2007-11-12 22:54:30 +03:00
# Select 32 or 64 bit
config 64BIT
2007-11-17 17:37:31 +03:00
bool "64-bit kernel" if ARCH = "x86"
2012-12-21 01:51:55 +04:00
default ARCH != "i386"
2009-02-05 18:21:53 +03:00
---help---
2007-11-12 22:54:30 +03:00
Say yes to build a 64-bit kernel - formerly known as x86_64
Say no to build a 32-bit kernel - formerly known as i386
config X86_32
2012-09-10 15:41:45 +04:00
def_bool y
depends on !64BIT
2011-05-08 21:55:19 +04:00
select CLKSRC_I8253
2012-10-09 03:28:08 +04:00
select HAVE_UID16
2007-11-12 22:54:30 +03:00
config X86_64
2012-09-10 15:41:45 +04:00
def_bool y
depends on 64BIT
2012-04-04 21:39:58 +04:00
select X86_DEV_DMA_OPS
lockref: implement lockless reference count updates using cmpxchg()
Instead of taking the spinlock, the lockless versions atomically check
that the lock is not taken, and do the reference count update using a
cmpxchg() loop. This is semantically identical to doing the reference
count update protected by the lock, but avoids the "wait for lock"
contention that you get when accesses to the reference count are
contended.
Note that a "lockref" is absolutely _not_ equivalent to an atomic_t.
Even when the lockref reference counts are updated atomically with
cmpxchg, the fact that they also verify the state of the spinlock means
that the lockless updates can never happen while somebody else holds the
spinlock.
So while "lockref_put_or_lock()" looks a lot like just another name for
"atomic_dec_and_lock()", and both optimize to lockless updates, they are
fundamentally different: the decrement done by atomic_dec_and_lock() is
truly independent of any lock (as long as it doesn't decrement to zero),
so a locked region can still see the count change.
The lockref structure, in contrast, really is a *locked* reference
count. If you hold the spinlock, the reference count will be stable and
you can modify the reference count without using atomics, because even
the lockless updates will see and respect the state of the lock.
In order to enable the cmpxchg lockless code, the architecture needs to
do three things:
(1) Make sure that the "arch_spinlock_t" and an "unsigned int" can fit
in an aligned u64, and have a "cmpxchg()" implementation that works
on such a u64 data type.
(2) define a helper function to test for a spinlock being unlocked
("arch_spin_value_unlocked()")
(3) select the "ARCH_USE_CMPXCHG_LOCKREF" config variable in its
Kconfig file.
This enables it for x86-64 (but not 32-bit, we'd need to make sure
cmpxchg() turns into the proper cmpxchg8b in order to enable it for
32-bit mode).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-09-02 23:12:15 +04:00
select ARCH_USE_CMPXCHG_LOCKREF
2007-11-06 23:35:08 +03:00
### Arch settings
2007-11-07 01:30:30 +03:00
config X86
2008-01-30 15:31:03 +03:00
def_bool y
2013-05-01 02:28:42 +04:00
select ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
2013-10-08 06:18:07 +04:00
select ARCH_MIGHT_HAVE_PC_PARPORT
2014-01-01 23:34:16 +04:00
select ARCH_MIGHT_HAVE_PC_SERIO
2008-06-17 15:19:34 +04:00
select HAVE_AOUT if X86_32
2008-05-06 01:19:50 +04:00
select HAVE_UNSTABLE_SCHED_CLOCK
2014-06-05 03:06:29 +04:00
select ARCH_SUPPORTS_NUMA_BALANCING if X86_64
2013-11-18 21:27:06 +04:00
select ARCH_SUPPORTS_INT128 if X86_64
2012-10-25 16:16:43 +04:00
select ARCH_WANTS_PROT_NUMA_PROT_NONE
2008-02-09 12:46:40 +03:00
select HAVE_IDE
2008-02-02 23:10:34 +03:00
select HAVE_OPROFILE
2011-06-01 22:05:09 +04:00
select HAVE_PCSPKR_PLATFORM
2010-11-16 23:49:01 +03:00
select HAVE_PERF_EVENTS
2008-07-24 08:27:05 +04:00
select HAVE_IOREMAP_PROT
2008-02-02 23:10:35 +03:00
select HAVE_KPROBES
2010-08-26 00:39:17 +04:00
select HAVE_MEMBLOCK
2011-07-14 13:44:23 +04:00
select HAVE_MEMBLOCK_NODE_MAP
2011-07-14 13:46:03 +04:00
select ARCH_DISCARD_MEMBLOCK
2008-07-26 15:52:50 +04:00
select ARCH_WANT_OPTIONAL_GPIOLIB
2009-01-07 13:05:10 +03:00
select ARCH_WANT_FRAME_POINTERS
2009-06-18 03:28:12 +04:00
select HAVE_DMA_ATTRS
2014-06-05 03:06:50 +04:00
select HAVE_DMA_CONTIGUOUS
2008-03-05 01:28:37 +03:00
select HAVE_KRETPROBES
2014-04-08 02:39:49 +04:00
select GENERIC_EARLY_IOREMAP
2010-02-25 16:34:46 +03:00
select HAVE_OPTPROBES
2012-09-28 12:15:20 +04:00
select HAVE_KPROBES_ON_FTRACE
2008-08-14 23:45:11 +04:00
select HAVE_FTRACE_MCOUNT_RECORD
ftrace/x86: Add support for -mfentry to x86_64
If the kernel is compiled with gcc 4.6.0 which supports -mfentry,
then use that instead of mcount.
With mcount, frame pointers are forced with the -pg option and we
get something like:
<can_vma_merge_before>:
55 push %rbp
48 89 e5 mov %rsp,%rbp
53 push %rbx
41 51 push %r9
e8 fe 6a 39 00 callq ffffffff81483d00 <mcount>
31 c0 xor %eax,%eax
48 89 fb mov %rdi,%rbx
48 89 d7 mov %rdx,%rdi
48 33 73 30 xor 0x30(%rbx),%rsi
48 f7 c6 ff ff ff f7 test $0xfffffffff7ffffff,%rsi
With -mfentry, frame pointers are no longer forced and the call looks
like this:
<can_vma_merge_before>:
e8 33 af 37 00 callq ffffffff81461b40 <__fentry__>
53 push %rbx
48 89 fb mov %rdi,%rbx
31 c0 xor %eax,%eax
48 89 d7 mov %rdx,%rdi
41 51 push %r9
48 33 73 30 xor 0x30(%rbx),%rsi
48 f7 c6 ff ff ff f7 test $0xfffffffff7ffffff,%rsi
This adds the ftrace hook at the beginning of the function before a
frame is set up, and allows the function callbacks to be able to access
parameters. As kprobes now can use function tracing (at least on x86)
this speeds up the kprobe hooks that are at the beginning of the
function.
Link: http://lkml.kernel.org/r/20120807194100.130477900@goodmis.org
Acked-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2011-02-09 21:32:18 +03:00
select HAVE_FENTRY if X86_64
2010-10-15 07:32:44 +04:00
select HAVE_C_RECORDMCOUNT
2008-05-17 08:01:36 +04:00
select HAVE_DYNAMIC_FTRACE
2012-09-28 12:15:17 +04:00
select HAVE_DYNAMIC_FTRACE_WITH_REGS
2008-10-07 03:06:12 +04:00
select HAVE_FUNCTION_TRACER
2008-12-02 02:20:39 +03:00
select HAVE_FUNCTION_GRAPH_TRACER
function-graph: add stack frame test
In case gcc does something funny with the stack frames, or the return
from function code, we would like to detect that.
An arch may implement passing of a variable that is unique to the
function and can be saved on entering a function and can be tested
when exiting the function. Usually the frame pointer can be used for
this purpose.
This patch also implements this for x86. Where it passes in the stack
frame of the parent function, and will test that frame on exit.
There was a case in x86_32 with optimize for size (-Os) where, for a
few functions, gcc would align the stack frame and place a copy of the
return address into it. The function graph tracer modified the copy and
not the actual return address. On return from the funtion, it did not go
to the tracer hook, but returned to the parent. This broke the function
graph tracer, because the return of the parent (where gcc did not do
this funky manipulation) returned to the location that the child function
was suppose to. This caused strange kernel crashes.
This test detected the problem and pointed out where the issue was.
This modifies the parameters of one of the functions that the arch
specific code calls, so it includes changes to arch code to accommodate
the new prototype.
Note, I notice that the parsic arch implements its own push_return_trace.
This is now a generic function and the ftrace_push_return_trace should be
used instead. This patch does not touch that code.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2009-06-18 20:45:08 +04:00
select HAVE_FUNCTION_GRAPH_FP_TEST
2008-11-06 00:05:44 +03:00
select HAVE_FUNCTION_TRACE_MCOUNT_TEST
2009-08-25 01:43:11 +04:00
select HAVE_SYSCALL_TRACEPOINTS
2012-10-09 03:28:16 +04:00
select SYSCTL_EXCEPTION_TRACE
2009-01-27 19:01:14 +03:00
select HAVE_KVM
2009-01-27 19:02:29 +03:00
select HAVE_ARCH_KGDB
2008-04-21 01:35:12 +04:00
select HAVE_ARCH_TRACEHOOK
2008-06-29 14:19:31 +04:00
select HAVE_GENERIC_DMA_COHERENT if X86_32
2008-07-25 12:45:33 +04:00
select HAVE_EFFICIENT_UNALIGNED_ACCESS
2008-11-23 13:39:08 +03:00
select USER_STACKTRACE_SUPPORT
2010-02-10 19:25:17 +03:00
select HAVE_REGS_AND_STACK_ACCESS_API
2009-01-09 17:13:15 +03:00
select HAVE_DMA_API_DEBUG
2009-01-05 02:41:25 +03:00
select HAVE_KERNEL_GZIP
select HAVE_KERNEL_BZIP2
select HAVE_KERNEL_LZMA
2011-01-13 04:01:24 +03:00
select HAVE_KERNEL_XZ
2010-01-09 01:42:45 +03:00
select HAVE_KERNEL_LZO
2013-07-09 03:01:48 +04:00
select HAVE_KERNEL_LZ4
2009-06-01 22:13:57 +04:00
select HAVE_HW_BREAKPOINT
2010-04-11 20:55:56 +04:00
select HAVE_MIXED_BREAKPOINTS_REGS
2009-12-17 03:33:54 +03:00
select PERF_EVENTS
2010-05-16 00:57:48 +04:00
select HAVE_PERF_EVENTS_NMI
2012-08-07 17:20:36 +04:00
select HAVE_PERF_REGS
2012-08-07 17:20:40 +04:00
select HAVE_PERF_USER_STACK_DUMP
2012-10-09 03:28:11 +04:00
select HAVE_DEBUG_KMEMLEAK
2009-12-17 03:33:54 +03:00
select ANON_INODES
2012-11-28 23:50:23 +04:00
select HAVE_ALIGNED_STRUCT_PAGE if SLUB
select HAVE_CMPXCHG_LOCAL
2012-01-13 05:17:33 +04:00
select HAVE_CMPXCHG_DOUBLE
2009-02-26 22:38:56 +03:00
select HAVE_ARCH_KMEMCHECK
2009-09-19 10:40:22 +04:00
select HAVE_USER_RETURN_NOTIFIER
2012-01-11 03:10:21 +04:00
select ARCH_BINFMT_ELF_RANDOMIZE_PIE
2010-09-23 07:10:23 +04:00
select HAVE_ARCH_JUMP_LABEL
2012-07-31 01:41:09 +04:00
select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
2011-10-12 22:53:17 +04:00
select SPARSE_IRQ
2011-03-08 12:24:26 +03:00
select GENERIC_FIND_FIRST_BIT
2010-09-27 16:46:02 +04:00
select GENERIC_IRQ_PROBE
select GENERIC_PENDING_IRQ if SMP
2010-12-16 19:59:57 +03:00
select GENERIC_IRQ_SHOW
2011-08-23 17:29:42 +04:00
select GENERIC_CLOCKEVENTS_MIN_ADJUST
2011-02-07 04:24:08 +03:00
select IRQ_FORCED_THREADING
2012-05-21 22:45:37 +04:00
select HAVE_BPF_JIT if X86_64
2012-10-09 03:30:04 +04:00
select HAVE_ARCH_TRANSPARENT_HUGEPAGE
2011-06-09 17:08:26 +04:00
select CLKEVT_I8253
2011-07-13 09:14:22 +04:00
select ARCH_HAVE_NMI_SAFE_CMPXCHG
2011-11-24 16:54:28 +04:00
select GENERIC_IOMAP
2012-05-03 21:16:43 +04:00
select DCACHE_WORD_ACCESS
2012-04-20 17:05:48 +04:00
select GENERIC_SMP_IDLE_THREAD
2012-07-31 01:42:46 +04:00
select ARCH_WANT_IPC_PARSE_VERSION if X86_32
2012-04-13 01:48:03 +04:00
select HAVE_ARCH_SECCOMP_FILTER
2012-04-24 22:23:15 +04:00
select BUILDTIME_EXTABLE_SORT
2012-05-18 20:45:44 +04:00
select GENERIC_CMOS_UPDATE
2014-06-05 03:08:16 +04:00
select HAVE_ARCH_SOFT_DIRTY if X86_64
2012-05-18 20:45:44 +04:00
select CLOCKSOURCE_WATCHDOG
select GENERIC_CLOCKEVENTS
2014-03-18 02:22:01 +04:00
select ARCH_CLOCKSOURCE_DATA
2012-05-18 20:45:44 +04:00
select GENERIC_CLOCKEVENTS_BROADCAST if X86_64 || (X86_32 && X86_LOCAL_APIC)
2014-03-18 02:22:01 +04:00
select GENERIC_TIME_VSYSCALL
2012-05-18 20:45:44 +04:00
select KTIME_SCALAR if X86_32
2012-05-26 21:14:39 +04:00
select GENERIC_STRNCPY_FROM_USER
2012-05-26 22:09:53 +04:00
select GENERIC_STRNLEN_USER
2012-11-27 22:33:25 +04:00
select HAVE_CONTEXT_TRACKING if X86_64
2012-09-09 16:56:31 +04:00
select HAVE_IRQ_TIME_ACCOUNTING
2013-03-07 08:48:16 +04:00
select VIRT_TO_BUS
2012-09-28 09:01:03 +04:00
select MODULES_USE_ELF_REL if X86_32
select MODULES_USE_ELF_RELA if X86_64
2012-10-23 06:34:11 +04:00
select CLONE_BACKWARDS if X86_32
2012-12-20 05:16:20 +04:00
select ARCH_USE_BUILTIN_BSWAP
2014-02-03 16:18:57 +04:00
select ARCH_USE_QUEUE_RWLOCK
2012-12-26 01:09:20 +04:00
select OLD_SIGSUSPEND3 if X86_32 || IA32_EMULATION
2012-12-26 04:14:55 +04:00
select OLD_SIGACTION if X86_32
select COMPAT_OLD_SIGACTION if IA32_EMULATION
x86: Do full rtc synchronization with ntp
Every 11 minutes ntp attempts to update the x86 rtc with the current
system time. Currently, the x86 code only updates the rtc if the system
time is within +/-15 minutes of the current value of the rtc. This
was done originally to avoid setting the RTC if the RTC was in localtime
mode (common with Windows dualbooting). Other architectures do a full
synchronization and now that we have better infrastructure to detect
when the RTC is in localtime, there is no reason that x86 should be
software limited to a 30 minute window.
This patch changes the behavior of the kernel to do a full synchronization
(year, month, day, hour, minute, and second) of the rtc when ntp requests
a synchronization between the system time and the rtc.
I've used the RTC library functions in this patchset as they do all the
required bounds checking.
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: x86@kernel.org
Cc: Matt Fleming <matt.fleming@intel.com>
Cc: David Vrabel <david.vrabel@citrix.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: linux-efi@vger.kernel.org
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
[jstultz: Tweak commit message, fold in build fix found by fengguang
Also add select RTC_LIB to X86, per new dependency, as found by prarit]
Signed-off-by: John Stultz <john.stultz@linaro.org>
2013-02-14 21:02:54 +04:00
select RTC_LIB
2013-07-02 00:04:42 +04:00
select HAVE_DEBUG_STACKOVERFLOW
2013-09-24 19:18:36 +04:00
select HAVE_IRQ_EXIT_ON_IRQ_STACK if X86_64
2013-12-19 23:35:58 +04:00
select HAVE_CC_STACKPROTECTOR
2014-02-08 16:34:10 +04:00
select GENERIC_CPU_AUTOPROBE
2014-02-25 13:16:24 +04:00
select HAVE_ARCH_AUDITSYSCALL
2008-02-10 10:16:28 +03:00
2009-06-06 15:58:12 +04:00
config INSTRUCTION_DECODER
2012-09-10 15:41:45 +04:00
def_bool y
depends on KPROBES || PERF_EVENTS || UPROBES
2009-06-06 15:58:12 +04:00
x86: unify arch/x86/boot/compressed/vmlinux_*.lds
Look at the:
diff -u arch/x86/boot/compressed/vmlinux_*.lds
output and realize that they're basially exactly the same except for
trivial naming differences, and the fact that the 64-bit version has a
"pgtable" thing.
So unify them.
There's some trivial cleanup there (make the output format a Kconfig thing
rather than doing #ifdef's for it, and unify both 32-bit and 64-bit BSS
end to "_ebss", where 32-bit used to use the traditional "_end"), but
other than that it's really very mindless and straigt conversion.
For example, I think we should aim to remove "startup_32" vs "startup_64",
and just call it "startup", and get rid of one more difference. I didn't
do that.
Also, notice the comment in the unified vmlinux.lds.S talks about
"head_64" and "startup_32" which is an odd and incorrect mix, but that was
actually what the old 64-bit only lds file had, so the confusion isn't
new, and now that mixing is arguably more accurate thanks to the
vmlinux.lds.S file being shared between the two cases ;)
[ Impact: cleanup, unification ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-26 21:12:47 +04:00
config OUTPUT_FORMAT
string
default "elf32-i386" if X86_32
default "elf64-x86-64" if X86_64
2008-05-26 01:03:18 +04:00
config ARCH_DEFCONFIG
2008-04-29 14:48:15 +04:00
string
2008-05-26 01:03:18 +04:00
default "arch/x86/configs/i386_defconfig" if X86_32
default "arch/x86/configs/x86_64_defconfig" if X86_64
2008-04-29 14:48:15 +04:00
2007-11-07 01:30:30 +03:00
config LOCKDEP_SUPPORT
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-07 01:30:30 +03:00
config STACKTRACE_SUPPORT
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-07 01:30:30 +03:00
2008-02-01 19:45:14 +03:00
config HAVE_LATENCYTOP_SUPPORT
def_bool y
2007-11-07 01:30:30 +03:00
config MMU
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-07 01:30:30 +03:00
config SBUS
bool
2010-03-11 02:23:22 +03:00
config NEED_DMA_MAP_STATE
2012-09-10 15:41:45 +04:00
def_bool y
depends on X86_64 || INTEL_IOMMU || DMA_API_DEBUG
2010-03-11 02:23:22 +03:00
2010-05-27 01:44:32 +04:00
config NEED_SG_DMA_LENGTH
2010-05-27 01:44:33 +04:00
def_bool y
2010-05-27 01:44:32 +04:00
2007-11-07 01:30:30 +03:00
config GENERIC_ISA_DMA
2012-09-10 15:41:45 +04:00
def_bool y
depends on ISA_DMA_API
2007-11-07 01:30:30 +03:00
config GENERIC_BUG
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-07 01:30:30 +03:00
depends on BUG
2008-12-16 14:40:27 +03:00
select GENERIC_BUG_RELATIVE_POINTERS if X86_64
config GENERIC_BUG_RELATIVE_POINTERS
bool
2007-11-07 01:30:30 +03:00
config GENERIC_HWEIGHT
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-07 01:30:30 +03:00
config ARCH_MAY_HAVE_PC_FDC
2012-09-10 15:41:45 +04:00
def_bool y
depends on ISA_DMA_API
2007-11-07 01:30:30 +03:00
2007-11-06 23:35:08 +03:00
config RWSEM_XCHGADD_ALGORITHM
2012-09-10 15:41:45 +04:00
def_bool y
2007-11-06 23:35:08 +03:00
config GENERIC_CALIBRATE_DELAY
def_bool y
2008-02-01 04:35:06 +03:00
config ARCH_HAS_CPU_RELAX
def_bool y
2008-04-28 13:12:22 +04:00
config ARCH_HAS_CACHE_LINE_SIZE
def_bool y
2008-01-30 15:33:32 +03:00
config HAVE_SETUP_PER_CPU_AREA
2009-01-27 06:56:48 +03:00
def_bool y
2008-01-30 15:32:51 +03:00
2009-08-14 10:00:49 +04:00
config NEED_PER_CPU_EMBED_FIRST_CHUNK
def_bool y
config NEED_PER_CPU_PAGE_FIRST_CHUNK
2009-02-20 10:29:09 +03:00
def_bool y
2007-12-08 04:12:39 +03:00
config ARCH_HIBERNATION_POSSIBLE
def_bool y
2007-12-08 04:14:00 +03:00
config ARCH_SUSPEND_POSSIBLE
def_bool y
2013-04-29 17:29:48 +04:00
config ARCH_WANT_HUGE_PMD_SHARE
def_bool y
2013-04-30 11:03:42 +04:00
config ARCH_WANT_GENERAL_HUGETLB
def_bool y
2007-11-07 01:30:30 +03:00
config ZONE_DMA32
bool
default X86_64
config AUDIT_ARCH
bool
default X86_64
2008-04-09 13:03:37 +04:00
config ARCH_SUPPORTS_OPTIMIZED_INLINING
def_bool y
2009-04-01 02:23:17 +04:00
config ARCH_SUPPORTS_DEBUG_PAGEALLOC
def_bool y
2009-09-02 05:25:07 +04:00
config HAVE_INTEL_TXT
def_bool y
2012-10-02 22:16:47 +04:00
depends on INTEL_IOMMU && ACPI
2009-09-02 05:25:07 +04:00
2008-01-30 15:32:27 +03:00
config X86_32_SMP
def_bool y
depends on X86_32 && SMP
config X86_64_SMP
def_bool y
depends on X86_64 && SMP
2007-11-07 01:30:30 +03:00
config X86_HT
2010-04-21 18:23:44 +04:00
def_bool y
2007-12-04 19:19:07 +03:00
depends on SMP
2007-11-07 01:30:30 +03:00
2009-02-09 16:17:40 +03:00
config X86_32_LAZY_GS
def_bool y
2009-02-09 16:17:40 +03:00
depends on X86_32 && !CC_STACKPROTECTOR
2009-02-09 16:17:40 +03:00
2010-03-05 19:34:46 +03:00
config ARCH_HWEIGHT_CFLAGS
string
default "-fcall-saved-ecx -fcall-saved-edx" if X86_32
default "-fcall-saved-rdi -fcall-saved-rsi -fcall-saved-rdx -fcall-saved-rcx -fcall-saved-r8 -fcall-saved-r9 -fcall-saved-r10 -fcall-saved-r11" if X86_64
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 13:26:42 +04:00
config ARCH_SUPPORTS_UPROBES
def_bool y
2014-04-19 02:19:54 +04:00
config FIX_EARLYCON_MEM
def_bool y
2007-11-09 23:56:54 +03:00
source "init/Kconfig"
2008-10-19 07:27:21 +04:00
source "kernel/Kconfig.freezer"
2007-11-07 01:30:30 +03:00
2007-11-09 23:56:54 +03:00
menu "Processor type and features"
2012-01-16 23:57:18 +04:00
config ZONE_DMA
bool "DMA memory allocation support" if EXPERT
default y
help
DMA memory allocation support allows devices with less than 32-bit
addressing to allocate within the first 16MB of address space.
