On 5/12/2021 11:13 AM, Michal Krawczyk wrote: > Hi, > > this version updates the driver to version 2.3.0, which fixes multiple > bugs, contains part of the work on making the ENA PMD fully MP aware, and > also updates HAL to the latest version. > > More detailed list of changes: > * memcpy mapping to the dpdk-optimized version. > * ena_com (HAL) update to the latest version. > * Bug fixes for the large LLQ headers and devargs parsing. > * Mbuf RSS hash presence indication. > * Bug fix for the default ring size. > * Various fixes for the SMP mode. > > Since v5, the series contains only 3, unmerged patches. > > v5: > * Indicate RSS hash offload in the Rx capabilities field. > * Changes made to 'making the ethdev references multiprocess safe: > - rework commit log, > - remove adapter::port_id field. > * Change made to "disable dev ops not supported by secondary process": > - rework commit log. > > v4: > * Remove patches adding shared memory for SMP: > - net/ena: move default RSS key to shared mem > - net/ena: support SMP for mz alloc counter > * Move release notes to the related patches. > * Squash HAL version update patch with the same commit for the core driver. > * Change names of the macros arguments in the "net/ena/base: destroy > multiple "wait events" patch. > * Removed call to the rte_panic(). > * Squashed ena_com version update with the driver version update. > * Fix multiple checkpatch warnings. > * Fix issues found in the commit logs. > > v3: > * Fix build of the PMD with the debug flags enabled. > * Fix typo in the commit log of the version upgrade patch. > > v2: > * Fix nested declaration of the rte_memcpy on arm64 architecture. > > Michal Krawczyk (1): > net/ena: indicate Rx RSS hash presence > > Stanislaw Kardach (2): > net/ena: make ethdev references multiprocess safe > net/ena: disable dev ops not supported by secondary process >
Series applied to dpdk-next-net/main, thanks.
