Nick Piggin [Tue, 3 Oct 2006 08:14:04 +0000 (01:14 -0700)]
[PATCH] sched: force /sbin/init off isolated cpus
Force /sbin/init off isolated cpus (unless every CPU is specified as an
isolcpu).
Users seem to think that the isolated CPUs shouldn't have much running on
them to begin with. That's fair enough: intuitive, I guess. It also means
that the cpu affinity masks of tasks will not include isolcpus by default,
which is also more intuitive, perhaps.
/sbin/init is spawned from the boot CPU's idle thread, and /sbin/init
starts the rest of userspace. So if the boot CPU is specified to be an
isolcpu, then prior to this patch, all of userspace will be run there.
(throw in a couple of plausible devinit -> cpuinit conversions I spotted
while we're here).
Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Dimitri Sivanich <sivanich@sgi.com> Acked-by: Paul Jackson <pj@sgi.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Bjorn Helgaas [Tue, 3 Oct 2006 08:14:00 +0000 (01:14 -0700)]
[PATCH] ia64: note requirement for 8250_pnp, now that 8250_acpi is gone
We removed 8250_acpi in 2.6.17. If we don't have PNPACPI turned on, we
won't find any ACPI serial devices, so mention this requirement in the
troubleshooting part of the documentation.
CONFIG_PNPACPI is already turned on in all the relevant defconfigs.
Corey Minyard [Tue, 3 Oct 2006 08:13:59 +0000 (01:13 -0700)]
[PATCH] IPMI: allow user to override the kernel IPMI daemon enable
After the previous patch to disable the kernel IPMI daemon if interrupts
were available, the issue of broken hardware was raised, and a reasonable
request to add an override was mode. So here it is.
Allow the user to force the kernel ipmi daemon on or off. This way,
hardware with broken interrupts or users that are not concerned with
performance can turn it on or off to their liking.
[akpm@osdl.org: save 4 bytes in vmlinux] Signed-off-by: Corey Minyard <minyard@acm.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Olaf Hering [Tue, 3 Oct 2006 08:13:56 +0000 (01:13 -0700)]
[PATCH] restore parport_pc probing on powermac
The last change for partport_pc did fix the common case for all PowerMacs,
but it broke the case for PCI multiport IO cards. In fact, the config
option CONFIG_PARPORT_PC_SUPERIO=y lead to a hard crash when cups probed
the parport driver. It enables the winbond and smsc probing.
Remove the PARPORT_BASE check again, parport_pc_find_nonpci_ports() will
take care of it. All powerpc configs should have
CONFIG_PARPORT_PC_SUPERIO=n, the code did not find anything on the chrp
boards we tested it on.
Tested on a G4/466 with a PCI card:
0001:10:13.0 Serial controller: Timedia Technology Co Ltd PCI2S550 (Dual 16550 UART) (rev 01) (prog-if 02 [16550])
Subsystem: Timedia Technology Co Ltd Unknown device 5079
Control: I/O+ Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping+ SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
Interrupt: pin A routed to IRQ 53
Region 0: I/O ports at f2000800 [size=32]
Region 2: I/O ports at f2000870 [size=8]
Region 3: I/O ports at f2000860 [size=8]
Signed-off-by: Olaf Hering <olaf@aepfle.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Adam Belay <ambx1@neo.rr.com> Cc: Dmitry Torokhov <dtor@mail.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Alan Cox [Tue, 3 Oct 2006 08:13:55 +0000 (01:13 -0700)]
[PATCH] ISDN warning fixes
Clean up warnings in drivers/isdn by using long not int for the values
where we pass void * and cast to integer types. The code is ok (ok passing
the stuff this way isn't pretty but the code is valid). In all the cases I
checked out the right thing happens anyway but this removes all the
warnings.
