- 28 Oct, 2010 40 commits
-
-
Mel Gorman authored
Task delay-accounting was identified as one means of determining how long a process spends in congestion_wait() without adding new statistics. For example, if the workload should not be doing IO, delay-accounting could reveal how long it was spending in unexpected IO or delays. Unfortunately, on closer examination it was clear that getdelays does not act as documented. Commit a3baf649 ("per-task-delay-accounting: documentation") added Documentation/accounting/getdelays.c with a -c switch that was documented to fork/exec a child and report statistics on it but for reasons that are unclear to me, commit 9e06d3f9 deleted support for this switch but did not update the documentation. It might be an oversight or it might be because the control flow of the program meant that accounting information would be printed once early in the lifetime of the program making it of limited use. This patch reimplements -c for getdelays.c to act as documented. Unlike the original version, it waits until the command completes before printing any information on it. An example of it being used looks like $ ./getdelays -d -c find /home/mel -name mel print delayacct stats ON /home/mel /home/mel/.notes-wine/drive_c/windows/profiles/mel /home/mel/.wine/drive_c/windows/profiles/mel /home/mel/git-configs/dot.kde/share/apps/konqueror/home/mel PID 5923 CPU count real total virtual total delay total 42779 5051232096 5164722692 564207988 IO count delay total 41727 97804147758 SWAP count delay total 0 0 RECLAIM count delay total 0 0 [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Mel Gorman <mel@csn.ul.ie> Acked-by:
Balbir Singh <balbir@in.ibm.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Daniel Lezcano authored
We have the namespaces as a menuconfig like the cgroup. The cgroup and the namespace are two base bricks for the containers. It is more logical to put the namespace menu right after the cgroup menu. Signed-off-by:
Daniel Lezcano <daniel.lezcano@free.fr> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: "Serge E. Hallyn" <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Paul Menage <menage@google.com> Signed-off-by:
-
Daniel Lezcano authored
This subsystem is merged since a long time now, I think we can consider it mature enough. Signed-off-by:
-
Daniel Lezcano authored
The different cgroup subsystems are under the cgroup submenu. The dependency between the cgroups and the menu subsystems is pointless. Signed-off-by:
Li Zefan <lizf@cn.fujitsu.com> Cc: "Serge E. Hallyn" <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Paul Menage <menage@google.com> Signed-off-by:
-
Daniel Lezcano authored
Make the namespaces config option a submenu. Signed-off-by:
-
Daniel Lezcano authored
As the different namespaces depend on 'CONFIG_NAMESPACES', it is logical to enable all the namespaces when we enable NAMESPACES. Signed-off-by:
Matt Helsley <matthltc@us.ibm.com> Acked-by:
Serge Hallyn <serge.hallyn@canonical.com> Signed-off-by:
-
Daniel Lezcano authored
The pid namespace is in the kernel since 2.6.27 and the net_ns since 2.6.29. They are enabled in the distro by default and used by userspace component. They are mature enough to remove the 'experimental' label. Signed-off-by:
-
Alexandre Bounine authored
Replace possibly damaging broadcast writes into the per-port SP_MODE registers with individual writes for each port. This will preserve individual port configurations in case if ports are configured differently. Signed-off-by:
Alexandre Bounine <alexandre.bounine@idt.com> Cc: Kumar Gala <galak@kernel.crashing.org> Cc: Matt Porter <mporter@kernel.crashing.org> Cc: Li Yang <leoli@freescale.com> Cc: Thomas Moll <thomas.moll@sysgo.com> Cc: Micha Nelissen <micha@neli.hopto.org> Signed-off-by:
-
Alexandre Bounine authored
Replace hardcoded maximum port number with actual reported number of switch ports. Signed-off-by:
-
Alexandre Bounine authored
RapidIO spec v.2.1 adds Idle Sequence 2 into LP-Serial Physical Layer. The fix ensures that corresponding bits are not corrupted during error handling. Signed-off-by:
-
Alexandre Bounine authored
Detects RIO link to the already enumerated device and properly sets links between device objects. Changes to the enumeration/discovery logic: 1. Use Master Enable bit to signal end of the enumeration - agents may start their discovery process as soon as they see this bit set (Component Tag register was used before for this purpose). 2. Enumerator sets Component Tag (!= 0) immediately during device setup. This allows to identify the device if the redundant route exists in a RIO system. Signed-off-by:
-
Alexandre Bounine authored
Add the RIO switch driver and definitions for IDT CPS-1848 and CPS-1616 Gen2 devices. Signed-off-by:
-
Alexandre Bounine authored
Add explicit device access check before performing device enumeration. This gives a chance to clear possible link error conditions by issuing safe maintenance read request(s). Signed-off-by:
-
Alexandre Bounine authored
