)]}' { "log": [ { "commit": "e4e364e865b382f9d99c7fc230ec2ce7df21257a", "tree": "9ff5ab54a0e40d7ad2b55d3ec48c6e175ebf50c7", "parents": [ "2741a559a01e1ba9bf87285569dc1a104d134ecf" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 31 02:30:52 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 31 12:18:55 2006 -0800" }, "message": "[PATCH] cpuset: memory migration interaction fix\n\nFix memory migration so that it works regardless of what cpuset the invoking\ntask is in.\n\nIf a task invoked a memory migration, by doing one of:\n\n 1) writing a different nodemask to a cpuset \u0027mems\u0027 file, or\n\n 2) writing a tasks pid to a different cpuset\u0027s \u0027tasks\u0027 file,\n where the cpuset had its \u0027memory_migrate\u0027 option turned on, then the\n allocation of the new pages for the migrated task(s) was constrained\n by the invoking tasks cpuset.\n\nIf this task wasn\u0027t in a cpuset that allowed the requested memory nodes, the\nmemory migration would happen to some other nodes that were in that invoking\ntasks cpuset. This was usually surprising and puzzling behaviour: Why didn\u0027t\nthe pages move? Why did the pages move -there-?\n\nTo fix this, temporarilly change the invoking tasks \u0027mems_allowed\u0027 task_struct\nfield to the nodes the migrating tasks is moving to, so that new pages can be\nallocated there.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "2741a559a01e1ba9bf87285569dc1a104d134ecf", "tree": "8c61321d5fa1a0c3517219eaa3089e223d5cd943", "parents": [ "4a01c8d5be628ac20cfd432c21808d76be5813e6" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 31 02:30:51 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 31 12:18:55 2006 -0800" }, "message": "[PATCH] cpuset: unsafe mm reference fix\n\nFix unsafe reference to a tasks mm struct, by moving the reference inside of a\nconvenient nearby properly guarded code block.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "4a01c8d5be628ac20cfd432c21808d76be5813e6", "tree": "040ae0d60df3993a752ce852d3e3dc036586b037", "parents": [ "2cf8d82d63807c2c68adf20bb28bf502496186dd" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 31 02:30:50 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 31 12:18:55 2006 -0800" }, "message": "[PATCH] cpuset: task_lock comment fix\n\nFix cpuset comment involving case of a tasks cpuset pointer being NULL.\nThanks to \"the_top_cpuset_hack\", this code no longer sees NULL task-\u003ecpuset\npointers.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "29afd49b72a9b2c26fa8c678bcf3976d0540446b", "tree": "4cc12f5bb3c5c3ee668226990e79115081609265", "parents": [ "b2455396be35383c4eebc6745cc718b1dd9e23df" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:12 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:24 2006 -0800" }, "message": "[PATCH] cpuset: remove useless local variable initialization\n\nRemove a useless variable initialization in cpuset __cpuset_zone_allowed().\n The local variable \u0027allowed\u0027 is unconditionally set before use, later on\nin the code, so does not need to be initialized.\n\nNot that it seems to matter to the code generated any, as the compiler\noptimizes out the superfluous assignment anyway.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "151a44202d097ae8b1bbaa6d8d2f97df30e3cd1e", "tree": "3752bba902e73001443bb75f40495cc3a1d24f54", "parents": [ "8488bc359d674baf710992e4b641513ea5ebd212" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:11 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:24 2006 -0800" }, "message": "[PATCH] cpuset: don\u0027t need to mark cpuset_mems_generation atomic\n\nDrop the atomic_t marking on the cpuset static global\ncpuset_mems_generation. Since all access to it is guarded by the global\nmanage_mutex, there is no need for further serialization of this value.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8488bc359d674baf710992e4b641513ea5ebd212", "tree": "28f6558f6b006c5a28a64a1beabec5498d9b77d4", "parents": [ "b0196009d8c3ecf6ea6ec080c63d2ccc146e7ad9" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:10 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:23 2006 -0800" }, "message": "[PATCH] cpuset: remove unnecessary NULL check\n\nRemove a no longer needed test for NULL cpuset pointer, with a little\ncomment explaining why the test isn\u0027t needed.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "825a46af5ac171f9f41f794a0a00165588ba1589", "tree": "b690fe9d809d7b047f0393097fc79892e1217d98", "parents": [ "8a39cc60bfa5a72f32d975729a354daca124f6de" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:03 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:22 2006 -0800" }, "message": "[PATCH] cpuset memory spread basic implementation\n\nThis patch provides the implementation and cpuset interface for an alternative\nmemory allocation policy that can be applied to certain kinds of memory\nallocations, such as the page cache (file system buffers) and some slab caches\n(such as inode caches).\n\nThe policy is called \"memory spreading.\" If enabled, it spreads out these\nkinds of memory allocations over all the nodes allowed to a task, instead of\npreferring to place them on the node where the task is executing.\n\nAll other kinds of allocations, including anonymous pages for a tasks stack\nand data regions, are not affected by this policy choice, and continue to be\nallocated preferring the node local to execution, as modified by the NUMA\nmempolicy.\n\nThere are two boolean flag files per cpuset that control where the kernel\nallocates pages for the file system buffers and related in kernel data\nstructures. They are called \u0027memory_spread_page\u0027 and \u0027memory_spread_slab\u0027.\n\nIf the per-cpuset boolean flag file \u0027memory_spread_page\u0027 is set, then the\nkernel will spread the file system buffers (page cache) evenly over all the\nnodes that the faulting task is allowed to use, instead of preferring to put\nthose pages on the node where the task is running.\n\nIf the per-cpuset boolean flag file \u0027memory_spread_slab\u0027 is set, then the\nkernel will spread some file system related slab caches, such as for inodes\nand dentries evenly over all the nodes that the faulting task is allowed to\nuse, instead of preferring to put those pages on the node where the task is\nrunning.\n\nThe implementation is simple. Setting the cpuset flags \u0027memory_spread_page\u0027\nor \u0027memory_spread_cache\u0027 turns on the per-process flags PF_SPREAD_PAGE or\nPF_SPREAD_SLAB, respectively, for each task that is in the cpuset or\nsubsequently joins that cpuset. In subsequent patches, the page allocation\ncalls for the affected page cache and slab caches are modified to perform an\ninline check for these flags, and if set, a call to a new routine\ncpuset_mem_spread_node() returns the node to prefer for the allocation.\n\nThe cpuset_mem_spread_node() routine is also simple. It uses the value of a\nper-task rotor cpuset_mem_spread_rotor to select the next node in the current\ntasks mems_allowed to prefer for the allocation.\n\nThis policy can provide substantial improvements for jobs that need to place\nthread local data on the corresponding node, but that need to access large\nfile system data sets that need to be spread across the several nodes in the\njobs cpuset in order to fit. Without this patch, especially for jobs that\nmight have one thread reading in the data set, the memory allocation across\nthe nodes in the jobs cpuset can become very uneven.\n\nA couple of Copyright year ranges are updated as well. And a couple of email\naddresses that can be found in the MAINTAINERS file are removed.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8a39cc60bfa5a72f32d975729a354daca124f6de", "tree": "c8a364e6fc5186293bca85152c442a4a23df4822", "parents": [ "7b5b9ef0e17d52c188fe73ea78e884fe67079e6c" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:01 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:22 2006 -0800" }, "message": "[PATCH] cpuset use combined atomic_inc_return calls\n\nReplace pairs of calls to \u003catomic_inc, atomic_read\u003e, with a single call\natomic_inc_return, saving a few bytes of source and kernel text.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7b5b9ef0e17d52c188fe73ea78e884fe67079e6c", "tree": "6f131575486edc261e20bbcdb2f73ec28d159e91", "parents": [ "0b1303fcf23678ee1785841fb0c770a35cd0833c" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri Mar 24 03:16:00 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Mar 24 07:33:22 2006 -0800" }, "message": "[PATCH] cpuset cleanup not not operators\n\nSince the test_bit() bit operator is boolean (return 0 or 1), the double not\n\"!!