doc/en/nvidia-guide.xml

<?xml version='1.0' encoding="UTF-8"?>
<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/nvidia-guide.xml,v 1.11 2005/02/25 09:25:33 swift Exp $ -->
<!DOCTYPE guide SYSTEM "/dtd/guide.dtd">

<guide link="/doc/en/nvidia-guide.xml">
<title>Gentoo Linux nVidia Guide</title>

<author title="Author">
  <mail link="swift@gentoo.org">Sven Vermeulen</mail>
</author>

<abstract>
Many Gentooists have an nVidia chipset on their system. nVidia provides specific
Linux drivers to boost the performance of your card. This guide informs you how
to install and configure these drivers.
</abstract>

<license/>

<version>1.9</version>
<date>2005-04-09</date>

<chapter>
<title>Configuring your Card</title>
<section>
<title>Installing the Appropriate Drivers</title>
<body>

<p>
The nVidia drivers include kernel modules that must integrate in your current
kernel. To accomplish this, your kernel must support the loading of kernel
modules. If you used <c>genkernel</c> to configure the kernel for you then
you're all set. If not, double check your kernel configuration so that this
support is enabled:
</p>

<pre caption="Enabling the Loading of Kernel Modules">
Loadable module support ---&gt;
  [*] Enable loadable module support
</pre>

<p>
You also need to enable <e>Memory Type Range Register</e> in your kernel:
</p>

<pre caption="Enabling MTRR">
Processor and Features ---&gt;
  [*] MTRR (Memory Type Range Register) support
</pre>

<p>
nVidia's modules and libraries are combined in two packages: <c>nvidia-glx</c> 
and <c>nvidia-kernel</c>. The former are the X11 GLX libraries while the latter 
are the kernel modules. Since <c>nvidia-glx</c> depends on <c>nvidia-kernel</c>,
installing <c>nvidia-glx</c> is sufficient:
</p>

<pre caption="Installing the nVidia modules">
# <i>emerge nvidia-glx</i>
</pre>

<p>
Once the installation has finished, run <c>modprobe nvidia</c> to load the
kernel module into memory.
</p>

<pre caption="Loading the kernel module">
# <i>modprobe nvidia</i>
</pre>

<p>
You probably want to have this done each time you boot your system, so edit
<path>/etc/modules.autoload.d/kernel-2.6</path> (or <path>kernel-2.4</path>) and
add <c>nvidia</c> to it. Don't forget to run <c>modules-update</c> afterwards!
</p>

<pre caption="Running modules-update">
# <i>modules-update</i>
</pre>

</body>
</section>
<section>
<title>Configuring the X Server</title>
<body>

<p>
Once the appropriate drivers are installed you need to configure your X Server
(XFree86 or Xorg) to use the <c>nvidia</c> driver instead of the default
<c>nv</c> driver.
</p>

<p>
Open <path>/etc/X11/XF86Config</path> (or <path>/etc/X11/xorg.conf</path>) with
your favorite editor (such as <c>nano</c> or <c>vim</c>) and go to the
<c>Device</c> section. In that section, change the <c>Driver</c> line:
</p>

<pre caption="Changing nv to nvidia in the X Server configuration">
Section "Device"
  Identifier "nVidia Inc. GeForce2"
  <i>Driver     "nvidia"</i>
  VideoRam   65536
EndSection
</pre>

<p>
Then go to the <c>Module</c> section and make sure the <c>glx</c> module gets
loaded while the <c>dri</c> module doesn't:
</p>

<pre caption="Updating the Module section">
Section "Module"
  <comment>(...)</comment>
  <i># Load  "dri"
  Load  "glx"</i>
  <comment>(...)</comment>
EndSection
</pre>

<p>
Next, in section <c>Screen</c>, make sure that either the <c>DefaultDepth</c>
directive is set to 16 or 24, or that you only have <c>Display</c> subsections 
with <c>Depth</c> settings of 16 or 24. Without it, the nvidia-glx extensions 
will not start.
</p>

