[m-rev.] diff: partial implementation of c_times on win32

[Peter Ross]

On Wed, Oct 8, 2008 at 3:58 PM, Julien Fischer
<juliensf at csse.unimelb.edu.au> wrote:
>
> On Wed, 8 Oct 2008, Peter Ross wrote:
>
>> Hi,
>>
>>
>> ===================================================================
>>
>>
>> Estimated hours taken: 2
>> Branches: main
>>
>> library/time.m:
>>        Partially implement c_times on win32 systems.
>>        We don't implement the child process times,
>>        just those of the current process.
>>
>> Index: library/time.m
>> ===================================================================
>> RCS file: /home/mercury1/repository/mercury/library/time.m,v
>> retrieving revision 1.60
>> diff -u -r1.60 time.m
>> --- library/time.m      13 Aug 2007 03:04:44 -0000      1.60
>> +++ library/time.m      8 Oct 2008 04:22:57 -0000
>> @@ -124,6 +124,8 @@
>>    % On non-POSIX systems that do not support this functionality,
>>    % this procedure may simply always throw an exception.
>>    %
>> +    % Currently on Win32 the child part of 'tms' is always zero.
>> +    %
>> :- pred time.times(tms::out, clock_t::out, io::di, io::uo) is det.
>>
>>    % time.clk_tck:
>> @@ -329,6 +331,16 @@
>>        Result = Ret
>>    ).
>>
>> +:- pragma foreign_decl(c, "
>> +#ifdef _WIN32
>
> The runtime defines the macro MR_WIN32 which should use here instead of
> _WIN32.
>
Done.
--------------------------------------------------------------------------
mercury-reviews mailing list
Post messages to:       mercury-reviews at csse.unimelb.edu.au
Administrative Queries: owner-mercury-reviews at csse.unimelb.edu.au
Subscriptions:          mercury-reviews-request at csse.unimelb.edu.au
--------------------------------------------------------------------------