1 |
I have without any changes to the Makefile or anything... just the |
2 |
straight up -r7 ebuild the following: |
3 |
|
4 |
airo.o airo_cs.o hermes.o netwave_cs.o orinoco.o orinoco_cs.o ray_cs.o |
5 |
wavelan_cs.o wvlan_cs.o wvlan_hcf.o wvlan_hcfio.o |
6 |
|
7 |
I have gcc-2.95 on my laptop. |
8 |
|
9 |
-Doug Goldstein |
10 |
|
11 |
Gentoo Laptop Developer http://www.gentoo.org/~cardoe/ |
12 |
|
13 |
-----Original Message----- |
14 |
From: gentoo-dev-admin@g.o [mailto:gentoo-dev-admin@g.o] |
15 |
On Behalf Of Ben Lutgens |
16 |
Sent: Monday, July 15, 2002 11:54 PM |
17 |
To: gentoo-dev@g.o |
18 |
Subject: [gentoo-dev] orinoco drivers not compiling: Guinea Pig Tester |
19 |
Request |
20 |
|
21 |
I'd like to solicit your help in testing something to make sure I'm not |
22 |
losing my mind :-) |
23 |
|
24 |
I was getting this wierd thing where no matter what I did, i was getting |
25 |
no |
26 |
orinoco drivers compiled or installed. A little investigation showed me |
27 |
the |
28 |
following code in the wireless/Makefile in the toplevel of the source |
29 |
tree. |
30 |
|
31 |
ifdef DO_ORINOCO |
32 |
MODULES += hermes.o orinoco.o orinoco_cs.o |
33 |
SRCS += hermes.c orinoco.c orinoco_cs.c |
34 |
endif |
35 |
|
36 |
Thanks to this little nugget the only way to get these .o files to |
37 |
compile |
38 |
is by a -DDO_ORINOCO or commenting the ifdef and endif lines out. Some |
39 |
folks have said that 3.1.34-r7 doesn't have this problem, and they a |
40 |
simple |
41 |
"ebuild pcmcia-cs-3.1.34-r7.ebuild merge" would net you some orinoco |
42 |
drivers. I do not believe this. |
43 |
|
44 |
Here's how to test. |
45 |
|
46 |
cd ${PORTDIR}/sys-apps/pcmcia-cs |
47 |
ebuild pcmcia-cs-3.1.34-r7.ebuild clean unpack compile |
48 |
cd /var/tmp/portage/pcmcia-cs-3.1.34-r7/work/pcmcia-3.1.34/wireless/ |
49 |
ls *.o |
50 |
# you should see (if i'm not smoking something) |
51 |
airo.o netwave_cs.o wavelan_cs.o wvlan_hcf.o |
52 |
airo_cs.o ray_cs.o wvlan_cs.o wvlan_hcfio.o |
53 |
|
54 |
Now, to further complete our test we'll fire up our favorite text editor |
55 |
on |
56 |
the Makefile in the same dir as we're in, and comment out ifdef in line |
57 |
21 |
58 |
and endif in line 24. Now if you run "make" you should see it compile |
59 |
some |
60 |
more modules. And another "ls *.o" and we see: |
61 |
airo.o hermes.o orinoco.o ray_cs.o wvlan_cs.o |
62 |
wvlan_hcfio.o |
63 |
airo_cs.o netwave_cs.o orinoco_cs.o wavelan_cs.o wvlan_hcf.o |
64 |
|
65 |
Now, to make sure that I'm not a complete dolt, please give this test a |
66 |
once over to and post your findings in this thread just to make an old |
67 |
man |
68 |
feel better and that he isn't losing his mind. I could continue to test |
69 |
this with more and more versions of this package but according to some |
70 |
conversations I've had this particular one is supposed to have been |
71 |
proven |
72 |
to "compile and install the orinoco drivers" and since I see no |
73 |
modifications to Makefiles or any any passing of a -DDO_ORINOCO in any |
74 |
patches or sed statements I can only assume that ALL packages using |
75 |
3.1.34 |
76 |
source are suceptible to the same wierdness. Thanks for your time, and |
77 |
may |
78 |
the schwartz be with you. |
79 |
|
80 |
-- |
81 |
Ben Lutgens | |
82 |
http://cvs.gentoo.org/~lamer/ |
83 |
Random Gentoo Developer | http://www.gentoo.org/ |
84 |
|
85 |
"I got a wife and kids too but you don't see me out here stealing |
86 |
Imperial |
87 |
Droids now do ya?" |