source: postlfs/config/firmware.xml@ 3ae81e1

10.0 10.1 11.0 11.1 11.2 11.3 12.0 12.1 7.10 7.8 7.9 8.0 8.1 8.2 8.3 8.4 9.0 9.1 basic bdubbs/svn elogind kea ken/TL2024 ken/inkscape-core-mods ken/tuningfonts lazarus lxqt nosym perl-modules plabs/newcss plabs/python-mods python3.11 qt5new rahul/power-profiles-daemon renodr/vulkan-addition trunk upgradedb xry111/intltool xry111/llvm18 xry111/soup3 xry111/test-20220226 xry111/xf86-video-removal
Last change on this file since 3ae81e1 was 3ae81e1, checked in by Bruce Dubbs <bdubbs@…>, 9 years ago

Add section on firmware

git-svn-id: svn://svn.linuxfromscratch.org/BLFS/trunk/BOOK@15851 af4574ff-66df-0310-9fd7-8a98e5e911e0

  • Property mode set to 100644
File size: 22.1 KB
Line 
1<?xml version="1.0" encoding="ISO-8859-1"?>
2<!DOCTYPE sect1 PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
3 "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd" [
4 <!ENTITY % general-entities SYSTEM "../../general.ent">
5 %general-entities;
6]>
7
8<sect1 id="postlfs-firmware" xreflabel="About Firmware">
9 <?dbhtml filename="firmware.html"?>
10
11 <sect1info>
12 <othername>$LastChangedBy: bdubbs $</othername>
13 <date>$Date: 2012-03-13 13:19:34 -0500 (Tue, 13 Mar 2012) $</date>
14 </sect1info>
15
16 <title>About Firmware</title>
17
18 <indexterm zone="postlfs-firmware">
19 <primary sortas="e-lib-firmware">/lib/firmware</primary>
20 </indexterm>
21
22 <para> On some recent PCs it can be necessary, or desirable, to load firmware
23 to make them work at their best. The kernel contains a directory, <filename
24 class="directory">/lib/firmware</filename>, where the kernel or kernel
25 drivers look for firmware images.</para>
26
27 <para>Preparing firmware for multiple different machines, as a distro would
28 do, is outside the scope of this book.</para>
29
30 <para>Currently, most firmware can be found at a <userinput>git</userinput>
31 repository: <ulink
32 url="http://git.kernel.org/cgit/linux/kernel/git/firmware/linux-firmware.git/tree/"/>.
33 For convenience, the LFS Project has created a mirror, updated daily, where
34 these firmware files can be accessed vau <userinput>wget</userinput> or a web
35 browser at <ulink
36 url="http://anduin.linuxfromscratch.org/sources/linux-firmware/"/>.</para>
37
38 <para>To get the firmware either point a browser to one of the above
39 repositories and then download the item(s) which you need, or install
40 <userinput>git</userinput> and clone that repository.</para>
41
42 <para>For some other firmware, particularly for Intel microcode and certain
43 wifi devices, the needed firmware is not available in the above repository.