\" operations needed to convert a scalar (zero or not zero) to a boolean are\nnot needed.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3d3f26a7baaa921a0e790b4c72d20f0de91a5d65", "tree": "f2ad04e2954f6c8430c27d98a3b3f658b13379ee", "parents": [ "6362e4d4eda61efb04ac1cdae32e48ac6d90b701" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Mar 23 03:00:18 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Mar 23 07:38:10 2006 -0800" }, "message": "[PATCH] kernel/cpuset.c, mutex conversion\n\nconvert cpuset.c\u0027s callback_sem and manage_sem to mutexes.\nBuild and boot tested by Ingo.\nBuild, boot, unit and stress tested by pj.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "06fed33849c13af637c4d09e9ba27828fac9edd5", "tree": "0dd2fa91503250edc94ec58b23fbb3e9ad6100f4", "parents": [ "651c29a17f7ea0204dacbc2a5042d57b1c9e2e37" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Wed Feb 15 15:17:38 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Feb 15 15:32:21 2006 -0800" }, "message": "[PATCH] cpuset: oops in exit on null cpuset fix\n\nFix a latent bug in cpuset_exit() handling. If a task tried to allocate\nmemory after calling cpuset_exit(), it oops\u0027d in\ncpuset_update_task_memory_state() on a NULL cpuset pointer.\n\nSo set the exiting tasks cpuset to the root cpuset instead of to NULL.\n\nA distro kernel hit this with an added kernel package that had just such a\nhook (allocating memory) in the exit code path.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fe85a998ca64a067e58ca9240ec54a95994d78ee", "tree": "6856b5b7e8aa8724abe9103737c71ef055e158c8", "parents": [ "778116920e89b91b19d2b488a0d6fe63cf43379c" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@xenotime.net", "time": "Fri Feb 03 03:04:23 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Feb 03 08:32:06 2006 -0800" }, "message": "[PATCH] cpuset: fix sparse warning\n\nkernel/cpuset.c:644:38: warning: non-ANSI function declaration of function \u0027cpuset_update_task_memory_state\u0027\n\nSigned-off-by: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "505970b96e3b7d22177c38e03435a68376628e7a", "tree": "5508317e391961355bf3d946a6aac05bb21569eb", "parents": [ "ed68cb3676bb179768529aeb808403d57295af56" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sat Jan 14 13:21:06 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Jan 14 18:27:10 2006 -0800" }, "message": "[PATCH] cpuset oom lock fix\n\nThe problem, reported in:\n\n http://bugzilla.kernel.org/show_bug.cgi?id\u003d5859\n\nand by various other email messages and lkml posts is that the cpuset hook\nin the oom (out of memory) code can try to take a cpuset semaphore while\nholding the tasklist_lock (a spinlock).\n\nOne must not sleep while holding a spinlock.\n\nThe fix seems easy enough - move the cpuset semaphore region outside the\ntasklist_lock region.\n\nThis required a few lines of mechanism to implement. The oom code where\nthe locking needs to be changed does not have access to the cpuset locks,\nwhich are internal to kernel/cpuset.c only. So I provided a couple more\ncpuset interface routines, available to the rest of the kernel, which\nsimple take and drop the lock needed here (cpusets callback_sem).\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "858119e159384308a5dde67776691a2ebf70df0f", "tree": "f360768f999d51edc0863917ce0bf79e88c0ec4c", "parents": [ "b0a9499c3dd50d333e2aedb7e894873c58da3785" ], "author": { "name": "Arjan van de Ven", "email": "arjan@infradead.org", "time": "Sat Jan 14 13:20:43 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Jan 14 18:27:06 2006 -0800" }, "message": "[PATCH] Unlinline a bunch of other functions\n\nRemove the \"inline\" keyword from a bunch of big functions in the kernel with\nthe goal of shrinking it by 30kb to 40kb\n\nSigned-off-by: Arjan van de Ven \u003carjan@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nAcked-by: Jeff Garzik \u003cjgarzik@pobox.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1b1dcc1b57a49136f118a0f16367256ff9994a69", "tree": "b0b36d4f41d28c9d6514fb309d33c1a084d6309b", "parents": [ "794ee1baee1c26be40410233e6c20bceb2b03c08" ], "author": { "name": "Jes Sorensen", "email": "jes@sgi.com", "time": "Mon Jan 09 15:59:24 2006 -0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@hera.kernel.org", "time": "Mon Jan 09 15:59:24 2006 -0800" }, "message": "[PATCH] mutex subsystem, semaphore to mutex: VFS, -\u003ei_sem\n\nThis patch converts the inode semaphore to a mutex. I have tested it on\nXFS and compiled as much as one can consider on an ia64. Anyway your\nluck with it might be different.\n\nModified-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n\n(finished the conversion)\n\nSigned-off-by: Jes Sorensen \u003cjes@sgi.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "5160ee6fc891a9ca114be0e90fa6655647bb64b2", "tree": "35d3740a777935582af1b78238f20d2c2971ed55", "parents": [ "21b6bf143d05d77c350d9c6764ae090a877b66ea" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Sun Jan 08 01:03:32 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:58 2006 -0800" }, "message": "[PATCH] shrink dentry struct\n\nSome long time ago, dentry struct was carefully tuned so that on 32 bits\nUP, sizeof(struct dentry) was exactly 128, ie a power of 2, and a multiple\nof memory cache lines.\n\nThen RCU was added and dentry struct enlarged by two pointers, with nice\nresults for SMP, but not so good on UP, because breaking the above tuning\n(128 + 8 \u003d 136 bytes)\n\nThis patch reverts this unwanted side effect, by using an union (d_u),\nwhere d_rcu and d_child are placed so that these two fields can share their\nmemory needs.\n\nAt the time d_free() is called (and d_rcu is really used), d_child is known\nto be empty and not touched by the dentry freeing.\n\nLockless lookups only access d_name, d_parent, d_lock, d_op, d_flags (so\nthe previous content of d_child is not needed if said dentry was unhashed\nbut still accessed by a CPU because of RCU constraints)\n\nAs dentry cache easily contains millions of entries, a size reduction is\nworth the extra complexity of the ugly C union.\n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nCc: Dipankar Sarma \u003cdipankar@in.ibm.com\u003e\nCc: Maneesh Soni \u003cmaneesh@in.ibm.com\u003e\nCc: Miklos Szeredi \u003cmiklos@szeredi.hu\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@us.ibm.com\u003e\nCc: Ian Kent \u003craven@themaw.net\u003e\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Al Viro \u003cviro@ftp.linux.org.uk\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: Neil Brown \u003cneilb@cse.unsw.edu.au\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nCc: Stephen Smalley \u003csds@epoch.ncsc.mil\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "03a285f58064b8e0af08383e082e383753d9c33e", "tree": "af458f3357d4e2d01dc04d2d41a7dd7d502c4755", "parents": [ "7edc59628b2f5d6516b4677b3b56f5f056e45cd9" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:02:04 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:45 2006 -0800" }, "message": "[PATCH] cpuset: skip rcu check if task is in root cpuset\n\nFor systems that aren\u0027t using cpusets, but have them CONFIG_CPUSET enabled in\ntheir kernel (eventually this may be most distribution kernels), this patch\nremoves even the minimal rcu_read_lock() from the memory page allocation path.\n\nActually, it removes that rcu call for any task that is in the root cpuset\n(top_cpuset), which on systems not actively using cpusets, is all tasks.\n\nWe don\u0027t need the rcu check for tasks in the top_cpuset, because the\ntop_cpuset is statically allocated, so at no risk of being freed out from\nunderneath us.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7edc59628b2f5d6516b4677b3b56f5f056e45cd9", "tree": "cc0f07bc07156c6f805a72c39ae2d5655a97fede", "parents": [ "6b9c2603ce07f70de9c7a8d335ecd028e8ff11f3" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:02:03 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:45 2006 -0800" }, "message": "[PATCH] cpuset: mark number_of_cpusets read_mostly\n\nMark cpuset global \u0027number_of_cpusets\u0027 as __read_mostly.\n\nThis global is accessed everytime a zone is considered in the zonelist loops\nbeneath __alloc_pages, looking for a free memory page. If number_of_cpusets\nis just one, then we can short circuit the mems_allowed check.\n\nSince this global is read alot on a hot path, and written rarely, it is an\nexcellent candidate for __read_mostly.\n\nThanks to Christoph Lameter for the suggestion.