This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: TSX lock elision for glibc v10


On Thu, Jun 13, 2013 at 10:08:43PM +0200, Andi Kleen wrote:
> > It seems the intent of this patch is to work around the issue by never
> > using elision unless it's explicitly requested. However, if I'm not
> > mistaken, there's an environment variable to request elision by
> > default, and this would DANGEROUSLY break applications using
> 
> The "danger" being the enforced deadlock not happening ?

Yes.

> Sorry this whole thing seems like a red herring to me. Even Torvalds
> agreed that it is more a bug in the POSIX standard, than anything else.
> Obviously it is.

You're welcome to argue this with the Austin Group, but the way the
standard is worded, it was clearly intentional at the time. But it
would have to be a change in the next version I think, not now, and
normally glibc aims to keep compatibility with outdated versions of
the standards over much stupider issues than this one.

> Only requiring opt-in would kill a lot of the benefits from 
> putting this into glibc.  Disabling any nesting would also kill a lot of the
> benefit.

Adding a separate constant for NORMAL type mutexes would fix the issue
completely and could be done right now without any controversy.

> We could ask the POSIX comittee for a clarification, but that would
> just prevent any chance of getting the feature in 2.18 and get
> some real testing.

2.18 is frozen anyway. I think there's plenty of time for 2.19.

Rich


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]