Add check for access to port-write (PW) message source device before processing the PW message. If source RIO device is not available (power down or RIO link failure) trace back to a last available switch/port on the PW message route and service failure at that point. Signed-off-by:
-
Alexandre Bounine authored
1. Change to create attribute "routes" only for switches. 2. Add a switch-specific callback to create/remove proprietary attributes. Signed-off-by:
-
Alexandre Bounine authored
The default error-stopped state handler provides recovery mechanism as defined by RIO specification. Signed-off-by:
-
Alexandre Bounine authored
Create back and forward links between RIO devices. These links are intended for use by error management and hot-plug extensions. Links for redundant RIO connections between switches are not set (will be fixed in a separate patch). Signed-off-by:
-
Alexandre Bounine authored
The switch port information is obtained and stored during RIO device setup. Therefore repeated reads from Switch Port Information CAR may be removed. Signed-off-by:
-
Alexandre Bounine authored
- Rearrange RIO port-write interrupt handling to perform message buffering as soon as possible. - Modify to disable port-write controller when clearing Transaction Error (TE) bit. Signed-off-by:
-
Alexandre Bounine authored
This set of RapidIO patches extends support for standard error recovery mechanism and adds new IDT Gen2 sRIO switch devices - CPS-1848 and CPS-1616. Implementation of the standard error-stopped state recovery mechanism (as defined by the RapidIO specification) is required for the new switches. Version 2 of this set of patches addresses received comments and fixes an error notification setup issue found in the idt_gen2.c after the first version was released. This patch: Make RapidIO devices appear in /sys/devices/rapidio directory instead of top of /sys/devices directory. Signed-off-by:
-
Rakib Mullick authored
Both mxser_disable_must_enchance_mode() and mxser_get_must_hardware_id() called from function CheckIsMoxaMust(), when CONFIG_PCI=y. So mark both the functions under CONFIG_PCI. We were warned by the following warning. drivers/char/mxser.c:306: warning: `mxser_disable_must_enchance_mode' defined but not used drivers/char/mxser.c:391: warning: `mxser_get_must_hardware_id' defined but not used Signed-off-by:
Rakib Mullick <rakib.mullick@gmail.com> Cc: Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
Tracey Dent authored
Changed <module>-objs to <module>-y in Makefile. Signed-off-by:
Tracey Dent <tdent48227@gmail.com> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Michal Marek <mmarek@suse.cz> Signed-off-by:
-
Tracey Dent authored
Changed <module>-objs to <module>-y in Makefile. Signed-off-by:
Jiri Kosina <jkosina@suse.cz> Acked-by:
David Sterba <dsterba@suse.cz> Signed-off-by:
-
Tracey Dent authored
Changed <module>-objs to <module>-y in Makefile and use ccflags-y option. Signed-off-by:
-
Tracey Dent authored
Changed <module>-objs to <module>-y in Makefile. Signed-off-by:
-
Tracey Dent authored
Changed <module>-objs to <module>-y in Makefile. Signed-off-by:
-
Paul Fulghum authored
Add support for extended byte synchronous mode feature of hardware. Signed-off-by:
Paul Fulghum <paulkf@microgate.com> Signed-off-by:
-
Vasiliy Kulikov authored
Structures tmp_params and new_line are copied to userland with some padding fields unitialized. It leads to leaking of stack memory. Signed-off-by:
Vasiliy Kulikov <segooon@gmail.com> Acked-by:
-
Vasiliy Kulikov authored
Structure par_timeout is copied to userland with some padding fields unitialized. Field tv_usec has type __kernel_suseconds_t, it differs from tv_sec's type on some architectures. It leads to leaking of stack memory. Signed-off-by:
-
Vasiliy Kulikov authored
Structure st_loc is copied to userland with some fields unitialized. It leads to leaking of stack memory. Signed-off-by:
-
Paul Fulghum authored
Fix a long standing bug with per device locking. Each device has an associated spinlock for synchronizing access to hardware and state information with the ISR. A single hardware card has one or more devices. Bug: Non ISR code correctly acquires and releases the per device lock. ISR incorrectly always acquires and releases the lock of the first device on the card. The decoupled and list based nature of the ISR and deferred processing interaction allowed this to work in normal operation. Exceptional events like an application forcing hardware shutdown, reset, or reconfiguration while active can trigger the bug. Fixed ISR to acquire and release the per device lock. One exception is manipulation of the GPIO card resource which is global and effectively owned by the first device of the card. Non-ISR access to GPIO resource is changed to use lock of first device on card. Signed-off-by:
-
Dimitri Sivanich authored
This patch for SGI Altix/IA64 eliminates interval long timer holdoffs in cases where we don't start an interval timer before the expiration time. This sometimes happens when a number of interval timers on the same shub with the same interval run simultaneously. Signed-off-by:
