I currently allocate a 64K contiguous buffer using the DMA operations
(Hal)AllocateCommonBuffer function. However, I have a customer that wants a
64Meg buffer, which I think is unwise since that will literally devour all
available non-paged pool in the system. The device is FibreChannel, and the
buffer is for RDMA. I’m sure that the customer believes that using RDMA will
be faster than using the normal Scatter/Gather DMA transfers. Our testing
for large buffers has shown otherwise, but you know what happens when
customers get a brainstorm.
What is the impact on the system? Hell, is it even doable?
Gary G. Little
Staff Engineer
Broadband Storage, Inc.
xxxxx@broadstor.com
You are currently subscribed to ntdev as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntdev-$subst(‘Recip.MemberIDChar’)@lists.osr.com