Signed-off-by: Alan Cox <alan@redhat.com> Cc: Jeff Garzik <jeff@garzik.org> Acked-by: Karsten Keil <kkeil@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jeff Garzik [Tue, 3 Oct 2006 08:13:53 +0000 (01:13 -0700)]
[PATCH] drivers/char/ip2: kill unused code, label
Kill warning:
drivers/char/ip2/ip2main.c: In function â\80\98ip2_loadmainâ\80\99:
drivers/char/ip2/ip2main.c:782: warning: label â\80\98out_classâ\80\99 defined but not used
This driver's initialization (and cleanup of errors during init) is
extremely convoluted, and could stand to be transformed into the standard
unwinding-goto style of error cleanup.
Signed-off-by: Jeff Garzik <jeff@garzik.org> Acked-by: Michael H. Warfield <mhw@wittsend.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jeff Garzik [Tue, 3 Oct 2006 08:13:52 +0000 (01:13 -0700)]
[PATCH] ipmi: fix uninitialized data bug
gcc issues the following warning:
drivers/char/ipmi/ipmi_si_intf.c: In function â\80\98init_ipmi_siâ\80\99:
drivers/char/ipmi/ipmi_si_intf.c:1729: warning: â\80\98data.irqâ\80\99 may be used uninitialized in this function
This is indeed a bug. data.irq is completely uninitialized in some code
paths. Worse than that, data from a previous decode_dmi() run can easily
leak through successive calls.
Signed-off-by: Jeff Garzik <jeff@garzik.org> Acked-by: Corey Minyard <minyard@acm.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
While reviewing the 'may be used uninitialized' bogus gcc warnings, I
noticed that an error code assignment was only needed if an error had
actually occured.
Signed-off-by: Jeff Garzik <jeff@garzik.org> Cc: Davide Libenzi <davidel@xmailserver.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
keios [Tue, 3 Oct 2006 08:13:49 +0000 (01:13 -0700)]
[PATCH] low performance of lib/sort.c
It is a non-standard heap-sort algorithm implementation because the index
of child node is wrong . The sort function still outputs right result, but
the performance is O( n * ( log(n) + 1 ) ) , about 10% ~ 20% worse than
standard algorithm.
Signed-off-by: keios <keios.cn@gmail.com> Acked-by: Matt Mackall <mpm@selenic.com> Acked-by: Zou Nan hai <nanhai.zou@intel.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Franck Bui-Huu [Tue, 3 Oct 2006 08:13:48 +0000 (01:13 -0700)]
[PATCH] Create kallsyms_lookup_size_offset()
Some uses of kallsyms_lookup() do not need to find out the name of a symbol
and its module's name it belongs. This is specially true in arch specific
code, which needs to unwind the stack to show the back trace during oops
(mips is an example). In this specific case, we just need to retreive the
function's size and the offset of the active intruction inside it.
Adds a new entry "kallsyms_lookup_size_offset()" This new entry does
exactly the same as kallsyms_lookup() but does not require any buffers to
store any names.
It returns 0 if it fails otherwise 1.
Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Howells [Tue, 3 Oct 2006 08:13:46 +0000 (01:13 -0700)]
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers
These patches make the kernel pass 64-bit inode numbers internally when
communicating to userspace, even on a 32-bit system. They are required
because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS
for example. The 64-bit inode numbers are then propagated to userspace
automatically where the arch supports it.
Problems have been seen with userspace (eg: ld.so) using the 64-bit inode
number returned by stat64() or getdents64() to differentiate files, and
failing because the 64-bit inode number space was compressed to 32-bits, and
so overlaps occur.
This patch:
Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit
inode number so that 64-bit inode numbers can be passed back to userspace.
The stat functions then returns the full 64-bit inode number where
available and where possible. If it is not possible to represent the inode
number supplied by the filesystem in the field provided by userspace, then
error EOVERFLOW will be issued.
Similarly, the getdents/readdir functions now pass the full 64-bit inode
number to userspace where possible, returning EOVERFLOW instead when a
directory entry is encountered that can't be properly represented.
Note that this means that some inodes will not be stat'able on a 32-bit
system with old libraries where they were before - but it does mean that
there will be no ambiguity over what a 32-bit inode number refers to.
Note similarly that directory scans may be cut short with an error on a
32-bit system with old libraries where the scan would work before for the
same reasons.
