The FXSAVE information leak patch introduced a bug in FP exception
handling: it clears FP exceptions only when there are already
none outstanding. Mikael Pettersson reported that causes problems
with the Erlang runtime and has tested this fix.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These aren't needed by glibc or klibc, and they're broken in some cases
anyway. The uClibc folks are apparently switching over to stop using
them too (now that we agreed that they should be dropped, at least).
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Proposed fix for ptep_get_and_clear_full PAE bug. Pte_clear had the same bug,
so use the same fix for both. Turns out pmd_clear had it as well, but pgds
are not affected.
The problem is rather intricate. Page table entries in PAE mode are 64-bits
wide, but the only atomic 8-byte write operation available in 32-bit mode is
cmpxchg8b, which is expensive (at least on P4), and thus avoided. But it can
happen that the processor may prefetch entries into the TLB in the middle of an
operation which clears a page table entry. So one must always clear the P-bit
in the low word of the page table entry first when clearing it.
Since the sequence *ptep = __pte(0) leaves the order of the write dependent on
the compiler, it must be coded explicitly as a clear of the low word followed
by a clear of the high word. Further, there must be a write memory barrier
here to enforce proper ordering by the compiler (and, in the future, by the
processor as well).
On > 4GB memory machines, the implementation of pte_clear for PAE was clearly
deficient, as it could leave virtual mappings of physical memory above 4GB
aliased to memory below 4GB in the TLB. The implementation of
ptep_get_and_clear_full has a similar bug, although not nearly as likely to
occur, since the mappings being cleared are in the process of being destroyed,
and should never be dereferenced again.
But, as luck would have it, it is possible to trigger bugs even without ever
dereferencing these bogus TLB mappings, even if the clear is followed fairly
soon after with a TLB flush or invalidation. The problem is that memory above
4GB may now be aliased into the first 4GB of memory, and in fact, may hit a
region of memory with non-memory semantics. These regions include AGP and PCI
space. As such, these memory regions are not cached by the processor. This
introduces the bug.
The processor can speculate memory operations, including memory writes, as long
as they are committed with the proper ordering. Speculating a memory write to
a linear address that has a bogus TLB mapping is possible. Normally, the
speculation is harmless. But for cached memory, it does leave the falsely
speculated cacheline unmodified, but in a dirty state. This cache line will be
eventually written back. If this cacheline happens to intersect a region of
memory that is not protected by the cache coherency protocol, it can corrupt
data in I/O memory, which is generally a very bad thing to do, and can cause
total system failure or just plain undefined behavior.
These bugs are extremely unlikely, but the severity is of such magnitude, and
the fix so simple that I think fixing them immediately is justified. Also,
they are nearly impossible to debug.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
sys_splice() moves data to/from pipes with a file input/output. sys_vmsplice()
moves data to a pipe, with the input being a user address range instead.
This uses an approach suggested by Linus, where we can hold partial ranges
inside the pages[] map. Hopefully this will be useful for network
receive support as well.
Signed-off-by: Jens Axboe <axboe@suse.de>
AMD K7/K8 CPUs only save/restore the FOP/FIP/FDP x87 registers in FXSAVE
when an exception is pending. This means the value leak through
context switches and allow processes to observe some x87 instruction
state of other processes.
This was actually documented by AMD, but nobody recognized it as
being different from Intel before.
The fix first adds an optimization: instead of unconditionally
calling FNCLEX after each FXSAVE test if ES is pending and skip
it when not needed. Then do a x87 load from a kernel variable to
clear FOP/FIP/FDP.
This means other processes always will only see a constant value
defined by the kernel in their FP state.
I took some pain to make sure to chose a variable that's already
in L1 during context switch to make the overhead of this low.
Also alternative() is used to patch away the new code on CPUs
who don't need it.
Patch for both i386/x86-64.
The problem was discovered originally by Jan Beulich. Richard
Brunner provided the basic code for the workarounds, with contribution
from Jan.
This is CVE-2006-1056
Cc: richard.brunner@amd.com
Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
atomic_add_return() if CONFIG_M386 can accidentally enable local interrupts.
Signed-off-by: Lepton Wu <ytht.net@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Basically an in-kernel implementation of tee, which uses splice and the
pipe buffers as an intelligent way to pass data around by reference.
Where the user space tee consumes the input and produces a stdout and
file output, this syscall merely duplicates the data inside a pipe to
another pipe. No data is copied, the output just grabs a reference to the
input pipe data.
Signed-off-by: Jens Axboe <axboe@suse.de>
* 'splice' of git://brick.kernel.dk/data/git/linux-2.6-block:
[PATCH] vfs: add splice_write and splice_read to documentation
[PATCH] Remove sys_ prefix of new syscalls from __NR_sys_*
[PATCH] splice: warning fix
[PATCH] another round of fs/pipe.c cleanups
[PATCH] splice: comment styles
[PATCH] splice: add Ingo as addition copyright holder
[PATCH] splice: unlikely() optimizations
[PATCH] splice: speedups and optimizations
[PATCH] pipe.c/fifo.c code cleanups
[PATCH] get rid of the PIPE_*() macros
[PATCH] splice: speedup __generic_file_splice_read
[PATCH] splice: add direct fd <-> fd splicing support
[PATCH] splice: add optional input and output offsets
[PATCH] introduce a "kernel-internal pipe object" abstraction
[PATCH] splice: be smarter about calling do_page_cache_readahead()
[PATCH] splice: optimize the splice buffer mapping
[PATCH] splice: cleanup __generic_file_splice_read()
[PATCH] splice: only call wake_up_interruptible() when we really have to
[PATCH] splice: potential !page dereference
[PATCH] splice: mark the io page as accessed
__NR_sys_kexec_load should be __NR_kexec_load. Mainly affects users of the
_syscallN() macros, and glibc is already checking for __NR_kexec_load.
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Current implementations define NODES_SHIFT in include/asm-xxx/numnodes.h for
each arch. Its definition is sometimes configurable. Indeed, ia64 defines 5
NODES_SHIFT values in the current git tree. But it looks a bit messy.
SGI-SN2(ia64) system requires 1024 nodes, and the number of nodes already has
been changeable by config. Suitable node's number may be changed in the
future even if it is other architecture. So, I wrote configurable node's
number.
This patch set defines just default value for each arch which needs multi
nodes except ia64. But, it is easy to change to configurable if necessary.
On ia64 the number of nodes can be already configured in generic ia64 and SN2
config. But, NODES_SHIFT is defined for DIG64 and HP'S machine too. So, I
changed it so that all platforms can be configured via CONFIG_NODES_SHIFT. It
would be simpler.
See also: http://marc.theaimsgroup.com/?l=linux-kernel&m=114358010523896&w=2
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Jack Steiner <steiner@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
On i386, we don't use sys_ prefix for __NR_*. This patch removes it
[FWIW, _syscall*() macros will generate foo() instead of sys_foo().]
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Jens Axboe <axboe@suse.de>
If the HPET timer is enabled, the clock can drift by ~3 seconds a day.
This is due to the HPET timer not being initialized with the correct
setting (still using PIT count).
If HZ changes, this drift can become even more pronounced.
HPET patch initializes tick_nsec with correct tick_nsec settings for
HPET timer.
