Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Paul Kocialkowski
What does the device actually use for debug port? The fex file in sunxi-boards says: This is not used at all by the kernel. The debug port is hardcoded in the defconfig and set up very early (in early ASM code). uboot could presumably include the debug uart specification it set up in the

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Luc Verhaegen
On Tue, Sep 23, 2014 at 07:06:54PM +0200, Paul Kocialkowski wrote: Le dimanche 21 septembre 2014 à 19:16 +0200, Luc Verhaegen a écrit : Crap. Let me see if i can work some heuristics into this: we assume that u-boot set up a debug uart correctly, and then use that one. Would a

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Henrik Nordström
tis 2014-09-23 klockan 19:03 +0200 skrev Paul Kocialkowski: As far as I know, the U-Boot debug UART output is hardcoded in the config as well (CONS_INDEX), so it could be handed to the kernel correctly. The u-boot console UART is only hardcoded in config because we haven't cared to implement

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Paul Kocialkowski
Specifying the early debug UART through cmdline is not standard behavior. Note that this is different from the console kernel cmdline parameter. This is instead about earlyprintk. It's possible and well defined to use kernel command line arguments to specify the earlycon console and

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Henrik Nordström
tis 2014-09-23 klockan 21:40 +0200 skrev Paul Kocialkowski: Alright. It is however not used on our sunxi-3.4 code, nor on the U-Boot we maintain. Our sunxi-3.4 kernel do use pinmappings from the FEX file, even for UART pinmappings. See drivers/tty/serial/8250/8250_sunxi.c for the UART parts.

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-23 Thread Paul Kocialkowski
Le mardi 23 septembre 2014 à 22:33 +0200, Henrik Nordström a écrit : tis 2014-09-23 klockan 21:40 +0200 skrev Paul Kocialkowski: Alright. It is however not used on our sunxi-3.4 code, nor on the U-Boot we maintain. Our sunxi-3.4 kernel do use pinmappings from the FEX file, even for UART

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-21 Thread Luc Verhaegen
On Sun, Sep 21, 2014 at 04:16:21PM +0200, Paul Kocialkowski wrote: UART0 with the default pinmux can never lead to a working UART on A13, as these pins are not available on the A13 package. When the A13 based device has its accessible UART on UART1, then the UART0 setting does not hurt, it

Re: [linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-21 Thread Michal Suchanek
On 21 September 2014 19:16, Luc Verhaegen l...@skynet.be wrote: On Sun, Sep 21, 2014 at 04:16:21PM +0200, Paul Kocialkowski wrote: UART0 with the default pinmux can never lead to a working UART on A13, as these pins are not available on the A13 package. When the A13 based device has its

[linux-sunxi] [PATCH 3/4] config:sunxi: set debug uart to 0 for sun5i defconfig

2014-09-11 Thread Luc Verhaegen
The sun5i defconfig default of CONFIG_SW_DEBUG_UART=1 is wrong on all devices except for those who use CONS_INDEX=2 in uboot. In others, both A10s as A13, it leads to a system which hardlocks. UART0 with the default pinmux can never lead to a working UART on A13, as these pins are not available