Lines Matching refs:by

25 queues to distribute processing among CPUs. The NIC distributes packets by
28 queue, which in turn can be processed by separate CPUs. This mechanism is
39 by masking out the low order seven bits of the computed hash for the
58 The indirection table of an RSS device, which resolves a queue by masked
59 hash, is usually programmed by the driver at initialization. The
106 above the interrupt handler. This is accomplished by placing the packet
121 associated flow of the packet. The hash is either provided by hardware
141 by default for SMP). Even when compiled in, RPS remains disabled until
177 during CPU contention by dropping packets from large flows slightly
180 queue exceeds half the maximum queue length (as set by sysctl
182 count over the last 256 packets. If a flow exceeds a set ratio (by
192 Flow limit is compiled in by default (CONFIG_NET_FLOW_LIMIT), but not
194 and cache contention) and toggled per CPU by setting the relevant bit
200 Per-flow rate is calculated by hashing each packet into a hashtable
230 application locality. This is accomplished by Receive Flow Steering
231 (RFS). The goal of RFS is to increase datacache hitrate by steering
237 In RFS, packets are not forwarded directly by the value of their hash,
271 entry i is actually selected by hash and multiple flows may hash to the
297 by default for SMP). The functionality remains disabled until explicitly
344 is maintained by the NIC driver. This is an auto-generated reverse map of
345 the IRQ affinity table shown by /proc/interrupts. Drivers can use
353 CONFIG_RFS_ACCEL and support is provided by the NIC device and driver.
356 configured for each receive queue by the driver, so no additional
380 XPS is configured per transmit queue by setting a bitmap of CPUs that
387 queues match, one is selected by using the flow hash to compute an index
405 XPS is only available if the kconfig symbol CONFIG_XPS is enabled (on by
427 These are rate-limitation mechanisms implemented by HW, where currently
428 a max-rate attribute is supported, by setting a Mbps value to
437 2.6.38. Original patches were submitted by Tom Herbert
441 submitted by Ben Hutchings (bwh@kernel.org)