The kernel's implementation of notifier chains is unsafe. There is no
protection against entries being added to or removed from a chain while the
chain is in use. The issues were discussed in this thread:
http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2
We noticed that notifier chains in the kernel fall into two basic usage
classes:
"Blocking" chains are always called from a process context
and the callout routines are allowed to sleep;
"Atomic" chains can be called from an atomic context and
the callout routines are not allowed to sleep.
We decided to codify this distinction and make it part of the API. Therefore
this set of patches introduces three new, parallel APIs: one for blocking
notifiers, one for atomic notifiers, and one for "raw" notifiers (which is
really just the old API under a new name). New kinds of data structures are
used for the heads of the chains, and new routines are defined for
registration, unregistration, and calling a chain. The three APIs are
explained in include/linux/notifier.h and their implementation is in
kernel/sys.c.
With atomic and blocking chains, the implementation guarantees that the chain
links will not be corrupted and that chain callers will not get messed up by
entries being added or removed. For raw chains the implementation provides no
guarantees at all; users of this API must provide their own protections. (The
idea was that situations may come up where the assumptions of the atomic and
blocking APIs are not appropriate, so it should be possible for users to
handle these things in their own way.)
There are some limitations, which should not be too hard to live with. For
atomic/blocking chains, registration and unregistration must always be done in
a process context since the chain is protected by a mutex/rwsem. Also, a
callout routine for a non-raw chain must not try to register or unregister
entries on its own chain. (This did happen in a couple of places and the code
had to be changed to avoid it.)
Since atomic chains may be called from within an NMI handler, they cannot use
spinlocks for synchronization. Instead we use RCU. The overhead falls almost
entirely in the unregister routine, which is okay since unregistration is much
less frequent that calling a chain.
Here is the list of chains that we adjusted and their classifications. None
of them use the raw API, so for the moment it is only a placeholder.
ATOMIC CHAINS
-------------
arch/i386/kernel/traps.c: i386die_chain
arch/ia64/kernel/traps.c: ia64die_chain
arch/powerpc/kernel/traps.c: powerpc_die_chain
arch/sparc64/kernel/traps.c: sparc64die_chain
arch/x86_64/kernel/traps.c: die_chain
drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list
kernel/panic.c: panic_notifier_list
kernel/profile.c: task_free_notifier
net/bluetooth/hci_core.c: hci_notifier
net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain
net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain
net/ipv6/addrconf.c: inet6addr_chain
net/netfilter/nf_conntrack_core.c: nf_conntrack_chain
net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain
net/netlink/af_netlink.c: netlink_chain
BLOCKING CHAINS
---------------
arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain
arch/s390/kernel/process.c: idle_chain
arch/x86_64/kernel/process.c idle_notifier
drivers/base/memory.c: memory_chain
drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list
drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list
drivers/macintosh/adb.c: adb_client_list
drivers/macintosh/via-pmu.c sleep_notifier_list
drivers/macintosh/via-pmu68k.c sleep_notifier_list
drivers/macintosh/windfarm_core.c wf_client_list
drivers/usb/core/notify.c usb_notifier_list
drivers/video/fbmem.c fb_notifier_list
kernel/cpu.c cpu_chain
kernel/module.c module_notify_list
kernel/profile.c munmap_notifier
kernel/profile.c task_exit_notifier
kernel/sys.c reboot_notifier_list
net/core/dev.c netdev_chain
net/decnet/dn_dev.c: dnaddr_chain
net/ipv4/devinet.c: inetaddr_chain
It's possible that some of these classifications are wrong. If they are,
please let us know or submit a patch to fix them. Note that any chain that
gets called very frequently should be atomic, because the rwsem read-locking
used for blocking chains is very likely to incur cache misses on SMP systems.
(However, if the chain's callout routines may sleep then the chain cannot be
atomic.)
The patch set was written by Alan Stern and Chandra Seetharaman, incorporating
material written by Keith Owens and suggestions from Paul McKenney and Andrew
Morton.
