- 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>
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>
Bitmap functions for the minix filesystem and the ext2 filesystem except
ext2_set_bit_atomic() and ext2_clear_bit_atomic() do not require the atomic
guarantees.
But these are defined by using atomic bit operations on several architectures.
(cris, frv, h8300, ia64, m32r, m68k, m68knommu, mips, s390, sh, sh64, sparc,
sparc64, v850, and xtensa)
This patch switches to non atomic bit operation.
Signed-off-by: Akinobu Mita <mita@miraclelinux.com>
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>
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>
Provide abstraction for generating type and size information of assembly
routines and data, while permitting architectures to override these
defaults.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: "Russell King" <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: "Andi Kleen" <ak@suse.de>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp>
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>
This option is no longer usable with supported compilers. It will be
replaced by usage of -msym32 in a separate patch.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Move function prototypes to asm/signal.h to detect trivial errors and
add some __user tags to get rid of sparse warnings. Generated code
should not be changed.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
1. Move ioswab*() and __mem_ioswab*() to mangle-port.h. This gets rid
of CONFIG_SGI_IP22 from include/asm-mips/io.h.
2. Pass a virtual address to *ioswab*(). Then we can provide
mach-specific *ioswab*() and can do every evil thing based on its
argument. It could be useful on machines which have regions with
different endian conversion scheme.
3. Call __swizzle_addr*() _after_ adding mips_io_port_base. This
unifies the meaning of the argument of __swizzle_addr*() (always
virtual address). Then mach-specific __swizzle_addr*() can do every
evil thing based on the argument.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
ALCHEMY: Add EHCI support for AU1200
Updated by removing the OHCI support
Signed-off-by: Jordan Crouse <jordan.crouse@amd.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
From Dave Johnson <djohnson+linuxmips@sw.starentnetworks.com>:
sb1250_gettimeoffset() simply reads the current cpu 0 timer remaining
value, however once this counter reaches 0 and the interrupt is raised,
it immediately resets and begins to count down again.
If sb1250_gettimeoffset() is called on cpu 1 via do_gettimeofday() after
the timer has reset but prior to cpu 0 processing the interrupt and
taking write_seqlock() in timer_interrupt() it will return a full value
(or close to it) causing time to jump backwards 1ms. Once cpu 0 handles
the interrupt and timer_interrupt() gets far enough along it will jump
forward 1ms.
Fix this problem by implementing mips_hpt_*() on sb1250 using a spare
timer unrelated to the existing periodic interrupt timers. It runs at
1Mhz with a full 23bit counter. This eliminated the custom
do_gettimeoffset() for sb1250 and allowed use of the generic
fixed_rate_gettimeoffset() using mips_hpt_*() and timerhi/timerlo.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
From Dave Johnson <djohnson+linuxmips@sw.starentnetworks.com>:
Field width should be 23 bits not 20 bits.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
If a call to set_io_port_base() was being followed by usage of
mips_io_port_base in the same function gcc was possibly using the old
value due to some clever abuse of const. Adding a barrier will keep
the optimization and result in correct code with latest gcc.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
If dcache_size != icache_size or dcache_size != scache_size, or
set-associative cache, icache/scache does not flushed properly. Make
blast_?cache_page_indexed() masks its index value correctly. Also,
use physical address for physically indexed pcache/scache.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The SB1 core has a three cycle interrupt disable hazard but we were
wrongly treating it as fully interlocked.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
This is gluing the build of cirrusfb but really the mess that would need
cleaning and fixing is <video/vga.h> and <linux/vt_buffer.h> ...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
A recent change requires cpu_possible_map to be initialized before
smp_sched_init() but most MIPS platforms were initializing their
processors in the prom_prepare_cpus callback of smp_prepare_cpus. The
simple fix of calling prom_prepare_cpus from one of the earlier SMP
initialization hooks doesn't work well either since IPIs may require
init_IRQ() to have completed, so bit the bullet and split
prom_prepare_cpus into two initialization functions, plat_smp_setup
which is called early from setup_arch and plat_prepare_cpus called where
prom_prepare_cpus used to be called.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
It seems current get_user() incorrectly sign-extend an unsigned int
value on 64bit kernel. I think this is because '(__typeof__(val))'
cast in final assignment. I suppose the cast should be
'(__typeof__(*(addr))'.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.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>
From Richard Sandiford <richard@codesourcery.com>:
This patch caused a miscompilation of the restore_gp_regs() block
in restore_sigcontext(). This was in a 32-bit kernel compiled with
GCC CVS head.
restore_gp_regs() copies 64-bit user fields into 32-bit variables,
and in this combination, the new __get_user_asm_ll32() clobbers too
many registers. It says:
/*
* Get a long long 64 using 32 bit registers.
*/
{ \
__asm__ __volatile__( \
"1: lw %1, (%3) \n" \
"2: lw %D1, 4(%3) \n" \
" move %0, $0 \n" \
"3: .section .fixup,\"ax\" \n" \
"4: li %0, %4 \n" \
" move %1, $0 \n" \
" move %D1, $0 \n" \
" j 3b \n" \
" .previous \n" \
" .section __ex_table,\"a\" \n" \
" " __UA_ADDR " 1b, 4b \n" \
" " __UA_ADDR " 2b, 4b \n" \
" .previous \n" \
: "=r" (__gu_err), "=&r" (val) \
: "0" (0), "r" (addr), "i" (-EFAULT)); \
}
and this requires val (%1) to be a 64-bit value. In the case I saw,
gcc was using $3 for the 32-bit val, and wasn't expecting $4 to be
clobbered.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Add blast_xxx_range(), protected_blast_xxx_range() etc. for common
use. They are built by __BUILD_BLAST_CACHE_RANGE().
Use protected_cache_op() macro for various protected_ routines.
Output code should be logically same.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
A pile of internal functions use only inside mips io.h has names starting
with mem_... and clashing with names in drivers; renamed to __mem_....
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
