while the current DTSpec says nothing about the chosen node possibly
having a unit-address, if i search the current linux kernel code base,
there are a few examples of just that:
$ git grep chosen@
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
arch/powerpc/boot/oflib.c: chosen = of_finddevice("/chosen@0");
drivers/of/base.c: of_chosen = of_find_node_by_path("/chosen@0");
drivers/of/fdt.c: offset = fdt_path_offset(fdt, "/chosen@0");
drivers/of/fdt.c: (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
is there some lingering value to the node name "chosen@0"? or should
this all be simplified to just "chosen"?
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca/dokuwiki
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
On Mon, Aug 6, 2018 at 6:34 AM Robert P. J. Day [off-list ref] wrote:
while the current DTSpec says nothing about the chosen node possibly
having a unit-address, if i search the current linux kernel code base,
there are a few examples of just that:
$ git grep chosen@
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
Nothing MIPS based should need this I think.
arch/powerpc/boot/oflib.c: chosen = of_finddevice("/chosen@0");
drivers/of/base.c: of_chosen = of_find_node_by_path("/chosen@0");
drivers/of/fdt.c: offset = fdt_path_offset(fdt, "/chosen@0");
drivers/of/fdt.c: (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
is there some lingering value to the node name "chosen@0"? or should
this all be simplified to just "chosen"?
These are needed for some PPC system(s) though I don't have any clue
which ones specifically.
Rob
On Mon, 6 Aug 2018, Rob Herring wrote:
On Mon, Aug 6, 2018 at 6:34 AM Robert P. J. Day [off-list ref] wrote:
quoted
while the current DTSpec says nothing about the chosen node
possibly having a unit-address, if i search the current linux
kernel code base, there are a few examples of just that:
$ git grep chosen@
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
arch/mips/generic/yamon-dt.c: chosen_off = fdt_path_offset(fdt, "/chosen@0");
Nothing MIPS based should need this I think.
i just submitted a patch to the MIPS list, and they're applying it
to mips-next for 4.19, so i just need to browbeat the PPC folks into
dropping it, and all references to "/chosen@0" can be tossed entirely.
i love deleting stuff.
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca/dokuwiki
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================