Messages in this thread | | | Date | Sat, 3 Apr 2010 17:08:29 +0100 | From | Russell King - ARM Linux <> | Subject | Re: [PATCH 4/5] kgdb: Use atomic operators which use barriers |
| |
On Fri, Apr 02, 2010 at 04:24:57PM -0700, Linus Torvalds wrote: > > > On Fri, 2 Apr 2010, Russell King - ARM Linux wrote: > > > > Actually, in future threads you end up agreeing with my position... > > I always agreed that it was not a memory barrier. > > In fact, the commit that extended on the "volatile-considered-harmful" > patch from you has this quote from me in the commit logs: > > Linus sayeth: > > : I don't think it was ever the intention that it would be seen as anything > : but a compiler barrier, although it is obviously implied that it might > : well perform some per-architecture actions that have "memory barrier-like" > : semantics. > : > : After all, the whole and only point of the "cpu_relax()" thing is to tell > : the CPU that we're busy-looping on some event. > : > : And that "event" might be (and often is) about reading the same memory > : location over and over until it changes to what we want it to be. So it's > : quite possible that on various architectures the "cpu_relax()" could be > : about making sure that such a tight loop on loads doesn't starve cache > : transactions, for example - and as such look a bit like a memory barrier > : from a CPU standpoint. > : > : But it's not meant to have any kind of architectural memory ordering > : semantics as far as the kernel is concerned - those must come from other > : sources. > > which I think is pretty clear. > > But that quote seems to be the one where you then think I "agree" with > you.
Yet again you read something into what I say that wasn't there.
Wait for me to return from holiday, as I said, and I'll respond further.
| |