svn commit: r324822 - head/sys/modules/dtb/allwinner [removal of sinovoip-bpi-m3.dts from sys/modules/dtb/allwinner/Makefile DTS list]

Mark Millard markmi at dsl-only.net
Sun Oct 22 03:05:30 UTC 2017


[Reporting what seems to be the case for
/boot/dtb/sinovoip-bpi-m3.dtb old and newer
and, so, its source files (old and newer).]

On 2017-Oct-21, at 4:09 PM, Mark Millard <markmi at dsl-only.net> wrote:

> On 2017-Oct-21, at 2:06 PM, Mark Millard <markmi at dsl-only.net> wrote:
> 
>> On 2017-Oct-21, at 11:43 AM, Emmanuel Vadot <manu at bidouilliste.com> wrote:
>> 
>>> On Sat, 21 Oct 2017 11:26:57 -0700
>>> Mark Millard <markmi at dsl-only.net> wrote:
>>> 
>>>> On 2017-Oct-21, at 10:58 AM, Emmanuel Vadot <manu at bidouilliste.com> wrote:
>>>> 
>>>>>>> . . .
>>>> 
>>>> If I understand the current status correctly, the
>>>> recent changes to use upstream DTS materials took
>>>> the BPI-M3 from having, e.g., USB working to USB
>>>> not working. -r323641 removed AWUSBPHY_TYPE_A83T
>>>> from enum awusbphy_type because of what was missing
>>>> from upstream DTS materials and such.
>>> 
>>> Uses of upstream DTS is active since ~2 years ago.
>>> I think I removed support for USB on A83T because I couldn't test, but
>>> maybe the code just work and adding the compatible string again might
>>> do it. Feel free to send a patch on phabricator if it does.
>> 
>> . . .
. . .

Using the .dtb files produced and using
dtc to convert back to .dts files (via
-s for allowing diffs) I get the
following (comparing to -r324743). The
list of types of changes is rather
short (many  instances, mostly not
shown below).

The types of differences may suggest
some about compatibility for going
forward.

# diff -u ~/bpi-m3-sorted*.dts | more
--- /root/bpi-m3-sorted-317015.dts      2017-10-22 01:47:30.424893000 -0700
+++ /root/bpi-m3-sorted-324743.dts      2017-10-22 02:17:24.545620000 -0700
@@ -7,6 +7,127 @@
        compatible = "sinovoip,bpi-m3", "allwinner,sun8i-a83t";
        interrupt-parent = <0x1>;
        model = "Sinovoip BananaPi M3 v1.2";
+       __local_fixups__ {
+
+               fixup = "/cpus/cpu at 0:clocks:0";
+               fixup = "/cpus/cpu at 0:cpu-supply:0";
. . .
+               fixup = "/leds/blue_led:gpios:0";
+               fixup = "/:interrupt-parent:0";
+       };

(lots of such fixup's) and lots of things like:

-               linux,phandle = <0x46>;

as well as some (17?) like:

-               phandle = <0x46>;

but other plain phandle's are instead unchanged.

There are 4 'function = "gpio_out";' related
differences as well (usb tied):

                        ahci_pwr_pin at 0 {
 
-                               allwinner,drive = <0x0>;
-                               allwinner,function = "gpio_out";
                                allwinner,pins = "PD25";
-                               allwinner,pull = <0x0>;
-                               linux,phandle = <0x31>;
+                               function = "gpio_out";
                                phandle = <0x31>;
+                               pins = "PB8";
                        };

and later:
                        usb0_vbus_pin at 0 {
 
-                               allwinner,drive = <0x0>;
-                               allwinner,function = "gpio_out";
-                               allwinner,pins = "PB9";
-                               allwinner,pull = <0x0>;
-                               linux,phandle = <0x32>;
+                               function = "gpio_out";
                                phandle = <0x32>;
+                               pins = "PB9";
                        };
                        usb1_vbus_pin at 0 {
 
-                               allwinner,drive = <0x0>;
-                               allwinner,function = "gpio_out";
                                allwinner,pins = "PD24";
-                               allwinner,pull = <0x0>;
-                               linux,phandle = <0x33>;
+                               function = "gpio_out";
                                phandle = <0x33>;
+                               pins = "PH6";
                        };
                        usb2_vbus_pin at 0 {
 
-                               allwinner,drive = <0x0>;
-                               allwinner,function = "gpio_out";
-                               allwinner,pins = "PH3";
-                               allwinner,pull = <0x0>;
-                               linux,phandle = <0x34>;
+                               function = "gpio_out";
                                phandle = <0x34>;
+                               pins = "PH3";
                        };

[Note the 'allwinner,pins = "PD25";' and the
'allwinner,pins = "PD24";' that were not deleted
when other "allwinner,pins" lines were deleted.]

And that is it for types of sinovoip-bpi-m3.dtb
changes as of -r324743 .

I do not know if any of that suggests a .dtb
handling incompatibility for going forward.

As near as I can tell:

/boot/dtb/sinovoip-bpi-m3.dtb

content is irrelevant to u-boot and
to ubldr, removing one of my worries.

[The above is my first use of dtc as far as I
remember. Interesting.]

===
Mark Millard
markmi at dsl-only.net



More information about the freebsd-arm mailing list