Vojtech comments:
"It's not entirely correct (it assumes the HPET ticks totally
exactly), but it's significantly better than assuming the PIT error
there."
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
AMD systems have a modern APIC that supports 8 bit IDs, but
don't have a XAPIC version number. Add a new "modern_apic"
subfunction that handles this correctly and use it (nearly)
everywhere where XAPIC is tested for.
I removed one wart: the code specified that external APICs
would use an 8bit APIC ID. But I checked a real 82093 data sheet
and it says clearly that they only use 4bit. So I removed
this special case since it would a bit awkward to implement now.
I removed the valid APIC tests in mptable parsing completely. On any modern
system they only check against the full field width (8bit) anyways
and are no-ops. This also fixes them doing the wrong thing
on >8 core Opterons.
This makes i386 boot again on 16 core Opterons.
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Introduce a e820_all_mapped() function which checks if the entire range
<start,end> is mapped with type.
This is done by moving the local start variable to the end of each
known-good region; if at the end of the function the start address is
still before end, there must be a part that's not of the correct type;
otherwise it's a good region.
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>
local_t's were defined to be unsigned. This increases confusion because
atomic_t's are signed. The patch goes through and changes all implementations
to use signed longs throughout.
Also, x86-64 was using 32-bit quantities for the value passed into local_add()
and local_sub(). Fixed.
All (actually, both) existing users have been audited.
(Also s/__inline__/inline/ in x86_64/local.h)
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the recently-added LINUX_FADV_ASYNC_WRITE and LINUX_FADV_WRITE_WAIT
fadvise() additions, do it in a new sys_sync_file_range() syscall instead.
Reasons:
- It's more flexible. Things which would require two or three syscalls with
fadvise() can be done in a single syscall.
- Using fadvise() in this manner is something not covered by POSIX.
The patch wires up the syscall for x86.
The sycall is implemented in the new fs/sync.c. The intention is that we can
move sys_fsync(), sys_fdatasync() and perhaps sys_sync() into there later.
Documentation for the syscall is in fs/sync.c.
A test app (sync_file_range.c) is in
http://www.zip.com.au/~akpm/linux/patches/stuff/ext3-tools.tar.gz.
The available-to-GPL-modules do_sync_file_range() is for knfsd: "A COMMIT can
say NFS_DATA_SYNC or NFS_FILE_SYNC. I can skip the ->fsync call for
NFS_DATA_SYNC which is hopefully the more common."
Note: the `async' writeout mode SYNC_FILE_RANGE_WRITE will turn synchronous if
the queue is congested. This is trivial to fix: add a new flag bit, set
wbc->nonblocking. But I'm not sure that we want to expose implementation
details down to that level.
Note: it's notable that we can sync an fd which wasn't opened for writing.
Same with fsync() and fdatasync()).
Note: the code takes some care to handle attempts to sync file contents
outside the 16TB offset on 32-bit machines. It makes such attempts appear to
succeed, for best 32-bit/64-bit compatibility. Perhaps it should make such
requests fail...
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Porting the patch I posted for x86_64 to i386.
http://marc.theaimsgroup.com/?l=linux-kernel&m=114178139610707&w=2
o While using kdump, after a system crash when second kernel boots, timer
vector gets (0x31) locked and CPU does not see timer interrupts
travelling from IOAPIC to APIC. Currently it does not lead to boot
failure in second kernel as timer interrupts continues to come as ExtInt
through LAPIC directly, but fixing it is good in case some boards do not
support the other mode.
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>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It's been disabled since v2.1.88
Signed-off-by: Brian Gerst <bgerst@didntduck.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This adds support for the sys_splice system call. Using a pipe as a
transport, it can connect to files or sockets (latter as output only).
From the splice.c comments:
"splice": joining two ropes together by interweaving their strands.
This is the "extended pipe" functionality, where a pipe is used as
an arbitrary in-memory buffer. Think of a pipe as a small kernel
buffer that you can use to transfer data from one end to the other.
The traditional unix read/write is extended with a "splice()" operation
that transfers data buffers to or from a pipe buffer.
Named by Larry McVoy, original implementation from Linus, extended by
Jens to support splicing to files and fixing the initial implementation
bugs.
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nowadays, even Debian stable ships a microcode_ctl utility recent enough to no
longer use this ioctl.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Tigran Aivazian <tigran_aivazian@symantec.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix up some RTC whitespace and style
Signed-off-by: Matt Mackall <mpm@selenic.com>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reading the CMOS clock on x86 and some other arches currently takes up to one
second because it synchronizes with the CMOS second tick-over. This delay
shows up at boot time as well a resume time.
This is the currently the most substantial boot time delay for machines that
are working towards instant-on capability. Also, a quick back of the envelope
calculation (.5sec * 2M users * 1 boot a day * 10 years) suggests it has cost
Linux users in the neighborhood of a million man-hours.
An earlier thread on this topic is here:
http://groups.google.com/group/linux.kernel/browse_frm/thread/8a24255215ff6151/2aa97e66a977653d?hl=en&lr=&ie=UTF-8&rnum=1&prev=/groups%3Fhl%3Den%26lr%3D%26ie%3DUTF-8%26selm%3D1To2R-2S7-11%40gated-at.bofh.it#2aa97e66a977653d
..from which the consensus seems to be that it's no longer desirable.
In my view, there are basically four cases to consider:
1) networked, need precise walltime: use NTP
2) networked, don't need precise walltime: use NTP anyway
3) not networked, don't need sub-second precision walltime: don't care
4) not networked, need sub-second precision walltime:
get a network or a radio time source because RTC isn't good enough anyway
So this patch series simply removes the synchronization in favor of a simple
seqlock-like approach using the seconds value.
Note that for purposes of timer accuracy on wakeup, this patch will cause us
to fire timers up to one second late. But as the current timer resume code
will already sync once (or more!), it's no worse for short timers.
Signed-off-by: Matt Mackall <mpm@selenic.com>
Cc: Andi Kleen <ak@muc.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
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>
- fix: initialize the robust list(s) to NULL in copy_process.
- doc update
- cleanup: rename _inuser to _inatomic
- __user cleanups and other small cleanups
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: 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>
i386: 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>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patchset provides a new (written from scratch) implementation of robust
futexes, called "lightweight robust futexes". We believe this new
implementation is faster and simpler than the vma-based robust futex solutions
presented before, and we'd like this patchset to be adopted in the upstream
kernel. This is version 1 of the patchset.
Background
----------
What are robust futexes? To answer that, we first need to understand what
futexes are: normal futexes are special types of locks that in the
noncontended case can be acquired/released from userspace without having to
enter the kernel.
A futex is in essence a user-space address, e.g. a 32-bit lock variable
field. If userspace notices contention (the lock is already owned and someone
else wants to grab it too) then the lock is marked with a value that says
"there's a waiter pending", and the sys_futex(FUTEX_WAIT) syscall is used to
wait for the other guy to release it. The kernel creates a 'futex queue'
internally, so that it can later on match up the waiter with the waker -
without them having to know about each other. When the owner thread releases
the futex, it notices (via the variable value) that there were waiter(s)
pending, and does the sys_futex(FUTEX_WAKE) syscall to wake them up. Once all
waiters have taken and released the lock, the futex is again back to
'uncontended' state, and there's no in-kernel state associated with it. The
kernel completely forgets that there ever was a futex at that address. This
method makes futexes very lightweight and scalable.
