Darwin boot loader level flags, for Darwin version 8.0

Darwin boot loader level flags, for Darwin version 8.0:

-v = verbose mode. Basically tells you wants happening during boot up.

-x = Safe mode. Basically boots your system with the bare minimum kexts.

-s = Single user mode. Command line only mode. Allows you to run commands as root to fix system.

-f = Tells the machine to reload all kext & dump the boot configuration cache, (kext cache found in: /System/Library/Extensions.mkext, you can delete it manually & the system will recreate it).

“Graphics Mode”= Tells the system what resolution width, height, color depth & refresh rate to boot the OS with.

Ex: “Graphics Mode”=”1024x768x32” WIDTHxHEIGHTxDEPTH

For VESA 3.0 graphics, you may append a refresh rate after an “@” character

Ex: “Graphics Mode”=”640x480x32@60” WIDTHxHEIGHTxDEPTH@REFRESHRATE

rd= This parameter state what is the boot disk to use (instead of using the boot menu appearing before the prompt) you state the drive & partition in here: diskXsY where X stands for the disk number (first disk, usually primary master in IDE or SATA) 0 second disk is 1 etc.) & Y stands for the partition on that disk starting with 1 as the first partition.

Ex: rd=disk0s1 If you have one disk & one partition the parameter will look like this.

You can moreover use rd=*<IODeviceTree path> for booting from a PCI RAID card for example. Ex: of this would be rd=*/PCI0@0/CHN0@0/@0:1

Platform= this parameter sets the platform to use at this boot time.

Examples of this flag are:

platform=ACPI (ACPI support)
platform=X86PC (non ACPI support)
platform=ACPI|86PC (try to support ACPI if fails do not support it)

?memory = this info screen display information approximately the memory on the machine

?video = this info screen display information approximately the video card supported graphic modes

ACPI Flags

acpi=off = Don’t enable ACPI

acpi=ht = Use ACPI boot table parsing, yet don’t enable ACPI interpreter

acpi=force = Force ACPI on (currently not needed)

acpi=strict = Disable out of spec ACPI workarounds.

acpi_sci={edge,level,high,low} Set up ACPI SCI interrupt. EX: acpi_sci=edge

acpi=noirq = Don’t route interrupts

Darwin flags most commonly added to com.apple.Boot.plist file:
Note: Any flag can be added to your boot.plist yet these are just the usual ones.

“Boot Graphics”=Yes = Yes or No. Use graphics mode or text mode when starting. Turns off vesa mode graphics at boot.

“Quiet Boot”=Yes = Yes or No. Use quiet boot mode (no messages or prompt). Same as adding -v option.

Timeout=3 = Any number 1-100. Number of seconds to pause at the boot prompt.

debug=0x144 When added as a boot flag to your com.apple.Boot.plist will donate you details approximately a kernel panic you have at any time when running OSX. I believe it is the same as debug=0x100, not sure.

Unknown Darwin Boot Loader Flags

Text Mode = Might be the same as -v not sure.

Kernel Flags = ?

MKext Cache = Might be to specify the mkext cache to use. Not sure.

Kernel = To specify kernel to use. Correct?

Kernel Cache = Possibly to specify where the kernel should cache?

Boot Device = Might be the same as rd= above. Not sure.

boot-uuid = Believed to be the same as for example rd=disk1s1. Not sure exactly how this is formatted & if it has any other special characteristics. Might be useful to some where the rd flag gives you trouble.

CD-ROM Prompt = ?

CD-ROM Option Key = ?

-F = Not sure could be the same as lower case -f flag.

Kernel level flags:

-l = The flag attempts to enable the L2 cache if not already enabled. Not sure if this works on hacks. If your having an L2 cache issue try this flag.

cpus= Tells the kernel how many cores there are in place. Ex: cpus=1 OR cpus=2

idlehalt= Lets you set two values ether 1 or 0 stating true or false, if set to true then at idle time the cpu will halt causing power saving & cooling of CPU, if set to 0 then the cpu will allways run even in idle time.

idlehalt=0
idlehalt=1

cpuidle= Lets you set two values ether 1 or 0 stating true or false. This flag is exactly the same as the one above I believe. Please correct me if I’m wrong.

cpuidle=0
cpuidle=1

-legacy = causes the system to load in 32 bit mode while running on 64 bit version of OS X

debug=0x100 = To show information approximately kernel panics & other useful info from system at startup. If you are getting a auto rebooting from offensive kernel or kext being loaded use this flag to see what it is. Will assist when posting information on this forum for diagnoses.

maxmem=xxxx = This allows you to specify maximum memory used by the system. Not sure if the rest of the memory is used for apps or not. Many people have to use this if they have 4GB of memory in a 32bit OS. Ex: maxmem=2048

Unknown kernel level flags

dtrace_dof_mode = ?

