Update few days before RC2. > 1. QAT no traffic after DPDK bump [0]. > I have put my findings so far into a quick ticket [1].
As mentioned in the ticket, I have a fix [12]. Other contributions related to dpdk/cryptodev do not seem to address any issue visible in CSIT, so I will test them later. > 2. Before that, OpenSSL bump [2] had introduced an IPsec HW regression on > 3n-srf. No clear fix, to be addressed in the next release cycle. > 3. New packed/natural buffer layout [4], on Xeon with huge pages it causes a > regression. I decided to change CSIT code [13] to explicitly set the “natural” layout, as that does not seem to have any negative effect (26.06 ignores unknown startup.conf options). > 4. Octeon driver got some merges, but now VPP usually [7] does not start [8] > in CSIT. As mentioned in e-mails below, the first issue has been fixed. As not mentioned yet, two more issues happened since then. The symbol visibility verification seems to depend on VPP platform, and octeon was apparently not tested. Here [14] is a minimal fix. Maintainers of similarly “enabled in verify jobs” plugins/drivers should check their symbol requirements. There seem to also be a cabling issue on 3n-oct testbed, not related to VPP. I am still working on two more CSIT issues, I will write an e-mail if they need some support on VPP side, or if they will affect release results. Assuming [12] and [14] will get merged to master and cherry-picked to stable/2610 before RC2, (and barring my possible future e-mails) I say VPP is ready (enough) for RC2 from CSIT point of view. Vratko. [12] 46976: build: tell dpdk where ipsec-mb is | https://gerrit.fd.io/r/c/vpp/+/46976 [13] 46985: feat(config): Apply natural layout on all testbeds | https://gerrit.fd.io/r/c/csit/+/46985 [14] 46998: vnet: export vnet TM symbols for octeon driver | https://gerrit.fd.io/r/c/vpp/+/46998 From: Vratko Polák -X (vrpolak - PANTHEON TECH SRO at Cisco) Sent: Wednesday, 23 September, 2026 10:26 To: [email protected] Cc: csit-dev <[email protected]> Subject: RE: [vpp-dev] notable issues before 2610-rc1 Thanks. Already merged, and initial testing suggests VPP startup is now stable in CSIT. Vratko. From: [email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>> On Behalf Of Monendra Singh Kushwaha via lists.fd.io Sent: Wednesday, 23 September, 2026 08:40 To: [email protected]<mailto:[email protected]> Cc: csit-dev <[email protected]<mailto:[email protected]>> Subject: Re: [vpp-dev] notable issues before 2610-rc1 Hi Vratko, >> 4. Octeon driver got some merges, but now VPP usually [7] does not start [8] >> in CSIT. I have raised following patch to gerrit to fix octeon driver issue https://gerrit.fd.io/r/c/vpp/+/46835 Thanks, Monendra From: [email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>> on behalf of Vratko Polak -X (vrpolak - PANTHEON TECHNOLOGIES at Cisco) via lists.fd.io <[email protected]<mailto:[email protected]>> Date: Tuesday, 22 September 2026 at 8:10 PM To: vpp-dev <[email protected]<mailto:[email protected]>> Cc: csit-dev <[email protected]<mailto:[email protected]>> Subject: [EXTERNAL] [vpp-dev] notable issues before 2610-rc1 Prioritize security for external emails: Confirm sender and content safety before clicking links or opening attachments Report Suspicious<https://us-phishalarm-ewt.proofpoint.com/EWT/v1/CRVmXkqW!ui3X-h-YrjX626H8GR36br_ETt5Z-w3Nynf1T7ucBvrwQKSMCS8KR59Alu8ndsK6pgR3nshGrejiI7cKCF4RqAATdyFpR6BiB3_s15DzYtIBwfRrDvbM5xbAIusOpGjVVa4rjyjddXYc$> I plan to discuss those on VPP semimonthly meeting in few minutes, this is both my agenda and a summary for people not joining that meeting. RC1 is this week and some issues are quite bad and may warrant some reverts, but as bugfixes are allowed until RC2 we have still few weeks time. 1. QAT no traffic after DPDK bump [0]. My bad, I got distracted by transient issues and forgot to re-run larger test selection. I am still actively investigating, I have put my findings so far into a quick ticket [1]. Kai was on CSIT meeting and he mentioned some compile flag for symmetric ciphers, but currently plugins/dpdk/cryptodev/ registers zero handlers, which cannot be correct. 