"Robustness" is about dealing with crashes while holding a lock: if a process
exits prematurely while holding a pthread_mutex_t lock that is also shared
with some other process (e.g. yum segfaults while holding a pthread_mutex_t,
or yum is kill -9-ed), then waiters for that lock need to be notified that the
last owner of the lock exited in some irregular way.
To solve such types of problems, "robust mutex" userspace APIs were created:
pthread_mutex_lock() returns an error value if the owner exits prematurely -
and the new owner can decide whether the data protected by the lock can be
recovered safely.
There is a big conceptual problem with futex based mutexes though: it is the
kernel that destroys the owner task (e.g. due to a SEGFAULT), but the kernel
cannot help with the cleanup: if there is no 'futex queue' (and in most cases
there is none, futexes being fast lightweight locks) then the kernel has no
information to clean up after the held lock! Userspace has no chance to clean
up after the lock either - userspace is the one that crashes, so it has no
opportunity to clean up. Catch-22.
In practice, when e.g. yum is kill -9-ed (or segfaults), a system reboot is
needed to release that futex based lock. This is one of the leading
bugreports against yum.
To solve this problem, 'Robust Futex' patches were created and presented on
lkml: the one written by Todd Kneisel and David Singleton is the most advanced
at the moment. These patches all tried to extend the futex abstraction by
registering futex-based locks in the kernel - and thus give the kernel a
chance to clean up.
E.g. in David Singleton's robust-futex-6.patch, there are 3 new syscall
variants to sys_futex(): FUTEX_REGISTER, FUTEX_DEREGISTER and FUTEX_RECOVER.
The kernel attaches such robust futexes to vmas (via
vma->vm_file->f_mapping->robust_head), and at do_exit() time, all vmas are
searched to see whether they have a robust_head set.
Lots of work went into the vma-based robust-futex patch, and recently it has
improved significantly, but unfortunately it still has two fundamental
problems left:
- they have quite complex locking and race scenarios. The vma-based
patches had been pending for years, but they are still not completely
reliable.
- they have to scan _every_ vma at sys_exit() time, per thread!
The second disadvantage is a real killer: pthread_exit() takes around 1
microsecond on Linux, but with thousands (or tens of thousands) of vmas every
pthread_exit() takes a millisecond or more, also totally destroying the CPU's
L1 and L2 caches!
This is very much noticeable even for normal process sys_exit_group() calls:
the kernel has to do the vma scanning unconditionally! (this is because the
kernel has no knowledge about how many robust futexes there are to be cleaned
up, because a robust futex might have been registered in another task, and the
futex variable might have been simply mmap()-ed into this process's address
space).
This huge overhead forced the creation of CONFIG_FUTEX_ROBUST, but worse than
that: the overhead makes robust futexes impractical for any type of generic
Linux distribution.
So it became clear to us, something had to be done. Last week, when Thomas
Gleixner tried to fix up the vma-based robust futex patch in the -rt tree, he
found a handful of new races and we were talking about it and were analyzing
the situation. At that point a fundamentally different solution occured to
me. This patchset (written in the past couple of days) implements that new
solution. Be warned though - the patchset does things we normally dont do in
Linux, so some might find the approach disturbing. Parental advice
recommended ;-)
New approach to robust futexes
------------------------------
At the heart of this new approach there is a per-thread private list of robust
locks that userspace is holding (maintained by glibc) - which userspace list
is registered with the kernel via a new syscall [this registration happens at
most once per thread lifetime]. At do_exit() time, the kernel checks this
user-space list: are there any robust futex locks to be cleaned up?
In the common case, at do_exit() time, there is no list registered, so the
cost of robust futexes is just a simple current->robust_list != NULL
comparison. If the thread has registered a list, then normally the list is
empty. If the thread/process crashed or terminated in some incorrect way then
the list might be non-empty: in this case the kernel carefully walks the list
[not trusting it], and marks all locks that are owned by this thread with the
FUTEX_OWNER_DEAD bit, and wakes up one waiter (if any).
The list is guaranteed to be private and per-thread, so it's lockless. There
is one race possible though: since adding to and removing from the list is
done after the futex is acquired by glibc, there is a few instructions window
for the thread (or process) to die there, leaving the futex hung. To protect
against this possibility, userspace (glibc) also maintains a simple per-thread
'list_op_pending' field, to allow the kernel to clean up if the thread dies
after acquiring the lock, but just before it could have added itself to the
list. Glibc sets this list_op_pending field before it tries to acquire the
futex, and clears it after the list-add (or list-remove) has finished.
That's all that is needed - all the rest of robust-futex cleanup is done in
userspace [just like with the previous patches].
Ulrich Drepper has implemented the necessary glibc support for this new
mechanism, which fully enables robust mutexes. (Ulrich plans to commit these
changes to glibc-HEAD later today.)
Key differences of this userspace-list based approach, compared to the vma
based method:
- it's much, much faster: at thread exit time, there's no need to loop
over every vma (!), which the VM-based method has to do. Only a very
simple 'is the list empty' op is done.
- no VM changes are needed - 'struct address_space' is left alone.
- no registration of individual locks is needed: robust mutexes dont need
any extra per-lock syscalls. Robust mutexes thus become a very lightweight
primitive - so they dont force the application designer to do a hard choice
between performance and robustness - robust mutexes are just as fast.
- no per-lock kernel allocation happens.
- no resource limits are needed.
- no kernel-space recovery call (FUTEX_RECOVER) is needed.
- the implementation and the locking is "obvious", and there are no
interactions with the VM.
Performance
-----------
I have benchmarked the time needed for the kernel to process a list of 1
million (!) held locks, using the new method [on a 2GHz CPU]:
- with FUTEX_WAIT set [contended mutex]: 130 msecs
- without FUTEX_WAIT set [uncontended mutex]: 30 msecs
I have also measured an approach where glibc does the lock notification [which
it currently does for !pshared robust mutexes], and that took 256 msecs -
clearly slower, due to the 1 million FUTEX_WAKE syscalls userspace had to do.
(1 million held locks are unheard of - we expect at most a handful of locks to
be held at a time. Nevertheless it's nice to know that this approach scales
nicely.)
Implementation details
----------------------
The patch adds two new syscalls: one to register the userspace list, and one
to query the registered list pointer:
asmlinkage long
sys_set_robust_list(struct robust_list_head __user *head,
size_t len);
asmlinkage long
sys_get_robust_list(int pid, struct robust_list_head __user **head_ptr,
size_t __user *len_ptr);
List registration is very fast: the pointer is simply stored in
current->robust_list. [Note that in the future, if robust futexes become
widespread, we could extend sys_clone() to register a robust-list head for new
threads, without the need of another syscall.]
So there is virtually zero overhead for tasks not using robust futexes, and
even for robust futex users, there is only one extra syscall per thread
lifetime, and the cleanup operation, if it happens, is fast and
straightforward. The kernel doesnt have any internal distinction between
robust and normal futexes.
If a futex is found to be held at exit time, the kernel sets the highest bit
of the futex word:
#define FUTEX_OWNER_DIED 0x40000000
and wakes up the next futex waiter (if any). User-space does the rest of
the cleanup.
Otherwise, robust futexes are acquired by glibc by putting the TID into the
futex field atomically. Waiters set the FUTEX_WAITERS bit:
#define FUTEX_WAITERS 0x80000000
and the remaining bits are for the TID.
Testing, architecture support
-----------------------------
I've tested the new syscalls on x86 and x86_64, and have made sure the parsing
of the userspace list is robust [ ;-) ] even if the list is deliberately
corrupted.
i386 and x86_64 syscalls are wired up at the moment, and Ulrich has tested the
new glibc code (on x86_64 and i386), and it works for his robust-mutex
testcases.
All other architectures should build just fine too - but they wont have the
new syscalls yet.
Architectures need to implement the new futex_atomic_cmpxchg_inuser() inline
function before writing up the syscalls (that function returns -ENOSYS right
now).
This patch:
Add placeholder futex_atomic_cmpxchg_inuser() implementations to every
architecture that supports futexes. It returns -ENOSYS.
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>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Just about every architecture defines some macros to do operations on pfns.
They're all virtually identical. This patch consolidates all of them.
One minor glitch is that at least i386 uses them in a very skeletal header
file. To keep away from #include dependency hell, I stuck the new
definitions in a new, isolated header.
Of all of the implementations, sh64 is the only one that varied by a bit.
It used some masks to ensure that any sign-extension got ripped away before
the arithmetic is done. This has been posted to that sh64 maintainers and
the development list.
Compiles on x86, x86_64, ia64 and ppc64.
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
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>
Current kprobe copies the original instruction at the probe point and replaces
it with a breakpoint instruction (int3). When the kernel hits the probe
point, kprobe handler is invoked. And the copied instruction is single-step
executed on the copied buffer (not on the original address) by kprobe. After
that, the kprobe checks registers and modify it (if need) as if the
instructions was executed on the original address.
My proposal is based on the fact there are many instructions which do NOT
require the register modification after the single-step execution. When the
copied instruction is a kind of them, kprobe just jumps back to the next
instruction after single-step execution. If so, why don't we execute those
instructions directly?
With kprobe-booster patch, kprobes will execute a copied instruction directly
and (if need) jump back to original code. This direct execution is executed
when the kprobe don't have both post_handler and break_handler, and the copied
instruction can be executed directly.
I sorted instructions which can be executed directly or not;
- Call instructions are NG(can not be executed directly).
We should correct the return address pushed into top of stack.
- Indirect instructions except for absolute indirect-jumps
are NG. Those instructions changes EIP randomly. We should
check EIP and correct it.
- Instructions that change EIP beyond the range of the
instruction buffer are NG.
- Instructions that change EIP to tail 5 bytes of the
instruction buffer (it is the size of a jump instruction).
We must write a jump instruction which backs to original
kernel code in the instruction buffer.
- Break point instruction is NG. We should not touch EIP and
pass to other handlers.
- Absolute direct/indirect jumps are OK.- Conditional Jumps are NG.
- Halt and software-interruptions are NG. Because it will stay on
the instruction buffer of kprobes.
- Prefixes are NG.
- Unknown/reserved opcode is NG.
- Other 1 byte instructions are OK. But those instructions need a
jump back code.
- 2 bytes instructions are mapped sparsely. So, in this release,
this patch don't boost those instructions.
>From Intel's IA-32 opcode map described in IA-32 Intel Architecture Software
Developer's Manual Vol.2 B, I determined that following opcodes are not
boostable.
- 0FH (2byte escape)
- 70H - 7FH (Jump on condition)
- 9AH (Call) and 9CH (Pushf)
- C0H-C1H (Grp 2: includes reserved opcode)
- C6H-C7H (Grp11: includes reserved opcode)
- CCH-CEH (Software-interrupt)
- D0H-D3H (Grp2: includes reserved opcode)
- D6H (Reserved)
- D8H-DFH (Coprocessor)
- E0H-E3H (loop/conditional jump)
- E8H (Call)
- F0H-F3H (Prefixes and reserved)
- F4H (Halt)
- F6H-F7H (Grp3: includes reserved opcode)
- FEH-FFH(Grp4,5: includes reserved opcode)
Kprobe-booster checks whether target instruction can be boosted (can be
executed directly) at arch_copy_kprobe() function. If the target instruction
can be boosted, it clears "boostable" flag. If not, it sets "boostable" flag
-1. This is disabled status. In resume_execution() function, If "boostable"
flag is cleared, kprobe-booster measures the size of the target instruction
and sets "boostable" flag 1.
In kprobe_handler(), kprobe checks the "boostable" flag. If the flag is 1, it
resets current kprobe and executes instruction buffer directly instead of
single stepping.
When unregistering a boosted kprobe, it calls synchronize_sched()
after "int3" is removed. So we can ensure followings after
the synchronize_sched() called.
- interrupt handlers are finished on all CPUs.
- instruction buffer is not executed on all CPUs.
And we can release the boosted kprobe safely.
And also, on preemptible kernel, the booster is not enabled where the kernel
preemption is enabled. So, there are no preempted threads on the instruction
buffer.
The description of kretprobe-booster:
====================================
In the normal operation, kretprobe make a target function return to trampoline
code. And a kprobe (called trampoline_probe) have been inserted at the
trampoline code. When the kernel hits this kprobe, it calls kretprobe's
handler and it returns to original return address.
Kretprobe-booster patch removes the trampoline_probe. It allows the
trampoline code to call kretprobe's handler directly instead of invoking
kprobe. And tranpoline code returns to original return address.
This new trampoline code stores and restores registers, so the kretprobe
handler is still able to access those registers.
Current kprobe has about 1.3 usec/probe(*) overhead, and kprobe-booster patch
reduces it to 0.6 usec/probe(*). Also current kretprobe has about 2.0
usec/probe(*) overhead. Kprobe-booster patch reduces it to 1.3 usec/probe(*),
and the combination of both kprobe-booster patch and kretprobe-booster patch
reduces it to 0.9 usec/probe(*).
I expect the combination of both patches can reduce half of a probing
overhead.
Performance numbers strongly depend on the processor model.
Andrew Morton wrote:
> These preempt tricks look rather nasty. Can you please describe what the
> problem is, precisely? And how this code avoids it? Perhaps we can find
> something cleaner.
The problem is how to remove the copied instructions of the
kprobe *safely* on the preemptable kernel (CONFIG_PREEMPT=y).
Kprobes basically executes the following actions;
(1)int3
(2)preempt_disable()
(3)kprobe_prehandler()
(4)copied instructioin(single step)
(5)kprobe_posthandler()
(6)preempt_enable()
(7)return to the original code
During the execution of copied instruction, preemption is
disabled (from step (2) to (6)).
When unregistering the probes, Kprobe waits for RCU
quiescent state by using synchronize_sched() after removing
int3 instruction.
Thus we can ensure the copied instruction is not executed.
On the other hand, kprobe-booster executes the following actions;
(1)int3
(2)preempt_disable()
(3)kprobe_prehandler()
(4)preempt_enable() <-- this one is added by my patch
(5)copied instruction(direct execution)
(6)jmp back to the original code
The problem is that we have no way to prevent preemption on
step (5) or (6). We cannot call preempt_disable() after step (6),
because there are no rooms to do that. Thus, some other
processes may be preempted at step(5) or (6) on preemptable kernel.
And I couldn't find the easy way to ensure that other processes'
stack do *not* have the address of them. (I thought some way
to do that, but those are very costly.)
