source: postlfs/config/firmware.xml@ 1c1bdba

10.0 10.1 11.0 11.1 8.2 8.3 8.4 9.0 9.1 basic bdubbs/svn elogind lazarus perl-modules qt5new trunk upgradedb xry111/intltool xry111/test-20220226
Last change on this file since 1c1bdba was 1c1bdba, checked in by Ken Moffat <ken@…>, 4 years ago

firmware: Update the intel microcode examples to the current, i.e. old, release and the 4.15.3 kernel. Nobody commented on the bogosity in my previous late-load example, but it is gone now.

git-svn-id: svn://svn.linuxfromscratch.org/BLFS/trunk/BOOK@19853 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>$LastChangedBy$</othername>
13 <date>$Date$</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. There is 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 via <userinput>wget</userinput> or a web
35 browser at <ulink
36 url="&sources-anduin-http;/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 mostly
60 apply to ATI devices : Radeons, the later AMD amdgpu chips, and Nvidia
61 Maxwell cards require firmware to be able to use KMS
62 (kernel modesetting - the preferred option) as well as for Xorg. For
63 earlier radeon chips (before the R600), the firmware is still in the
64 kernel.</para>
65 </listitem>
66 <listitem>
67 <para>Firmware updates for wired network ports. Mostly they work even
68 without the updates, but one must assume that they will work better with
69 the updated firmware.</para>
70 </listitem>
71 <listitem>
72 <para>Firmware for other devices, such as wifi. These devices are not
73 required for the PC to boot, but need the firmware before these devices
74 can be used.</para>
75 </listitem>
76 </itemizedlist>
77
78 <note><para>Although not needed to load a firmware blob, the following
79 tools may be useful for determining, obtaining, or preparing the needed
80 firmware in order to load it into the system:
81 <xref linkend="cpio"/>,
82 <xref linkend="git"/>,
83 <xref linkend="pciutils"/>, and
84 <xref linkend="wget"/></para></note>
85
86 <para condition="html" role="usernotes">User Notes:
87 <ulink url="&blfs-wiki;/aboutfirmware"/></para>
88
89 <sect2 id="cpu-microcode">
90 <title>Microcode updates for CPUs</title>
91
92 <para>In general, microcode can be loaded by the BIOS or UEFI, and it might
93 be updated by upgrading to a newer version of those. On linux, you can also
94 load the microcode from the kernel if you are using an AMD family 10h or
95 later processor (first introduced late 2007), or an Intel processor from
96 1998 and later (Pentium4, Core, etc), if updated microcode has been
97 released. These updates only last until the machine is powered off, so they
98 need to be applied on every boot.</para>
99
100 <para>Intel provide frequent updates of their microcode. It is not uncommon
101 to find a newer version of microcode for an Intel processor even two years
102 after its release. New versions of AMD firmware are less common.</para>
103
104 <para>There used to be two ways of loading the microcode, described as 'early'
105 and 'late'. Early loading happens before userspace has been started, late
106 loading happens after userspace has started. Not surprisingly, early loading
107 was preferred, (see e.g. an explanatory comment in a kernel commit noted at
108 <ulink url="https://lwn.net/Articles/530346/">x86/microcode: Early load
109 microcode </ulink> on LWN.) Indeed, it is needed to work around one
110 particular erratum in early Intel Haswell processors which had TSX enabled.
111 (See <ulink
112 url="http://www.anandtech.com/show/8376/intel-disables-tsx-instructions-erratum-found-in-haswell-haswelleep-broadwellyi/">
113 Intel Disables TSX Instructions: Erratum Found in Haswell, Haswell-E/EP,
114 Broadwell-Y</ulink>.) Without this update glibc can do the wrong thing in
115 uncommon situations. </para>
116
117 <para>As a result, early loading is now expected, although for the moment
118 (4.11 kernels) it is still possible to manually force late loading of
119 microcode for testing. You will need to reconfigure your kernel for either
120 method. The instructions here will create a kernel
121 <filename>.config</filename> to suite early loading, before forcing late
122 loading to see if there is any microcode. If there is, the instructions
123 then show you how to create an initrd for early loading.</para>
124
125 <para>To confirm what processor(s) you have (if more than one, they will be
126 identical) look in /proc/cpuinfo.</para>
127
128 <sect3 id="intel-microcode">
129 <title>Intel Microcode for the CPU</title>
130
131 <para>The first step is to get the most recent version of the Intel
132 microcode. This must be done by navigating to
133 <ulink url='https://downloadcenter.intel.com/download/27337/Linux-Processor-Microcode-Data-File'/>
134 and following the instructions there. As of this writing the most recent
135 version of the microcode is <filename>microcode-20171117.tgz</filename>.