<pre caption="Updating the Screen section">
Section "Screen"
  <comment>(...)</comment>
  <i>DefaultDepth 16</i>
  Subsection "Display"
  <comment>(...)</comment>
EndSection
</pre>

<p>
Run <c>opengl-update</c> so that the X Server uses the nVidia GLX libraries:
</p>

<pre caption="Running opengl-update">
# <i>opengl-update nvidia</i>
</pre>

</body>
</section>
<section>
<title>Adding your Users to the video Group</title>
<body>

<p>
You have to add your user to the <c>video</c> group so he has access to the
nvidia device files:
</p>

<pre caption="Adding your user to the video group">
# <i>gpasswd -a youruser video</i>
</pre>

<p>
This might not be totally necessary if you aren't used <c>udev</c> but it
doesn't hurt either and makes your system future-proof :)
</p>

</body>
</section>
<section>
<title>Testing your Card</title>
<body>

<p>
To test your nVidia card, fire up X and run the <c>glxinfo | grep direct</c>
command. It should say that direct rendering is activated:
</p>

<pre caption="Checking the direct rendering status">
$ <i>glxinfo | grep direct</i>
direct rendering: Yes
</pre>

<p>
To monitor your FPS, run <c>glxgears</c>. 
</p>

</body>
</section>
<section>
<title>Enabling nvidia Support</title>
<body>

<p>
Some tools, such as <c>mplayer</c> and <c>xine-lib</c>, use a local USE flag
called "nvidia" which enables XvMCNVIDIA support, useful when watching high
resolution movies. Add in "nvidia" in your USE variable in
<path>/etc/make.conf</path> or add it as USE flag to <c>media-video/mplayer</c>
and/or <c>media-libs/xine-lib</c> in <path>/etc/portage/package.use</path>.
</p>

<p>
Then, run <c>emerge -uD --newuse world</c> to rebuild the applications that
benefit from the USE flag change.
</p>

</body>
</section>
</chapter>

<chapter>
<title>Troubleshooting</title>
<section>
<title>Getting 2D to work on machines with 4Gb or more memory</title>
<body>

<p>
If you are having troubles with the nVidia 2D acceleration it is likely that you
are unable to set up a write-combining range with MTRR. To verify, check the
contents of <path>/proc/mtrr</path>:
</p>

<pre caption="Checking if you have write-combining enabled">
# <i>cat /proc/mtrr</i>
</pre>

<p>
Every line should contain "write-back" or "write-combining". If you see a line
with "uncachable" in it you will need to change a BIOS setting to fix this.
</p>

<p>
Reboot and enter the BIOS, then find the MTRR settings (probably under "CPU
Settings"). Change the setting from "continuous" to "discrete" and boot back
into Linux. You will now find out that there is no "uncachable" entry anymore
and 2D acceleration now works without any glitches.
</p>

</body>
</section>
<section>
<title>I receive warnings about unsupported 4K stack sizes</title>
<body>

<p>
<c>nvidia-kernel</c> packages older than 1.0.6106 only support kernels using an
8K stack size. More recent kernels (2.6.6 and higher) have support for 4K stack
size's as well. Do not select 4K stack size in your kernel configuration if you
are using such an <c>nvidia-kernel</c> package. You can find this option in the
section <c>Kernel Hacking</c>.
</p>

