Messages in this thread | | | Date | Fri, 9 Apr 2021 14:03:46 +0200 | From | Peter Zijlstra <> | Subject | Re: static_branch/jump_label vs branch merging |
| |
On Fri, Apr 09, 2021 at 07:55:42AM -0400, David Malcolm wrote: > On Fri, 2021-04-09 at 13:12 +0200, Peter Zijlstra wrote: > > On Fri, Apr 09, 2021 at 11:57:22AM +0200, Ard Biesheuvel wrote: > > > On Thu, 8 Apr 2021 at 18:53, Peter Zijlstra <peterz@infradead.org> > > > wrote: > > > > > > Is there *any* way in which we can have the compiler recognise > > > > that the > > > > asm_goto only depends on its arguments and have it merge the > > > > branches > > > > itself? > > > > > > > > I do realize that asm-goto being volatile this is a fairly huge > > > > ask, but > > > > I figured I should at least raise the issue, if only to raise > > > > awareness. > > > > > > > > > > Wouldn't that require the compiler to interpret the contents of the > > > asm() block? > > > > Yeah, this is more or less asking for ponies :-) One option would be > > some annotation that conveys the desired semantics without it having > > to > > untangle the mess in the asm block. > > > > The thing the compiler needs to know is that the branch is constant > > for > > any @key, and hence allow the obvious optimizations. I'm not sure if > > this is something compiler folks would be even willing to consider, > > but > > I figured asking never hurts. > > > > Sorry if this is a dumb question, but does the function attribute: > __attribute__ ((pure)) > help here? It's meant to allow multiple calls to a predicate to be > merged - though I'd be nervous of using it here, the predicate isn't > 100% pure, since AIUI the whole point of what you've built is for > predicates that very rarely change - but can change occasionally.
I actually tried that, but it doesn't seem to work. Given the function arguments are all compile time constants it should DTRT AFAICT, but alas.
| |