Remove old (and non-shared) VA addresses from the mappings
in arch/arm/mach-s3c2410/map.h and anywhere they are being
mapped in arch/arm/mach-s3c2410
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fix address-space conversion errors from passing addresses
generated from include/asm-arm/arch-s3c2410/map.h by adding
an __force argument to the `void __iomem *` for all the
virtual addresses.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fix address-space conversion errors from passing addresses
generated from include/asm-arm/arch-s3c2410/map.h by adding
an __force argument to the `void __iomem *` for all the
virtual addresses.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fix copyright notices in include/asm-arm/arch-s3c2410
to actually have `Copyright` in the line. This patch
deals with all the core files.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fix copyright notices in include/asm-arm/arch-s3c2410
to actually have `Copyright` in the line. This patch
deals with all the core files.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add:
sys_unshare
sys_set_robust_list
sys_get_robust_list
sys_splice
sys_arm_sync_file_range
sys_tee
sys_vmsplice
sys_move_pages
sys_getcpu
Special note about sys_arm_sync_file_range(), which is implemented as:
asmlinkage long sys_arm_sync_file_range(int fd, unsigned int flags,
loff_t offset, loff_t nbytes)
{
return sys_sync_file_range(fd, offset, nbytes, flags);
}
We can't export sys_sync_file_range() directly on ARM because the
argument list someone picked does not fit in the available registers.
Would be nice if... there was an arch maintainer review mechanism for
new syscalls before they hit the kernel.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
On architectures where the atomicity of the bit operations is handled by
external means (ie a separate spinlock to protect concurrent accesses),
just doing a direct assignment on the workqueue data field (as done by
commit 4594bf159f) can cause the
assignment to be lost due to lack of serialization with the bitops on
the same word.
So we need to serialize the assignment with the locks on those
architectures (notably older ARM chips, PA-RISC and sparc32).
So rather than using an "unsigned long", let's use "atomic_long_t",
which already has a safe assignment operation (atomic_long_set()) on
such architectures.
This requires that the atomic operations use the same atomicity locks as
the bit operations do, but that is largely the case anyway. Sparc32
will probably need fixing.
Architectures (including modern ARM with LL/SC) that implement sane
atomic operations for SMP won't see any of this matter.
Cc: Russell King <rmk+lkml@arm.linux.org.uk>
Cc: David Howells <dhowells@redhat.com>
Cc: David Miller <davem@davemloft.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Linux Arch Maintainers <linux-arch@vger.kernel.org>
Cc: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nobody uses it, but it was still wrong. Using the macro argument name
'work' meant that when we used 'work' as a member name, that would also
get replaced by the macro argument.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It has caused more problems than it ever really solved, and is
apparently not getting cleaned up and fixed. We can put it back when
it's stable and isn't likely to make warning or bug events worse.
In the meantime, enable frame pointers for more readable stack traces.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reboot hangs on LPARs without diag308 support. The reason for this is,
that before the reboot is done, the channel subsystem is shut down.
During the reset on each possible subchannel a "store subchannel" is
done. This operation can end in a program check interruption, if the
specified subchannel set is not implemented by the hardware. During
the reset, currently we do not have a program check handler, which
leads to the described kernel bug. We install now a new program check
handler for the reboot code to fix this problem.
Signed-off-by: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
A HiperSocket multicast queue works asynchronously. When sending
buffers, the buffer state change from PRIMED to EMPTY may happen
delayed. Reschedule the checking for changes in the outbound queue,
if there are still PRIMED buffers.
Signed-off-by: Ursula Braun <braunu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
At least on PPC, the "op ? op : dma" construct causes a compile failure
because the dma_* is a do{}while(0) macro.
This turns all of them into proper if/else to avoid this problem.
Signed-off-by: Ben Collins <bcollins@ubuntu.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the deferred hooks and all related code as scheduled in
feature-removal-schedule.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make fib6_node 'subtree' depend on IPV6_SUBTREES.