</body>
</section>
</chapter>

</guide>
1	<?xml version='1.0' encoding="UTF-8"?>
2	<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/nvidia-guide.xml,v 1.11 2005/02/25 09:25:33 swift Exp $ -->
3	<!DOCTYPE guide SYSTEM "/dtd/guide.dtd">
4
5	<guide link="/doc/en/nvidia-guide.xml">
6	<title>Gentoo Linux nVidia Guide</title>
7
8	<author title="Author">
9	<mail link="swift@gentoo.org">Sven Vermeulen</mail>
10	</author>
11
12	<abstract>
13	Many Gentooists have an nVidia chipset on their system. nVidia provides specific
14	Linux drivers to boost the performance of your card. This guide informs you how
15	to install and configure these drivers.
16	</abstract>
17
18	<license/>
19
20	<version>1.9</version>
21	<date>2005-04-09</date>
22
23	<chapter>
24	<title>Configuring your Card</title>
25	<section>
26	<title>Installing the Appropriate Drivers</title>
27	<body>
28
29	<p>
30	The nVidia drivers include kernel modules that must integrate in your current
31	kernel. To accomplish this, your kernel must support the loading of kernel
32	modules. If you used <c>genkernel</c> to configure the kernel for you then
33	you're all set. If not, double check your kernel configuration so that this
34	support is enabled:
35	</p>
36
37	<pre caption="Enabling the Loading of Kernel Modules">
38	Loadable module support --->
39	[*] Enable loadable module support
40	</pre>
41
42	<p>
43	You also need to enable <e>Memory Type Range Register</e> in your kernel:
44	</p>
45
46	<pre caption="Enabling MTRR">
47	Processor and Features --->
48	[*] MTRR (Memory Type Range Register) support
49	</pre>
50
51	<p>
52	nVidia's modules and libraries are combined in two packages: <c>nvidia-glx</c>
53	and <c>nvidia-kernel</c>. The former are the X11 GLX libraries while the latter
54	are the kernel modules. Since <c>nvidia-glx</c> depends on <c>nvidia-kernel</c>,
55	installing <c>nvidia-glx</c> is sufficient:
56	</p>
57
58	<pre caption="Installing the nVidia modules">
59	# <i>emerge nvidia-glx</i>
60	</pre>
61
62	<p>
63	Once the installation has finished, run <c>modprobe nvidia</c> to load the
64	kernel module into memory.
65	</p>
66
67	<pre caption="Loading the kernel module">
68	# <i>modprobe nvidia</i>
69	</pre>
70
71	<p>
72	You probably want to have this done each time you boot your system, so edit
73	<path>/etc/modules.autoload.d/kernel-2.6</path> (or <path>kernel-2.4</path>) and
74	add <c>nvidia</c> to it. Don't forget to run <c>modules-update</c> afterwards!
75	</p>
76
77	<pre caption="Running modules-update">
78	# <i>modules-update</i>
79	</pre>
80
81	</body>
82	</section>
83	<section>
84	<title>Configuring the X Server</title>
85	<body>
86
87	<p>
88	Once the appropriate drivers are installed you need to configure your X Server
89	(XFree86 or Xorg) to use the <c>nvidia</c> driver instead of the default
90	<c>nv</c> driver.
91	</p>
92
93	<p>
94	Open <path>/etc/X11/XF86Config</path> (or <path>/etc/X11/xorg.conf</path>) with
95	your favorite editor (such as <c>nano</c> or <c>vim</c>) and go to the
96	<c>Device</c> section. In that section, change the <c>Driver</c> line:
97	</p>
98
99	<pre caption="Changing nv to nvidia in the X Server configuration">
100	Section "Device"
101	Identifier "nVidia Inc. GeForce2"
102	<i>Driver "nvidia"</i>
103	VideoRam 65536
104	EndSection
105	</pre>
106
107	<p>
108	Then go to the <c>Module</c> section and make sure the <c>glx</c> module gets
109	loaded while the <c>dri</c> module doesn't:
110	</p>
111
112	<pre caption="Updating the Module section">
113	Section "Module"
114	<comment>(...)</comment>
115	<i># Load "dri"
116	Load "glx"</i>
117	<comment>(...)</comment>
118	EndSection
119	</pre>
120
121	<p>
122	Next, in section <c>Screen</c>, make sure that either the <c>DefaultDepth</c>
123	directive is set to 16 or 24, or that you only have <c>Display</c> subsections
