Re: [PATCH] supporting the x86_64 architecture

[Andy Polyakov]

> appro> Well, *if* this is supposed to be the last beta, then the only question
> appro> is if we *dare* to merge the code directly into the final version, i.e.
> appro> without exposing it in beta. I consider that we can dare to do so as
> appro> long as SuSE Labs promise that they would double-check couple of
> appro> snapshots upon explicit request. Peter?
> 
> Well, to be perfectly honest, we can't say in advance if a beta is
> going to be final or not.  We can only say that we hope it is or
> expect it to be.  For that reason, I'd prefer things like that to go
> in *now*, so they have a chance for exposure.
> 
> The other option is to have it wait until 0.9.7a...
> 
> Your call.

If it's up to me, then I already answered. To roll out x86_64
accelerated support is trivial. But I think it's more important to
expose DES PIC code so I'd wait till it's committed. At least till the
weekend or so... A.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Richard Levitte - VMS Whacker]

In message <[EMAIL PROTECTED]> on Thu, 12 Dec 2002 15:00:41 +0100, Andy 
Polyakov <[EMAIL PROTECTED]> said:

appro> Well, *if* this is supposed to be the last beta, then the only question
appro> is if we *dare* to merge the code directly into the final version, i.e.
appro> without exposing it in beta. I consider that we can dare to do so as
appro> long as SuSE Labs promise that they would double-check couple of
appro> snapshots upon explicit request. Peter?

Well, to be perfectly honest, we can't say in advance if a beta is
going to be final or not.  We can only say that we hope it is or
expect it to be.  For that reason, I'd prefer things like that to go
in *now*, so they have a chance for exposure.

The other option is to have it wait until 0.9.7a...

Your call.

-- 
Richard Levitte   \ Spannvägen 38, II \ [EMAIL PROTECTED]
Redakteur@Stacken  \ S-168 35  BROMMA  \ T: +46-8-26 52 47
\  SWEDEN   \ or +46-708-26 53 44
Procurator Odiosus Ex Infernis-- [EMAIL PROTECTED]
Member of the OpenSSL development team: http://www.openssl.org/

Unsolicited commercial email is subject to an archival fee of $400.
See  for more info.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Tim Rice]

On Thu, 12 Dec 2002, Richard Levitte - VMS Whacker wrote:

> In message <[EMAIL PROTECTED]> on Wed, 11 Dec 2002 23:03:03 +0100, 
>Andy Polyakov <[EMAIL PROTECTED]> said:
> 
> appro> > I have been working on
> appro> > BN assembler aided implementation that would need some benchmarking. It
> appro> > should give around 3x speed-up...
> appro> 
> appro> Preliminary patch relative to 0.9.6h is available at
> appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
> appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.
> 
> Hmm.  Supposedly, we're producing a final beta today.  Shall I
> postpone?

Will my Makefile.ssl patch to support assembler on SCO OpenServer and
UnixWare make it into the final beta?

> 
> 

-- 
Tim RiceMultitalents(707) 887-1469
[EMAIL PROTECTED]


__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Peter Poeml]

On Thu, Dec 12, 2002 at 03:00:41PM +0100, Andy Polyakov wrote:
> > appro> > I have been working on
> > appro> > BN assembler aided implementation that would need some benchmarking. It
> > appro> > should give around 3x speed-up...
> > appro>
> > appro> Preliminary patch relative to 0.9.6h is available at
> > appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
> > appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.
> > 
> > Hmm.  Supposedly, we're producing a final beta today.  Shall I
> > postpone?
> 
> Well, *if* this is supposed to be the last beta, then the only question
> is if we *dare* to merge the code directly into the final version, i.e.
> without exposing it in beta. I consider that we can dare to do so as
> long as SuSE Labs promise that they would double-check couple of
> snapshots upon explicit request. Peter?

I'll do what I can to assist you.

Peter

-- 
Thought is limitation. Free your mind.



msg14330/pgp0.pgp
Description: PGP signature

Re: [PATCH] supporting the x86_64 architecture

[Andy Polyakov]

> appro> > I have been working on
> appro> > BN assembler aided implementation that would need some benchmarking. It
> appro> > should give around 3x speed-up...
> appro>
> appro> Preliminary patch relative to 0.9.6h is available at
> appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
> appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.
> 
> Hmm.  Supposedly, we're producing a final beta today.  Shall I
> postpone?

Well, *if* this is supposed to be the last beta, then the only question
is if we *dare* to merge the code directly into the final version, i.e.
without exposing it in beta. I consider that we can dare to do so as
long as SuSE Labs promise that they would double-check couple of
snapshots upon explicit request. Peter?

However! I also have this IA-32 DES PIC-ified code coming in. The code
already passes the test and I'm now striving for getting it into those
perlasm things. As the idea is to roll that code into *all* IA-32 [Unix]
platforms (as already mentioned it's *not* Solaris specific problem,
it's just that Solaris is the only one that brings it up) it's
definitely good idea to expose it in beta. But it seems that I'd need
some help there... More about it in separate mail...

So that *my* answer to the question is "yes, please, but it has little
to do with x86_64." But that's only *my* opinion:-)

A.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Richard Levitte - VMS Whacker]

In message <[EMAIL PROTECTED]> on Thu, 12 Dec 2002 12:10:38 +0100, Peter 
Poeml <[EMAIL PROTECTED]> said:

poeml> > appro> Preliminary patch relative to 0.9.6h is available at
poeml> > appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
poeml> > appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.
poeml> > 
poeml> > Hmm.  Supposedly, we're producing a final beta today.  Shall I
poeml> > postpone?
poeml> 
poeml> I have just tested Andy's patch for 0.9.6h, and can confirm that it
poeml> works fine.

