[nflug] Destructive Makefile???

[Roelant Ossewaarde]

Well, yes, I can confirm that 2.6.8 is, what shall we say, not quite  
production-ready :) I needed debian for something, and I had an old  
DVD with this kernel laying around, that's why I am using this  
kernel. My mistake.

I can guarantee that nothing works as it should with this kernel. As  
per the common linux-strategy: there are multiple ways of doing  
things, and they are all broken in this kernel.

I just wanted my network card to work. Ndiswrapper doesn't want to  
build. The two device driver packages that are supposed to support  
the card show the destructive behavior I emailed about.

Tonight, 2.6.8 is going byebye. I hope to have more luck then.

On Feb 28, 2008, at 7:35 AM, Cyber Source wrote:

> The first thing that screams out is, 2.6.8* kernel??!!
> I'm far from knowing alot about kernel stuff but that does seem to  
> be the common denominator in your problems and I do remember that  
> that particular kernel was fraught with problems, especially in the  
> 4k vs 8k thing and would have give fits to build processes. Why are  
> you running such an ancient, known problem series, kernel?
>
> Roelant Ossewaarde wrote:
>> Hi all,
>>
>> I have a problem that has me scratching my head in disbelief.
>>
>> Here's the scenario: I'm trying to build a RTL-8180 driver for my  
>> wireless card. I've installed linux-sources (I'm running a  
>> 2.6.8-18 kernel).
>>
>> I should say I've never had to build drivers before; it doesn't  
>> look hard, but for some reason I get an error message.
>>
>> Here's the funky part: `make' will actually REMOVE all of my  
>> source files when it encounters the error:
>>
>> miep at debian:~$ ls rtl8180-0.21
>> AUTHORS ieee80211_crypt.h ieee80211_rx.c Makefile module_unload24  
>> r8180_gct.h r8180_pm.c r8180_wx.h
>> CHANGES ieee80211_crypt_wep.c ieee80211_tx.c Makefile26  
>> r8180_93cx6.c r8180.h r8180_pm.h README
>> compat24.h ieee802_11.h ieee80211_wx.c module_load r8180_93cx6.h  
>> r8180_hw.h r8180_sa2400.c README.adhoc
>> COPYING ieee80211.h INSTALL module_load24 r8180_core.c  
>> r8180_max2820.c r8180_sa2400.h README.master
>> ieee80211_crypt.c ieee80211_module.c LICENSE module_unload  
>> r8180_gct.c r8180_max2820.h r8180_wx.c
>> miep at debian:~$ cd rtl8180-0.21/
>> miep at debian:~/rtl8180-0.21$ make
>> make -C /lib/modules/2.6.18-5-486/build SUBDIRS=/home/miep/ 
>> rtl8180-0.21 MODVERDIR=/home/miep/rtl8180-0.21 modules
>> make[1]: Entering directory `/usr/src/linux-source-2.6.18'
>>
>> WARNING: Symbol version dump /usr/src/linux-source-2.6.18/ 
>> Module.symvers
>> is missing; modules will have no dependencies and modversions.
>>
>> scripts/Makefile.build:17: /home/miep/rtl8180-0.21/Makefile: No  
>> such file or directory
>> make[2]: *** No rule to make target `/home/miep/rtl8180-0.21/ 
>> Makefile'. Stop.
>> make[1]: *** [_module_/home/miep/rtl8180-0.21] Error 2
>> make[1]: Leaving directory `/usr/src/linux-source-2.6.18'
>> make: *** [2.6] Error 2
>> miep at debian:~/rtl8180-0.21$ ls
>> miep at debian:~/rtl8180-0.21$
>>
>> Note the last command: the directory is now EMPTY.
>>
>> This has nothing to do with the driver itself; before I installed  
>> the kernel sources, I got the same error, but no removal of files.  
>> Also, the same behavior occurs when I'm trying to build a  
>> different driver, in a different directory. I have the feeling  
>> something within the kernel build system does this, but I cannot  
>> believe that anyone would ever code such a destructive procedure.
>>
>> I have no clue what's happening here. Anyone else?
>>
>> Thanks,
>>
>> Roelant.
>>
>> _______________________________________________
>> nflug mailing list
>> nflug at nflug.org
>> http://www.nflug.org/mailman/listinfo/nflug
> _______________________________________________
> nflug mailing list
> nflug at nflug.org
> http://www.nflug.org/mailman/listinfo/nflug
>