[openib-general] FW: [PATCH 1 of 3] mad: large RMPP support

[Roland Dreier]

    Jack> I'm concerned that an allocation of a 4K buffer may fail in
    Jack> a situation where lots of small allocations of around 256
    Jack> bytes would succeed.  Is your point that if we fail to
    Jack> allocate a 4K buffer, we're in deep trouble already?  Note
    Jack> that I've only considered a 1000 host cluster.

    Sean> Yes - if we can't allocate a 4k buffer, it seems highly
    Sean> unlikely that we'd be able to allocate 1000 256-byte
    Sean> buffers.

My rule of thumb is that we shouldn't rely on being able to allocate a
contiguous buffer bigger than 4 KB, but assuming we can allocate 4 KB
is fine.  4 KB is the lowest page size of any real architecture, and
if the kernel is out of free pages then any allocation is likely to
fail.  Allocations of larger buffers may fail because of memory
fragmentation, even with plenty of free memory.

That is: a 4 KB buffer is fine.

 - R.