DisableFBT = ?

IgnoreFBTBlacklist = ?

-s = May sever the same purpose as -s at darwin. Otherwise how would you enter this. May not be a correct flag.

-b = ?

-x = May sever the same purpose as -x at darwin. Otherwise how would you enter this. May not be a correct flag.

srv = ?

ncl = ?

nbuf = ?

kmem = ?

trace = ?

msgbuf = ?

rp = ?

mcache_flags = ?

mbuf_debug = ?

initmcl = ?

socket_debug = ?

net_affinity = ?

rte_debug = ?

-rwroot_hack = ?

keepsyms = ?

mseg = ?

dart = ?

io = ?

nvram_paniclog = ?

pmsafe_debug = ?

preempt = ?

unsafe = = ?

poll = ?

yield = ?

panic_io_port = ?

_fpu = ?

diag = ?

serial = ?

himemory_mode = ?

immediate_NMI = ?

lcks = ?

novmx = ?

max_valid_dma_addr = ?

maxbouncepool = ?

maxloreserve = ?

mtxspin = ?

npvhash = ?

wpkernel = ?

-no_shared_cr3 = ?

-pmap_trace = ?

_panicd_ip = ?

_router_ip = ?

panicd_port = ?

-zc = ?

vmmforce = ?

zsize = ?

colors = ?

fill = ?

serialbaud = ?

net_affinity = ?

rte_debug = ?

fn=x This flag can increase performance & forces OSX to use the users input for CPU/Fan/Idle control. “fn” stands for “forcenap” & provides different implementions for handling CPU idle time with the goal of CPU heat/fan control.

There are five values:
fn=0 : Sets this to off I believe
fn=1 : Not sure what this one does?
fn=2 : Uses CPU instructions “monitor/mwait” (will crash on SSE2 CPUs)
fn=3 : Original implemention of the previous Mac OS X x86 builds. This option sets your energy level to high & causes the fans to run full. This is helpful to keep your processor running fast, yet chews up your battery life. Some users find this option gives the fastest system performance on desktops.
fn=4 : Uses CPU instructions “monitor/mwait” (will crash on SSE2 CPUs)

Kernel level flags used for hacked kernels ONLY 

kernel name “mach_kernel” This flag simply tells the system to boot from another kernel available in / (aka root).

fsb=<mhz> = Most of the hacked kernels include the possibility at boot to chosen FSB frequency. These flags DO NOT work with any known vanilla kernel. Do testing with the below values to see what works for you. The default value is 200Mhz. If you want different value, you have a few possibilities.

Ex: kernel name “netkasSS_kernel”

Ex: fsb=800 or use one of these flags below:

-g = For frequency’s multiplied by 100 Mhz
-y = For frequency’s multiplied by 133 Mhz
-z = For frequency’s multiplied by 166 Mhz

* You can easily add either the -y, -g, or -z options to your boot.plist file or use the fsb=<mhz> flag to donate it an exact figure. I have not tested the fsb=<mhz> flag yet the -y gave me the correct 800fsb (100mhz x 8) & boosted my performance to where it should be in OSX. This may not alter the FSB in approximately this mac on a hack, it did not for me. You would need to test with a benchmark app to see the gain. Hopefully hardcoded front side bus speeds can be added to SMBIOS files at some point in the future by netkas or Mac.Nub.

-vmware = Force vmware support, with hacked kernels. Needs testing.