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "6b9c2603ce07f70de9c7a8d335ecd028e8ff11f3", "tree": "38b009da71a1f93bbda8621c7d2721c18913260d", "parents": [ "c417f0242ebe578924a30d4e53d35b5059fed4e7" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:02:02 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:45 2006 -0800" }, "message": "[PATCH] cpuset: use rcu directly optimization\n\nOptimize the cpuset impact on page allocation, the most performance critical\ncpuset hook in the kernel.\n\nOn each page allocation, the cpuset hook needs to check for a possible change\nin the current tasks cpuset. It can now handle the common case, of no change,\nwithout taking any spinlock or semaphore, thanks to RCU.\n\nConvert a spinlock on the current task to an rcu_read_lock(), saving\napproximately a memory barrier and an atomic op, depending on architecture.\n\nThis is done by adding rcu_assign_pointer() and synchronize_rcu() calls to the\nwrite side of the task-\u003ecpuset pointer, in cpuset.c:attach_task(), to delay\nfreeing up a detached cpuset until after any critical sections referencing\nthat pointer.\n\nThanks to Andi Kleen, Nick Piggin and Eric Dumazet for ideas.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "c417f0242ebe578924a30d4e53d35b5059fed4e7", "tree": "3058c7c79aedb11e7013f5faca34eb07e9a761bd", "parents": [ "04c19fa6f16047abff2288ddbc1f0798ede5a849" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:02:01 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: remove test for null cpuset from alloc code path\n\nRemove a couple of more lines of code from the cpuset hooks in the page\nallocation code path.\n\nThere was a check for a NULL cpuset pointer in the routine\ncpuset_update_task_memory_state() that was only needed during system boot,\nafter the memory subsystem was initialized, before the cpuset subsystem was\ninitialized, to catch a NULL task-\u003ecpuset pointer.\n\nAdd a cpuset_init_early() routine, just before the mem_init() call in\ninit/main.c, that sets up just enough of the init tasks cpuset structure to\nrender cpuset_update_task_memory_state() calls harmless.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "04c19fa6f16047abff2288ddbc1f0798ede5a849", "tree": "4c4f109d919042b300ac907a8fde64b822faa7aa", "parents": [ "4225399a66b315d4d1fb1cb61b75dda201c832e3" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:02:00 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: migrate all tasks in cpuset at once\n\nGiven the mechanism in the previous patch to handle rebinding the per-vma\nmempolicies of all tasks in a cpuset that changes its memory placement, it is\nnow easier to handle the page migration requirements of such tasks at the same\ntime.\n\nThe previous code didn\u0027t actually attempt to migrate the pages of the tasks in\na cpuset whose memory placement changed until the next time each such task\ntried to allocate memory. This was undesirable, as users invoking memory page\nmigration exected to happen when the placement changed, not some unspecified\ntime later when the task needed more memory.\n\nIt is now trivial to handle the page migration at the same time as the per-vma\nrebinding is done.\n\nThe routine cpuset.c:update_nodemask(), which handles changing a cpusets\nmemory placement (\u0027mems\u0027) now checks for the special case of being asked to\nwrite a placement that is the same as before. It was harmless enough before\nto just recompute everything again, even though nothing had changed. But page\nmigration is a heavy weight operation - moving pages about. So now it is\nworth avoiding that if asked to move a cpuset to its current location.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "4225399a66b315d4d1fb1cb61b75dda201c832e3", "tree": "c8bd976bc6590c5fe859c6129abb93072d99cfa8", "parents": [ "202f72d5d1b5c2c084f63ef996c736d208b447b5" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:59 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: rebind vma mempolicies fix\n\nFix more of longstanding bug in cpuset/mempolicy interaction.\n\nNUMA mempolicies (mm/mempolicy.c) are constrained by the current tasks cpuset\nto just the Memory Nodes allowed by that cpuset. The kernel maintains\ninternal state for each mempolicy, tracking what nodes are used for the\nMPOL_INTERLEAVE, MPOL_BIND or MPOL_PREFERRED policies.\n\nWhen a tasks cpuset memory placement changes, whether because the cpuset\nchanged, or because the task was attached to a different cpuset, then the\ntasks mempolicies have to be rebound to the new cpuset placement, so as to\npreserve the cpuset-relative numbering of the nodes in that policy.\n\nAn earlier fix handled such mempolicy rebinding for mempolicies attached to a\ntask.\n\nThis fix rebinds mempolicies attached to vma\u0027s (address ranges in a tasks\naddress space.) Due to the need to hold the task-\u003emm-\u003emmap_sem semaphore while\nupdating vma\u0027s, the rebinding of vma mempolicies has to be done when the\ncpuset memory placement is changed, at which time mmap_sem can be safely\nacquired. The tasks mempolicy is rebound later, when the task next attempts\nto allocate memory and notices that its task-\u003ecpuset_mems_generation is\nout-of-date with its cpusets mems_generation.\n\nBecause walking the tasklist to find all tasks attached to a changing cpuset\nrequires holding tasklist_lock, a spinlock, one cannot update the vma\u0027s of the\naffected tasks while doing the tasklist scan. In general, one cannot acquire\na semaphore (which can sleep) while already holding a spinlock (such as\ntasklist_lock). So a list of mm references has to be built up during the\ntasklist scan, then the tasklist lock dropped, then for each mm, its mmap_sem\nacquired, and the vma\u0027s in that mm rebound.\n\nOnce the tasklist lock is dropped, affected tasks may fork new tasks, before\ntheir mm\u0027s are rebound. A kernel global \u0027cpuset_being_rebound\u0027 is set to\npoint to the cpuset being rebound (there can only be one; cpuset modifications\nare done under a global \u0027manage_sem\u0027 semaphore), and the mpol_copy code that\nis used to copy a tasks mempolicies during fork catches such forking tasks,\nand ensures their children are also rebound.\n\nWhen a task is moved to a different cpuset, it is easier, as there is only one\ntask involved. It\u0027s mm-\u003evma\u0027s are scanned, using the same\nmpol_rebind_policy() as used above.\n\nIt may happen that both the mpol_copy hook and the update done via the\ntasklist scan update the same mm twice. This is ok, as the mempolicies of\neach vma in an mm keep track of what mems_allowed they are relative to, and\nsafely no-op a second request to rebind to the same nodes.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "202f72d5d1b5c2c084f63ef996c736d208b447b5", "tree": "f50551f9588f9090fee17130614e17a2dd64c656", "parents": [ "74cb21553f4bf244185b9bec4c26e4e3169ad55e" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:57 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: number_of_cpusets optimization\n\nEasy little optimization hack to avoid actually having to call\ncpuset_zone_allowed() and check mems_allowed, in the main page allocation\nroutine, __alloc_pages(). This saves several CPU cycles per page allocation\non systems not using cpusets.\n\nA counter is updated each time a cpuset is created or removed, and whenever\nthere is only one cpuset in the system, it must be the root cpuset, which\ncontains all CPUs and all Memory Nodes. In that case, when the counter is\none, all allocations are allowed.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "74cb21553f4bf244185b9bec4c26e4e3169ad55e", "tree": "3f8f13e8dacc8f0876b01f62765a123ce1722b17", "parents": [ "909d75a3b77bdd8baa9429bad3b69a654d2954ce" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:56 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: numa_policy_rebind cleanup\n\nCleanup, reorganize and make more robust the mempolicy.c code to rebind\nmempolicies relative to the containing cpuset after a tasks memory placement\nchanges.\n\nThe real motivator for this cleanup patch is to lay more groundwork for the\nupcoming patch to correctly rebind NUMA mempolicies that are attached to vma\u0027s\nafter the containing cpuset memory placement changes.\n\nNUMA mempolicies are constrained by the cpuset their task is a member of.\nWhen either (1) a task is moved to a different cpuset, or (2) the \u0027mems\u0027\nmems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded\nnode numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to\nbe recalculated, relative to their new cpuset placement.