Disable if no such devices will be used.
If unsure, say Y.
2007-11-09 23:56:54 +03:00
config SMP
bool "Symmetric multi-processing support"
---help---
This enables support for systems with more than one CPU. If you have
2014-01-24 03:55:29 +04:00
a system with only one CPU, say N. If you have a system with more
than one CPU, say Y.
2007-11-09 23:56:54 +03:00
2014-01-24 03:55:29 +04:00
If you say N here, the kernel will run on uni- and multiprocessor
2007-11-09 23:56:54 +03:00
machines, but will use only one CPU of a multiprocessor machine. If
you say Y here, the kernel will run on many, but not all,
2014-01-24 03:55:29 +04:00
uniprocessor machines. On a uniprocessor machine, the kernel
2007-11-09 23:56:54 +03:00
will run faster if you say N here.
Note that if you say Y here and choose architecture "586" or
"Pentium" under "Processor family", the kernel will not work on 486
architectures. Similarly, multiprocessor kernels for the "PPro"
architecture may not work on all Pentium based boards.
People using multiprocessor machines who say Y here should also say
Y to "Enhanced Real Time Clock Support", below. The "Advanced Power
Management" code will be disabled if you say Y here.
2011-08-15 04:02:26 +04:00
See also <file:Documentation/x86/i386/IO-APIC.txt>,
2007-11-09 23:56:54 +03:00
<file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at
<http://www.tldp.org/docs.html#howto>.
If you don't know what to do here, say N.
2009-02-17 04:29:58 +03:00
config X86_X2APIC
bool "Support x2apic"
2011-08-24 04:05:25 +04:00
depends on X86_LOCAL_APIC && X86_64 && IRQ_REMAP
2009-02-17 04:29:58 +03:00
---help---
This enables x2apic support on CPUs that have this feature.
This allows 32-bit apic IDs (so it can support very large systems),
and accesses the local apic via MSRs not via mmio.
If you don't know what to do here, say N.
2008-06-19 23:13:09 +04:00
config X86_MPPARSE
2012-10-25 20:35:44 +04:00
bool "Enable MPS table" if ACPI || SFI
2008-10-30 13:38:24 +03:00
default y
2008-07-10 16:42:03 +04:00
depends on X86_LOCAL_APIC
2009-02-05 18:21:53 +03:00
---help---
2008-06-19 23:13:09 +04:00
For old smp systems that do not have proper acpi support. Newer systems
(esp with 64bit cpus) with acpi support, MADT and DSDT will override it
2009-01-30 01:19:22 +03:00
config X86_BIGSMP
bool "Support for big SMP systems with more than 8 CPUs"
depends on X86_32 && SMP
2009-02-05 18:21:53 +03:00
---help---
2009-01-30 01:19:22 +03:00
This option is needed for the systems that have more than 8 CPUs
2007-11-09 23:56:54 +03:00
2013-01-21 21:23:09 +04:00
config GOLDFISH
def_bool y
depends on X86_GOLDFISH
2009-02-21 03:59:11 +03:00
if X86_32
2009-02-10 05:18:14 +03:00
config X86_EXTENDED_PLATFORM
bool "Support for extended (non-PC) x86 platforms"
default y
2009-02-05 18:21:53 +03:00
---help---
2009-01-27 20:11:43 +03:00
If you disable this option then the kernel will only support
standard PC platforms. (which covers the vast majority of
systems out there.)
2009-02-21 03:59:11 +03:00
If you enable this option then you'll be able to select support
for the following (non-PC) 32 bit x86 platforms:
2013-06-24 04:05:25 +04:00
Goldfish (Android emulator)
2009-02-21 03:59:11 +03:00
AMD Elan
RDC R-321x SoC
SGI 320/540 (Visual Workstation)
2012-04-04 21:40:21 +04:00
STA2X11-based (e.g. Northville)
2009-08-29 16:54:20 +04:00
Moorestown MID devices
2009-01-27 20:11:43 +03:00
If you have one of these systems, or if you want to build a
generic distribution kernel, say Y here - otherwise say N.
2009-02-21 03:59:11 +03:00
endif
2009-01-27 20:11:43 +03:00
2009-02-21 03:59:11 +03:00
if X86_64
config X86_EXTENDED_PLATFORM
bool "Support for extended (non-PC) x86 platforms"
default y
---help---
If you disable this option then the kernel will only support
standard PC platforms. (which covers the vast majority of
systems out there.)
If you enable this option then you'll be able to select support
for the following (non-PC) 64 bit x86 platforms:
2011-12-05 20:07:26 +04:00
Numascale NumaChip
2009-02-21 03:59:11 +03:00
ScaleMP vSMP
SGI Ultraviolet
If you have one of these systems, or if you want to build a
generic distribution kernel, say Y here - otherwise say N.
endif
2009-02-10 05:18:14 +03:00
# This is an alphabetically sorted list of 64 bit extended platforms
# Please maintain the alphabetic order if and when there are additions
2011-12-05 20:07:26 +04:00
config X86_NUMACHIP
bool "Numascale NumaChip"
depends on X86_64
depends on X86_EXTENDED_PLATFORM
depends on NUMA
depends on SMP
depends on X86_X2APIC
2012-12-08 01:24:32 +04:00
depends on PCI_MMCONFIG
2011-12-05 20:07:26 +04:00
---help---
Adds support for Numascale NumaChip large-SMP systems. Needed to
enable more than ~168 cores.
If you don't have one of these, you should say N here.
2007-11-09 23:56:54 +03:00
2009-02-10 05:18:14 +03:00
config X86_VSMP
bool "ScaleMP vSMP"
2013-03-05 00:20:21 +04:00
select HYPERVISOR_GUEST
2009-02-10 05:18:14 +03:00
select PARAVIRT
depends on X86_64 && PCI
depends on X86_EXTENDED_PLATFORM
2012-04-16 11:39:35 +04:00
depends on SMP
2009-02-05 18:21:53 +03:00
---help---
2009-02-10 05:18:14 +03:00
Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is
supposed to run on these EM64T-based machines. Only choose this option
if you have one of these machines.
2008-01-30 15:33:36 +03:00
2009-01-20 06:36:04 +03:00
config X86_UV
bool "SGI Ultraviolet"
depends on X86_64
2009-02-10 05:18:14 +03:00
depends on X86_EXTENDED_PLATFORM
2009-04-04 00:39:42 +04:00
depends on NUMA
2009-04-21 00:02:31 +04:00
depends on X86_X2APIC
2009-02-05 18:21:53 +03:00
---help---
2009-01-20 06:36:04 +03:00
This option is needed in order to support SGI Ultraviolet systems.
If you don't have one of these, you should say N here.
2009-02-10 05:18:14 +03:00
# Following is an alphabetically sorted list of 32 bit extended platforms
# Please maintain the alphabetic order if and when there are additions
2007-11-09 23:56:54 +03:00
2013-01-21 21:23:09 +04:00
config X86_GOLDFISH
bool "Goldfish (Virtual Platform)"
2013-06-24 04:05:25 +04:00
depends on X86_EXTENDED_PLATFORM
2013-01-21 21:23:09 +04:00
---help---
Enable support for the Goldfish virtual platform used primarily
for Android development. Unless you are building for the Android
Goldfish emulator say N here.
2010-11-09 23:08:04 +03:00
config X86_INTEL_CE
bool "CE4100 TV platform"
depends on PCI
depends on PCI_GODIRECT
depends on X86_32
depends on X86_EXTENDED_PLATFORM
2010-11-09 23:08:08 +03:00
select X86_REBOOTFIXUPS
2011-02-22 23:07:37 +03:00
select OF
select OF_EARLY_FLATTREE
2011-12-17 02:50:17 +04:00
select IRQ_DOMAIN
2010-11-09 23:08:04 +03:00
---help---
Select for the Intel CE media processor (CE4100) SOC.
This option compiles in support for the CE4100 SOC for settop
boxes and media devices.
2013-12-17 05:37:26 +04:00
config X86_INTEL_MID
2011-07-12 20:49:29 +04:00
bool "Intel MID platform support"
depends on X86_32
depends on X86_EXTENDED_PLATFORM
2014-01-21 22:41:39 +04:00
depends on X86_PLATFORM_DEVICES
2011-11-10 17:29:14 +04:00
depends on PCI
depends on PCI_GOANY
depends on X86_IO_APIC
2011-12-29 18:43:16 +04:00
select SFI
2013-12-17 05:37:26 +04:00
select I2C
2011-12-29 18:43:16 +04:00
select DW_APB_TIMER
2011-11-10 17:29:14 +04:00
select APB_TIMER
select INTEL_SCU_IPC
2012-01-26 21:35:05 +04:00
select MFD_INTEL_MSIC
2011-11-10 17:29:14 +04:00
---help---
2013-12-17 05:37:26 +04:00
Select to build a kernel capable of supporting Intel MID (Mobile
Internet Device) platform systems which do not have the PCI legacy
interfaces. If you are building for a PC class system say N here.
2011-11-10 17:29:14 +04:00
2013-12-17 05:37:26 +04:00
Intel MID platforms are based on an Intel processor and chipset which
consume less power than most of the x86 derivatives.
2011-07-12 20:49:29 +04:00
2013-01-18 17:45:59 +04:00
config X86_INTEL_LPSS
bool "Intel Low Power Subsystem Support"
depends on ACPI
select COMMON_CLK
2013-09-13 18:02:29 +04:00
select PINCTRL
2013-01-18 17:45:59 +04:00
---help---
Select to build support for Intel Low Power Subsystem such as
found on Intel Lynxpoint PCH. Selecting this option enables
2013-09-13 18:02:29 +04:00
things like clock tree (common clock framework) and pincontrol
which are needed by the LPSS peripheral drivers.
2013-01-18 17:45:59 +04:00
2009-02-10 05:18:14 +03:00
config X86_RDC321X
bool "RDC R-321x SoC"
2007-11-09 23:56:54 +03:00
depends on X86_32
2009-02-10 05:18:14 +03:00
depends on X86_EXTENDED_PLATFORM
select M486
select X86_REBOOTFIXUPS
---help---
This option is needed for RDC R-321x system-on-chip, also known
as R-8610-(G).
If you don't have one of these chips, you should say N here.
2009-01-27 20:43:09 +03:00
config X86_32_NON_STANDARD
2009-01-27 20:24:57 +03:00
bool "Support non-standard 32-bit SMP architectures"
depends on X86_32 && SMP
2009-02-10 05:18:14 +03:00
depends on X86_EXTENDED_PLATFORM
2009-02-05 18:21:53 +03:00
---help---
2014-02-26 00:14:06 +04:00
This option compiles in the bigsmp and STA2X11 default
subarchitectures. It is intended for a generic binary
kernel. If you select them all, kernel will probe it one by
one and will fallback to default.
2008-06-09 05:31:54 +04:00
2009-02-10 05:18:14 +03:00
# Alphabetically sorted list of Non standard 32 bit platforms
2008-06-09 05:31:54 +04:00
2009-09-26 20:35:07 +04:00
config X86_SUPPORTS_MEMORY_FAILURE
2010-04-21 18:23:44 +04:00
def_bool y
2009-09-26 20:35:07 +04:00
# MCE code calls memory_failure():
depends on X86_MCE
# On 32-bit this adds too big of NODES_SHIFT and we run out of page flags:
# On 32-bit SPARSEMEM adds too big of SECTIONS_WIDTH:
depends on X86_64 || !SPARSEMEM
select ARCH_SUPPORTS_MEMORY_FAILURE
2012-04-04 21:40:21 +04:00
config STA2X11
bool "STA2X11 Companion Chip Support"
depends on X86_32_NON_STANDARD && PCI
select X86_DEV_DMA_OPS
select X86_DMA_REMAP
select SWIOTLB
select MFD_STA2X11
select ARCH_REQUIRE_GPIOLIB
default n
---help---
This adds support for boards based on the STA2X11 IO-Hub,
a.k.a. "ConneXt". The chip is used in place of the standard
PC chipset, so all "standard" peripherals are missing. If this
option is selected the kernel will still be able to boot on
standard PC machines.
2010-09-25 08:06:57 +04:00
config X86_32_IRIS
tristate "Eurobraille/Iris poweroff module"
depends on X86_32
---help---
The Iris machines from EuroBraille do not have APM or ACPI support
to shut themselves down properly. A special I/O sequence is
needed to do so, which is what this module does at
kernel shutdown.
This is only for Iris machines from EuroBraille.
If unused, say N.
2008-11-11 11:05:16 +03:00
config SCHED_OMIT_FRAME_POINTER
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Single-depth WCHAN output"
2008-11-06 22:10:49 +03:00
depends on X86
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Calculate simpler /proc/<PID>/wchan values. If this option
is disabled then wchan values will recurse back to the
caller function. This provides more accurate wchan values,
at the expense of slightly more scheduling overhead.
If in doubt, say "Y".
2013-03-05 00:20:21 +04:00
menuconfig HYPERVISOR_GUEST
bool "Linux guest support"
2009-02-05 18:21:53 +03:00
---help---
2013-03-05 00:20:21 +04:00
Say Y here to enable options for running Linux under various hyper-
visors. This option enables basic hypervisor detection and platform
setup.
2007-11-09 23:56:54 +03:00
2013-03-05 00:20:21 +04:00
If you say N, all options in this submenu will be skipped and
disabled, and Linux guest support won't be built in.
2007-11-09 23:56:54 +03:00
2013-03-05 00:20:21 +04:00
if HYPERVISOR_GUEST
2007-11-09 23:56:54 +03:00
2008-01-30 15:33:32 +03:00
config PARAVIRT
bool "Enable paravirtualization code"
2009-02-05 18:21:53 +03:00
---help---
2008-01-30 15:33:32 +03:00
This changes the kernel so it can modify itself when it is run
under a hypervisor, potentially improving performance significantly
over full virtualization. However, when run without a hypervisor
the kernel is theoretically slower and slightly larger.
2013-03-05 00:20:21 +04:00
config PARAVIRT_DEBUG
bool "paravirt-ops debugging"
depends on PARAVIRT && DEBUG_KERNEL
---help---
Enable to debug paravirt_ops internals. Specifically, BUG if
a paravirt_op is missing when it is called.
x86: Fix performance regression caused by paravirt_ops on native kernels
Xiaohui Xin and some other folks at Intel have been looking into what's
behind the performance hit of paravirt_ops when running native.
It appears that the hit is entirely due to the paravirtualized
spinlocks introduced by:
| commit 8efcbab674de2bee45a2e4cdf97de16b8e609ac8
| Date: Mon Jul 7 12:07:51 2008 -0700
|
| paravirt: introduce a "lock-byte" spinlock implementation
The extra call/return in the spinlock path is somehow
causing an increase in the cycles/instruction of somewhere around 2-7%
(seems to vary quite a lot from test to test). The working theory is
that the CPU's pipeline is getting upset about the
call->call->locked-op->return->return, and seems to be failing to
speculate (though I haven't seen anything definitive about the precise
reasons). This doesn't entirely make sense, because the performance
hit is also visible on unlock and other operations which don't involve
locked instructions. But spinlock operations clearly swamp all the
other pvops operations, even though I can't imagine that they're
nearly as common (there's only a .05% increase in instructions
executed).
If I disable just the pv-spinlock calls, my tests show that pvops is
identical to non-pvops performance on native (my measurements show that
it is actually about .1% faster, but Xiaohui shows a .05% slowdown).
Summary of results, averaging 10 runs of the "mmperf" test, using a
no-pvops build as baseline:
nopv Pv-nospin Pv-spin
CPU cycles 100.00% 99.89% 102.18%
instructions 100.00% 100.10% 100.15%
CPI 100.00% 99.79% 102.03%
cache ref 100.00% 100.84% 100.28%
cache miss 100.00% 90.47% 88.56%
cache miss rate 100.00% 89.72% 88.31%
branches 100.00% 99.93% 100.04%
branch miss 100.00% 103.66% 107.72%
branch miss rt 100.00% 103.73% 107.67%
wallclock 100.00% 99.90% 102.20%
The clear effect here is that the 2% increase in CPI is
directly reflected in the final wallclock time.
(The other interesting effect is that the more ops are
out of line calls via pvops, the lower the cache access
and miss rates. Not too surprising, but it suggests that
the non-pvops kernel is over-inlined. On the flipside,
the branch misses go up correspondingly...)
So, what's the fix?
Paravirt patching turns all the pvops calls into direct calls, so
_spin_lock etc do end up having direct calls. For example, the compiler
generated code for paravirtualized _spin_lock is:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq *0xffffffff805a5b30
<_spin_lock+22>: retq
The indirect call will get patched to:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq <__ticket_spin_lock>
<_spin_lock+20>: nop; nop /* or whatever 2-byte nop */
<_spin_lock+22>: retq
One possibility is to inline _spin_lock, etc, when building an
optimised kernel (ie, when there's no spinlock/preempt
instrumentation/debugging enabled). That will remove the outer
call/return pair, returning the instruction stream to a single
call/return, which will presumably execute the same as the non-pvops
case. The downsides arel 1) it will replicate the
preempt_disable/enable code at eack lock/unlock callsite; this code is
fairly small, but not nothing; and 2) the spinlock definitions are
already a very heavily tangled mass of #ifdefs and other preprocessor
magic, and making any changes will be non-trivial.
The other obvious answer is to disable pv-spinlocks. Making them a
separate config option is fairly easy, and it would be trivial to
enable them only when Xen is enabled (as the only non-default user).
But it doesn't really address the common case of a distro build which
is going to have Xen support enabled, and leaves the open question of
whether the native performance cost of pv-spinlocks is worth the
performance improvement on a loaded Xen system (10% saving of overall
system CPU when guests block rather than spin). Still it is a
reasonable short-term workaround.
[ Impact: fix pvops performance regression when running native ]
Analysed-by: "Xin Xiaohui" <xiaohui.xin@intel.com>
Analysed-by: "Li Xin" <xin.li@intel.com>
Analysed-by: "Nakajima Jun" <jun.nakajima@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <4A0B62F7.5030802@goop.org>
[ fixed the help text ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-14 04:16:55 +04:00
config PARAVIRT_SPINLOCKS
bool "Paravirtualization layer for spinlocks"
2012-10-02 22:16:47 +04:00
depends on PARAVIRT && SMP
2013-08-09 18:21:50 +04:00
select UNINLINE_SPIN_UNLOCK
x86: Fix performance regression caused by paravirt_ops on native kernels
Xiaohui Xin and some other folks at Intel have been looking into what's
behind the performance hit of paravirt_ops when running native.
It appears that the hit is entirely due to the paravirtualized
spinlocks introduced by:
| commit 8efcbab674de2bee45a2e4cdf97de16b8e609ac8
| Date: Mon Jul 7 12:07:51 2008 -0700
|
| paravirt: introduce a "lock-byte" spinlock implementation
The extra call/return in the spinlock path is somehow
causing an increase in the cycles/instruction of somewhere around 2-7%
(seems to vary quite a lot from test to test). The working theory is
that the CPU's pipeline is getting upset about the
call->call->locked-op->return->return, and seems to be failing to
speculate (though I haven't seen anything definitive about the precise
reasons). This doesn't entirely make sense, because the performance
hit is also visible on unlock and other operations which don't involve
locked instructions. But spinlock operations clearly swamp all the
other pvops operations, even though I can't imagine that they're
nearly as common (there's only a .05% increase in instructions
executed).
If I disable just the pv-spinlock calls, my tests show that pvops is
identical to non-pvops performance on native (my measurements show that
it is actually about .1% faster, but Xiaohui shows a .05% slowdown).