So currently, I simply boost the kprobe only when the probe
point is already preemption disabled.
> Also, the patch adds a preempt_enable() but I don't see a corresponding
> preempt_disable(). Am I missing something?
It is corresponding to the preempt_disable() in the top of
kprobe_handler().
I copied the code of kprobe_handler() here:
static int __kprobes kprobe_handler(struct pt_regs *regs)
{
struct kprobe *p;
int ret = 0;
kprobe_opcode_t *addr = NULL;
unsigned long *lp;
struct kprobe_ctlblk *kcb;
/*
* We don't want to be preempted for the entire
* duration of kprobe processing
*/
preempt_disable(); <-- HERE
kcb = get_kprobe_ctlblk();
Signed-off-by: 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>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add blkcnt_t as the type of inode.i_blocks. This enables you to make the size
of blkcnt_t either 4 bytes or 8 bytes on 32 bits architecture with CONFIG_LSF.
- CONFIG_LSF
Add new configuration parameter.
- blkcnt_t
On h8300, i386, mips, powerpc, s390 and sh that define sector_t,
blkcnt_t is defined as u64 if CONFIG_LSF is enabled; otherwise it is
defined as unsigned long.
On other architectures, it is defined as unsigned long.
- inode.i_blocks
Change the type from sector_t to blkcnt_t.
Signed-off-by: Takashi Sato <sho@tnes.nec.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch series fixes the following problems on 32 bits architecture.
o stat64 returns the lower 32 bits of blocks, although userland st_blocks
has 64 bits, because i_blocks has only 32 bits. The ioctl with FIOQSIZE has
the same problem.
o As Dave Kleikamp said, making >2TB file on JFS results in writing an
invalid block number to disk inode. The cause is the same as above too.
o In generic quota code dquot_transfer(), the file usage is calculated from
i_blocks via inode_get_bytes(). If the file is over 2TB, the change of
usage is less than expected. The cause is the same as above too.
o As Trond Myklebust said, statfs64's entries related to blocks are invalid
on statfs64 for a network filesystem which has more than 2^32-1 blocks with
CONFIG_LBD disabled. [PATCH 3/3]
We made patches to fix problems that occur when handling a large filesystem
and a large file. It was discussed on the mails titled "stat64 for over 2TB
file returned invalid st_blocks".
Signed-off-by: Takashi Sato <sho@tnes.nec.co.jp>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Jan Kara <jack@ucw.cz>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
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>
Implement the half-closed devices notifiation, by adding a new POLLRDHUP
(and its alias EPOLLRDHUP) bit to the existing poll/select sets. Since the
existing POLLHUP handling, that does not report correctly half-closed
devices, was feared to be changed, this implementation leaves the current
POLLHUP reporting unchanged and simply add a new bit that is set in the few
places where it makes sense. The same thing was discussed and conceptually
agreed quite some time ago:
http://lkml.org/lkml/2003/7/12/116
Since this new event bit is added to the existing Linux poll infrastruture,
even the existing poll/select system calls will be able to use it. As far
as the existing POLLHUP handling, the patch leaves it as is. The
pollrdhup-2.6.16.rc5-0.10.diff defines the POLLRDHUP for all the existing
archs and sets the bit in the six relevant files. The other attached diff
is the simple change required to sys/epoll.h to add the EPOLLRDHUP
definition.
There is "a stupid program" to test POLLRDHUP delivery here:
http://www.xmailserver.org/pollrdhup-test.c
It tests poll(2), but since the delivery is same epoll(2) will work equally.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
include/linux/platform.h contained nothing that was actually used except
the default_idle() prototype, and is therefore removed by this patch.
This patch does the following with the platform specific default_idle()
functions on different architectures:
- remove the unused function:
- parisc
- sparc64
- make the needlessly global function static:
- arm
- h8300
- m68k
- m68knommu
- s390
- v850
- x86_64
- add a prototype in asm/system.h:
- cris
- i386
- ia64
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Patrick Mochel <mochel@digitalimplant.org>
Acked-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
unused isa_...() helpers removed.
Adrian Bunk:
The asm-sh part was rediffed due to unrelated changes.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Without branch hints, the very unlikely chance of the loop repeating due to
cmpxchg failure is unrolled with gcc-4 that I have tested.
Improve this for architectures with a native cas/cmpxchg. llsc archs
should try to implement this natively.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Andi Kleen <ak@muc.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Seems like needless clutter having a bunch of #if defined(CONFIG_$ARCH) in
include/linux/cache.h. Move the per architecture section definition to
asm/cache.h, and keep the if-not-defined dummy case in linux/cache.h to
catch architectures which don't implement the section.
Verified that symbols still go in .data.read_mostly on parisc,
and the compile doesn't break.
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Gcc reserves %ebx when compiling position-independent-code on i386. This
means, the _syscallX() macros in include/asm-i386/unistd.h will not
compile. This patch is changes the existing macros to take special care to
preserve %ebx.
The bug can be tracked at http://bugzilla.kernel.org/show_bug.cgi?id=6204
Signed-off-by: Markus Gutschke <markus@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
_raw_spin_lock_flags() is entered with interrupts disabled. If it cannot
obtain a spinlock, it checks the flags that were passed and re-enables
interrupts before spinning if that's how the flags are set. When the
spinlock might be available, it disables interrupts (even if they are
already disabled) before trying to get the lock. Change that so interrupts
are only disabled if they have been enabled. This costs nine bytes of
duplicated spinloop code.
Fastpath before patch:
jle <keep looping> not-taken conditional jump
cli disable interrupts
jmp <try for lock> unconditional jump
Fastpath after patch, if interrupts were not enabled:
jg <try for lock> taken conditional branch
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
arch/i386/kernel/cpu/centaur.c: In function `centaur_mcr_insert':
arch/i386/kernel/cpu/centaur.c:33: warning: implicit declaration of function `mtrr_centaur_report_mcr'
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
>commit 76381fee7e
>Author: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
>Date: Thu Jun 23 00:08:46 2005 -0700
>
> [PATCH] xen: x86_64: use more usermode macro
>
> Make use of the user_mode macro where it's possible. This is useful for Xen
> because it will need only to redefine only the macro to a hypervisor call.
I am of the opinion that the above changeset is incomplete, i.e. it missed
converting some previous uses of user_mode to user_mode_vm. While most of
them could be considered just cosmetical, at least the one in die_nmi
doesn't appear to be.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Cc: Zachary Amsden <zach@vmware.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Registering a callback handler through register_die_notifier() is obviously
primarily intended for use by modules. However, the way these currently
get called it is basically impossible for them to actually be used by
modules, as there is, on non-PAE configurationes, a good chance (the larger
the module, the better) for the system to crash as a result.
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.
