File: clrjmp1.ms

package info (click to toggle)
gdb-doc 13.1-1
  • links: PTS, VCS
  • area: non-free
  • in suites: bookworm
  • size: 234,284 kB
  • sloc: ansic: 1,988,072; asm: 373,465; exp: 187,579; cpp: 75,697; makefile: 69,432; sh: 24,829; yacc: 11,654; python: 9,602; ada: 6,680; xml: 6,073; perl: 5,077; pascal: 3,357; f90: 2,555; tcl: 1,902; lisp: 1,578; cs: 879; lex: 759; sed: 228; awk: 154; objc: 137; fortran: 57
file content (36 lines) | stat: -rw-r--r-- 578 bytes parent folder | download | duplicates (37) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
 
# mach: crisv3 crisv8 crisv10 crisv32
# output: ffffff00\n

; A bug resulting in a non-effectual clear.b discovered running the GCC
; testsuite; jump actually wrote to p0.

 .include "testutils.inc"

 start
 jump 1f
 nop
 .p2align 8
1:
 move.d y,r4

 .if 0 == ..asm.arch.cris.v32
; There was a bug causing this insn to set special register p0
; (byte-clear) to 8 (low 8 bits of location after insn).
 jump [r4+]
 .endif

1:
 move.d 0f,r4

; The corresponding bug would cause this insn too, to set p0.
 jump r4
 nop
 quit
0:
 moveq -1,r3
 clear.b r3
 dumpr3
 quit

y:
 .dword 1b