Summary of results, averaging 10 runs of the "mmperf" test, using a
no-pvops build as baseline:
nopv Pv-nospin Pv-spin
CPU cycles 100.00% 99.89% 102.18%
instructions 100.00% 100.10% 100.15%
CPI 100.00% 99.79% 102.03%
cache ref 100.00% 100.84% 100.28%
cache miss 100.00% 90.47% 88.56%
cache miss rate 100.00% 89.72% 88.31%
branches 100.00% 99.93% 100.04%
branch miss 100.00% 103.66% 107.72%
branch miss rt 100.00% 103.73% 107.67%
wallclock 100.00% 99.90% 102.20%
The clear effect here is that the 2% increase in CPI is
directly reflected in the final wallclock time.
(The other interesting effect is that the more ops are
out of line calls via pvops, the lower the cache access
and miss rates. Not too surprising, but it suggests that
the non-pvops kernel is over-inlined. On the flipside,
the branch misses go up correspondingly...)
So, what's the fix?
Paravirt patching turns all the pvops calls into direct calls, so
_spin_lock etc do end up having direct calls. For example, the compiler
generated code for paravirtualized _spin_lock is:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq *0xffffffff805a5b30
<_spin_lock+22>: retq
The indirect call will get patched to:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq <__ticket_spin_lock>
<_spin_lock+20>: nop; nop /* or whatever 2-byte nop */
<_spin_lock+22>: retq
One possibility is to inline _spin_lock, etc, when building an
optimised kernel (ie, when there's no spinlock/preempt
instrumentation/debugging enabled). That will remove the outer
call/return pair, returning the instruction stream to a single
call/return, which will presumably execute the same as the non-pvops
case. The downsides arel 1) it will replicate the
preempt_disable/enable code at eack lock/unlock callsite; this code is
fairly small, but not nothing; and 2) the spinlock definitions are
already a very heavily tangled mass of #ifdefs and other preprocessor
magic, and making any changes will be non-trivial.
The other obvious answer is to disable pv-spinlocks. Making them a
separate config option is fairly easy, and it would be trivial to
enable them only when Xen is enabled (as the only non-default user).
But it doesn't really address the common case of a distro build which
is going to have Xen support enabled, and leaves the open question of
whether the native performance cost of pv-spinlocks is worth the
performance improvement on a loaded Xen system (10% saving of overall
system CPU when guests block rather than spin). Still it is a
reasonable short-term workaround.
[ Impact: fix pvops performance regression when running native ]
Analysed-by: "Xin Xiaohui" <xiaohui.xin@intel.com>
Analysed-by: "Li Xin" <xin.li@intel.com>
Analysed-by: "Nakajima Jun" <jun.nakajima@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <4A0B62F7.5030802@goop.org>
[ fixed the help text ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-14 04:16:55 +04:00
---help---
Paravirtualized spinlocks allow a pvops backend to replace the
spinlock implementation with something virtualization-friendly
(for example, block the virtual CPU rather than spinning).
2013-10-21 20:05:08 +04:00
It has a minimal impact on native kernels and gives a nice performance
benefit on paravirtualized KVM / Xen kernels.
x86: Fix performance regression caused by paravirt_ops on native kernels
Xiaohui Xin and some other folks at Intel have been looking into what's
behind the performance hit of paravirt_ops when running native.
It appears that the hit is entirely due to the paravirtualized
spinlocks introduced by:
| commit 8efcbab674de2bee45a2e4cdf97de16b8e609ac8
| Date: Mon Jul 7 12:07:51 2008 -0700
|
| paravirt: introduce a "lock-byte" spinlock implementation
The extra call/return in the spinlock path is somehow
causing an increase in the cycles/instruction of somewhere around 2-7%
(seems to vary quite a lot from test to test). The working theory is
that the CPU's pipeline is getting upset about the
call->call->locked-op->return->return, and seems to be failing to
speculate (though I haven't seen anything definitive about the precise
reasons). This doesn't entirely make sense, because the performance
hit is also visible on unlock and other operations which don't involve
locked instructions. But spinlock operations clearly swamp all the
other pvops operations, even though I can't imagine that they're
nearly as common (there's only a .05% increase in instructions
executed).
If I disable just the pv-spinlock calls, my tests show that pvops is
identical to non-pvops performance on native (my measurements show that
it is actually about .1% faster, but Xiaohui shows a .05% slowdown).
Summary of results, averaging 10 runs of the "mmperf" test, using a
no-pvops build as baseline:
nopv Pv-nospin Pv-spin
CPU cycles 100.00% 99.89% 102.18%
instructions 100.00% 100.10% 100.15%
CPI 100.00% 99.79% 102.03%
cache ref 100.00% 100.84% 100.28%
cache miss 100.00% 90.47% 88.56%
cache miss rate 100.00% 89.72% 88.31%
branches 100.00% 99.93% 100.04%
branch miss 100.00% 103.66% 107.72%
branch miss rt 100.00% 103.73% 107.67%
wallclock 100.00% 99.90% 102.20%
The clear effect here is that the 2% increase in CPI is
directly reflected in the final wallclock time.
(The other interesting effect is that the more ops are
out of line calls via pvops, the lower the cache access
and miss rates. Not too surprising, but it suggests that
the non-pvops kernel is over-inlined. On the flipside,
the branch misses go up correspondingly...)
So, what's the fix?
Paravirt patching turns all the pvops calls into direct calls, so
_spin_lock etc do end up having direct calls. For example, the compiler
generated code for paravirtualized _spin_lock is:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq *0xffffffff805a5b30
<_spin_lock+22>: retq
The indirect call will get patched to:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq <__ticket_spin_lock>
<_spin_lock+20>: nop; nop /* or whatever 2-byte nop */
<_spin_lock+22>: retq
One possibility is to inline _spin_lock, etc, when building an
optimised kernel (ie, when there's no spinlock/preempt
instrumentation/debugging enabled). That will remove the outer
call/return pair, returning the instruction stream to a single
call/return, which will presumably execute the same as the non-pvops
case. The downsides arel 1) it will replicate the
preempt_disable/enable code at eack lock/unlock callsite; this code is
fairly small, but not nothing; and 2) the spinlock definitions are
already a very heavily tangled mass of #ifdefs and other preprocessor
magic, and making any changes will be non-trivial.
The other obvious answer is to disable pv-spinlocks. Making them a
separate config option is fairly easy, and it would be trivial to
enable them only when Xen is enabled (as the only non-default user).
But it doesn't really address the common case of a distro build which
is going to have Xen support enabled, and leaves the open question of
whether the native performance cost of pv-spinlocks is worth the
performance improvement on a loaded Xen system (10% saving of overall
system CPU when guests block rather than spin). Still it is a
reasonable short-term workaround.
[ Impact: fix pvops performance regression when running native ]
Analysed-by: "Xin Xiaohui" <xiaohui.xin@intel.com>
Analysed-by: "Li Xin" <xin.li@intel.com>
Analysed-by: "Nakajima Jun" <jun.nakajima@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <4A0B62F7.5030802@goop.org>
[ fixed the help text ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-14 04:16:55 +04:00
2013-10-21 20:05:08 +04:00
If you are unsure how to answer this question, answer Y.
x86: Fix performance regression caused by paravirt_ops on native kernels
Xiaohui Xin and some other folks at Intel have been looking into what's
behind the performance hit of paravirt_ops when running native.
It appears that the hit is entirely due to the paravirtualized
spinlocks introduced by:
| commit 8efcbab674de2bee45a2e4cdf97de16b8e609ac8
| Date: Mon Jul 7 12:07:51 2008 -0700
|
| paravirt: introduce a "lock-byte" spinlock implementation
The extra call/return in the spinlock path is somehow
causing an increase in the cycles/instruction of somewhere around 2-7%
(seems to vary quite a lot from test to test). The working theory is
that the CPU's pipeline is getting upset about the
call->call->locked-op->return->return, and seems to be failing to
speculate (though I haven't seen anything definitive about the precise
reasons). This doesn't entirely make sense, because the performance
hit is also visible on unlock and other operations which don't involve
locked instructions. But spinlock operations clearly swamp all the
other pvops operations, even though I can't imagine that they're
nearly as common (there's only a .05% increase in instructions
executed).
If I disable just the pv-spinlock calls, my tests show that pvops is
identical to non-pvops performance on native (my measurements show that
it is actually about .1% faster, but Xiaohui shows a .05% slowdown).
Summary of results, averaging 10 runs of the "mmperf" test, using a
no-pvops build as baseline:
nopv Pv-nospin Pv-spin
CPU cycles 100.00% 99.89% 102.18%
instructions 100.00% 100.10% 100.15%
CPI 100.00% 99.79% 102.03%
cache ref 100.00% 100.84% 100.28%
cache miss 100.00% 90.47% 88.56%
cache miss rate 100.00% 89.72% 88.31%
branches 100.00% 99.93% 100.04%
branch miss 100.00% 103.66% 107.72%
branch miss rt 100.00% 103.73% 107.67%
wallclock 100.00% 99.90% 102.20%
The clear effect here is that the 2% increase in CPI is
directly reflected in the final wallclock time.
(The other interesting effect is that the more ops are
out of line calls via pvops, the lower the cache access
and miss rates. Not too surprising, but it suggests that
the non-pvops kernel is over-inlined. On the flipside,
the branch misses go up correspondingly...)
So, what's the fix?
Paravirt patching turns all the pvops calls into direct calls, so
_spin_lock etc do end up having direct calls. For example, the compiler
generated code for paravirtualized _spin_lock is:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq *0xffffffff805a5b30
<_spin_lock+22>: retq
The indirect call will get patched to:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq <__ticket_spin_lock>
<_spin_lock+20>: nop; nop /* or whatever 2-byte nop */
<_spin_lock+22>: retq
One possibility is to inline _spin_lock, etc, when building an
optimised kernel (ie, when there's no spinlock/preempt
instrumentation/debugging enabled). That will remove the outer
call/return pair, returning the instruction stream to a single
call/return, which will presumably execute the same as the non-pvops
case. The downsides arel 1) it will replicate the
preempt_disable/enable code at eack lock/unlock callsite; this code is
fairly small, but not nothing; and 2) the spinlock definitions are
already a very heavily tangled mass of #ifdefs and other preprocessor
magic, and making any changes will be non-trivial.
The other obvious answer is to disable pv-spinlocks. Making them a
separate config option is fairly easy, and it would be trivial to
enable them only when Xen is enabled (as the only non-default user).
But it doesn't really address the common case of a distro build which
is going to have Xen support enabled, and leaves the open question of
whether the native performance cost of pv-spinlocks is worth the
performance improvement on a loaded Xen system (10% saving of overall
system CPU when guests block rather than spin). Still it is a
reasonable short-term workaround.
[ Impact: fix pvops performance regression when running native ]
Analysed-by: "Xin Xiaohui" <xiaohui.xin@intel.com>
Analysed-by: "Li Xin" <xin.li@intel.com>
Analysed-by: "Nakajima Jun" <jun.nakajima@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <4A0B62F7.5030802@goop.org>
[ fixed the help text ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-14 04:16:55 +04:00
2013-03-05 00:20:21 +04:00
source "arch/x86/xen/Kconfig"
2008-06-03 18:17:29 +04:00
2013-03-05 00:20:21 +04:00
config KVM_GUEST
bool "KVM Guest support (including kvmclock)"
depends on PARAVIRT
select PARAVIRT_CLOCK
default y
2009-02-05 18:21:53 +03:00
---help---
2013-03-05 00:20:21 +04:00
This option enables various optimizations for running under the KVM
hypervisor. It includes a paravirtualized clock, so that instead
of relying on a PIT (or probably other) emulation by the
underlying device model, the host provides the guest with
timing infrastructure such as time of day, and system time
2007-11-09 23:56:54 +03:00
2013-08-09 18:22:01 +04:00
config KVM_DEBUG_FS
bool "Enable debug information for KVM Guests in debugfs"
depends on KVM_GUEST && DEBUG_FS
default n
---help---
This option enables collection of various statistics for KVM guest.
Statistics are displayed in debugfs filesystem. Enabling this option
may incur significant overhead.
2013-03-05 00:20:21 +04:00
source "arch/x86/lguest/Kconfig"
config PARAVIRT_TIME_ACCOUNTING
bool "Paravirtual steal time accounting"
depends on PARAVIRT
default n
2009-02-05 18:21:53 +03:00
---help---
2013-03-05 00:20:21 +04:00
Select this option to enable fine granularity task steal time
accounting. Time spent executing other tasks in parallel with
the current vCPU is discounted from the vCPU power. To account for
that, there can be a small performance impact.
If in doubt, say N here.
config PARAVIRT_CLOCK
bool
2008-06-25 08:19:14 +04:00
2013-03-05 00:20:21 +04:00
endif #HYPERVISOR_GUEST
2008-06-25 08:19:14 +04:00
2010-02-10 12:20:20 +03:00
config NO_BOOTMEM
2010-08-26 00:39:18 +04:00
def_bool y
2010-02-10 12:20:20 +03:00
2008-04-19 04:49:15 +04:00
config MEMTEST
bool "Memtest"
2009-02-05 18:21:53 +03:00
---help---
2008-03-22 04:56:19 +03:00
This option adds a kernel parameter 'memtest', which allows memtest
2008-04-19 04:49:15 +04:00
to be set.
2009-02-05 18:21:53 +03:00
memtest=0, mean disabled; -- default
memtest=1, mean do 1 test pattern;
...
memtest=4, mean do 4 test patterns.
2008-07-15 16:48:48 +04:00
If you are unsure how to answer this question, answer N.
2007-11-09 23:56:54 +03:00
source "arch/x86/Kconfig.cpu"
config HPET_TIMER
2008-01-30 15:31:03 +03:00
def_bool X86_64
2007-11-09 23:56:54 +03:00
prompt "HPET Timer Support" if X86_32
2009-02-05 18:21:53 +03:00
---help---
Use the IA-PC HPET (High Precision Event Timer) to manage
time in preference to the PIT and RTC, if a HPET is
present.
HPET is the next generation timer replacing legacy 8254s.
The HPET provides a stable time base on SMP
systems, unlike the TSC, but it is more expensive to access,
as it is off-chip. You can find the HPET spec at
<http://www.intel.com/hardwaredesign/hpetspec_1.pdf>.
2007-11-09 23:56:54 +03:00
2009-02-05 18:21:53 +03:00
You can safely choose Y here. However, HPET will only be
activated if the platform and the BIOS support this feature.
Otherwise the 8254 will be used for timing services.
2007-11-09 23:56:54 +03:00
2009-02-05 18:21:53 +03:00
Choose N to continue using the legacy 8254 timer.
2007-11-09 23:56:54 +03:00
config HPET_EMULATE_RTC
2008-01-30 15:31:03 +03:00
def_bool y
2008-02-06 12:38:52 +03:00
depends on HPET_TIMER && (RTC=y || RTC=m || RTC_DRV_CMOS=m || RTC_DRV_CMOS=y)
2007-11-09 23:56:54 +03:00
2009-09-02 18:37:17 +04:00
config APB_TIMER
2011-12-17 21:43:40 +04:00
def_bool y if X86_INTEL_MID
prompt "Intel MID APB Timer Support" if X86_INTEL_MID
2011-06-06 15:43:07 +04:00
select DW_APB_TIMER
2011-12-18 01:57:25 +04:00
depends on X86_INTEL_MID && SFI
2009-09-02 18:37:17 +04:00
help
APB timer is the replacement for 8254, HPET on X86 MID platforms.
The APBT provides a stable time base on SMP
systems, unlike the TSC, but it is more expensive to access,
as it is off-chip. APB timers are always running regardless of CPU
C states, they are used as per CPU clockevent device when possible.
2011-01-21 01:44:16 +03:00
# Mark as expert because too many people got it wrong.
2007-11-09 23:56:54 +03:00
# The code disables itself when not needed.
2008-04-28 13:14:14 +04:00
config DMI
default y
2014-01-24 03:54:39 +04:00
select DMI_SCAN_MACHINE_NON_EFI_FALLBACK
2011-01-21 01:44:16 +03:00
bool "Enable DMI scanning" if EXPERT
2009-02-05 18:21:53 +03:00
---help---
2008-04-28 13:14:14 +04:00
Enabled scanning of DMI to identify machine quirks. Say Y
here unless you have verified that your setup is not
affected by entries in the DMI blacklist. Required by PNP
BIOS code.
2007-11-09 23:56:54 +03:00
config GART_IOMMU
2013-10-05 01:37:56 +04:00
bool "Old AMD GART IOMMU support"
2007-11-09 23:56:54 +03:00
select SWIOTLB
2010-09-17 20:03:43 +04:00
depends on X86_64 && PCI && AMD_NB
2009-02-05 18:21:53 +03:00
---help---
2013-10-06 13:45:20 +04:00
Provides a driver for older AMD Athlon64/Opteron/Turion/Sempron
GART based hardware IOMMUs.
The GART supports full DMA access for devices with 32-bit access
limitations, on systems with more than 3 GB. This is usually needed
for USB, sound, many IDE/SATA chipsets and some other devices.
Newer systems typically have a modern AMD IOMMU, supported via
the CONFIG_AMD_IOMMU=y config option.
In normal configurations this driver is only active when needed:
there's more than 3 GB of memory and the system contains a
32-bit limited device.
If unsure, say Y.
2007-11-09 23:56:54 +03:00
config CALGARY_IOMMU
bool "IBM Calgary IOMMU support"
select SWIOTLB
2012-10-02 22:16:47 +04:00
depends on X86_64 && PCI
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Support for hardware IOMMUs in IBM's xSeries x366 and x460
systems. Needed to run systems with more than 3GB of memory
properly with 32-bit PCI devices that do not support DAC
(Double Address Cycle). Calgary also supports bus level
isolation, where all DMAs pass through the IOMMU. This
prevents them from going anywhere except their intended
destination. This catches hard-to-find kernel bugs and
mis-behaving drivers and devices that do not use the DMA-API
properly to set up their DMA buffers. The IOMMU can be
turned off at boot time with the iommu=off parameter.
Normally the kernel will make the right choice by itself.
If unsure, say Y.
config CALGARY_IOMMU_ENABLED_BY_DEFAULT
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Should Calgary be enabled by default?"
2007-11-09 23:56:54 +03:00
depends on CALGARY_IOMMU
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Should Calgary be enabled by default? if you choose 'y', Calgary
will be used (if it exists). If you choose 'n', Calgary will not be
used even if it exists. If you choose 'n' and would like to use
Calgary anyway, pass 'iommu=calgary' on the kernel command line.
If unsure, say Y.
# need this always selected by IOMMU for the VIA workaround
config SWIOTLB
2008-11-18 14:44:21 +03:00
def_bool y if X86_64
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Support for software bounce buffers used on x86-64 systems
2012-09-03 04:38:20 +04:00
which don't have a hardware IOMMU. Using this PCI devices
which can only access 32-bits of memory can be used on systems
with more than 3 GB of memory.
If unsure, say Y.
2007-11-09 23:56:54 +03:00
2008-04-29 11:59:36 +04:00
config IOMMU_HELPER
2012-09-10 15:41:45 +04:00
def_bool y
depends on CALGARY_IOMMU || GART_IOMMU || SWIOTLB || AMD_IOMMU
2008-08-26 01:15:38 +04:00
2008-05-12 23:21:12 +04:00
config MAXSMP
2010-08-21 23:32:41 +04:00
bool "Enable Maximum number of SMP Processors and NUMA Nodes"
2012-10-02 22:16:47 +04:00
depends on X86_64 && SMP && DEBUG_KERNEL
2008-12-17 04:33:51 +03:00
select CPUMASK_OFFSTACK
2009-02-05 18:21:53 +03:00
---help---
2010-08-21 23:32:41 +04:00
Enable maximum number of CPUS and NUMA Nodes for this architecture.
2008-05-12 23:21:12 +04:00
If unsure, say N.
2007-11-09 23:56:54 +03:00
config NR_CPUS
2008-12-17 04:33:51 +03:00
int "Maximum number of CPUs" if SMP && !MAXSMP
2009-04-22 05:44:48 +04:00
range 2 8 if SMP && X86_32 && !X86_BIGSMP
2013-11-05 18:37:29 +04:00
range 2 512 if SMP && !MAXSMP && !CPUMASK_OFFSTACK
2013-11-05 18:38:16 +04:00
range 2 8192 if SMP && !MAXSMP && CPUMASK_OFFSTACK && X86_64
2008-12-17 04:34:00 +03:00
default "1" if !SMP
2013-11-05 18:38:16 +04:00
default "8192" if MAXSMP
2014-02-26 00:14:06 +04:00
default "32" if SMP && X86_BIGSMP
2008-12-17 04:34:00 +03:00
default "8" if SMP
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This allows you to specify the maximum number of CPUs which this
2013-11-05 18:37:29 +04:00
kernel will support. If CPUMASK_OFFSTACK is enabled, the maximum
supported value is 4096, otherwise the maximum value is 512. The
2007-11-09 23:56:54 +03:00
minimum value which makes sense is 2.
This is purely to save memory - each supported CPU adds
approximately eight kilobytes to the kernel image.
config SCHED_SMT
bool "SMT (Hyperthreading) scheduler support"
2008-02-28 00:16:30 +03:00
depends on X86_HT
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
SMT scheduler support improves the CPU scheduler's decision making
when dealing with Intel Pentium 4 chips with HyperThreading at a
cost of slightly increased overhead in some places. If unsure say
N here.
config SCHED_MC
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Multi-core scheduler support"
2008-02-28 00:16:30 +03:00
depends on X86_HT
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Multi-core scheduler support improves the CPU scheduler's decision
making when dealing with multi-core CPU chips at a cost of slightly
increased overhead in some places. If unsure say N here.
source "kernel/Kconfig.preempt"
config X86_UP_APIC
bool "Local APIC support on uniprocessors"
x86, build, pci: Fix PCI_MSI build on !SMP
Commit ebd97be635 ('PCI: remove ARCH_SUPPORTS_MSI kconfig option')
removed the ARCH_SUPPORTS_MSI option which architectures could select
to indicate that they support MSI. Now, all architectures are supposed
to build fine when MSI support is enabled: instead of having the
architecture tell *when* MSI support can be used, it's up to the
architecture code to ensure that MSI support can be enabled.
