PrintAssembly output - Please help
Ulf Zibis
Ulf.Zibis at gmx.de
Wed Dec 9 22:30:14 PST 2009
Ok thanks, but I'm wondering, that from PrintOptoAssembly I get other
but valid JMP targets (all B114 here) to the end of the code, not to itself:
....
6c6 B110: # B114 <- B6 Freq: 9.99938e-006
6c6 # exception oop is in EAX; no code emitted
6c6 MOV ECX,EAX
6c8 JMP,s B114
6c8
6ca B111: # B114 <- B4 Freq: 9.99959e-006
6ca # exception oop is in EAX; no code emitted
6ca MOV ECX,EAX
6cc JMP,s B114
6cc
6ce B112: # B114 <- B3 Freq: 9.99979e-006
6ce # exception oop is in EAX; no code emitted
6ce MOV ECX,EAX
6d0 JMP,s B114
6d0
6d2 B113: # B114 <- B2 Freq: 9.99999e-006
6d2 # exception oop is in EAX; no code emitted
6d2 MOV ECX,EAX
6d2
6d4 B114: # N1111 <- B79 B113 B112 B111 B110 B109 B108 B103
B102 B101 B100 B93 B92 B91 B90 B87 B86 B85 B84 B83 B82 B81 B80 B107 B106
B105 B104 B78 B77 B76 B75 B99 Freq: 7.11172e-005
6d4 ADD ESP,72 # Destroy frame
POPL EBP
6d8 JMP rethrow_stub
6d8
-Ulf
Am 09.12.2009 23:05, Tom Rodriguez schrieb:
> static stubs are used as trampolines when calling from compiled code into the interpreter. That's just the initial state. You can simply ignore them.
>
> tom
>
> On Dec 9, 2009, at 1:55 PM, Ulf Zibis wrote:
>
>
>> From PrintAssembly I see some code like this:
>>
>> ....
>> 0x00b9bb0c: mov $0x0,%ebx ;...bb000000 00
>> ; {static_stub}
>> 0x00b9bb11: jmp 0x00b9bb11 ;...e9fbffff ff
>> ; {runtime_call}
>> 0x00b9bb16: mov $0x0,%ebx ;...bb000000 00
>> ; {static_stub}
>> 0x00b9bb1b: jmp 0x00b9bb1b ;...e9fbffff ff
>> ; {runtime_call}
>> 0x00b9bb20: mov $0x0,%ebx ;...bb000000 00
>> ; {static_stub}
>> 0x00b9bb25: jmp 0x00b9bb25 ;...e9fbffff ff
>> ; {runtime_call}
>> [Exception Handler]
>> 0x00b9bb2a: jmp 0x00b90f00 ;...e9d153ff ff
>> ; {runtime_call}
>> 0x00b9bb2f: push $0xb9bb2f ;...682fbbb9 00
>> ; {section_word}
>> 0x00b9bb34: jmp 0x00b6bb40 ;...e90700fd ff
>>
>>
>> What does that mean:
>> 0x00b9bb11: jmp 0x00b9bb11
>>
>> Jump to itself ?
>> Please explain me!
>>
>> Thanks,
>>
>> -Ulf
>>
>>
>>
>
>
>
More information about the hotspot-compiler-dev
mailing list