It is judged unlikely that this situation will occur because modern glibc
uses 64-bit capable versions of stat and getdents class functions
exclusively, and that older systems are unlikely to encounter
unrepresentable inode numbers anyway.
[akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jeff Garzik [Tue, 3 Oct 2006 01:08:22 +0000 (21:08 -0400)]
[PATCH] hp100: fix conditional compilation mess
The previous hp100 changeset attempted to kill warnings, but was only
tested on !CONFIG_ISA platforms. The correct conditional compilation
setup involves tested CONFIG_ISA rather than just MODULE.
Fixes link on CONFIG_ISA platforms (i386) in current -git.
Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Move the call to ib_register_device() later, since a device should not
be registered until it is completely read to be used. This fixes
crashes that occur if an upper-layer driver such as IPoIB is loaded
before the ehca module.
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Reorganize code relating to cma_get_net_info() and rdam_create_id() to
optimize error case handling (no need to alloc memory/etc. as part of
rdma_create_id() if input parameters are wrong).
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
A process : cma_process_remove() calls cma_remove_id_dev(),
which sets id state to CMA_DEVICE_REMOVAL and
calls wait_event(dev_remove).
B process : cma_req_handler() had incremented dev_remove,
and calls cma_acquire_ib_dev() and on failure
calls cma_release_remove(), which does a
wake_up of cma_process_remove(). Then
cma_req_handler() calls rdma_destroy_id();
A Process : cma_remove_id_dev() gets woken and checks the
state of id, and since it is still (wrongly)
CMA_DEVICE_REMOVAL, it calls notify_user(id)
and if that fails, the caller - cma_process_remove()
calls rdma_destroy_id(id). Two processes can
call rdma_destroy_id(), resulting in one
de-referencing kfreed id_priv.
Fix is for process B to set CMA_DESTROYING in cma_req_handler()
so that process A will return instead of doing a rdma_destroy_id().
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com> Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Samuel Tardieu [Sat, 9 Sep 2006 15:34:31 +0000 (17:34 +0200)]
[WATCHDOG] use ENOTTY instead of ENOIOCTLCMD in ioctl()
Return ENOTTY instead of ENOIOCTLCMD in user-visible ioctl() results
The watchdog drivers used to return ENOIOCTLCMD for bad ioctl() commands.
ENOIOCTLCMD should not be visible by the user, so use ENOTTY instead.
Signed-off-by: Samuel Tardieu <sam@rfc1149.net> Signed-off-by: Wim Van Sebroeck <wim@iguana.be> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org>
Ben Dooks [Wed, 6 Sep 2006 11:24:35 +0000 (12:24 +0100)]
[WATCHDOG] s3c24XX nowayout
If the driver is not configured for `no way out`,
then the open method should not automatically allow
the setting of allow_close to CLOSE_STATE_ALLOW.
The setting of allow_close nullifies the use of
the magic close via the write path. It means that
in the default state, the watchdog will shut-down
even if the magic close has not been issued.
Add io spinlocks to prevent possible race
conditions between start and stop operations
that are issued from different child processes
where the master process opened /dev/watchdog.
Linus Torvalds [Mon, 2 Oct 2006 21:05:20 +0000 (14:05 -0700)]
Add prototype for sigset_from_compat()
Duh. I screwed up editing David Howells patch in commit 3f2e05e90e0846c42626e3d272454f26be34a1bc, and the actual declaration for
the sigset_from_compat() function went missing. My bad.
Olaf Hering saved the day and noticed that I'm a moron.
Linus Torvalds [Mon, 2 Oct 2006 15:56:58 +0000 (08:56 -0700)]
Merge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6
* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: (37 commits)
[netdrvr] hp100: encapsulate all non-module code
drivers/net/wireless/{airo,ipw2100}: fix error handling bugs
[netdrvr] phy: Fix bugs in error handling
[PATCH] spidernet: Use pci_dma_mapping_error()
[PATCH] sky2: version 1.9
[PATCH] sky2: fragmented receive for large MTU
[PATCH] sky2: use netif_tx_lock instead of LLTX
[PATCH] sky2: incremental transmit completion
[PATCH] sky2: name irq after eth for irqbalance
[PATCH] sky2: workarounds for some 88e806x chips
[PATCH] sky2: use standard pci register capabilties for error register
[PATCH] sky2: gigabit full duplex negotiation
e100, e1000, ixgb: increment version numbers
ixgb: convert to netdev_priv(netdev)
ixgb: combine more rx descriptors to improve performance
e1000: possible memory leak in e1000_set_ringparam
e1000: Janitor: Use #defined values for literals
e1000: don't strip vlan ID if 8021q claims it
e1000: rework polarity, NVM, eeprom code and fixes.
e1000: driver state fixes (race fix)
...
