[parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[Randolph Chung]

(cc'ed to the list)

> Thanks for the pointers.  From the document, I was able to add prefetch support
> for PA-RISC 2.0 to ATLAS's header file. 

Cool! Be aware that, if atlas relies on being able to prefetch invalid
addresses and have that not fault, then it will only work on some PA 
processors (mostly pa20) and not all of them.

randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[John David Anglin]

> > Thanks for the pointers.  From the document, I was able to add prefetch support
> > for PA-RISC 2.0 to ATLAS's header file. 
> 
> Cool! Be aware that, if atlas relies on being able to prefetch invalid
> addresses and have that not fault, then it will only work on some PA 
> processors (mostly pa20) and not all of them.

Are you sure this doesn't apply to all pa20 processors?  See page 6-11
in PA-RISC 2.0 Architecture book.

Dave
-- 
J. David Anglin                                  dave.anglin@nrc-cnrc.gc.ca
National Research Council of Canada              (613) 990-0752 (FAX: 952-6602)
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[Randolph Chung]

> > Cool! Be aware that, if atlas relies on being able to prefetch invalid
> > addresses and have that not fault, then it will only work on some PA 
> > processors (mostly pa20) and not all of them.
> 
> Are you sure this doesn't apply to all pa20 processors?  See page 6-11
> in PA-RISC 2.0 Architecture book.

It does work on all pa20 processors, afaik. Let me rephrase:

"Be aware that, if atlas relies on being able to prefetch invalid
addresses and have that not fault, then it will only work on some PA 
processors but not all of them. All pa20 processors should work.
pa11 processors will not."

We ran into this while trying to get PA7300LC to do prefetching in the
kernel.

randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[R Clint Whaley]

Randolph,

>It does work on all pa20 processors, afaik. Let me rephrase:
>
>"Be aware that, if atlas relies on being able to prefetch invalid
>addresses and have that not fault, then it will only work on some PA 
>processors but not all of them. All pa20 processors should work.
>pa11 processors will not."
>
>We ran into this while trying to get PA7300LC to do prefetching in the
>kernel.

My understanding from the docs is that pa11 doesn't define any prefetch
instructions, which means these formulations are loads to an invalid
register, which, since they are loads and not prefetches, could have
side effects (eg., seg fault).  Therefore, you need to use them only
on pa20, even if they compile on pa11.  I take it this is what you are saying?

Thanks,
Clint
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] Re: atlas3 on hppa: unexpected reloc type

[Matthew Wilcox]

On Mon, Jun 14, 2004 at 04:53:46PM -0400, R Clint Whaley wrote:
> My understanding from the docs is that pa11 doesn't define any prefetch
> instructions, which means these formulations are loads to an invalid
> register, which, since they are loads and not prefetches, could have
> side effects (eg., seg fault).  Therefore, you need to use them only
> on pa20, even if they compile on pa11.  I take it this is what you are saying?

That's correct.

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux