summaryrefslogtreecommitdiffstats
path: root/Documentation/kernel-parameters.txt
Commit message (Collapse)AuthorAgeFilesLines
* [PATCH] scheduler cache-hot-autodetectakpm@osdl.org2006-01-121-0/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ) From: Ingo Molnar <mingo@elte.hu> This is the latest version of the scheduler cache-hot-auto-tune patch. The first problem was that detection time scaled with O(N^2), which is unacceptable on larger SMP and NUMA systems. To solve this: - I've added a 'domain distance' function, which is used to cache measurement results. Each distance is only measured once. This means that e.g. on NUMA distances of 0, 1 and 2 might be measured, on HT distances 0 and 1, and on SMP distance 0 is measured. The code walks the domain tree to determine the distance, so it automatically follows whatever hierarchy an architecture sets up. This cuts down on the boot time significantly and removes the O(N^2) limit. The only assumption is that migration costs can be expressed as a function of domain distance - this covers the overwhelming majority of existing systems, and is a good guess even for more assymetric systems. [ People hacking systems that have assymetries that break this assumption (e.g. different CPU speeds) should experiment a bit with the cpu_distance() function. Adding a ->migration_distance factor to the domain structure would be one possible solution - but lets first see the problem systems, if they exist at all. Lets not overdesign. ] Another problem was that only a single cache-size was used for measuring the cost of migration, and most architectures didnt set that variable up. Furthermore, a single cache-size does not fit NUMA hierarchies with L3 caches and does not fit HT setups, where different CPUs will often have different 'effective cache sizes'. To solve this problem: - Instead of relying on a single cache-size provided by the platform and sticking to it, the code now auto-detects the 'effective migration cost' between two measured CPUs, via iterating through a wide range of cachesizes. The code searches for the maximum migration cost, which occurs when the working set of the test-workload falls just below the 'effective cache size'. I.e. real-life optimized search is done for the maximum migration cost, between two real CPUs. This, amongst other things, has the positive effect hat if e.g. two CPUs share a L2/L3 cache, a different (and accurate) migration cost will be found than between two CPUs on the same system that dont share any caches. (The reliable measurement of migration costs is tricky - see the source for details.) Furthermore i've added various boot-time options to override/tune migration behavior. Firstly, there's a blanket override for autodetection: migration_cost=1000,2000,3000 will override the depth 0/1/2 values with 1msec/2msec/3msec values. Secondly, there's a global factor that can be used to increase (or decrease) the autodetected values: migration_factor=120 will increase the autodetected values by 20%. This option is useful to tune things in a workload-dependent way - e.g. if a workload is cache-insensitive then CPU utilization can be maximized by specifying migration_factor=0. I've tested the autodetection code quite extensively on x86, on 3 P3/Xeon/2MB, and the autodetected values look pretty good: Dual Celeron (128K L2 cache): --------------------- migration cost matrix (max_cache_size: 131072, cpu: 467 MHz): --------------------- [00] [01] [00]: - 1.7(1) [01]: 1.7(1) - --------------------- cacheflush times [2]: 0.0 (0) 1.7 (1784008) --------------------- Here the slow memory subsystem dominates system performance, and even though caches are small, the migration cost is 1.7 msecs. Dual HT P4 (512K L2 cache): --------------------- migration cost matrix (max_cache_size: 524288, cpu: 2379 MHz): --------------------- [00] [01] [02] [03] [00]: - 0.4(1) 0.0(0) 0.4(1) [01]: 0.4(1) - 0.4(1) 0.0(0) [02]: 0.0(0) 0.4(1) - 0.4(1) [03]: 0.4(1) 0.0(0) 0.4(1) - --------------------- cacheflush times [2]: 0.0 (33900) 0.4 (448514) --------------------- Here it can be seen that there is no migration cost between two HT siblings (CPU#0/2 and CPU#1/3 are separate physical CPUs). A fast memory system makes inter-physical-CPU migration pretty cheap: 0.4 msecs. 8-way P3/Xeon [2MB L2 cache]: --------------------- migration cost matrix (max_cache_size: 2097152, cpu: 700 MHz): --------------------- [00] [01] [02] [03] [04] [05] [06] [07] [00]: - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) [01]: 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) [02]: 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) [03]: 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) [04]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) [05]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) [06]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) [07]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - --------------------- cacheflush times [2]: 0.0 (0) 19.2 (19281756) --------------------- This one has huge caches and a relatively slow memory subsystem - so the migration cost is 19 msecs. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Ashok Raj <ashok.raj@intel.com> Signed-off-by: Ken Chen <kenneth.w.chen@intel.com> Cc: <wilder@us.ibm.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] kdump: x86_64: add elfcorehdr command line optionVivek Goyal2006-01-101-3/+4
| | | | | | | | | | | | | | | | - elfcorehdr= specifies the location of elf core header stored by the crashed kernel. This command line option will be passed by the kexec-tools to capture kernel. Changes in this version : - Added more comments in kernel-parameters.txt and in code. Signed-off-by: Murali M Chakravarthy <muralim@in.ibm.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Cc: Andi Kleen <ak@muc.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] kdump: x86_64: add memmmap command line optionakpm@osdl.org2006-01-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | ) From: Vivek Goyal <vgoyal@in.ibm.com> - This patch introduces the memmap option for x86_64 similar to i386. - memmap=exactmap enables setting of an exact E820 memory map, as specified by the user. Changes in this version: - Used e820_end_of_ram() to find the max_pfn as suggested by Andi kleen. - removed PFN_UP & PFN_DOWN macros - Printing the user defined map also. Signed-off-by: Murali M Chakravarthy <muralim@in.ibm.com> Signed-off-by: Hariprasad Nellitheertha <nharipra@gmail.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Cc: Andi Kleen <ak@muc.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [SERIAL] Make the number of UARTs registered configurable.Dave Jones2006-01-071-0/+2
| | | | | | | | | | | | Also add a nr_uarts module option to the 8250 code to override this, up to a maximum of CONFIG_SERIAL_8250_NR_UARTS This should appease people who complain about a proliferation of /dev/ttyS & /sysfs nodes whilst at the same time allowing a single kernel image to support the rarer occasions of lots of devices. Signed-off-by: Dave Jones <davej@redhat.com>
* Auto-update from upstreamLen Brown2006-01-071-0/+8
|\
| * NFSv4: Allow entries in the idmap cache to expireTrond Myklebust2006-01-061-0/+4
| | | | | | | | | | | | | | | | | | | | If someone changes the uid/gid mapping in userland, then we do eventually want those changes to be propagated to the kernel. Currently the kernel assumes that it may cache entries forever. Add an expiration time + garbage collector for idmap entries. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
| * NFSv4: Allow user to set the port used by the NFSv4 callback channelTrond Myklebust2006-01-061-0/+4
| | | | | | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* | Auto-update from upstreamLen Brown2006-01-061-0/+8
|\ \ | |/
| * [PATCH] add boot option to control Intel SATA/PATA combined modeJesse Barnes2005-12-131-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Combined mode sucks. Especially when both libata and the legacy IDE drivers try to drive ports on the same device, since that makes DMA rather difficult. This patch addresses the problem by allowing the user to control which driver binds to the ports in a combined mode configuration. In many cases, they'll probably want the libata driver to control both ports since it can use DMA for talking with ATAPI devices (when libata.atapi_enabled=1 of course). It also allows the user to get old school behavior by letting the legacy IDE driver bind to both ports. But neither is forced, the patch doesn't change current behavior unless one of combined_mode=ide or combined_mode=libata is passed on the boot line. Either of those options may require you to access your devices via different device nodes (/dev/hd* in the ide case and /dev/sd* in the libata case), though of course if you have udev installed nicely you may not notice anything. :) Let me know if the documentation is too cryptic, I'd be happy to expand on it if necessary. I think most users will want to boot with 'combined_mode=libata' and add 'options libata atapi_enabled=1' to their modules.conf to get good DVD playing and disk behavior (haven't tested CD or DVD writing though). I'd much rather things behave sanely by default (i.e. DMA for devices on both ports), but apparently that's difficult given the various chip bugs and hardware configs out there (not to mention that people's drives may suddenly change from /dev/hdc to /dev/sdb), so this boot option may be the correct long term fix. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