Signed-off-by: Kim Nordlund <kim.nordlund@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ivan Skytte Jorgensen <isj-sctp@i1.dk>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently in SCTP, we maintain a local address list by rebuilding the whole
list from the device list whenever we get a address add/delete event.
This patch fixes it by only adding/deleting the address for which we
receive the event.
Also removed the sctp_local_addr_lock() which is no longer needed as we
now use list_for_each_safe() to traverse this list. This fixes the bugs
in sctp_copy_laddrs_xxx() routines where we do copy_to_user() while
holding this lock.
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To do that, this makes nf_ct_l3proto_try_module_{get,put} compatible
functions. As a result we can remove '#ifdef' surrounds and direct call of
need_conntrack().
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
platform_device_add_data() makes a copy of the data that is given to it,
and thus the parameter can be const. This removes a warning when data
from get_property() on powerpc is handed to platform_device_add_data(),
as get_property() returns a const pointer.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
IA64 is in a tiny minority providing these defines in pci.h.
Almost everyone else has them in scatterlist.h
Signed-off-by: Tony Luck <tony.luck@intel.com>
Support to change MX1 CPU frequency at runtime.
Tested on PiKRON's PiMX1 board and seems to be fully
stable up to 200 MHz end even as low as 8 MHz.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
i.MX needs to tweak the control register to support CPU frequency
scaling. Rather than have folk blindly try and change the control
register by writing to it and then wondering why it doesn't work,
provide a method (which is safe for UP only, and therefore only
available for UP) to achieve this.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
optimization
The gcc manual says:
|`-fdelete-null-pointer-checks'
| Use global dataflow analysis to identify and eliminate useless
| checks for null pointers. The compiler assumes that dereferencing
| a null pointer would have halted the program. If a pointer is
| checked after it has already been dereferenced, it cannot be null.
| Enabled at levels `-O2', `-O3', `-Os'.
Now the problem can be seen with this test case:
#include <linux/prefetch.h>
extern void bar(char *x);
void foo(char *x)
{
prefetch(x);
if (x)
bar(x);
}
Because the constraint to the inline asm used in the prefetch() macro is
a memory operand, gcc assumes that the asm code does dereference the
pointer and the delete-null-pointer-checks optimization kicks in.
Inspection of generated assembly for the above example shows that bar()
is indeed called unconditionally without any test on the value of x.
Of course in the prefetch case there is no real dereference and it
cannot be assumed that a null pointer would have been caught at that
point. This causes kernel oopses with constructs like
hlist_for_each_entry() where the list's 'next' content is prefetched
before the pointer is tested against NULL, and only when gcc feels like
applying this optimization which doesn't happen all the time with more
complex code.
It appears that the way to prevent delete-null-pointer-checks
optimization to occur in this case is to make prefetch() into a static
inline function instead of a macro. At least this is what is done on
x86_64 where a similar inline asm memory operand is used (I presume they
would have seen the same problem if it didn't work) and resulting code
for the above example confirms that.
An alternative would consist of replacing the memory operand by a
register operand containing the pointer, and use the addressing mode
explicitly in the asm template. But that would be less optimal than an
offsettable memory reference.
Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Virtually index, physically tagged cache architectures can get away
without cache flushing when forking. This patch adds a new cache
flushing function flush_cache_dup_mm(struct mm_struct *) which for the
moment I've implemented to do the same thing on all architectures
except on MIPS where it's a no-op.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Provide a custom copy_user_highpage() to deal with aliasing issues on
MIPS. It uses kmap_coherent() to map an user page for kernel with same
color. Rewrite copy_to_user_page() and copy_from_user_page() with the
new interfaces to avoid extra cache flushing.
The main part of this patch was originally written by Ralf Baechle;
Atushi Nemoto did the the debugging.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
To allow a more effective copy_user_highpage() on certain architectures,
a vma argument is added to the function and cow_user_page() allowing
the implementation of these functions to check for the VM_EXEC bit.
