Lines Matching +full:cpu +full:- +full:read

1 .\"-
31 .Nd "SMP-friendly kernel counter implementation"
82 can be used in any non-interrupt context.
90 is considered suitable for accounting in the performance-critical
92 .Bl -tag -width indent
94 Allocate a new 64-bit unsigned counter.
150 The function is a multiprocessor-friendly version of
155 Returns non-negative value if the rate is not yet reached during the current
190 A read of the oid returns value obtained through
202 A read of the oid returns value obtained through
217 A read of the oid returns len-sized array of
234 A read of the oid returns len-sized array of
243 is implemented using per-CPU data fields that are specially aligned
244 in memory, to avoid inter-CPU bus traffic due to shared use
250 The update operation only touches the field that is private to current CPU.
251 Fetch operation loops through all per-CPU fields and obtains a snapshot
257 operating on the private data for the current CPU,
262 The multi-instruction sequence provides the same guarantees as the
263 amd64 single-instruction implementation.
271 .Bd -literal -offset indent
292 .An -nosplit