This is my mistake, I'll do a patch to use the appropriate PRI macro.
On 19/08/2010 15:37, "Tim Deegan" <Tim.Deegan@xxxxxxxxxx> wrote:
> At 15:13 +0100 on 19 Aug (1282230825), Ian Jackson wrote:
>> We'd prefer to avoid PRI macros in xen itself
> I don't think we would; we even define a bunch of our own. :)
>> , so instead we cast the
>> value to long to match the printf format. Without this, the build
>> fails on some compilers.
> This is the wrong fix; on PAE builds the value will be truncated from 64
> bits to 32.
>> Signed-off-by: Ian Jackson <ian.jackson@xxxxxxxxxxxxx>
>> diff -r a05a2d559931 xen/common/timer.c
>> --- a/xen/common/timer.c Thu Aug 19 14:46:32 2010 +0100
>> +++ b/xen/common/timer.c Thu Aug 19 15:12:36 2010 +0100
>> @@ -500,7 +500,7 @@ static void dump_timer(struct timer *t,
>> static void dump_timer(struct timer *t, s_time_t now)
>> printk(" ex=%8ldus timer=%p cb=%p(%p)",
>> - (t->expires - now) / 1000, t, t->function, t->data);
>> + (long)((t->expires - now) / 1000), t, t->function, t->data);
>> print_symbol(" %s\n", (unsigned long)t->function);
>> Xen-devel mailing list
Xen-devel mailing list