Re: [Qemu-devel] [PULL 2/2] hw/audio/sb16.c: change dolog() to qemu_log_mask()

[Peter Maydell]

On 2 February 2018 at 07:50, Gerd Hoffmann <address@hidden> wrote:
> From: John Arbuckle <address@hidden>
>
> Changes all the occurrances of dolog() to qemu_log_mask().
>
> Signed-off-by: John Arbuckle <address@hidden>
> Message-id: address@hidden
> Signed-off-by: Gerd Hoffmann <address@hidden>
> ---

> @@ -735,9 +742,8 @@ static void complete (SB16State *s)
>              break;
>
>          case 0x42:              /* FT2 sets output freq with this, go figure 
> */
> -#if 0
> -            dolog ("cmd 0x42 might not do what it think it should\n");
> -#endif
> +            qemu_log_mask(LOG_UNIMP, "cmd 0x42 might not do what it think it"
> +                          " should\n");
>          case 0x41:
>              s->freq = dsp_get_hilo (s);
>              ldebug ("set freq %d\n", s->freq);

Hi. The removal of the #if 0 here means that Coverity reports a
new warning (CID 1385841) about a potential missing break.

The case 0x42 should end either with a "break;" or with a comment
"/* fall through */".

http://homepages.cae.wisc.edu/~brodskye/sb16doc/sb16doc.html#SamplingRate
suggests that "/* fall through */" is correct, since 0x42 is 'set
input sample rate', 0x41 is 'set output sample rate', and supposedly
the two are equivalent on the hardware.

I suspect that may also mean that this log should be a LOG_GUEST_ERROR
rather than LOG_UNIMP (or perhaps not a warning at all, since we
can't tell if the guest really was trying to program the input
frequency).

thanks
-- PMM