1 |
Mansour Al Akeel <mansour.alakeel@×××××.com> posted |
2 |
2a21586d0904121007y7c8d5862r5e6a883600076a4d@××××××××××.com, excerpted |
3 |
below, on Sun, 12 Apr 2009 14:07:50 -0300: |
4 |
|
5 |
> Still nothing is working for me. Here's my make.conf |
6 |
> |
7 |
> CFLAGS="-march=athlon64 -Os -pipe" |
8 |
> CXXFLAGS="${CFLAGS}" |
9 |
> CHOST="x86_64-pc-linux-gnu" |
10 |
> MAKEOPT="-j3" |
11 |
> #FEATURES="ccache" |
12 |
> #FEATURES="-sandbox emerge sandbox" |
13 |
|
14 |
This isn't related to the problem you're having ATM, but that second |
15 |
commented features line is evidence that you misunderstood someone's |
16 |
instructions at some point when you were having problems with portage/ |
17 |
sandbox. |
18 |
|
19 |
What they meant was to put on the command line |
20 |
|
21 |
FEATURES="-sandbox" emerge sandbox |
22 |
|
23 |
What that does when put all on one line as a command is assign a value |
24 |
"-sandbox" to the FEATURES variable for ONLY the emerge sandbox command. |
25 |
Obviously, you misunderstood and put it in your make.conf. |
26 |
|
27 |
Since it's commented it won't hurt anything, but you might as well delete |
28 |
it. |
29 |
|
30 |
> USE="X acpi aiglx alsa aoss apache2 asf bzip2 cairo d fam firefox gdm |
31 |
> gif glitz gphoto2 gstreamer gtk imap jpeg json laptop mad mng mp3 mp4 |
32 |
> mpeg mysql nptl nptlonly ntfs nvidia opengl pdf png pop pop3d postfix |
33 |
> quicktime raw remote samba scp screen scsh smime smp smtp spell sqlite |
34 |
> svg symlink syslog threads tiff tk truetype unicode x264 xcb xcomposite |
35 |
> xine xml xorg xscreensaver xterm-color xv xvid -arts -dbus -eds -esd |
36 |
> -evo -games -ipv6 -kde -langpacks -mono -nautilus -nspr -nss -php |
37 |
> -qt3 -qt4 -servletapi -sesame2 -session -gnome hal" |
38 |
> |
39 |
> VIDEO_CARDS="nv vesa fbdev nvidia" |
40 |
> INPUT_DEVICES="evdev keyboard mouse synaptics" |
41 |
> #PORTDIR_OVERLAY=/usr/portage-overlay/ |
42 |
|
43 |
I think hal uses dbus too, you if you are using hal you may need dbus |
44 |
enabled to. However, with the Option AllowEmptyInput "False" in xorg |
45 |
below, it shouldn't matter whether hal is enabled or not since it should |
46 |
ignore it. |
47 |
|
48 |
> And I have this at the top of my xorg.conf: |
49 |
|
50 |
> Section "ServerLayout" |
51 |
> Identifier "X.org Configured" |
52 |
> Screen 0 "Screen0" 0 0 |
53 |
> InputDevice "Mouse0" "CorePointer" |
54 |
> InputDevice "Synaptics Touchpad" "CorePointer" |
55 |
> InputDevice "Keyboard0" "CoreKeyboard" |
56 |
> EndSection |
57 |
|
58 |
That section looks reasonable. |
59 |
|
60 |
> Section "ServerFlags" |
61 |
> Option "AutoAddDevices" "true" |
62 |
> Option "AutoEnableDevices" "true" |
63 |
> Option "AllowEmptyInput" "false" |
64 |
> EndSection |
65 |
|
66 |
Since those two auto* entries default to true, you shouldn't need them. |
67 |
They just add extra stuff that doesn't do anything. But since you do NOT |
68 |
want the hal autodetect stuff, you'll want to keep the allowemptyinput |
69 |
one as you have it. |
70 |
|
71 |
> Section "Files" |
72 |
> ModulePath "/usr/lib64/xorg/modules" |
73 |
> FontPath "/usr/share/fonts/misc/" |
74 |
> FontPath "/usr/share/fonts/TTF/" |
75 |
> FontPath "/usr/share/fonts/OTF" |
76 |
> FontPath "/usr/share/fonts/Type1/" |
77 |
> FontPath "/usr/share/fonts/100dpi/" |
78 |
> FontPath "/usr/share/fonts/75dpi/" |
79 |
> EndSection |
80 |
|
81 |
Again, these are the defaults, so you should be able to simply delete |
82 |
this entire section. |
83 |
|
84 |
You said the "top" of your xorg.conf. You do still have the InputDevice |
85 |
sections as you originally posted, right? You could delete them *IF* you |
86 |
were using hal, but since you have it turned of, you need to have them. |
87 |
I'm guessing you do, but am just making sure. |
88 |
|
89 |
> My kernel has the event inteface support. I was not able to properly |
