RE: y2038 branch

["Steve Hay" <SteveHay@planit.com>]

Michael G Schwern wrote:
> Steve Hay wrote:
>> I've made a couple of minor corrections for BCC and VC9 (#34449): the
>> (current) Configure test programs show that the ranges are slightly
>> different for LOCALTIME compared to GMTIME for those compilers.
>> 
>> I've also found that the programs break using VC8/VC9 (where time_t
>> has size 8) if I include <windows.h> because that defines MINLONG and
>> MAXLONG (as -2147483648 and 2147483647 respectively): I end up with
>> results like 
>> 
>> sGMTIME_min 2147483648 [sic]
>> sGMTIME_max 2147483647
>> 
>> instead of
>> 
>> sGMTIME_min 0
>> sGMTIME_max 32535244799
>> 
>> (I see there are discussions on-going about changing these programs
>> anyway, so this may not be relevant in due course.)
> 
> Give
> http://code.google.com/p/y2038/source/browse/trunk/bin/check_max.c a
> shot please? 

That gives different results for each of my compilers, and I'm losing
sight of what results I'm expecting now.

Here's what the current Configure programs give me:

VC6, VC7, GCC, BCC(localtime): 0 -> 2147483647
BCC(gmtime): -2147483648 -> 2147483647
VC8, VC9(localtime): 0 -> 32535244799
VC9(gmtime): -43200 -> 32535291599

and here's what your program gives me:

VC6, VC7, GCC, BCC(localtime): -2147483648 -> 2147483647
BCC(gmtime): -1073741824 -> 2147483647
VC8, VC9: 0 -> 9223372036854775800

---

Follow-Ups from:

Michael G Schwern <schwern@pobox.com>
Michael G Schwern <schwern@pobox.com>

---

[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index][Thread Index][Top&Search][Original]

---