\n\nThe old code used an unreliable method of determining what was the old\nmems_allowed constraining the mempolicy. It just looked at the tasks\nmems_allowed value. This sort of worked with the present code, that just\nrebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,\nreferring to the old nodes. But in an upcoming patch, the vma mempolicies\nwill be rebound as well. Then the order in which the various task and vma\nmempolicies are updated will no longer be deterministic, and one can no longer\ncount on the task-\u003emems_allowed holding the old value for as long as needed.\nIt\u0027s not even clear if the current code was guaranteed to work reliably for\ntask mempolicies.\n\nSo I added a mems_allowed field to each mempolicy, stating exactly what\nmems_allowed the policy is relative to, and updated synchronously and reliably\nanytime that the mempolicy is rebound.\n\nAlso removed a useless wrapper routine, numa_policy_rebind(), and had its\ncaller, cpuset_update_task_memory_state(), call directly to the rewritten\npolicy_rebind() routine, and made that rebind routine extern instead of\nstatic, and added a \"mpol_\" prefix to its name, making it\nmpol_rebind_policy().\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "909d75a3b77bdd8baa9429bad3b69a654d2954ce", "tree": "f9955ff697b7569fc75e5b8683d886315f34ac49", "parents": [ "cf2a473c4089aa41c26f653200673f5a4cc25047" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:55 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:44 2006 -0800" }, "message": "[PATCH] cpuset: implement cpuset_mems_allowed\n\nProvide a cpuset_mems_allowed() method, which the sys_migrate_pages() code\nneeded, to obtain the mems_allowed vector of a cpuset, and replaced the\nworkaround in sys_migrate_pages() to call this new method.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "cf2a473c4089aa41c26f653200673f5a4cc25047", "tree": "0bce21f4684a382b13e93ba5b85409cf5eab1c2c", "parents": [ "b4b2641843db124637fa3d2cb2101982035dcc82" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:54 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:43 2006 -0800" }, "message": "[PATCH] cpuset: combine refresh_mems and update_mems\n\nThe important code paths through alloc_pages_current() and alloc_page_vma(),\nby which most kernel page allocations go, both called\ncpuset_update_current_mems_allowed(), which in turn called refresh_mems().\n-Both- of these latter two routines did a tasklock, got the tasks cpuset\npointer, and checked for out of date cpuset-\u003emems_generation.\n\nThat was a silly duplication of code and waste of CPU cycles on an important\ncode path.\n\nConsolidated those two routines into a single routine, called\ncpuset_update_task_memory_state(), since it updates more than just\nmems_allowed.\n\nChanged all callers of either routine to call the new consolidated routine.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b4b2641843db124637fa3d2cb2101982035dcc82", "tree": "fe8ed223f2e1828a6d14090711bad1c864aded09", "parents": [ "59dac16fb95f09253b8086134443abeb439703cd" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:53 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:43 2006 -0800" }, "message": "[PATCH] cpuset: fork hook fix\n\nFix obscure, never seen in real life, cpuset fork race. The cpuset_fork()\ncall in fork.c was setting up the correct task-\u003ecpuset pointer after the\ntasklist_lock was dropped, which briefly exposed the newly forked process with\nan unsafe (copied from parent without locks or usage counter increment) cpuset\npointer.\n\nIn theory, that exposed cpuset pointer could have been pointing at a cpuset\nthat was already freed and removed, and in theory another task that had been\nsitting on the tasklist_lock waiting to scan the task list could have raced\ndown the entire tasklist, found our new child at the far end, and dereferenced\nthat bogus cpuset pointer.\n\nTo fix, setup up the correct cpuset pointer in the new child by calling\ncpuset_fork() before the new task is linked into the tasklist, and with that,\nadd a fork failure case, to dereference that cpuset, if the fork fails along\nthe way, after cpuset_fork() was called.\n\nHad to remove a BUG_ON() from cpuset_exit(), because it was no longer valid -\nthe call to cpuset_exit() from a failed fork would not have PF_EXITING set.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "59dac16fb95f09253b8086134443abeb439703cd", "tree": "25491bdcff6f980edfed259d7edadeb273a27ee3", "parents": [ "c5b2aff89635495064592dc90da595f8a880ee87" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:52 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:43 2006 -0800" }, "message": "[PATCH] cpuset: update_nodemask code reformat\n\nRestructure code layout of the kernel/cpuset.c update_nodemask() routine,\nremoving embedded returns and nested if\u0027s in favor of goto completion labels.\nThis is being done in anticipation of adding more logic to this routine, which\nwill favor the goto style structure.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "c5b2aff89635495064592dc90da595f8a880ee87", "tree": "2a796593bd675979bbad88ddd4de1a11f120bf62", "parents": [ "90c9cc4043fd8454d11886379f841f70e176698e" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:51 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:43 2006 -0800" }, "message": "[PATCH] cpuset: minor spacing initializer fixes\n\nFour trivial cpuset fixes: remove extra spaces, remove useless initializers,\nmark one __read_mostly.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3e0d98b9f1eb757fc98efc84e74e54a08308aa73", "tree": "7cf1c75994f734ede7ec89373de640c4a58b237a", "parents": [ "5966514db662fb24c9bb43226a80106bcffd51f8" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:49 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:42 2006 -0800" }, "message": "[PATCH] cpuset: memory pressure meter\n\nProvide a simple per-cpuset metric of memory pressure, tracking the -rate-\nthat the tasks in a cpuset call try_to_free_pages(), the synchronous\n(direct) memory reclaim code.\n\nThis enables batch managers monitoring jobs running in dedicated cpusets to\nefficiently detect what level of memory pressure that job is causing.\n\nThis is useful both on tightly managed systems running a wide mix of\nsubmitted jobs, which may choose to terminate or reprioritize jobs that are\ntrying to use more memory than allowed on the nodes assigned them, and with\ntightly coupled, long running, massively parallel scientific computing jobs\nthat will dramatically fail to meet required performance goals if they\nstart to use more memory than allowed to them.\n\nThis patch just provides a very economical way for the batch manager to\nmonitor a cpuset for signs of memory pressure. It\u0027s up to the batch\nmanager or other user code to decide what to do about it and take action.\n\n\u003d\u003d\u003e Unless this feature is enabled by writing \"1\" to the special file\n /dev/cpuset/memory_pressure_enabled, the hook in the rebalance\n code of __alloc_pages() for this metric reduces to simply noticing\n that the cpuset_memory_pressure_enabled flag is zero. So only\n systems that enable this feature will compute the metric.\n\nWhy a per-cpuset, running average:\n\n Because this meter is per-cpuset, rather than per-task or mm, the\n system load imposed by a batch scheduler monitoring this metric is\n sharply reduced on large systems, because a scan of the tasklist can be\n avoided on each set of queries.\n\n Because this meter is a running average, instead of an accumulating\n counter, a batch scheduler can detect memory pressure with a single\n read, instead of having to read and accumulate results for a period of\n time.\n\n Because this meter is per-cpuset rather than per-task or mm, the\n batch scheduler can obtain the key information, memory pressure in a\n cpuset, with a single read, rather than having to query and accumulate\n results over all the (dynamically changing) set of tasks in the cpuset.\n\nA per-cpuset simple digital filter (requires a spinlock and 3 words of data\nper-cpuset) is kept, and updated by any task attached to that cpuset, if it\nenters the synchronous (direct) page reclaim code.\n\nA per-cpuset file provides an integer number representing the recent\n(half-life of 10 seconds) rate of direct page reclaims caused by the tasks\nin the cpuset, in units of reclaims attempted per second, times 1000.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "5966514db662fb24c9bb43226a80106bcffd51f8", "tree": "9c6d8f4f6fee0d6574de7e225141d37b28811dc3", "parents": [ "96b7f34143c2c823a6a750fcb758fc66c44945d2" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:01:47 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:13:42 2006 -0800" }, "message": "[PATCH] cpuset: mempolicy one more nodemask conversion\n\nFinish converting mm/mempolicy.c from bitmaps to nodemasks. The previous\nconversion had left one routine using bitmaps, since it involved a\ncorresponding change to kernel/cpuset.c\n\nFix that interface by replacing with a simple macro that calls nodes_subset(),\nor if !CONFIG_CPUSET, returns (1).