On 19/03/2021 17:00, Robin Murphy wrote:
On 2021-03-19 13:25, John Garry wrote:
Add a function to allow the max size which we want to optimise DMA
mappings
for.
It seems neat in theory - particularly for packet-based interfaces that
might have a known fixed size of data unit that they're working on at
any given time - but aren't there going to be many cases where the
driver has no idea because it depends on whatever size(s) of request
userspace happens to throw at it? Even if it does know the absolute
maximum size of thing it could ever transfer, that could be
impractically large in areas like video/AI/etc., so it could still be
hard to make a reasonable decision.
So if you consider the SCSI stack, which is my interest, we know the max
segment size and we know the max number of segments per request, so we
should know the theoretical upper limit of the actual IOVA length we can
get.
Indeed, from my experiment on my SCSI host, max IOVA len is found to be
507904, which is PAGE_SIZE * 124 (that is max sg ents there).
Incidentally that means that we want RCACHE RANGE MAX of 8, not 6.
Being largely workload-dependent is why I still think this should be a
command-line or sysfs tuneable - we could set the default based on how
much total memory is available, but ultimately it's the end user who
knows what the workload is going to be and what they care about
optimising for.
If that hardware is only found in a server, then the extra memory cost
would be trivial, so setting to max is standard approach.
Another thought (which I'm almost reluctant to share) is that I would
*love* to try implementing a self-tuning strategy that can detect high
contention on particular allocation sizes and adjust the caches on the
fly, but I can easily imagine that having enough inherent overhead to
end up being an impractical (but fun) waste of time.
For now, I just want to recover the performance lost recently :)
Thanks,
John
