1 .. SPDX-License-Identifier: GPL-2.0
3 ===========================
4 AMD64 Specific Boot Options
5 ===========================
7 There are many others (usually documented in driver documentation), but
8 only the AMD64 specific ones are listed here.
12 Please see Documentation/x86/x86_64/machinecheck.rst for sysfs runtime tunables.
17 Disable CMCI(Corrected Machine Check Interrupt) that
18 Intel processor supports. Usually this disablement is
19 not recommended, but it might be handy if your hardware
21 Note that you'll get more problems without CMCI than with
22 due to the shared banks, i.e. you might get duplicated
25 Don't make logs for corrected errors. All events reported
26 as corrected are silently cleared by OS.
27 This option will be useful if you have no interest in any
30 Disable features for corrected errors, e.g. polling timer
31 and CMCI. All events reported as corrected are not cleared
32 by OS and remained in its error banks.
33 Usually this disablement is not recommended, however if
34 there is an agent checking/clearing corrected errors
35 (e.g. BIOS or hardware monitoring applications), conflicting
36 with OS's error handling, and you cannot deactivate the agent,
37 then this option will be a help.
39 Do not opt-in to Local MCE delivery. Use legacy method
42 Enable logging of machine checks left over from booting.
43 Disabled by default on AMD Fam10h and older because some BIOS
45 If your BIOS doesn't do that it's a good idea to enable though
46 to make sure you log even machine check events that result
47 in a reboot. On Intel systems it is enabled by default.
49 Disable boot machine check logging.
50 mce=tolerancelevel[,monarchtimeout] (number,number)
52 0: always panic on uncorrected errors, log corrected errors
53 1: panic or SIGBUS on uncorrected errors, log corrected errors
54 2: SIGBUS or log uncorrected errors, log corrected errors
55 3: never panic or SIGBUS, log all errors (for testing only)
57 Can be also set using sysfs which is preferable.
59 Sets the time in us to wait for other CPUs on machine checks. 0
61 mce=bios_cmci_threshold
62 Don't overwrite the bios-set CMCI threshold. This boot option
63 prevents Linux from overwriting the CMCI threshold set by the
64 bios. Without this option, Linux always sets the CMCI
65 threshold to 1. Enabling this may make memory predictive failure
66 analysis less effective if the bios sets thresholds for memory
67 errors since we will not see details for all errors.
69 Force-enable recoverable machine check code paths
71 nomce (for compatibility with i386)
74 Everything else is in sysfs now.
83 Don't use the IO-APIC.
86 Don't use the local APIC
89 Don't use the local APIC (alias for i386 compatibility)
92 See Documentation/x86/i386/IO-APIC.rst
95 Don't set up the APIC timer
98 Don't check the IO-APIC timer. This can work around
99 problems with incorrect timer initialization on some boards.
102 Do APIC timer calibration using the pmtimer. Implies
103 apicmaintimer. Useful when your PIT timer is totally broken.
109 Deprecated, use tsc=unstable instead.
112 Don't use the HPET timer.
118 Don't do power saving in the idle loop using HLT, but poll for rescheduling
119 event. This will make the CPUs eat a lot more power, but may be useful
120 to get slightly better performance in multiprocessor benchmarks. It also
121 makes some profiling using performance counters more accurate.
122 Please note that on systems with MONITOR/MWAIT support (like Intel EM64T
123 CPUs) this option has no performance advantage over the normal idle loop.
124 It may also interact badly with hyperthreading.
129 reboot=b[ios] | t[riple] | k[bd] | a[cpi] | e[fi] [, [w]arm | [c]old]
131 Use the CPU reboot vector for warm reset
133 Don't set the cold reboot flag
135 Set the cold reboot flag
137 Force a triple fault (init)
139 Use the keyboard controller. cold reset (default)
141 Use the ACPI RESET_REG in the FADT. If ACPI is not configured or
142 the ACPI reset does not work, the reboot path attempts the reset
143 using the keyboard controller.
145 Use efi reset_system runtime service. If EFI is not configured or
146 the EFI reset does not work, the reboot path attempts the reset using
147 the keyboard controller.
149 Using warm reset will be much faster especially on big memory
150 systems because the BIOS will not go through the memory check.
151 Disadvantage is that not all hardware will be completely reinitialized
152 on reboot so there may be boot problems on some systems.
155 Don't stop other CPUs on reboot. This can make reboot more reliable
158 Non Executable Mappings
159 =======================
171 Only set up a single NUMA node spanning all memory.
174 Don't parse the SRAT table for NUMA setup
177 Don't parse the HMAT table for NUMA setup, or soft-reserved memory
181 If given as a memory unit, fills all system RAM with nodes of
182 size interleaved over physical nodes.
