In message <[EMAIL PROTECTED]>, Richard Bass
<[EMAIL PROTECTED]> writes
>robert w hall wrote:
>>
>> I keep seeing occasional reports that 'wine does not run... on a
>> win4lin-enabled kernel'. Though this is contrary to my experience and
>> sounds like finger-trouble, I wonder if anyone has any better/hard
>> info?.
>>
>> (The one possibility for serious interaction would appear to be the
>> dosmod/vm86 area? - which I don't personally use)
>
>There was a problem with some versions of Wine with win4lin kernels.
>This was related to the DOSEMU/DPMI problem. The source of the bug
>was the kernel patches, which introduced an unnecessary incompatibility
>in the GDT. Most applications don't care, but dosemu and wine seemed
>to care.
Following Alexandre's reply to my similar post in wine-devel (which I've
posted to this group earlier), I now understand what you're at here. ie
the changed contents of the kernel header file (include/asm/segment.h?),
and in particular USER_CS, USER_DS.
This is going to be a double whammy of course, because people installing
stock kernels from rpm don't get the header file updated!
To avoid it you need to compile BOTH your kernel and wine from source I
think.
(Fortunate indeed are we users of Slack and other 'non-mainstream
distributions)
Bob
PS Question to Richard - I note that there are also changes to
KERNEL_CS, KERNEL_DS (??? I don't read diffs with conditional compiles
too well at gone midnight!) These would scupper loadlin of course....
(unless recompiled!)
>
--
robert w hall
_______________________________________________
Win4Lin-users mailing list
[EMAIL PROTECTED]
http://lists.trelos.com/mailman/listinfo/win4lin-users
