Lines Matching +full:average +full:- +full:on
19 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 .Nm pmc.cmn-600
29 .Nd Library for accessing the Arm CoreLink CMN-600 Coherent Mesh Network
36 CMN-600 PMU counters may be configured to count any one of a defined set of
38 Unlike other performance counters, counters for the CMN-600 require the
52 Some BIOS versions of dual-socket machines have no NUMA domain information in
57 the CPU package containing that CMN-600 controller.
64 Arm CoreLink CMN-600 Coherent Mesh Network Controller performance counters are
67 .%B "Arm CoreLink CMN-600 Coherent Mesh Network Technical Reference Manual"
73 CMN-600 PMU counters support the following capabilities:
74 .Bl -column "PMC_CAP_INTERRUPT" "Support"
91 .Bl -tag -width indent
111 .Bl -tag -width indent
114 This includes all the sub-types include TLB invalidate, Branch predictor
144 .Ss HN-F node events
145 .Bl -tag -width indent
173 Counts number of times a HighHigh priority request is protocol-retried at the
174 HN-F
176 Counts the POCQ occupancy in HN-F.
190 Counts number of times HN-F has a pending TXDAT flit but no credits to upload
192 Counts number of times HN-F has a pending TXRSP flit but no credits to upload
216 .Ss HN-I node events
217 .Bl -tag -width indent
253 Non-PCIe serialization event
258 .Bl -tag -width indent
260 Number of flits transmitted on a specified port and CHI channel.
262 Note: On device ports, this event also includes link flit transfers.
266 This is a measure of the flit traffic congestion on the mesh and at the flit
270 RN-F_CHIA port.
279 .Bl -tag -width indent
315 .Ss RN-D node events
316 .Bl -tag -width indent
318 Number of RData beats, RVALID and RREADY, dispatched on port 0.
321 Number of RData beats, RVALID and RREADY, dispatched on port 1.
324 Number of RData beats, RVALID and RREADY, dispatched on port 2.
352 Number of WData beats, WVALID and WREADY, dispatched on port 0.
353 This is a measure of write bandwidth on AXI port 0.
355 Number of WData beats, WVALID and WREADY, dispatched on port 1.
356 This is a measure of write bandwidth on AXI port 1.
358 Number of WData beats, WVALID and WREADY, dispatched on port 2.
359 This is a measure of write bandwidth on AXI port 2.
376 .Ss RN-I node events
377 .Bl -tag -width indent
379 Number of RData beats, RVALID and RREADY, dispatched on port 0.
382 Number of RData beats, RVALID and RREADY, dispatched on port 1.
385 Number of RData beats, RVALID and RREADY, dispatched on port 2.
412 Number of WData beats, WVALID and WREADY, dispatched on port 0.
413 This is a measure of write bandwidth on AXI port 0.
415 Number of WData beats, WVALID and WREADY, dispatched on port 1.
416 This is a measure of write bandwidth on AXI port 1.
418 Number of WData beats, WVALID and WREADY, dispatched on port 2.
419 This is a measure of write bandwidth on AXI port 2.
438 .Bl -tag -width indent
446 Number of Snoop Pcrd Stall on Link 0
448 Number of Snoop Pcrd Stall on Link 1
450 Number of Snoop Pcrd Stall on Link 2
467 .Bl -tag -width indent
483 Average size of request chains, only for chain sizes larger than one
506 .Bl -tag -width indent
532 Average RX TLP size in DWs
534 Average TX TLP size in DWs
536 Average RX TLP size in CCIX messages
538 Average TX TLP size in CCIX messages
540 Average size of RX CXS in DWs within a beat
542 Average size of TX CXS in DWs within a beat
550 Average latency to process an RX TLP
552 Average latency to form a TX TLP
577 .Nm pmc.cmn-600
581 .An -nosplit
588 The CMN-600 PMU driver was sponsored by ARM Ltd.