Hi Daniel,
Great. I already posted a patch to openssl-dev, it's at;
http://marc.info/?l=openssl-dev&m=139923556631116&w=2
If you want to apply it ("git apply" or "git am") directly, the raw email
patch can be downloaded using;
http://marc.info/?l=openssl-dev&m=139923556631116&q=raw
Thanks in advance for any feedback on this.
Cheers,
Geoff
On Mon, May 5, 2014 at 1:07 PM, Daniel Sands <[email protected]> wrote:
> I'm using the RHEL6 standard distro. I can certainly test a patch for
> you.
>
> On Sat, 2014-05-03 at 00:01 -0400, Geoffrey Thorpe wrote:
> > I hadn't noticed this serialisation before, thanks. I'll try to send a
> > patch over the weekend some time, in case you're able to test? If so,
> > what version (or git branch) are you using?
> >
> >
> > Cheers,
> > Geoff
> >
> >
> >
> >
> > On Fri, May 2, 2014 at 6:54 PM, Daniel Sands <[email protected]>
> > wrote:
> > Examining performance bottlenecks on a busy server, we
> > discovered that
> > connections are being forced to serialize due to calls to
> > DH_generate_key. I looked through the source, and if I
> > understand it
> > correctly, the code locks a common DH mutex when it uses
> > Montgomery
> > Multiplication, and due to the way it sets a flag, it always
> > does.
> >
> > But I have not yet found any reason why this requires critical
> > section
> > protection. I do not see any global variables that are being
> > manipulated by the call, nor any resource contention. Is it
> > possible
> > that the mutex lock is a holdover from earlier code? Could
> > this locking
> > behavior be removed?
> >
> >
> >
> ______________________________________________________________________
> > OpenSSL Project
> > http://www.openssl.org
> > Development Mailing List
> > [email protected]
> > Automated List Manager
> > [email protected]
> >
> >
> >
>
>
> ______________________________________________________________________
> OpenSSL Project http://www.openssl.org
> Development Mailing List [email protected]
> Automated List Manager [email protected]
>
>