\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Christoph Lameter \u003cchristoph@lameter.com\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "45b07ef31d1182d2cfde7711327e3afb268bb1ac", "tree": "3bf820531f920b43d4ed963643b1f565c82bcaa4", "parents": [ "d0d963281ccb22e6f339bfdd75c6b2e31351929f" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Jan 08 01:00:56 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:12:43 2006 -0800" }, "message": "[PATCH] cpusets: swap migration interface\n\nAdd a boolean \"memory_migrate\" to each cpuset, represented by a file\ncontaining \"0\" or \"1\" in each directory below /dev/cpuset.\n\nIt defaults to false (file contains \"0\"). It can be set true by writing\n\"1\" to the file.\n\nIf true, then anytime that a task is attached to the cpuset so marked, the\npages of that task will be moved to that cpuset, preserving, to the extent\npractical, the cpuset-relative placement of the pages.\n\nAlso anytime that a cpuset so marked has its memory placement changed (by\nwriting to its \"mems\" file), the tasks in that cpuset will have their pages\nmoved to the cpusets new nodes, preserving, to the extent practical, the\ncpuset-relative placement of the moved pages.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Christoph Lameter \u003cchristoph@lameter.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "5563e77078d85c4f107a0a673500c43ce57cf702", "tree": "b9cacaa2f6ad595d6ade9e69f8a84d7c9b063048", "parents": [ "3c8d61bcf2d762fb84dbf741df400c833cada18a" ], "author": { "name": "Bob Picco", "email": "bob.picco@hp.com", "time": "Sun Nov 13 16:06:35 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Nov 13 18:14:11 2005 -0800" }, "message": "[PATCH] cpuset: fix return without releasing semaphore\n\nIt is wrong to acquire the semaphore and then return from\ncpuset_zone_allowed without releasing it.\n\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "68860ec10bcc07ab4f89f9d940e3b77ae5ca13b3", "tree": "a411f9623a113d72f23da38cf2de51f66c177873", "parents": [ "fb5eeeee44edb248b4837416966f19731f497f79" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Oct 30 15:02:36 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:22 2005 -0800" }, "message": "[PATCH] cpusets: automatic numa mempolicy rebinding\n\nThis patch automatically updates a tasks NUMA mempolicy when its cpuset\nmemory placement changes. It does so within the context of the task,\nwithout any need to support low level external mempolicy manipulation.\n\nIf a system is not using cpusets, or if running on a system with just the\nroot (all-encompassing) cpuset, then this remap is a no-op. Only when a\ntask is moved between cpusets, or a cpusets memory placement is changed\ndoes the following apply. Otherwise, the main routine below,\nrebind_policy() is not even called.\n\nWhen mixing cpusets, scheduler affinity, and NUMA mempolicies, the\nessential role of cpusets is to place jobs (several related tasks) on a set\nof CPUs and Memory Nodes, the essential role of sched_setaffinity is to\nmanage a jobs processor placement within its allowed cpuset, and the\nessential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs\nmemory placement within its allowed cpuset.\n\nHowever, CPU affinity and NUMA memory placement are managed within the\nkernel using absolute system wide numbering, not cpuset relative numbering.\n\nThis is ok until a job is migrated to a different cpuset, or what\u0027s the\nsame, a jobs cpuset is moved to different CPUs and Memory Nodes.\n\nThen the CPU affinity and NUMA memory placement of the tasks in the job\nneed to be updated, to preserve their cpuset-relative position. This can\nbe done for CPU affinity using sched_setaffinity() from user code, as one\ntask can modify anothers CPU affinity. This cannot be done from an\nexternal task for NUMA memory placement, as that can only be modified in\nthe context of the task using it.\n\nHowever, it easy enough to remap a tasks NUMA mempolicy automatically when\na task is migrated, using the existing cpuset mechanism to trigger a\nrefresh of a tasks memory placement after its cpuset has changed. All that\nis needed is the old and new nodemask, and notice to the task that it needs\nto rebind its mempolicy. The tasks mems_allowed has the old mask, the\ntasks cpuset has the new mask, and the existing\ncpuset_update_current_mems_allowed() mechanism provides the notice. The\nbitmap/cpumask/nodemask remap operators provide the cpuset relative\ncalculations.\n\nThis patch leaves open a couple of issues:\n\n 1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:\n\n These mempolicies may reference nodes outside of those allowed to\n the current task by its cpuset. Tasks are migrated as part of jobs,\n which reside on what might be several cpusets in a subtree. When such\n a job is migrated, all NUMA memory policy references to nodes within\n that cpuset subtree should be translated, and references to any nodes\n outside that subtree should be left untouched. A future patch will\n provide the cpuset mechanism needed to mark such subtrees. With that\n patch, we will be able to correctly migrate these other memory policies\n across a job migration.\n\n 2) Updating cpuset, affinity and memory policies in user space:\n\n This is harder. Any placement state stored in user space using\n system-wide numbering will be invalidated across a migration. More\n work will be required to provide user code with a migration-safe means\n to manage its cpuset relative placement, while preserving the current\n API\u0027s that pass system wide numbers, not cpuset relative numbers across\n the kernel-user boundary.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "18a19cb3047e454ee5ecbc35d7acf3f8e09e0466", "tree": "ad91d0024d886bdd207ba72e875e736e833de616", "parents": [ "053199edf54f685e7dea765b60d4d5e9070dadec" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Oct 30 15:02:31 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:21 2005 -0800" }, "message": "[PATCH] cpusets: simple rename\n\nAdd support for renaming cpusets. Only allow simple rename of cpuset\ndirectories in place. Don\u0027t allow moving cpusets elsewhere in hierarchy or\nrenaming the special cpuset files in each cpuset directory.\n\nThe usefulness of this simple rename became apparent when developing task\nmigration facilities. It allows building a second cpuset hierarchy using\nnew names and containing new CPUs and Memory Nodes, moving tasks from the\nold to the new cpusets, removing the old cpusets, and then renaming the new\ncpusets to be just like the old names, so that any knowledge that the tasks\nhad of their cpuset names will still be valid.\n\nLeaf node cpusets can be migrated to other CPUs or Memory Nodes by just\nupdating their \u0027cpus\u0027 and \u0027mems\u0027 files, but because no cpuset can contain\nCPUs or Nodes not in its parent cpuset, one cannot do this in a cpuset\nhierarchy without first expanding all the non-leaf cpusets to contain the\nunion of both the old and new CPUs and Nodes, which would obfuscate the\none-to-one migration of a task from one cpuset to another required to\ncorrectly migrate the physical page frames currently allocated to that\ntask.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "053199edf54f685e7dea765b60d4d5e9070dadec", "tree": "a2d12a8b7f07b59048da992e7ae9405bc4ee292b", "parents": [ "5aa15b5f27fc2c404530c6c8eabdb8437deb3163" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Oct 30 15:02:30 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:21 2005 -0800" }, "message": "[PATCH] cpusets: dual semaphore locking overhaul\n\nOverhaul cpuset locking. Replace single semaphore with two semaphores.\n\nThe suggestion to use two locks was made by Roman Zippel.\n\nBoth locks are global. Code that wants to modify cpusets must first\nacquire the exclusive manage_sem, which allows them read-only access to\ncpusets, and holds off other would-be modifiers. Before making actual\nchanges, the second semaphore, callback_sem must be acquired as well. Code\nthat needs only to query cpusets must acquire callback_sem, which is also a\nglobal exclusive lock.\n\nThe earlier problems with double tripping are avoided, because it is\nallowed for holders of manage_sem to nest the second callback_sem lock, and\nonly callback_sem is needed by code called from within __alloc_pages(),\nwhere the double tripping had been possible.