136 Extract this file in the normal way to create an <filename>intel-ucode</filename>
137 directory, containing various blobs with names in the form XX-YY-ZZ.</para>
138
139 <note><para>The above URL may not be the latest page. If it is not,
140 a line at the top of the list will direct you to the latest page.
141 </para></note>
142
143 <para>Now you need to determine your processor's identity to see if there
144 is any microcode for it. Determine the decimal values of the cpu family,
145 model and stepping by running the following command (it will also report
146 the current microcode version):</para>
147
148<screen><userinput>head -n7 /proc/cpuinfo</userinput></screen>
149
150 <para>Convert the cpu family, model and stepping to pairs of hexadecimal
151 digits. For a Haswell i7-4790 (described as Intel(R) Core(TM) i7-4790
152 CPU) the relevant values are cpu family 6, model 60, stepping 3 so in
153 this case the required identification is 06-3c-03. A look at the blobs
154 will show that there is one for this CPU (although it might
155 have already been applied by the BIOS). If there is a blob for your
156 system then test if it will be applied by copying it (replace &lt;XX-YY-ZZ&gt;
157 by the identifier for your machine) to where the kernel can find it:</para>
158
159<screen><userinput>mkdir -pv /lib/firmware/intel-ucode
160cp -v intel-ucode/&lt;XX-YY-ZZ&gt; /lib/firmware/intel-ucode</userinput></screen>
161
162 <para>Now that the Intel microcode has been prepared, use the following
163 options when you configure the kernel to load Intel
164 microcode:</para>
165
166<screen><literal>General Setup ---&gt;
167 [y] Initial RAM filesystem and RAM disk (initramfs/initrd) support [CONFIG_BLK_DEV_INITRD]
168Processor type and features ---&gt;
169 [y] CPU microcode loading support [CONFIG_MICROCODE]
170 [y] Intel microcode loading support [CONFIG_MICROCODE_INTEL]</literal></screen>
171
172 <para>After you have successfully booted the new system, force late loading by
173 using the command:</para>
174
175<screen><userinput>echo 1 > /sys/devices/system/cpu/microcode/reload</userinput></screen>
176
177 <para>Then use the following command to see if anything was loaded:</para>
178
179<screen><userinput>dmesg | grep -e 'microcode' -e 'Linux version' -e 'Command line'</userinput></screen>
180
181 <para>This example from the Haswell i7 which was released in Q2 2014 and is
182 not affected by the TSX errata shows it has been updated from revision 0x19
183 in the BIOS/UEFI (which this version of hte kernel now complains about) to
184 revision 0x22. Unlike in older kernels, the individual CPUs are not separately
185 reported:</para>
186
187<screen><literal>[ 0.000000] Linux version 4.15.3 (ken@plexi) (gcc version 7.3.0 (GCC))
188 #2 SMP PREEMPT Fri Feb 23 03:13:53 GMT 2018
189[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.15.3-sda6 root=/dev/sda6 ro
190[ 0.000000] [Firmware Bug]: TSC_DEADLINE disabled due to Errata;
191 please update microcode to version: 0x22 (or later)
192[ 0.482712] microcode: sig=0x306c3, pf=0x2, revision=0x19
193[ 0.482937] microcode: Microcode Update Driver: v2.2.
194[ 402.668612] microcode: updated to revision 0x22, date = 2017-01-27</literal></screen>
195
196 <para>If the microcode was not updated, there is no new microcode for
197 this system's processor. If it did get updated, you can now proceed to <xref
198 linkend='early-microcode'/>.</para>
199
200 </sect3>
201
202 <sect3 id="and-microcode">
203 <title>AMD Microcode for the CPU</title>
204
205 <para>Begin by downloading a container of firmware for your CPU family
206 from <ulink
207 url='&sources-anduin-http;/linux-firmware/amd-ucode/'/>.
208 The family is always specified in hex. Families 10h to 14h (16 to 20)
209 are in microcode_amd.bin. Families 15h and 16h have their own containers.
