On Wed, Jan 21, 2026 at 2:39 PM Akhil Goyal <[email protected]> wrote:
> IMO, it varies from platform to platform. Adding Kai Ji for details on QAT. > > However, you can use as many VFs as you want. Generally, we use 1VF per > core. > > > > Do you plan to include other platforms as well? > > How about openssl PMD? > > > I'll jump in as Andrew is clocked out for today. Yes we will include testcases for the OpenSSL vdev PMD. For instance in Andrew's RFC the first testcase is "openssl_aead_aes_gcm_encrypt." I guess the only decision to be made there is whether we want to group that under a separate testsuite (software based cryptodev testing) or whether we keep HW/SW testing on the same testsuite (this sounds fine to me). Yes we should add in other valid options like openssl aes-ccm (andrew see https://doc.dpdk.org/guides/cryptodevs/overview.html) and those will be available in the final version for DPDK 26.03. Yes, the DTS testsuite should support all platforms which dpdk-crypto-perf can, not just QAT. We are using QAT as an example (and using it to run the testsuite) because we have a QAT 8970 on one of our servers at UNH. We also have a CN10K with a crypto device but have had issues with VF creation on it in the past and are only running ethdev testing with that board. I can't tell whether there is a software or hardware issue but once the testsuite is ready I would like to rebuild the Marvell SDK and the ubuntu buildroot OS and "clean install" the board to see if we can get Andrew's cryptodev testsuite running on CN10K for testing. But, in general one should be able to install any crypto device on their system and run this testsuite (the testsuite is not device specific). If a cipher algorithm is not supported by the device, the testcase skips. If it is supported, it will run and return a pass or fail.