185 If given as an integer, fills all system RAM with N fake nodes
186 interleaved over physical nodes.
189 If given as an integer followed by 'U', it will divide each
190 physical node into N emulated nodes.
198 Use ACPI boot table parsing, but don't enable ACPI interpreter
200 Force ACPI on (currently not needed)
202 Disable out of spec ACPI workarounds.
203 acpi_sci={edge,level,high,low}
204 Set up ACPI SCI interrupt.
206 Don't route interrupts
208 Disable firmware first mode for corrected errors. This
209 disables parsing the HEST CMC error source to check if
210 firmware has set the FF flag. This may result in
211 duplicate corrected error reports.
227 Set PCI interrupt mask to MASK
229 Scan up to NUMBER busses, no matter what the mptable says.
231 Don't use ACPI to set up PCI interrupt routing.
233 IOMMU (input/output memory management unit)
234 ===========================================
235 Multiple x86-64 PCI-DMA mapping implementations exist, for example:
237 1. <kernel/dma/direct.c>: use no hardware/software IOMMU at all
238 (e.g. because you have < 3 GB memory).
239 Kernel boot message: "PCI-DMA: Disabling IOMMU"
241 2. <arch/x86/kernel/amd_gart_64.c>: AMD GART based hardware IOMMU.
242 Kernel boot message: "PCI-DMA: using GART IOMMU"
244 3. <arch/x86_64/kernel/pci-swiotlb.c> : Software IOMMU implementation. Used
245 e.g. if there is no hardware IOMMU in the system and it is need because
246 you have >3GB memory or told the kernel to us it (iommu=soft))
247 Kernel boot message: "PCI-DMA: Using software bounce buffering
250 4. <arch/x86_64/pci-calgary.c> : IBM Calgary hardware IOMMU. Used in IBM
251 pSeries and xSeries servers. This hardware IOMMU supports DMA address
252 mapping with memory protection, etc.
253 Kernel boot message: "PCI-DMA: Using Calgary IOMMU"
257 iommu=[<size>][,noagp][,off][,force][,noforce]
258 [,memaper[=<order>]][,merge][,fullflush][,nomerge]
259 [,noaperture][,calgary]
261 General iommu options:
264 Don't initialize and use any kind of IOMMU.
266 Don't force hardware IOMMU usage when it is not needed. (default).
268 Force the use of the hardware IOMMU even when it is
269 not actually needed (e.g. because < 3 GB memory).
271 Use software bounce buffering (SWIOTLB) (default for
272 Intel machines). This can be used to prevent the usage
273 of an available hardware IOMMU.
275 iommu options only relevant to the AMD GART hardware IOMMU:
278 Set the size of the remapping area in bytes.
280 Overwrite iommu off workarounds for specific chipsets.
282 Flush IOMMU on each allocation (default).
284 Don't use IOMMU fullflush.
286 Allocate an own aperture over RAM with size 32MB<<order.
287 (default: order=1, i.e. 64MB)
289 Do scatter-gather (SG) merging. Implies "force" (experimental).
291 Don't do scatter-gather (SG) merging.
293 Ask the IOMMU not to touch the aperture for AGP.
295 Don't initialize the AGP driver and use full aperture.
297 Always panic when IOMMU overflows.
299 Use the Calgary IOMMU if it is available
301 iommu options only relevant to the software bounce buffering (SWIOTLB) IOMMU
304 swiotlb=<pages>[,force]
306 Prereserve that many 128K pages for the software IO bounce buffering.
308 Force all IO through the software TLB.
310 Settings for the IBM Calgary hardware IOMMU currently found in IBM
311 pSeries and xSeries machines
313 calgary=[64k,128k,256k,512k,1M,2M,4M,8M]
314 Set the size of each PCI slot's translation table when using the
315 Calgary IOMMU. This is the size of the translation table itself
316 in main memory. The smallest table, 64k, covers an IO space of
317 32MB; the largest, 8MB table, can cover an IO space of 4GB.
318 Normally the kernel will make the right choice by itself.
319 calgary=[translate_empty_slots]
320 Enable translation even on slots that have no devices attached to
321 them, in case a device will be hotplugged in the future.
322 calgary=[disable=<PCI bus number>]
323 Disable translation on a given PHB. For
324 example, the built-in graphics adapter resides on the first bridge
325 (PCI bus number 0); if translation (isolation) is enabled on this
326 bridge, X servers that access the hardware directly from user
327 space might stop working. Use this option if you have devices that
328 are accessed from userspace directly on some PCI host bridge.
330 Always panic when IOMMU overflows
337 Do not use GB pages for kernel direct mappings.
339 Use GB pages for kernel direct mappings.