210 Create the required directory and put the firmware you downloaded into
211 it as the <systemitem class="username">root</systemitem> user:</para>
212
213<screen><userinput>mkdir -pv /lib/firmware/amd-ucode
214cp -v microcode_amd* /lib/firmware/amd-ucode</userinput></screen>
215
216 <para>When you configure the kernel, use the following options
217 to load AMD microcode:</para>
218
219<screen><literal>General Setup ---&gt;
220 [y] Initial RAM filesystem and RAM disk (initramfs/initrd) support [CONFIG_BLK_DEV_INITRD]
221Processor type and features ---&gt;
222 [y] CPU microcode loading support [CONFIG_MICROCODE]
223 [y] AMD microcode loading support [CONFIG_MICROCODE_AMD]</literal></screen>
224
225 <para>After you have successfully booted the new system, force late loading by
226 using the command:</para>
227
228<screen><userinput>echo 1 > /sys/devices/system/cpu/microcode/reload</userinput></screen>
229
230 <para>Then use the following command to see if anything was loaded:</para>
231
232<screen><userinput>dmesg | grep -e 'microcode' -e 'Linux version' -e 'Command line'</userinput></screen>
233 <para>This example from an old Athlon(tm) II X2 shows it has been updated.
234 For the moment, all CPUs are still reported in the microcode details on AMD
235 machines:</para>
236
237<screen><literal>[ 0.000000] Linux version 4.15.3 (ken@testserver) (gcc version 7.3.0 (GCC))
238 #1 SMP Sun Feb 18 02:08:12 GMT 2018
239[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.15.3-sda5 root=/dev/sda5 ro
240[ 0.307619] microcode: CPU0: patch_level=0x010000b6
241[ 0.307671] microcode: CPU1: patch_level=0x010000b6
242[ 0.307743] microcode: Microcode Update Driver: v2.2.
243[ 187.928891] microcode: CPU0: new patch_level=0x010000c8
244[ 187.928899] microcode: CPU1: new patch_level=0x010000c8</literal></screen>
245
246 <para>If the microcode was not updated, there is no new microcode for
247 this system's processor. If it did get updated, you can now proceed to <xref
248 linkend='early-microcode'/>.</para>
249
250 </sect3>
251
252 <sect3 id="early-microcode">
253 <title>Early loading of microcode</title>
254
255 <para>If you have established that updated microcode is available for
256 your system, it is time to prepare it for early loading. This requires
257 an additional package, <xref linkend='cpio'/> and the creation of an
258 initrd which will need to be added to grub.cfg.</para>
259
260 <para>It does not matter where you prepare the initrd, and once it is
261 working you can apply the same initrd to later LFS systems or newer
262 kernels on this same machine, at least until any newer microcode is
263 released. Use the following commands:</para>
264
265<screen><userinput>mkdir -p initrd/kernel/x86/microcode
266cd initrd</userinput></screen>
267
268 <para>For an AMD machine, use the following command (replace
269 &lt;MYCONTAINER&gt; with the name of the container for your CPU's
270 family):</para>
271
272<screen><userinput>cp -v /lib/firmware/amd-ucode/&lt;MYCONTAINER&gt; kernel/x86/microcode/AuthenticAMD.bin</userinput></screen>
273
274 <para>Or for an Intel machine copy the appropriate blob using this command:</para>
275
276<screen><userinput>cp -v /lib/firmware/intel-ucode/&lt;XX-YY-ZZ&gt; kernel/x86/microcode/GenuineIntel.bin</userinput></screen>
277
278 <para>Now prepare the initrd:</para>
279
280<screen><userinput>find . | cpio -o -H newc &gt; /boot/microcode.img</userinput></screen>
281
282 <para>You now need to add a new entry to /boot/grub/grub.cfg and
283 here you should add a new line after the linux line within the stanza.
284 If /boot is a separate mountpoint: </para>
285
286<screen><userinput>initrd /microcode.img</userinput></screen>
287
288 <para>or this if it is not:</para>
289
290<screen><userinput>initrd /boot/microcode.img</userinput></screen>
291
292 <para>If you are already booting with an initrd (see <xref
293 linkend="initramfs"/>) you must specify the microcode initrd first, using
294 a line such as <userinput>initrd /microcode.img
295 /other-initrd.img</userinput> (adapt that as above if /boot is not a
296 separate mountpoint).</para>
297
298 <para>You can now reboot with the added initrd, and then use the same
299 command to check that the early load worked.</para>
300
301<screen><userinput>dmesg | grep -e 'microcode' -e 'Linux version' -e 'Command line'</userinput></screen>
302
303 <para>The places and times where early loading happens are very different
304 in AMD and Intel machines. First, an Intel example from an updated
305 kernel, showing that the first notification comes before the kernel version
306 is mentioned:</para>
307
308<screen><literal>[ 0.000000] microcode: microcode updated early to revision 0x22, date = 2017-01-27
309[ 0.000000] Linux version 4.15.3 (ken@plexi) (gcc version 7.3.0 (GCC))
310 #3 SMP PREEMPT Fri Feb 23 05:06:50 GMT 2018
311[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.15.3-sda6 root=/dev/sda6 ro
312[ 0.489478] microcode: sig=0x306c3, pf=0x2, revision=0x22
313[ 0.489747] microcode: Microcode Update Driver: v2.2.</literal></screen>
314
315 <para>An AMD example:</para>
316
317<screen><literal>[ 0.000000] Linux version 4.15.3 (ken@testserver) (gcc version 7.3.0 (GCC))
318 #2 SMP Sun Feb 18 02:32:03 GMT 2018
319[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.15.3-sda5 root=/dev/sda5 ro
320[ 0.307619] microcode: microcode updated early to new patch_level=0x010000c8
321[ 0.307678] microcode: CPU0: patch_level=0x010000c8
322[ 0.307723] microcode: CPU1: patch_level=0x010000c8
323[ 0.307795] microcode: Microcode Update Driver: v2.2.</literal></screen>
324
325 </sect3>
326
327 </sect2>
328
329 <sect2 id="video-firmware">
330 <title>Firmware for Video Cards</title>
331
332 <sect3 id="ati-video-firmware">
333 <title>Firmware for ATI video chips (R600 and later)</title>
334
335 <para>These instructions do NOT apply to old radeons before the R600
336 family. For those, the firmware is in the kernel's <filename
337 class='directory'>/lib/firmware/</filename> directory. Nor do they apply if
338 you intend to avoid a graphical setup such as Xorg and are content to use
339 the default 80x25 display rather than a framebuffer. </para>
340
341 <para> Early radeon devices only needed a single 2K blob of firmware.
342 Recent devices need several different blobs, and some of them are much
343 bigger. The total size of the radeon firmware directory is over 500K &mdash; on a
344 large modern system you can probably spare the space, but it is still
345 redundant to install all the unused files each time you build a system.</para>
346
347 <para>A better approach is to install <xref linkend='pciutils'/> and then
348 use <userinput>lspci</userinput> to identify which VGA controller is
349 installed.</para>
350
351 <para>With that information, check the RadeonFeature page of the Xorg wiki
352 for <ulink url="http://wiki.x.org/wiki/RadeonFeature/#index5h2">Decoder
353 ring for engineering vs marketing names</ulink> to identify the family (you
354 may need to know this for the Xorg driver in BLFS &mdash; Southern Islands and
355 Sea Islands use the radeonsi driver) and the specific model.</para>
356
357 <para>Now that you know which controller you are using, consult the
358 <ulink url="https://wiki.gentoo.org/wiki/Radeon#Firmware">Radeon</ulink> page
359 of the Gentoo wiki which has a table listing the required firmware blobs
360 for the various chipsets. Note that Southern Islands and Sea Islands chips
361 use different firmware for kernel 3.17 and later compared to earlier
362 kernels. Identify and download the required blobs then install them:</para>
363
364<screen><userinput>mkdir -pv /lib/firmware/radeon
365cp -v &lt;YOUR_BLOBS&gt; /lib/firmware/radeon</userinput></screen>
366
367 <para>There are actually two ways of installing this firmware. BLFS, in the
368 'Kernel Configuration for additional firmware' section part of the <xref
369 linkend="xorg-ati-driver"/> section gives an example of compiling the
