Hi Marcel, > -----Original Message----- > From: Marcel Holtmann [mailto:marcel@xxxxxxxxxxxx] > Sent: Wednesday, October 3, 2018 12:26 PM > To: Phulari, Mallikarjun <mallikarjun.phulari@xxxxxxxxx> > Cc: linux-bluetooth@xxxxxxxxxxxxxxx > Subject: Re: [PATCH] Bluetooth : Errata Service Release 8, Erratum 3253 > > Hi Malikarjun, > > >>> L2CAP: New result values > >>> 0x0006 - Connection refused – Invalid Source CID > >>> 0x0007 - Connection refused – Source CID already allocated > >>> > >>> As per the ESR08_V1.0.0, 1.11.2 Erratum 3253, Page No. 54, "Remote > >>> CID invalid Issue". > >>> Applies to Core Specification versions: V5.0, V4.2, v4.1, v4.0, and > >>> v3.0 + HS Vol 3, Part A, Section 4.2, 4.3, 4.14, 4.15. > >>> > >>> Core Specification Version 5.0, Page No.1753, Table 4.6 and Page No. > >>> 1767, Table 4.14 > >>> > >>> New result values are added to l2cap connect/create channel response > >>> as > >>> 0x0006 - Connection refused – Invalid Source CID > >>> 0x0007 - Connection refused – Source CID already allocated > >>> > >>> Signed-off-by: Mallikarjun Phulari <mallikarjun.phulari@xxxxxxxxx> > >>> --- > >>> include/net/bluetooth/l2cap.h | 6 ++++++ > >>> net/bluetooth/l2cap_core.c | 17 ++++++++++++++++- > >>> 2 files changed, 22 insertions(+), 1 deletion(-) > >>> > >>> diff --git a/include/net/bluetooth/l2cap.h > >>> b/include/net/bluetooth/l2cap.h index 0697fd4..c7f97cc 100644 > >>> --- a/include/net/bluetooth/l2cap.h > >>> +++ b/include/net/bluetooth/l2cap.h > >>> @@ -284,6 +284,12 @@ struct l2cap_conn_rsp { > >>> #define L2CAP_CR_INVALID_SCID 0x0009 > >>> #define L2CAP_CR_SCID_IN_USE 0x000A > >>> > >>> +/* connect/create channel results > >>> + * New result codesAs per ESR08_V1.0.0, Erratum 3253 */ > >>> +#define L2CAP_CR_BREDR_INVALID_SCID 0x0006 > >>> +#define L2CAP_CR_BREDR_SCID_IN_USE 0x0007 > >>> + > >> > >> I am confused here. We currently have this list: > >> > >> /* connect/create channel results */ > >> #define L2CAP_CR_SUCCESS 0x0000 > >> #define L2CAP_CR_PEND 0x0001 > >> #define L2CAP_CR_BAD_PSM 0x0002 > >> #define L2CAP_CR_SEC_BLOCK 0x0003 > >> #define L2CAP_CR_NO_MEM 0x0004 > >> #define L2CAP_CR_BAD_AMP 0x0005 > >> #define L2CAP_CR_AUTHENTICATION 0x0005 #define > L2CAP_CR_AUTHORIZATION > >> 0x0006 > >> #define L2CAP_CR_BAD_KEY_SIZE 0x0007 > >> #define L2CAP_CR_ENCRYPTION 0x0008 > >> #define L2CAP_CR_INVALID_SCID 0x0009 > >> #define L2CAP_CR_SCID_IN_USE 0x000A > >> > >> Now here we have 0x0006 and 0x0007 taken. Have these been changed now. > >> The L2CAP_CR_AUTHORIZATION we actually do use in LE and if this table > >> is not treated with independent name spaces for BR/EDR vs LE, then we > >> need to split this one first and get the LE only ones proper names and fix their > usage. > >> > > The current name spaces that you have mentioned above, have the combined > result values for LE Credit base Connection (VOL 3, Part A, 4.23, Table 4.20) and > for L2CAP Create/Connect (VOL3, Part A, 4.3 /4.15, Table 4.6/4.14). > > To create separate name spaces for each set of error codes for L2CAP channel > request, L2CAP create Connection and LE credit based create connection will > add more rework to replace the error codes where these are being used. > > > > Therefore, to avoid the rework , I had used the result values as: > > > > #define L2CAP_CR_BREDR_INVALID_SCID 0x0006 > > #define L2CAP_CR_BREDR_SCID_IN_USE 0x0007 > > > > If you suggest to separate the name spaces for LE and BREDR result values, I > shall re-send the v3 patch with the changes. > > if the currently 0x0006-0x000A are LE only, then we should change their names > to L2CAP_CR_LE_* and convert their usages. And then add the new BR/EDR only > 0x0006-0x0007 into the list (without _BREDR in the name). I will change the names of the result values and their usages as you suggested and I will send you the new patch. > > Regards > > Marcel Thanks & Regards Mallikarjun Phulari
