Hello.
On 04/06/2016 01:51 PM, Geert Uytterhoeven wrote:
I've got a Porter board (Revision B) which stopped showing serial
output since the patch mentioned in the subject. The terrible clearly
wrong hack below gets serial back (reverting to just the fck clock for
scif0) on both linux-next and linux master.
And I have the same problem on my Porter. Once the SCIF driver probes
ttySC0, all console output stops.
Thanks for your report!
I assume something is wrong with the SCIF_CLK input.
Hence I guess the issue is also fixed by removing the "&scif_clk" block in
arch/arm/boot/dts/r8a7791-porter.dts?
Unfortunately I don't have enough documentation on the board/hardware
to diagnose this further, but hopefully someone else does ;)
I don't have a Porter, so I based my patch on the documentation, which states:
The SCIF_CLK pin of the R-CarM2 is connected to the crystal oscillator
(X4) on the R-Car M2 Application Development Board, which
supplies a clock frequency of 14.7456 MHz.
which matches r8a7791-porter.dts, if I'm not mistaken.
Perhaps the documentation applies to another version of the board than
Revision B?
Questions:
- Does your board have X4 populated?
Mine doesn't (as I told you on IRC). The older board called Henninger B
REV2 does have X4 (on the bottom).
X4 seems to be unpopulated on http://elinux.org/R-Car/Boards/Porter
(which is Revision A, look close to the fan, at 10 o'clock).
This photo seems to match my current board.
On Silk, it is populated on http://elinux.org/R-Car/Boards/SILK
(look close to the SoC, at 1 o'clock).
Populated on my SILK as well.
- If X4 is populated, does it help if you change the scif_clk_pins pinctrl
group in arch/arm/boot/dts/r8a7791-porter.dts from "scif_clk" to
"scif_clk_b"?
- Does anyone have schematics?
I have 3 versions of the Porter schematics, and all 3 have X4 framed with
a comment "DNS" (do not stuff?) at the top.
Thanks!
Gr{oetje,eeting}s,
Geert
MBR, Sergei