On x86, commit ebd97be635 removed the following line:
select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC)
Which meant that MSI support was only available when the local APIC
and I/O APIC were enabled. While this is always true on SMP or x86-64,
it is not necessarily the case on i386 !SMP.
The below patch makes sure that the local APIC and I/O APIC support is
always enabled when MSI support is enabled. To do so, it:
* Ensures the X86_UP_APIC option is not visible when PCI_MSI is
enabled. This is the option that allows, on UP machines, to enable
or not the APIC support. It is already not visible on SMP systems,
or x86-64 systems, for example. We're simply also making it
invisible on i386 MSI systems.
* Ensures that the X86_LOCAL_APIC and X86_IO_APIC options are 'y'
when PCI_MSI is enabled.
Notice that this change requires a change in drivers/iommu/Kconfig to
avoid a recursive Kconfig dependencey. The AMD_IOMMU option selects
PCI_MSI, but was depending on X86_IO_APIC. This dependency is no
longer needed: as soon as PCI_MSI is selected, the presence of
X86_IO_APIC is guaranteed. Moreover, the AMD_IOMMU already depended on
X86_64, which already guaranteed that X86_IO_APIC was enabled, so this
dependency was anyway redundant.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Link: http://lkml.kernel.org/r/1380794354-9079-1-git-send-email-thomas.petazzoni@free-electrons.com
Reported-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2013-10-03 13:59:14 +04:00
depends on X86_32 && !SMP && !X86_32_NON_STANDARD && !PCI_MSI
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
A local APIC (Advanced Programmable Interrupt Controller) is an
integrated interrupt controller in the CPU. If you have a single-CPU
system which has a processor with a local APIC, you can say Y here to
enable and use it. If you say Y here even though your machine doesn't
have a local APIC, then the kernel will still run with no slowdown at
all. The local APIC supports CPU-generated self-interrupts (timer,
performance counters), and the NMI watchdog which detects hard
lockups.
config X86_UP_IOAPIC
bool "IO-APIC support on uniprocessors"
depends on X86_UP_APIC
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an
SMP-capable replacement for PC-style interrupt controllers. Most
SMP systems and many recent uniprocessor systems have one.
If you have a single-CPU system with an IO-APIC, you can say Y here
to use it. If you say Y here even though your machine doesn't have
an IO-APIC, then the kernel will still run with no slowdown at all.
config X86_LOCAL_APIC
2008-01-30 15:31:03 +03:00
def_bool y
x86, build, pci: Fix PCI_MSI build on !SMP
Commit ebd97be635 ('PCI: remove ARCH_SUPPORTS_MSI kconfig option')
removed the ARCH_SUPPORTS_MSI option which architectures could select
to indicate that they support MSI. Now, all architectures are supposed
to build fine when MSI support is enabled: instead of having the
architecture tell *when* MSI support can be used, it's up to the
architecture code to ensure that MSI support can be enabled.
On x86, commit ebd97be635 removed the following line:
select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC)
Which meant that MSI support was only available when the local APIC
and I/O APIC were enabled. While this is always true on SMP or x86-64,
it is not necessarily the case on i386 !SMP.
The below patch makes sure that the local APIC and I/O APIC support is
always enabled when MSI support is enabled. To do so, it:
* Ensures the X86_UP_APIC option is not visible when PCI_MSI is
enabled. This is the option that allows, on UP machines, to enable
or not the APIC support. It is already not visible on SMP systems,
or x86-64 systems, for example. We're simply also making it
invisible on i386 MSI systems.
* Ensures that the X86_LOCAL_APIC and X86_IO_APIC options are 'y'
when PCI_MSI is enabled.
Notice that this change requires a change in drivers/iommu/Kconfig to
avoid a recursive Kconfig dependencey. The AMD_IOMMU option selects
PCI_MSI, but was depending on X86_IO_APIC. This dependency is no
longer needed: as soon as PCI_MSI is selected, the presence of
X86_IO_APIC is guaranteed. Moreover, the AMD_IOMMU already depended on
X86_64, which already guaranteed that X86_IO_APIC was enabled, so this
dependency was anyway redundant.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Link: http://lkml.kernel.org/r/1380794354-9079-1-git-send-email-thomas.petazzoni@free-electrons.com
Reported-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2013-10-03 13:59:14 +04:00
depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC || PCI_MSI
2007-11-09 23:56:54 +03:00
config X86_IO_APIC
2008-01-30 15:31:03 +03:00
def_bool y
x86, build, pci: Fix PCI_MSI build on !SMP
Commit ebd97be635 ('PCI: remove ARCH_SUPPORTS_MSI kconfig option')
removed the ARCH_SUPPORTS_MSI option which architectures could select
to indicate that they support MSI. Now, all architectures are supposed
to build fine when MSI support is enabled: instead of having the
architecture tell *when* MSI support can be used, it's up to the
architecture code to ensure that MSI support can be enabled.
On x86, commit ebd97be635 removed the following line:
select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC)
Which meant that MSI support was only available when the local APIC
and I/O APIC were enabled. While this is always true on SMP or x86-64,
it is not necessarily the case on i386 !SMP.
The below patch makes sure that the local APIC and I/O APIC support is
always enabled when MSI support is enabled. To do so, it:
* Ensures the X86_UP_APIC option is not visible when PCI_MSI is
enabled. This is the option that allows, on UP machines, to enable
or not the APIC support. It is already not visible on SMP systems,
or x86-64 systems, for example. We're simply also making it
invisible on i386 MSI systems.
* Ensures that the X86_LOCAL_APIC and X86_IO_APIC options are 'y'
when PCI_MSI is enabled.
Notice that this change requires a change in drivers/iommu/Kconfig to
avoid a recursive Kconfig dependencey. The AMD_IOMMU option selects
PCI_MSI, but was depending on X86_IO_APIC. This dependency is no
longer needed: as soon as PCI_MSI is selected, the presence of
X86_IO_APIC is guaranteed. Moreover, the AMD_IOMMU already depended on
X86_64, which already guaranteed that X86_IO_APIC was enabled, so this
dependency was anyway redundant.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Link: http://lkml.kernel.org/r/1380794354-9079-1-git-send-email-thomas.petazzoni@free-electrons.com
Reported-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2013-10-03 13:59:14 +04:00
depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_IOAPIC || PCI_MSI
2014-05-07 19:44:06 +04:00
select GENERIC_IRQ_LEGACY_ALLOC_HWIRQ
2007-11-09 23:56:54 +03:00
2008-07-15 15:48:55 +04:00
config X86_REROUTE_FOR_BROKEN_BOOT_IRQS
bool "Reroute for broken boot IRQs"
depends on X86_IO_APIC
2009-02-05 18:21:53 +03:00
---help---
2008-07-15 15:48:55 +04:00
This option enables a workaround that fixes a source of
spurious interrupts. This is recommended when threaded
interrupt handling is used on systems where the generation of
superfluous "boot interrupts" cannot be disabled.
Some chipsets generate a legacy INTx "boot IRQ" when the IRQ
entry in the chipset's IO-APIC is masked (as, e.g. the RT
kernel does during interrupt handling). On chipsets where this
boot IRQ generation cannot be disabled, this workaround keeps
the original IRQ line masked so that only the equivalent "boot
IRQ" is delivered to the CPUs. The workaround also tells the
kernel to set up the IRQ handler on the boot IRQ line. In this
way only one interrupt is delivered to the kernel. Otherwise
the spurious second interrupt may cause the kernel to bring
down (vital) interrupt lines.
Only affects "broken" chipsets. Interrupt sharing may be
increased on these systems.
2007-11-09 23:56:54 +03:00
config X86_MCE
2009-07-09 02:31:38 +04:00
bool "Machine Check / overheating reporting"
2011-09-13 17:23:21 +04:00
default y
2007-11-09 23:56:54 +03:00
---help---
2009-07-09 02:31:38 +04:00
Machine Check support allows the processor to notify the
kernel if it detects a problem (e.g. overheating, data corruption).
2007-11-09 23:56:54 +03:00
The action the kernel takes depends on the severity of the problem,
2009-07-09 02:31:38 +04:00
ranging from warning messages to halting the machine.
x86, mce: use 64bit machine check code on 32bit
The 64bit machine check code is in many ways much better than
the 32bit machine check code: it is more specification compliant,
is cleaner, only has a single code base versus one per CPU,
has better infrastructure for recovery, has a cleaner way to communicate
with user space etc. etc.
Use the 64bit code for 32bit too.
This is the second attempt to do this. There was one a couple of years
ago to unify this code for 32bit and 64bit. Back then this ran into some
trouble with K7s and was reverted.
I believe this time the K7 problems (and some others) are addressed.
I went over the old handlers and was very careful to retain
all quirks.
But of course this needs a lot of testing on old systems. On newer
64bit capable systems I don't expect much problems because they have been
already tested with the 64bit kernel.
I made this a CONFIG for now that still allows to select the old
machine check code. This is mostly to make testing easier,
if someone runs into a problem we can ask them to try
with the CONFIG switched.
The new code is default y for more coverage.
Once there is confidence the 64bit code works well on older hardware
too the CONFIG_X86_OLD_MCE and the associated code can be easily
removed.
This causes a behaviour change for 32bit installations. They now
have to install the mcelog package to be able to log
corrected machine checks.
The 64bit machine check code only handles CPUs which support the
standard Intel machine check architecture described in the IA32 SDM.
The 32bit code has special support for some older CPUs which
have non standard machine check architectures, in particular
WinChip C3 and Intel P5. I made those a separate CONFIG option
and kept them for now. The WinChip variant could be probably
removed without too much pain, it doesn't really do anything
interesting. P5 is also disabled by default (like it
was before) because many motherboards have it miswired, but
according to Alan Cox a few embedded setups use that one.
Forward ported/heavily changed version of old patch, original patch
included review/fixes from Thomas Gleixner, Bert Wesarg.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-04-28 21:07:31 +04:00
2007-11-09 23:56:54 +03:00
config X86_MCE_INTEL
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Intel MCE features"
2009-07-09 02:31:41 +04:00
depends on X86_MCE && X86_LOCAL_APIC
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Additional support for intel specific MCE features such as
the thermal monitor.
config X86_MCE_AMD
2008-01-30 15:31:03 +03:00
def_bool y
prompt "AMD MCE features"
2009-07-09 02:31:41 +04:00
depends on X86_MCE && X86_LOCAL_APIC
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Additional support for AMD specific MCE features such as
the DRAM Error Threshold.
x86, mce: use 64bit machine check code on 32bit
The 64bit machine check code is in many ways much better than
the 32bit machine check code: it is more specification compliant,
is cleaner, only has a single code base versus one per CPU,
has better infrastructure for recovery, has a cleaner way to communicate
with user space etc. etc.
Use the 64bit code for 32bit too.
This is the second attempt to do this. There was one a couple of years
ago to unify this code for 32bit and 64bit. Back then this ran into some
trouble with K7s and was reverted.
I believe this time the K7 problems (and some others) are addressed.
I went over the old handlers and was very careful to retain
all quirks.
But of course this needs a lot of testing on old systems. On newer
64bit capable systems I don't expect much problems because they have been
already tested with the 64bit kernel.
I made this a CONFIG for now that still allows to select the old
machine check code. This is mostly to make testing easier,
if someone runs into a problem we can ask them to try
with the CONFIG switched.
The new code is default y for more coverage.
Once there is confidence the 64bit code works well on older hardware
too the CONFIG_X86_OLD_MCE and the associated code can be easily
removed.
This causes a behaviour change for 32bit installations. They now
have to install the mcelog package to be able to log
corrected machine checks.
The 64bit machine check code only handles CPUs which support the
standard Intel machine check architecture described in the IA32 SDM.
The 32bit code has special support for some older CPUs which
have non standard machine check architectures, in particular
WinChip C3 and Intel P5. I made those a separate CONFIG option
and kept them for now. The WinChip variant could be probably
removed without too much pain, it doesn't really do anything
interesting. P5 is also disabled by default (like it
was before) because many motherboards have it miswired, but
according to Alan Cox a few embedded setups use that one.
Forward ported/heavily changed version of old patch, original patch
included review/fixes from Thomas Gleixner, Bert Wesarg.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-04-28 21:07:31 +04:00
config X86_ANCIENT_MCE
2010-04-21 18:23:44 +04:00
bool "Support for old Pentium 5 / WinChip machine checks"
2009-07-09 02:31:37 +04:00
depends on X86_32 && X86_MCE
2009-05-27 11:57:31 +04:00
---help---
Include support for machine check handling on old Pentium 5 or WinChip
2013-11-30 16:38:43 +04:00
systems. These typically need to be enabled explicitly on the command
2009-05-27 11:57:31 +04:00
line.
x86, mce: use 64bit machine check code on 32bit
The 64bit machine check code is in many ways much better than
the 32bit machine check code: it is more specification compliant,
is cleaner, only has a single code base versus one per CPU,
has better infrastructure for recovery, has a cleaner way to communicate
with user space etc. etc.
Use the 64bit code for 32bit too.
This is the second attempt to do this. There was one a couple of years
ago to unify this code for 32bit and 64bit. Back then this ran into some
trouble with K7s and was reverted.
I believe this time the K7 problems (and some others) are addressed.
I went over the old handlers and was very careful to retain
all quirks.
But of course this needs a lot of testing on old systems. On newer
64bit capable systems I don't expect much problems because they have been
already tested with the 64bit kernel.
I made this a CONFIG for now that still allows to select the old
machine check code. This is mostly to make testing easier,
if someone runs into a problem we can ask them to try
with the CONFIG switched.
The new code is default y for more coverage.
Once there is confidence the 64bit code works well on older hardware
too the CONFIG_X86_OLD_MCE and the associated code can be easily
removed.
This causes a behaviour change for 32bit installations. They now
have to install the mcelog package to be able to log
corrected machine checks.
The 64bit machine check code only handles CPUs which support the
standard Intel machine check architecture described in the IA32 SDM.
The 32bit code has special support for some older CPUs which
have non standard machine check architectures, in particular
WinChip C3 and Intel P5. I made those a separate CONFIG option
and kept them for now. The WinChip variant could be probably
removed without too much pain, it doesn't really do anything
interesting. P5 is also disabled by default (like it
was before) because many motherboards have it miswired, but
according to Alan Cox a few embedded setups use that one.
Forward ported/heavily changed version of old patch, original patch
included review/fixes from Thomas Gleixner, Bert Wesarg.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-04-28 21:07:31 +04:00
2009-02-12 15:49:31 +03:00
config X86_MCE_THRESHOLD
depends on X86_MCE_AMD || X86_MCE_INTEL
2010-04-21 18:23:44 +04:00
def_bool y
2009-02-12 15:49:31 +03:00
2009-04-29 21:31:00 +04:00
config X86_MCE_INJECT
2009-07-09 02:31:41 +04:00
depends on X86_MCE
2009-04-29 21:31:00 +04:00
tristate "Machine check injector support"
---help---
Provide support for injecting machine checks for testing purposes.
If you don't know what a machine check is and you don't do kernel
QA it is safe to say n.
x86, mce: use 64bit machine check code on 32bit
The 64bit machine check code is in many ways much better than
the 32bit machine check code: it is more specification compliant,
is cleaner, only has a single code base versus one per CPU,
has better infrastructure for recovery, has a cleaner way to communicate
with user space etc. etc.
Use the 64bit code for 32bit too.
This is the second attempt to do this. There was one a couple of years
ago to unify this code for 32bit and 64bit. Back then this ran into some
trouble with K7s and was reverted.
I believe this time the K7 problems (and some others) are addressed.
I went over the old handlers and was very careful to retain
all quirks.
But of course this needs a lot of testing on old systems. On newer
64bit capable systems I don't expect much problems because they have been
already tested with the 64bit kernel.
I made this a CONFIG for now that still allows to select the old
machine check code. This is mostly to make testing easier,
if someone runs into a problem we can ask them to try
with the CONFIG switched.
The new code is default y for more coverage.
Once there is confidence the 64bit code works well on older hardware
too the CONFIG_X86_OLD_MCE and the associated code can be easily
removed.
This causes a behaviour change for 32bit installations. They now
have to install the mcelog package to be able to log
corrected machine checks.
The 64bit machine check code only handles CPUs which support the
standard Intel machine check architecture described in the IA32 SDM.
The 32bit code has special support for some older CPUs which
have non standard machine check architectures, in particular
WinChip C3 and Intel P5. I made those a separate CONFIG option
and kept them for now. The WinChip variant could be probably
removed without too much pain, it doesn't really do anything
interesting. P5 is also disabled by default (like it
was before) because many motherboards have it miswired, but
according to Alan Cox a few embedded setups use that one.
Forward ported/heavily changed version of old patch, original patch
included review/fixes from Thomas Gleixner, Bert Wesarg.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-04-28 21:07:31 +04:00
config X86_THERMAL_VECTOR
def_bool y
2009-07-09 02:31:39 +04:00
depends on X86_MCE_INTEL
x86, mce: use 64bit machine check code on 32bit
The 64bit machine check code is in many ways much better than
the 32bit machine check code: it is more specification compliant,
is cleaner, only has a single code base versus one per CPU,
has better infrastructure for recovery, has a cleaner way to communicate
with user space etc. etc.
Use the 64bit code for 32bit too.
This is the second attempt to do this. There was one a couple of years
ago to unify this code for 32bit and 64bit. Back then this ran into some
trouble with K7s and was reverted.
I believe this time the K7 problems (and some others) are addressed.
I went over the old handlers and was very careful to retain
all quirks.
But of course this needs a lot of testing on old systems. On newer
64bit capable systems I don't expect much problems because they have been
already tested with the 64bit kernel.
I made this a CONFIG for now that still allows to select the old
machine check code. This is mostly to make testing easier,
if someone runs into a problem we can ask them to try
with the CONFIG switched.
The new code is default y for more coverage.
Once there is confidence the 64bit code works well on older hardware
too the CONFIG_X86_OLD_MCE and the associated code can be easily
removed.
This causes a behaviour change for 32bit installations. They now
have to install the mcelog package to be able to log
corrected machine checks.
The 64bit machine check code only handles CPUs which support the
standard Intel machine check architecture described in the IA32 SDM.
The 32bit code has special support for some older CPUs which
have non standard machine check architectures, in particular
WinChip C3 and Intel P5. I made those a separate CONFIG option
and kept them for now. The WinChip variant could be probably
removed without too much pain, it doesn't really do anything
interesting. P5 is also disabled by default (like it
was before) because many motherboards have it miswired, but
according to Alan Cox a few embedded setups use that one.
Forward ported/heavily changed version of old patch, original patch
included review/fixes from Thomas Gleixner, Bert Wesarg.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-04-28 21:07:31 +04:00
2007-11-09 23:56:54 +03:00
config VM86
2011-01-21 01:44:16 +03:00
bool "Enable VM86 support" if EXPERT
2007-11-09 23:56:54 +03:00
default y
depends on X86_32
2009-02-05 18:21:53 +03:00
---help---
2014-05-04 21:36:22 +04:00
This option is required by programs like DOSEMU to run
16-bit real mode legacy code on x86 processors. It also may
be needed by software like XFree86 to initialize some video
cards via BIOS. Disabling this option saves about 6K.
config X86_16BIT
bool "Enable support for 16-bit segments" if EXPERT
default y
---help---
This option is required by programs like Wine to run 16-bit
protected mode legacy code on x86 processors. Disabling
this option saves about 300 bytes on i386, or around 6K text
plus 16K runtime memory on x86-64,
config X86_ESPFIX32
def_bool y
depends on X86_16BIT && X86_32
2007-11-09 23:56:54 +03:00
2014-05-04 21:00:49 +04:00
config X86_ESPFIX64
def_bool y
2014-05-04 21:36:22 +04:00
depends on X86_16BIT && X86_64
2007-11-09 23:56:54 +03:00
config TOSHIBA
tristate "Toshiba Laptop support"
depends on X86_32
---help---
This adds a driver to safely access the System Management Mode of
the CPU on Toshiba portables with a genuine Toshiba BIOS. It does
not work on models with a Phoenix BIOS. The System Management Mode
is used to set the BIOS and power saving options on Toshiba portables.
For information on utilities to make use of this driver see the
Toshiba Linux utilities web site at:
<http://www.buzzard.org.uk/toshiba/>.
Say Y if you intend to run this kernel on a Toshiba portable.
Say N otherwise.
config I8K
tristate "Dell laptop support"
2011-05-25 22:43:33 +04:00
select HWMON
2007-11-09 23:56:54 +03:00
---help---
This adds a driver to safely access the System Management Mode
of the CPU on the Dell Inspiron 8000. The System Management Mode
is used to read cpu temperature and cooling fan status and to
control the fans on the I8K portables.
This driver has been tested only on the Inspiron 8000 but it may
also work with other Dell laptops. You can force loading on other
models by passing the parameter `force=1' to the module. Use at
your own risk.
For information on utilities to make use of this driver see the
I8K Linux utilities web site at:
<http://people.debian.org/~dz/i8k/>
Say Y if you intend to run this kernel on a Dell Inspiron 8000.