-force64 = Force 64bit mode for AMD 64 bit cpu’s. Needs testing.

cpu=x = Number of physical cpu’s installed NOT CORES. Also needs testing

Unix Flags that might work in OSX. See subcategories for details

This list is mostly AMD specific boot options & for 64bit CPU’s. Might moreover assist with Intel CPU’s. This list needs extensive testing & feedback yet most of these flags do work in OSX. Once more info is gathered They will be categorized into the lists above

Machine check Options Flags

mce=off = Disable machine check. For compatibility with i386. Might assist boot some AMD systems. nomce does the same same as mce=off

mce=bootlog = Enable logging of machine checks left over from booting.
Disabled by default on AMD because some BIOS leave bogus logs. If your BIOS does not do that it’s a satisfactory idea to enable this log to make sure you log every machine check event that result in a reboot. On Intel systems it is enabled by default.

mce=nobootlog = Disable boot machine check logging.

mce=tolerancelevel (number from list below)

0: always panic on uncorrected errors, log corrected errors
1: panic or SIGBUS on uncorrected errors, log corrected errors. Default is 1
2: SIGBUS or log uncorrected errors, log corrected errors
3: never panic or SIGBUS, log all errors (for testing only)

APIC Flags

apic = Use IO-APIC. Default

noapic = Don’t use the IO-APIC.

disableapic = Don’t use the local APIC

nolapic = Don’t use the local APIC (alias for i386 compatibility)

noapictimer = Don’t set up the APIC timer

no_timer_check = Don’t check the IO-APIC timer. This can work around problems with incorrect timer initialization on some boards.

apicmaintimer = Run time keeping from the local APIC timer instead of using the PIT/HPET interrupt for this. This is useful when the PIT/HPET interrupts are unreliable.

noapicmaintimer = Don’t do time keeping using the APIC timer. Useful when this option was auto selected, yet doesn’t work.

apicpmtimer=1 Either set to 0 or 1. Same as apicmaintimer=1/0. Do APIC timer calibration using the pmtimer. Implies apicmaintimer. Useful when your PIT timer is totally broken. Sometimes there are timer routing problems on some Nvidia & ATI chipsets. Assuming you’re using 64bit then you can try apicmaintimer or apicpmtimer. On 32bit you can try pci=noacpi noapic. Needs testing.

disable_8254_timer / enable_8254_timer = Enable interrupt 0 timer routing over the 8254 in addition to over
the IO-APIC. The kernel tries to set a sensible default.

PCI Flags

pci=off = Don’t use PCI

pci=conf1 = Use conf1 access.

pci=conf2 = Use conf2 access.

pci=rom = Assign ROMs.

pci=assign-busses = Assign busses

pci=noacpi = Don’t use ACPI to set up PCI interrupt routing.

IOMMU is Input/output memory management unit Flags

Currently four x86-64 PCI-DMA mapping implementations exist:

iommu=[1,2,3,4][<size>][,noagp][,off][,force][,noforce][,leak[=<nr_of_leak_pages>]
[,memaper[=<order>]][,merge][,forcesac][,fullflush][,nomerge]
[,noaperture][,calgary]

1-4 consist of the following combined options:

1. <arch/x86_64/kernel/pci-nommu.c>: use no hardware/software IOMMU at all (e.g. because you have < 3 GB memory).
Kernel boot message: “PCI-DMA: Disabling IOMMU”

2. <arch/x86_64/kernel/pci-gart.c>: AMD GART based hardware IOMMU.
Kernel boot message: “PCI-DMA: using GART IOMMU”

3. <arch/x86_64/kernel/pci-swiotlb.c> : Software IOMMU implementation. Used e.g. if there is no hardware IOMMU in the system & it is need because you have >3GB memory or told the kernel to us it (iommu=soft))
Kernel boot message: “PCI-DMA: Using software bounce buffering for IO (SWIOTLB)”

4. <arch/x86_64/pci-calgary.c> : IBM Calgary hardware IOMMU. Used in IBM pSeries & xSeries servers. This hardware IOMMU supports DMA address mapping with memory protection, etc.
Kernel boot message: “PCI-DMA: Using Calgary IOMMU”

General iommu options:

off = Don’t initialize & use any kind of IOMMU.

noforce = Don’t force hardware IOMMU usage when it is not needed. (default).

force = Force the use of the hardware IOMMU even when it is not actually needed (e.g. because < 3 GB memory).

soft = Use software bounce buffering (SWIOTLB) (default for Intel machines). This can be used to prevent the usage
of an available hardware IOMMU.

iommu options only relevant to the AMD GART hardware IOMMU:

<size> Set the size of the remapping area in bytes.

allowed Overwrite iommu off workarounds for specific chipsets.

fullflush Flush IOMMU on each allocation (default).

nofullflush Don’t use IOMMU fullflush.

