Re: [PATCH v4 01/23] perf capstone: Fix arm64 jump/adrp disassembly mismatch with objdump

[Shuai Xue <xueshuai@linux.alibaba.com>]

On 8/11/26 10:27 AM, Tengda Wu wrote:
> Hi Shuai, thank you for your time.
> 
> On 2026/8/10 21:08, Shuai Xue wrote:
>>
>>
>> On 8/8/26 8:23 PM, Tengda Wu wrote:
>>> The jump and adrp instructions parsed by libcapstone currently lack
>>> symbolic representation and use a '#' prefix for addresses. This
>>> format is inconsistent with objdump's output, which causes subsequent
>>> parsing in jump__parse() and arm64_mov__parse() to fail.
>>>
>>> Example mismatch:
>>>     Current: b #0xffff8000800114c8
>>>     Fix:     b ffff8000800114c8 <el0t_64_sync+0x108>
>>>
>>>     Current: adrp x18, #0xffff800081f5f000
>>>     Fix:     adrp x18, ffff800081f5f000 <this_cpu_vector>
>>>
>>> Fix this by implementing extended formatting for these arm64
>>> instructions during symbol__disassemble_capstone(). This ensures
>>> the output matches objdump's expected style, including the raw
>>> address and the associated <symbol+offset> suffix.
>>>
>>> Signed-off-by: Tengda Wu <wutengda@huaweicloud.com>
>>> ---
>>>    tools/perf/util/capstone.c | 136 +++++++++++++++++++++++++++++++++----
>>>    tools/perf/util/disasm.c   |   5 ++
>>>    tools/perf/util/disasm.h   |   1 +
>>>    3 files changed, 130 insertions(+), 12 deletions(-)
>>>
>>> diff --git a/tools/perf/util/capstone.c b/tools/perf/util/capstone.c
>>> index 74213daf8786..fb8a2bc5558f 100644
>>> --- a/tools/perf/util/capstone.c
>>> +++ b/tools/perf/util/capstone.c
>>> @@ -3,6 +3,7 @@
>>>      #include <errno.h>
>>>    #include <inttypes.h>
>>> +#include <stdlib.h>
>>>    #include <string.h>
>>>      #include <dlfcn.h>
>>> @@ -31,6 +32,10 @@
>>>    #define CS_MODE_RISCVC   4
>>>    #endif
>>>    +#if CS_VERSION_MAJOR < 4
>>> +#define ARM64_GRP_BRANCH_RELATIVE  7
>>
>> Please add a comment explaining where '7' comes from
>> (CS_GRP_BRANCH_RELATIVE in capstone v3), otherwise it reads like an
>> arbitrary magic number.
>>
> 
> Um, this was done following Namhyung's approach. That said, adding a
> comment would certainly make this definition clearer -- will add it.
> 
> Also, I couldn't find CS_GRP_BRANCH_RELATIVE in v3. From what I can see,
> it was originally introduced in v4, together with ARM64_GRP_BRANCH_RELATIVE
> (see https://github.com/capstone-engine/capstone/commit/a09a81813c83).
> So the comment might look something like this:
> 
>    #define ARM64_GRP_BRANCH_RELATIVE  7  /* = CS_GRP_BRANCH_RELATIVE */
> 
> Is this acceptable?
> 

LGTM.

Thanks.
Shuai