Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | AMD64 specific boot options |
| 2 | |
| 3 | There are many others (usually documented in driver documentation), but |
| 4 | only the AMD64 specific ones are listed here. |
| 5 | |
| 6 | Machine check |
| 7 | |
| 8 | mce=off disable machine check |
Andi Kleen | d5172f2 | 2005-08-07 09:42:07 -0700 | [diff] [blame] | 9 | mce=bootlog Enable logging of machine checks left over from booting. |
Andi Kleen | e583538 | 2005-11-05 17:25:54 +0100 | [diff] [blame] | 10 | Disabled by default on AMD because some BIOS leave bogus ones. |
Andi Kleen | d5172f2 | 2005-08-07 09:42:07 -0700 | [diff] [blame] | 11 | If your BIOS doesn't do that it's a good idea to enable though |
| 12 | to make sure you log even machine check events that result |
Andi Kleen | e583538 | 2005-11-05 17:25:54 +0100 | [diff] [blame] | 13 | in a reboot. On Intel systems it is enabled by default. |
| 14 | mce=nobootlog |
| 15 | Disable boot machine check logging. |
Andi Kleen | 8c566ef | 2005-09-12 18:49:24 +0200 | [diff] [blame] | 16 | mce=tolerancelevel (number) |
| 17 | 0: always panic, 1: panic if deadlock possible, |
| 18 | 2: try to avoid panic, 3: never panic or exit (for testing) |
| 19 | default is 1 |
| 20 | Can be also set using sysfs which is preferable. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 21 | |
| 22 | nomce (for compatibility with i386): same as mce=off |
| 23 | |
| 24 | Everything else is in sysfs now. |
| 25 | |
| 26 | APICs |
| 27 | |
| 28 | apic Use IO-APIC. Default |
| 29 | |
| 30 | noapic Don't use the IO-APIC. |
| 31 | |
| 32 | disableapic Don't use the local APIC |
| 33 | |
| 34 | nolapic Don't use the local APIC (alias for i386 compatibility) |
| 35 | |
| 36 | pirq=... See Documentation/i386/IO-APIC.txt |
| 37 | |
| 38 | noapictimer Don't set up the APIC timer |
| 39 | |
Andi Kleen | 14d98ca | 2005-05-20 14:27:59 -0700 | [diff] [blame] | 40 | no_timer_check Don't check the IO-APIC timer. This can work around |
| 41 | problems with incorrect timer initialization on some boards. |
| 42 | |
Andi Kleen | 73dea47 | 2006-02-03 21:50:50 +0100 | [diff] [blame] | 43 | apicmaintimer Run time keeping from the local APIC timer instead |
| 44 | of using the PIT/HPET interrupt for this. This is useful |
| 45 | when the PIT/HPET interrupts are unreliable. |
| 46 | |
| 47 | noapicmaintimer Don't do time keeping using the APIC timer. |
| 48 | Useful when this option was auto selected, but doesn't work. |
| 49 | |
Andi Kleen | 0c3749c | 2006-02-03 21:51:41 +0100 | [diff] [blame] | 50 | apicpmtimer |
| 51 | Do APIC timer calibration using the pmtimer. Implies |
| 52 | apicmaintimer. Useful when your PIT timer is totally |
| 53 | broken. |
| 54 | |
Andi Kleen | ab9b32e | 2006-02-26 04:18:52 +0100 | [diff] [blame] | 55 | disable_8254_timer / enable_8254_timer |
| 56 | Enable interrupt 0 timer routing over the 8254 in addition to over |
| 57 | the IO-APIC. The kernel tries to set a sensible default. |
| 58 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 59 | Early Console |
| 60 | |
| 61 | syntax: earlyprintk=vga |
| 62 | earlyprintk=serial[,ttySn[,baudrate]] |
| 63 | |
| 64 | The early console is useful when the kernel crashes before the |
| 65 | normal console is initialized. It is not enabled by |
| 66 | default because it has some cosmetic problems. |
| 67 | Append ,keep to not disable it when the real console takes over. |
| 68 | Only vga or serial at a time, not both. |
| 69 | Currently only ttyS0 and ttyS1 are supported. |
| 70 | Interaction with the standard serial driver is not very good. |
| 71 | The VGA output is eventually overwritten by the real console. |
| 72 | |
| 73 | Timing |
| 74 | |
| 75 | notsc |
| 76 | Don't use the CPU time stamp counter to read the wall time. |
| 77 | This can be used to work around timing problems on multiprocessor systems |
Andi Kleen | ef4d7cb | 2005-07-28 21:15:34 -0700 | [diff] [blame] | 78 | with not properly synchronized CPUs. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 79 | |
| 80 | report_lost_ticks |
| 81 | Report when timer interrupts are lost because some code turned off |
| 82 | interrupts for too long. |
| 83 | |
| 84 | nmi_watchdog=NUMBER[,panic] |
| 85 | NUMBER can be: |
| 86 | 0 don't use an NMI watchdog |
| 87 | 1 use the IO-APIC timer for the NMI watchdog |
| 88 | 2 use the local APIC for the NMI watchdog using a performance counter. Note |
