[Rd] wine and build difference between R.2.4.0 and R 2.4.1 windows binaries?
hin-tak.leung at cimr.cam.ac.uk
Fri Jan 12 15:44:08 CET 2007
Duncan Murdoch wrote:
> On 1/11/2007 6:52 AM, Hin-Tak Leung wrote:
>> Duncan Murdoch wrote:
>>> On 1/10/2007 2:29 PM, Hin-Tak Leung wrote:
>>>> Does anybody (most probably the core team) know if there is
>>>> any difference in how the official 2.4.0 and 2.4.1 binaries are
>>>> Problem is, 2.4.0 loads with the wine (I tried a few recent
>>>> versions, and also used 2.3.x under wine from time to time),
>>>> but 2.4.1 won't.
>>> Yes, there were changes to the MinGW run-time library between those
>>> two releases. See http://www.murdoch-sutherland.com/Rtools/.
>>> If you can be more specific about the problem, it's possible the Wine
>>> or MinGW people could fix it.
>> With 2.4.1, "wine Rgui.exe" launches the main Rgui window, then crashes
>> and dropped with debug traces without being able to get to the
>> Rconsole tab. Logically it is a wine "bug", since the same binary
>> runs under genuine windows, but it looks like it is due to change
>> from mingw, so I guess I should involve both parties.
> I'd suggest trying a build with debug information ("make distclean; make
> DEBUG=T"), so that the debug trace has symbol names and line information
> in it. I could easily do such a build of R-patched or R-devel for you
> if you want to try this.
I tried loading R-patched and R2.5dev (they both have a time stamp for
11 Jan early morning):
Neither of them loads under wine. I am inclined to fix wine itself
rather than requiring changes in mingw or R to cope. So as a first
thing I'd like to try build the older 2.4.0 with the new tools (this is
not a request for help - I think I'll try the cross-tools myself) -
because that should eliminate any possibillity that some newly added
code triggers difference between wine and real windows.
I'd like to know one thing though - was there any specific reason for
the mingw upgrade after 2.4.0, or was it just a routine
More information about the R-devel