Besides the modular aspect, similar problems would even arise for in-
kernel consumers of the API if they touched ioremap()ed or vmalloc()ed
memory inside their handlers.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The history is that -mm kernels do not work for me for a few months
already. The things started from crashing somewhere after starting init,
and for the last month - no boot at all, just "Uncompressing... OK,
booting kernel", and silence. Early console didn't work too. With the
latest releases this degraded into an infinite stream of the "Unknown
interrupt or fault" messages. So today my patience ran out and I started
to think how can I collect at least some info for the bug-report. Attached
is the patch that allows to gather some valueable debug info on the problem
by making an early console more useable. I can't properly test the patch,
as the kernel still doesn't boot, so I'll explain it in details in a hope
someone else can justify the intrusive changes.
arch_hooks.h: added prototypes for setup_early_printk() and early_printk().
setup.c: killed wrong setup_early_printk() prototype. Moved
setup_early_printk() a bit earlier, as it was not "early enough" to cover
the bug I was fighting with.
early_printk.c: made it to start printing from the bottom of the screen,
otherwise the messages interfere with the ones of the boot-loader, so you
can't read them.
Signed-off-by: Stas Sergeev <stsp@aknet.ru>
Cc: Andi Kleen <ak@muc.de>
Cc: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
ES7000 platform code clean up for compilation errors and a warning.
Ifdef'd the ACPI related parts in the ES7000 platform code. They were
causing compile errors in certain configuration (without ACPI defined). I
think this approach would be best (as opposed to Kconfig changes) since it
only touches the subarch...
Signed-off-by: <Natalie.Protasevich@unisys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In some code I am developing I had occasion to change the type of a
variable. This made the value put_user was putting to user space wrong.
But the code continued to build cleanly without errors.
Introducing a temporary fixes this problem and at least with gcc-3.3.5 does
not cause gcc any problems with optimizing out the temporary. gcc-4.x
using SSA internally ought to be even better at optimizing out temporaries,
so I don't expect a temporary to become a problem. Especially because in
all correct cases the types on both sides of the assignment to the
temporary are the same.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Implement SMP alternatives, i.e. switching at runtime between different
code versions for UP and SMP. The code can patch both SMP->UP and UP->SMP.
The UP->SMP case is useful for CPU hotplug.
With CONFIG_CPU_HOTPLUG enabled the code switches to UP at boot time and
when the number of CPUs goes down to 1, and switches to SMP when the number
of CPUs goes up to 2.
Without CONFIG_CPU_HOTPLUG or on non-SMP-capable systems the code is
patched once at boot time (if needed) and the tables are released
afterwards.
The changes in detail:
* The current alternatives bits are moved to a separate file,
the SMP alternatives code is added there.
* The patch adds some new elf sections to the kernel:
.smp_altinstructions
like .altinstructions, also contains a list
of alt_instr structs.
.smp_altinstr_replacement
like .altinstr_replacement, but also has some space to
save original instruction before replaving it.
.smp_locks
list of pointers to lock prefixes which can be nop'ed
out on UP.
The first two are used to replace more complex instruction
sequences such as spinlocks and semaphores. It would be possible
to deal with the lock prefixes with that as well, but by handling
them as special case the table sizes become much smaller.
* The sections are page-aligned and padded up to page size, so they
can be free if they are not needed.
* Splitted the code to release init pages to a separate function and
use it to release the elf sections if they are unused.
Signed-off-by: Gerd Hoffmann <kraxel@suse.de>
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2.6.16-rc3 uses hugetlb on-demand paging, but it doesn_t support hugetlb
mprotect.
From: David Gibson <david@gibson.dropbear.id.au>
Remove a test from the mprotect() path which checks that the mprotect()ed
range on a hugepage VMA is hugepage aligned (yes, really, the sense of
is_aligned_hugepage_range() is the opposite of what you'd guess :-/).
In fact, we don't need this test. If the given addresses match the
beginning/end of a hugepage VMA they must already be suitably aligned. If
they don't, then mprotect_fixup() will attempt to split the VMA. The very
first test in split_vma() will check for a badly aligned address on a
hugepage VMA and return -EINVAL if necessary.
From: "Chen, Kenneth W" <kenneth.w.chen@intel.com>
On i386 and x86-64, pte flag _PAGE_PSE collides with _PAGE_PROTNONE. The
identify of hugetlb pte is lost when changing page protection via mprotect.
A page fault occurs later will trigger a bug check in huge_pte_alloc().
The fix is to always make new pte a hugetlb pte and also to clean up
legacy code where _PAGE_PRESENT is forced on in the pre-faulting day.
Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
check_acpi_pci() is called from arch/i386/kernel/setup.c even if
CONFIG_ACPI is not defined, but the code in include/asm/acpi.h doesn't
provide it in this case.
Signed-off-by: Herbert Pötzl <herbert@13thfloor.at>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
ATI chipsets tend to generate double timer interrupts for the local APIC
timer when both the 8254 and the IO-APIC timer pins are enabled. This is
because they route it to both and the result is anded together and the CPU
ends up processing it twice.
This patch changes check_timer to disable the 8254 routing for interrupt 0.
I think it would be safe on all chipsets actually (i tested it on a couple
and it worked everywhere) and Windows seems to do it in a similar way, but
to be conservative this patch only enables this mode on ATI (and adds
options to enable/disable too)
Ported over from a similar x86-64 change.
I reused the ACPI earlyquirk infrastructure for the ATI bridge check, but
tweaked it a bit to work even without ACPI.
Inspired by a patch from Chuck Ebbert, but redone.
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Recent GDT changes broke the SMP boot sequence if the booting CPU is
numbered anything other than zero. There's also a subtle source of error
in that the boot time CPU now uses cpu_gdt_table (which is actually the GDT
for booting CPUs in head.S). This patch fixes both problems by making GDT
descriptors themselves allocated from a per_cpu area and switching to them
in cpu_init(), which now means that cpu_gdt_table is exclusively used for
booting CPUs again.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix a problem seen on i686 machine with NX support where the instruction
could not be single stepped because of NX bit set on the memory pages
allocated by kprobes module. This patch provides allocation of instruction
solt so that the processor can execute the instruction from that location
similar to x86_64 architecture. Thanks to Bibo and Masami for testing this
patch.
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>
Do not mask TIF_SINGLESTEP bit in _TIF_WORK_MASK. Masking this stopped
do_notify_resume() from being called when it should have been.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make new MADV_REMOVE, MADV_DONTFORK, MADV_DOFORK consistent across all
arches. The idea is to make it possible to use them portably even before
distros include them in libc headers.
Move common flags to asm-generic/mman.h
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently, copy-on-write may change the physical address of a page even if the
user requested that the page is pinned in memory (either by mlock or by
get_user_pages). This happens if the process forks meanwhile, and the parent
writes to that page. As a result, the page is orphaned: in case of
get_user_pages, the application will never see any data hardware DMA's into
this page after the COW. In case of mlock'd memory, the parent is not getting
the realtime/security benefits of mlock.
In particular, this affects the Infiniband modules which do DMA from and into
user pages all the time.
This patch adds madvise options to control whether memory range is inherited
across fork. Useful e.g. for when hardware is doing DMA from/into these
pages. Could also be useful to an application wanting to speed up its forks
by cutting large areas out of consideration.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The correct way to export hyperthreading based functions is to predicate
them on CONFIG_X86_HT. Without this, the topology exporting patch breaks
the build on all non-PC x86 subarchitectures.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The *at patches introduced fstatat and, due to inusfficient research, I
used the newfstat functions generally as the guideline. The result is that
on 32-bit platforms we don't have all the information needed to implement
fstatat64.
This patch modifies the code to pass up 64-bit information if
__ARCH_WANT_STAT64 is defined. I renamed the syscall entry point to make
this clear. Other archs will continue to use the existing code. On x86-64
the compat code is implemented using a new sys32_ function. this is what
is done for the other stat syscalls as well.