44 Some of this will be addressed below, but a search of the Internet for needed
45 firmware is sometimes necessary.</para>
46
47 <para>Firmware files are conventionally referred to as blobs because you cannot
48 determine what they will do. Note that firmware is distributed under various
49 different licenses which do not permit disassembly or reverse-engineering.</para>
50
51 <para>Firmware for PCs falls into four categories:</para>
52
53 <itemizedlist spacing="compact">
54 <listitem>
55 <para>Updates to the CPU to work around errata, usually referred to as
56 microcode.</para>
57 </listitem>
58 <listitem>
59 <para>Firmware for video controllers. On x86 machines this seems to only
60 apply to ATI devices : Radeons require firmware to be able to use KMS
61 (kernel modesetting - the preferred option) as well as for Xorg. For
62 earlier radeon chips (before the R600), the firmware is still in the
63 kernel.</para>
64 </listitem>
65 <listitem>
66 <para>Firmware updates for wired network ports. Mostly they work even
67 without the updates, but one must assume that they will work better with
68 the updated firmware.</para>
69 </listitem>
70 <listitem>
71 <para>Firmware for other devices, such as wifi. These devices are not
72 required for the PC to boot, but need the firmware before these devices
73 can be used.</para>
74 </listitem>
75 </itemizedlist>
76
77 <note><para>Although not needed to load a firmware blob, the following
78 tools may be useful for determining, obtaining, or preparing the needed
79 firmware in order to load it into the system:
80 <xref linkend="cpio"/>,
81 <xref linkend="git"/>,
82 <xref linkend="pciutils"/>, and
83 <xref linkend="wget"/></para></note>
84
85 <para condition="html" role="usernotes">User Notes:
86 <ulink url="&blfs-wiki;/aboutfirmware"/></para>
87
88 <sect2 id="cpu-microcode">
89 <title>Microcode updates for CPUs</title>
90
91 <para>In general, microcode can be loaded by the BIOS or UEFI, and it might
92 be updated by upgrading to a newer version of those. On linux, you can also
93 load the microcode from the kernel if you are using an AMD family 10h or
94 later processor (first introduced late 2007), or an intel processor from
95 1998 and later (Pentium4, Core, etc), if updated microcode has been
96 released. These updates only last until the machine is powered off, so they
97 need to be applied on every boot.</para>
98
99 <para>There are two ways of loading the microcode, described as 'early' and
100 'late'. Early loading happens before userspace has been started, late
101 loading happens when userspace has started. Not surprisingly, early loading
102 is preferred, (see e.g. an explanatory comment in a kernel commit noted at
103 <ulink url="https://lwn.net/Articles/530346/">x86/microcode: Early load
104 microcode </ulink> on LWN.) Indeed, it is needed to work around one
105 particular erratum in early intel Haswell processors which had TSX enabled.
106 (See <ulink
107 url="http://www.anandtech.com/show/8376/intel-disables-tsx-instructions-erratum-found-in-haswell-haswelleep-broadwellyi/">
108 Intel Disables TSX Instructions: Erratum Found in Haswell, Haswell-E/EP,
109 Broadwell-Y</ulink>.) Without this update glibc can do the wrong thing in
110 uncommon situations.</para>
111
112 <para>It is much simpler to begin by building a kernel which boots on
113 your hardware, try late microcode loading to see if there is an update (in
114 many cases the BIOS or UEFI will have already applied any update), and then
115 take the extra steps required for early loading.</para>
116
117 <para>This means you will be reconfiguring your kernel if you use early
118 loading, so keep the built source around to minimise what gets rebuilt, and
119 if you are at all uncertain, add your own identifier (A,B, etc) to the end
120 of the EXTRAVERSION in the kernel configuration, e.g. "EXTRAVERSION -A" if
121 nothing was set.</para>
122
123 <para>To confirm what processor(s) you have (if more than one, they will be
124 identical) look in /proc/cpufinfo.</para>
125
126 <sect3 id="intel-microcode">
127 <title>Intel Microcode for the CPU</title>
128
129 <bridgehead renderas="sect4">Required Package</bridgehead>
130 <para role='required'>
131 <ulink url='http://fedorahosted.org/released/microcode_ctl/'/></para>
132
133 <para>For intel CPUs an extra package, microcode_ctl, is required. The
134 package chosen is the version hosted at fedora &mdash; there is an
135 alternative version at github from the same packager, but that still
136 includes a redundant old version of an AMD microcode container, and also
137 requires the unzip package.</para>
138
139 <para>Download the latest version from the link above; when last checked,
140 this was 2.1-7 and is updated when intel releases new microcode.</para>
141
142 <para>This package reformats the microcode supplied by intel into a
143 format which the kernel can apply. The program
144 <userinput>intel-microcode2ucode</userinput> is built and invoked by the
145 Makefile to create the individual firmware blobs, so there is no reason
146 to install it.</para>
147
148 <para>Begin by extracting the tarball and changing to the directory it created.
