Vratko Polak

Expose m_tx_drop in stats

2

Hello. We (at FD.io CSIT project) are using TRex (currently v2.73) as a traffic generator for performance testing. Sometimes we see a strange TRex behavior, when a burst of traffic...

This become active after #4117 fix. See [0] for `show hardware verbose` output. That runs also contains GTPUsw test for comparison, which has no issue sending the GTPU-encapsulated jumbo packet...

Zero traffic in GTPUhw jumbo tests

2

Contrary to other jumbo issues, this is not specific to dev_iavf plugin. Also, GTPUsw tests are passing also with jumbo. Telemetry [0] shows symptoms, first fragmentation happens, then the fragments...

ip6 regressions end of November 2025

2

While results with AVF driver are also affected by #4109, looking at trending [0] this might be two different regressions, 2n-zn2 being affected few days sooner than others. 2n-grc is...

2n-zn2 iavf: two-band structure in 1c mrr tests

4

This is interesting, because 2c tests use 4 workers. Looking at trending [0], 1c tests have similar stdev but no band structure. Comparing good [1] and bad [2] telemetry I...

jumbo+iavf buffer alloc error: loadbalancer

1

Previously reported as one of symptoms of generic jumbo+avf ticket [0]. Still present [0] in rls2506. Tracking as a separate bug, because similar issues ([2] for vhost and [3] for...

Despite similar timing, this is different from #4111 by only affecting ip6 tests but across multiple platforms. Contrary to #4114 this regression is generally less than 5%. The best signal...

These are expected as ARM contributors optimize VPP build process for specific hardware. [Trending](https://csit.fd.io/trending/#eNrtlMsOgjAQRb8GN2YSqSArFz7-w9QyAkkpTVsQ_XoLIQ4uNMZE3Ljp6870Tk8mta4yeLAo10G8DZJtwJIi9UOw3Mz91GgNTEFmBLDFIkOmQ9EmDXIHpWxjKHR0cLVSKC2soiOEAtDl_hSGuU51hgobhNDHdfFHbhFKYzoTtutM0to9OJKi8wspL-qgDG6QU8qoPApxaEd-75RLuSfDS7TFFekC_2zShadJUigeXd1Fj9SBQbLvIyZDr9M_-oHBp-gl64lHPXG_a8WUTS3ZC6D3aj4FF32zZ5-Am6glfwmOOi7ez1Rlyv6_9WubV2dwpuDSDkc3JPuQgQ) shows both progressions and regressions all caused by this [0] change. [0] 41259: vppinfra: add ARM...

iavf: rare bus error when creating interface

2

In rls2502 we executed many AVF tests, this is after migrating CSIT to use plugins/dev_iavf instead of the previous plugins/avf. Most tests have no issues, but I have spotted one...

The symptom visible in console logs is `Failed to set flow control on {pf_eth}!`, but log.html [0] shows the true issue is a command like `basename /sys/bus/pci/devices/0000:01:00.1/net/*` returning an empty...