Return-Path: From: Ilya Faenson To: Marcel Holtmann CC: Guenter Roeck , "linux-kernel@vger.kernel.org" , Frederic Danis , BlueZ development , "linux-serial@vger.kernel.org" Subject: RE: Build error -in -next due to 'Bluetooth: hci_uart: Support operational speed during setup' Date: Mon, 15 Jun 2015 16:51:12 +0000 Message-ID: References: <557ED81D.7090908@roeck-us.net> <7CF3D545-DAE4-4510-9726-5D40740DDC5D@holtmann.org> <557EE045.2090503@roeck-us.net> <7F15059F-B180-4722-8167-864C4A24CE9B@holtmann.org> In-Reply-To: <7F15059F-B180-4722-8167-864C4A24CE9B@holtmann.org> Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 List-ID: Hi Marcel, -----Original Message----- From: Marcel Holtmann [mailto:marcel@holtmann.org]=20 Sent: Monday, June 15, 2015 10:43 AM To: Ilya Faenson Cc: Guenter Roeck; linux-kernel@vger.kernel.org; Frederic Danis; BlueZ deve= lopment; linux-serial@vger.kernel.org Subject: Re: Build error -in -next due to 'Bluetooth: hci_uart: Support ope= rational speed during setup' Hi Ilya, >>> alpha:allmodconfig fails to build in -next with >>>=20 >>> drivers/bluetooth/hci_ldisc.c: In function 'hci_uart_set_baudrate': >>> drivers/bluetooth/hci_ldisc.c:276:22: error: 'BOTHER' undeclared >>>=20 >>> BOTHER is not defined for all architectures and otherwise only >>> used with '#ifdef BOTHER'. >>=20 >> thanks for the report. Is this something that should be #ifdef at the us= ing side or maybe just fixed for the other architectures. >>=20 >=20 > So far it has been fixed with #ifdef, and this is the first code which > uses it outside drivers/tty. So my first tendency would be to fix it > on the user side. >=20 > However, looking into it, I think only alpha and avr32 are affected. > Let me submit patches against those and see where it goes. >=20 > IF: The BOTHER flag should not be involved in setting the baud rate at al= l. I've published a patch last week that implements the hci_uart_set_baudra= te properly (in my opinion). can you extract this and just submit that version. We might just take that = into hci_ldisc.c right away. IF: I've published my version now, thanks. However fixing the #ifdef around BOTHER is useful as well since they look u= nneeded. At least I can not see a reason why a platform would not define th= em. Regards Marcel