\n\nThis is not quite the same as a normal read/write semaphore, because\nobtaining read-only access with intent to change must hold off other such\nattempts, while allowing read-only access w/o such intention. Changing\ncpusets involves several related checks and changes, which must be done\nwhile allowing read-only queries (to avoid the double trip), but while\nensuring nothing changes (holding off other would be modifiers.)\n\nThis overhaul of cpuset locking also makes careful use of task_lock() to\nguard access to the task-\u003ecpuset pointer, closing a couple of race\nconditions noticed while reading this code (thanks, Roman). I\u0027ve never\nseen these races fail in any use or test.\n\nSee further the comments in the code.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "5aa15b5f27fc2c404530c6c8eabdb8437deb3163", "tree": "914f0b33f5190bd0183cde2e9f6da552d3d1d7aa", "parents": [ "f35f31d7ed0150f9865619f21b5050c91b46c03f" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Oct 30 15:02:28 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:21 2005 -0800" }, "message": "[PATCH] cpusets: remove depth counted locking hack\n\nRemove a rather hackish depth counter on cpuset locking. The depth counter\nwas avoiding a possible double trip on the global cpuset_sem semaphore. It\nworked, but now an improved version of cpuset locking is available, to come\nin the next patch, using two global semaphores.\n\nThis patch reverses \"cpuset semaphore depth check deadlock fix\"\n\nThe kernel still works, even after this patch, except for some rare and\ndifficult to reproduce race conditions when agressively creating and\ndestroying cpusets marked with the notify_on_release option, on very large\nsystems.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f35f31d7ed0150f9865619f21b5050c91b46c03f", "tree": "fb33df4eb302de0d249682e77d3d07bd6f6611ba", "parents": [ "e9543659715602e3180f00a227bb6db34141ac41" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sun Oct 30 15:02:27 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:21 2005 -0800" }, "message": "[PATCH] cpuset cleanup\n\nRemove one more useless line from cpuset_common_file_read().\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "dd0fc66fb33cd610bc1a5db8a5e232d34879b4d7", "tree": "51f96a9db96293b352e358f66032e1f4ff79fafb", "parents": [ "3b0e77bd144203a507eb191f7117d2c5004ea1de" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Fri Oct 07 07:46:04 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 08 15:00:57 2005 -0700" }, "message": "[PATCH] gfp flags annotations - part 1\n\n - added typedef unsigned int __nocast gfp_t;\n\n - replaced __nocast uses for gfp flags with gfp_t - it gives exactly\n the same warnings as far as sparse is concerned, doesn\u0027t change\n generated code (from gcc point of view we replaced unsigned int with\n typedef) and documents what\u0027s going on far better.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "eacaa1f5aa4a41a48349f55abcd9258506943e76", "tree": "6b0e7e82f0ba4144639bb30eeb82dc8588cfd3f5", "parents": [ "46d7031ecb8a8360b0022abd8014f38cc1197166" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Fri Sep 30 03:26:43 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Sep 30 08:42:24 2005 -0700" }, "message": "[PATCH] cpuset crapectomy\n\nSwitched cpuset_common_file_read() to simple_read_from_buffer(), killed\na bunch of useless (and not quite correct - e.g. min(size_t,ssize_t))\ncode.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "5134fc15b643dc36eb9aa77e4318b886844a9ac5", "tree": "170339651303da0bb530c407300f09d3cd39caa2", "parents": [ "2dd3c1df95fb29e9227f16ccd7d786d129e2b34d" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Wed Sep 28 06:42:24 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Sep 28 07:58:51 2005 -0700" }, "message": "[PATCH] cpuset read past eof memory leak fix\n\nDon\u0027t leak a page of memory if user reads a cpuset file past eof.\n\nSigned-off-by: KUROSAWA Takahiro \u003ckurosawa@valinux.co.jp\u003e\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b3426599af9524104be6938bcb1fcaab314781c7", "tree": "c6d354bddb5b8cd298d139b60a9257ebd8323b90", "parents": [ "f24ec7f6c6278c0ea4c00efe96d50b1e66796c44" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Mon Sep 12 04:30:30 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Sep 12 09:16:27 2005 -0700" }, "message": "[PATCH] cpuset semaphore depth check optimize\n\nOptimize the deadlock avoidance check on the global cpuset\nsemaphore cpuset_sem. Instead of adding a depth counter to the\ntask struct of each task, rather just two words are enough, one\nto store the depth and the other the current cpuset_sem holder.\n\nThanks to Nikita Danilov for the idea.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\n\n[ We may want to change this further, but at least it\u0027s now\n a totally internal decision to the cpusets code ]\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "4247bdc60048018b98f71228b45cfbc5f5270c86", "tree": "6f6abbd10685af84c97e661da6771726a12209ac", "parents": [ "fb1c8f93d869b34cacb8b8932e2b83d96a19d720" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Sat Sep 10 00:26:06 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Sep 10 10:06:21 2005 -0700" }, "message": "[PATCH] cpuset semaphore depth check deadlock fix\n\nThe cpusets-formalize-intermediate-gfp_kernel-containment patch\nhas a deadlock problem.\n\nThis patch was part of a set of four patches to make more\nextensive use of the cpuset \u0027mem_exclusive\u0027 attribute to\nmanage kernel GFP_KERNEL memory allocations and to constrain\nthe out-of-memory (oom) killer.\n\nA task that is changing cpusets in particular ways on a system\nwhen it is very short of free memory could double trip over\nthe global cpuset_sem semaphore (get the lock and then deadlock\ntrying to get it again).\n\nThe second attempt to get cpuset_sem would be in the routine\ncpuset_zone_allowed(). This was discovered by code inspection.\nI can not reproduce the problem except with an artifically\nhacked kernel and a specialized stress test.\n\nIn real life you cannot hit this unless you are manipulating\ncpusets, and are very unlikely to hit it unless you are rapidly\nmodifying cpusets on a memory tight system. Even then it would\nbe a rare occurence.\n\nIf you did hit it, the task double tripping over cpuset_sem\nwould deadlock in the kernel, and any other task also trying\nto manipulate cpusets would deadlock there too, on cpuset_sem.\nYour batch manager would be wedged solid (if it was cpuset\nsavvy), but classic Unix shells and utilities would work well\nenough to reboot the system.\n\nThe unusual condition that led to this bug is that unlike most\nsemaphores, cpuset_sem _can_ be acquired while in the page\nallocation code, when __alloc_pages() calls cpuset_zone_allowed.\nSo it easy to mistakenly perform the following sequence:\n 1) task makes system call to alter a cpuset\n 2) take cpuset_sem\n 3) try to allocate memory\n 4) memory allocator, via cpuset_zone_allowed, trys to take cpuset_sem\n 5) deadlock\n\nThe reason that this is not a serious bug for most users\nis that almost all calls to allocate memory don\u0027t require\ntaking cpuset_sem. Only some code paths off the beaten\ntrack require taking cpuset_sem -- which is good. Taking\na global semaphore on the main code path for allocating\nmemory would not scale well.\n\nThis patch fixes this deadlock by wrapping the up() and down()\ncalls on cpuset_sem in kernel/cpuset.c with code that tracks\nthe nesting depth of the current task on that semaphore, and\nonly does the real down() if the task doesn\u0027t hold the lock\nalready, and only does the real up() if the nesting depth\n(number of unmatched downs) is exactly one.\n\nThe previous required use of refresh_mems(), anytime that\nthe cpuset_sem semaphore was acquired and the code executed\nwhile holding that semaphore might try to allocate memory, is\nno longer required. Two refresh_mems() calls were removed\nthanks to this. This is a good change, as failing to get\nall the necessary refresh_mems() calls placed was a primary\nsource of bugs in this cpuset code. The only remaining call\nto refresh_mems() is made while doing a memory allocation,\nif certain task memory placement data needs to be updated\nfrom its cpuset, due to the cpuset having been changed behind\nthe tasks back.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "73a358d1892a8233801e3fd54668075b52ec42da", "tree": "851250c42eb890c6fa3afe59b3398e17ee850bae", "parents": [ "ac0b1bc1edbe81c0cb36cad7e7f5b91f4d9e12ed" ], "author": { "name": "KUROSAWA Takahiro", "email": "kurosawa@valinux.co.jp", "time": "Fri Sep 09 13:02:10 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Sep 09 13:57:32 2005 -0700" }, "message": "[PATCH] fix for cpusets minor problem\n\nThis patch fixes minor problem that the CPUSETS have when files in the\ncpuset filesystem are read after lseek()-ed beyond the EOF.