Re: [RFC -tip] x86: introduce ENTRY(KPROBE)_X86 assembly helpers to catch unbalanced declaration

[Cyrill Gorcunov <gorcunov@gmail.com>]

[Cyrill Gorcunov - Sun, Nov 23, 2008 at 08:58:46PM +0300]
| [Sam Ravnborg - Sun, Nov 23, 2008 at 06:51:25PM +0100]
| | On Sun, Nov 23, 2008 at 07:57:11PM +0300, Cyrill Gorcunov wrote:
| | > It's usefull to catch unbalanced, missed or mixed declarations of ENTRY and
| | > KPROBES. These macros would help a bit (at least I hope so).
| | > 
| | > For example the following code would compile without problems
| | > 
| | >         ENTRY_X86(mcount)
| | >                 retq
| | >         END_X86(mcount)
| | > 
| | > But if you forget and mix the following form
| | > 
| | >         ENTRY_X86(mcount)
| | >                 retq
| | >         END(mcount)
| | > 
| | >         ENTRY_X86(ftrace_caller)
| | > 
| | > The assembler will issue the following message:
| | > Error: ENTRY_X86/KPROBE_X86 unbalanced,missed,mixed
| | > 
| | > Actually the checking is performed at every _X86 macro
| | > so maybe it's good idea to put ENTRY_KPROBE_FINAL_X86
| | > at the end of .S file to be sure you didn't miss anything.
| | 
| | Could we at least try this out in -next before we decide to make 
| | this X86 only?
| | I am aware that binutils can be a bit fragile but -next testing should
| | make a good check on this.
| | 
| | 	Sam
| | 
| 
| I don't have -next tree on my laptop, neither cross-compile tools but
| if someone could test it -- it would be great. But I used gas macros
| here -- i doubt other architectures has the same syntax. At least
| PDP-11 would beat us with ';' symbol :)
| 
| 		- Cyrill -

On the other hand, if this feature show 'good' behaviour on x86 we could
propagate it on other arch's. If we just turn it on by default -- lots of
errors will be.

		- Cyrill -