Linus Torvalds [Mon, 2 Oct 2006 15:27:09 +0000 (08:27 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy/jfs-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy/jfs-2.6:
JFS: White space cleanup
[PATCH] JFS: return correct error when i-node allocation failed
JFS: Remove shadow variable from fs/jfs/jfs_txnmgr.c:xtLog()
Linus Torvalds [Mon, 2 Oct 2006 15:22:17 +0000 (08:22 -0700)]
Merge git://git.infradead.org/mtd-2.6
* git://git.infradead.org/mtd-2.6:
[MTD] Cleanup of 'ioremap balanced with iounmap for drivers/mtd subsystem'
[MTD] fix nftl_write warning
[MTD] fix printk warning
[MTD ONENAND] Check OneNAND lock scheme & all block unlock command support
[MTD ONENAND] Remove unused MTD_ONENAND_SYNC_READ configuration
[MTD ONENAND] Fix OneNAND probe
[MTD NAND] Provide prototype for newly-exported nand_wait_ready()
[MTD] Remove #ifndef __KERNEL__ hack in <mtd/mtd-abi.h>
[MTD NAND] Allow override of page read and write functions.
[MTD NAND] Allocate chip->buffers separately to allow it to be overridden
[MTD NAND] Split nand_scan() into two parts; allow board driver to intervene
[MTD NAND] Export nand_wait_ready() for use by board drivers
Linus Torvalds [Mon, 2 Oct 2006 15:20:33 +0000 (08:20 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input: (35 commits)
Input: wistron - add support for Acer TravelMate 2424NWXCi
Input: wistron - fix setting up special buttons
Input: add KEY_BLUETOOTH and KEY_WLAN definitions
Input: add new BUS_VIRTUAL bus type
Input: add driver for stowaway serial keyboards
Input: make input_register_handler() return error codes
Input: remove cruft that was needed for transition to sysfs
Input: fix input module refcounting
Input: constify input core
Input: libps2 - rearrange exports
Input: atkbd - support Microsoft Natural Elite Pro keyboards
Input: i8042 - disable MUX mode on Toshiba Equium A110
Input: i8042 - get rid of polling timer
Input: send key up events at disconnect
Input: constify psmouse driver
Input: i8042 - add Amoi to the MUX blacklist
Input: logips2pp - add sugnature 56 (Cordless MouseMan Wheel), cleanup
Input: add driver for Touchwin serial touchscreens
Input: add driver for Touchright serial touchscreens
Input: add driver for Penmount serial touchscreens
...
David Howells [Mon, 2 Oct 2006 13:12:31 +0000 (14:12 +0100)]
[PATCH] BLOCK: Revert patch to hack around undeclared sigset_t in linux/compat.h
Revert Andrew Morton's patch to temporarily hack around the lack of a
declaration of sigset_t in linux/compat.h to make the block-disablement
patches build on IA64. This got accidentally pushed to Linus and should
be fixed in a different manner.
Also make linux/compat.h #include asm/signal.h to gain a definition of
sigset_t so that it can externally declare sigset_from_compat().
This has been compile-tested for i386, x86_64, ia64, mips, mips64, frv, ppc and
ppc64 and run-tested on frv.
Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There are a few places in the kernel where the init task is signaled. The
ctrl+alt+del sequence is one them. It kills a task, usually init, using a
cached pid (cad_pid).
