From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Subject: Re: [PATCH v5 0/6] bcm2835: auxiliar device support for spi References: <1441359711-2800-1-git-send-email-kernel@martin.sperl.org> <87h9n4weg8.fsf@eliezer.anholt.net> <55F1A651.5090102@tronnes.org> <8B777380-A2A1-475E-8A1C-942BF28EE160@martin.sperl.org> From: Phil Elwell Message-ID: <55F1B69F.1010505@raspberrypi.org> Date: Thu, 10 Sep 2015 17:58:07 +0100 MIME-Version: 1.0 In-Reply-To: <8B777380-A2A1-475E-8A1C-942BF28EE160@martin.sperl.org> Content-Type: multipart/alternative; boundary="------------070201010405030201070908" To: Martin Sperl , =?UTF-8?Q?Noralf_Tr=c3=b8nnes?= Cc: Eric Anholt , Stephen Warren , Lee Jones , Russell King , Mark Brown , Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , Kumar Gala , devicetree@vger.kernel.org, linux-rpi-kernel@lists.infradead.org, linux-spi@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org List-ID: --------------070201010405030201070908 Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: 8bit Noralf pointed me at fixed-factor-clock, and that works in our (downstream) environment: soc: soc { ... uart1: uart@7e215040 { compatible = "brcm,bcm2835-aux-uart", "ns16550"; reg = <0x7e215040 0x40>; interrupts = <1 29>; clocks = <&clk_uart1>; reg-shift = <2>; no-loopback-test; status = "disabled"; }; }; clocks: clocks { ... clk_core: clock@2 { compatible = "fixed-clock"; reg = <2>; #clock-cells = <0>; clock-output-names = "core"; clock-frequency = <250000000>; }; ... clk_uart1: clock@6 { compatible = "fixed-factor-clock"; clocks = <&clk_core>; #clock-cells = <0>; clock-div = <1>; clock-mult = <2>; }; }; Phil On 10/09/2015 16:57, Martin Sperl wrote: >> On 10.09.2015, at 17:48, Noralf Tr�nnes wrote: >> >> This looks interesting. >> But there's a challenge with the uart1 and the 8250 driver. >> >> Phil Elwell has this to say: >> This means that that UART1 isn't an exact clone of a 8250 UART. >> In a particular, the clock divisor is calculated differently. >> A standard 8250 derives the baud rate as clock/(divisor16), >> whereas the BCM2835 mini UART uses clock/(divisor8). This means >> that if you want to use the standard driver then you need to lie >> about the clock frequency, providing a value is twice the real >> value, in order for a suitable divisor to be calculated. >> >> Ref: https://github.com/raspberrypi/linux/pull/1008#issuecomment-139234607 >> >> So either we need a new uart1 driver or a doubled clock freq. somehow. > Found out the same thing and communicated it to Eric - not > knowing about the different divider� > > Martin --------------070201010405030201070908 Content-Type: text/html; charset="windows-1252" Content-Transfer-Encoding: 8bit Noralf pointed me at fixed-factor-clock, and that works in our (downstream) environment:

�� soc: soc {�� ...�� uart1: uart@7e215040 {�� compatible = "brcm,bcm2835-aux-uart", "ns16550";�� reg = <0x7e215040 0x40>;�� interrupts = <1 29>;�� clocks = <&clk_uart1>;�� reg-shift = <2>;�� no-loopback-test;�� status = "disabled";�� };�� };�� clocks: clocks {�� ... �� clk_core: clock@2 { �� compatible = "fixed-clock"; �� reg = <2>; �� #clock-cells = <0>; �� clock-output-names = "core"; �� clock-frequency = <250000000>; �� }; �� ...�� clk_uart1: clock@6 {�� compatible = "fixed-factor-clock";�� clocks = <&clk_core>;�� #clock-cells = <0>;�� clock-div = <1>;�� clock-mult = <2>;�� };�� };Phil

On 10/09/2015 16:57, Martin Sperl wrote:

On 10.09.2015, at 17:48, Noralf Tr�nnes <noralf@tronnes.org> wrote:

This looks interesting.
But there's a challenge with the uart1 and the 8250 driver.

Phil Elwell has this to say:
This means that that UART1 isn't an exact clone of a 8250 UART.
In a particular, the clock divisor is calculated differently.
A standard 8250 derives the baud rate as clock/(divisor16),
whereas the BCM2835 mini UART uses clock/(divisor8). This means
that if you want to use the standard driver then you need to lie
about the clock frequency, providing a value is twice the real
value, in order for a suitable divisor to be calculated.

Ref: https://github.com/raspberrypi/linux/pull/1008#issuecomment-139234607

So either we need a new uart1 driver or a doubled clock freq. somehow.

Found out the same thing and communicated it to Eric - not 
knowing about the different divider�

Martin

--------------070201010405030201070908--