\n\nSigned-off-by: KUROSAWA Takahiro \u003ckurosawa@valinux.co.jp\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "0811bab24ff1eecab38110eda7ea7847db95c64e", "tree": "1de9a8a04c8c3dec7f9cdf46f2db6b6609757127", "parents": [ "d1b551386a5f3f50a5003b691f819b07f8e6f034" ], "author": { "name": "John Hawkes", "email": "hawkes@sgi.com", "time": "Tue Sep 06 15:18:15 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Sep 07 16:57:41 2005 -0700" }, "message": "[PATCH] cpusets: re-enable \"dynamic sched domains\"\n\nRevert the hack introduced last week.\n\nSigned-off-by: John Hawkes \u003chawkes@sgi.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "ef08e3b4981aebf2ba9bd7025ef7210e8eec07ce", "tree": "3b5386e011c87dde384115c8eb0d6961c2536025", "parents": [ "9bf2229f8817677127a60c177aefce1badd22d7b" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Tue Sep 06 15:18:13 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Sep 07 16:57:40 2005 -0700" }, "message": "[PATCH] cpusets: confine oom_killer to mem_exclusive cpuset\n\nNow the real motivation for this cpuset mem_exclusive patch series seems\ntrivial.\n\nThis patch keeps a task in or under one mem_exclusive cpuset from provoking an\noom kill of a task under a non-overlapping mem_exclusive cpuset. Since only\ninterrupt and GFP_ATOMIC allocations are allowed to escape mem_exclusive\ncontainment, there is little to gain from oom killing a task under a\nnon-overlapping mem_exclusive cpuset, as almost all kernel and user memory\nallocation must come from disjoint memory nodes.\n\nThis patch enables configuring a system so that a runaway job under one\nmem_exclusive cpuset cannot cause the killing of a job in another such cpuset\nthat might be using very high compute and memory resources for a prolonged\ntime.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "9bf2229f8817677127a60c177aefce1badd22d7b", "tree": "06e95863a26b197233081db1dafd869dfd231950", "parents": [ "f90b1d2f1aaaa40c6519a32e69615edc25bb97d5" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Tue Sep 06 15:18:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Sep 07 16:57:40 2005 -0700" }, "message": "[PATCH] cpusets: formalize intermediate GFP_KERNEL containment\n\nThis patch makes use of the previously underutilized cpuset flag\n\u0027mem_exclusive\u0027 to provide what amounts to another layer of memory placement\nresolution. With this patch, there are now the following four layers of\nmemory placement available:\n\n 1) The whole system (interrupt and GFP_ATOMIC allocations can use this),\n 2) The nearest enclosing mem_exclusive cpuset (GFP_KERNEL allocations can use),\n 3) The current tasks cpuset (GFP_USER allocations constrained to here), and\n 4) Specific node placement, using mbind and set_mempolicy.\n\nThese nest - each layer is a subset (same or within) of the previous.\n\nLayer (2) above is new, with this patch. The call used to check whether a\nzone (its node, actually) is in a cpuset (in its mems_allowed, actually) is\nextended to take a gfp_mask argument, and its logic is extended, in the case\nthat __GFP_HARDWALL is not set in the flag bits, to look up the cpuset\nhierarchy for the nearest enclosing mem_exclusive cpuset, to determine if\nplacement is allowed. The definition of GFP_USER, which used to be identical\nto GFP_KERNEL, is changed to also set the __GFP_HARDWALL bit, in the previous\ncpuset_gfp_hardwall_flag patch.\n\nGFP_ATOMIC and GFP_KERNEL allocations will stay within the current tasks\ncpuset, so long as any node therein is not too tight on memory, but will\nescape to the larger layer, if need be.\n\nThe intended use is to allow something like a batch manager to handle several\njobs, each job in its own cpuset, but using common kernel memory for caches\nand such. Swapper and oom_kill activity is also constrained to Layer (2). A\ntask in or below one mem_exclusive cpuset should not cause swapping on nodes\nin another non-overlapping mem_exclusive cpuset, nor provoke oom_killing of a\ntask in another such cpuset. Heavy use of kernel memory for i/o caching and\nsuch by one job should not impact the memory available to jobs in other\nnon-overlapping mem_exclusive cpusets.\n\nThis patch enables providing hardwall, inescapable cpusets for memory\nallocations of each job, while sharing kernel memory allocations between\nseveral jobs, in an enclosing mem_exclusive cpuset.\n\nLike Dinakar\u0027s patch earlier to enable administering sched domains using the\ncpu_exclusive flag, this patch also provides a useful meaning to a cpuset flag\nthat had previously done nothing much useful other than restrict what cpuset\nconfigurations were allowed.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "212d6d2237f60bc28c1518f8abf9d3ed6c17574a", "tree": "693ac9cdb0693acb0f935d4b61a34e6a12056add", "parents": [ "ca2f3daf779f5e89d14e9783fcfd7920842df9e9" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Thu Aug 25 12:47:56 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Aug 26 16:38:47 2005 -0700" }, "message": "[PATCH] completely disable cpu_exclusive sched domain\n\nAt the suggestion of Nick Piggin and Dinakar, totally disable\nthe facility to allow cpu_exclusive cpusets to define dynamic\nsched domains in Linux 2.6.13, in order to avoid problems\nfirst reported by John Hawkes (corrupt sched data structures\nand kernel oops).\n\nThis has been built for ppc64, i386, ia64, x86_64, sparc, alpha.\nIt has been built, booted and tested for cpuset functionality\non an SN2 (ia64).\n\nDinakar or Nick - could you verify that it for sure does avoid\nthe problems Hawkes reported. Hawkes is out of town, and I don\u0027t\nhave the recipe to reproduce what he found.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "ca2f3daf779f5e89d14e9783fcfd7920842df9e9", "tree": "07ce4cc60957d842dac1cb3d44dd5441071cf90f", "parents": [ "13142341ac867bb67e88204cbfcb8d90f9a861b7" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Thu Aug 25 12:47:50 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Aug 26 16:38:46 2005 -0700" }, "message": "[PATCH] undo partial cpu_exclusive sched domain disabling\n\nThe partial disabling of Dinakar\u0027s new facility to allow\ncpu_exclusive cpusets to define dynamic sched domains\ndoesn\u0027t go far enough. At the suggestion of Nick Piggin\nand Dinakar, let us instead totally disable this facility\nfor 2.6.13, in order to avoid problems first reported\nby John Hawkes (corrupt sched data structures and kernel oops).\n\nThis patch removes the partial disabling code in 2.6.13-rc7,\nin anticipation of the next patch, which will totally disable\nit instead.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3725822f7c7134249addcd4549aff086950c8090", "tree": "d7db18d3d5f75fe4309ddc7aa373f3213f845b41", "parents": [ "40bb0c3ef52d872de348e10000eb5432a43a147d" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Wed Aug 24 04:15:10 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Aug 24 09:40:45 2005 -0700" }, "message": "[PATCH] cpu_exclusive sched domains build fix\n\nAs reported by Paul Mackerras \u003cpaulus@samba.org\u003e, the previous patch\n\"cpu_exclusive sched domains fix\" broke the ppc64 build with\nCONFIC_CPUSET, yielding error messages:\n\nkernel/cpuset.c: In function \u0027update_cpu_domains\u0027:\nkernel/cpuset.c:648: error: invalid lvalue in unary \u0027\u0026\u0027\nkernel/cpuset.c:648: error: invalid lvalue in unary \u0027\u0026\u0027\n\nOn some arch\u0027s, the node_to_cpumask() is a function, returning\na cpumask_t. But the for_each_cpu_mask() requires an lvalue mask.\n\nThe following patch fixes this build failure by making a copy\nof the cpumask_t on the stack.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d10689b68aff7b48e3de1a3f7fcd6567bd2905af", "tree": "c81c261274011d301dfbcfd1a3e13480b93c167e", "parents": [ "ae75784bc576a1af70509c2f3ba2b70bb65a0c58" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Tue Aug 23 01:04:27 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Aug 23 20:02:52 2005 -0700" }, "message": "[PATCH] cpu_exclusive sched domains on partial nodes temp fix\n\nThis keeps the kernel/cpuset.c routine update_cpu_domains() from\ninvoking the sched.c routine partition_sched_domains() if the cpuset in\nquestion doesn\u0027t fall on node boundaries.\n\nI have boot tested this on an SN2, and with the help of a couple of ad\nhoc printk\u0027s, determined that it does indeed avoid calling the\npartition_sched_domains() routine on partial nodes.