This patch replaces the pid_t by a struct pid to avoid pid wrap around
problem. The struct pid is initialized at boot time in init() and can be
modified through systctl with
/proc/sys/kernel/cad_pid
[ I haven't found any distro using it ? ]
It also introduces a small helper routine kill_cad_pid() which is used
where it seemed ok to use cad_pid instead of pid 1.
[akpm@osdl.org: cleanups, build fix] Signed-off-by: Cedric Le Goater <clg@fr.ibm.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Oleg Nesterov [Mon, 2 Oct 2006 09:18:53 +0000 (02:18 -0700)]
[PATCH] proc: convert do_task_stat() to use lock_task_sighand()
Drop tasklist_lock. ->siglock protects almost all interesting data
(including sub-threads traversal) except:
->signal->tty
protected by tty_mutex
->real_parent
the task can't be unhashed while we are holding
->siglock, so ->real_parent can change from under us
but we can safely dereference it under rcu_read_lock()
->pgrp/->session
we can get inconsistent numbers if the task does
sys_setsid/daemonize at the same time. I hope this
is acceptable.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] proc: Use pid_task instead of open coding it
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>
[PATCH] proc: Merge proc_tid_attr and proc_tgid_attr
The implementation is exactly the same and there is currently nothing to
distinguish proc_tid_attr, and proc_tgid_attr. So it is pointless to have two
separate implementations.
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>
The hard coded inode numbers in proc currently limit its maintainability,
its flexibility, and what can be done with the rest of system. /proc limits
pid-max to 32768 on 32 bit systems it limits fd-max to 32768 on all systems,
and placing the pid in the inode number really gets in the way of implementing
subdirectories of per process information.
Ever since people started adding to the middle of the file type enumeration we
haven't been maintaing the historical inode numbers, all we have really
succeeded in doing is keeping the pid in the proc inode number. The pid is
already available in the directory name so no information is lost removing it
from the inode number.
So if something in user space cares if we remove the inode number from the
/proc inode it is almost certainly broken.
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>
[PATCH] proc: Factor out an instantiate method from every lookup method
To remove the hard coded proc inode numbers it is necessary to be able to
create the proc inodes during readdir. The instantiate methods are the subset
of lookup that is needed to accomplish that.
This first step just splits the lookup methods into 2 functions.
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>
[PATCH] proc: Make the generation of the self symlink table driven
This patch generalizes the concept of files in /proc that are related to
processes but live in the root directory of /proc
Ideally this would reuse infrastructure from the rest of the process specific
parts of proc but unfortunately security_task_to_inode must not be called on
files that are not strictly per process. security_task_to_inode really needs
to be reexamined as the security label can change in important places that we
are not currently catching, but I'm not certain that simplifies this problem.
By at least matching the structure of the rest of proc we get more idiom reuse
and it becomes easier to spot problems in the way things are put together.
Later things like /proc/mounts are likely to be moved into proc_base as well.
If union mounts are ever supported we may be able to make /proc a union mount,
and properly split it into 2 filesystems.
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>
Move execve() into arch/avr32/kernel/sys_avr32.c, rename it to
kernel_execve() and return the syscall return value directly without
setting errno.
This also gets rid of the __KERNEL_SYSCALLS__ stuff from unistd.h and
expands #ifdef __KERNEL__ to cover everything in unistd.h except the
__NR_foo definitions.
Arnd Bergmann [Mon, 2 Oct 2006 09:18:44 +0000 (02:18 -0700)]
[PATCH] remove remaining errno and __KERNEL_SYSCALLS__ references
The last in-kernel user of errno is gone, so we should remove the definition
and everything referring to it. This also removes the now-unused lib/execve.c
file that was introduced earlier.
Also remove every trace of __KERNEL_SYSCALLS__ that still remained in the
kernel.
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arnd Bergmann [Mon, 2 Oct 2006 09:18:41 +0000 (02:18 -0700)]
[PATCH] sh64: remove the use of kernel syscalls
sh64 is using system call macros to call some functions from the kernel.
The old debug code can simply be removed, since we don't really have that much
of a need for it anymore, it was mostly something that was handy during the
initial bringup. This also brings us closer to something that looks like
readable code again..
