Re: [PATCH] RISC-V: virt: This is a "sifive,test1" test finisher

[Palmer Dabbelt <palmerdabbelt@google.com>]

On Thu, 21 Nov 2019 10:55:32 PST (-0800), alistair23@gmail.com wrote:
> On Wed, Nov 20, 2019 at 6:40 PM Palmer Dabbelt <palmerdabbelt@google.com> wrote:
>>
>> On Sun, 10 Nov 2019 13:10:33 PST (-0800), david@gibson.dropbear.id.au wrote:
>> > On Fri, Nov 08, 2019 at 10:13:16AM -0800, Palmer Dabbelt wrote:
>> >> On Fri, 08 Nov 2019 10:04:47 PST (-0800), Peter Maydell wrote:
>> >> > On Fri, 8 Nov 2019 at 17:15, Alistair Francis <alistair23@gmail.com> wrote:
>> >> > >
>> >> > > On Fri, Nov 8, 2019 at 9:05 AM Palmer Dabbelt <palmer@sifive.com> wrote:
>> >> > > >
>> >> > > > The test finisher implements the reset command, which means it's a
>> >> > > > "sifive,test1" device.  This is a backwards compatible change, so it's
>> >> > > > also a "sifive,test0" device.  I copied the odd idiom for adding a
>> >> > > > two-string compatible field from the ARM virt board.
>> >> > > >
>> >> > > > Fixes: 9a2551ed6f ("riscv: sifive_test: Add reset functionality")
>> >> > > > Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
>> >> > > > Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
>> >> > > > ---
>> >> > > >  hw/riscv/virt.c | 5 ++++-
>> >> > > >  1 file changed, 4 insertions(+), 1 deletion(-)
>> >> > > >
>> >> > > > diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
>> >> > > > index 23f340df19..74f2dce81c 100644
>> >> > > > --- a/hw/riscv/virt.c
>> >> > > > +++ b/hw/riscv/virt.c
>> >> > > > @@ -359,7 +359,10 @@ static void create_fdt(RISCVVirtState *s, const struct MemmapEntry *memmap,
>> >> > > >      nodename = g_strdup_printf("/test@%lx",
>> >> > > >          (long)memmap[VIRT_TEST].base);
>> >> > > >      qemu_fdt_add_subnode(fdt, nodename);
>> >> > > > -    qemu_fdt_setprop_string(fdt, nodename, "compatible", "sifive,test0");
>> >> > > > +    {
>> >> > > > +        const char compat[] = "sifive,test1\0sifive,test0";
>> >> > >
>> >> > > Does this really work? Why not use qemu_fdt_setprop_cells()?
>> >> > >
>> >> > > Alistair
>> >> > >
>> >> > > > +        qemu_fdt_setprop(fdt, nodename, "compatible", compat, sizeof(compat));
>> >> > > > +    }
>> >> >
>> >> > qemu_fdt_setprop_cells() is for "set this property to
>> >> > contain this list of 32-bit integers" (and it does a byteswap
>> >> > of each 32-bit value from host to BE). That's not what
>> >> > you want for a string (or a string list, which is what
>> >> > we have here).
>> >> >
>> >> > Cc'ing David Gibson who's our device tree expert to see if there's
>> >> > a nicer way to write this. Oddly, given that it's used in the
>> >> > ubiquitous 'compatible' prop, the dtc Documentation/manual.txt
>> >> > doesn't say anything about properties being able to be
>> >> > 'string lists', only 'strings', '32 bit numbers', 'lists of
>> >> > 32-bit numbers' and 'byte sequences'. You have to dig through
>> >> > the header file comments to deduce that a string list is
>> >> > represented by a string with embedded NULs separating
>> >> > each list item.
>> >>
>> >> I copied this from hw/arm/virt.c, but messed up.  There they use
>> >>
>> >>        const char compat[] = "arm,armv8-timer\0arm,armv7-timer";
>> >>        qemu_fdt_setprop(vms->fdt, "/timer", "compatible",
>> >>                         compat, sizeof(compat));
>> >
>> > I'm not sure what you're saying is messed up.  AFAICT, this matches
>> > the code you have above, and both should be correct.
>>
>> Sorry, I must have been hallucinating.  For some reason I though I wrote
>> qemu_fdt_setprop_string(... compat).
>>
>> I'd like to take this for 4.2 if possible, but I don't think I have a
>> reviewed-by (I just got my email set up on my Google computer, so it might be
>> messy for a bit).  I'm happy to submit the cleaner valist version after 4.2, as
>> per Peter's suggestion.
>>
>> Alistair: are you OK with this?
>
> Yeah, that works for me.
>
> For 5.0 we can then merge Anup's patch and your series improving the
> multi compat support.

Sounds good.  I think the syscon driver is the right way to go, but I don't
think it's 4.2 material as it's all a bit late.  This one at least fixes the
bug where the DTS and hardware don't match.

> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
>
> Alistair
>
>>
>> >> I'll send a v2, but I'd be happy to add some sort of setprop_stringlist
>> >> function.  Maybe we just indicate the length with two '\0's?  IIRC that's
>> >> how other similar-looking data structures are encoded.
>> >>
>> >
>> > --
>> > David Gibson                  | I'll have my music baroque, and my code
>> > david AT gibson.dropbear.id.au        | minimalist, thank you.  NOT _the_ _other_
>> >                               | _way_ _around_!
>> > http://www.ozlabs.org/~dgibson