- 03 Apr, 2009 40 commits
-
-
KAMEZAWA Hiroyuki authored
This patch tries to fix OOM Killer problems caused by hierarchy. Now, memcg itself has OOM KILL function (in oom_kill.c) and tries to kill a task in memcg. But, when hierarchy is used, it's broken and correct task cannot be killed. For example, in following cgroup /groupA/ hierarchy=1, limit=1G, 01 nolimit 02 nolimit All tasks' memory usage under /groupA, /groupA/01, groupA/02 is limited to groupA's 1Gbytes but OOM Killer just kills tasks in groupA. This patch provides makes the bad process be selected from all tasks under hierarchy. BTW, currently, oom_jiffies is updated against groupA in above case. oom_jiffies of tree should be updated. To see how oom_jiffies is used, please check mem_cgroup_oom_called() callers. [akpm@linux-foundation.org: build fix] [akpm@linux-foundation.org: const fix] Signed-off-by:
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Cc: David Rientjes <rientjes@google.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
KAMEZAWA Hiroyuki authored
As pointed out, shrinking memcg's limit should return -EBUSY after reasonable retries. This patch tries to fix the current behavior of shrink_usage. Before looking into "shrink should return -EBUSY" problem, we should fix hierarchical reclaim code. It compares current usage and current limit, but it only makes sense when the kernel reclaims memory because hit limits. This is also a problem. What this patch does are. 1. add new argument "shrink" to hierarchical reclaim. If "shrink==true", hierarchical reclaim returns immediately and the caller checks the kernel should shrink more or not. (At shrinking memory, usage is always smaller than limit. So check for usage < limit is useless.) 2. For adjusting to above change, 2 changes in "shrink"'s retry path. 2-a. retry_count depends on # of children because the kernel visits the children under hierarchy one by one. 2-b. rather than checking return value of hierarchical_reclaim's progress, compares usage-before-shrink and usage-after-shrink. If usage-before-shrink <= usage-after-shrink, retry_count is decremented. Reported-by:Li Zefan <lizf@cn.fujitsu.com> Signed-off-by:
-
KAMEZAWA Hiroyuki authored
Clean up memory.stat file routine and show "total" hierarchical stat. This patch does - renamed get_all_zonestat to be get_local_zonestat. - remove old mem_cgroup_stat_desc, which is only for per-cpu stat. - add mcs_stat to cover both of per-cpu/per-lru stat. - add "total" stat of hierarchy (*) - add a callback system to scan all memcg under a root. == "total" is added. [kamezawa@localhost ~]$ cat /opt/cgroup/xxx/memory.stat cache 0 rss 0 pgpgin 0 pgpgout 0 inactive_anon 0 active_anon 0 inactive_file 0 active_file 0 unevictable 0 hierarchical_memory_limit 50331648 hierarchical_memsw_limit 9223372036854775807 total_cache 65536 total_rss 192512 total_pgpgin 218 total_pgpgout 155 total_inactive_anon 0 total_active_anon 135168 total_inactive_file 61440 total_active_file 4096 total_unevictable 0 == (*) maybe the user can do calc hierarchical stat by his own program in userland but if it can be written in clean way, it's worth to be shown, I think. Signed-off-by:
-
KAMEZAWA Hiroyuki authored
Assigning CSS ID for each memcg and use css_get_next() for scanning hierarchy. Assume folloing tree. group_A (ID=3) /01 (ID=4) /0A (ID=7) /02 (ID=10) group_B (ID=5) and task in group_A/01/0A hits limit at group_A. reclaim will be done in following order (round-robin). group_A(3) -> group_A/01 (4) -> group_A/01/0A (7) -> group_A/02(10) -> group_A -> ..... Round robin by ID. The last visited cgroup is recorded and restart from it when it start reclaim again. (More smart algorithm can be implemented..) No cgroup_mutex or hierarchy_mutex is required. Signed-off-by:
-
Li Zefan authored
There is nothing special that has to be protected by cgroup_lock, so introduce devcgroup_mtuex for it's own use. Signed-off-by:
Serge Hallyn <serue@us.ibm.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by:
-
Li Zefan authored
