On 2015/5/21 11:07, Eric Blake wrote:
> On 05/20/2015 08:54 PM, arei.gong...@huawei.com wrote:
>> From: Gonglei <arei.gong...@huawei.com>
>>
>> before:
>> (qemu) 2015-05-21T02:07:43.695065Z qemu-system-x86_64: test....
>
> Ah, but the trailing Z is UTC,
>
>> after applying this patch:
>> (qemu) [2015-05-21T02:21:46.023619Z UTC] qemu-system-x86_64: test....
>
> so this representation is now redundant for specifying the same timezone
> twice. Basically, 'Z', '+0000', and ' UTC' are synonyms; I don't know
> if any one form is more popular than others, but I don't think we need
> this patch.
>
Yes, I checked it too, please ignore the noise.
Regards,
-Gonglei
>>
>> Cc: Eric Blake <ebl...@redhat.com>
>> Signed-off-by: Gonglei <arei.gong...@huawei.com>
>> ---
>> util/qemu-error.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/util/qemu-error.c b/util/qemu-error.c
>> index 9bba5f5..0b91cbc 100644
>> --- a/util/qemu-error.c
>> +++ b/util/qemu-error.c
>> @@ -212,7 +212,7 @@ void error_vreport(const char *fmt, va_list ap)
>> if (enable_timestamp_msg) {
>> g_get_current_time(&tv);
>> timestr = g_time_val_to_iso8601(&tv);
>
> Since glib already provided a trailing Z, it's actually more work to
> strip it and provide an alternative of ' UTC'; so I'm inclined to NACK
> this patch, unless anyone else has a strong opinion in favor of it.
>
>> - error_printf("%s ", timestr);
>> + error_printf("[%s UTC] ", timestr);
>> g_free(timestr);
>> }
>>
>>
>
