mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* binary portability
@ 2006-06-08 10:01 Majumder, Rajib
  2006-06-08 10:42 ` Jan Engelhardt
  0 siblings, 1 reply; 5+ messages in thread
From: Majumder, Rajib @ 2006-06-08 10:01 UTC (permalink / raw)
  To: 'linux-kernel@vger.kernel.org'

Hello,

I have some binaries built on a RHEL 3/EM64T platform. Can I directly port them to an RHEL 3/AMD64 platform and run without any issue? 

I know that EM64T and AMD64 are ISA compatible, but there could be some differences in ELF32 between these 2 processor architectures. 

Any input will be appreciated. 

Thanks

Rajib  


==============================================================================
Please access the attached hyperlink for an important electronic communications disclaimer: 

http://www.credit-suisse.com/legal/en/disclaimer_email_ib.html
==============================================================================


^ permalink raw reply	[flat|nested] 5+ messages in thread
* RE: binary portability
@ 2006-06-14  8:34 Majumder, Rajib
  2006-06-14 10:11 ` Alan Cox
  0 siblings, 1 reply; 5+ messages in thread
From: Majumder, Rajib @ 2006-06-14  8:34 UTC (permalink / raw)
  To: 'Alan Cox', Jan Engelhardt
  Cc: Majumder, Rajib, 'linux-kernel@vger.kernel.org'

Hi,

Thanks for the clarifications. Just had 1 more question. Can I port binaries built on RHEL 3/Opteron(2.4.21) to SLES 9/Opteron (2.6.5-7.252) and run without any issues?  

Thanks

Rajib

-----Original Message-----
From: Alan Cox [mailto:alan@lxorguk.ukuu.org.uk]
Sent: 08 June 2006 22:51
To: Jan Engelhardt
Cc: Majumder, Rajib; 'linux-kernel@vger.kernel.org'
Subject: Re: binary portability


Ar Iau, 2006-06-08 am 12:42 +0200, ysgrifennodd Jan Engelhardt:
> >I know that EM64T and AMD64 are ISA compatible, but there could be some 
> >differences in ELF32 between these 2 processor architectures. 
> >
> What differences? (Apart from MMXEXT and SSE2,SSE3)

There are multiple ISA differences that affect ring 0 (kernel code), but
only one nasty that hit user code with early Intel clones. The early
Intel clones didn't implement the prefetch instructions that are
mandatory in x86-64. This broke a few apps early on but they got
workarounds very fast.

If the code is built for the generic instruction set (as is the case
unless you try very hard) then it should be perfect on both.

Alan


==============================================================================
Please access the attached hyperlink for an important electronic communications disclaimer: 

http://www.credit-suisse.com/legal/en/disclaimer_email_ib.html
==============================================================================


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2006-06-14  9:55 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-06-08 10:01 binary portability Majumder, Rajib
2006-06-08 10:42 ` Jan Engelhardt
2006-06-08 14:50   ` Alan Cox
2006-06-14  8:34 Majumder, Rajib
2006-06-14 10:11 ` Alan Cox

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome