[PATCH v4 0/9] ARM/arm64: dts: rcar: Enable SCIF_CLK frequency and pins
horms at verge.net.au
Tue Feb 2 05:16:47 PST 2016
On Tue, Feb 02, 2016 at 11:21:33AM +0100, Simon Horman wrote:
> On Fri, Jan 29, 2016 at 11:17:17AM +0100, Geert Uytterhoeven wrote:
> > Hi Simon, Magnus,
> > This patch series adds support for the external crystal feeding SCIF_CLK
> > to (H)SCIF on the various Renesas R-Car Gen1, Gen2, and Gen3 development
> > boards. This increases the range and accuracy of supported baud rates.
> > Support for this has been accepted in the driver in v4.5-rc1.
> > Changes compared to v3:
> > - Change one-line summary prefix to match current arm-soc practices,
> > - Tested on r8a7794/alt,
> > - Rebased,
> > Changes compared to v2:
> > - Add support for r8a7778/bockw, r8a7779/marzen, r8a7790/lager,
> > r8a7791/porter, r8a7793/gose, r8a7794/alt, and r8a7794/silk,
> > - Add Reviewed-by.
> > Testing:
> > - This was tested locally on r8a7791/koelsch and r8a7795/salvator-x,
> > - This was tested remotely on r8a7778/bockw, r8a7779/marzen, and
> > r8a7794/alt,
> > - This has been in renesas-drivers for a while, so I assume it has
> > received basic testing on other boards,
> > - Hints for various levels of testing:
> > - Change the "dev_dbg" after "done:" in
> > drivers/tty/serial/sh-sci.c to "dev_info" to print clock and bit
> > rate,
> > - "yes U | tr -d '\n' > /dev/ttySCx" outputs a square wave with a
> > frequency of half the bit rate on TXD,
> > - Modifying the frequency of SCIF_CLK causes garbage on the serial
> > console, unless you compensate by changing the bit rate in
> > /chosen/stdout-path,
> > - If you disable the SCIF_CLK, the brg_int clock will be used. If
> > you change that clock, you'll get garbage again.
> > Dependencies:
> > - renesas-devel-20160129-v4.5-rc1,
> > - series "[PATCH v4 0/7] ARM: dts: R-Car: Add SCIF fallback
> > compatibility strings",
> > - series "[PATCH v4 00/11] ARM: dts: shmobile: Rename the serial port
> > clock to fck",
> > - series "[PATCH v4 0/7] ARM: dts: r-car: Add BRG support for
> > (H)SCIF".
> > Thanks for applying!
> Is this safe to queue up in the light of the dependency on
Discussed elsewhere; I have queued this up.
More information about the linux-arm-kernel