Say N otherwise.
config X86_REBOOTFIXUPS
2008-10-16 09:01:38 +04:00
bool "Enable X86 board specific fixups for reboot"
depends on X86_32
2007-11-09 23:56:54 +03:00
---help---
This enables chipset and/or board specific fixups to be done
in order to get reboot to work correctly. This is only needed on
some combinations of hardware and BIOS. The symptom, for which
this config is intended, is when reboot ends with a stalled/hung
system.
Currently, the only fixup is for the Geode machines using
2008-01-30 15:33:36 +03:00
CS5530A and CS5536 chipsets and the RDC R-321x SoC.
2007-11-09 23:56:54 +03:00
Say Y if you want to enable the fixup. Currently, it's safe to
enable this option even if you don't need it.
Say N otherwise.
config MICROCODE
2012-08-01 21:17:01 +04:00
tristate "CPU microcode loading support"
2013-10-13 20:36:29 +04:00
depends on CPU_SUP_AMD || CPU_SUP_INTEL
2007-11-09 23:56:54 +03:00
select FW_LOADER
---help---
2012-08-01 21:17:01 +04:00
2007-11-09 23:56:54 +03:00
If you say Y here, you will be able to update the microcode on
2008-07-28 20:44:22 +04:00
certain Intel and AMD processors. The Intel support is for the
2012-08-01 21:17:01 +04:00
IA32 family, e.g. Pentium Pro, Pentium II, Pentium III, Pentium 4,
Xeon etc. The AMD support is for families 0x10 and later. You will
obviously need the actual microcode binary data itself which is not
shipped with the Linux kernel.
2007-11-09 23:56:54 +03:00
2008-07-28 20:44:21 +04:00
This option selects the general module only, you need to select
at least one vendor specific module as well.
2007-11-09 23:56:54 +03:00
2012-08-01 21:17:01 +04:00
To compile this driver as a module, choose M here: the module
will be called microcode.
2007-11-09 23:56:54 +03:00
2008-07-28 20:44:21 +04:00
config MICROCODE_INTEL
2012-08-01 21:17:01 +04:00
bool "Intel microcode loading support"
2009-02-05 18:21:53 +03:00
depends on MICROCODE
default MICROCODE
select FW_LOADER
---help---
This options enables microcode patch loading support for Intel
processors.
2014-01-20 22:01:56 +04:00
For the current Intel microcode data package go to
<https://downloadcenter.intel.com> and search for
'Linux Processor Microcode Data File'.
2008-07-28 20:44:21 +04:00
2008-07-28 20:44:22 +04:00
config MICROCODE_AMD
2012-08-01 21:17:01 +04:00
bool "AMD microcode loading support"
2009-02-05 18:21:53 +03:00
depends on MICROCODE
select FW_LOADER
---help---
If you select this option, microcode patch loading support for AMD
processors will be enabled.
2008-07-28 20:44:22 +04:00
2009-02-05 18:21:53 +03:00
config MICROCODE_OLD_INTERFACE
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-09 23:56:54 +03:00
depends on MICROCODE
2012-12-21 11:44:32 +04:00
config MICROCODE_INTEL_EARLY
2013-05-30 23:09:19 +04:00
def_bool n
config MICROCODE_AMD_EARLY
def_bool n
config MICROCODE_EARLY
2012-12-21 11:44:32 +04:00
bool "Early load microcode"
2013-05-31 10:53:24 +04:00
depends on MICROCODE=y && BLK_DEV_INITRD
2013-05-30 23:09:19 +04:00
select MICROCODE_INTEL_EARLY if MICROCODE_INTEL
select MICROCODE_AMD_EARLY if MICROCODE_AMD
2012-12-21 11:44:32 +04:00
default y
help
This option provides functionality to read additional microcode data
at the beginning of initrd image. The data tells kernel to load
microcode to CPU's as early as possible. No functional change if no
microcode data is glued to the initrd, therefore it's safe to say Y.
2007-11-09 23:56:54 +03:00
config X86_MSR
tristate "/dev/cpu/*/msr - Model-specific register support"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This device gives privileged processes access to the x86
Model-Specific Registers (MSRs). It is a character device with
major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr.
MSR accesses are directed to a specific CPU on multi-processor
systems.
config X86_CPUID
tristate "/dev/cpu/*/cpuid - CPU information support"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This device gives processes access to the x86 CPUID instruction to
be executed on a specific processor. It is a character device
with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
/dev/cpu/31/cpuid.
choice
prompt "High Memory Support"
2010-04-21 18:23:44 +04:00
default HIGHMEM4G
2007-11-09 23:56:54 +03:00
depends on X86_32
config NOHIGHMEM
bool "off"
---help---
Linux can use up to 64 Gigabytes of physical memory on x86 systems.
However, the address space of 32-bit x86 processors is only 4
Gigabytes large. That means that, if you have a large amount of
physical memory, not all of it can be "permanently mapped" by the
kernel. The physical memory that's not permanently mapped is called
"high memory".
If you are compiling a kernel which will never run on a machine with
more than 1 Gigabyte total physical RAM, answer "off" here (default
choice and suitable for most users). This will result in a "3GB/1GB"
split: 3GB are mapped so that each process sees a 3GB virtual memory
space and the remaining part of the 4GB virtual memory space is used
by the kernel to permanently map as much physical memory as
possible.
If the machine has between 1 and 4 Gigabytes physical RAM, then
answer "4GB" here.
If more than 4 Gigabytes is used then answer "64GB" here. This
selection turns Intel PAE (Physical Address Extension) mode on.
PAE implements 3-level paging on IA32 processors. PAE is fully
supported by Linux, PAE mode is implemented on all recent Intel
processors (Pentium Pro and better). NOTE: If you say "64GB" here,
then the kernel will not boot on CPUs that don't support PAE!
The actual amount of total physical memory will either be
auto detected or can be forced by using a kernel command line option
such as "mem=256M". (Try "man bootparam" or see the documentation of
your boot loader (lilo or loadlin) about how to pass options to the
kernel at boot time.)
If unsure, say "off".
config HIGHMEM4G
bool "4GB"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Select this if you have a 32-bit processor and between 1 and 4
gigabytes of physical RAM.
config HIGHMEM64G
bool "64GB"
2012-11-28 23:50:23 +04:00
depends on !M486
2007-11-09 23:56:54 +03:00
select X86_PAE
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Select this if you have a 32-bit processor and more than 4
gigabytes of physical RAM.
endchoice
choice
2011-01-21 01:44:16 +03:00
prompt "Memory split" if EXPERT
2007-11-09 23:56:54 +03:00
default VMSPLIT_3G
depends on X86_32
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Select the desired split between kernel and user memory.
If the address range available to the kernel is less than the
physical memory installed, the remaining memory will be available
as "high memory". Accessing high memory is a little more costly
than low memory, as it needs to be mapped into the kernel first.
Note that increasing the kernel address space limits the range
available to user programs, making the address space there
tighter. Selecting anything other than the default 3G/1G split
will also likely make your kernel incompatible with binary-only
kernel modules.
If you are not absolutely sure what you are doing, leave this
option alone!
config VMSPLIT_3G
bool "3G/1G user/kernel split"
config VMSPLIT_3G_OPT
depends on !X86_PAE
bool "3G/1G user/kernel split (for full 1G low memory)"
config VMSPLIT_2G
bool "2G/2G user/kernel split"
config VMSPLIT_2G_OPT
depends on !X86_PAE
bool "2G/2G user/kernel split (for full 2G low memory)"
config VMSPLIT_1G
bool "1G/3G user/kernel split"
endchoice
config PAGE_OFFSET
hex
default 0xB0000000 if VMSPLIT_3G_OPT
default 0x80000000 if VMSPLIT_2G
default 0x78000000 if VMSPLIT_2G_OPT
default 0x40000000 if VMSPLIT_1G
default 0xC0000000
depends on X86_32
config HIGHMEM
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-09 23:56:54 +03:00
depends on X86_32 && (HIGHMEM64G || HIGHMEM4G)
config X86_PAE
2008-10-16 09:01:38 +04:00
bool "PAE (Physical Address Extension) Support"
2007-11-09 23:56:54 +03:00
depends on X86_32 && !HIGHMEM4G
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
PAE is required for NX support, and furthermore enables
larger swapspace support for non-overcommit purposes. It
has the cost of more pagetable lookup overhead, and also
consumes more pagetable space per process.
2008-09-11 12:31:45 +04:00
config ARCH_PHYS_ADDR_T_64BIT
2012-09-10 15:41:45 +04:00
def_bool y
depends on X86_64 || X86_PAE
2008-09-11 12:31:45 +04:00
2010-10-21 02:55:35 +04:00
config ARCH_DMA_ADDR_T_64BIT
2012-09-10 15:41:45 +04:00
def_bool y
depends on X86_64 || HIGHMEM64G
2010-10-21 02:55:35 +04:00
2008-10-22 14:33:16 +04:00
config DIRECT_GBPAGES
2011-01-21 01:44:16 +03:00
bool "Enable 1GB pages for kernel pagetables" if EXPERT
2008-10-22 14:33:16 +04:00
default y
depends on X86_64
2009-02-05 18:21:53 +03:00
---help---
2008-10-22 14:33:16 +04:00
Allow the kernel linear mapping to use 1GB pages on CPUs that
support it. This can improve the kernel's performance a tiny bit by
reducing TLB pressure. If in doubt, say "Y".
2007-11-09 23:56:54 +03:00
# Common NUMA Features
config NUMA
2008-11-04 20:27:19 +03:00
bool "Numa Memory Allocation and Scheduler Support"
2007-11-09 23:56:54 +03:00
depends on SMP
2014-02-26 00:14:06 +04:00
depends on X86_64 || (X86_32 && HIGHMEM64G && X86_BIGSMP)
default y if X86_BIGSMP
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Enable NUMA (Non Uniform Memory Access) support.
2008-11-04 20:27:19 +03:00
2007-11-09 23:56:54 +03:00
The kernel will try to allocate memory used by a CPU on the
local memory controller of the CPU and add some more
NUMA awareness to the kernel.
2008-11-08 15:29:45 +03:00
For 64-bit this is recommended if the system is Intel Core i7
2008-11-04 20:27:19 +03:00
(or later), AMD Opteron, or EM64T NUMA.
2014-02-26 00:14:06 +04:00
For 32-bit this is only needed if you boot a 32-bit
2014-02-12 06:11:13 +04:00
kernel on a 64-bit NUMA platform.
2008-11-04 20:27:19 +03:00
Otherwise, you should say N.
2007-11-09 23:56:54 +03:00
2010-10-29 19:14:30 +04:00
config AMD_NUMA
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Old style AMD Opteron NUMA detection"
2011-07-11 12:34:32 +04:00
depends on X86_64 && NUMA && PCI
2009-02-05 18:21:53 +03:00
---help---
2010-10-29 19:14:30 +04:00
Enable AMD NUMA node topology detection. You should say Y here if
you have a multi processor AMD system. This uses an old method to
read the NUMA configuration directly from the builtin Northbridge
of Opteron. It is recommended to use X86_64_ACPI_NUMA instead,
which also takes priority if both are compiled in.
2007-11-09 23:56:54 +03:00
config X86_64_ACPI_NUMA
2008-01-30 15:31:03 +03:00
def_bool y
prompt "ACPI NUMA detection"
2007-11-09 23:56:54 +03:00
depends on X86_64 && NUMA && ACPI && PCI
select ACPI_NUMA
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Enable ACPI SRAT based node topology detection.
2008-03-25 20:14:35 +03:00
# Some NUMA nodes have memory ranges that span
# other nodes. Even though a pfn is valid and
# between a node's start and end pfns, it may not
# reside on that node. See memmap_init_zone()
# for details.
config NODES_SPAN_OTHER_NODES
def_bool y
depends on X86_64_ACPI_NUMA
2007-11-09 23:56:54 +03:00
config NUMA_EMU
bool "NUMA emulation"
2011-05-02 19:24:48 +04:00
depends on NUMA
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Enable NUMA emulation. A flat machine will be split
into virtual nodes when booted with "numa=fake=N", where N is the
number of nodes. This is only useful for debugging.
config NODES_SHIFT
2008-08-26 01:15:38 +04:00
int "Maximum NUMA Nodes (as a power of 2)" if !MAXSMP
2010-03-26 01:39:27 +03:00
range 1 10
default "10" if MAXSMP
2007-11-09 23:56:54 +03:00
default "6" if X86_64
default "3"
depends on NEED_MULTIPLE_NODES
2009-02-05 18:21:53 +03:00
---help---
2008-05-12 23:21:12 +04:00
Specify the maximum number of NUMA Nodes available on the target
2009-01-26 13:12:25 +03:00
system. Increases memory reserved to accommodate various tables.
2007-11-09 23:56:54 +03:00
config ARCH_HAVE_MEMORY_PRESENT
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-09 23:56:54 +03:00
depends on X86_32 && DISCONTIGMEM
config NEED_NODE_MEMMAP_SIZE
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-09 23:56:54 +03:00
depends on X86_32 && (DISCONTIGMEM || SPARSEMEM)
config ARCH_FLATMEM_ENABLE
def_bool y
2011-04-01 13:15:12 +04:00
depends on X86_32 && !NUMA
2007-11-09 23:56:54 +03:00
config ARCH_DISCONTIGMEM_ENABLE
def_bool y
2008-01-30 15:30:47 +03:00
depends on NUMA && X86_32
2007-11-09 23:56:54 +03:00
config ARCH_DISCONTIGMEM_DEFAULT
def_bool y
2008-01-30 15:30:47 +03:00
depends on NUMA && X86_32
2007-11-09 23:56:54 +03:00
config ARCH_SPARSEMEM_ENABLE
def_bool y
2012-10-02 22:16:47 +04:00
depends on X86_64 || NUMA || X86_32 || X86_32_NON_STANDARD
2007-11-09 23:56:54 +03:00
select SPARSEMEM_STATIC if X86_32
select SPARSEMEM_VMEMMAP_ENABLE if X86_64
2011-04-01 13:15:12 +04:00
config ARCH_SPARSEMEM_DEFAULT
def_bool y
depends on X86_64
2007-11-09 23:56:54 +03:00
config ARCH_SELECT_MEMORY_MODEL
def_bool y
2008-01-30 15:30:47 +03:00
depends on ARCH_SPARSEMEM_ENABLE
2007-11-09 23:56:54 +03:00
config ARCH_MEMORY_PROBE
2013-07-19 21:47:48 +04:00
bool "Enable sysfs memory/probe interface"
2012-09-10 15:41:45 +04:00
depends on X86_64 && MEMORY_HOTPLUG
2013-07-19 21:47:48 +04:00
help
This option enables a sysfs memory/probe interface for testing.
See Documentation/memory-hotplug.txt for more information.
If you are unsure how to answer this question, answer N.
2007-11-09 23:56:54 +03:00
2011-04-01 13:15:12 +04:00
config ARCH_PROC_KCORE_TEXT
def_bool y
depends on X86_64 && PROC_KCORE
2010-01-10 17:28:09 +03:00
config ILLEGAL_POINTER_VALUE
hex
default 0 if X86_32
default 0xdead000000000000 if X86_64
2007-11-09 23:56:54 +03:00
source "mm/Kconfig"
config HIGHPTE
bool "Allocate 3rd-level pagetables from highmem"
2010-04-21 18:23:44 +04:00
depends on HIGHMEM
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
The VM uses one page table entry for each page of physical memory.
For systems with a lot of RAM, this can be wasteful of precious
low memory. Setting this option will put user-space page table
entries in high memory.
2008-09-07 12:51:34 +04:00
config X86_CHECK_BIOS_CORRUPTION
2009-02-05 18:21:53 +03:00
bool "Check for low memory corruption"
---help---
Periodically check for memory corruption in low memory, which
is suspected to be caused by BIOS. Even when enabled in the
configuration, it is disabled at runtime. Enable it by
setting "memory_corruption_check=1" on the kernel command
line. By default it scans the low 64k of memory every 60
seconds; see the memory_corruption_check_size and
memory_corruption_check_period parameters in
Documentation/kernel-parameters.txt to adjust this.
When enabled with the default parameters, this option has
almost no overhead, as it reserves a relatively small amount
of memory and scans it infrequently. It both detects corruption
and prevents it from affecting the running system.
It is, however, intended as a diagnostic tool; if repeatable
BIOS-originated corruption always affects the same memory,
you can use memmap= to prevent the kernel from using that
memory.
2008-09-07 12:51:34 +04:00
2008-09-07 13:37:32 +04:00
config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK
2009-02-05 18:21:53 +03:00
bool "Set the default setting of memory_corruption_check"
2008-09-07 13:37:32 +04:00
depends on X86_CHECK_BIOS_CORRUPTION
default y
2009-02-05 18:21:53 +03:00
---help---
Set whether the default state of memory_corruption_check is
on or off.
2008-09-07 13:37:32 +04:00
2010-08-26 03:38:20 +04:00
config X86_RESERVE_LOW
2010-08-25 04:32:04 +04:00
int "Amount of low memory, in kilobytes, to reserve for the BIOS"
default 64
range 4 640
2009-02-05 18:21:53 +03:00
---help---
2010-08-25 04:32:04 +04:00
Specify the amount of low memory to reserve for the BIOS.
The first page contains BIOS data structures that the kernel
must not use, so that page must always be reserved.
By default we reserve the first 64K of physical RAM, as a
number of BIOSes are known to corrupt that memory range
during events such as suspend/resume or monitor cable
insertion, so it must not be used by the kernel.
2008-09-16 12:07:34 +04:00
2010-08-25 04:32:04 +04:00
You can set this to 4 if you are absolutely sure that you
trust the BIOS to get all its memory reservations and usages
right. If you know your BIOS have problems beyond the
default 64K area, you can set this to 640 to avoid using the
entire low memory range.
2008-09-16 12:07:34 +04:00
2010-08-25 04:32:04 +04:00
If you have doubts about the BIOS (e.g. suspend/resume does
not work or there's kernel crashes after certain hardware
hotplug events) then you might want to enable
X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check
typical corruption patterns.
2008-09-16 12:07:34 +04:00
2010-08-25 04:32:04 +04:00
Leave this to the default value of 64 if you are unsure.
2008-09-16 12:07:34 +04:00
2007-11-09 23:56:54 +03:00
config MATH_EMULATION
bool
prompt "Math emulation" if X86_32
---help---
Linux can emulate a math coprocessor (used for floating point
operations) if you don't have one. 486DX and Pentium processors have
a math coprocessor built in, 486SX and 386 do not, unless you added
a 487DX or 387, respectively. (The messages during boot time can
give you some hints here ["man dmesg"].) Everyone needs either a
coprocessor or this emulation.
If you don't have a math coprocessor, you need to say Y here; if you
say Y here even though you have a coprocessor, the coprocessor will
be used nevertheless. (This behavior can be changed with the kernel
command line option "no387", which comes handy if your coprocessor
is broken. Try "man bootparam" or see the documentation of your boot
loader (lilo or loadlin) about how to pass options to the kernel at
boot time.) This means that it is a good idea to say Y here if you
intend to use this kernel on different machines.
More information about the internals of the Linux math coprocessor
emulation can be found in <file:arch/x86/math-emu/README>.
If you are not sure, say Y; apart from resulting in a 66 KB bigger
kernel, it won't hurt.
config MTRR
2010-04-21 18:23:44 +04:00
def_bool y
2011-01-21 01:44:16 +03:00
prompt "MTRR (Memory Type Range Register) support" if EXPERT
2007-11-09 23:56:54 +03:00
---help---
On Intel P6 family processors (Pentium Pro, Pentium II and later)
the Memory Type Range Registers (MTRRs) may be used to control
processor access to memory ranges. This is most useful if you have
a video (VGA) card on a PCI or AGP bus. Enabling write-combining
allows bus write transfers to be combined into a larger transfer
before bursting over the PCI/AGP bus. This can increase performance
of image write operations 2.5 times or more. Saying Y here creates a
/proc/mtrr file which may be used to manipulate your processor's
MTRRs. Typically the X server should use this.
This code has a reasonably generic interface so that similar
control registers on other processors can be easily supported
as well:
The Cyrix 6x86, 6x86MX and M II processors have Address Range
Registers (ARRs) which provide a similar functionality to MTRRs. For
these, the ARRs are used to emulate the MTRRs.
The AMD K6-2 (stepping 8 and above) and K6-3 processors have two
MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing
write-combining. All of these processors are supported by this code
and it makes sense to say Y here if you have one of them.
Saying Y here also fixes a problem with buggy SMP BIOSes which only
set the MTRRs for the boot CPU and not for the secondary CPUs. This
can lead to all sorts of problems, so it's good to say Y here.
You can safely say Y even if your machine doesn't have MTRRs, you'll
just add about 9 KB to your kernel.
2008-07-27 04:54:22 +04:00
See <file:Documentation/x86/mtrr.txt> for more information.
2007-11-09 23:56:54 +03:00
2008-04-29 14:52:33 +04:00
config MTRR_SANITIZER
x86: change MTRR_SANITIZER to def_bool y
This option has been added in v2.6.26 as a default-disabled
feature and went through several revisions since then.
The feature fixes a wide range of MTRR setup problems that BIOSes
leave us with: slow system, slow Xorg, slow system when adding lots
of RAM, etc., so we want to enable it by default for v2.6.28.