| 89 | This will use one performance counter and the local APIC's performance |
| 90 | vector. |
| 91 | When panic is specified panic when an NMI watchdog timeout occurs. |
| 92 | This is useful when you use a panic=... timeout and need the box |
| 93 | quickly up again. |
| 94 | |
| 95 | nohpet |
| 96 | Don't use the HPET timer. |
| 97 | |
| 98 | Idle loop |
| 99 | |
| 100 | idle=poll |
| 101 | Don't do power saving in the idle loop using HLT, but poll for rescheduling |
| 102 | event. This will make the CPUs eat a lot more power, but may be useful |
| 103 | to get slightly better performance in multiprocessor benchmarks. It also |
| 104 | makes some profiling using performance counters more accurate. |
Andi Kleen | ef4d7cb | 2005-07-28 21:15:34 -0700 | [diff] [blame] | 105 | Please note that on systems with MONITOR/MWAIT support (like Intel EM64T |
| 106 | CPUs) this option has no performance advantage over the normal idle loop. |
| 107 | It may also interact badly with hyperthreading. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 108 | |
| 109 | Rebooting |
| 110 | |
| 111 | reboot=b[ios] | t[riple] | k[bd] [, [w]arm | [c]old] |
Matt LaPlante | d6bc8ac | 2006-10-03 22:54:15 +0200 | [diff] [blame] | 112 | bios Use the CPU reboot vector for warm reset |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 113 | warm Don't set the cold reboot flag |
| 114 | cold Set the cold reboot flag |
| 115 | triple Force a triple fault (init) |
| 116 | kbd Use the keyboard controller. cold reset (default) |
| 117 | |
| 118 | Using warm reset will be much faster especially on big memory |
| 119 | systems because the BIOS will not go through the memory check. |
| 120 | Disadvantage is that not all hardware will be completely reinitialized |
| 121 | on reboot so there may be boot problems on some systems. |
| 122 | |
| 123 | reboot=force |
| 124 | |
| 125 | Don't stop other CPUs on reboot. This can make reboot more reliable |
| 126 | in some cases. |
| 127 | |
| 128 | Non Executable Mappings |
| 129 | |
| 130 | noexec=on|off |
| 131 | |
| 132 | on Enable(default) |
| 133 | off Disable |
| 134 | |
| 135 | SMP |
| 136 | |
| 137 | nosmp Only use a single CPU |
| 138 | |
| 139 | maxcpus=NUMBER only use upto NUMBER CPUs |
| 140 | |
| 141 | cpumask=MASK only use cpus with bits set in mask |
| 142 | |
Andi Kleen | 420f8f6 | 2005-11-05 17:25:54 +0100 | [diff] [blame] | 143 | additional_cpus=NUM Allow NUM more CPUs for hotplug |
Andi Kleen | f62a91f | 2006-01-11 22:42:35 +0100 | [diff] [blame] | 144 | (defaults are specified by the BIOS, see Documentation/x86_64/cpu-hotplug-spec) |
Andi Kleen | 420f8f6 | 2005-11-05 17:25:54 +0100 | [diff] [blame] | 145 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 146 | NUMA |
| 147 | |
| 148 | numa=off Only set up a single NUMA node spanning all memory. |
| 149 | |
| 150 | numa=noacpi Don't parse the SRAT table for NUMA setup |
| 151 | |
| 152 | numa=fake=X Fake X nodes and ignore NUMA setup of the actual machine. |
| 153 | |
Andi Kleen | 68a3a7f | 2006-04-07 19:49:18 +0200 | [diff] [blame] | 154 | numa=hotadd=percent |
| 155 | Only allow hotadd memory to preallocate page structures upto |
| 156 | percent of already available memory. |
| 157 | numa=hotadd=0 will disable hotadd memory. |
| 158 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 159 | ACPI |
| 160 | |
| 161 | acpi=off Don't enable ACPI |
| 162 | acpi=ht Use ACPI boot table parsing, but don't enable ACPI |
| 163 | interpreter |
| 164 | acpi=force Force ACPI on (currently not needed) |
| 165 | |
| 166 | acpi=strict Disable out of spec ACPI workarounds. |
| 167 | |
| 168 | acpi_sci={edge,level,high,low} Set up ACPI SCI interrupt. |
| 169 | |
| 170 | acpi=noirq Don't route interrupts |
| 171 | |
| 172 | PCI |
| 173 | |
| 174 | pci=off Don't use PCI |
| 175 | pci=conf1 Use conf1 access. |
| 176 | pci=conf2 Use conf2 access. |
| 177 | pci=rom Assign ROMs. |
| 178 | pci=assign-busses Assign busses |
| 179 | pci=irqmask=MASK Set PCI interrupt mask to MASK |
| 180 | pci=lastbus=NUMBER Scan upto NUMBER busses, no matter what the mptable says. |
| 181 | pci=noacpi Don't use ACPI to set up PCI interrupt routing. |
| 182 | |
| 183 | IOMMU |
| 184 | |
| 185 | iommu=[size][,noagp][,off][,force][,noforce][,leak][,memaper[=order]][,merge] |
| 186 | [,forcesac][,fullflush][,nomerge][,noaperture] |
| 187 | size set size of iommu (in bytes) |