[jes@sgi.com: restructure the notifier chain initialization macros]
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com>
Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
x86_64: add the futex_atomic_cmpxchg_inuser() assembly implementation, and
wire up the new syscalls.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Acked-by: Ulrich Drepper <drepper@redhat.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a new sched domain for representing multi-core with shared caches
between cores. Consider a dual package system, each package containing two
cores and with last level cache shared between cores with in a package. If
there are two runnable processes, with this appended patch those two
processes will be scheduled on different packages.
On such systems, with this patch we have observed 8% perf improvement with
specJBB(2 warehouse) benchmark and 35% improvement with CFP2000 rate(with 2
users).
This new domain will come into play only on multi-core systems with shared
caches. On other systems, this sched domain will be removed by domain
degeneration code. This new domain can be also used for implementing power
savings policy (see OLS 2005 CMP kernel scheduler paper for more details..
I will post another patch for power savings policy soon)
Most of the arch/* file changes are for cpu_coregroup_map() implementation.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Provide proper kprobes fault handling, if a user-specified pre/post handlers
tries to access user address space, through copy_from_user(), get_user() etc.
The user-specified fault handler gets called only if the fault occurs while
executing user-specified handlers. In such a case user-specified handler is
allowed to fix it first, later if the user-specifed fault handler does not fix
it, we try to fix it by calling fix_exception().
The user-specified handler will not be called if the fault happens when single
stepping the original instruction, instead we reset the current probe and
allow the system page fault handler to fix it up.
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently kprobe handler traps only happen in kernel space, so function
kprobe_exceptions_notify should skip traps which happen in user space.
This patch modifies this, and it is based on 2.6.16-rc4.
Signed-off-by: bibo mao <bibo.mao@intel.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: "Keshavamurthy, Anil S" <anil.s.keshavamurthy@intel.com>
Cc: <hiramatu@sdl.hitachi.co.jp>
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When kretprobe probes the schedule() function, if the probed process exits
then schedule() will never return, so some kretprobe instances will never
be recycled.
In this patch the parent process will recycle retprobe instances of the
probed function and there will be no memory leak of kretprobe instances.
Signed-off-by: bibo mao <bibo.mao@intel.com>
Cc: Masami Hiramatsu <hiramatu@sdl.hitachi.co.jp>
Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Create compat_sys_adjtimex and use it an all appropriate places.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We had a copy of the compatibility version of struct timex in each 64 bit
architecture. This patch just creates a global one and replaces all the
usages of the old ones.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Kyle McMartin <kyle@parisc-linux.org>
Acked-by: Tony Luck <tony.luck@intel.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a "make isoimage" to i386 and x86-64, which allows the automatic
creation of a bootable CD image. It also adds an option FDINITRD= to
include an initrd of the user's choice in generated floppy- or CD boot
images. Finally, some minor cleanups of the image generation code.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
tcsh is not happy with the -9999 error code.
Suggested by Ernie Petrides
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
No need to restrict to power of two here.
TBD needs more double checking
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
pfn_to_page() and others need to access both memnode_shift and the very
first bytes of memnodemap[]. If we force memnode_shift to be just before the
memnodemap array, we can reduce the memory footprint to one cache line
instead of two for most setups. This patch introduce a 'memnode' structure
where shift and map[] are carefully placed.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
register_die_notifier is exported twice, once in traps.c and once in
x8664_ksyms.c. This results in a warning on build.
Signed-off-by: Kevin Winchester <kwin@ns.sympatico.ca>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
arch/x86_64/kernel/aperture.c: The search for the AGP bridge has been
extended to search for all the 256 buses instead of the first 32. This
is required since on a some systems, the bridge may be located on a bus
much farther than the first 32. By searching all 256 buses, we guarantee
that the search succeeds on such systems.
arch/x86_64/kernel/pci-gart.c: The search for the Northbridge is not
limited to just bus 0 anymore. This is required because on certain
systems, we may not find one on bus 0.
Signed-off-by: Navin Boppuri <navin.boppuri@newisys.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Have the GART_IOMMU help text specify that this is the hardware IOMMU in
amd64 processors. This will be significant if/when other IOMMUs are
added to the x86-64 architecture. :-)
Also, note that the previous help text stated that IOMMU was needed for
>3GB memory instead of >4GB. This is fixed in the newer version.
