Re: XDP socket rings, and LKMM litmus tests

From: maranget <luc.maranget@inria.fr>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: "Paul E. McKenney" <paulmck@kernel.org>,
	"Björn Töpel" <bjorn.topel@gmail.com>, bpf <bpf@vger.kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	"Andrea Parri" <parri.andrea@gmail.com>,
	"Will Deacon" <will@kernel.org>,
	"Peter Zijlstra" <peterz@infradead.org>,
	"Boqun Feng" <boqun.feng@gmail.com>,
	"Nicholas Piggin" <npiggin@gmail.com>,
	"David Howells" <dhowells@redhat.com>,
	"Alglave, Jade" <j.alglave@ucl.ac.uk>,
	"Akira Yokosawa" <akiyks@gmail.com>,
	"Daniel Lustig" <dlustig@nvidia.com>,
	joel@joelfernandes.org,
	"Toke Høiland-Jørgensen" <toke@redhat.com>,
	"Karlsson, Magnus" <magnus.karlsson@intel.com>
Subject: Re: XDP socket rings, and LKMM litmus tests
Date: Thu, 4 Mar 2021 16:44:34 +0100	[thread overview]
Message-ID: <EF3F87BF-2AA1-4F96-A2A0-EA8A9D6FC8F7@inria.fr> (raw)
In-Reply-To: <20210303202246.GC1582185@rowland.harvard.edu>

> On 3 Mar 2021, at 21:22, Alan Stern <stern@rowland.harvard.edu> wrote:
> 
>>> 
>>> Local variables absolutely should be treated just like CPU registers, if 
>>> possible.  In fact, the compiler has the option of keeping local 
>>> variables stored in registers.
>>> 
>>> (Of course, things may get complicated if anyone writes a litmus test 
>>> that uses a pointer to a local variable,  Especially if the pointer 
>>> could hold the address of a local variable in one execution and a 
>>> shared variable in another!  Or if the pointer is itself a shared 
>>> variable and is dereferenced in another thread!)
>> 
>> Good point!  I did miss this complication.  ;-)
> 
> I suspect it wouldn't be so bad if herd7 disallowed taking addresses of 
> local variables.
> 
> 

Herd7 does disallow taking addresses of local variables.

However, such  tests can still be run on machine, provided function bodies are accepted by the C compiler.

—Luc