1AMD64 specific boot options
2
3There are many others (usually documented in driver documentation), but
4only the AMD64 specific ones are listed here.
5
6Machine check
7
8   Please see Documentation/x86/x86_64/machinecheck for sysfs runtime tunables.
9
10   mce=off
11		Disable machine check
12   mce=no_cmci
13		Disable CMCI(Corrected Machine Check Interrupt) that
14		Intel processor supports.  Usually this disablement is
15		not recommended, but it might be handy if your hardware
16		is misbehaving.
17		Note that you'll get more problems without CMCI than with
18		due to the shared banks, i.e. you might get duplicated
19		error logs.
20   mce=dont_log_ce
21		Don't make logs for corrected errors.  All events reported
22		as corrected are silently cleared by OS.
23		This option will be useful if you have no interest in any
24		of corrected errors.
25   mce=ignore_ce
26		Disable features for corrected errors, e.g. polling timer
27		and CMCI.  All events reported as corrected are not cleared
28		by OS and remained in its error banks.
29		Usually this disablement is not recommended, however if
30		there is an agent checking/clearing corrected errors
31		(e.g. BIOS or hardware monitoring applications), conflicting
32		with OS's error handling, and you cannot deactivate the agent,
33		then this option will be a help.
34   mce=bootlog
35		Enable logging of machine checks left over from booting.
36		Disabled by default on AMD because some BIOS leave bogus ones.
37		If your BIOS doesn't do that it's a good idea to enable though
38		to make sure you log even machine check events that result
39		in a reboot. On Intel systems it is enabled by default.
40   mce=nobootlog
41		Disable boot machine check logging.
42   mce=tolerancelevel[,monarchtimeout] (number,number)
43		tolerance levels:
44		0: always panic on uncorrected errors, log corrected errors
45		1: panic or SIGBUS on uncorrected errors, log corrected errors
46		2: SIGBUS or log uncorrected errors, log corrected errors
47		3: never panic or SIGBUS, log all errors (for testing only)
48		Default is 1
49		Can be also set using sysfs which is preferable.
50		monarchtimeout:
51		Sets the time in us to wait for other CPUs on machine checks. 0
52		to disable.
53
54   nomce (for compatibility with i386): same as mce=off
55
56   Everything else is in sysfs now.
57
58APICs
59
60   apic		 Use IO-APIC. Default
61
62   noapic	 Don't use the IO-APIC.
63
64   disableapic	 Don't use the local APIC
65
66   nolapic	 Don't use the local APIC (alias for i386 compatibility)
67
68   pirq=...	 See Documentation/x86/i386/IO-APIC.txt
69
70   noapictimer	 Don't set up the APIC timer
71
72   no_timer_check Don't check the IO-APIC timer. This can work around
73		 problems with incorrect timer initialization on some boards.
74
75   apicmaintimer Run time keeping from the local APIC timer instead
76                 of using the PIT/HPET interrupt for this. This is useful
77                 when the PIT/HPET interrupts are unreliable.
78
79   noapicmaintimer  Don't do time keeping using the APIC timer.
80		 Useful when this option was auto selected, but doesn't work.
81
82   apicpmtimer
83		 Do APIC timer calibration using the pmtimer. Implies
84		 apicmaintimer. Useful when your PIT timer is totally
85		 broken.
86
87Early Console
88
89   syntax: earlyprintk=vga
90           earlyprintk=serial[,ttySn[,baudrate]]
91
92   The early console is useful when the kernel crashes before the
93   normal console is initialized. It is not enabled by
94   default because it has some cosmetic problems.
95   Append ,keep to not disable it when the real console takes over.
96   Only vga or serial at a time, not both.
97   Currently only ttyS0 and ttyS1 are supported.
98   Interaction with the standard serial driver is not very good.
99   The VGA output is eventually overwritten by the real console.
100
101Timing
102
103  notsc
104  Don't use the CPU time stamp counter to read the wall time.
105  This can be used to work around timing problems on multiprocessor systems
106  with not properly synchronized CPUs.
107
108  report_lost_ticks
109  Report when timer interrupts are lost because some code turned off
110  interrupts for too long.