124	with <c>Depth</c> settings of 16 or 24. Without it, the nvidia-glx extensions
125	will not start.
126	</p>
127
128	<pre caption="Updating the Screen section">
129	Section "Screen"
130	<comment>(...)</comment>
131	<i>DefaultDepth 16</i>
132	Subsection "Display"
133	<comment>(...)</comment>
134	EndSection
135	</pre>
136
137	<p>
138	Run <c>opengl-update</c> so that the X Server uses the nVidia GLX libraries:
139	</p>
140
141	<pre caption="Running opengl-update">
142	# <i>opengl-update nvidia</i>
143	</pre>
144
145	</body>
146	</section>
147	<section>
148	<title>Adding your Users to the video Group</title>
149	<body>
150
151	<p>
152	You have to add your user to the <c>video</c> group so he has access to the
153	nvidia device files:
154	</p>
155
156	<pre caption="Adding your user to the video group">
157	# <i>gpasswd -a youruser video</i>
158	</pre>
159
160	<p>
161	This might not be totally necessary if you aren't used <c>udev</c> but it
162	doesn't hurt either and makes your system future-proof :)
163	</p>
164
165	</body>
166	</section>
167	<section>
168	<title>Testing your Card</title>
169	<body>
170
171	<p>
172	To test your nVidia card, fire up X and run the <c>glxinfo \| grep direct</c>
173	command. It should say that direct rendering is activated:
174	</p>
175
176	<pre caption="Checking the direct rendering status">
177	$ <i>glxinfo \| grep direct</i>
178	direct rendering: Yes
179	</pre>
180
181	<p>
182	To monitor your FPS, run <c>glxgears</c>.
183	</p>
184
185	</body>
186	</section>
187	<section>
188	<title>Enabling nvidia Support</title>
189	<body>
190
191	<p>
192	Some tools, such as <c>mplayer</c> and <c>xine-lib</c>, use a local USE flag
193	called "nvidia" which enables XvMCNVIDIA support, useful when watching high
194	resolution movies. Add in "nvidia" in your USE variable in
195	<path>/etc/make.conf</path> or add it as USE flag to <c>media-video/mplayer</c>
196	and/or <c>media-libs/xine-lib</c> in <path>/etc/portage/package.use</path>.
197	</p>
198
199	<p>
200	Then, run <c>emerge -uD --newuse world</c> to rebuild the applications that
201	benefit from the USE flag change.
202	</p>
203
204	</body>
205	</section>
206	</chapter>
207
208	<chapter>
209	<title>Troubleshooting</title>
210	<section>
211	<title>Getting 2D to work on machines with 4Gb or more memory</title>
212	<body>
213
214	<p>
215	If you are having troubles with the nVidia 2D acceleration it is likely that you
216	are unable to set up a write-combining range with MTRR. To verify, check the
217	contents of <path>/proc/mtrr</path>:
218	</p>
219
220	<pre caption="Checking if you have write-combining enabled">
221	# <i>cat /proc/mtrr</i>
222	</pre>
223
224	<p>
225	Every line should contain "write-back" or "write-combining". If you see a line
226	with "uncachable" in it you will need to change a BIOS setting to fix this.
227	</p>
228
229	<p>
230	Reboot and enter the BIOS, then find the MTRR settings (probably under "CPU
231	Settings"). Change the setting from "continuous" to "discrete" and boot back
232	into Linux. You will now find out that there is no "uncachable" entry anymore
233	and 2D acceleration now works without any glitches.
234	</p>
235
236	</body>
237	</section>
238	<section>
239	<title>I receive warnings about unsupported 4K stack sizes</title>
240	<body>
241
242	<p>
243	<c>nvidia-kernel</c> packages older than 1.0.6106 only support kernels using an
244	8K stack size. More recent kernels (2.6.6 and higher) have support for 4K stack
245	size's as well. Do not select 4K stack size in your kernel configuration if you
246	are using such an <c>nvidia-kernel</c> package. You can find this option in the
247	section <c>Kernel Hacking</c>.
248	</p>
249
250	</body>
251	</section>
252	</chapter>
253
254	</guide>