soc/tegra: Fix terminating condition

Clang warns:

drivers/soc/tegra/common.c:27:16: error: address of array
'match->compatible' will always evaluate to 'true'
[-Werror,-Wpointer-bool-conversion]
        while (match->compatible) {
        ~~~~~  ~~~~~~~^~~~~~~~~~
1 error generated.

Whoops, we have an infinite loop and QEMU no longer boots...

https://travis-ci.com/ClangBuiltLinux/continuous-integration/jobs/160242918

Check that the first character of the string isn't null so that the loop
properly terminates.

Fixes: c57eff9503a5 ("soc/tegra: refactor soc_is_tegra()")
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
---
 drivers/soc/tegra/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

On Thu, Nov 22, 2018 at 3:53 PM Nathan Chancellor
<natechancellor@gmail.com> wrote:
>
> Clang warns:
>
> drivers/soc/tegra/common.c:27:16: error: address of array
> 'match->compatible' will always evaluate to 'true'
> [-Werror,-Wpointer-bool-conversion]
>         while (match->compatible) {
>         ~~~~~  ~~~~~~~^~~~~~~~~~
> 1 error generated.
>
> Whoops, we have an infinite loop and QEMU no longer boots...
>
> https://travis-ci.com/ClangBuiltLinux/continuous-integration/jobs/160242918
>
> Check that the first character of the string isn't null so that the loop
> properly terminates.
>
> Fixes: c57eff9503a5 ("soc/tegra: refactor soc_is_tegra()")
> Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
> ---
>  drivers/soc/tegra/common.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/soc/tegra/common.c b/drivers/soc/tegra/common.c
> index 8a538b968fe9..54627ca957e8 100644
> --- a/drivers/soc/tegra/common.c
> +++ b/drivers/soc/tegra/common.c
> @@ -24,7 +24,7 @@ bool soc_is_tegra(void)
>  {
>         const struct of_device_id *match = tegra_machine_match;
>
> -       while (match->compatible) {
> +       while (match->compatible[0]) {
How about this?
            while (*match->compatible) {
：-）
Acked-by: Yangtao Li <tiny.windzz@gmail.com>
>                 if (of_machine_is_compatible(match->compatible))
>                         return true;
>
> --
> 2.20.0.rc1
>

On Thu, Nov 22, 2018 at 12:52:44AM -0700, Nathan Chancellor wrote:
> Clang warns:
> 
> drivers/soc/tegra/common.c:27:16: error: address of array
> 'match->compatible' will always evaluate to 'true'
> [-Werror,-Wpointer-bool-conversion]
>         while (match->compatible) {
>         ~~~~~  ~~~~~~~^~~~~~~~~~
> 1 error generated.
> 
> Whoops, we have an infinite loop and QEMU no longer boots...
> 
> https://travis-ci.com/ClangBuiltLinux/continuous-integration/jobs/160242918
> 
> Check that the first character of the string isn't null so that the loop
> properly terminates.
> 
> Fixes: c57eff9503a5 ("soc/tegra: refactor soc_is_tegra()")
> Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
> ---
>  drivers/soc/tegra/common.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Wow... that's what I get from not testing this properly. Apologies for
the inconvenience.

Applied, thanks.

Thierry

Hi Thierry:

How about change it to this?
            while (*match->compatible) {
This is more beautiful.    :-)

Thanks,
Yangtao
On Thu, Nov 22, 2018 at 4:50 PM Thierry Reding <thierry.reding@gmail.com> wrote:
>
> On Thu, Nov 22, 2018 at 12:52:44AM -0700, Nathan Chancellor wrote:
> > Clang warns:
> >
> > drivers/soc/tegra/common.c:27:16: error: address of array
> > 'match->compatible' will always evaluate to 'true'
> > [-Werror,-Wpointer-bool-conversion]
> >         while (match->compatible) {
> >         ~~~~~  ~~~~~~~^~~~~~~~~~
> > 1 error generated.
> >
> > Whoops, we have an infinite loop and QEMU no longer boots...
> >
> > https://travis-ci.com/ClangBuiltLinux/continuous-integration/jobs/160242918
> >
> > Check that the first character of the string isn't null so that the loop
> > properly terminates.
> >
> > Fixes: c57eff9503a5 ("soc/tegra: refactor soc_is_tegra()")
> > Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
> > ---
> >  drivers/soc/tegra/common.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
>
> Wow... that's what I get from not testing this properly. Apologies for
> the inconvenience.
>
> Applied, thanks.
>
> Thierry