See:
[Bug 10508] Upgrade to 4GB of RAM messes up MTRRs
http://bugzilla.kernel.org/show_bug.cgi?id=10508
and the test results in:
http://lkml.org/lkml/2008/9/29/273
1. hpa
reg00: base=0xc0000000 (3072MB), size=1024MB: uncachable, count=1
reg01: base=0x13c000000 (5056MB), size= 64MB: uncachable, count=1
reg02: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg03: base=0x100000000 (4096MB), size=1024MB: write-back, count=1
reg04: base=0xbf700000 (3063MB), size= 1MB: uncachable, count=1
reg05: base=0xbf800000 (3064MB), size= 8MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 128M num_reg: 6 lose RAM: 0M
range0: 0000000000000000 - 00000000c0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
hole: 00000000bf700000 - 00000000c0000000
Setting variable MTRR 2, base: 3063MB, range: 1MB, type UC
Setting variable MTRR 3, base: 3064MB, range: 8MB, type UC
range0: 0000000100000000 - 0000000140000000
Setting variable MTRR 4, base: 4096MB, range: 1024MB, type WB
hole: 000000013c000000 - 0000000140000000
Setting variable MTRR 5, base: 5056MB, range: 64MB, type UC
2. Dylan Taft
reg00: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg01: base=0x100000000 (4096MB), size= 512MB: write-back, count=1
reg02: base=0x120000000 (4608MB), size= 256MB: write-back, count=1
reg03: base=0xd0000000 (3328MB), size= 256MB: uncachable, count=1
reg04: base=0xe0000000 (3584MB), size= 512MB: uncachable, count=1
reg05: base=0xc7e00000 (3198MB), size= 2MB: uncachable, count=1
reg06: base=0xc8000000 (3200MB), size= 128MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 4M num_reg: 6 lose RAM: 0M
range0: 0000000000000000 - 00000000c8000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
Setting variable MTRR 2, base: 3072MB, range: 128MB, type WB
hole: 00000000c7e00000 - 00000000c8000000
Setting variable MTRR 3, base: 3198MB, range: 2MB, type UC
rangeX: 0000000100000000 - 0000000130000000
Setting variable MTRR 4, base: 4096MB, range: 512MB, type WB
Setting variable MTRR 5, base: 4608MB, range: 256MB, type WB
3. Gabriel
reg00: base=0xd0000000 (3328MB), size= 256MB: uncachable, count=1
reg01: base=0xe0000000 (3584MB), size= 512MB: uncachable, count=1
reg02: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg03: base=0x100000000 (4096MB), size= 512MB: write-back, count=1
reg04: base=0x120000000 (4608MB), size= 128MB: write-back, count=1
reg05: base=0x128000000 (4736MB), size= 64MB: write-back, count=1
reg06: base=0xcf600000 (3318MB), size= 2MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 16M num_reg: 7 lose RAM: 0M
range0: 0000000000000000 - 00000000d0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
Setting variable MTRR 2, base: 3072MB, range: 256MB, type WB
hole: 00000000cf600000 - 00000000cf800000
Setting variable MTRR 3, base: 3318MB, range: 2MB, type UC
rangeX: 0000000100000000 - 000000012c000000
Setting variable MTRR 4, base: 4096MB, range: 512MB, type WB
Setting variable MTRR 5, base: 4608MB, range: 128MB, type WB
Setting variable MTRR 6, base: 4736MB, range: 64MB, type WB
4. Mika Fischer
reg00: base=0xc0000000 (3072MB), size=1024MB: uncachable, count=1
reg01: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg02: base=0x100000000 (4096MB), size=1024MB: write-back, count=1
reg03: base=0xbf700000 (3063MB), size= 1MB: uncachable, count=1
reg04: base=0xbf800000 (3064MB), size= 8MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 16M num_reg: 5 lose RAM: 0M
range0: 0000000000000000 - 00000000c0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
hole: 00000000bf700000 - 00000000c0000000
Setting variable MTRR 2, base: 3063MB, range: 1MB, type UC
Setting variable MTRR 3, base: 3064MB, range: 8MB, type UC
rangeX: 0000000100000000 - 0000000140000000
Setting variable MTRR 4, base: 4096MB, range: 1024MB, type WB
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-01 03:29:40 +04:00
def_bool y
2008-04-29 14:52:33 +04:00
prompt "MTRR cleanup support"
depends on MTRR
2009-02-05 18:21:53 +03:00
---help---
2008-07-15 16:48:48 +04:00
Convert MTRR layout from continuous to discrete, so X drivers can
add writeback entries.
2008-04-29 14:52:33 +04:00
2008-07-15 16:48:48 +04:00
Can be disabled with disable_mtrr_cleanup on the kernel command line.
2009-01-26 13:12:25 +03:00
The largest mtrr entry size for a continuous block can be set with
2008-07-15 16:48:48 +04:00
mtrr_chunk_size.
2008-04-29 14:52:33 +04:00
x86: change MTRR_SANITIZER to def_bool y
This option has been added in v2.6.26 as a default-disabled
feature and went through several revisions since then.
The feature fixes a wide range of MTRR setup problems that BIOSes
leave us with: slow system, slow Xorg, slow system when adding lots
of RAM, etc., so we want to enable it by default for v2.6.28.
See:
[Bug 10508] Upgrade to 4GB of RAM messes up MTRRs
http://bugzilla.kernel.org/show_bug.cgi?id=10508
and the test results in:
http://lkml.org/lkml/2008/9/29/273
1. hpa
reg00: base=0xc0000000 (3072MB), size=1024MB: uncachable, count=1
reg01: base=0x13c000000 (5056MB), size= 64MB: uncachable, count=1
reg02: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg03: base=0x100000000 (4096MB), size=1024MB: write-back, count=1
reg04: base=0xbf700000 (3063MB), size= 1MB: uncachable, count=1
reg05: base=0xbf800000 (3064MB), size= 8MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 128M num_reg: 6 lose RAM: 0M
range0: 0000000000000000 - 00000000c0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
hole: 00000000bf700000 - 00000000c0000000
Setting variable MTRR 2, base: 3063MB, range: 1MB, type UC
Setting variable MTRR 3, base: 3064MB, range: 8MB, type UC
range0: 0000000100000000 - 0000000140000000
Setting variable MTRR 4, base: 4096MB, range: 1024MB, type WB
hole: 000000013c000000 - 0000000140000000
Setting variable MTRR 5, base: 5056MB, range: 64MB, type UC
2. Dylan Taft
reg00: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg01: base=0x100000000 (4096MB), size= 512MB: write-back, count=1
reg02: base=0x120000000 (4608MB), size= 256MB: write-back, count=1
reg03: base=0xd0000000 (3328MB), size= 256MB: uncachable, count=1
reg04: base=0xe0000000 (3584MB), size= 512MB: uncachable, count=1
reg05: base=0xc7e00000 (3198MB), size= 2MB: uncachable, count=1
reg06: base=0xc8000000 (3200MB), size= 128MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 4M num_reg: 6 lose RAM: 0M
range0: 0000000000000000 - 00000000c8000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
Setting variable MTRR 2, base: 3072MB, range: 128MB, type WB
hole: 00000000c7e00000 - 00000000c8000000
Setting variable MTRR 3, base: 3198MB, range: 2MB, type UC
rangeX: 0000000100000000 - 0000000130000000
Setting variable MTRR 4, base: 4096MB, range: 512MB, type WB
Setting variable MTRR 5, base: 4608MB, range: 256MB, type WB
3. Gabriel
reg00: base=0xd0000000 (3328MB), size= 256MB: uncachable, count=1
reg01: base=0xe0000000 (3584MB), size= 512MB: uncachable, count=1
reg02: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg03: base=0x100000000 (4096MB), size= 512MB: write-back, count=1
reg04: base=0x120000000 (4608MB), size= 128MB: write-back, count=1
reg05: base=0x128000000 (4736MB), size= 64MB: write-back, count=1
reg06: base=0xcf600000 (3318MB), size= 2MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 16M num_reg: 7 lose RAM: 0M
range0: 0000000000000000 - 00000000d0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
Setting variable MTRR 2, base: 3072MB, range: 256MB, type WB
hole: 00000000cf600000 - 00000000cf800000
Setting variable MTRR 3, base: 3318MB, range: 2MB, type UC
rangeX: 0000000100000000 - 000000012c000000
Setting variable MTRR 4, base: 4096MB, range: 512MB, type WB
Setting variable MTRR 5, base: 4608MB, range: 128MB, type WB
Setting variable MTRR 6, base: 4736MB, range: 64MB, type WB
4. Mika Fischer
reg00: base=0xc0000000 (3072MB), size=1024MB: uncachable, count=1
reg01: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg02: base=0x100000000 (4096MB), size=1024MB: write-back, count=1
reg03: base=0xbf700000 (3063MB), size= 1MB: uncachable, count=1
reg04: base=0xbf800000 (3064MB), size= 8MB: uncachable, count=1
will get
Found optimal setting for mtrr clean up
gran_size: 1M chunk_size: 16M num_reg: 5 lose RAM: 0M
range0: 0000000000000000 - 00000000c0000000
Setting variable MTRR 0, base: 0MB, range: 2048MB, type WB
Setting variable MTRR 1, base: 2048MB, range: 1024MB, type WB
hole: 00000000bf700000 - 00000000c0000000
Setting variable MTRR 2, base: 3063MB, range: 1MB, type UC
Setting variable MTRR 3, base: 3064MB, range: 8MB, type UC
rangeX: 0000000100000000 - 0000000140000000
Setting variable MTRR 4, base: 4096MB, range: 1024MB, type WB
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-01 03:29:40 +04:00
If unsure, say Y.
2008-04-29 14:52:33 +04:00
config MTRR_SANITIZER_ENABLE_DEFAULT
2008-04-30 07:25:58 +04:00
int "MTRR cleanup enable value (0-1)"
range 0 1
default "0"
2008-04-29 14:52:33 +04:00
depends on MTRR_SANITIZER
2009-02-05 18:21:53 +03:00
---help---
2008-04-30 07:25:58 +04:00
Enable mtrr cleanup default value
2008-04-29 14:52:33 +04:00
2008-05-02 13:40:22 +04:00
config MTRR_SANITIZER_SPARE_REG_NR_DEFAULT
int "MTRR cleanup spare reg num (0-7)"
range 0 7
default "1"
depends on MTRR_SANITIZER
2009-02-05 18:21:53 +03:00
---help---
2008-05-02 13:40:22 +04:00
mtrr cleanup spare entries default, it can be changed via
2008-07-15 16:48:48 +04:00
mtrr_spare_reg_nr=N on the kernel command line.
2008-05-02 13:40:22 +04:00
2008-03-19 03:00:14 +03:00
config X86_PAT
2010-04-21 18:23:44 +04:00
def_bool y
2011-01-21 01:44:16 +03:00
prompt "x86 PAT support" if EXPERT
2008-04-26 12:26:52 +04:00
depends on MTRR
2009-02-05 18:21:53 +03:00
---help---
2008-03-19 03:00:14 +03:00
Use PAT attributes to setup page level cache control.
2008-03-25 00:22:35 +03:00
2008-03-19 03:00:14 +03:00
PATs are the modern equivalents of MTRRs and are much more
flexible than MTRRs.
Say N here if you see bootup problems (boot crash, boot hang,
2008-03-25 00:22:35 +03:00
spontaneous reboots) or a non-working video driver.
2008-03-19 03:00:14 +03:00
If unsure, say Y.
2009-07-10 20:57:37 +04:00
config ARCH_USES_PG_UNCACHED
def_bool y
depends on X86_PAT
x86, random: Architectural inlines to get random integers with RDRAND
Architectural inlines to get random ints and longs using the RDRAND
instruction.
Intel has introduced a new RDRAND instruction, a Digital Random Number
Generator (DRNG), which is functionally an high bandwidth entropy
source, cryptographic whitener, and integrity monitor all built into
hardware. This enables RDRAND to be used directly, bypassing the
kernel random number pool.
For technical documentation, see:
http://software.intel.com/en-us/articles/download-the-latest-bull-mountain-software-implementation-guide/
In this patch, this is *only* used for the nonblocking random number
pool. RDRAND is a nonblocking source, similar to our /dev/urandom,
and is therefore not a direct replacement for /dev/random. The
architectural hooks presented in the previous patch only feed the
kernel internal users, which only use the nonblocking pool, and so
this is not a problem.
Since this instruction is available in userspace, there is no reason
to have a /dev/hw_rng device driver for the purpose of feeding rngd.
This is especially so since RDRAND is a nonblocking source, and needs
additional whitening and reduction (see the above technical
documentation for details) in order to be of "pure entropy source"
quality.
The CONFIG_EXPERT compile-time option can be used to disable this use
of RDRAND.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Originally-by: Fenghua Yu <fenghua.yu@intel.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "Theodore Ts'o" <tytso@mit.edu>
2011-08-01 00:59:29 +04:00
config ARCH_RANDOM
def_bool y
prompt "x86 architectural random number generator" if EXPERT
---help---
Enable the x86 architectural RDRAND instruction
(Intel Bull Mountain technology) to generate random numbers.
If supported, this is a high bandwidth, cryptographically
secure hardware random number generator.
2012-09-21 23:43:10 +04:00
config X86_SMAP
def_bool y
prompt "Supervisor Mode Access Prevention" if EXPERT
---help---
Supervisor Mode Access Prevention (SMAP) is a security
feature in newer Intel processors. There is a small
performance cost if this enabled and turned on; there is
also a small increase in the kernel size if this is enabled.
If unsure, say Y.
2007-11-09 23:56:54 +03:00
config EFI
2008-10-16 09:01:38 +04:00
bool "EFI runtime service support"
2008-01-30 15:31:19 +03:00
depends on ACPI
2013-04-16 18:31:08 +04:00
select UCS2_STRING
2007-11-09 23:56:54 +03:00
---help---
2009-02-05 18:21:53 +03:00
This enables the kernel to use EFI runtime services that are
available (such as the EFI variable services).
2007-11-09 23:56:54 +03:00
2009-02-05 18:21:53 +03:00
This option is only useful on systems that have EFI firmware.
In addition, you should use the latest ELILO loader available
at <http://elilo.sourceforge.net> in order to take advantage
of EFI runtime services. However, even with this option, the
resultant kernel should continue to boot on existing non-EFI
platforms.
2007-11-09 23:56:54 +03:00
x86, efi: EFI boot stub support
There is currently a large divide between kernel development and the
development of EFI boot loaders. The idea behind this patch is to give
the kernel developers full control over the EFI boot process. As
H. Peter Anvin put it,
"The 'kernel carries its own stub' approach been very successful in
dealing with BIOS, and would make a lot of sense to me for EFI as
well."
This patch introduces an EFI boot stub that allows an x86 bzImage to
be loaded and executed by EFI firmware. The bzImage appears to the
firmware as an EFI application. Luckily there are enough free bits
within the bzImage header so that it can masquerade as an EFI
application, thereby coercing the EFI firmware into loading it and
jumping to its entry point. The beauty of this masquerading approach
is that both BIOS and EFI boot loaders can still load and run the same
bzImage, thereby allowing a single kernel image to work in any boot
environment.
The EFI boot stub supports multiple initrds, but they must exist on
the same partition as the bzImage. Command-line arguments for the
kernel can be appended after the bzImage name when run from the EFI
shell, e.g.
Shell> bzImage console=ttyS0 root=/dev/sdb initrd=initrd.img
v7:
- Fix checkpatch warnings.
v6:
- Try to allocate initrd memory just below hdr->inird_addr_max.
v5:
- load_options_size is UTF-16, which needs dividing by 2 to convert
to the corresponding ASCII size.
v4:
- Don't read more than image->load_options_size
v3:
- Fix following warnings when compiling CONFIG_EFI_STUB=n
arch/x86/boot/tools/build.c: In function ‘main’:
arch/x86/boot/tools/build.c:138:24: warning: unused variable ‘pe_header’
arch/x86/boot/tools/build.c:138:15: warning: unused variable ‘file_sz’
- As reported by Matthew Garrett, some Apple machines have GOPs that
don't have hardware attached. We need to weed these out by
searching for ones that handle the PCIIO protocol.
- Don't allocate memory if no initrds are on cmdline
- Don't trust image->load_options_size
Maarten Lankhorst noted:
- Don't strip first argument when booted from efibootmgr
- Don't allocate too much memory for cmdline
- Don't update cmdline_size, the kernel considers it read-only
- Don't accept '\n' for initrd names
v2:
- File alignment was too large, was 8192 should be 512. Reported by
Maarten Lankhorst on LKML.
- Added UGA support for graphics
- Use VIDEO_TYPE_EFI instead of hard-coded number.
- Move linelength assignment until after we've assigned depth
- Dynamically fill out AddressOfEntryPoint in tools/build.c
- Don't use magic number for GDT/TSS stuff. Requested by Andi Kleen
- The bzImage may need to be relocated as it may have been loaded at
a high address address by the firmware. This was required to get my
macbook booting because the firmware loaded it at 0x7cxxxxxx, which
triggers this error in decompress_kernel(),
if (heap > ((-__PAGE_OFFSET-(128<<20)-1) & 0x7fffffff))
error("Destination address too large");
Cc: Mike Waychison <mikew@google.com>
Cc: Matthew Garrett <mjg@redhat.com>
Tested-by: Henrik Rydberg <rydberg@euromail.se>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Link: http://lkml.kernel.org/r/1321383097.2657.9.camel@mfleming-mobl1.ger.corp.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2011-12-13 01:27:52 +04:00
config EFI_STUB
bool "EFI stub support"
depends on EFI
---help---
This kernel feature allows a bzImage to be loaded directly
by EFI firmware without the use of a bootloader.
2013-09-23 02:45:25 +04:00
See Documentation/efi-stub.txt for more information.
2012-03-16 16:03:13 +04:00
2014-01-10 22:52:06 +04:00
config EFI_MIXED
bool "EFI mixed-mode support"
depends on EFI_STUB && X86_64
---help---
Enabling this feature allows a 64-bit kernel to be booted
on a 32-bit firmware, provided that your CPU supports 64-bit
mode.
Note that it is not possible to boot a mixed-mode enabled
kernel via the EFI boot stub - a bootloader that supports
the EFI handover protocol must be used.
If unsure, say N.
2007-11-09 23:56:54 +03:00
config SECCOMP
2008-01-30 15:31:03 +03:00
def_bool y
prompt "Enable seccomp to safely compute untrusted bytecode"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This kernel feature is useful for number crunching applications
that may need to compute untrusted bytecode during their
execution. By using pipes or other transports made available to
the process as file descriptors supporting the read/write
syscalls, it's possible to isolate those applications in
their own address space using seccomp. Once seccomp is
2008-09-09 11:01:31 +04:00
enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
2007-11-09 23:56:54 +03:00
and the task is only allowed to execute a few safe syscalls
defined by each seccomp mode.
If unsure, say Y. Only embedded should say N here.
source kernel/Kconfig.hz
config KEXEC
bool "kexec system call"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
kexec is a system call that implements the ability to shutdown your
current kernel, and to start another kernel. It is like a reboot
but it is independent of the system firmware. And like a reboot
you can start any kernel with it, not just Linux.
The name comes from the similarity to the exec system call.
It is an ongoing process to be certain the hardware in a machine
is properly shutdown, so do not be surprised if this code does not
2013-08-20 23:38:03 +04:00
initially work for you. As of this writing the exact hardware
interface is strongly in flux, so no good recommendation can be
made.
2007-11-09 23:56:54 +03:00
config CRASH_DUMP
2008-08-14 19:16:50 +04:00
bool "kernel crash dumps"
2007-11-09 23:56:54 +03:00
depends on X86_64 || (X86_32 && HIGHMEM)
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
Generate crash dump after being started by kexec.
This should be normally only set in special crash dump kernels
which are loaded in the main kernel with kexec-tools into
a specially reserved region and then later executed after
a crash by kdump/kexec. The crash dump kernel must be compiled
to a memory address not used by the main kernel or BIOS using
PHYSICAL_START, or it must be built as a relocatable image
(CONFIG_RELOCATABLE=y).
For more details see Documentation/kdump/kdump.txt
2008-07-26 06:45:07 +04:00
config KEXEC_JUMP
2012-10-02 22:16:47 +04:00
bool "kexec jump"
2009-03-10 05:57:16 +03:00
depends on KEXEC && HIBERNATION
2009-02-05 18:21:53 +03:00
---help---
kexec jump: save/restore device state
This patch implements devices state save/restore before after kexec.
This patch together with features in kexec_jump patch can be used for
following:
- A simple hibernation implementation without ACPI support. You can kexec a
hibernating kernel, save the memory image of original system and shutdown
the system. When resuming, you restore the memory image of original system
via ordinary kexec load then jump back.
- Kernel/system debug through making system snapshot. You can make system
snapshot, jump back, do some thing and make another system snapshot.
- Cooperative multi-kernel/system. With kexec jump, you can switch between
several kernels/systems quickly without boot process except the first time.
This appears like swap a whole kernel/system out/in.
- A general method to call program in physical mode (paging turning
off). This can be used to invoke BIOS code under Linux.
The following user-space tools can be used with kexec jump:
- kexec-tools needs to be patched to support kexec jump. The patches
and the precompiled kexec can be download from the following URL:
source: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-src_git_kh10.tar.bz2
patches: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-patches_git_kh10.tar.bz2
binary: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec_git_kh10
- makedumpfile with patches are used as memory image saving tool, it
can exclude free pages from original kernel memory image file. The
patches and the precompiled makedumpfile can be download from the
following URL:
source: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-src_cvs_kh10.tar.bz2
patches: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-patches_cvs_kh10.tar.bz2
binary: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile_cvs_kh10
- An initramfs image can be used as the root file system of kexeced
kernel. An initramfs image built with "BuildRoot" can be downloaded
from the following URL:
initramfs image: http://khibernation.sourceforge.net/download/release_v10/initramfs/rootfs_cvs_kh10.gz
All user space tools above are included in the initramfs image.