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It was a failed experiment - all benchmarks done with it on both AMD
and Intel showed it was a loss. That was probably because the store
buffers of the CPUs for write combining traffic weren't large enough.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
free_bootmem_node expects a physical address to be passed in, but
__alloc_bootmem_node returns a virtual one. That address needs to be
translated to physical.
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
o check_timer() routine fails while second kernel is booting after a crash
on an opetron box. Problem happens because timer vector (0x31) seems to be
locked.
o After a system crash, it is not safe to service interrupts any more, hence
interrupts are disabled. This leads to pending interrupts at LAPIC. LAPIC
sends these interrupts to the CPU during early boot of second kernel. Other
pending interrupts are discarded saying unexpected trap but timer interrupt
is serviced and CPU does not issue an LAPIC EOI because it think this
interrupt came from i8259 and sends ack to 8259. This leads to vector 0x31
locking as LAPIC does not clear respective ISR and keeps on waiting for
EOI.
o This patch issues extra EOI for the pending interrupts who have ISR set.
o Though today only timer seems to be the special case because in early
boot it thinks interrupts are coming from i8259 and uses
mask_and_ack_8259A() as ack handler and does not issue LAPIC EOI. But
probably doing it in generic manner for all vectors makes sense.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
cpu_vm_mask is of type cpumask_t, so use the proper bitops.
Signed-off-by: Brian Gerst <bgerst@didntduck.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This fixes problems with very large nodes (over 128GB) filling up all of
the first 4GB with their mem_map and not leaving enough space for the
swiotlb.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This means i386 processes compiled with a recent compiler will get non
executable heap by default now. This is the same default as a 32bit PAE
kernel would use on a NX enabled CPU.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Because 256 causes overflows in some code that stores them in 8 bit
fields and the x86 APIC architecture cannot handle more than 255
anyways.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When x86_64 timer init messages were changed to use apic verbosity
levels, two messages were missed and one got the wrong level. This
causes the last word of a suppressed message to print on a line by
itself. Fix that so either the entire message prints or none of it
does.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch puts the infrastructure in place to allow for a reordering of
functions based inside the vmlinux. The general idea is that it is possible
to put all "common" functions into the first 2Mb of the code, so that they
are covered by one TLB entry. This as opposed to the current situation where
a typical vmlinux covers about 3.5Mb (on x86-64) and thus 2 TLB entries.
This is done by enabling the -ffunction-sections flag in gcc, which puts
each function in its own ELF section, so that the linker can then order them
in a way defined by the linker script.
As per previous discussions, Linus said he wanted a "static" list for this,
eg a list provided by the kernel tarbal, so that most people have the same
ordering at least. A script is provided to create this list based on
readprofile(1) output. The included list is provisional, and entirely biased
on my own testbox and me running a few kernel compiles and some other
things.
I think that to get to a better list we need to invite people to submit
their own profiles, and somehow add those all up and base the final list on
that. I'm willing to do that effort if this is ends up being the prefered
approach. Such an effort probably needs to be repeated like once a year or
so to adopt to the changing nature of the kernel.
Made it a CONFIG with default n because it increases link times
dramatically.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I tested it on a couple of chipsets and it worked everywhere so it
should be ok as default for now.
So far I haven't done the great purge of the useless old check_timer
code yet though.
Can be overwritten with enable_8254_timer in the worst case
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There is a fallback logic, so it's better to not use the OOM killer
in the allocations.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
ACPIv2 has an official but optional way to get a date >2100. Use it.
But all the platforms I tested didn't seem to support it. But anyways
the x86-64 kernel should be ready for the 22nd century now. Actually i
shouldn't care about this because I will be dead by then @)
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch puts the code from head.S in a special .bootstrap.text
section.
I'm working on a patch to reorder the functions in the kernel (I'll post
that later), but for x86-64 at least the kernel bootstrap requires that
the head.S functions are on the very first page/pages of the kernel
text. This is understandable since the bootstrap is complex enough
already and not a problem at all, it just means they aren't allowed to
be reordered. This patch puts these special functions into a separate
section to document this, and to guarantee this in the light of possibly
reordering the rest later.