111
112  nohpet
113  Don't use the HPET timer.
114
115Idle loop
116
117  idle=poll
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.
125
126Rebooting
127
128   reboot=b[ios] | t[riple] | k[bd] | a[cpi] | e[fi] [, [w]arm | [c]old]
129   bios	  Use the CPU reboot vector for warm reset
130   warm   Don't set the cold reboot flag
131   cold   Set the cold reboot flag
132   triple Force a triple fault (init)
133   kbd    Use the keyboard controller. cold reset (default)
134   acpi   Use the ACPI RESET_REG in the FADT. If ACPI is not configured or the
135          ACPI reset does not work, the reboot path attempts the reset using
136          the keyboard controller.
137   efi    Use efi reset_system runtime service. If EFI is not configured or the
138          EFI reset does not work, the reboot path attempts the reset using
139          the keyboard controller.
140
141   Using warm reset will be much faster especially on big memory
142   systems because the BIOS will not go through the memory check.
143   Disadvantage is that not all hardware will be completely reinitialized
144   on reboot so there may be boot problems on some systems.
145
146   reboot=force
147
148   Don't stop other CPUs on reboot. This can make reboot more reliable
149   in some cases.
150
151Non Executable Mappings
152
153  noexec=on|off
154
155  on      Enable(default)
156  off     Disable
157
158SMP
159
160  additional_cpus=NUM Allow NUM more CPUs for hotplug
161		 (defaults are specified by the BIOS, see Documentation/x86/x86_64/cpu-hotplug-spec)
162
163NUMA
164
165  numa=off	Only set up a single NUMA node spanning all memory.
166
167  numa=noacpi   Don't parse the SRAT table for NUMA setup
168
169  numa=fake=<size>[MG]
170		If given as a memory unit, fills all system RAM with nodes of
171		size interleaved over physical nodes.
172
173  numa=fake=<N>
174		If given as an integer, fills all system RAM with N fake nodes
175		interleaved over physical nodes.
176
177ACPI
178
179  acpi=off	Don't enable ACPI
180  acpi=ht	Use ACPI boot table parsing, but don't enable ACPI
181		interpreter
182  acpi=force	Force ACPI on (currently not needed)
183
184  acpi=strict   Disable out of spec ACPI workarounds.
185
186  acpi_sci={edge,level,high,low}  Set up ACPI SCI interrupt.
187
188  acpi=noirq	Don't route interrupts
189
190PCI
191
192  pci=off		Don't use PCI
193  pci=conf1		Use conf1 access.
194  pci=conf2		Use conf2 access.
195  pci=rom		Assign ROMs.
196  pci=assign-busses	Assign busses
197  pci=irqmask=MASK	Set PCI interrupt mask to MASK
198  pci=lastbus=NUMBER	Scan up to NUMBER busses, no matter what the mptable says.
199  pci=noacpi		Don't use ACPI to set up PCI interrupt routing.
200
201IOMMU (input/output memory management unit)
202
203 Currently four x86-64 PCI-DMA mapping implementations exist:
204
205   1. <arch/x86_64/kernel/pci-nommu.c>: use no hardware/software IOMMU at all
206      (e.g. because you have < 3 GB memory).
207      Kernel boot message: "PCI-DMA: Disabling IOMMU"
208
209   2. <arch/x86/kernel/amd_gart_64.c>: AMD GART based hardware IOMMU.
210      Kernel boot message: "PCI-DMA: using GART IOMMU"
211
212   3. <arch/x86_64/kernel/pci-swiotlb.c> : Software IOMMU implementation. Used
213      e.g. if there is no hardware IOMMU in the system and it is need because
214      you have >3GB memory or told the kernel to us it (iommu=soft))
215      Kernel boot message: "PCI-DMA: Using software bounce buffering
216      for IO (SWIOTLB)"
217
218   4. <arch/x86_64/pci-calgary.c> : IBM Calgary hardware IOMMU. Used in IBM
219      pSeries and xSeries servers. This hardware IOMMU supports DMA address
220      mapping with memory protection, etc.
