On 13.06.2011 11:31, Jiri Denemark wrote:
On Thu, Jun 09, 2011 at 16:06:48 +0200, Michal Privoznik wrote:
> On 07.06.2011 15:01, Jiri Denemark wrote:
>> ---
>> src/libvirt_private.syms | 1 +
>> src/util/util.c | 24 ++++++++++++++++++++++++
>> src/util/util.h | 2 ++
>> 3 files changed, 27 insertions(+), 0 deletions(-)
>>
>> diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms
>> index e6ab870..2415a89 100644
>> --- a/src/libvirt_private.syms
>> +++ b/src/libvirt_private.syms
>> @@ -1027,6 +1027,7 @@ virStrToLong_ul;
>> virStrToLong_ull;
>> virStrcpy;
>> virStrncpy;
>> +virTimeMs;
>> virTimestamp;
>> virVasprintf;
>>
>> diff --git a/src/util/util.c b/src/util/util.c
>> index e221abe..d758cf9 100644
>> --- a/src/util/util.c
>> +++ b/src/util/util.c
>> @@ -3117,6 +3117,30 @@ virTimestamp(void)
>> return timestamp;
>> }
>>
>> +#define timeval_to_ms(tv) (((tv).tv_sec * 1000ull) + ((tv).tv_usec / 1000))
> What about making this in util.h so we can have only one definition and
> then cleanup src/qemu/qemu_domain.c:48, src/qemu/qemu_driver.c:116,
> src/qemu/qemu_migration.c:49?
Nope, one of the reasons for virTimeMs is to get rid of timeval_to_ms usage in
those files (which is done by patch 2/6).
Yeah, I realized when I replied to this and opened 2/6.
>> --- a/src/util/util.h
>> +++ b/src/util/util.h
>> @@ -286,6 +286,8 @@ int virBuildPathInternal(char **path, ...)
ATTRIBUTE_SENTINEL;
>>
>> char *virTimestamp(void);
>>
>> +int virTimeMs(unsigned long long *ms);
> I'd add ATTRIBUTE_NONNULL(1); here.
Added and pushed, thanks.
Jirka