2. Before that, OpenSSL bump [2] had introduced an IPsec HW regression on 3n-srf. Both issues are visible in trending [3]. 3. New packed/natural buffer layout [4], on Xeon with huge pages it causes a regression. I opened a ticket [5] in favor of keeping old behavior (all natural) as default [6], but if most of users like the new default I can adapt CSIT to set “natural” in startup.conf explicitly instead. 4. Octeon driver got some merges, but now VPP usually [7] does not start [8] in CSIT. I am not sure if it is an infra issue or VPP source issue. I will try to investigate before RC2. Vratko. [0] 46433: dpdk: bump to DPDK 26.07 and rdma-core 64.0 | https://gerrit.fd.io/r/c/vpp/+/46433<https://urldefense.com/v3/__https:/gerrit.fd.io/r/c/vpp/*/46433__;Kw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsIWLRbMQ$> [1] No cryprodev handler registers since DPDK bump to 26.07 | https://github.com/FDio/vpp/issues/3740<https://urldefense.com/v3/__https:/github.com/FDio/vpp/issues/3740__;!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7Us2qdmvQ4$> [2] 46462: build: update to openssl 4.0.1 in optional deps | https://gerrit.fd.io/r/c/vpp/+/46462<https://urldefense.com/v3/__https:/gerrit.fd.io/r/c/vpp/*/46462__;Kw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsF9KYTFc$> [3] https://csit.fd.io/trending/#eNrtVctOwzAQ_JpwQRvZSUzKoYeW_Ady7W1jESfGa6jC15PHwcmhEiDxUMXFjx2vZ7yjlSl0Hh8Jm20i9km5T7LS6GFI8t3tML06B3kLaD1wxk6YOY4bztQzaKefQPnehQ644JsDcAUYauMK4whVEdqmPkvqWwVD7CAJwbQBJFIm7k7KgvV-JMoeRiL9ElasEXF1H5HLWmKC9ChjxiwxogFpwfQZwfGOo5cWybxhvGgqQTyhhqouQLXmD71boKs61F2UJ0Y3OEtzlgqeFiyeisFsDIoqKasJ-lEDnf43cHqg_n4DyR__TAde0HL9HXj_i159sdmuzquPNtvCK1HdtJ230w83rKnuzhC8kQ1t56x5Mye8A2ma-UA<https://urldefense.com/v3/__https:/csit.fd.io/trending/*eNrtVctOwzAQ_JpwQRvZSUzKoYeW_Ady7W1jESfGa6jC15PHwcmhEiDxUMXFjx2vZ7yjlSl0Hh8Jm20i9km5T7LS6GFI8t3tML06B3kLaD1wxk6YOY4bztQzaKefQPnehQ644JsDcAUYauMK4whVEdqmPkvqWwVD7CAJwbQBJFIm7k7KgvV-JMoeRiL9ElasEXF1H5HLWmKC9ChjxiwxogFpwfQZwfGOo5cWybxhvGgqQTyhhqouQLXmD71boKs61F2UJ0Y3OEtzlgqeFiyeisFsDIoqKasJ-lEDnf43cHqg_n4DyR__TAde0HL9HXj_i159sdmuzquPNtvCK1HdtJ230w83rKnuzhC8kQ1t56x5Mye8A2ma-UA__;Iw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsCxcjlmE$> [4] 46182: buffers: fix pool allocation on small pages | https://gerrit.fd.io/r/c/vpp/+/46182<https://urldefense.com/v3/__https:/gerrit.fd.io/r/c/vpp/*/46182__;Kw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsoKCGZx0$> [5] Packed buffer layout as new default has worse performance with huge pages in CSIT | https://github.com/FDio/vpp/issues/3739<https://urldefense.com/v3/__https:/github.com/FDio/vpp/issues/3739__;!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7Us8uQD25U$> [6] 46703: buffers: make natural layout always default | https://gerrit.fd.io/r/c/vpp/+/46703<https://urldefense.com/v3/__https:/gerrit.fd.io/r/c/vpp/*/46703__;Kw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsnGyDdgQ$> [7] https://csit.fd.io/stats/#eNrLyk-yTS7OLNEtKyjQLUgtStPNLSrSTUnMzKnUzU0sLkkt0jXO081PLgEARDUPQQ<https://urldefense.com/v3/__https:/csit.fd.io/stats/*eNrLyk-yTS7OLNEtKyjQLUgtStPNLSrSTUnMzKnUzU0sLkkt0jXO081PLgEARDUPQQ__;Iw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7UsEU9i6NU$> [8] https://logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-mrr-daily-master-3n-oct/35663534320/log.html.gz#s1-s1-s1-k1-k6<https://urldefense.com/v3/__https:/logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-mrr-daily-master-3n-oct/35663534320/log.html.gz*s1-s1-s1-k1-k6__;Iw!!CRVmXkqW!Sa68gFLj0uEVLRxcLtd64r8zwi5TRYhLMw5cSzl0J2OHwRsN9c-Ty1GR8c56AJfBziCLAC_DR37ZJypNRuunz7Us4ZMGGvM$>
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#27244): https://lists.fd.io/g/vpp-dev/message/27244 Mute This Topic: https://lists.fd.io/mt/121375868/21656 Group Owner: [email protected] Unsubscribe: https://lists.fd.io/g/vpp-dev/leave/14379924/21656/631435203/xyzzy [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
