[PATCH 1/1] efi_loader: compliance Simple Text Input Ex Protocol

[Simon Glass]

Hi Heinrich,

On Thu, 1 Sept 2022 at 18:36, Heinrich Schuchardt
<heinrich.schuchardt at canonical.com> wrote:
>
>
>
> On 9/2/22 01:52, Simon Glass wrote:
> > On Thu, 1 Sept 2022 at 16:07, Heinrich Schuchardt
> > <heinrich.schuchardt at canonical.com> wrote:
> >>
> >> We cannot expect the buffers passed to the input protocols to be zero
> >> filled. If only modifier keys are pressed, we have to return EFI_NOT_READY
> >> but we still have to fill the key structure.
> >>
> >> Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt at canonical.com>
> >> ---
> >>   lib/efi_loader/efi_console.c | 1 +
> >>   1 file changed, 1 insertion(+)
> >
> > Reviewed-by: Simon Glass <sjg at chromium.org>
> >
> >> diff --git a/lib/efi_loader/efi_console.c b/lib/efi_loader/efi_console.c
> >> index 3164fd484e..1fcaabe1c4 100644
> >> --- a/lib/efi_loader/efi_console.c
> >> +++ b/lib/efi_loader/efi_console.c
> >> @@ -987,6 +987,7 @@ static efi_status_t EFIAPI efi_cin_read_key_stroke_ex(
> >>          efi_cin_check();
> >>
> >>          if (!key_available) {
> >> +               memset(key_data, 0, sizeof(struct efi_key_data));
> >
> > '\0'
>
> memset() wants int here not char.
> (Yes, the compiler would add the missing conversion for '\0'.)

Sure, but it is actually a char, right? It is not good practice to
have function arguments be non-natural types due to the masking the
compiler must then do (increases code size and lower performance). So
that is why memset() uses a natural type.

Regards,
Simon