Hi Akhil, Please see inline.
Thanks, Anoob > -----Original Message----- > From: dev <[email protected]> On Behalf Of Anoob Joseph > Sent: Friday, January 31, 2020 9:04 PM > To: Akhil Goyal <[email protected]>; Declan Doherty > <[email protected]>; Thomas Monjalon <[email protected]> > Cc: Ankur Dwivedi <[email protected]>; Jerin Jacob Kollanukkaran > <[email protected]>; Narayana Prasad Raju Athreya > <[email protected]>; Kiran Kumar Kokkilagadda > <[email protected]>; Nithin Kumar Dabilpuram > <[email protected]>; Pavan Nikhilesh Bhagavatula > <[email protected]>; Archana Muniganti <[email protected]>; > Tejasree Kondoj <[email protected]>; Vamsi Krishna Attunuru > <[email protected]>; Lukas Bartosik <[email protected]>; > [email protected] > Subject: Re: [dpdk-dev] [PATCH v3 14/15] net/octeontx2: add inline Tx path > changes > > Hi Akhil, > > Please see inline. > > Thanks, > Anoob > > > -----Original Message----- > > From: Akhil Goyal <[email protected]> > > Sent: Friday, January 31, 2020 7:06 PM > > To: Anoob Joseph <[email protected]>; Declan Doherty > > <[email protected]>; Thomas Monjalon <[email protected]> > > Cc: Ankur Dwivedi <[email protected]>; Jerin Jacob Kollanukkaran > > <[email protected]>; Narayana Prasad Raju Athreya > > <[email protected]>; Kiran Kumar Kokkilagadda > > <[email protected]>; Nithin Kumar Dabilpuram > > <[email protected]>; Pavan Nikhilesh Bhagavatula > > <[email protected]>; Archana Muniganti > <[email protected]>; > > Tejasree Kondoj <[email protected]>; Vamsi Krishna Attunuru > > <[email protected]>; Lukas Bartosik <[email protected]>; > > [email protected] > > Subject: [EXT] RE: [PATCH v3 14/15] net/octeontx2: add inline Tx path > > changes > > > > External Email > > > > ---------------------------------------------------------------------- > > Hi Anoob, > > > > > > > > From: Ankur Dwivedi <[email protected]> > > > > > > Adding pre-processing required for inline IPsec outbound packets. > > > > > > Signed-off-by: Ankur Dwivedi <[email protected]> > > > Signed-off-by: Anoob Joseph <[email protected]> > > > Signed-off-by: Archana Muniganti <[email protected]> > > > Signed-off-by: Tejasree Kondoj <[email protected]> > > > Signed-off-by: Vamsi Attunuru <[email protected]> > > > --- > > > drivers/event/octeontx2/meson.build | 3 +- > > > drivers/event/octeontx2/otx2_worker.h | 6 + > > > drivers/net/octeontx2/otx2_ethdev_sec.c | 82 +++++++++++++ > > > drivers/net/octeontx2/otx2_ethdev_sec.h | 19 +++ > > > drivers/net/octeontx2/otx2_ethdev_sec_tx.h | 181 > > > +++++++++++++++++++++++++++++ > > > 5 files changed, 290 insertions(+), 1 deletion(-) create mode > > > 100644 drivers/net/octeontx2/otx2_ethdev_sec_tx.h > > > > > > diff --git a/drivers/event/octeontx2/meson.build > > > b/drivers/event/octeontx2/meson.build > > > index 56febb8..be4b47a 100644 > > > --- a/drivers/event/octeontx2/meson.build > > > +++ b/drivers/event/octeontx2/meson.build > > > @@ -31,6 +31,7 @@ foreach flag: extra_flags > > > endif > > > endforeach > > > > > > -deps += ['bus_pci', 'common_octeontx2', 'mempool_octeontx2', > > > 'pmd_octeontx2'] > > > +deps += ['bus_pci', 'common_octeontx2', 'cryptodev', > > > +'mempool_octeontx2', > > > 'pmd_octeontx2', > > > + 'security'] > > > > > > includes += include_directories('../../crypto/octeontx2') > > > > > > Why would you need cryptodev in event driver meson.build? > > If you really need it, then why only in meson.build, Makefile should > > also have some change. > > [Anoob] On our platform, it's the eventdev which does packet I/O during inline > processing. So the Tx and Rx paths are embedded into event_dequeue_burst() > and tx_adapter_enqueue(). Hence the need for such dependency. > > I'll have the meson and Makefiles synced. Can send the updated series once you > are done with the rest of the reviews. The dependency was added in > meson.build file when some build failure was observed. [Anoob] You are right. The above change is not necessary. I've confirmed that the build is not failing even without the above. Do you want me to send a new version with this fix? > > > > > Please make sure that all individual patches compile for both meson > > and Makefile. > > I have not run the test yet. Will run it on Monday. > > [Anoob] Sure. From our end, we have all the patches passing the build stages. > In > case if there is any failure, I'll have it addressed with the next version. > > > > > Regards, > > Akhil