Usage example of simple hibernation:
1. Compile and install patched kernel with following options selected:
CONFIG_X86_32=y
CONFIG_RELOCATABLE=y
CONFIG_KEXEC=y
CONFIG_CRASH_DUMP=y
CONFIG_PM=y
CONFIG_HIBERNATION=y
CONFIG_KEXEC_JUMP=y
2. Build an initramfs image contains kexec-tool and makedumpfile, or
download the pre-built initramfs image, called rootfs.gz in
following text.
3. Prepare a partition to save memory image of original kernel, called
hibernating partition in following text.
4. Boot kernel compiled in step 1 (kernel A).
5. In the kernel A, load kernel compiled in step 1 (kernel B) with
/sbin/kexec. The shell command line can be as follow:
/sbin/kexec --load-preserve-context /boot/bzImage --mem-min=0x100000
--mem-max=0xffffff --initrd=rootfs.gz
6. Boot the kernel B with following shell command line:
/sbin/kexec -e
7. The kernel B will boot as normal kexec. In kernel B the memory
image of kernel A can be saved into hibernating partition as
follow:
jump_back_entry=`cat /proc/cmdline | tr ' ' '\n' | grep kexec_jump_back_entry | cut -d '='`
echo $jump_back_entry > kexec_jump_back_entry
cp /proc/vmcore dump.elf
Then you can shutdown the machine as normal.
8. Boot kernel compiled in step 1 (kernel C). Use the rootfs.gz as
root file system.
9. In kernel C, load the memory image of kernel A as follow:
/sbin/kexec -l --args-none --entry=`cat kexec_jump_back_entry` dump.elf
10. Jump back to the kernel A as follow:
/sbin/kexec -e
Then, kernel A is resumed.
Implementation point:
To support jumping between two kernels, before jumping to (executing)
the new kernel and jumping back to the original kernel, the devices
are put into quiescent state, and the state of devices and CPU is
saved. After jumping back from kexeced kernel and jumping to the new
kernel, the state of devices and CPU are restored accordingly. The
devices/CPU state save/restore code of software suspend is called to
implement corresponding function.
Known issues:
- Because the segment number supported by sys_kexec_load is limited,
hibernation image with many segments may not be load. This is
planned to be eliminated by adding a new flag to sys_kexec_load to
make a image can be loaded with multiple sys_kexec_load invoking.
Now, only the i386 architecture is supported.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-26 06:45:10 +04:00
Jump between original kernel and kexeced kernel and invoke
code in physical address mode via KEXEC
2008-07-26 06:45:07 +04:00
2007-11-09 23:56:54 +03:00
config PHYSICAL_START
2011-01-21 01:44:16 +03:00
hex "Physical address where the kernel is loaded" if (EXPERT || CRASH_DUMP)
2009-05-12 03:12:16 +04:00
default "0x1000000"
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This gives the physical address where the kernel is loaded.
If kernel is a not relocatable (CONFIG_RELOCATABLE=n) then
bzImage will decompress itself to above physical address and
run from there. Otherwise, bzImage will run from the address where
it has been loaded by the boot loader and will ignore above physical
address.
In normal kdump cases one does not have to set/change this option
as now bzImage can be compiled as a completely relocatable image
(CONFIG_RELOCATABLE=y) and be used to load and run from a different
address. This option is mainly useful for the folks who don't want
to use a bzImage for capturing the crash dump and want to use a
vmlinux instead. vmlinux is not relocatable hence a kernel needs
to be specifically compiled to run from a specific memory area
(normally a reserved region) and this option comes handy.
2009-05-12 03:12:16 +04:00
So if you are using bzImage for capturing the crash dump,
leave the value here unchanged to 0x1000000 and set
CONFIG_RELOCATABLE=y. Otherwise if you plan to use vmlinux
for capturing the crash dump change this value to start of
the reserved region. In other words, it can be set based on
the "X" value as specified in the "crashkernel=YM@XM"
command line boot parameter passed to the panic-ed
kernel. Please take a look at Documentation/kdump/kdump.txt
for more details about crash dumps.
2007-11-09 23:56:54 +03:00
Usage of bzImage for capturing the crash dump is recommended as
one does not have to build two kernels. Same kernel can be used
as production kernel and capture kernel. Above option should have
gone away after relocatable bzImage support is introduced. But it
is present because there are users out there who continue to use
vmlinux for dump capture. This option should go away down the
line.
Don't change this unless you know what you are doing.
config RELOCATABLE
2009-05-08 01:19:34 +04:00
bool "Build a relocatable kernel"
default y
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This builds a kernel image that retains relocation information
so it can be loaded someplace besides the default 1MB.
The relocations tend to make the kernel binary about 10% larger,
but are discarded at runtime.
One use is for the kexec on panic case where the recovery kernel
must live at a different physical address than the primary
kernel.
Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address
it has been loaded at and the compile time physical address
2013-10-11 04:18:14 +04:00
(CONFIG_PHYSICAL_START) is used as the minimum location.
2007-11-09 23:56:54 +03:00
2013-10-11 04:18:14 +04:00
config RANDOMIZE_BASE
bool "Randomize the address of the kernel image"
depends on RELOCATABLE
depends on !HIBERNATION
default n
---help---
Randomizes the physical and virtual address at which the
kernel image is decompressed, as a security feature that
deters exploit attempts relying on knowledge of the location
of kernel internals.
2013-11-12 02:28:39 +04:00
Entropy is generated using the RDRAND instruction if it is
supported. If RDTSC is supported, it is used as well. If
neither RDRAND nor RDTSC are supported, then randomness is
read from the i8254 timer.
2013-10-11 04:18:14 +04:00
The kernel will be offset by up to RANDOMIZE_BASE_MAX_OFFSET,
2013-11-12 02:28:39 +04:00
and aligned according to PHYSICAL_ALIGN. Since the kernel is
built using 2GiB addressing, and PHYSICAL_ALGIN must be at a
minimum of 2MiB, only 10 bits of entropy is theoretically
possible. At best, due to page table layouts, 64-bit can use
9 bits of entropy and 32-bit uses 8 bits.
2013-10-11 04:18:14 +04:00
2013-12-11 00:27:45 +04:00
If unsure, say N.
2013-10-11 04:18:14 +04:00
config RANDOMIZE_BASE_MAX_OFFSET
2013-12-11 00:27:45 +04:00
hex "Maximum kASLR offset allowed" if EXPERT
2013-10-11 04:18:14 +04:00
depends on RANDOMIZE_BASE
2013-10-11 04:18:18 +04:00
range 0x0 0x20000000 if X86_32
default "0x20000000" if X86_32
range 0x0 0x40000000 if X86_64
default "0x40000000" if X86_64
2013-10-11 04:18:14 +04:00
---help---
2013-12-11 00:27:45 +04:00
The lesser of RANDOMIZE_BASE_MAX_OFFSET and available physical
memory is used to determine the maximal offset in bytes that will
be applied to the kernel when kernel Address Space Layout
Randomization (kASLR) is active. This must be a multiple of
PHYSICAL_ALIGN.
On 32-bit this is limited to 512MiB by page table layouts. The
default is 512MiB.
2013-10-11 04:18:18 +04:00
2013-12-11 00:27:45 +04:00
On 64-bit this is limited by how the kernel fixmap page table is
positioned, so this cannot be larger than 1GiB currently. Without
RANDOMIZE_BASE, there is a 512MiB to 1.5GiB split between kernel
and modules. When RANDOMIZE_BASE_MAX_OFFSET is above 512MiB, the
modules area will shrink to compensate, up to the current maximum
1GiB to 1GiB split. The default is 1GiB.
2013-10-11 04:18:18 +04:00
2013-12-11 00:27:45 +04:00
If unsure, leave at the default value.
2013-10-11 04:18:14 +04:00
# Relocation on x86 needs some additional build support
2009-05-06 08:20:51 +04:00
config X86_NEED_RELOCS
def_bool y
2013-10-11 04:18:14 +04:00
depends on RANDOMIZE_BASE || (X86_32 && RELOCATABLE)
2009-05-06 08:20:51 +04:00
2007-11-09 23:56:54 +03:00
config PHYSICAL_ALIGN
2013-07-08 20:15:17 +04:00
hex "Alignment value to which kernel should be aligned"
2013-10-11 04:18:14 +04:00
default "0x200000"
2013-07-08 20:15:17 +04:00
range 0x2000 0x1000000 if X86_32
range 0x200000 0x1000000 if X86_64
2009-02-05 18:21:53 +03:00
---help---
2007-11-09 23:56:54 +03:00
This value puts the alignment restrictions on physical address
where kernel is loaded and run from. Kernel is compiled for an
address which meets above alignment restriction.
If bootloader loads the kernel at a non-aligned address and
CONFIG_RELOCATABLE is set, kernel will move itself to nearest
address aligned to above value and run from there.
If bootloader loads the kernel at a non-aligned address and
CONFIG_RELOCATABLE is not set, kernel will ignore the run time
load address and decompress itself to the address it has been
compiled for and run from there. The address for which kernel is
compiled already meets above alignment restrictions. Hence the
end result is that kernel runs from a physical address meeting
above alignment restrictions.
2013-07-08 20:15:17 +04:00
On 32-bit this value must be a multiple of 0x2000. On 64-bit
this value must be a multiple of 0x200000.
2007-11-09 23:56:54 +03:00
Don't change this unless you know what you are doing.
config HOTPLUG_CPU
2008-08-11 19:46:46 +04:00
bool "Support for hot-pluggable CPUs"
2013-05-21 07:49:35 +04:00
depends on SMP
2007-11-09 23:56:54 +03:00
---help---
2008-08-11 19:46:46 +04:00
Say Y here to allow turning CPUs off and on. CPUs can be
controlled through /sys/devices/system/cpu.
( Note: power management support will enable this option
automatically on SMP systems. )
Say N if you want to disable CPU hotplug.
2007-11-09 23:56:54 +03:00
2012-11-13 23:32:39 +04:00
config BOOTPARAM_HOTPLUG_CPU0
bool "Set default setting of cpu0_hotpluggable"
default n
2013-01-23 01:01:19 +04:00
depends on HOTPLUG_CPU
2012-11-13 23:32:39 +04:00
---help---
Set whether default state of cpu0_hotpluggable is on or off.
Say Y here to enable CPU0 hotplug by default. If this switch
is turned on, there is no need to give cpu0_hotplug kernel
parameter and the CPU0 hotplug feature is enabled by default.
Please note: there are two known CPU0 dependencies if you want
to enable the CPU0 hotplug feature either by this switch or by
cpu0_hotplug kernel parameter.
First, resume from hibernate or suspend always starts from CPU0.
So hibernate and suspend are prevented if CPU0 is offline.
Second dependency is PIC interrupts always go to CPU0. CPU0 can not
offline if any interrupt can not migrate out of CPU0. There may
be other CPU0 dependencies.
Please make sure the dependencies are under your control before
you enable this feature.
Say N if you don't want to enable CPU0 hotplug feature by default.
You still can enable the CPU0 hotplug feature at boot by kernel
parameter cpu0_hotplug.
2012-11-13 23:32:51 +04:00
config DEBUG_HOTPLUG_CPU0
def_bool n
prompt "Debug CPU0 hotplug"
2013-01-23 01:01:19 +04:00
depends on HOTPLUG_CPU
2012-11-13 23:32:51 +04:00
---help---
Enabling this option offlines CPU0 (if CPU0 can be offlined) as
soon as possible and boots up userspace with CPU0 offlined. User
can online CPU0 back after boot time.
To debug CPU0 hotplug, you need to enable CPU0 offline/online
feature by either turning on CONFIG_BOOTPARAM_HOTPLUG_CPU0 during
compilation or giving cpu0_hotplug kernel parameter at boot.
If unsure, say N.
2007-11-09 23:56:54 +03:00
config COMPAT_VDSO
2014-03-14 03:01:26 +04:00
def_bool n
prompt "Disable the 32-bit vDSO (needed for glibc 2.3.3)"
2008-01-30 15:30:43 +03:00
depends on X86_32 || IA32_EMULATION
2009-02-05 18:21:53 +03:00
---help---
2014-03-14 03:01:26 +04:00
Certain buggy versions of glibc will crash if they are
presented with a 32-bit vDSO that is not mapped at the address
indicated in its segment table.
2009-11-11 02:46:52 +03:00
2014-03-14 03:01:26 +04:00
The bug was introduced by f866314b89d56845f55e6f365e18b31ec978ec3a
and fixed by 3b3ddb4f7db98ec9e912ccdf54d35df4aa30e04a and
49ad572a70b8aeb91e57483a11dd1b77e31c4468. Glibc 2.3.3 is
the only released version with the bug, but OpenSUSE 9
contains a buggy "glibc 2.3.2".
2007-11-09 23:56:54 +03:00
2014-03-14 03:01:26 +04:00
The symptom of the bug is that everything crashes on startup, saying:
dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
Saying Y here changes the default value of the vdso32 boot
option from 1 to 0, which turns off the 32-bit vDSO entirely.
This works around the glibc bug but hurts performance.
If unsure, say N: if you are compiling your own kernel, you
are unlikely to be using a buggy version of glibc.
2007-11-09 23:56:54 +03:00
2008-08-12 23:52:36 +04:00
config CMDLINE_BOOL
bool "Built-in kernel command line"
2009-02-05 18:21:53 +03:00
---help---
2008-08-12 23:52:36 +04:00
Allow for specifying boot arguments to the kernel at
build time. On some systems (e.g. embedded ones), it is
necessary or convenient to provide some or all of the
kernel boot arguments with the kernel itself (that is,
to not rely on the boot loader to provide them.)
To compile command line arguments into the kernel,
set this option to 'Y', then fill in the
the boot arguments in CONFIG_CMDLINE.
Systems with fully functional boot loaders (i.e. non-embedded)
should leave this option set to 'N'.
config CMDLINE
string "Built-in kernel command string"
depends on CMDLINE_BOOL
default ""
2009-02-05 18:21:53 +03:00
---help---
2008-08-12 23:52:36 +04:00
Enter arguments here that should be compiled into the kernel
image and used at boot time. If the boot loader provides a
command line at boot time, it is appended to this string to
form the full kernel command line, when the system boots.
However, you can use the CONFIG_CMDLINE_OVERRIDE option to
change this behavior.
In most cases, the command line (whether built-in or provided
by the boot loader) should specify the device for the root
file system.
config CMDLINE_OVERRIDE
bool "Built-in command line overrides boot loader arguments"
depends on CMDLINE_BOOL
2009-02-05 18:21:53 +03:00
---help---
2008-08-12 23:52:36 +04:00
Set this option to 'Y' to have the kernel ignore the boot loader
command line, and use ONLY the built-in command line.
This is used to work around broken boot loaders. This should
be set to 'N' under normal conditions.
2007-11-09 23:56:54 +03:00
endmenu
config ARCH_ENABLE_MEMORY_HOTPLUG
def_bool y
depends on X86_64 || (X86_32 && HIGHMEM)
2008-10-31 20:52:03 +03:00
config ARCH_ENABLE_MEMORY_HOTREMOVE
def_bool y
depends on MEMORY_HOTPLUG
2010-05-27 01:44:58 +04:00
config USE_PERCPU_NUMA_NODE_ID
2011-01-23 16:37:40 +03:00
def_bool y
2010-05-27 01:44:58 +04:00
depends on NUMA
2013-11-15 02:31:10 +04:00
config ARCH_ENABLE_SPLIT_PMD_PTLOCK
def_bool y
depends on X86_64 || X86_PAE
2014-06-05 03:05:35 +04:00
config ARCH_ENABLE_HUGEPAGE_MIGRATION
def_bool y
depends on X86_64 && HUGETLB_PAGE && MIGRATION
2008-11-05 22:37:27 +03:00
menu "Power management and ACPI options"
2007-11-06 22:41:05 +03:00
config ARCH_HIBERNATION_HEADER
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-06 22:41:05 +03:00
depends on X86_64 && HIBERNATION
source "kernel/power/Kconfig"
source "drivers/acpi/Kconfig"
2009-08-14 23:23:29 +04:00
source "drivers/sfi/Kconfig"
2008-01-30 15:32:49 +03:00
config X86_APM_BOOT
2010-04-21 18:23:44 +04:00
def_bool y
2011-11-17 14:41:31 +04:00
depends on APM
2008-01-30 15:32:49 +03:00
2007-11-06 22:41:05 +03:00
menuconfig APM
tristate "APM (Advanced Power Management) BIOS support"
2008-07-10 18:09:50 +04:00
depends on X86_32 && PM_SLEEP
2007-11-06 22:41:05 +03:00
---help---
APM is a BIOS specification for saving power using several different
techniques. This is mostly useful for battery powered laptops with
APM compliant BIOSes. If you say Y here, the system time will be
reset after a RESUME operation, the /proc/apm device will provide
battery status information, and user-space programs will receive
notification of APM "events" (e.g. battery status change).
If you select "Y" here, you can disable actual use of the APM
BIOS by passing the "apm=off" option to the kernel at boot time.
Note that the APM support is almost completely disabled for
machines with more than one CPU.
In order to use APM, you will need supporting software. For location
2011-07-09 01:11:16 +04:00
and more information, read <file:Documentation/power/apm-acpi.txt>
and the Battery Powered Linux mini-HOWTO, available from
2007-11-06 22:41:05 +03:00
<http://www.tldp.org/docs.html#howto>.
This driver does not spin down disk drives (see the hdparm(8)
manpage ("man 8 hdparm") for that), and it doesn't turn off
VESA-compliant "green" monitors.
This driver does not support the TI 4000M TravelMate and the ACER
486/DX4/75 because they don't have compliant BIOSes. Many "green"
desktop machines also don't have compliant BIOSes, and this driver
may cause those machines to panic during the boot phase.
Generally, if you don't have a battery in your machine, there isn't
much point in using this driver and you should say N. If you get
random kernel OOPSes or reboots that don't seem to be related to
anything, try disabling/enabling this option (or disabling/enabling
APM in your BIOS).
Some other things you should try when experiencing seemingly random,
"weird" problems:
1) make sure that you have enough swap space and that it is
enabled.
2) pass the "no-hlt" option to the kernel
3) switch on floating point emulation in the kernel and pass
the "no387" option to the kernel
4) pass the "floppy=nodma" option to the kernel
5) pass the "mem=4M" option to the kernel (thereby disabling
all but the first 4 MB of RAM)
6) make sure that the CPU is not over clocked.
7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/>
8) disable the cache from your BIOS settings
9) install a fan for the video card or exchange video RAM
10) install a better fan for the CPU
11) exchange RAM chips
12) exchange the motherboard.
To compile this driver as a module, choose M here: the
module will be called apm.
if APM
config APM_IGNORE_USER_SUSPEND
bool "Ignore USER SUSPEND"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
This option will ignore USER SUSPEND requests. On machines with a
compliant APM BIOS, you want to say N. However, on the NEC Versa M
series notebooks, it is necessary to say Y because of a BIOS bug.
config APM_DO_ENABLE
bool "Enable PM at boot time"
---help---
Enable APM features at boot time. From page 36 of the APM BIOS
specification: "When disabled, the APM BIOS does not automatically
power manage devices, enter the Standby State, enter the Suspend
State, or take power saving steps in response to CPU Idle calls."
This driver will make CPU Idle calls when Linux is idle (unless this
feature is turned off -- see "Do CPU IDLE calls", below). This
should always save battery power, but more complicated APM features
will be dependent on your BIOS implementation. You may need to turn
this option off if your computer hangs at boot time when using APM
support, or if it beeps continuously instead of suspending. Turn
this off if you have a NEC UltraLite Versa 33/C or a Toshiba
T400CDT. This is off by default since most machines do fine without
this feature.
config APM_CPU_IDLE
2013-02-10 06:10:04 +04:00
depends on CPU_IDLE
2007-11-06 22:41:05 +03:00
bool "Make CPU Idle calls when idle"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop.
On some machines, this can activate improved power savings, such as
a slowed CPU clock rate, when the machine is idle. These idle calls
are made after the idle loop has run for some length of time (e.g.,
333 mS). On some machines, this will cause a hang at boot time or
whenever the CPU becomes idle. (On machines with more than one CPU,
this option does nothing.)
config APM_DISPLAY_BLANK
bool "Enable console blanking using APM"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
Enable console blanking using the APM. Some laptops can use this to
turn off the LCD backlight when the screen blanker of the Linux
virtual console blanks the screen. Note that this is only used by
the virtual console screen blanker, and won't turn off the backlight
when using the X Window system. This also doesn't have anything to
do with your VESA-compliant power-saving monitor. Further, this
option doesn't work for all laptops -- it might not turn off your
backlight at all, or it might print a lot of errors to the console,
especially if you are using gpm.
config APM_ALLOW_INTS
bool "Allow interrupts during APM BIOS calls"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
Normally we disable external interrupts while we are making calls to
the APM BIOS as a measure to lessen the effects of a badly behaving
BIOS implementation. The BIOS should reenable interrupts if it
needs to. Unfortunately, some BIOSes do not -- especially those in
many of the newer IBM Thinkpads. If you experience hangs when you
suspend, try setting this to Y. Otherwise, say N.
endif # APM
2011-05-20 02:51:07 +04:00
source "drivers/cpufreq/Kconfig"
2007-11-06 22:41:05 +03:00
source "drivers/cpuidle/Kconfig"
2008-10-09 22:45:22 +04:00
source "drivers/idle/Kconfig"
2007-11-06 22:41:05 +03:00
endmenu
menu "Bus options (PCI etc.)"
config PCI
2008-05-19 16:10:14 +04:00
bool "PCI support"
2008-01-30 15:32:32 +03:00
default y
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
Find out whether you have a PCI motherboard. PCI is the name of a
bus system, i.e. the way the CPU talks to the other stuff inside
your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
VESA. If you have PCI, say Y, otherwise N.
choice
prompt "PCI access mode"
2008-07-10 18:09:50 +04:00
depends on X86_32 && PCI
2007-11-06 22:41:05 +03:00
default PCI_GOANY
---help---
On PCI systems, the BIOS can be used to detect the PCI devices and
determine their configuration. However, some old PCI motherboards
have BIOS bugs and may crash if this is done. Also, some embedded
PCI-based systems don't have any BIOS at all. Linux can also try to
detect the PCI hardware directly without using the BIOS.