Thank you!

-- 
Richard Levitte   \ Spannvägen 38, II \ [EMAIL PROTECTED]
Redakteur@Stacken  \ S-168 35  BROMMA  \ T: +46-8-26 52 47
\  SWEDEN   \ or +46-708-26 53 44
Procurator Odiosus Ex Infernis-- [EMAIL PROTECTED]
Member of the OpenSSL development team: http://www.openssl.org/

Unsolicited commercial email is subject to an archival fee of $400.
See  for more info.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Peter Poeml]

On Thu, Dec 12, 2002 at 11:44:04AM +0100, Richard Levitte - VMS Whacker wrote:
> In message <[EMAIL PROTECTED]> on Wed, 11 Dec 2002 23:03:03 +0100, 
>Andy Polyakov <[EMAIL PROTECTED]> said:
> 
> appro> > I have been working on
> appro> > BN assembler aided implementation that would need some benchmarking. It
> appro> > should give around 3x speed-up...
> appro> 
> appro> Preliminary patch relative to 0.9.6h is available at
> appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
> appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.
> 
> Hmm.  Supposedly, we're producing a final beta today.  Shall I
> postpone?

I have just tested Andy's patch for 0.9.6h, and can confirm that it
works fine.

Peter

-- 
Thought is limitation. Free your mind.



msg14327/pgp0.pgp
Description: PGP signature

Re: [PATCH] supporting the x86_64 architecture

[Richard Levitte - VMS Whacker]

In message <[EMAIL PROTECTED]> on Wed, 11 Dec 2002 23:03:03 +0100, Andy 
Polyakov <[EMAIL PROTECTED]> said:

appro> > I have been working on
appro> > BN assembler aided implementation that would need some benchmarking. It
appro> > should give around 3x speed-up...
appro> 
appro> Preliminary patch relative to 0.9.6h is available at
appro> http://www.openssl.org/~appro/. Once it's confirmed to be working on
appro> real hardware, it will be ported/merged to 0.9.7 and HEAD branches.

Hmm.  Supposedly, we're producing a final beta today.  Shall I
postpone?

-- 
Richard Levitte   \ Spannvägen 38, II \ [EMAIL PROTECTED]
Redakteur@Stacken  \ S-168 35  BROMMA  \ T: +46-8-26 52 47
\  SWEDEN   \ or +46-708-26 53 44
Procurator Odiosus Ex Infernis-- [EMAIL PROTECTED]
Member of the OpenSSL development team: http://www.openssl.org/

Unsolicited commercial email is subject to an archival fee of $400.
See  for more info.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Andy Polyakov]

> I have been working on
> BN assembler aided implementation that would need some benchmarking. It
> should give around 3x speed-up...

Preliminary patch relative to 0.9.6h is available at
http://www.openssl.org/~appro/. Once it's confirmed to be working on
real hardware, it will be ported/merged to 0.9.7 and HEAD branches.

Cheers. A.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]

Re: [PATCH] supporting the x86_64 architecture

[Peter Poeml]

On Wed, Nov 27, 2002 at 12:53:31AM +0100, Andy Polyakov wrote:
> > to facilitate building openssl on the x86_64 platform I suggest to apply
> > the attached patch.
> 
> > +"linux-x86_64", "gcc:-DL_ENDIAN -DNO_ASM:
> 
> Linux/x86_64 suports two ABIs. As far as I understand it's perfectly
> possible to compile gcc so that it supports both. Which one will be
> default? To ensure that intended ABI is chosen I'd recommend to add -m64
> to command line.

The 64-bit environment is the default. The compiler seems to know... we
don't use -m64 explicitely.

> Why don't you use -O? I had no problem with -O3 and gcc-3.2...

We use -O2 but I hand it over on the command line, because compilers are
unsteady guys I have heard bad things about -O3, like making binaries
slower and larger. -O2 is what has been tested most, it seems.

> As for NO_ASM. Do you have real x86_64 hardware? I have been working on
> BN assembler aided implementation that would need some benchmarking. It
> should give around 3x speed-up... If you have real hardware is it
> possible to get an account?

I'll see how I can get you in contact with someone,

Peter

-- 
Gravity is an unforgiving motherfucker.



msg14175/pgp0.pgp
Description: PGP signature

Re: [PATCH] supporting the x86_64 architecture

[Andy Polyakov]

> to facilitate building openssl on the x86_64 platform I suggest to apply
> the attached patch.

> +"linux-x86_64", "gcc:-DL_ENDIAN -DNO_ASM:

Linux/x86_64 suports two ABIs. As far as I understand it's perfectly
possible to compile gcc so that it supports both. Which one will be
default? To ensure that intended ABI is chosen I'd recommend to add -m64
to command line.

Why don't you use -O? I had no problem with -O3 and gcc-3.2...

As for NO_ASM. Do you have real x86_64 hardware? I have been working on
BN assembler aided implementation that would need some benchmarking. It
should give around 3x speed-up... If you have real hardware is it
possible to get an account?

A.
__
OpenSSL Project http://www.openssl.org
Development Mailing List   [EMAIL PROTECTED]
Automated List Manager   [EMAIL PROTECTED]