> -----Original Message----- > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon > Sent: Monday, January 16, 2017 2:42 PM > To: Azarewicz, PiotrX T <piotrx.t.azarew...@intel.com> > Cc: dev@dpdk.org; De Lara Guarch, Pablo <pablo.de.lara.gua...@intel.com>; > Doherty, Declan <declan.dohe...@intel.com> > Subject: Re: [dpdk-dev] [PATCH v5] crypto/aesni_gcm: migration from MB > library to ISA-L > > 2017-01-16 13:37, Piotr Azarewicz: > > v5 changes: > > - rebase on top of dpdk-next-crypto > > - remove the perftest output from commit message > > - correction in aesni_gcm.rst > > - fix typo > > Why have you removed the perftest output? > Because it is not in favor of this patch?
Hi, We removed the performance results because we are not allowed to publish them publically without legal approval. Also, and more importantly, the perftest results aren't reliable which is one of the reasons that we are developing the new dpdk-test-crypto-perf app (patch submitted). Using the newer app we are seeing more of less the same performance, slightly higher for some packets sizes, slightly lower for some others. John.
