [in0] = f18, 0x10
- st8.nta [r10] = r21, 8
- mov r17 = ar.lc
- ;;
- stf.spill.nta [in0] = f19, 0x10
- st8.nta [r10] = r22, 8
- ;;
- stf.spill.nta [in0] = f20, 0x10
- st8.nta [r10] = r23, 8
- ;;
- stf.spill.nta [in0] = f21, 0x10
- st8.nta [r10] = r24, 8
- ;;
- stf.spill.nta [in0] = f22, 0x10
- st8.nta [r10] = r25, 8
- ;;
- stf.spill.nta [in0] = f23, 0x10
- mov r18 = ar.unat
- ;;
- stf.spill.nta [in0] = f24, 0x10
- st8.nta [r10] = r14, 8
- ;;
- stf.spill.nta [in0] = f25, 0x10
- st8.nta [r10] = r18, 8
- ;;
- stf.spill.nta [in0] = f26, 0x10
- st8.nta [r10] = loc0, 8
- ;;
- stf.spill.nta [in0] = f27, 0x10
- st8.nta [r10] = r15, 8
- mov r8 = 0
- ;;
- stf.spill.nta [in0] = f28, 0x10
- mov r19 = ar.fpsr
- ;;
- stf.spill.nta [in0] = f29, 0x10
- st8.nta [r10] = r16, 8
- mov ar.pfs = loc0
- ;;
- stf.spill.nta [in0] = f30, 0x10
- st8.nta [r10] = r17, 8
- mov b0 = loc1
- ;;
- stf.spill.nta [in0] = f31, 0x10
- st8.nta [r10] = r19
- ;;
- mov ar.unat = r14
- br.ret.sptk b0
- ;;
- .globl longjmp
- .type longjmp, @function
- .regstk 2, 0, 0, 0
- longjmp:
- add r10 = 0x10*20 + 8*14, in0
- movl r2 = ~((((1<<14) - 1) << 16) | 3)
- ;;
- ld8.nt1 r14 = [r10], -8*2
- mov r15 = ar.bspstore
- ;;
- ld8.nt1 r17 = [r10], -8
- mov r16 = ar.rsc
- cmp.leu p6 = r14, r15
- ;;
- ld8.nt1 r18 = [r10], -8
- ld8.nt1 r25 = [r10], -8
- and r2 = r16, r2
- ;;
- ldf.fill.nt1 f2 = [in0], 0x10
- ld8.nt1 r24 = [r10], -8
- mov b5 = r25
- ;;
- mov ar.rsc = r2
- ld8.nt1 r23 = [r10], -8
- mov b4 = r24
- ;;
- ldf.fill.nt1 f3 = [in0], 0x10
- mov ar.unat = r17
- (p6) br.spnt.many _skip_flushrs
- ;;
- flushrs
- mov r15 = ar.bsp
- ;;
- _skip_flushrs:
- mov r31 = ar.rnat
- loadrs
- ;;
- ldf.fill.nt1 f4 = [in0], 0x10
- ld8.nt1 r22 = [r10], -8
- dep r2 = -1, r14, 3, 6
- ;;
- ldf.fill.nt1 f5 = [in0], 0x10
- ld8.nt1 r21 = [r10], -8
- cmp.ltu p6 = r2, r15
- ;;
- ld8.nt1 r20 = [r10], -0x10
- (p6) ld8.nta r31 = [r2]
- mov b3 = r23
- ;;
- ldf.fill.nt1 f16 = [in0], 0x10
- ld8.fill.nt1 r7 = [r10], -8
- mov b2 = r22
- ;;
- ldf.fill.nt1 f17 = [in0], 0x10
- ld8.fill.nt1 r6 = [r10], -8
- mov b1 = r21
- ;;
- ldf.fill.nt1 f18 = [in0], 0x10
- ld8.fill.nt1 r5 = [r10], -8
- mov b0 = r20
- ;;
- ldf.fill.nt1 f19 = [in0], 0x10
- ld8.fill.nt1 r4 = [r10], 8*13
- ;;
- ldf.fill.nt1 f20 = [in0], 0x10
- ld8.nt1 r19 = [r10], 0x10
- ;;
- ldf.fill.nt1 f21 = [in0], 0x10
- ld8.nt1 r26 = [r10], 8
- mov ar.pfs = r19
- ;;
- ldf.fill.nt1 f22 = [in0], 0x10
- ld8.nt1 r27 = [r10], 8
- mov pr = r26, -1
- ;;
- ldf.fill.nt1 f23 = [in0], 0x10
- ld8.nt1 r28 = [r10], -17*8 - 0x10
- mov ar.lc = r27
- ;;
- ldf.fill.nt1 f24 = [in0], 0x10
- ldf.fill.nt1 f25 = [in0], 0x10
- mov r8 = in1
- ;;
- ldf.fill.nt1 f26 = [in0], 0x10
- ldf.fill.nt1 f31 = [r10], -0x10
- ;;
- ldf.fill.nt1 f27 = [in0], 0x10
- ldf.fill.nt1 f30 = [r10], -0x10
- ;;
- ldf.fill.nt1 f28 = [in0]
- ldf.fill.nt1 f29 = [r10], 0x10*3 + 8*4
- ;;
- ld8.fill.nt1 sp = [r10]
- mov ar.unat = r18
- ;;
- mov ar.bspstore = r14
- mov ar.rnat = r31
- ;;
- invala
- mov ar.rsc = r16
- br.ret.sptk b0
- #if defined(__ELF__) && defined(__linux__)
- .section .note.GNU-stack,"",%progbits
- #endif
|