Re: [Qemu-devel] [PATCH] tpm_tis: fix format for 64bit variable

[Stefan Berger <stefanb@linux.vnet.ibm.com>]

On 03/31/2015 10:15 AM, Eric Blake wrote:
> On 03/31/2015 08:09 AM, Eric Blake wrote:
>> On 03/31/2015 07:27 AM, Stefan Berger wrote:
>>> Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
>>> ---
>>>   hw/tpm/tpm_tis.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/hw/tpm/tpm_tis.c b/hw/tpm/tpm_tis.c
>>> index 815c8ea..52e0148 100644
>>> --- a/hw/tpm/tpm_tis.c
>>> +++ b/hw/tpm/tpm_tis.c
>>> @@ -814,7 +814,7 @@ static void tpm_tis_mmio_write_intern(void *opaque, hwaddr addr,
>>>               tis->loc[locty].state == TPM_TIS_STATE_COMPLETION) {
>>>               /* drop the byte */
>>>           } else {
>>> -            DPRINTF("tpm_tis: Data to send to TPM: %08x (size=%d)\n",
>>> +            DPRINTF("tpm_tis: Data to send to TPM: %08lx (size=%d)\n",
>> NACK.  When printing uint64_t val, you HAVE to use PRIx64, not lx, for
>> the sake of 32-bit platforms.
> Furthermore, you could have caught this much sooner if you fix the
> definition of DPRINTF.  This file used:
>
> #ifdef DEBUG_TIS
> #define DPRINTF(fmt, ...) \
>      do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0)
> #else
> #define DPRINTF(fmt, ...) \
>      do { } while (0)
> #endif
>
> which is bad, because the compiler CAN'T see the type mismatch unless
> you turn debugging on (alas, we have a LOT of files in the code base
> with similar bad patterns).  A better example is hw/display/cg3.c:
>
> /* Change to 1 to enable debugging */
> #define DEBUG_CG3 0
> ...
> #define DPRINTF(fmt, ...) do { \
>      if (DEBUG_CG3) { \
>          printf("CG3: " fmt , ## __VA_ARGS__); \
>      } \
> } while (0);

I'll move it towards that after 2.3 is out.

      Stefan