I also added a sane kernel_thread() implementation that gets away from this,
so that should take care of sh64 at least.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arnd Bergmann [Mon, 2 Oct 2006 09:18:37 +0000 (02:18 -0700)]
[PATCH] Remove the use of _syscallX macros in UML
User mode linux uses _syscallX() to call into the host kernel. The
recommended way to do this is to use the syscall() function from libc.
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arnd Bergmann [Mon, 2 Oct 2006 09:18:34 +0000 (02:18 -0700)]
[PATCH] provide kernel_execve on all architectures
This adds the new kernel_execve function on all architectures that were using
_syscall3() to implement execve.
The implementation uses code from the _syscall3 macros provided in the
unistd.h header file. I don't have cross-compilers for any of these
architectures, so the patch is untested with the exception of i386.
Most architectures can probably implement this in a nicer way in assembly or
by combining it with the sys_execve implementation itself, but this should do
it for now.
[bunk@stusta.de: m68knommu build fix]
[markh@osdl.org: build fix]
[bero@arklinux.org: build fix]
[ralf@linux-mips.org: mips fix]
[schwidefsky@de.ibm.com: s390 fix] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: Bernhard Rosenkraenzer <bero@arklinux.org> Signed-off-by: Mark Haverkamp <markh@osdl.org> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arnd Bergmann [Mon, 2 Oct 2006 09:18:31 +0000 (02:18 -0700)]
[PATCH] rename the provided execve functions to kernel_execve
Some architectures provide an execve function that does not set errno, but
instead returns the result code directly. Rename these to kernel_execve to
get the right semantics there. Moreover, there is no reasone for any of these
architectures to still provide __KERNEL_SYSCALLS__ or _syscallN macros, so
remove these right away.
[akpm@osdl.org: build fix]
[bunk@stusta.de: build fix] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Acked-by: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Adrian Bunk <bunk@stusta.de> Cc: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Arnd Bergmann [Mon, 2 Oct 2006 09:18:26 +0000 (02:18 -0700)]
[PATCH] introduce kernel_execve
The use of execve() in the kernel is dubious, since it relies on the
__KERNEL_SYSCALLS__ mechanism that stores the result in a global errno
variable. As a first step of getting rid of this, change all users to a
global kernel_execve function that returns a proper error code.
This function is a terrible hack, and a later patch removes it again after the
kernel syscalls are gone.
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Matt Helsley [Mon, 2 Oct 2006 09:18:25 +0000 (02:18 -0700)]
[PATCH] ipc: replace kmalloc and memset in get_undo_list with kzalloc
Simplify get_undo_list() by dropping the unnecessary cast, removing the
size variable, and switching to kzalloc() instead of a kmalloc() followed
by a memset().
This cleanup was split then modified from Jes Sorenson's Task Notifiers
patches.
Signed-off-by: Matt Helsley <matthltc@us.ibm.com> Cc: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Pavel [Mon, 2 Oct 2006 09:18:24 +0000 (02:18 -0700)]
[PATCH] nsproxy cloning error path fix
This patch fixes copy_namespaces()'s error path.
when new nsproxy (new_ns) is created pointers to namespaces (ipc, uts) are
copied from the old nsproxy. Later in copy_utsname, copy_ipcs, etc.
according namespaces are get-ed. On error path needed namespaces are
put-ed, so there's no need to put new nsproxy itelf as it woud cause
putting namespaces for the second time.
Found when incorporating namespaces into OpenVZ kernel.
Signed-off-by: Pavel Emelianov <xemul@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Kirill Korotaev [Mon, 2 Oct 2006 09:18:19 +0000 (02:18 -0700)]
[PATCH] IPC namespace core
This patch set allows to unshare IPCs and have a private set of IPC objects
(sem, shm, msg) inside namespace. Basically, it is another building block of
containers functionality.