149 Then change to the source diirectory and run:</para>
150
151<screen><userinput>make</userinput></screen>
152
153 <para>This creates various blobs with names in the form XX-YY-ZZ. Now you
154 need to determine your processor's identity, to see if there is any
155 microcode for it. Determine the decimal values of the cpu family, model
156 and stepping by running:</para>
157
158<screen><userinput>head -n7 /proc/cpuinfo</userinput></screen>
159
160 <para>Now convert the cpu family, model and stepping to pairs of hexadecimal
161 digits. For a SandyBridge i3-2120 (described as Intel(R) Core(TM) i3-2120
162 CPU) the relevant values are cpu family 6, model 42, stepping 7 so in
163 this case the required identification is 06-2a-07. A look at the blobs
164 will show that there is one for this CPU (although it might
165 have already been applied by the BIOS). If there is a blob for your
166 system then test if it will be applied by copying it (replace &lt;XX-YY-ZZ&gt;
167 by the identifier for your machine) and then procede through Late loading
168 of microcode on Intel Processors:</para>
169
170<screen><userinput>mkdir -pv /lib/firmware/intel-ucode
171cp -v &lt;XX-YY-ZZ&gt; /lib/firmware/intel-ucode</userinput></screen>
172
173 <para>Now that the intel microcode has been prepared, use the following
174 options when you configure the kernel to try late loading of the intel
175 microcode:</para>
176
177<screen><literal>Processor type and features ---&gt;
178 &lt;M&gt; CPU microcode loading support [CONFIG_MICROCODE]
179 [*] Intel microcode loading support [CONFIG_MICROCODE_INTEL]</literal></screen>
180
181 <para>After you have successfully booted the new system, use the command
182 <userinput>dmesg | grep microcode</userinput> and study the results to
183 see if the message new patch_level appears. This example from the
184 SandyBridge i3:</para>
185
186<screen><literal>[ 0.059906] perf_event_intel: PEBS disabled due to CPU errata, please upgrade microcode
187[ 2.603083] microcode: CPU0 sig=0x206a7, pf=0x2, revision=0x23
188[ 2.669378] microcode: CPU0 sig=0x206a7, pf=0x2, revision=0x23
189[ 2.669994] microcode: CPU0 updated to revision 0x29, date = 2013-06-12
190[ 2.670069] microcode: CPU1 sig=0x206a7, pf=0x2, revision=0x23
191[ 2.670139] microcode: CPU1 sig=0x206a7, pf=0x2, revision=0x23
192[ 2.670501] microcode: CPU1 updated to revision 0x29, date = 2013-06-12
193[ 2.670509] microcode: CPU2 sig=0x206a7, pf=0x2, revision=0x23
194[ 2.670540] microcode: CPU2 sig=0x206a7, pf=0x2, revision=0x23
195[ 2.670917] microcode: CPU2 updated to revision 0x29, date = 2013-06-12
196[ 2.670955] microcode: CPU3 sig=0x206a7, pf=0x2, revision=0x23
197[ 2.670988] microcode: CPU3 sig=0x206a7, pf=0x2, revision=0x23
198[ 2.671348] microcode: CPU3 updated to revision 0x29, date = 2013-06-12
199[ 2.671356] perf_event_intel: PEBS enabled due to microcode update
200[ 2.671412] microcode: Microcode Update Driver: v2.00 &lt;tigran@aivazian.fsnet.co.uk&gt;, Peter Oruba</literal></screen>
201
202 <para>If the microcode was not updated, there is no new microcode for
203 this system's processor. If it did get updated, you can now proceed to <xref
204 linkend='early-microcode'/>.</para>
205
206 </sect3>
207
208 <sect3 id="and-microcode">
209 <title>AMD Microcode for the CPU</title>
210
211 <para>Begin by downloading a container of firmware for your CPU family
212 from <ulink
213 url='http://anduin.linuxfromscratch.org/sources/linux-firmware/amd-ucode/'/>.
214 The family is always specified in hex. Families 10h to 14h (16 to 20)
215 are in microcode_amd.bin. Families 15h and 16h have their own containers.