Dimitri Sivanich <sivanich@sgi.com> Cc: "Luck, Tony" <tony.luck@intel.com> Signed-off-by:
-
Andrew Morton authored
This doesn't do anything. Cc: Timur Tabi <timur@freescale.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Amit Shah <amit.shah@redhat.com> Cc: Greg Kroah-Hartman <gregkh@suse.de> Cc: Kumar Gala <galak@gate.crashing.org> Signed-off-by:
-
Dan Carpenter authored
There was a release_region() missing on the error path. Signed-off-by:
Dan Carpenter <error27@gmail.com> Signed-off-by:
-
Corey Minyard authored
__init and __exit belong after the return type on functions, not before. Signed-off-by:
Corey Minyard <cminyard@mvista.com> Signed-off-by:
-
Dan Rosenberg authored
This takes care of leaking uninitialized kernel stack memory to userspace from non-zeroed fields in structs in compat ipc functions. Signed-off-by:
Dan Rosenberg <drosenberg@vsecurity.com> Cc: Manfred Spraul <manfred@colorfullife.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: <stable@kernel.org> Signed-off-by:
-
Helge Deller authored
The kernel currently provides no functionality to analyze the RSS and swap space usage of each individual sysvipc shared memory segment. This patch adds this info for each existing shm segment by extending the output of /proc/sysvipc/shm by two columns for RSS and swap. Since shmctl(SHM_INFO) already provides a similiar calculation (it currently sums up all RSS/swap info for all segments), I did split out a static function which is now used by the /proc/sysvipc/shm output and shmctl(SHM_INFO). SAP products (esp. the SAP Netweaver ABAP Kernel) uses lots of big shared memory segments (we often have Linux systems with >= 16GB shm usage). Sometimes we get customer reports about "slow" system responses and while looking into their configurations we often find massive swapping activity on the system. With this patch it's now easy to see from the command line if and which shm segments gets swapped out (and how much) and can more easily give recommendations for system tuning. Without the patch it's currently not possible to do such shm analysis at all. Also... Add some spaces in front of the "size" field for 64bit kernels to get the columns correct if you cat the contents of the file. In sysvipc_shm_proc_show() the kernel prints the size value in "SPEC_SIZE" format, which is defined like this: #if BITS_PER_LONG <= 32 #define SIZE_SPEC "%10lu" #else #define SIZE_SPEC "%21lu" #endif So, if the header is not adjusted, the columns are not correctly aligned. I actually tested this on 32- and 64-bit and it seems correct now. Signed-off-by:
Helge Deller <deller@gmx.de> Cc: Manfred Spraul <manfred@colorfullife.com> Acked-by:
Hugh Dickins <hughd@google.com> Signed-off-by:
-
KOSAKI Motohiro authored
Presently do_execve() turns PF_KTHREAD off before search_binary_handler(). THis has a theorical risk of PF_KTHREAD getting lost. We don't have to turn PF_KTHREAD off in the ENOEXEC case. This patch moves this flag modification to after the finding of the executable file. This is only a theorical issue because kthreads do not call do_execve() directly. But fixing would be better. Signed-off-by:
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Acked-by:
Roland McGrath <roland@redhat.com> Acked-by:
Oleg Nesterov <oleg@redhat.com> Signed-off-by:
-
KAMEZAWA Hiroyuki authored
In /proc/stat, the number of per-IRQ event is shown by making a sum each irq's events on all cpus. But we can make use of kstat_irqs(). kstat_irqs() do the same calculation, If !CONFIG_GENERIC_HARDIRQ, it's not a big cost. (Both of the number of cpus and irqs are small.) If a system is very big and CONFIG_GENERIC_HARDIRQ, it does for_each_irq() for_each_cpu() - look up a radix tree - read desc->irq_stat[cpu] This seems not efficient. This patch adds kstat_irqs() for CONFIG_GENRIC_HARDIRQ and change the calculation as for_each_irq() look up radix tree for_each_cpu() - read desc->irq_stat[cpu] This reduces cost. A test on (4096cpusp, 256 nodes, 4592 irqs) host (by Jack Steiner) %time cat /proc/stat > /dev/null Before Patch: 2.459 sec After Patch : .561 sec [akpm@linux-foundation.org: unexport kstat_irqs, coding-style tweaks] [akpm@linux-foundation.org: fix unused variable 'per_irq_sum'] Signed-off-by:
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Tested-by:
Jack Steiner <steiner@sgi.com> Acked-by:
-
KAMEZAWA Hiroyuki authored
/proc/stat shows the total number of all interrupts to each cpu. But when the number of IRQs are very large, it take very long time and 'cat /proc/stat' takes more than 10 secs. This is because sum of all irq events are counted when /proc/stat is read. This patch adds "sum of all irq" counter percpu and reduce read costs. The cost of reading /proc/stat is important because it's used by major applications as 'top', 'ps', 'w', etc.... A test on a mechin (4096cpu, 256 nodes, 4592 irqs) shows %time cat /proc/stat > /dev/null Before Patch: 12.627 sec After Patch: 2.459 sec Signed-off-by:
-