leak Turn on simple iommu leak tracing (only when CONFIG_IOMMU_LEAK is on). Default number of leak pages is 20.

memaper[=<order>] Allocate an own aperture over RAM with size 32MB<<order.
(default: order=1, i.e. 64MB)

merge Do scatter-gather (SG) merging. Implies “force” (experimental).

nomerge Don’t do scatter-gather (SG) merging.

noaperture Ask the IOMMU not to touch the aperture for AGP.

forcesac Force single-address cycle (SAC) mode for masks <40bits (experimental).

noagp Don’t initialize the AGP driver & use full aperture.

allowdac Allow double-address cycle (DAC) mode, i.e. DMA >4GB. DAC is used with 32-bit PCI to push a 64-bit address in
two cycles. When off all DMA over >4GB is forced through an IOMMU or software bounce buffering.

nodac Forbid DAC mode, i.e. DMA >4GB.

panic Always panic when IOMMU overflows.

calgary Use the Calgary IOMMU if it is available

Early Console Flags, not sure if this exists in OSX

earlyprintk=vga

earlyprintk=serial[,ttySn[,baudrate]] 

EX: earlyprintk=serial,ttyS0,115200 OR earlyprintk=serial,ttyS1,9600

The early console is useful when the kernel crashes before the normal console is initialized. It is not enabled by default because it has some cosmetic problems. Append keep to not disable it when the real console takes over. Only use vga or serial, not both. Currently only ttyS0 & ttyS1 are supported. Interaction with the standard serial driver is not very good. The VGA output is eventually overwritten by the real console.

Timing Flags

notsc = Don’t use the CPU time stamp counter to read the wall time. This can be used to work around timing problems on multiprocessor systems with not properly synchronized CPUs.

report_lost_ticks = Report when timer interrupts are lost because some code turned off interrupts for too long.

nmi_watchdog=“NUMBER”

nmi_watchdog=NUMBER[,panic]

NUMBER can be:
0 = don’t use an NMI watchdog
1 = use the IO-APIC timer for the NMI watchdog
2 = use the local APIC for the NMI watchdog using a performance counter. Note This will use one performance counter & the local APIC’s performance vector.

When ,panic is specified panic when an NMI watchdog timeout occurs. This is useful when you use a ,panic=… timeout & need the box
quickly up again. EX nmi_watchdog=2,panic=20

nohpet = Don’t use the HPET timer.

Idle loop Flags

idle=poll = Don’t do power saving in the idle loop using HLT, yet poll for rescheduling event. This will make the CPUs eat a lot more power, yet may be useful to obtain slightly better performance in multiprocessor benchmarks. It moreover makes some profiling using performance counters more accurate. Please note that on systems with MONITOR/MWAIT support (like Intel EM64T CPUs) this option has no performance advantage over the normal idle loop. It may moreover interact badly with hyperthreading.

Rebooting flags, to control how the system reboots

reboot= b[ios] | t[riple] | k[bd] [, w[arm] | c[old]

bios = Use the CPU reboot vector for warm reset
warm = Don’t set the cold reboot flag
cold = Set the cold reboot flag
triple = Force a triple fault (init)
kbd = Use the keyboard controller. cold reset (default)

Using warm reset will be much faster especially on huge memory systems because the BIOS will not go through the memory check. Disadvantage is that not all hardware will be completely reinitialized on reboot so there may be boot problems on some systems.

reboot=force Don’t stop other CPUs on reboot. This can make reboot more reliable in some cases.

Non Executable Mappings flags

noexec=on/off

on Enable(default)
off Disable

Debugging Flags

oops=panic = Always panic on oopses. Default is to just kill the process, yet there is a small probability of deadlocking the machine.
This will moreover cause panics on machine check exceptions. Useful together with panic=30 to trigger a reboot.

kstack=N = Print N words from the kernel stack in oops dumps.

pagefaulttrace = Dump all page faults. Only useful for extreme debugging & will create a lot of output.

call_trace= [old|both|newfallback|new]

old: use old inexact backtracer
new: use new exact dwarf2 unwinder
both: print entries from both
newfallback: use new unwinder yet fall back to old if it gets stuck (default)

Example of some flags from above that can assist to obtain speedstep to work on a nonhpet AMD system with the ToH 9.2.0 SS kernel

apicpmtimer=1 notsc=1 nohpet=1 cpus=2