This patch might break some other archs (those which define
__ARCH_WANT_STAT64 and which already wired up the syscall). Yet others
might need changes to accomodate the compatibility mode. I really don't
want to do that work because all this stat handling is a mess (more so in
glibc, but the kernel is also affected). It should be done by the arch
maintainers. I'll provide some stand-alone test shortly. Those who are
eager could compile glibc and run 'make check' (no installation needed).
The patch below has been tested on x86 and x86-64.
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Registers system call for the i386 architecture.
Signed-off-by: Janak Desai <janak@us.ibm.com>
Cc: Al Viro <viro@ftp.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix gcc4.1 compile warnings "value computed is not used" with
set_current_state() and set_task_state() on i386/SMP and x86-64.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Enable selection of different user/kernel VM splits for i386, including an
optimized mode for 1GB physical RAM, which gives the kernel a direct (non
HIGHMEM) mapping to the entire 1GB rather than just the first 896MB.
There is a similarly a similarly optimized mode for machines with exactly 2GB
of physical RAM.
This can speed up the kernel by avoiding having to create/destroy temporary
HIGHMEM mappings, and by not having to include HIGHMEM support at all on such
machines. The flip side is that there's less virtual addressing left for
userspace in these alternatives, and some binary-only kernel modules may
misbehave unless rebuilt with the same VMSPLIT option as the main kernel
image.
Original idea/patch from Jens Axboe, modified based on suggestions from Linus
et al.
Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is a subset of the bluesmoke project core code, stripped of the NMI work
which isn't ready to merge and some of the "interesting" proc functionality
that needs reworking or just has no place in kernel. It requires no core
kernel changes except the added scrub functions already posted.
The goal is to merge further functionality only after the core code is
accepted and proven in the base kernel, and only at the point the upstream
extras are really ready to merge.
From: doug thompson <norsk5@xmission.com>
This converts EDAC to sysfs and is the final chunk neccessary before EDAC
has a stable user space API and can be considered for submission into the
base kernel.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: doug thompson <norsk5@xmission.com>
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
EDAC requires a way to scrub memory if an ECC error is found and the chipset
does not do the work automatically. That means rewriting memory locations
atomically with respect to all CPUs _and_ bus masters. That means we can't
use atomic_add(foo, 0) as it gets optimised for non-SMP
This adds a function to include/asm-foo/atomic.h for the platforms currently
supported which implements a scrub of a mapped block.
It also adjusts a few other files include order where atomic.h is included
before types.h as this now causes an error as atomic_scrub uses u32.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add the sys_pselect6() and sys_poll() calls to the i386 syscall table.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Handle TIF_RESTORE_SIGMASK as added by David Woodhouse's patch entitled:
[PATCH] 2/3 Add TIF_RESTORE_SIGMASK support for arch/powerpc
[PATCH] 3/3 Generic sys_rt_sigsuspend
It does the following:
(1) Declares TIF_RESTORE_SIGMASK for i386.
(2) Invokes it over to do_signal() when TIF_RESTORE_SIGMASK is set.
(3) Makes do_signal() support TIF_RESTORE_SIGMASK, using the signal mask saved
in current->saved_sigmask.
(4) Discards sys_rt_sigsuspend() from the arch, using the generic one instead.
(5) Makes sys_sigsuspend() save the signal mask and set TIF_RESTORE_SIGMASK
rather than attempting to fudge the return registers.
(6) Makes sys_sigsuspend() return -ERESTARTNOHAND rather than looping
intrinsically.
(7) Makes setup_frame(), setup_rt_frame() and handle_signal() return 0 or
-EFAULT rather than true/false to be consistent with the rest of the
kernel.
Due to the fact do_signal() is then only called from one place:
(8) Makes do_signal() no longer have a return value is it was just being
ignored; force_sig() takes care of this.
(9) Discards the old sigmask argument to do_signal() as it's no longer
necessary.
(10) Makes do_signal() static.
(11) Marks the second argument to do_notify_resume() as unused. The unused
argument should remain in the middle as the arguments are passed in as
registers, and the ordering is specific in entry.S
Given the way do_signal() is now no longer called from sys_{,rt_}sigsuspend(),
they no longer need access to the exception frame, and so can just take
arguments normally.
This patch depends on sys_rt_sigsuspend patch.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Wire up the x86 syscalls
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I added this line to share this file with UML, but now it's no longer
shared so remove this useless leftover.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arjan van de Ven <arjan@infradead.org>
Mark a number of functions as 'must inline'. The functions affected by this
patch need to be inlined because they use knowledge that their arguments are
constant so that most of the function optimizes away. At this point this
patch does not change behavior, it's for documentation only (and for future
patches in the inline series)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
{get,put}_thread_info() were introduced in 2.5.4 and never
had been called by anything in the tree.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
)
From: Al Viro <viro@ftp.linux.org.uk>
task_pt_regs() needs the same offset-by-8 to match copy_thread()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
)
From: Ingo Molnar <mingo@elte.hu>
This is the latest version of the scheduler cache-hot-auto-tune patch.
The first problem was that detection time scaled with O(N^2), which is
unacceptable on larger SMP and NUMA systems. To solve this:
- I've added a 'domain distance' function, which is used to cache
measurement results. Each distance is only measured once. This means
that e.g. on NUMA distances of 0, 1 and 2 might be measured, on HT
distances 0 and 1, and on SMP distance 0 is measured. The code walks
the domain tree to determine the distance, so it automatically follows
whatever hierarchy an architecture sets up. This cuts down on the boot
time significantly and removes the O(N^2) limit. The only assumption
is that migration costs can be expressed as a function of domain
distance - this covers the overwhelming majority of existing systems,
and is a good guess even for more assymetric systems.
[ People hacking systems that have assymetries that break this
assumption (e.g. different CPU speeds) should experiment a bit with
the cpu_distance() function. Adding a ->migration_distance factor to
the domain structure would be one possible solution - but lets first
see the problem systems, if they exist at all. Lets not overdesign. ]
Another problem was that only a single cache-size was used for measuring
the cost of migration, and most architectures didnt set that variable
up. Furthermore, a single cache-size does not fit NUMA hierarchies with
L3 caches and does not fit HT setups, where different CPUs will often
have different 'effective cache sizes'. To solve this problem:
- Instead of relying on a single cache-size provided by the platform and
sticking to it, the code now auto-detects the 'effective migration
cost' between two measured CPUs, via iterating through a wide range of
cachesizes. The code searches for the maximum migration cost, which
occurs when the working set of the test-workload falls just below the
'effective cache size'. I.e. real-life optimized search is done for
the maximum migration cost, between two real CPUs.
This, amongst other things, has the positive effect hat if e.g. two
CPUs share a L2/L3 cache, a different (and accurate) migration cost
will be found than between two CPUs on the same system that dont share
any caches.
(The reliable measurement of migration costs is tricky - see the source
for details.)
Furthermore i've added various boot-time options to override/tune
migration behavior.
Firstly, there's a blanket override for autodetection:
migration_cost=1000,2000,3000
will override the depth 0/1/2 values with 1msec/2msec/3msec values.
