Sufficient 1.2 release criterion: passing all tests on all platforms?

Dan Kegel dank at kegel.com
Sun Mar 8 17:35:42 CDT 2009


On Sun, Mar 8, 2009 at 11:17 PM, Austin English <austinenglish at gmail.com> wrote:
>> For Windows conformance test validation:
>> 1st tier:  Win XP 32 bit, Win 2003 32 bit, Win Vista 32 and 64 bit,
>> Win 2008 32 bit
>> 2nd tier: Win XP 16 bit, Win 95, Win 98, Win ME, Win 7 32 and 64 bit
>> 3rd tier: Win 3.1, DOS
>
> Not sure exactly what you mean by Win XP 16-bit? The Win16 test suite on XP?

Right.  (I forgot to specify, but I would not run the 16 bit tests on any
other version of windows except perhaps tier 3's win 3.1.  Win XP is
probably enough.)

>> For CPUs:
>> 1st tier: whatever our developers use, but mostly < 2 year old Intel
>> and AMD chips, running apps in all three modes, 16, 32, and 64 bit (as
>> supported by hw)
>> 2nd tier: none
>> 3rd tier: power pc, sparc, other less-common pentium-compatible chips
>
> No argument there. Perhaps move 64-bit to 2nd tier, and move it up to
> 1st once we've got better support for it.

We'll see how Alexandre feels about it.

>> For host OS:
>> 1st tier: Linux
>> 2nd tier: Mac OS X
>> 3rd tier: Solaris, FreeBSD
>
> Having tested these often, I'd say OS X is more broken than FreeBSD.
> I'd swap those two around to be honest. Solaris/OpenBSD/NetBSD are
> tier 3 though.

You're probably right, but I'll bet most wine developers would probably
leave FreeBSD in tier 3.  So maybe tier 2 will be empty for this release;
it depends on how strong the support is from FreeBSD advocates.

Also, about +heap: that's probably just a plain old release criterion
for tier one
host os platforms.  That and Valgrind.
- Dan



More information about the wine-devel mailing list