Since we are in cgroup write handler, so the cgrp is valid, so we don't have to hold cgroup_mutex when calling cgroup_task_count(). One similar example is in cgroup_tasks_open(). Signed-off-by:
Paul Menage <menage@google.com> Signed-off-by:
-
Li Zefan authored
Remount can fail in either case: - wrong mount options is specified, or option 'noprefix' is changed. - a to-be-added subsys is already mounted/active. When using remount to change 'release_agent', for the above former failure case, remount will return errno with release_agent unchanged, but for the latter case, remount will return EBUSY with relase_agent changed, which is unexpected I think: # mount -t cgroup -o cpu xxx /cgrp1 # mount -t cgroup -o cpuset,release_agent=agent1 yyy /cgrp2 # cat /cgrp2/release_agent agent1 # mount -t cgroup -o remount,cpuset,noprefix,release_agent=agent2 yyy /cgrp2 mount: /cgrp2 not mounted already, or bad option # cat /cgrp2/release_agent agent1 <-- ok # mount -t cgroup -o remount,cpu,cpuset,release_agent=agent2 yyy /cgrp2 mount: /cgrp2 is busy # cat /cgrp2/release_agent agent2 <-- unexpected! Signed-off-by:
-
Li Zefan authored
We have some read-only files and write-only files, but currently they are all set to 0644, which is counter-intuitive and cause trouble for some cgroup tools like libcgroup. This patch adds 'mode' to struct cftype to allow cgroup subsys to set it's own files' file mode, and for the most cases cft->mode can be default to 0 and cgroup will figure out proper mode. Acked-by:
-
Li Zefan authored
This won't remove cpuacct from the mounted hierachy: # mount -t cgroup -o cpu,cpuacct xxx /mnt # mount -o remount,cpu /mnt Because for this usage mount(8) will append the new options to the original options. And this will get you right: # mount [-t cgroup] -o remount,cpu xxx /mnt Also document how to specify or change release_agent. Signed-off-by:
-
Jesper Juhl authored
Reduces object file size a bit: Before: $ size kernel/cgroup.o text data bss dec hex filename 21593 7804 4924 34321 8611 kernel/cgroup.o After: $ size kernel/cgroup.o text data bss dec hex filename 21537 7744 4924 34205 859d kernel/cgroup.o Signed-off-by:
Jesper Juhl <jj@chaosbits.net> Cc: Paul Menage <menage@google.com> Signed-off-by:
-
KAMEZAWA Hiroyuki authored
In following situation, with memory subsystem, /groupA use_hierarchy==1 /01 some tasks /02 some tasks /03 some tasks /04 empty When tasks under 01/02/03 hit limit on /groupA, hierarchical reclaim is triggered and the kernel walks tree under groupA. In this case, rmdir /groupA/04 fails with -EBUSY frequently because of temporal refcnt from the kernel. In general. cgroup can be rmdir'd if there are no children groups and no tasks. Frequent fails of rmdir() is not useful to users. (And the reason for -EBUSY is unknown to users.....in most cases) This patch tries to modify above behavior, by - retries if css_refcnt is got by someone. - add "return value" to pre_destroy() and allows subsystem to say "we're really busy!" Signed-off-by:
-
KAMEZAWA Hiroyuki authored
Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by:Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by:
-
Grzegorz Nosek authored
The ns_proxy cgroup allows moving processes to child cgroups only one level deep at a time. This commit relaxes this restriction and makes it possible to attach tasks directly to grandchild cgroups, e.g.: ($pid is in the root cgroup) echo $pid > /cgroup/CG1/CG2/tasks Previously this operation would fail with -EPERM and would have to be performed as two steps: echo $pid > /cgroup/CG1/tasks echo $pid > /cgroup/CG1/CG2/tasks Also, the target cgroup no longer needs to be empty to move a task there. Signed-off-by:
Grzegorz Nosek <root@localdomain.pl> Acked-by:
-
Paul Menage authored
Fix the style of some multi-line comments in cgroup.h to match Documentation/CodingStyle Signed-off-by:
-
Li Xiaodong authored
Previous description about system parameter in /proc/sys/net/unix/ is wrong (or missed). Simply add a new description about unix_dgram_qlen according to latest kernel. Signed-off-by:
Li Xiaodong <lixd@cn.fujitsu.com> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by:
-
Shen Feng authored
Now /proc/sys is described in many places and much information is redundant. This patch updates the proc.txt and move the /proc/sys desciption out to the files in Documentation/sysctls. Details are: merge - 2.1 /proc/sys/fs - File system data - 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem - 2.17 /proc/sys/fs/epoll - Configuration options for the epoll interface with Documentation/sysctls/fs.txt. remove - 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats since it's not better then the Documentation/binfmt_misc.txt. merge - 2.3 /proc/sys/kernel - general kernel parameters with Documentation/sysctls/kernel.txt remove - 2.5 /proc/sys/dev - Device specific parameters since it's obsolete the sysfs is used now. remove - 2.6 /proc/sys/sunrpc - Remote procedure calls since it's not better then the Documentation/sysctls/sunrpc.txt move - 2.7 /proc/sys/net - Networking stuff - 2.9 Appletalk - 2.10 IPX to newly created Documentation/sysctls/net.txt. remove - 2.8 /proc/sys/net/ipv4 - IPV4 settings since it's not better then the Documentation/networking/ip-sysctl.txt. add - Chapter 3 Per-Process Parameters to descibe /proc/<pid>/xxx parameters. Signed-off-by:
Shen Feng <shen@cn.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by:
-
Henrik Austad authored
When using 'make pdfdocs', auto-generated files should be ignored Signed-off-by:
Henrik Austad <henrik@austad.us> Cc: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
Roel Kluin authored
hppfs_read_file() may return (ssize_t) -ENOMEM, or -EFAULT. When stored in size_t 'count', these errors will not be noticed, a large value will be added to *ppos. Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
-
Jan Kara authored
Sometimes block_write_begin() can map buffers in a page but later we fail to copy data into those buffers (because the source page has been paged out in the mean time). We then end up with !uptodate mapped buffers. To add a bit more to the confusion, block_write_end() does not commit any data (and thus does not any mark buffers as uptodate) if we didn't succeed with copying all the data. Commit f4fc66a8 (ext3: convert to new aops) missed these cases and thus we were inserting non-uptodate buffers to transaction's list which confuses JBD code and it reports IO errors, aborts a transaction and generally makes users afraid about their data ;-P. This patch fixes the problem by reorganizing ext3_..._write_end() code to first call block_write_end() to mark buffers with valid data uptodate and after that we file only uptodate buffers to transaction's lists. We also fix a problem where we could leave blocks allocated beyond i_size (i_disksize in fact) because of failed write. We now add inode to orphan list when write fails (to be safe in case we crash) and then truncate blocks beyond i_size in a separate transaction. Signed-off-by:
Jan Kara <jack@suse.cz> Reviewed-by:
Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: <linux-ext4@vger.kernel.org> Signed-off-by:
-
Bryan Donlan authored
ext3_iget() returns -ESTALE if invoked on a deleted inode, in order to report errors to NFS properly. However, in ext[234]_lookup(), this -ESTALE can be propagated to userspace if the filesystem is corrupted such that a directory entry references a deleted inode. This leads to a misleading error message - "Stale NFS file handle" - and confusion on the part of the admin. The bug can be easily reproduced by creating a new filesystem, making a link to an unused inode using debugfs, then mounting and attempting to ls -l said link. This patch thus changes ext3_lookup to return -EIO if it receives -ESTALE from ext3_iget(), as ext3 does for other filesystem metadata corruption; and also invokes the appropriate ext*_error functions when this case is detected. Signed-off-by:
Bryan Donlan <bdonlan@gmail.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by:
-
Wei Yongjun authored
Use unsigned instead of int for the parameter which carries a blocksize. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Wei Yongjun <yjwei@cn.fujitsu.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by:
-
Jan Kara authored
On 32-bit system with CONFIG_LBD getblk can fail because provided block number is too big. Make JBD gracefully handle that. Signed-off-by:
-
Cyrus Massoumi authored
Reformat ext3/ioctl.c to make it look more like ext4/ioctl.c and remove the BKL around ext3_ioctl(). Signed-off-by:
Cyrus Massoumi <cyrusm@gmx.net> Cc: <linux-ext4@vger.kernel.org> Acked-by:
Jan Kara <jack@ucw.cz> Signed-off-by:
-
Erik Ekman authored
- Error code from kthread_run() is now returned in pnpbios_thread_init() - Remove variable which always was 0. Signed-off-by:
Erik Ekman <erik@kryo.se> Cc: Bjorn Helgaas <bjorn.helgaas@hp.com> Signed-off-by:
-
Erik Ekman authored
drivers/pnp/pnpbios/core.c: In function 'pnpbios_thread_init': drivers/pnp/pnpbios/core.c:578: warning: unused variable 'task' Signed-off-by:
-
Michael Buesch authored
Change spi-gpio so that it is possible to drive SPI communications over GPIO without the need for a chipselect signal. This is useful in very small setups where there's only one slave device on the bus. This patch does not affect existing setups. I use this for a tiny communication channel between an embedded device and a microcontroller. There are not enough GPIOs available for chipselect and it's not needed anyway in this case. Signed-off-by:
Michael Buesch <mb@bu3sch.de> Cc: David Brownell <david-b@pacbell.net> Signed-off-by:
-
David Brownell authored
We want to phase out the GPIO "autorequest" mechanism in gpiolib and require all callers to use gpio_request(). - Update feature-removal-schedule - Update the documentation now - Convert the relevant pr_warning() in gpiolib to a WARN() so folk using this mechanism get a noisy stack dump Some drivers and board init code will probably need to change. Implementations not using gpiolib will still be fine; they are already required to implement gpio_{request,free}() stubs. Signed-off-by:David Brownell <dbrownell@users.sourceforge.net> Signed-off-by:
-
Daniel Silverstone authored
Allow GPIOs in GPIOLIB chips to be named. This name is then used when the GPIO is exported to sysfs, although it could be used elsewhere if deemed useful. Signed-off-by:
Daniel Silverstone <dsilvers@simtec.co.uk> Cc: David Brownell <david-b@pacbell.net> Signed-off-by:
-
Daniel Glockner authored
Compared to the other supported chips, the m41t62 uses a different register to set the square wave frequency. Signed-off-by:
Daniel Glockner <dg@emlix.com> Cc: Chris Zankel <chris@zankel.net> Cc: David Brownell <david-b@pacbell.net> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by:
-
Mike Rapoport authored
The v3020 RTC can be connected to GPIOs as well as to memory-like interface. Add ability to use GPIO bit-bang for v3020 read-write access. [akpm@linux-foundation.org: fix off-by-one in error path] Signed-off-by:
Mike Rapoport <mike@compulab.co.il> Acked-by:
Alessandro Zummo <a.zummo@towertech.it> Cc: David Brownell <david-b@pacbell.net> Signed-off-by:
-
Simon Kitching authored
initramfs uses printk without a linefeed, then does some work, then uses printk to finish the message off. However if some other code does a printk in between, then the messages get mixed together. Better for each message to be an independent line... Example of problem that this fixes: checking if image is initramfs...<7>Switched to high resolution mode on CPU 1 Switched to high resolution mode on CPU 0 it is Signed-off-by:Simon Kitching <skitching@apache.org> Signed-off-by:
-
Alexey Dobriyan authored
First argument unused since 2.3.11. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Cc: <linux-arch@vger.kernel.org> Signed-off-by:
-
David Brownell authored
- Define new setup() hook to export the accessor - Implement accessor methods Moves some error checking out of the sysfs interface code into the layer below it, which is now shared by both sysfs and memory access code. Signed-off-by:
Kevin Hilman <khilman@deeprootsystems.com> Cc: Jean Delvare <khali@linux-fr.org> Signed-off-by:
-
Kevin Hilman authored
