Vladimir Smirnov
Vladimir Smirnov
At least in DPDK 23.03 (and later) xstats_names might return more names than xstats_values and that is considered normal. See https://doc.dpdk.org/api-23.03/rte__ethdev_8h.html#a300d75b583c1f5acfe5b162a5d8c0ac1 In this case it seems appropriate to force names...
In DPDK 23.03 (and later), it is possible that Length of xstats_name doesn't match lenth of xstats_values: https://doc.dpdk.org/api-23.03/rte__ethdev_8h.html#a300d75b583c1f5acfe5b162a5d8c0ac1 ``` There is an assumption that 'xstat_names' and 'xstats' arrays are matched...
I understand that there is a consideration about smaller terminals. However, I don't see any reason why that shouldn't be configurable. E.x. running on 4k monitor displaying stats from more...
I've ran huffshell and get this suggestion: ``` bash for 30: 'for' => f I 30: 'for I' => fi in 30: 'for I in' => fii ``` this is...
I've got folowing output for huffshell: ``` bash Most common commands: ssh 284 executer 74 echo 64 61 ls 43 for 30 ./checker.sh 29 POST 27 cd 26 sudo 25...
As OSX is known to completely ignore all affinity hints (on normal devices) and then pin tasks to random cores, those making results utterly useless, what if implement a mode...
waitFinished was never actually used (initialized and then assigned from within a goroutine, but value was never checked). Not catching that was a bug in Golang that was fixed in...
* Update bundled DPDK to 24.03 I've tested that it works on my mellanox cards, but I don't have neither intel nor broadcoms so I can't actually verify that it...
clustering * Fix dpdk_setup_ports numa detection code - it now support sub-numa clustering * Remove artificial limitation on 63 max cores in dpdk_setup_ports * Change RTE_MAX_LCORE in build options for...
I have a new test bench setup that accidentally have 8 NUMA nodes. ``` # numactl -H available: 8 nodes (0-7) node 0 cpus: 0 1 2 3 4 5...