The main part of this patch was originally written by Ralf Baechle;
Atushi Nemoto did the the debugging.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Problem:
1. There is a process containing two thread (T1 and T2). The
thread T1 calls fork(). Then dup_mmap() function called on T1 context.
static inline int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
...
flush_cache_mm(current->mm);
... /* A */
(write-protect all Copy-On-Write pages)
... /* B */
flush_tlb_mm(current->mm);
...
2. When preemption happens between A and B (or on SMP kernel), the
thread T2 can run and modify data on COW pages without page fault
(modified data will stay in cache).
3. Some time after fork() completed, the thread T2 may cause a page
fault by write-protect on a COW page.
4. Then data of the COW page will be copied to newly allocated
physical page (copy_cow_page()). It reads data via kernel mapping.
The kernel mapping can have different 'color' with user space
mapping of the thread T2 (dcache aliasing). Therefore
copy_cow_page() will copy stale data. Then the modified data in
cache will be lost.
In order to allow architecture code to deal with this problem allow
architecture code to override copy_user_highpage() by defining
__HAVE_ARCH_COPY_USER_HIGHPAGE in <asm/page.h>.
The main part of this patch was originally written by Ralf Baechle;
Atushi Nemoto did the the debugging.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Run this:
#!/bin/sh
for f in $(grep -Erl "\([^\)]*\) *k[cmz]alloc" *) ; do
echo "De-casting $f..."
perl -pi -e "s/ ?= ?\([^\)]*\) *(k[cmz]alloc) *\(/ = \1\(/" $f
done
And then go through and reinstate those cases where code is casting pointers
to non-pointers.
And then drop a few hunks which conflicted with outstanding work.
Cc: Russell King <rmk@arm.linux.org.uk>, Ian Molton <spyro@f2s.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Greg KH <greg@kroah.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Paul Fulghum <paulkf@microgate.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Karsten Keil <kkeil@suse.de>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Ian Kent <raven@themaw.net>
Cc: Steven French <sfrench@us.ibm.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Neil Brown <neilb@cse.unsw.edu.au>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Modify the sstfb (Voodoo1/2) driver:
- fix a memleak when removing the sstfb module
- fix sstfb to use the fbdev default videomode database
- add module option "mode_option" to set initial screen mode
- add sysfs-interface to turn VGA-passthrough on/off via
/sys/class/graphics/fbX/vgapass
- remove old debug functions from ioctl interface
Signed-off-by: Helge Deller <deller@gmx.de>
Acked-By: James Simmons <jsimmons@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Define an op descriptor struct, use it to simplify nfsd4_proc_compound().
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Tuck away the replay_owner in the cstate while we're at it.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Pass the saved and current filehandles together into all the nfsd4 compound
operations.
I want a unified interface to these operations so we can just call them by
pointer and throw out the huge switch statement.
Also I'll eventually want a structure like this--that holds the state used
during compound processing--for deferral.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A comment here incorrectly states that "slack_space" is measured in words, not
bytes. Remove the comment, and adjust a variable name and a few comments to
clarify the situation.
This is pure cleanup; there should be no change in functionality.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The BLK_DEV_SWIM_IOP driver has:
- already been marked as BROKEN in 2.6.0 three years ago and
- is still marked as BROKEN.
Drivers that had been marked as BROKEN for such a long time seem to be
unlikely to be revived in the forseeable future.
But if anyone wants to ever revive this driver, the code is still
present in the older kernel releases.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Jens Axboe <jens.axboe@oracle.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch converts the tracking of the user space watchdog process from using
a pid_t to use struct pid. This makes us safe from pid wrap around issues and
prepares the way for the pid namespace.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Petr Vandrovec <VANDROVE@vc.cvut.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
smbfs keeps track of the user space server process in conn_pid. This converts
that track to use a struct pid instead of pid_t. This keeps us safe from pid
wrap around issues and prepares the way for the pid namespace.