* | [ACPI] document processor.nocst parameterLen Brown2005-12-281-0/+4
|/ | | | Signed-off-by: Len Brown <len.brown@intel.com>
* [PATCH] mm: m68k kill stram swapHugh Dickins2005-10-291-2/+0
| | | | | | | | | | | | | | | | | | | | | Please, please now delete the Atari CONFIG_STRAM_SWAP code. It may be excellent and ingenious code, but its reference to swap_vfsmnt betrays that it hasn't been built since 2.5.1 (four years old come December), it's delving deep into matters which are the preserve of core mm code, its only purpose is to give the more conscientious mm guys an anxiety attack from time to time; yet we keep on breaking it more and more. If you want to use RAM for swap, then if the MTD driver does not already provide just what you need, I'm sure David could be persuaded to add the extra. But you'd also like to be able to allocate extents of that swap for other use: we can give you a core interface for that if you need. But unbuilt for four years suggests to me that there's no need at all. I cannot swear the patch below won't break your build, but believe so. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] USB: Always do usb-handoffAlan Stern2005-10-281-2/+0
| | | | | | | | | | | | This revised patch (as586b) makes usb-handoff permanently true and no longer a kernel boot parameter. It also removes the piix3_usb quirk code; that was nothing more than an early version of the USB handoff code (written at a time when Intel's PIIX3 was about the only motherboard with USB support). And it adds identifiers for the three PCI USB controller classes to pci_ids.h. Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* [PATCH] kernel-parameters cleanupRandy Dunlap2005-10-231-236/+260
| | | | | | | | | | | | | Fix typos & trailing whitespace. Add blank lines in a few places. Remove "AM53C974=" option: driver does not exist. Restrict to < 80 columns in most places (but don't split formatted command-line arguments). Add a few option arguments for completeness. Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] x86-64: i386/x86-64: Fix time going twice as fast problem on ATI ↵Chuck Ebbert2005-09-121-0/+9
| | | | | | | | | | | Xpress chipsets Original patch from Bertro Simul This is probably still not quite correct, but seems to be the best solution so far. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* Manual merge with LinusDmitry Torokhov2005-09-091-0/+5
|\
| * [PATCH] Add rdinit parameter to pick early userspace initOlof Johansson2005-09-071-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since early userspace was added, there's no way to override which init to run from it. Some people tack on an extra cpio archive with a link from /init depending on what they want to run, but that's sometimes impractical. Changing the "init=" to also override the early userspace isn't feasible, since it is still used to indicate what init to run from disk when early userspace has completed doing whatever it's doing (i.e. load filesystem modules and drivers). Instead, introduce "rdinit=" and make it override the default "/init" if specified. Signed-off-by: Olof Johansson <olof@lixom.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | Input: i8042 - add i8042.nokbd module option to allow supressingDmitry Torokhov2005-09-041-0/+1
|/ | | | | | creation of keyboard port. Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
* Merge ../to-linus-stable/Len Brown2005-08-151-0/+5
|\
| * [ACPI] re-enable platform-specific hotkey drivers by defaultLuming Yu2005-08-151-0/+5
| | | | | | | | | | | | | | | | | | | | When both platform-specific and generic drivers exist, enable generic over-ride with "acpi_generic_hotkey". http://bugzilla.kernel.org/show_bug.cgi?id=4953 Signed-off-by: Luming Yu <luming.yu@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
* | [PATCH] Documentation/kernel-parameters.txt: fix a typoAdrian Bunk2005-07-121-1/+1
| | | | | | | | | | | | Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | [PATCH] DocumentationBob Picco2005-07-071-0/+3
| | | | | | | | | | | | | | This is a small documentation patch for a boot time parameter. Signed-off-by: Bob Picco <bob.picco@hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
* | Remove duplicate mention of "edd" in Documentation/kernel-parameters.txtLinus Torvalds2005-06-291-4/+0
| |
* | [PATCH] irqpollAlan Cox2005-06-281-0/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Anyone reporting a stuck IRQ should try these options. Its effectiveness varies we've found in the Fedora case. Quite a few systems with misdescribed IRQ routing just work when you use irqpoll. It also fixes up the VIA systems although thats now fixed with the VIA quirk (which we could just make default as its what Redmond OS does but Linus didn't like it historically). A small number of systems have jammed IRQ sources or misdescribes that cause an IRQ that we have no handler registered anywhere for. In those cases it doesn't help. Signed-off-by: Alan Cox <number6@the-village.bc.nu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | Merge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6Greg KH2005-06-271-1/+1
|\ \
| * \ Merge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/dtor/input.git manuallyLinus Torvalds2005-06-271-1/+1
| |\ \ | | | | | | | | | | | | Some manual fixups required due to clashes with the PF_FREEZE cleanups.
| | * | Input: Add Fujitsu Lifebook B-series touchscreen driver.Kenan Esau2005-05-291-1/+1
| | |/ | | | | | | | | | | | | | | | From: Kenan Esau <kenan.esau@conan.de> Signed-off-by: Vojtech Pavlik <vojtech@suse.cz> Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
* | | [PATCH] PCI Allow OutOfRange PIRQ table addressjayalk@intworks.biz2005-06-271-0/+4
|/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I updated this to remove unnecessary variable initialization, make check_routing be inline only and not __init, switch to strtoul, and formatting fixes as per Randy Dunlap's recommendations. I updated this to change pirq_table_addr to a long, and to add a warning msg if the PIRQ table wasn't found at the specified address, as per thread with Matthew Wilcox. In our hardware situation, the BIOS is unable to store or generate it's PIRQ table in the F0000h-100000h standard range. This patch adds a pci kernel parameter, pirqaddr to allow the bootloader (or BIOS based loader) to inform the kernel where the PIRQ table got stored. A beneficial side-effect is that, if one's BIOS uses a static address each time for it's PIRQ table, then pirqaddr can be used to avoid the $pirq search through that address block each time at boot for normal PIRQ BIOSes. Signed-off-by: Jaya Kumar <jayalk@intworks.biz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* | [PATCH] Retrieve elfcorehdr address from command lineVivek Goyal2005-06-251-0/+4
| | | | | | | | | | | | | | | | | | This patch adds support for retrieving the address of elf core header if one is passed in command line. Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | [PATCH] kexec: add kexec syscallsEric W. Biederman2005-06-251-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch introduces the architecture independent implementation the sys_kexec_load, the compat_sys_kexec_load system calls. Kexec on panic support has been integrated into the core patch and is relatively clean. In addition the hopefully architecture independent option crashkernel=size@location has been docuemented. It's purpose is to reserve space for the panic kernel to live, and where no DMA transfer will ever be setup to access. Signed-off-by: Eric Biederman <ebiederm@xmission.com> Signed-off-by: Alexander Nyberg <alexn@telia.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | [PATCH] I8K: use standard DMI interfaceDmitry Torokhov2005-06-251-0/+3
|/ | | | | | | | | | | | I8K: Change to use stock dmi infrastructure instead of homegrown parsing code. The driver now requires box's DMI data to match list of supported models so driver can be safely compiled-in by default without fear of it poking into random SMM BIOS code. DMI checks can be ignored with i8k.ignore_dmi option. Signed-off-by: Dmitry Torokhov <dtor@mail.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* Linux-2.6.12-rc2v2.6.12-rc2Linus Torvalds2005-04-161-0/+1511
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
OpenPOWER on IntegriCloud