This patch implements core IPC namespace changes:
- ipc_namespace structure
- new config option CONFIG_IPC_NS
- adds CLONE_NEWIPC flag
- unshare support
[clg@fr.ibm.com: small fix for unshare of ipc namespace]
[akpm@osdl.org: build fix] Signed-off-by: Pavel Emelianov <xemul@openvz.org> Signed-off-by: Kirill Korotaev <dev@openvz.org> Signed-off-by: Cedric Le Goater <clg@fr.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Serge Hallyn [Mon, 2 Oct 2006 09:18:18 +0000 (02:18 -0700)]
[PATCH] uts: copy nsproxy only when needed
The nsproxy was being copied in unshare() when anything was being unshared,
even if it was something not referenced from nsproxy. This should end up
in some cases with far more memory usage than necessary.
Signed-off-by: Serge Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Serge E. Hallyn [Mon, 2 Oct 2006 09:18:15 +0000 (02:18 -0700)]
[PATCH] namespaces: utsname: sysctl
Sysctl uts patch. This will need to be done another way, but since sysctl
itself needs to be container aware, 'the right thing' is a separate patchset.
[akpm@osdl.org: ia64 build fix]
[sam.vilain@catalyst.net.nz: cleanup]
[sam.vilain@catalyst.net.nz: add proc_do_utsns_string] Signed-off-by: Serge E. Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Serge E. Hallyn [Mon, 2 Oct 2006 09:18:13 +0000 (02:18 -0700)]
[PATCH] namespaces: utsname: use init_utsname when appropriate
In some places, particularly drivers and __init code, the init utsns is the
appropriate one to use. This patch replaces those with a the init_utsname
helper.
Changes: Removed several uses of init_utsname(). Hope I picked all the
right ones in net/ipv4/ipconfig.c. These are now changed to
utsname() (the per-process namespace utsname) in the previous
patch (2/7)
[akpm@osdl.org: CIFS fix] Signed-off-by: Serge E. Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Cc: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Define utsname() and init_utsname() which return &system_utsname. Users of
system_utsname will be changed to use these helpers, after which
system_utsname will disappear.
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
exit_task_namespaces() has replaced the former exit_namespace(). It
invalidates task->nsproxy and associated namespaces. This is an issue for
the (futur) pid namespace which is required to be valid in exit_notify().
This patch moves exit_task_namespaces() after exit_notify() to keep nsproxy
valid.
Signed-off-by: Cedric Le Goater <clg@fr.ibm.com> Cc: Serge E. Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Serge E. Hallyn [Mon, 2 Oct 2006 09:18:08 +0000 (02:18 -0700)]
[PATCH] namespaces: incorporate fs namespace into nsproxy
This moves the mount namespace into the nsproxy. The mount namespace count
now refers to the number of nsproxies point to it, rather than the number of
tasks. As a result, the unshare_namespace() function in kernel/fork.c no
longer checks whether it is being shared.
Signed-off-by: Serge Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Serge E. Hallyn [Mon, 2 Oct 2006 09:18:06 +0000 (02:18 -0700)]
[PATCH] namespaces: add nsproxy
This patch adds a nsproxy structure to the task struct. Later patches will
move the fs namespace pointer into this structure, and introduce a new utsname
namespace into the nsproxy.
The vserver and openvz functionality, then, would be implemented in large part
by virtualizing/isolating more and more resources into namespaces, each
contained in the nsproxy.
Sam Vilain [Mon, 2 Oct 2006 09:18:04 +0000 (02:18 -0700)]
[PATCH] proc: sysctl: add _proc_do_string helper
The logic in proc_do_string is worth re-using without passing in a
ctl_table structure (say, we want to calculate a pointer and pass that in
instead); pass in the two fields it uses from that structure as explicit
arguments.
Signed-off-by: Sam Vilain <sam.vilain@catalyst.net.nz> Cc: Serge E. Hallyn <serue@us.ibm.com> Cc: Kirill Korotaev <dev@openvz.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Herbert Poetzl <herbert@13thfloor.at> Cc: Andrey Savochkin <saw@sw.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Peter Zijlstra [Mon, 2 Oct 2006 09:18:03 +0000 (02:18 -0700)]
[PATCH] nfsd: lockdep annotation
while doing a kernel make modules_install install over an NFS mount.
=============================================
[ INFO: possible recursive locking detected ]
---------------------------------------------
nfsd/9550 is trying to acquire lock:
(&inode->i_mutex){--..}, at: [<c034c845>] mutex_lock+0x1c/0x1f
but task is already holding lock:
(&inode->i_mutex){--..}, at: [<c034c845>] mutex_lock+0x1c/0x1f
other info that might help us debug this:
2 locks held by nfsd/9550:
#0: (hash_sem){..--}, at: [<cc895223>] exp_readlock+0xd/0xf [nfsd]
#1: (&inode->i_mutex){--..}, at: [<c034c845>] mutex_lock+0x1c/0x1f
=============================================
[ INFO: possible recursive locking detected ]
---------------------------------------------
nfsd/9580 is trying to acquire lock:
(&inode->i_mutex){--..}, at: [<c034cc1d>] mutex_lock+0x1c/0x1f
but task is already holding lock:
(&inode->i_mutex){--..}, at: [<c034cc1d>] mutex_lock+0x1c/0x1f
other info that might help us debug this:
2 locks held by nfsd/9580:
#0: (hash_sem){..--}, at: [<cc89522b>] exp_readlock+0xd/0xf [nfsd]
#1: (&inode->i_mutex){--..}, at: [<c034cc1d>] mutex_lock+0x1c/0x1f
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Neil Brown <neilb@suse.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg Banks [Mon, 2 Oct 2006 09:18:02 +0000 (02:18 -0700)]
[PATCH] knfsd: allow admin to set nthreads per node
Add /proc/fs/nfsd/pool_threads which allows the sysadmin (or a userspace
daemon) to read and change the number of nfsd threads in each pool. The
format is a list of space-separated integers, one per pool.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg Banks [Mon, 2 Oct 2006 09:18:01 +0000 (02:18 -0700)]
[PATCH] knfsd: make rpc threads pools numa aware
Actually implement multiple pools. On NUMA machines, allocate a svc_pool per
NUMA node; on SMP a svc_pool per CPU; otherwise a single global pool. Enqueue
sockets on the svc_pool corresponding to the CPU on which the socket bh is run
(i.e. the NIC interrupt CPU). Threads have their cpu mask set to limit them
to the CPUs in the svc_pool that owns them.
This is the patch that allows an Altix to scale NFS traffic linearly
beyond 4 CPUs and 4 NICs.
Incorporates changes and feedback from Neil Brown, Trond Myklebust, and
Christoph Hellwig.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg Banks [Mon, 2 Oct 2006 09:17:59 +0000 (02:17 -0700)]
[PATCH] knfsd: add svc_set_num_threads
Currently knfsd keeps its own list of all nfsd threads in nfssvc.c; add a new
way of managing the list of all threads in a svc_serv. Add
svc_create_pooled() to allow creation of a svc_serv whose threads are managed
by the sunrpc code. Add svc_set_num_threads() to manage the number of threads
in a service, either per-pool or globally across the service.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg Banks [Mon, 2 Oct 2006 09:17:58 +0000 (02:17 -0700)]
[PATCH] knfsd: split svc_serv into pools
Split out the list of idle threads and pending sockets from svc_serv into a
new svc_pool structure, and allocate a fixed number (in this patch, 1) of
pools per svc_serv. The new structure contains a lock which takes over
several of the duties of svc_serv->sv_lock, which is now relegated to
protecting only sv_tempsocks, sv_permsocks, and sv_tmpcnt in svc_serv.
The point is to move the hottest fields out of svc_serv and into svc_pool,
allowing a following patch to arrange for a svc_pool per NUMA node or per CPU.
This is a major step towards making the NFS server NUMA-friendly.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg Banks [Mon, 2 Oct 2006 09:17:57 +0000 (02:17 -0700)]
[PATCH] knfsd: test and set SK_BUSY atomically
The SK_BUSY bit in svc_sock->sk_flags ensures that we do not attempt to
enqueue a socket twice. Currently, setting and clearing the bit is protected
by svc_serv->sv_lock. As I intend to reduce the data that the lock protects
so it's not held when svc_sock_enqueue() tests and sets SK_BUSY, that test and
set needs to be atomic.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
projects / linux-2.6 / log