[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] Re: [PATCH RFC V2 3/5] jump_label: if a key has already been initialized, don't nop it out



On Thu, Oct 06, 2011 at 11:29:25AM -0700, H. Peter Anvin wrote:
> On 10/06/2011 11:26 AM, Steven Rostedt wrote:
> > On Thu, 2011-10-06 at 14:10 -0400, Jason Baron wrote:
> > 
> >>> Looks like jmp2 is about 5% faster than jmp5 on Sandybridge with this
> >>> benchmark.
> >>>
> >>> But insignificant difference on Nehalem.
> >>>
> >>>     J
> >>
> >> It would be cool if we could make the total width 2-bytes, when
> >> possible.  It might be possible by making the initial 
> >> 'JUMP_LABEL_INITIAL_NOP'
> >> as a 'jmp' to the 'l_yes' label. And then patching that with a no-op at 
> >> boot
> >> time or link time - letting the compiler pick the width. In that way we 
> >> could
> >> get the optimal width...
> > 
> > Why not just do it?
> > 
> > jump_label is encapsulated in arch_static_branch() which on x86 looks
> > like:
> > 
> > static __always_inline bool arch_static_branch(struct jump_label_key *key)
> > {
> >     asm goto("1:"
> >             JUMP_LABEL_INITIAL_NOP
> >             ".pushsection __jump_table,  \"aw\" \n\t"
> >             _ASM_ALIGN "\n\t"
> >             _ASM_PTR "1b, %l[l_yes], %c0 \n\t"
> >             ".popsection \n\t"
> >             : :  "i" (key) : : l_yes);
> >     return false;
> > l_yes:
> >     return true;
> > }
> > 
> > 
> > That jmp to l_yes should easily be a two byte jump.

remember the compiler is moving the l_yes out of line, so its not
necessarily always a two byte jump. Also, I plan to look at a possible
'cold' label for the 'l_yes' branch, so that it can moved to a separate
'cold' section, but we might only want that for some cases...

> > 
> > If not I'm sure it would be easy to catch it before modifying the code.
> > And then complain real loudly about it.
> > 
> 
> The important thing is that it requires the build-time elimination of
> jumps.  It's just work.
> 
>       -hpa
> 

Right, its certainly doable, but I'm not sure its so simple, since we'll
need a pass to eliminate the jumps - which can be keyed off the
'__jump_table' section. 

thanks,

-Jason

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.