1 |
In reverse order... |
2 |
|
3 |
The % means it's a new use flag added to the ebuild. It did not exist when |
4 |
the package was originally emerged. |
5 |
|
6 |
dDo not remove the old gcc (3.3.4) until you've got everything rebuilt with |
7 |
3.4.6. |
8 |
|
9 |
Here's everything you need to know on upgrading the compiler version: |
10 |
|
11 |
http://www.gentoo.org/doc/en/gcc-upgrading.xml |
12 |
|
13 |
|
14 |
dcm |
15 |
|
16 |
On 7/27/06, Mark Knecht <markknecht@×××××.com> wrote: |
17 |
> |
18 |
> Hi, |
19 |
> I have two 15-month old Pundit-R machines that I've been updating. |
20 |
> One machine (myth12) I rebuilt from scratch. It works. On the second |
21 |
> machine (myth14) I went the update path since it also servers as a NFS |
22 |
> server and was hard to take off-line. It is not working. The |
23 |
> overriding problem right now is that the ati-driver does not load |
24 |
> correctly on Myth14. ati-drivers is required as I use the TV output, |
25 |
> not the VGA to drive the TV inputs. |
26 |
> |
27 |
> I've been going through the machines with a fine tooth comb looking |
28 |
> for differences and they are now very close except for a couple of |
29 |
> things that I would very much like to get some info on. Thanks in |
30 |
> advance |
31 |
> |
32 |
> 1) Myth12, which works, only uses gcc-3.4.6-r1. Myth14, which was |
33 |
> updated, has both gcc-3.4.6-r1 as well as gcc-3.3.6. |
34 |
> |
35 |
> 2) Myth12, which works, has libstdc++ installed while Myth14 does not. |
36 |
> |
37 |
> myth12 ~ # eix -I libstdc |
38 |
> * sys-libs/libstdc++-v3 |
39 |
> Available versions: ~3.3.3-r1 3.3.4 ~3.3.6 |
40 |
> Installed: 3.3.4 |
41 |
> Homepage: http://gcc.gnu.org/libstdc++/ |
42 |
> Description: Compatibility package for running binaries |
43 |
> linked against a pre gcc 3.4 libstdc++ |
44 |
> |
45 |
> myth14 ~ # eix -I libstdc |
46 |
> |
47 |
> Found 0 matches |
48 |
> myth14 ~ # |
49 |
> |
50 |
> 3) Myth14 seems to use the older compiler: |
51 |
> |
52 |
> myth14 ~ # gcc-config -l |
53 |
> [1] i686-pc-linux-gnu-3.3.6 * |
54 |
> [2] i686-pc-linux-gnu-3.3.6-hardened |
55 |
> [3] i686-pc-linux-gnu-3.3.6-hardenednopie |
56 |
> [4] i686-pc-linux-gnu-3.3.6-hardenednopiessp |
57 |
> [5] i686-pc-linux-gnu-3.3.6-hardenednossp |
58 |
> [6] i686-pc-linux-gnu-3.4.6 |
59 |
> [7] i686-pc-linux-gnu-3.4.6-hardened |
60 |
> [8] i686-pc-linux-gnu-3.4.6-hardenednopie |
61 |
> [9] i686-pc-linux-gnu-3.4.6-hardenednopiessp |
62 |
> [10] i686-pc-linux-gnu-3.4.6-hardenednossp |
63 |
> myth14 ~ # |
64 |
> |
65 |
> myth12 ~ # gcc-config -l |
66 |
> [1] i686-pc-linux-gnu-3.4.6 * |
67 |
> [2] i686-pc-linux-gnu-3.4.6-hardened |
68 |
> [3] i686-pc-linux-gnu-3.4.6-hardenednopie |
69 |
> [4] i686-pc-linux-gnu-3.4.6-hardenednopiessp |
70 |
> [5] i686-pc-linux-gnu-3.4.6-hardenednossp |
71 |
> myth12 ~ # |
72 |
> |
73 |
> 4) If I diff all the packages that would be installed by emerge -DuN |
74 |
> --emptytree world on both machines I see some differences with '%' in |
75 |
> the way the results are reported. What would cause this? |
76 |
> |
77 |
> myth14 ~ # diff myth12-test-build-list myth14-test-build-list |
78 |
> 38,39c38,39 |
79 |
> < [ebuild N ] sys-libs/libstdc++-v3-3.3.4 USE="nls nptl -build" |
80 |
> 22,784 kB |
81 |
> < [ebuild N ] sys-kernel/linux-headers-2.6.11-r2 USE="-gcc64%" 36,470 |
82 |
> kB |
83 |
> --- |
84 |
> > [ebuild N ] sys-devel/gcc-3.3.6 USE="fortran gtk nls -bootstrap |
85 |
> -boundschecking -build -doc -gcj -hardened -ip28 -ip32r10k -multislot -nocxx |
86 |
> -nopie -nossp -objc -vanilla" 23,522 kB |
87 |
> > [ebuild N ] sys-kernel/linux-headers-2.6.11-r2 USE="-gcc64" 36,470 |
88 |
> kB |
89 |
> 52c52 |
90 |
> < [ebuild N ] dev-perl/Locale-gettext-1.05 USE="perl% -minimal%" 7 kB |
91 |
> --- |
92 |
> > [ebuild N ] dev-perl/Locale-gettext-1.05 USE="perl -minimal" 7 kB |
93 |
> 219a220,221 |
94 |
> > [ebuild N ] net-nds/portmap-5b-r9 USE="tcpd" 18 kB |
95 |
> > [ebuild N ] net-fs/nfs-utils-1.0.6-r6 USE="tcpd" 259 kB |
96 |
> 285c287 |
97 |
> < Total size of downloads: 429,413 kB |
98 |
> --- |
99 |
> > Total size of downloads: 430,429 kB |
100 |
> myth14 ~ # |
101 |
> |
102 |
> QUESTION: How would I make Myth14 work like Myth12? |
103 |
> |
104 |
> 1) Can I safely emerge libstdc++ on Myth14, use gcc-config to twitch |
105 |
> to the new compiler, and then emerge -C gcc-3.3.4? Will Myth14 still |
106 |
> function? |
107 |
> |
108 |
> 2) What is causing the differences in the output of emerge with the |
109 |
> '%' on someuse flags? How can I make those identical? |
110 |
> |
111 |
> What else might I need to do? |
112 |
> |
113 |
> Thanks, |
114 |
> Mark |
115 |
> -- |
116 |
> gentoo-user@g.o mailing list |
117 |
> |
118 |
> |