221      Kernel boot message: "PCI-DMA: Using Calgary IOMMU"
222
223 iommu=[<size>][,noagp][,off][,force][,noforce][,leak[=<nr_of_leak_pages>]
224	[,memaper[=<order>]][,merge][,forcesac][,fullflush][,nomerge]
225	[,noaperture][,calgary]
226
227  General iommu options:
228    off                Don't initialize and use any kind of IOMMU.
229    noforce            Don't force hardware IOMMU usage when it is not needed.
230                       (default).
231    force              Force the use of the hardware IOMMU even when it is
232                       not actually needed (e.g. because < 3 GB memory).
233    soft               Use software bounce buffering (SWIOTLB) (default for
234                       Intel machines). This can be used to prevent the usage
235                       of an available hardware IOMMU.
236
237  iommu options only relevant to the AMD GART hardware IOMMU:
238    <size>             Set the size of the remapping area in bytes.
239    allowed            Overwrite iommu off workarounds for specific chipsets.
240    fullflush          Flush IOMMU on each allocation (default).
241    nofullflush        Don't use IOMMU fullflush.
242    leak               Turn on simple iommu leak tracing (only when
243                       CONFIG_IOMMU_LEAK is on). Default number of leak pages
244                       is 20.
245    memaper[=<order>]  Allocate an own aperture over RAM with size 32MB<<order.
246                       (default: order=1, i.e. 64MB)
247    merge              Do scatter-gather (SG) merging. Implies "force"
248                       (experimental).
249    nomerge            Don't do scatter-gather (SG) merging.
250    noaperture         Ask the IOMMU not to touch the aperture for AGP.
251    forcesac           Force single-address cycle (SAC) mode for masks <40bits
252                       (experimental).
253    noagp              Don't initialize the AGP driver and use full aperture.
254    allowdac           Allow double-address cycle (DAC) mode, i.e. DMA >4GB.
255                       DAC is used with 32-bit PCI to push a 64-bit address in
256                       two cycles. When off all DMA over >4GB is forced through
257                       an IOMMU or software bounce buffering.
258    nodac              Forbid DAC mode, i.e. DMA >4GB.
259    panic              Always panic when IOMMU overflows.
260    calgary            Use the Calgary IOMMU if it is available
261
262  iommu options only relevant to the software bounce buffering (SWIOTLB) IOMMU
263  implementation:
264    swiotlb=<pages>[,force]
265    <pages>            Prereserve that many 128K pages for the software IO
266                       bounce buffering.
267    force              Force all IO through the software TLB.
268
269  Settings for the IBM Calgary hardware IOMMU currently found in IBM
270  pSeries and xSeries machines:
271
272    calgary=[64k,128k,256k,512k,1M,2M,4M,8M]
273    calgary=[translate_empty_slots]
274    calgary=[disable=<PCI bus number>]
275    panic              Always panic when IOMMU overflows
276
277    64k,...,8M - Set the size of each PCI slot's translation table
278    when using the Calgary IOMMU. This is the size of the translation
279    table itself in main memory. The smallest table, 64k, covers an IO
280    space of 32MB; the largest, 8MB table, can cover an IO space of
281    4GB. Normally the kernel will make the right choice by itself.
282
283    translate_empty_slots - Enable translation even on slots that have
284    no devices attached to them, in case a device will be hotplugged
285    in the future.
286
287    disable=<PCI bus number> - Disable translation on a given PHB. For
288    example, the built-in graphics adapter resides on the first bridge
289    (PCI bus number 0); if translation (isolation) is enabled on this
290    bridge, X servers that access the hardware directly from user
291    space might stop working. Use this option if you have devices that
292    are accessed from userspace directly on some PCI host bridge.
293
294Debugging
295
296  kstack=N	Print N words from the kernel stack in oops dumps.
297
298  pagefaulttrace  Dump all page faults. Only useful for extreme debugging
299		and will create a lot of output.
300
301  call_trace=[old|both|newfallback|new]
302		old: use old inexact backtracer
303		new: use new exact dwarf2 unwinder
304 		both: print entries from both
305		newfallback: use new unwinder but fall back to old if it gets
306			stuck (default)
307
308Miscellaneous
309
310	nogbpages
311		Do not use GB pages for kernel direct mappings.
312	gbpages
313		Use GB pages for kernel direct mappings.
314