Secondly, there's a global factor that can be used to increase (or
decrease) the autodetected values:
migration_factor=120
will increase the autodetected values by 20%. This option is useful to
tune things in a workload-dependent way - e.g. if a workload is
cache-insensitive then CPU utilization can be maximized by specifying
migration_factor=0.
I've tested the autodetection code quite extensively on x86, on 3
P3/Xeon/2MB, and the autodetected values look pretty good:
Dual Celeron (128K L2 cache):
---------------------
migration cost matrix (max_cache_size: 131072, cpu: 467 MHz):
---------------------
[00] [01]
[00]: - 1.7(1)
[01]: 1.7(1) -
---------------------
cacheflush times [2]: 0.0 (0) 1.7 (1784008)
---------------------
Here the slow memory subsystem dominates system performance, and even
though caches are small, the migration cost is 1.7 msecs.
Dual HT P4 (512K L2 cache):
---------------------
migration cost matrix (max_cache_size: 524288, cpu: 2379 MHz):
---------------------
[00] [01] [02] [03]
[00]: - 0.4(1) 0.0(0) 0.4(1)
[01]: 0.4(1) - 0.4(1) 0.0(0)
[02]: 0.0(0) 0.4(1) - 0.4(1)
[03]: 0.4(1) 0.0(0) 0.4(1) -
---------------------
cacheflush times [2]: 0.0 (33900) 0.4 (448514)
---------------------
Here it can be seen that there is no migration cost between two HT
siblings (CPU#0/2 and CPU#1/3 are separate physical CPUs). A fast memory
system makes inter-physical-CPU migration pretty cheap: 0.4 msecs.
8-way P3/Xeon [2MB L2 cache]:
---------------------
migration cost matrix (max_cache_size: 2097152, cpu: 700 MHz):
---------------------
[00] [01] [02] [03] [04] [05] [06] [07]
[00]: - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[01]: 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[02]: 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[03]: 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[04]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1)
[05]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1)
[06]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1)
[07]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) -
---------------------
cacheflush times [2]: 0.0 (0) 19.2 (19281756)
---------------------
This one has huge caches and a relatively slow memory subsystem - so the
migration cost is 19 msecs.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Cc: <wilder@us.ibm.com>
Signed-off-by: John Hawkes <hawkes@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add per-arch sched_cacheflush() which is a write-back cacheflush used by
the migration-cost calibration code at bootup time.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Passing random input values in eax to cpuid is not a good idea
because the CPU will GPF for unknown ones.
Use the correct x86-64 version that exists for a longer time too.
This also adds a memory barrier to prevent the optimizer from
reordering.
Cc: tigran@veritas.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As suggested by Linus.
This catches driver bugs that could cause corruption on IOMMU architectures.
Also I converted the BUGs to out_of_line_bug()s to save a bit
of text space.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Whenever we see that a CPU is capable of C3 (during ACPI cstate init), we
disable local APIC timer and switch to using a broadcast from external timer
interrupt (IRQ 0). This is needed because Intel CPUs stop the local
APIC timer in C3. This is currently only enabled for Intel CPUs.
Patch below adds the code for i386 and also the ACPI hunk.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Some people need it now on 64bit so reuse the i386 code for
x86-64. This will be also useful for future bug workarounds.
It is a bit simplified there because there is no need
to do it very early on x86-64. This means it doesn't need
early ioremap et.al. We run it as a core initcall right now.
I hope it's not needed for early setup.
I added a general CONFIG_DMI symbol in case IA64 or someone
else wants to reuse the code later too.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I noticed that some lowlevel send_IPI_mask helpers had a hotplug/preempt
race whereupon the cpu_online_map was read before disabling preemption;
...
cpumask_t mask = cpu_online_map;
int cpu = get_cpu();
cpu_clear(cpu, mask);
...
But then i realised that there is no need for these lowlevel functions to
be going through all this trouble when all the callers are already made
hotplug/preempt safe.
Signed-off-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Print bits for RDTSCP, SVM, CR8-LEGACY.
Also now print power flags on i386 like x86-64 always did.
This will add a new line in the 386 cpuinfo, but that shouldn't
be an issue - did that in the past too and I haven't heard
of any breakage.
I shrunk some of the fields in the i386 cpuinfo_x86 to chars
to make up for the new int "x86_power" field. Overall it's
smaller than before.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Define it for i386 too.
This is a synthetic flag that signifies that the CPU's TSC runs
at a constant P state invariant frequency.
Fix up the logic on x86-64/i386 to set it on all known CPUs.
Use the AMD defined bit to set it on future AMD CPUs.
Cc: venkatesh.pallipadi@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Noticed by Arjan originally on x86-64, then Ingo on x86, and finally me
grepping for it in the generic version.
Bad parenthesis nesting.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Call the mutex slowpath more conservatively - e.g. FRAME_POINTERS can
change the calling convention, in which case a direct branch to the
slowpath becomes illegal. Bug found by Hugh Dickins.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following patch (against 2.6.15-rc5-mm3) fixes a kprobes build break
due to changes introduced in the kprobe locking in 2.6.15-rc5-mm3. In
addition, the patch reverts back the open-coding of kprobe_mutex.
Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently arch_remove_kprobes() is only implemented/required for x86_64 and
powerpc. All other architecture like IA64, i386 and sparc64 implementes a
dummy function which is being called from arch independent kprobes.c file.
This patch removes the dummy functions and replaces it with
#define arch_remove_kprobe(p, s) do { } while(0)
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The arch specific kprobes.h files never gets included when CONFIG_KPROBES is
turned off. Hence check for CONFIG_KPROBES is not appropriate here in this
arch specific kprobes.h files.
Also the below defined function kprobes_exception_notify() is not needed when
CONFIG_KPROBES is off.
Compile tested for both CONFIG_KPROBES=y and N.
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Most arches copied the i386 ioctl.h. Combine them into a generic header.
Signed-off-by: Brian Gerst <bgerst@didntduck.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- If system panics then cpu register states are captured through funciton
crash_get_current_regs(). This is not a inline function hence a stack frame
is pushed on to the stack and then cpu register state is captured. Later
this frame is popped and new frames are pushed (machine_kexec).
- In theory this is not very right as we are capturing register states for a
frame and that frame is no more valid. This seems to have created back
trace problems for ppc64.
- This patch fixes it up. The very first thing it does after entering
crash_kexec() is to capture the register states. Anyway we don't want the
back trace beyond crash_kexec(). crash_get_current_regs() has been made
inline
- crash_setup_regs() is the top architecture dependent function which should
be responsible for capturing the register states as well as to do some
architecture dependent tricks. For ex. fixing up ss and esp for i386.
crash_setup_regs() has also been made inline to ensure no new call frame is
pushed onto stack.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- In case of system crash, current state of cpu registers is saved in memory
in elf note format. So far memory for storing elf notes was being allocated
statically for NR_CPUS.
- This patch introduces dynamic allocation of memory for storing elf notes.
It uses alloc_percpu() interface. This should lead to better memory usage.
- Introduced based on Andi Kleen's and Eric W. Biederman's suggestions.
- This patch also moves memory allocation for elf notes from architecture
dependent portion to architecture independent portion. Now crash_notes is
architecture independent. The whole idea is that size of memory to be
allocated per cpu (MAX_NOTE_BYTES) can be architecture dependent and
allocation of this memory can be architecture independent.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