With this option, you can specify how Linux should detect the
PCI devices. If you choose "BIOS", the BIOS will be used,
if you choose "Direct", the BIOS won't be used, and if you
choose "MMConfig", then PCI Express MMCONFIG will be used.
If you choose "Any", the kernel will try MMCONFIG, then the
direct access method and falls back to the BIOS if that doesn't
work. If unsure, go with the default, which is "Any".
config PCI_GOBIOS
bool "BIOS"
config PCI_GOMMCONFIG
bool "MMConfig"
config PCI_GODIRECT
bool "Direct"
2008-04-29 11:59:53 +04:00
config PCI_GOOLPC
2010-09-23 20:28:04 +04:00
bool "OLPC XO-1"
2008-04-29 11:59:53 +04:00
depends on OLPC
2008-06-06 01:14:41 +04:00
config PCI_GOANY
bool "Any"
2007-11-06 22:41:05 +03:00
endchoice
config PCI_BIOS
2008-01-30 15:31:03 +03:00
def_bool y
2008-07-10 18:09:50 +04:00
depends on X86_32 && PCI && (PCI_GOBIOS || PCI_GOANY)
2007-11-06 22:41:05 +03:00
# x86-64 doesn't support PCI BIOS access from long mode so always go direct.
config PCI_DIRECT
2008-01-30 15:31:03 +03:00
def_bool y
2011-05-27 10:59:39 +04:00
depends on PCI && (X86_64 || (PCI_GODIRECT || PCI_GOANY || PCI_GOOLPC || PCI_GOMMCONFIG))
2007-11-06 22:41:05 +03:00
config PCI_MMCONFIG
2008-01-30 15:31:03 +03:00
def_bool y
2009-08-14 23:37:50 +04:00
depends on X86_32 && PCI && (ACPI || SFI) && (PCI_GOMMCONFIG || PCI_GOANY)
2007-11-06 22:41:05 +03:00
2008-04-29 11:59:53 +04:00
config PCI_OLPC
2008-06-06 01:14:41 +04:00
def_bool y
depends on PCI && OLPC && (PCI_GOOLPC || PCI_GOANY)
2008-04-29 11:59:53 +04:00
2010-03-18 23:31:34 +03:00
config PCI_XEN
def_bool y
depends on PCI && XEN
select SWIOTLB_XEN
2007-11-06 22:41:05 +03:00
config PCI_DOMAINS
2008-01-30 15:31:03 +03:00
def_bool y
2007-11-06 22:41:05 +03:00
depends on PCI
config PCI_MMCONFIG
bool "Support mmconfig PCI config space access"
depends on X86_64 && PCI && ACPI
2010-04-01 22:43:30 +04:00
config PCI_CNB20LE_QUIRK
2011-01-21 01:44:16 +03:00
bool "Read CNB20LE Host Bridge Windows" if EXPERT
2012-10-02 22:16:47 +04:00
depends on PCI
2010-04-01 22:43:30 +04:00
help
Read the PCI windows out of the CNB20LE host bridge. This allows
PCI hotplug to work on systems with the CNB20LE chipset which do
not have ACPI.
2011-01-06 20:12:30 +03:00
There's no public spec for this chipset, and this functionality
is known to be incomplete.
You should say N unless you know you need this.
2007-11-06 22:41:05 +03:00
source "drivers/pci/pcie/Kconfig"
source "drivers/pci/Kconfig"
2011-03-23 02:34:59 +03:00
# x86_64 have no ISA slots, but can have ISA-style DMA.
2007-11-06 22:41:05 +03:00
config ISA_DMA_API
2011-03-23 02:34:59 +03:00
bool "ISA-style DMA support" if (X86_64 && EXPERT)
default y
help
Enables ISA-style DMA support for devices requiring such controllers.
If unsure, say Y.
2007-11-06 22:41:05 +03:00
if X86_32
config ISA
bool "ISA support"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
Find out whether you have ISA slots on your motherboard. ISA is the
name of a bus system, i.e. the way the CPU talks to the other stuff
inside your box. Other bus systems are PCI, EISA, MicroChannel
(MCA) or VESA. ISA is an older system, now being displaced by PCI;
newer boards don't support it. If you have ISA, say Y, otherwise N.
config EISA
bool "EISA support"
depends on ISA
---help---
The Extended Industry Standard Architecture (EISA) bus was
developed as an open alternative to the IBM MicroChannel bus.
The EISA bus provided some of the features of the IBM MicroChannel
bus while maintaining backward compatibility with cards made for
the older ISA bus. The EISA bus saw limited use between 1988 and
1995 when it was made obsolete by the PCI bus.
Say Y here if you are building a kernel for an EISA-based machine.
Otherwise, say N.
source "drivers/eisa/Kconfig"
config SCx200
tristate "NatSemi SCx200 support"
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
This provides basic support for National Semiconductor's
(now AMD's) Geode processors. The driver probes for the
PCI-IDs of several on-chip devices, so its a good dependency
for other scx200_* drivers.
If compiled as a module, the driver is named scx200.
config SCx200HR_TIMER
tristate "NatSemi SCx200 27MHz High-Resolution Timer Support"
2010-07-14 04:56:20 +04:00
depends on SCx200
2007-11-06 22:41:05 +03:00
default y
2009-02-05 18:21:53 +03:00
---help---
2007-11-06 22:41:05 +03:00
This driver provides a clocksource built upon the on-chip
27MHz high-resolution timer. Its also a workaround for
NSC Geode SC-1100's buggy TSC, which loses time when the
processor goes idle (as is done by the scheduler). The
other workaround is idle=poll boot option.
2008-04-29 11:59:53 +04:00
config OLPC
bool "One Laptop Per Child support"
2011-02-23 11:50:15 +03:00
depends on !X86_PAE
2009-12-15 05:00:36 +03:00
select GPIOLIB
2011-02-23 12:08:31 +03:00
select OF
2011-03-13 18:10:17 +03:00
select OF_PROMTREE
2011-12-17 02:50:17 +04:00
select IRQ_DOMAIN
2009-02-05 18:21:53 +03:00
---help---
2008-04-29 11:59:53 +04:00
Add support for detecting the unique features of the OLPC
XO hardware.
2011-06-25 20:34:10 +04:00
config OLPC_XO1_PM
bool "OLPC XO-1 Power Management"
2011-06-25 20:34:11 +04:00
depends on OLPC && MFD_CS5535 && PM_SLEEP
2011-06-25 20:34:10 +04:00
select MFD_CORE
2010-10-10 13:40:32 +04:00
---help---
2011-06-25 20:34:11 +04:00
Add support for poweroff and suspend of the OLPC XO-1 laptop.
2010-10-10 13:40:32 +04:00
2011-06-25 20:34:17 +04:00
config OLPC_XO1_RTC
bool "OLPC XO-1 Real Time Clock"
depends on OLPC_XO1_PM && RTC_DRV_CMOS
---help---
Add support for the XO-1 real time clock, which can be used as a
programmable wakeup source.
2011-06-25 20:34:12 +04:00
config OLPC_XO1_SCI
bool "OLPC XO-1 SCI extras"
2011-07-24 21:33:21 +04:00
depends on OLPC && OLPC_XO1_PM
2012-12-19 00:22:17 +04:00
depends on INPUT=y
2011-07-24 21:33:21 +04:00
select POWER_SUPPLY
2011-06-25 20:34:12 +04:00
select GPIO_CS5535
select MFD_CORE
---help---
Add support for SCI-based features of the OLPC XO-1 laptop:
2011-06-25 20:34:14 +04:00
- EC-driven system wakeups
2011-06-25 20:34:12 +04:00
- Power button
2011-06-25 20:34:14 +04:00
- Ebook switch
2011-06-25 20:34:15 +04:00
- Lid switch
2011-06-25 20:34:16 +04:00
- AC adapter status updates
- Battery status updates
2011-06-25 20:34:12 +04:00
2011-06-25 20:34:18 +04:00
config OLPC_XO15_SCI
bool "OLPC XO-1.5 SCI extras"
2011-07-24 21:33:21 +04:00
depends on OLPC && ACPI
select POWER_SUPPLY
2011-06-25 20:34:18 +04:00
---help---
Add support for SCI-based features of the OLPC XO-1.5 laptop:
- EC-driven system wakeups
- AC adapter status updates
- Battery status updates
2010-10-10 13:40:32 +04:00
2011-09-21 01:00:12 +04:00
config ALIX
bool "PCEngines ALIX System Support (LED setup)"
select GPIOLIB
---help---
This option enables system support for the PCEngines ALIX.
At present this just sets up LEDs for GPIO control on
ALIX2/3/6 boards. However, other system specific setup should
get added here.
Note: You must still enable the drivers for GPIO and LED support
(GPIO_CS5535 & LEDS_GPIO) to actually use the LEDs
Note: You have to set alix.force=1 for boards with Award BIOS.
2012-03-06 03:05:15 +04:00
config NET5501
bool "Soekris Engineering net5501 System Support (LEDS, GPIO, etc)"
select GPIOLIB
---help---
This option enables system support for the Soekris Engineering net5501.
2012-01-14 12:45:39 +04:00
config GEOS
bool "Traverse Technologies GEOS System Support (LEDS, GPIO, etc)"
select GPIOLIB
depends on DMI
---help---
This option enables system support for the Traverse Technologies GEOS.
2013-01-05 01:18:14 +04:00
config TS5500
bool "Technologic Systems TS-5500 platform support"
depends on MELAN
select CHECK_SIGNATURE
select NEW_LEDS
select LEDS_CLASS
---help---
This option enables system support for the Technologic Systems TS-5500.
2007-11-07 01:10:39 +03:00
endif # X86_32
2010-09-17 20:03:43 +04:00
config AMD_NB
2007-11-06 22:41:05 +03:00
def_bool y
2010-03-12 17:43:03 +03:00
depends on CPU_SUP_AMD && PCI
2007-11-06 22:41:05 +03:00
source "drivers/pcmcia/Kconfig"
source "drivers/pci/hotplug/Kconfig"
2011-03-24 02:43:03 +03:00
config RAPIDIO
2013-07-04 02:08:56 +04:00
tristate "RapidIO support"
2011-03-24 02:43:03 +03:00
depends on PCI
default n
help
2013-07-04 02:08:56 +04:00
If enabled this option will include drivers and the core
2011-03-24 02:43:03 +03:00
infrastructure code to support RapidIO interconnect devices.
source "drivers/rapidio/Kconfig"
x86: provide platform-devices for boot-framebuffers
The current situation regarding boot-framebuffers (VGA, VESA/VBE, EFI) on
x86 causes troubles when loading multiple fbdev drivers. The global
"struct screen_info" does not provide any state-tracking about which
drivers use the FBs. request_mem_region() theoretically works, but
unfortunately vesafb/efifb ignore it due to quirks for broken boards.
Avoid this by creating a platform framebuffer devices with a pointer
to the "struct screen_info" as platform-data. Drivers can now create
platform-drivers and the driver-core will refuse multiple drivers being
active simultaneously.
We keep the screen_info available for backwards-compatibility. Drivers
can be converted in follow-up patches.
Different devices are created for VGA/VESA/EFI FBs to allow multiple
drivers to be loaded on distro kernels. We create:
- "vesa-framebuffer" for VBE/VESA graphics FBs
- "efi-framebuffer" for EFI FBs
- "platform-framebuffer" for everything else
This allows to load vesafb, efifb and others simultaneously and each
picks up only the supported FB types.
Apart from platform-framebuffer devices, this also introduces a
compatibility option for "simple-framebuffer" drivers which recently got
introduced for OF based systems. If CONFIG_X86_SYSFB is selected, we
try to match the screen_info against a simple-framebuffer supported
format. If we succeed, we create a "simple-framebuffer" device instead
of a platform-framebuffer.
This allows to reuse the simplefb.c driver across architectures and also
to introduce a SimpleDRM driver. There is no need to have vesafb.c,
efifb.c, simplefb.c and more just to have architecture specific quirks
in their setup-routines.
Instead, we now move the architecture specific quirks into x86-setup and
provide a generic simple-framebuffer. For backwards-compatibility (if
strange formats are used), we still allow vesafb/efifb to be loaded
simultaneously and pick up all remaining devices.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Link: http://lkml.kernel.org/r/1375445127-15480-4-git-send-email-dh.herrmann@gmail.com
Tested-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2013-08-02 16:05:22 +04:00
config X86_SYSFB
bool "Mark VGA/VBE/EFI FB as generic system framebuffer"
help
Firmwares often provide initial graphics framebuffers so the BIOS,
bootloader or kernel can show basic video-output during boot for
user-guidance and debugging. Historically, x86 used the VESA BIOS
Extensions and EFI-framebuffers for this, which are mostly limited
to x86.
This option, if enabled, marks VGA/VBE/EFI framebuffers as generic
framebuffers so the new generic system-framebuffer drivers can be
used on x86. If the framebuffer is not compatible with the generic
modes, it is adverticed as fallback platform framebuffer so legacy
drivers like efifb, vesafb and uvesafb can pick it up.
If this option is not selected, all system framebuffers are always
marked as fallback platform framebuffers as usual.
Note: Legacy fbdev drivers, including vesafb, efifb, uvesafb, will
not be able to pick up generic system framebuffers if this option
is selected. You are highly encouraged to enable simplefb as
replacement if you select this option. simplefb can correctly deal
with generic system framebuffers. But you should still keep vesafb
and others enabled as fallback if a system framebuffer is
incompatible with simplefb.
If unsure, say Y.
2007-11-06 22:41:05 +03:00
endmenu
menu "Executable file formats / Emulations"
source "fs/Kconfig.binfmt"
config IA32_EMULATION
bool "IA32 Emulation"
depends on X86_64
2013-06-18 23:33:40 +04:00
select BINFMT_ELF
2008-01-30 15:31:55 +03:00
select COMPAT_BINFMT_ELF
2012-10-09 03:28:08 +04:00
select HAVE_UID16
2009-02-05 18:21:53 +03:00
---help---
2012-02-19 22:40:03 +04:00
Include code to run legacy 32-bit programs under a
64-bit kernel. You should likely turn this on, unless you're
100% sure that you don't have any 32-bit programs left.
2007-11-06 22:41:05 +03:00
config IA32_AOUT
2009-02-05 18:21:53 +03:00
tristate "IA32 a.out support"
depends on IA32_EMULATION
---help---
Support old a.out binaries in the 32bit emulation.
2007-11-06 22:41:05 +03:00
2012-02-28 02:09:10 +04:00
config X86_X32
2012-10-02 22:16:47 +04:00
bool "x32 ABI for 64-bit mode"
depends on X86_64 && IA32_EMULATION
2012-02-19 22:40:03 +04:00
---help---
Include code to run binaries for the x32 native 32-bit ABI
for 64-bit processors. An x32 process gets access to the
full 64-bit register file and wide data path while leaving
pointers at 32 bits for smaller memory footprint.
You will need a recent binutils (2.22 or later) with
elf32_x86_64 support enabled to compile a kernel with this
option set.
2007-11-06 22:41:05 +03:00
config COMPAT
2008-01-30 15:31:03 +03:00
def_bool y
2012-02-28 02:09:10 +04:00
depends on IA32_EMULATION || X86_X32
[PATCH v3] ipc: provide generic compat versions of IPC syscalls
When using the "compat" APIs, architectures will generally want to
be able to make direct syscalls to msgsnd(), shmctl(), etc., and
in the kernel we would want them to be handled directly by
compat_sys_xxx() functions, as is true for other compat syscalls.
However, for historical reasons, several of the existing compat IPC
syscalls do not do this. semctl() expects a pointer to the fourth
argument, instead of the fourth argument itself. msgsnd(), msgrcv()
and shmat() expect arguments in different order.
This change adds an ARCH_WANT_OLD_COMPAT_IPC config option that can be
set to preserve this behavior for ports that use it (x86, sparc, powerpc,
s390, and mips). No actual semantics are changed for those architectures,
and there is only a minimal amount of code refactoring in ipc/compat.c.
Newer architectures like tile (and perhaps future architectures such
as arm64 and unicore64) should not select this option, and thus can
avoid having any IPC-specific code at all in their architecture-specific
compat layer. In the same vein, if this option is not selected, IPC_64
mode is assumed, since that's what the <asm-generic> headers expect.
The workaround code in "tile" for msgsnd() and msgrcv() is removed
with this change; it also fixes the bug that shmat() and semctl() were
not being properly handled.
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
2012-03-15 21:13:38 +04:00
select ARCH_WANT_OLD_COMPAT_IPC
2007-11-06 22:41:05 +03:00
2012-09-10 15:41:45 +04:00
if COMPAT
2007-11-06 22:41:05 +03:00
config COMPAT_FOR_U64_ALIGNMENT
2012-09-10 15:41:45 +04:00
def_bool y
2007-11-06 22:41:05 +03:00
config SYSVIPC_COMPAT
2008-01-30 15:31:03 +03:00
def_bool y
2012-09-10 15:41:45 +04:00
depends on SYSVIPC
2007-11-06 22:41:05 +03:00
2011-03-07 18:06:20 +03:00
config KEYS_COMPAT
2012-09-10 15:41:45 +04:00
def_bool y
depends on KEYS
endif
2011-03-07 18:06:20 +03:00
2007-11-06 22:41:05 +03:00
endmenu
2008-11-03 20:21:45 +03:00
config HAVE_ATOMIC_IOMAP
def_bool y
depends on X86_32
2012-04-04 21:39:58 +04:00
config X86_DEV_DMA_OPS
bool
2012-04-04 21:40:21 +04:00
depends on X86_64 || STA2X11
2012-04-04 21:39:58 +04:00
2012-04-04 21:40:10 +04:00
config X86_DMA_REMAP
bool
2012-04-04 21:40:21 +04:00
depends on STA2X11
2012-04-04 21:40:10 +04:00
arch: x86: New MailBox support driver for Intel SOC's
Current Intel SOC cores use a MailBox Interface (MBI) to provide access to
configuration registers on devices (called units) connected to the system
fabric. This is a support driver that implements access to this interface on
those platforms that can enumerate the device using PCI. Initial support is for
BayTrail, for which port definitons are provided. This is a requirement for
implementing platform specific features (e.g. RAPL driver requires this to
perform platform specific power management using the registers in PUNIT).
Dependant modules should select IOSF_MBI in their respective Kconfig
configuraiton. Serialized access is handled by all exported routines with
spinlocks.
The API includes 3 functions for access to unit registers:
int iosf_mbi_read(u8 port, u8 opcode, u32 offset, u32 *mdr)
int iosf_mbi_write(u8 port, u8 opcode, u32 offset, u32 mdr)
int iosf_mbi_modify(u8 port, u8 opcode, u32 offset, u32 mdr, u32 mask)
port: indicating the unit being accessed
opcode: the read or write port specific opcode
offset: the register offset within the port
mdr: the register data to be read, written, or modified
mask: bit locations in mdr to change
Returns nonzero on error
Note: GPU code handles access to the GFX unit. Therefore access to that unit
with this driver is disallowed to avoid conflicts.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1389216471-734-1-git-send-email-david.e.box@linux.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Rafael J. Wysocki <rjw@rjwysocki.net>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
2014-01-09 01:27:51 +04:00
config IOSF_MBI
2014-05-10 00:44:05 +04:00
tristate
default m
arch: x86: New MailBox support driver for Intel SOC's
Current Intel SOC cores use a MailBox Interface (MBI) to provide access to
configuration registers on devices (called units) connected to the system
fabric. This is a support driver that implements access to this interface on
those platforms that can enumerate the device using PCI. Initial support is for
BayTrail, for which port definitons are provided. This is a requirement for
implementing platform specific features (e.g. RAPL driver requires this to
perform platform specific power management using the registers in PUNIT).
Dependant modules should select IOSF_MBI in their respective Kconfig
configuraiton. Serialized access is handled by all exported routines with
spinlocks.
The API includes 3 functions for access to unit registers:
int iosf_mbi_read(u8 port, u8 opcode, u32 offset, u32 *mdr)
int iosf_mbi_write(u8 port, u8 opcode, u32 offset, u32 mdr)
int iosf_mbi_modify(u8 port, u8 opcode, u32 offset, u32 mdr, u32 mask)
port: indicating the unit being accessed
opcode: the read or write port specific opcode
offset: the register offset within the port
mdr: the register data to be read, written, or modified
mask: bit locations in mdr to change
Returns nonzero on error
Note: GPU code handles access to the GFX unit. Therefore access to that unit
with this driver is disallowed to avoid conflicts.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1389216471-734-1-git-send-email-david.e.box@linux.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Rafael J. Wysocki <rjw@rjwysocki.net>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
2014-01-09 01:27:51 +04:00
depends on PCI
2007-11-06 22:41:05 +03:00
source "net/Kconfig"
source "drivers/Kconfig"
source "drivers/firmware/Kconfig"
source "fs/Kconfig"
source "arch/x86/Kconfig.debug"
source "security/Kconfig"
source "crypto/Kconfig"
2007-12-16 12:02:48 +03:00
source "arch/x86/kvm/Kconfig"
2007-11-06 22:41:05 +03:00
source "lib/Kconfig"