216 Create the required directory and put the firmware you downloaded into
217 it as the <systemitem class="username">root</systemitem> user:</para>
218
219<screen><userinput>mkdir -pv /lib/firmware/amd-ucode
220cp -v microcode_amd* /lib/firmware/amd-ucode</userinput></screen>
221
222 <para>When you configure the kernel, use the following options to try
223 late loading of AMD microcode:</para>
224
225<screen><literal>Processor type and features ---&gt;
226 &lt;M&gt; CPU microcode loading support [CONFIG_MICROCODE]
227 [*] AMD microcode loading support [CONFIG_MICROCODE_AMD]</literal></screen>
228
229 <para>After you have successfully booted the new system, use the command
230 <userinput>dmesg | grep microcode</userinput> and study the results to see
231 if the message new patch_level appears, as in this example from an old
232 Athlon(tm) II X2:</para>
233
234<screen><literal>[ 4.183907] microcode: CPU0: patch_level=0x010000b6
235[ 4.184271] microcode: CPU0: new patch_level=0x010000c8
236[ 4.184278] microcode: CPU1: patch_level=0x010000b6
237[ 4.184283] microcode: CPU1: new patch_level=0x010000c8
238[ 4.184359] microcode: Microcode Update Driver: v2.00 &lt;tigran@aivazian.fsnet.co.uk&gt;, Peter Oruba</literal></screen>
239
240 <para>If the microcode was not updated, there is no new microcode for
241 this system's processor. If it did get updated, you can now proceed to <xref
242 linkend='early-microcode'/>.</para>
243
244 </sect3>
245
246 <sect3 id="early-microcode">
247 <title>Early loading of microcode</title>
248
249 <para>If you have established that updated microcode is available for
250 your system, it is time to prepare it for early loading. This requires
251 an additional package, <xref linkend='cpio'/>, as well as changes to
252 the kernel config and the creation of an initrd which will need to be
253 added to grub.cfg.</para>
254
255 <para>It does not matter where you prepare the initrd, and once it is
256 working you can apply the same initrd to later LFS systems or newer
257 kernels on this same machine, at least until any newer microcode is
258 released. Use the following commands:</para>
259
260<screen><userinput>mkdir -p initrd/kernel/x86/microcode
261cd initrd</userinput></screen>
262
263 <para>For an AMD machine, use the following command (replace
264 &lt;MYCONTAINER&gt; with the name of the container for your CPU's
265 family):</para>
266
267<screen><userinput>cp -v /lib/firmware/amd_ucode/&lt;MYCONTAINER&gt; kernel/x86/microcode/AuthenticAMD.bin</userinput></screen>
268
269 <para>Or for an Intel machine copy the appropriate blob using this command:</para>
270
271<screen><userinput>cp -v /lib/firmware/intel-ucode/&lt;XX-YY-ZZ&gt; kernel/x86/microcode/GenuineIntel.bin</userinput></screen>
272
273 <para>Now prepare the initrd:</para>
274
275<screen><userinput>find . | cpio -o -H newc &gt; /boot/microcode.img</userinput></screen>
276
277 <para>You will now need to reconfigure and rebuild your kernel. It is
278 safer to either add/change the EXTRAVERSION in the kernel's configuration
279 and install the newer kernel with a new name, or else (unless you have a
280 machine which requires an early firmware update) wait for the next
281 SUBLEVEL kernel release so that you can fall back to the existing kernel
282 in the event that something goes wrong.</para>
283
284 <para>You will also need to add a new entry to /boot/grub/grub.cfg and
285 here you should add a new line after the linux line within the stanza.
286 If /boot is a separate mountpoint: </para>
287
288<screen><userinput>initrd /microcode.img</userinput></screen>
289
290 <para>or this if it is not:</para>
291
292<screen><userinput>initrd /boot/microcode.img</userinput></screen>
293
294 <para>You must also change the kernel config:</para>
295
296<screen><literal>General Setup ---&gt;
297 [y] Initial RAM filesystem and RAM disk (initramfs/initrd) support [CONFIG_BLK_DEV_INITRD]
298 [y] CPU microcode loading support [CONFIG_MICROCODE]</literal></screen>
299
300 <para>Retain the setting for INTEL or AMD microcode. When you have saved
301 the .config file, either CONFIG_MICROCODE_INTEL_EARLY=y or
302 CONFIG_MICROCODE_AMD_EARLY=y should be set, together with
303 CONFIG_MICROCODE_EARLY=y.</para>
304
305 <para>When you have installed and booted this kernel, you should check
306 the output of dmesg to confirm that the early load worked. The places and
307 times where this happens are very different in AMD and Intel machines.