| 188 | noagp don't initialize the AGP driver and use full aperture. |
| 189 | off don't use the IOMMU |
| 190 | leak turn on simple iommu leak tracing (only when CONFIG_IOMMU_LEAK is on) |
| 191 | memaper[=order] allocate an own aperture over RAM with size 32MB^order. |
| 192 | noforce don't force IOMMU usage. Default. |
| 193 | force Force IOMMU. |
| 194 | merge Do SG merging. Implies force (experimental) |
| 195 | nomerge Don't do SG merging. |
| 196 | forcesac For SAC mode for masks <40bits (experimental) |
| 197 | fullflush Flush IOMMU on each allocation (default) |
| 198 | nofullflush Don't use IOMMU fullflush |
| 199 | allowed overwrite iommu off workarounds for specific chipsets. |
| 200 | soft Use software bounce buffering (default for Intel machines) |
| 201 | noaperture Don't touch the aperture for AGP. |
Andi Kleen | ece6684 | 2006-09-30 01:47:55 +0200 | [diff] [blame] | 202 | allowdac Allow DMA >4GB |
| 203 | When off all DMA over >4GB is forced through an IOMMU or bounce |
| 204 | buffering. |
| 205 | nodac Forbid DMA >4GB |
Andi Kleen | d802ab9 | 2006-09-30 01:47:55 +0200 | [diff] [blame] | 206 | panic Always panic when IOMMU overflows |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 207 | |
| 208 | swiotlb=pages[,force] |
| 209 | |
| 210 | pages Prereserve that many 128K pages for the software IO bounce buffering. |
| 211 | force Force all IO through the software TLB. |
| 212 | |
Jon Mason | e465058 | 2006-06-26 13:58:14 +0200 | [diff] [blame] | 213 | calgary=[64k,128k,256k,512k,1M,2M,4M,8M] |
| 214 | calgary=[translate_empty_slots] |
| 215 | calgary=[disable=<PCI bus number>] |
| 216 | |
| 217 | 64k,...,8M - Set the size of each PCI slot's translation table |
| 218 | when using the Calgary IOMMU. This is the size of the translation |
| 219 | table itself in main memory. The smallest table, 64k, covers an IO |
| 220 | space of 32MB; the largest, 8MB table, can cover an IO space of |
| 221 | 4GB. Normally the kernel will make the right choice by itself. |
| 222 | |
| 223 | translate_empty_slots - Enable translation even on slots that have |
| 224 | no devices attached to them, in case a device will be hotplugged |
| 225 | in the future. |
| 226 | |
| 227 | disable=<PCI bus number> - Disable translation on a given PHB. For |
| 228 | example, the built-in graphics adapter resides on the first bridge |
| 229 | (PCI bus number 0); if translation (isolation) is enabled on this |
| 230 | bridge, X servers that access the hardware directly from user |
| 231 | space might stop working. Use this option if you have devices that |
| 232 | are accessed from userspace directly on some PCI host bridge. |
| 233 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 234 | Debugging |
| 235 | |
| 236 | oops=panic Always panic on oopses. Default is to just kill the process, |
| 237 | but there is a small probability of deadlocking the machine. |
| 238 | This will also cause panics on machine check exceptions. |
| 239 | Useful together with panic=30 to trigger a reboot. |
| 240 | |
| 241 | kstack=N Print that many words from the kernel stack in oops dumps. |
| 242 | |
Andi Kleen | 9e43e1b | 2005-11-05 17:25:54 +0100 | [diff] [blame] | 243 | pagefaulttrace Dump all page faults. Only useful for extreme debugging |
| 244 | and will create a lot of output. |
| 245 | |
Andi Kleen | b783fd9 | 2006-07-28 14:44:54 +0200 | [diff] [blame] | 246 | call_trace=[old|both|newfallback|new] |
| 247 | old: use old inexact backtracer |
| 248 | new: use new exact dwarf2 unwinder |
| 249 | both: print entries from both |
| 250 | newfallback: use new unwinder but fall back to old if it gets |
| 251 | stuck (default) |
| 252 | |
Andi Kleen | 538b5b4 | 2006-09-26 10:52:28 +0200 | [diff] [blame] | 253 | call_trace=[old|both|newfallback|new] |
| 254 | old: use old inexact backtracer |
| 255 | new: use new exact dwarf2 unwinder |
| 256 | both: print entries from both |
| 257 | newfallback: use new unwinder but fall back to old if it gets |
| 258 | stuck (default) |
| 259 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 260 | Misc |
| 261 | |
Adrian Bunk | 575c968 | 2006-01-15 02:00:17 +0100 | [diff] [blame] | 262 | noreplacement Don't replace instructions with more appropriate ones |
Andi Kleen | ef4d7cb | 2005-07-28 21:15:34 -0700 | [diff] [blame] | 263 | for the CPU. This may be useful on asymmetric MP systems |
| 264 | where some CPU have less capabilities than the others. |