This new helper function encodes the idiom used by the ARM virt board to
set a string array.  I don't currently have a working ARM userspace, so I haven't tested
this, but I made the helper function because I wanted to use it for the
RISC-V virt board where I have tested it.

Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
---
 hw/arm/virt.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index d4bedc2607..4dc00f54d5 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -304,9 +304,8 @@ static void fdt_add_timer_nodes(const VirtMachineState *vms)
 
     armcpu = ARM_CPU(qemu_get_cpu(0));
     if (arm_feature(&armcpu->env, ARM_FEATURE_V8)) {
-        const char compat[] = "arm,armv8-timer\0arm,armv7-timer";
-        qemu_fdt_setprop(vms->fdt, "/timer", "compatible",
-                         compat, sizeof(compat));
+        qemu_fdt_setprop_strings(vms->fdt, "/timer", "compatible",
+                                 "arm,armv8-timer\0arm,armv7-timer\0");
     } else {
         qemu_fdt_setprop_string(vms->fdt, "/timer", "compatible",
                                 "arm,armv7-timer");
-- 
2.21.0

On Fri, 8 Nov 2019 at 19:48, Palmer Dabbelt <palmer@dabbelt.com> wrote:
>
> This new helper function encodes the idiom used by the ARM virt board to
> set a string array.  I don't currently have a working ARM userspace, so I haven't tested
> this, but I made the helper function because I wanted to use it for the
> RISC-V virt board where I have tested it.
>
> Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
> ---
>  hw/arm/virt.c | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> index d4bedc2607..4dc00f54d5 100644
> --- a/hw/arm/virt.c
> +++ b/hw/arm/virt.c
> @@ -304,9 +304,8 @@ static void fdt_add_timer_nodes(const VirtMachineState *vms)
>
>      armcpu = ARM_CPU(qemu_get_cpu(0));
>      if (arm_feature(&armcpu->env, ARM_FEATURE_V8)) {
> -        const char compat[] = "arm,armv8-timer\0arm,armv7-timer";
> -        qemu_fdt_setprop(vms->fdt, "/timer", "compatible",
> -                         compat, sizeof(compat));
> +        qemu_fdt_setprop_strings(vms->fdt, "/timer", "compatible",
> +                                 "arm,armv8-timer\0arm,armv7-timer\0");
>      } else {


This seems to be changing the property we put in -- in
the old code it is 'foo\0bar\0', but in the new code
there will end up being two \0 at the end: 'foo\0bar\0\0'.

thanks
-- PMM

On Sat, Nov 09, 2019 at 03:56:21PM +0000, Peter Maydell wrote:
> On Fri, 8 Nov 2019 at 19:48, Palmer Dabbelt <palmer@dabbelt.com> wrote:
> >
> > This new helper function encodes the idiom used by the ARM virt board to
> > set a string array.  I don't currently have a working ARM userspace, so I haven't tested
> > this, but I made the helper function because I wanted to use it for the
> > RISC-V virt board where I have tested it.
> >
> > Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
> > ---
> >  hw/arm/virt.c | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> >
> > diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> > index d4bedc2607..4dc00f54d5 100644
> > --- a/hw/arm/virt.c
> > +++ b/hw/arm/virt.c
> > @@ -304,9 +304,8 @@ static void fdt_add_timer_nodes(const VirtMachineState *vms)
> >
> >      armcpu = ARM_CPU(qemu_get_cpu(0));
> >      if (arm_feature(&armcpu->env, ARM_FEATURE_V8)) {
> > -        const char compat[] = "arm,armv8-timer\0arm,armv7-timer";
> > -        qemu_fdt_setprop(vms->fdt, "/timer", "compatible",
> > -                         compat, sizeof(compat));
> > +        qemu_fdt_setprop_strings(vms->fdt, "/timer", "compatible",
> > +                                 "arm,armv8-timer\0arm,armv7-timer\0");
> >      } else {
> 
> 
> This seems to be changing the property we put in -- in
> the old code it is 'foo\0bar\0', but in the new code
> there will end up being two \0 at the end: 'foo\0bar\0\0'.

In fact it's not because the setprop_strings() helper just uses the
\0\0 to detect the end, but truncates what it actually puts into the
dtb at the first \0.  But I agree this is confusing enough not to
really be an improvement over the original version.

-- 
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