308 First, an Intel example where a development kernel is being tested,
309 showing that the first notification comes before the kernel version is
310 mentioned:</para>
311
312<screen><literal>[ 0.000000] CPU0 microcode updated early to revision 0x29, date = 2013-06-12
313[ 0.000000] Linux version 4.0.0-rc6 (ken@jtm1) (gcc version 4.9.2 (GCC) )
314 #3 SMP PREEMPT Mon Mar 30 21:26:02 BST 2015
315[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.0.0-rc6-sda13 root=/dev/sda13 ro
316...
317[ 0.103091] CPU1 microcode updated early to revision 0x29, date = 2013-06-12
318[ 0.113241] #2
319[ 0.134631] #3
320[ 0.147821] x86: Booted up 1 node, 4 CPUs
321[ 0.147936] smpboot: Total of 4 processors activated (26338.66 BogoMIPS)
322...
323[ 0.272643] microcode: CPU0 sig=0x206a7, pf=0x2, revision=0x29
324[ 0.272709] microcode: CPU1 sig=0x206a7, pf=0x2, revision=0x29
325[ 0.272775] microcode: CPU2 sig=0x206a7, pf=0x2, revision=0x29
326[ 0.272842] microcode: CPU3 sig=0x206a7, pf=0x2, revision=0x29
327[ 0.272941] microcode: Microcode Update Driver: v2.00 &lt;tigran@aivazian.fsnet.co.uk&gt;, Peter Oruba</literal></screen>
328
329 <para>A second AMD example is where the machine was running a stable
330 kernel on an older version of LFS. Note that here there is no mention of
331 the previous microcode version &mdash; compare this output to the AMD
332 late loading messages (above) from the same machine:</para>
333
334<screen><literal>[ 0.000000] Linux version 3.18.11 (ken@milliways) (gcc version 4.9.1 (GCC) )
335 #4 SMP Thu Apr 9 21:51:05 BST 2015
336[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-3.18.11-sda5 root=/dev/sda5 video=800x600 ro
337...
338[ 0.584009] Trying to unpack rootfs image as initramfs...
339[ 0.584092] microcode: updated early to new patch_level=0x010000c8
340...
341[ 0.586733] microcode: CPU0: patch_level=0x010000c8
342[ 0.586778] microcode: CPU1: patch_level=0x010000c8
343[ 0.586866] microcode: Microcode Update Driver: v2.00 &lt;tigran@aivazian.fsnet.co.uk&gt;, Peter Oruba</literal></screen>
344
345 </sect3>
346
347 </sect2>
348
349 <sect2 id="ati-video-firmware">
350 <title>Firmware for ATI video chips (R600 and later)</title>
351
352 <para>These instructions do NOT apply to old radeons before the R600
353 family. For those, the firmware is in the kernel's <filename
354 class='directory'>/lib/firmware/</filename> directory. Nor do they apply if
355 you intend to avoid a graphical setup such as Xorg and are content to use
356 the default 80x25 display rather than a framebuffer. </para>
357
358 <para> Early radeon devices only needed a single 2K blob of firmware.