(So this patch doesn't fix a bug per se, but makes things more robust by
making the order of these functions explicit)
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There are more and more cases where we need to know DMI information
early to work around bugs. i386 already had early DMI scanning, but
x86-64 didn't. Implement this now.
This required some cleanup in the i386 code.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As suggested by Andi (and Alan), move the default kernel location
from 1Mb to 2Mb, to align to the start of a TLB entry.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In a micro-benchmark that stresses the pagefault path, the down_read_trylock
on the mmap_sem showed up quite high on the profile. Turns out this lock is
bouncing between cpus quite a bit and thus is cache-cold a lot. This patch
prefetches the lock (for write) as early as possible (and before some other
somewhat expensive operations). With this patch, the down_read_trylock
basically fell out of the top of profile.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
phys_proc_id[] on AMD boxes is right now populated with the initial
apic id, obtained by the cpuid instruction. But, the initial apic id
need not be the local apic id on clustered APIC systems (see comment at
x86_64/kernel/genapic_cluster.c, line 110). On vSMPowered with AMD
CPUs the cpu_to_node will turn out to be incorrect (as apicid_to_node[] is
indexed by the initial apic id rather than the local apic id).
On vSMPowered boxes with Intel CPUs this is working correctly as
phys_proc_id[] is initialized correctly in detect_ht().
This fixes AMD boot path according to specification, to use the correct
routines for local apic id and socket ids. We use
hard_smp_processor_id() to read the local apic id, and phys_pkg_id() to
determine socket id for phys_proc_id[]
Patch tested on Tyan multicore boxes as well as vSMPowered boxes.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- adjust limits of GDT/IDT pseudo-descriptors (some were off by one)
- move empty_zero_page into .bss.page_aligned
- move cpu_gdt_table into .data.page_aligned
- move idt_table into .bss
- align inital_code and init_rsp
- eliminate pointless (re-)declaration of idt_table in traps.c
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It needs num_physpages, so initialize it early. It's later overwritten
again.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Attached is a small code style cleanup patch that resulted from my
skimming through the arch/x86_64/kernel/traps.c code to figure out what
went haywire.
Signed-off-by: Roberto Nibali <ratz@drugphish.ch>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We've always had the problem that arguments only did a prefix match,
which resulted e.g. in noapic and noapictimer getting confused.
Fix the early argument parsing code to always check that arguments are
whole words (except for those that take additional arguments of course)
I factored out the checking code for that while also makes the code
easier to maintain.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
While the modular aspect of the respective i386 patch doesn't apply to
x86-64 (as the top level page directory entry is shared between modules
and the base kernel), handlers registered with register_die_notifier()
are still under similar constraints for touching ioremap()ed or
vmalloc()ed memory. The likelihood of this problem becoming visible is
of course significantly lower, as the assigned virtual addresses would
have to cross a 2**39 byte boundary. This is because the callback gets
invoked
(a) in the page fault path before the top level page table propagation
gets carried out (hence a fault to propagate the top level page table
entry/entries mapping to module's code/data would nest infinitly) and
(b) in the NMI path, where nested faults must absolutely not happen,
since otherwise the IRET from the nested fault re-enables NMIs,
potentially resulting in nested NMI occurences.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The code waits for the GART to clear the TLB flush bit. Use cpu_relax
in this time to allow hypervisors to yield the CPU in this time.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Matt Mackall
Alan Stern
Ingo Molnar
KAMEZAWA Hiroyuki
Siddha, Suresh B
Akinobu Mita
Prasanna S Panchamukhi
bibo,mao
bibo mao
Stephen Rothwell
H. Peter Anvin
Andi Kleen
Eric Dumazet
Kevin Winchester
Navin Boppuri
Jon Mason
Vivek Goyal
Brian Gerst
Chuck Ebbert
Arjan van de Ven
Arjan van de Ven
Dave Jones
Arjan van de Ven
Ravikiran G Thirumalai
Jan Beulich
Roberto Nibali