\n\nI did not directly verify that this avoids setting up bogus sched\ndomains or avoids the oops that Hawkes saw.\n\nThis patch imposes a silent artificial constraint on which cpusets can\nbe used to define dynamic sched domains.\n\nThis patch should allow proceeding with this new feature in 2.6.13 for\nthe configurations in which it is useful (node alligned sched domains)\nwhile avoiding trying to setup sched domains in the less useful cases\nthat can cause the kernel corruption and oops.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nAcked-by: Dinakar Guniguntala \u003cdino@in.ibm.com\u003e\nAcked-by: John Hawkes \u003chawkes@sgi.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3077a260e9f316b611436b1506eec9cc5c4f8aa6", "tree": "43b7d5faa5f204904c713c463015792d9ff56b01", "parents": [ "a242b44da6feb604c4c659b78f63dedb69b2d4a3" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Tue Aug 09 10:07:59 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Aug 09 12:08:22 2005 -0700" }, "message": "[PATCH] cpuset release ABBA deadlock fix\n\nFix possible cpuset_sem ABBA deadlock if \u0027notify_on_release\u0027 set.\n\nFor a particular usage pattern, creating and destroying cpusets fairly\nfrequently using notify_on_release, on a very large system, this deadlock\ncan be seen every few days. If you are not using the cpuset\nnotify_on_release feature, you will never see this deadlock.\n\nThe existing code, on task exit (or cpuset deletion) did:\n\n get cpuset_sem\n if cpuset marked notify_on_release and is ready to release:\n compute cpuset path relative to /dev/cpuset mount point\n call_usermodehelper() forks /sbin/cpuset_release_agent with path\n drop cpuset_sem\n\nUnfortunately, the fork in call_usermodehelper can allocate memory, and\nallocating memory can require cpuset_sem, if the mems_generation values\nchanged in the interim. This results in an ABBA deadlock, trying to obtain\ncpuset_sem when it is already held by the current task.\n\nTo fix this, I put the cpuset path (which must be computed while holding\ncpuset_sem) in a temporary buffer, to be used in the call_usermodehelper\ncall of /sbin/cpuset_release_agent only _after_ dropping cpuset_sem.\n\nSo the new logic is:\n\n get cpuset_sem\n if cpuset marked notify_on_release and is ready to release:\n compute cpuset path relative to /dev/cpuset mount point\n stash path in kmalloc\u0027d buffer\n drop cpuset_sem\n call_usermodehelper() forks /sbin/cpuset_release_agent with path\n free path\n\nThe sharp eyed reader might notice that this patch does not contain any\ncalls to kmalloc. The existing code in the check_for_release() routine was\nalready kmalloc\u0027ing a buffer to hold the cpuset path. In the old code, it\njust held the buffer for a few lines, over the cpuset_release_agent() call\nthat in turn invoked call_usermodehelper(). In the new code, with the\napplication of this patch, it returns that buffer via the new char\n**ppathbuf parameter, for later use and freeing in cpuset_release_agent(),\nwhich is called after cpuset_sem is dropped. Whereas the old code has just\none call to cpuset_release_agent(), right in the check_for_release()\nroutine, the new code has three calls to cpuset_release_agent(), from the\nvarious places that a cpuset can be released.\n\nThis patch has been build and booted on SN2, and passed a stress test that\npreviously hit the deadlock within a few seconds.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d9fd8a6d443b509147280f058d4e59f0b796a323", "tree": "427385e403fc0d4282be19732a00ddfd224c1461", "parents": [ "207a7ba8dc000e1b13acac97f3736810dd86e8e2" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@xenotime.net", "time": "Wed Jul 27 11:45:11 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Jul 27 16:26:06 2005 -0700" }, "message": "[PATCH] kernel/cpuset.c: add kerneldoc, fix typos\n\nAdd kerneldoc to kernel/cpuset.c\n\nFix cpuset typos in init/Kconfig\n\nSigned-off-by: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "85d7b94981e2e919697bc235aad7367b33c3864b", "tree": "79d7705a897abde11b5f2f967bf24487aaeea354", "parents": [ "1a20ff27ef75d866730ee796acd811a925af762f" ], "author": { "name": "Dinakar Guniguntala", "email": "dino@in.ibm.com", "time": "Sat Jun 25 14:57:34 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Jun 25 16:24:45 2005 -0700" }, "message": "[PATCH] Dynamic sched domains: cpuset changes\n\nAdds the core update_cpu_domains code and updated cpusets documentation\n\nSigned-off-by: Dinakar Guniguntala \u003cdino@in.ibm.com\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "5f45f1a78fbac3cc859ec10c5366e97d20d40fa2", "tree": "bafacf11adbfb4eebf2d2d80d92e1fe2d9c02b18", "parents": [ "ac20427ef6aa63da663bdc88b71d16f7394f5e23" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Thu Jun 23 00:09:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:20 2005 -0700" }, "message": "[PATCH] remove duplicate get_dentry functions in various places\n\nVarious filesystem drivers have grown a get_dentry() function that\u0027s a\nduplicate of lookup_one_len, except that it doesn\u0027t take a maximum length\nargument and doesn\u0027t check for \\0 or / in the passed in filename.\n\nSwitch all these places to use lookup_one_len.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Greg KH \u003cgreg@kroah.com\u003e\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "2efe86b809d97debaaf9fcc13b041aedf15bd3d2", "tree": "87e039397918f4c5b0a21d798589a8ce517bfa2d", "parents": [ "88c1834633341bbb94e315433067496338bff4ad" ], "author": { "name": "Paul Jackson", "email": "pj@sgi.com", "time": "Fri May 27 02:02:43 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Fri May 27 08:07:26 2005 -0700" }, "message": "[PATCH] cpuset exit NULL dereference fix\n\nThere is a race in the kernel cpuset code, between the code\nto handle notify_on_release, and the code to remove a cpuset.\nThe notify_on_release code can end up trying to access a\ncpuset that has been removed. In the most common case, this\ncauses a NULL pointer dereference from the routine cpuset_path.\nHowever all manner of bad things are possible, in theory at least.\n\nThe existing code decrements the cpuset use count, and if the\ncount goes to zero, processes the notify_on_release request,\nif appropriate. However, once the count goes to zero, unless we\nare holding the global cpuset_sem semaphore, there is nothing to\nstop another task from immediately removing the cpuset entirely,\nand recycling its memory.\n\nThe obvious fix would be to always hold the cpuset_sem\nsemaphore while decrementing the use count and dealing with\nnotify_on_release. However we don\u0027t want to force a global\nsemaphore into the mainline task exit path, as that might create\na scaling problem.\n\nThe actual fix is almost as easy - since this is only an issue\nfor cpusets using notify_on_release, which the top level big\ncpusets don\u0027t normally need to use, only take the cpuset_sem\nfor cpusets using notify_on_release.\n\nThis code has been run for hours without a hiccup, while running\na cpuset create/destroy stress test that could crash the existing\nkernel in seconds. This patch applies to the current -linus\ngit kernel.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Simon Derr \u003csimon.derr@bull.net\u003e\nAcked-by: Dinakar Guniguntala \u003cdino@in.ibm.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "9a8488965dc4c42a4a1f84cab907c7d6c5cf1563", "tree": "58581a02cc06bb1a2991209c9e4d559353cbec6f", "parents": [ "b52402c783d8c16b11f146a244bb21086a94bf84" ], "author": { "name": "Benoit Boissinot", "email": "benoit.boissinot@ens-lyon.org", "time": "Sat Apr 16 15:25:59 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:25:59 2005 -0700" }, "message": "[PATCH] cpuset: remove function attribute const\n\ngcc-4 warns with\ninclude/linux/cpuset.h:21: warning: type qualifiers ignored on function\nreturn type\n\ncpuset_cpus_allowed is declared with const\nextern const cpumask_t cpuset_cpus_allowed(const struct task_struct *p);\n\nFirst const should be __attribute__((const)), but the gcc manual\nexplains that:\n\n\"Note that a function that has pointer arguments and examines the data\npointed to must not be declared const. Likewise, a function that calls a\nnon-const function usually must not be const. It does not make sense for\na const function to return void.\"\n\nThe following patch remove const from the function declaration.\n\nSigned-off-by: Benoit Boissinot \u003cbenoit.boissinot@ens-lyon.org\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2", "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d", "parents": [], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n" } ] }