359 Recent devices need several different blobs, and some of them are much
360 bigger. The total size of the radeon firmware directory is over 500K &mdash; on a
361 large modern system you can probably spare the space, but it is still
362 redundant to install all the unused files each time you build a system.</para>
363
364 <para>A better approach is to install <xref linkend='pciutils'/> and then
365 use <userinput>lspci</userinput> to identify which VGA controller is
366 installed.</para>
367
368 <para>With that information, check the RadeonFeature page of the Xorg wiki
369 for <ulink url="http://wiki.x.org/wiki/RadeonFeature/#index5h2">Decoder
370 ring for engineering vs marketing names</ulink> to identify the family (you
371 may need to know this for the Xorg driver in BLFS &mdash; Southern Islands and
372 Sea Islands use the radeonsi driver) and the specific model.</para>
373
374 <para>Now that you know which controller you are using, consult the
375 <ulink url="https://wiki.gentoo.org/wiki/Radeon#Firmware">Radeon</ulink> page
376 of the Gentoo wiki which has a table listing the required firmware blobs
377 for the various chipsets. Note that Southern Islands and Sea Islands chips
378 use different firmware for kernel 3.17 and later compared to earlier
379 kernels. Identify and download the required blobs then install them:</para>
380
381<screen><userinput>mkdir -pv /lib/firmware/radeon
382cp -v &lt;YOUR_BLOBS&gt; /lib/firmware/radeon</userinput></screen>
383
384 <para>There are actually two ways of installing this firmware. BLFS, in the
385 'Kernel Configuration for additional firmware' section part of the <xref
386 linkend="xorg-ati-driver"/> section gives an example of compiling the
387 firmware into the kernel - that is slightly faster to load, but uses more
388 kernel memory. Here we will use the alternative method of making the radeon
389 driver a module. In your kernel config set the following: </para>
390
391<screen><literal>Device Drivers ---&gt;
392 Graphics support ---&gt;
393 Direct Rendering Manager ---&gt;
394 &lt;*&gt; Direct Rendering Manager (XFree86 ... support) [CONFIG_DRM]
395 &lt;m&gt; ATI Radeon [CONFIG_DRM_RADEON]</literal></screen>
396
397 <para>Loading several large blobs from /lib/firmware takes a noticeable
398 time, during which the screen will be blank. If you do not enable the
399 penguin framebuffer logo, or change the console size by using a bigger
400 font, that probably does not matter. If desired, you can slightly
401 reduce the time if you follow the alternate method of specifying 'y' for
402 CONFIG_DRM_RADEON covered in BLFS at the link above &mdash; you must specify each
403 needed radeon blob if you do that.</para>
404
405 </sect2>
406
407 <sect2 id="nic-firmware">
408 <title>Firmware for Network Interfaces</title>
409
410 <para>The kernel likes to load firmware for some network drivers,
411 particularly those from Realtek (the /lib/linux-firmware/rtl_nic/) directory,
412 but they generally appear to work without it. Therefore, you can boot the
413 kernel, check dmesg for messages about this missing firmware, and if
414 necessary download the firmware and put it in the specified directory in
415 /lib/firmware so that it will be found on subsequent boots. Note that with
416 current kernels this works whether or not the driver is compiled in or
417 built as a module, there is no need to build this firmware into the kernel.
418 Here is an example where the R8169 driver has been compiled in but the
419 firmware was not made available. Once the firmware had been provided, there
420 was no mention of it on later boots. </para>
421
422<screen><literal>dmesg | grep firmware | grep r8169
423[ 7.018028] r8169 0000:01:00.0: Direct firmware load for rtl_nic/rtl8168g-2.fw failed with error -2
424[ 7.018036] r8169 0000:01:00.0 eth0: unable to load firmware patch rtl_nic/rtl8168g-2.fw (-2)</literal></screen>
425
426 </sect2>
427
428 <sect2 id="other-firmware">
429 <title>Firmware for Other Devices</title>
430
431 <para> Identifying the correct firmware will typically require you to
432 install <xref linkend='pciutils'/>, and then use
433 <userinput>lspci</userinput> to identify the device. You should then search
434 online to check which module it uses, which firmware, and where to obtain
435 the firmware &mdash; not all of it is in linux-firmware.</para>
436
437 <para>If possible, you should begin by using a wired connection when you
438 first boot your LFS system. To use a wireless connection you will need to
439 use a network tools such as <xref linkend='wireless_tools'/> and <xref
440 linkend='wpa_supplicant'/>.</para>
441
442 <para>Firmware may also be needed for other devices such as some SCSI
443 controllers, bluetooth adaptors, or TV recorders. The same principles
444 apply.</para>
445
446 </sect2>
447
448</sect1>
Note: See TracBrowser for help on using the repository browser.