In the case of at24, the platform code registers a 'setup' callback with the at24_platform_data. When the at24 driver detects an EEPROM, it fills out the read and write functions of the memory_accessor and calls the setup callback passing the memory_accessor struct. The platform code can then use the read/write functions in the memory_accessor struct for reading and writing the EEPROM. Signed-off-by:
-
Kevin Hilman authored
Add an interface by which other kernel code can read/write persistent memory such as I2C or SPI EEPROMs, or devices which provide NVRAM. Use cases include storage of board-specific configuration data like Ethernet addresses and sensor calibrations. Original idea, review and improvement suggestions by David Brownell. Acked-by:
-
Jean Delvare authored
It is a fairly common operation to have a pointer to a work and to need a pointer to the delayed work it is contained in. In particular, all delayed works which want to rearm themselves will have to do that. So it would seem fair to offer a helper function for this operation. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Jean Delvare <khali@linux-fr.org> Acked-by:
Ingo Molnar <mingo@elte.hu> Cc: "David S. Miller" <davem@davemloft.net> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Greg KH <greg@kroah.com> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by:
-
Miklos Szeredi authored
Fix the following warnings: arch/um/kernel/syscall.c: In function 'kernel_execve': arch/um/kernel/syscall.c:130: warning: passing argument 1 of 'um_execve' discards qualifiers from pointer target type arch/um/kernel/syscall.c:130: warning: passing argument 2 of 'um_execve' discards qualifiers from pointer target type arch/um/kernel/syscall.c:130: warning: passing argument 3 of 'um_execve' discards qualifiers from pointer target type Signed-off-by:
Miklos Szeredi <mszeredi@suse.cz> Cc: Jeff Dike <jdike@addtoit.com> Cc: WANG Cong <xiyou.wangcong@gmail.com> Signed-off-by:
-
Miklos Szeredi authored
Fix the following link error: arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x11c): undefined reference to `ptregs_fork' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x140): undefined reference to `ptregs_execve' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x2cc): undefined reference to `ptregs_iopl' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x2d8): undefined reference to `ptregs_vm86old' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x2f0): undefined reference to `ptregs_sigreturn' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x2f4): undefined reference to `ptregs_clone' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x3ac): undefined reference to `ptregs_vm86' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x3c8): undefined reference to `ptregs_rt_sigreturn' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x3fc): undefined reference to `ptregs_sigaltstack' arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x40c): undefined reference to `ptregs_vfork' This was introduced by commit 253f29a4, "x86: pass in pt_regs pointer for syscalls that need it" Signed-off-by:
WANG Cong <xiyou.wangcong@gmail.com> Signed-off-by:
-
Miklos Szeredi authored
Fix the following compile error: arch/um/drivers/net_kern.c: In function 'uml_inetaddr_event': arch/um/drivers/net_kern.c:760: error: 'struct net_device' has no member named 'open' This was introduced by commit 8bb95b39, "uml: convert network device to netdevice ops". Signed-off-by:
-
Scott James Remnant authored
The missing device table means that the floppy module is not auto-loaded, even when the appropriate PNP device (0700) is found. We don't actually use the table in the module, since the device doesn't have a struct pnp_driver, but it's sufficient to cause an alias in the module that udev/modprobe will use. Signed-off-by:
Scott James Remnant <scott@canonical.com> Signed-off-by:
Tim Gardner <tim.gardner@canonical.com> Cc: Bjorn Helgaas <bjorn.helgaas@hp.com> Cc: Philippe De Muyter <phdm@macqel.be> Acked-by:
Kay Sievers <kay.sievers@vrfy.org> Signed-off-by:
-
