From mboxrd@z Thu Jan 1 00:00:00 1970 From: "christophe.ricard" Subject: Re: [PATCH v1 27/34] NFC: st21nfcb: Add HCI protocol over NCI protocol support Date: Tue, 27 Jan 2015 23:16:32 +0100 Message-ID: <54C80E40.7020707@gmail.com> References: <1418072919-10535-1-git-send-email-christophe-h.ricard@st.com> <1418072919-10535-28-git-send-email-christophe-h.ricard@st.com> <20150127004958.GB31396@ribalta.ger.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20150127004958.GB31396-41CF7WKNp/FoDWY/xQGDymt3HXsI98Cx0E9HWUfgJXw@public.gmane.org> Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Samuel Ortiz Cc: linux-nfc-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org, christophe-h.ricard-qxv4g6HH51o@public.gmane.org, devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: devicetree@vger.kernel.org Hi Samuel, The main reason for making an st21nfcb_hci_core is because the hci network initialization procedure is proprietary. It is almost using only NCI standard commands but some id's and commands could be proprietary. I would ideally rely on the hci core framework but i run out of idea when trying this. However i think "exporting the relevant API for drivers to send HCI commands/events through their NCI connections." is a good idea i need to dig into. It will then needs to hci/nci hook for the proprietary stuff. I have just one worry, when do you think those updates (patch 20 & 27) needs to be ready for your 3.20 pull request ? Best Regards Christophe On 27/01/2015 01:49, Samuel Ortiz wrote: > Christophe, > > On Mon, Dec 08, 2014 at 10:08:32PM +0100, Christophe Ricard wrote: >> Some st21nfcb features are only available through HCI commands. Those >> HCI commands can be address over NCI by sending data using a dynamic >> conn_id. This is useful for example for Secure Element communication. >> >> The HCI core brings the minimal HCI functions required to communicate with >> a secure element. >> >> Signed-off-by: Christophe Ricard >> --- >> drivers/nfc/st21nfcb/Makefile | 2 +- >> drivers/nfc/st21nfcb/st21nfcb_hci_core.c | 783 +++++++++++++++++++++++++++++++ >> drivers/nfc/st21nfcb/st21nfcb_hci_core.h | 134 ++++++ >> 3 files changed, 918 insertions(+), 1 deletion(-) >> create mode 100644 drivers/nfc/st21nfcb/st21nfcb_hci_core.c >> create mode 100644 drivers/nfc/st21nfcb/st21nfcb_hci_core.h > I may be missing some details, but I believe most of this code could be > made part of the core NCI implementation. I don't mind creating an > hci_dev from there and exporting the relevant API for drivers to send > HCI commands/events through their NCI connections. > You would not need to bring chunks of the HCI core into a driver, and > as a side effect, patch #21 would no longer be needed. > > Cheers, > Samuel. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html