[pvrusb2] make trace...

[Dhimiter Bozo]

What I did different this time was that I just ran "make" without any
args inside the driver dir. I did some more exploring last night after
I sent that email, and I found that the kernel source does not come
with mythbuntu 9.10. The build dir is pointing to the kernel-header.
I am not in fron of my machine now so I can't verify the paths.
How do I get the kernel source and compile that?
Would it be better if I whipe out mythbuntu and install ubuntu + mythtv?

Thanks,
Dhimiter


On 12/21/09, Mike Isely <isely at isely.net> wrote:
>
> You must have done something different this time.  The error you were
> pointing out before was a complete failure of the make program to
> recognize the "modules" target.  This time you obviously got past that
> because it's compiling things now.  (What did you do?)  Read on...
>
> On Mon, 21 Dec 2009, Dhimiter Bozo wrote:
>
>> hi , i posted earlier about this make problem and i'm not sure how to fix
>> it. below is the trace log when i execute make:
>>
>> dhimiter at laptop:~/Downloads/pvrusb2-mci-20091124/driver$ make
>> make INSTALL_MOD_DIR=pvrusb2 -C /lib/modules/2.6.31-16-generic/build
>> M=/home/dhimiter/Downloads/pvrusb2-mci-20091124/driver
>> CONFIG_VIDEO_PVRUSB2=m CONFIG_VIDEO_PVRUSB2_24XXX=y
>> CONFIG_VIDEO_PVRUSB2_SYSFS=y CONFIG_VIDEO_PVRUSB2_DVB=y
>> CONFIG_VIDEO_PVRUSB2_DEBUGIFC=y CONFIG_VIDEO_ADV_DEBUG=y modules
>> make[1]: Entering directory `/usr/src/linux-headers-2.6.31-16-generic'
>>   CC [M]
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-std.o
>>   CC [M]
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-compat.o
>>   CC [M]
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-ctrl.o
>>   CC [M]
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-hdw.o
>>   CC [M]
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.o
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:42:22:
>> error: tda18271.h: No such file or directory
>
> That missing header is a part of the kernel source tree.  If it isn't
> there then the source tree has to be incomplete.  For a "complete"
> 2.6.31 source tree, tda18271.h can be found in this path (relative to
> the root of the tree):
>
> drivers/media/common/tuners/tda18271.h
>
> Do you see this file anywhere?
>
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:43:21:
>> error: tda8290.h: No such file or directory
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:44:26:
>> error: tuner-simple.h: No such file or directory
>
> Well tuner-simple.h should be in the same place as tda18271.h (and
> probably also tda8290.h though I didn't bother to check that one).  Odds
> are this is the same problem.
>
>    [...]
>
>
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c: In
>> function ‘pvr2_lgh06xf_attach’:
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:208:
>> error: ‘simple_tuner_attach’ undeclared (first use in this function)
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:208:
>> error: (Each undeclared identifier is reported only once
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:208:
>> error: for each function it appears in.)
>> /home/dhimiter/Downloads/pvrusb2-mci-20091124/driver/pvrusb2-devattr.c:208:
>> warning: type defaults to ‘int’ in declaration of ‘__a’
>
>    [...]
>
> The rest of these errors are most likely references to things that those
> missing headers would have defined.
>
> There seems to be a rash here of people having trouble due to
> compilation against source trees which are apparently incomplete.
> Honestly, the build process in the standalone pvrusb2 driver - except
> for the issue last June - has not changed for several years now.  And
> the simple fact is that these particular headers are REQUIRED pieces for
> the build to work.  These headers define important interfaces that the
> driver has to use.  There's just no way around this.  I test using (lots
> of) vanilla kernel trees from kernel.org and as I've said before it's a
> practical impossibility for me to also test all the combinations of what
> other distributions are doing.  It's just too much.  I'm dismayed that
> this is apparently happening but there isn't a lot I can do about it.
> I'm open to suggestions from others.
>
> Perhaps others would like to volunteer effort to solve and support their
> favorite distribution?  I will try to roll such fixes back if I can
> without impacting everyone else.
>
>   -Mike
>
> --
>
> Mike Isely
> isely @ isely (dot) net
> PGP: 03 54 43 4D 75 E5 CC 92 71 16 01 E2 B5 F5 C1 E8