This is the mail archive of the cygwin mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: How to correctly rebase?

Dr Rainer Woitok writes:
> I locally built a product from source (Gdal), and while running the test
> programs, I got
>    python2.7 8800 child_info_fork::abort: unable to remap _gdalconst.dll
>    to same address as parent (0x260000) - try running rebaseall

This address is really low, so either some BLODA interferes with the
loading of the DLL or you're having far too many packages installed than
can possibly coexist on a 32bit installation.

If it's a real-time virus scanner, especially of the heuristical sort,
you need to exempt all installation paths for DLL used by Cygwin from
getting picked up.

> After doing so, I rebooted and tested again, but this time I got
>    python2.7 8136 child_info_fork::abort: address space needed by
>    '_io.dll' (0x370000) is already occupied
>    2324 [main] python2.7 9104 child_info_fork::abort: unable to remap
>    time.dll to same address as parent (0x270000) - try running rebaseall

So the rebase map changed and you're now having the same problem with
different libraries, interestingly enough in about the same address
range.  Unless the image base of those two libraries really now is
0x370000 and 0x270000 (which I doubt, there are many more libraries that
come _after_ those in a full rebase), then that is again hinting at
BLODA.  Do you use any (perhaps commercial) software that is based on
Cygwin or MSys under the hood?  Their DLL will have a high propensity of
interfering with Cygwin unless you include them in the rebase or block
out the range of image addresses that they use.

> PS: Please also reply by personal mail to me,  as I am not subscribed to
> this mailing list.

You can read the answers in the mailing list archive or on 

+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Wavetables for the Terratec KOMPLEXER:

Problem reports:
Unsubscribe info:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]