Re: InfiniBand Implementations

In article <[EMAIL PROTECTED]>,
del cecchi <[EMAIL PROTECTED]> wrote:
>> My specific question is about the software designs, and the way that
>> the hardware interfaces intend that the software should use them.  In
>> particular, that dreaded question of user-space DMA, with its related
>> issues of whether the user can provide arbitrary locations for DMA
>> use, the scope of pinning of such locations, what support there is for
>> memory migration or swapping and so on.
>
>Has the Linux code for the existing (plug into PCI) HCA cards been made
>public?  Perhaps that would answer your question, or is there something
>about PCI that would cloud the issue?

It wouldn't really help, but not for that reason.  I am asking about
design policy, and all that would tell me is about what a particular
implementation does.  At worst, it might be positively misleading,
because it did it all the "non-strategic" way to bypass temporary
problems (remember the swing?)


Regards,
Nick Maclaren.