370 firmware into the kernel - that is slightly faster to load, but uses more
371 kernel memory. Here we will use the alternative method of making the radeon
372 driver a module. In your kernel config set the following: </para>
373
374<screen><literal>Device Drivers ---&gt;
375 Graphics support ---&gt;
376 Direct Rendering Manager ---&gt;
377 &lt;*&gt; Direct Rendering Manager (XFree86 ... support) [CONFIG_DRM]
378 &lt;m&gt; ATI Radeon [CONFIG_DRM_RADEON]</literal></screen>
379
380 <para>Loading several large blobs from /lib/firmware takes a noticeable
381 time, during which the screen will be blank. If you do not enable the
382 penguin framebuffer logo, or change the console size by using a bigger
383 font, that probably does not matter. If desired, you can slightly
384 reduce the time if you follow the alternate method of specifying 'y' for
385 CONFIG_DRM_RADEON covered in BLFS at the link above &mdash; you must specify each
386 needed radeon blob if you do that.</para>
387
388 </sect3>
389
390 <sect3 id="nvidia-video-firmware">
391 <title>Firmware for Nvidia video chips</title>
392
393 <para>Some Nvidia graphics chips need firmware updates to take advantage
394 of all the card's capability. These are generally the GeForce 8, 9, 9300,
395 and 200-900 series chips. For more exact information, see <ulink
396 url="https://nouveau.freedesktop.org/wiki/VideoAcceleration/#firmware">
397 https://nouveau.freedesktop.org/wiki/VideoAcceleration/#firmware</ulink>.</para>
398
399 <para>First, the kernel Nvidia driver must be activated:</para>
400
401<screen><literal>Device Drivers ---&gt;
402 Graphics support ---&gt;
403 Direct Rendering Manager ---&gt;
404 &lt;*&gt; Direct Rendering Manager (XFree86 ... support) [CONFIG_DRM]
405 &lt;*/m&gt; Nouveau (NVIDIA) cards [CONFIG_DRM_NOUVEAU]</literal></screen>
406
407 <para>The steps to install the Nvidia firmware are:</para>
408
409<screen><userinput>wget https://raw.github.com/imirkin/re-vp2/master/extract_firmware.py
410wget http://us.download.nvidia.com/XFree86/Linux-x86/325.15/NVIDIA-Linux-x86-325.15.run
411sh NVIDIA-Linux-x86-325.15.run --extract-only
412python extract_firmware.py
413mkdir -p /lib/firmware/nouveau
414cp -d nv* vuc-* /lib/firmware/nouveau/</userinput></screen>
415
416 </sect3>
417 </sect2>
418
419 <sect2 id="nic-firmware">
420 <title>Firmware for Network Interfaces</title>
421
422 <para>The kernel likes to load firmware for some network drivers,
423 particularly those from Realtek (the /lib/linux-firmware/rtl_nic/) directory,
424 but they generally appear to work without it. Therefore, you can boot the
425 kernel, check dmesg for messages about this missing firmware, and if
426 necessary download the firmware and put it in the specified directory in
427 /lib/firmware so that it will be found on subsequent boots. Note that with
428 current kernels this works whether or not the driver is compiled in or
429 built as a module, there is no need to build this firmware into the kernel.
430 Here is an example where the R8169 driver has been compiled in but the
431 firmware was not made available. Once the firmware had been provided, there
432 was no mention of it on later boots. </para>
433
434<screen><literal>dmesg | grep firmware | grep r8169
435[ 7.018028] r8169 0000:01:00.0: Direct firmware load for rtl_nic/rtl8168g-2.fw failed with error -2
436[ 7.018036] r8169 0000:01:00.0 eth0: unable to load firmware patch rtl_nic/rtl8168g-2.fw (-2)</literal></screen>
437
438 </sect2>
439
440 <sect2 id="other-firmware">
441 <title>Firmware for Other Devices</title>
442
443 <para> Identifying the correct firmware will typically require you to
444 install <xref linkend='pciutils'/>, and then use
445 <userinput>lspci</userinput> to identify the device. You should then search
446 online to check which module it uses, which firmware, and where to obtain
447 the firmware &mdash; not all of it is in linux-firmware.</para>
448
449 <para>If possible, you should begin by using a wired connection when you
450 first boot your LFS system. To use a wireless connection you will need to
451 use a network tools such as <xref linkend='wireless_tools'/> and <xref
452 linkend='wpa_supplicant'/>.</para>
453
454 <para>Firmware may also be needed for other devices such as some SCSI
455 controllers, bluetooth adaptors, or TV recorders. The same principles
456 apply.</para>
457
458 </sect2>
459
460</sect1>
Note: See TracBrowser for help on using the repository browser.