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>
Currently this driver tracks user space clients it should send signals to. In
the presenct of file descriptor passing this is appears susceptible to
confusion from pid wrap around issues.
Replacing this with a struct pid prevents us from getting confused, and
prepares for a pid namespace implementation.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Annotated, all places switched to keeping status net-endian.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
All kcalloc() calls of the form "kcalloc(1,...)" are converted to the
equivalent kzalloc() calls, and a few kcalloc() calls with the incorrect
ordering of the first two arguments are fixed.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Adam Belay <ambx1@neo.rr.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Greg KH <greg@kroah.com>
Cc: Mark Fasheh <mark.fasheh@oracle.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When we print an assert due to scheduling-in-atomic bugs, and if lockdep
is enabled, then the IRQ tracing information of lockdep can be printed
to pinpoint the code location that disabled interrupts. This saved me
quite a bit of debugging time in cases where the backtrace did not
identify the irq-disabling site well enough.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Most distributions enable sysrq support but set it to 0 by default. Add a
sysrq_always_enabled boot option to always-enable sysrq keys. Useful for
debugging - without having to modify the disribution's config files (which
might not be possible if the kernel is on a live CD, etc.).
Also, while at it, clean up the sysrq interfaces.
[bunk@stusta.de: make sysrq_always_enabled_setup() static]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Implement block device specific .direct_IO method instead of going through
generic direct_io_worker for block device.
direct_io_worker() is fairly complex because it needs to handle O_DIRECT on
file system, where it needs to perform block allocation, hole detection,
extents file on write, and tons of other corner cases. The end result is
that it takes tons of CPU time to submit an I/O.
For block device, the block allocation is much simpler and a tight triple
loop can be written to iterate each iovec and each page within the iovec in
order to construct/prepare bio structure and then subsequently submit it to
the block layer. This significantly speeds up O_D on block device.
[akpm@osdl.org: small speedup]
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Zach Brown <zach.brown@oracle.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add "relatime" (relative atime) support. Relative atime only updates the
atime if the previous atime is older than the mtime or ctime. Like
noatime, but useful for applications like mutt that need to know when a
file has been read since it was last modified.
A corresponding patch against mount(8) is available at
http://userweb.kernel.org/~akpm/mount-relative-atime.txt
Signed-off-by: Valerie Henson <val_henson@linux.intel.com>
Cc: Mark Fasheh <mark.fasheh@oracle.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Karel Zak <kzak@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The kernel termios (ktermios) changes were somehow missed for Xtensa. This
patch adds the ktermios structure and also includes some minor file name
fix that was missed in the syscall patch.
Signed-off-by: Chris Zankel <chris@zankel.net>
Acked-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently, to tell a task that it should go to the refrigerator, we set the
PF_FREEZE flag for it and send a fake signal to it. Unfortunately there
are two SMP-related problems with this approach. First, a task running on
another CPU may be updating its flags while the freezer attempts to set
PF_FREEZE for it and this may leave the task's flags in an inconsistent
state. Second, there is a potential race between freeze_process() and
refrigerator() in which freeze_process() running on one CPU is reading a
task's PF_FREEZE flag while refrigerator() running on another CPU has just
set PF_FROZEN for the same task and attempts to reset PF_FREEZE for it. If
the refrigerator wins the race, freeze_process() will state that PF_FREEZE
hasn't been set for the task and will set it unnecessarily, so the task
will go to the refrigerator once again after it's been thawed.
To solve first of these problems we need to stop using PF_FREEZE to tell
tasks that they should go to the refrigerator. Instead, we can introduce a
special TIF_*** flag and use it for this purpose, since it is allowed to
change the other tasks' TIF_*** flags and there are special calls for it.
