Re: FInal kprobes rollup patches

[Harvey Harrison <harvey.harrison@gmail.com>]

On Sat, 2007-12-15 at 09:50 +0100, Ingo Molnar wrote:
> * Harvey Harrison <harvey.harrison@gmail.com> wrote:
> 
> > Further unification work.  There is a possible behavior change on 
> > X86_32 here.
> > 
> > is_IF_modifier(p->opcode)
> > 
> > to
> > 
> > is_IF_modifier(p->ainsn.insn)
> > 
> > Which should be equivalent, but is not purely cosmetic as the rest of 
> > the unification so far.
> 
> hm, could you split this into two, the pure-equivalence and the 
> possibly-modifying patch? (that way any potential breakage becomes 
> bisectable) Same end result, just two commits instead of one.
> 
Sure, I'll go back through and see if the series can be cleaned up a bit
as well as expand the commit message a little bit.

> also, did you manage to run/test kprobes (on 32-bit or 64-bit x86), and 
> it worked fine?
> 

Sorry, I should have predicated the whole series with RFC.  Currently
this is compile-tested only.  There is only the one patch that has
any behavioral change.  I believe the series also pointed out an
existing bug in the 32-bit version...which I've preserved but will
note in the commit messages in the respun series.

In case you're interested, from the patch which unifies the definition
of MAX_INSTRUCTION_SIZE:

memcpy(p->ainsn.insn, p->addr, (MAX_INSN_SIZE + 1) *
sizeof(kprobe_opcode_t));

If you compare this memcpy from arch_prepare_kprobe in 32/64 bit I'm
almost sure the X86_32 version should be

... + sizeof(kprobe_opcode_t)

not

... * sizeof(kprobe_opcode_t)

Cheers,

Harvey