90 |
> install HAL, and this is the only possibility of the problems, I am |
91 |
> getting a file colisiont problem. Since I didn't have hal enabled (I |
92 |
> don't see why I need it), I did enable it now, and emerge update. |
93 |
> This is what I get at the end, then it stops there "emerge --update |
94 |
> --newuse --deep world" |
95 |
> |
96 |
> * /etc/udev/rules.d/90-hal.rules |
97 |
> * /sbin/umount.hal |
98 |
> * /lib/udev/hal_unmount |
99 |
> * /usr/lib64/libhal.so.1 |
100 |
> * /usr/lib64/libhal.so |
101 |
> * /usr/lib64/libhal-storage.so.1 |
102 |
> * /usr/lib64/libhal-storage.so |
103 |
> * /usr/lib64/hal/scripts/linux/hal-system-wol-enable-linux |
104 |
> * /usr/lib64/hal/scripts/linux/hal-system-wol-enabled-linux |
105 |
> * /usr/lib64/hal/scripts/linux/hal-system-wol-supported-linux |
106 |
> * |
107 |
> * Searching all installed packages for file collisions... |
108 |
> * |
109 |
> * Press Ctrl-C to Stop |
110 |
> * |
111 |
> * None of the installed packages claim the file(s). |
112 |
> * |
113 |
> * Package 'sys-apps/hal-0.5.11-r8' merged despite file collisions. If |
114 |
> * necessary, refer to your elog messages for the whole content of the |
115 |
> * above message. |
116 |
> |
117 |
> Any idea ? |
118 |
|
119 |
If you notice, it says the package was merged despite the collisions. It |
120 |
was just a warning. If that's at the end, it's just the replay of all |
121 |
the messages at the end of the merge. Either everything else was fine |
122 |
and it all merged and that's the end, or something else caused an error |
123 |
somewhere and that's just at the end as portage now replays all the |
124 |
messages other packages spit out earlier when it finishes, regardless of |
125 |
whether it finished successfully or not. However, if it failed, the |
126 |
failed package and its error message should always be printed last. |
127 |
|
128 |
So that doesn't seem to give /me/ any hints of the problem, at least |
129 |
(unless you had indeed deleted the InputDevice Sections). |
130 |
|
131 |
|
132 |
Oh! That shared memory error from your first post! |
133 |
|
134 |
I don't know what the synaptics driver requires or doesn't require or |
135 |
anything about anything special it needs in xorg.conf including that |
136 |
SHMConfig option, so I've no idea what that's about in regard to |
137 |
Synaptics. |
138 |
|
139 |
HOWEVER, I know what it refers to in general! Perhaps that's the problem? |
140 |
|
141 |
SHM is the former name of the tmpfs filesystem time in the kernel, so if |
142 |
it's enabled for your synaptics driver, make sure you have tmpfs enabled |
143 |
in your kernel. That is, let's see, File systems > Pseudo filesystems > |
144 |
Virtual memory file system support (formerly shm fs). The config symbol |
145 |
for it (under help) is CONFIG_TMPFS. |
146 |
|
147 |
Once you've verified you're running a kernel with tmpfs, you also need an |
148 |
fstab entry for the LSB standard location one, /dev/shm. Here's the |
149 |
fstab entry I have: |
150 |
|
151 |
shm /dev/shm tmpfs size=20m,noexec,nodev,nosuid 0 0 |
152 |
|
153 |
If you wish you can simply use defaults where I have the fancy options, |
154 |
as so: |
155 |
|
156 |
shm /dev/shm tmpfs defaults 0 0 |
157 |
|
158 |
Then (as root) run mount /dev/shm, and it should mount |
159 |
|
160 |
You can verify what's mounted by running mount with no parameters. Once |
161 |
it says you have a tmpfs mounted at /dev/shm, try X again. Maybe that'll |
162 |
do it. |
163 |
|
164 |
If it still doesn't work, try posting your xorg.0.log file (in /var/log) |
165 |
and hopefully someone can help you decipher it. |
166 |
|
167 |
-- |
168 |
Duncan - List replies preferred. No HTML msgs. |
169 |
"Every nonfree program has a lord, a master -- |
170 |
and if you use the program, he is your master." Richard Stallman |