To avoid the freeze_process()-refrigerator() race we can make
freeze_process() to always check the task's PF_FROZEN flag after it's read
its "freeze" flag. We should also make sure that refrigerator() will
always reset the task's "freeze" flag after it's set PF_FROZEN for it.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: David Howells <dhowells@redhat.com>
Cc: Andi Kleen <ak@muc.de>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When some objects are allocated by one CPU but freed by another CPU we can
consume lot of cycles doing divides in obj_to_index().
(Typical load on a dual processor machine where network interrupts are
handled by one particular CPU (allocating skbufs), and the other CPU is
running the application (consuming and freeing skbufs))
Here on one production server (dual-core AMD Opteron 285), I noticed this
divide took 1.20 % of CPU_CLK_UNHALTED events in kernel. But Opteron are
quite modern cpus and the divide is much more expensive on oldest
architectures :
On a 200 MHz sparcv9 machine, the division takes 64 cycles instead of 1
cycle for a multiply.
Doing some math, we can use a reciprocal multiplication instead of a divide.
If we want to compute V = (A / B) (A and B being u32 quantities)
we can instead use :
V = ((u64)A * RECIPROCAL(B)) >> 32 ;
where RECIPROCAL(B) is precalculated to ((1LL << 32) + (B - 1)) / B
Note :
I wrote pure C code for clarity. gcc output for i386 is not optimal but
acceptable :
mull 0x14(%ebx)
mov %edx,%eax // part of the >> 32
xor %edx,%edx // useless
mov %eax,(%esp) // could be avoided
mov %edx,0x4(%esp) // useless
mov (%esp),%ebx
[akpm@osdl.org: small cleanups]
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Elaborate the API for calling cpuset_zone_allowed(), so that users have to
explicitly choose between the two variants:
cpuset_zone_allowed_hardwall()
cpuset_zone_allowed_softwall()
Until now, whether or not you got the hardwall flavor depended solely on
whether or not you or'd in the __GFP_HARDWALL gfp flag to the gfp_mask
argument.
If you didn't specify __GFP_HARDWALL, you implicitly got the softwall
version.
Unfortunately, this meant that users would end up with the softwall version
without thinking about it. Since only the softwall version might sleep,
this led to bugs with possible sleeping in interrupt context on more than
one occassion.
The hardwall version requires that the current tasks mems_allowed allows
the node of the specified zone (or that you're in interrupt or that
__GFP_THISNODE is set or that you're on a one cpuset system.)
The softwall version, depending on the gfp_mask, might allow a node if it
was allowed in the nearest enclusing cpuset marked mem_exclusive (which
requires taking the cpuset lock 'callback_mutex' to evaluate.)
This patch removes the cpuset_zone_allowed() call, and forces the caller to
explicitly choose between the hardwall and the softwall case.
If the caller wants the gfp_mask to determine this choice, they should (1)
be sure they can sleep or that __GFP_HARDWALL is set, and (2) invoke the
cpuset_zone_allowed_softwall() routine.
This adds another 100 or 200 bytes to the kernel text space, due to the few
lines of nearly duplicate code at the top of both cpuset_zone_allowed_*
routines. It should save a few instructions executed for the calls that
turned into calls of cpuset_zone_allowed_hardwall, thanks to not having to
set (before the call) then check (within the call) the __GFP_HARDWALL flag.
For the most critical call, from get_page_from_freelist(), the same
instructions are executed as before -- the old cpuset_zone_allowed()
routine it used to call is the same code as the
cpuset_zone_allowed_softwall() routine that it calls now.
Not a perfect win, but seems worth it, to reduce this chance of hitting a
sleeping with irq off complaint again.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
More cleanups for slab.h
1. Remove tabs from weird locations as suggested by Pekka
2. Drop the check for NUMA and SLAB_DEBUG from the fallback section
as suggested by Pekka.
3. Uses static inline for the fallback defs as also suggested by Pekka.
4. Make kmem_ptr_valid take a const * argument.
5. Separate the NUMA fallback definitions from the kmalloc_track fallback
definitions.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ben Dooks