[Crash-utility] function arguments

[Jay Lan]

Dave Anderson wrote:
> Shahar Luxenburg wrote:
>> Hi,
>>
>>  
>>
>> I'm using crash utility on RHEL 5 on x86. According to the TODO list
>> 'bt' command should show function arguments: "Display arguments in the
>> stack trace. At present, we do not have support for PPC64 and x86_64".
>> Unfortunately, no function arguments are presented using 'bt' command.
>> Am I missing something?
>>
>>  
>>
>> If that's the case, I'd like to add that feature (for x86 architecture
>> at first).  Code from kdb can be ported to the crash utility for that
>> purpose. kdb is analyzing the code using basic blocks, understanding
>> where the registers are being saved (either to other registers or
>> memory location) and if their values can be retrieved, it prints them.

The basic blocks analysis was coded by Keith Owens last year in 2.6.23
time frame. It was a "best effort" try and would revert back to the
old style (ie, no arguments) if encounters problems.

Kernel has been changing, and the basic block analysis code needs fixes
when a new kernel breaks it from time to time. I fixed most of x86_32 bt
problem in 2.6.26-rc1 but it was broken again in 2.6.27-rc[n]. Worse
yet, 2.6.27 also broke x86_64 in my testing. I still need to fix the
new breakage.

I would not recommand porting the code to crash for that reason. The
code can be broken on every new kernel release.

Regards,
 - jay



>> Can I use debug information for that purpose? I can look at the
>> eh_frame section to understand where ebx, edi, esi and ebp are saved
>> but I can not find information, for each frame, what is the fate of
>> eax, edx and ecx – If I'd know that, then either they were saved in
>> memory (stack location) – it's easy to get them, or they were saved in
>> another register, than I can use the debug information to understand
>> if they were saved on the stack at some 'younger' frame.
>>
>>  
>>
>> Any hint will be welcomed.
>>
>>  
>>
>> Regards,
>>
>> Shahar.
>>
> 
> You can go back in the archives and find several more discussions regarding
> your question, but this is the most recent:
> 
>   https://www.redhat.com/archives/crash-utility/2008-July/msg00000.html
> 
> Basically the ia64 is the only architecture that can display function
> arguments, and that's because the capability of doing so is built
> into the kernel, and so I essentially ported the kernel ia64 unwind
> facility into the crash utility, and with that in place, accessing
> function arguments came along with it.
> 
> For all other arches, all bets are off.  And in fact, the backtraces
> themselves are pretty much educated guesses because the stack frame
> size determination is a best-guess when -fomit-frame-pointer is in
> place.  So until that functionality is absolutely rock-solid, getting
> function arguments is secondary.
> 
> Hence, the you-got-to-be-able-to-walk-before-you-run analogy...
> 
> Anyway, nothing's impossible, but nobody's come up with a working
> strategy yet.  And I welcome all attempts...
> 
> Dave
> 
> -- 
> Crash-utility mailing list
> Crash-utility at redhat.com
> https://www.redhat.com/mailman/listinfo/crash-utility