[PATCH v4 07/39] serial: msm: add debug UART

[Sumit Garg]

On Tue, 20 Feb 2024 at 17:09, Caleb Connolly <caleb.connolly at linaro.org> wrote:
>
>
>
> On 20/02/2024 06:08, Sumit Garg wrote:
> > On Fri, 16 Feb 2024 at 02:22, Caleb Connolly <caleb.connolly at linaro.org> wrote:
> >>
> >> Introduce support for early debugging. This relies on the previous stage
> >> bootloader to initialise the UART clocks, when running with U-Boot as
> >> the primary bootloader this feature doesn't work. It will require a way
> >> to configure the clocks before the driver model is available.
> >>
> >> Signed-off-by: Caleb Connolly <caleb.connolly at linaro.org>
> >> ---
> >>  drivers/serial/Kconfig      |  8 ++++++++
> >>  drivers/serial/serial_msm.c | 37 +++++++++++++++++++++++++++++++++++++
> >>  2 files changed, 45 insertions(+)
> >>
> >> diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
> >> index 26460c4e0cab..fbd351a47859 100644
> >> --- a/drivers/serial/Kconfig
> >> +++ b/drivers/serial/Kconfig
> >> @@ -319,6 +319,14 @@ config DEBUG_UART_S5P
> >>           will need to provide parameters to make this work. The driver will
> >>           be available until the real driver-model serial is running.
> >>
> >> +config DEBUG_UART_MSM
> >> +       bool "Qualcomm QUP UART debug"
> >> +       depends on ARCH_SNAPDRAGON
> >
> > Since this debug UART only works for chainloaded configuration, can we
> > somehow add explicit dependency here? Something like !REMAKE_ELF?
>
> With a small patch (which didn't make it into v4 apparently) the
> apq8016_clk_init_uart() function from clock-apq8016 can be adjusted to
> just take a base address rather than "struct msm_clk_priv". It can then
> be called from debug_uart_init() and allows for debug UART to be used
> when U-Boot is running as the first stage.
>
> This is definitely not ideal (although fwiw if the GPLLs were configured
> right then this same function could maybe work on QCS404 as well

QCS404 is chainloaded config too, so the debug UART should work there.

> - the
> RCGs are at the same physical addresses), but I don't think gating it
> behind REMAKE_ELF or something is a great solution here.

I don't have a strong opinion here and I could live with just a
documentation update for debug UART too.

-Sumit