)]}' { "log": [ { "commit": "879669961b11e7f40b518784863a259f735a72bf", "tree": "9bff5392e365caf656c9dd9be38f7471c182278c", "parents": [ "eb96c925152fc289311e5d7e956b919e9b60ab53" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Jun 17 11:25:59 2011 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jun 17 09:40:48 2011 -0700" }, "message": "KEYS/DNS: Fix ____call_usermodehelper() to not lose the session keyring\n\n____call_usermodehelper() now erases any credentials set by the\nsubprocess_inf::init() function. The problem is that commit\n17f60a7da150 (\"capabilites: allow the application of capability limits\nto usermode helpers\") creates and commits new credentials with\nprepare_kernel_cred() after the call to the init() function. This wipes\nall keyrings after umh_keys_init() is called.\n\nThe best way to deal with this is to put the init() call just prior to\nthe commit_creds() call, and pass the cred pointer to init(). That\nmeans that umh_keys_init() and suchlike can modify the credentials\n_before_ they are published and potentially in use by the rest of the\nsystem.\n\nThis prevents request_key() from working as it is prevented from passing\nthe session keyring it set up with the authorisation token to\n/sbin/request-key, and so the latter can\u0027t assume the authority to\ninstantiate the key. This causes the in-kernel DNS resolver to fail\nwith ENOKEY unconditionally.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Eric Paris \u003ceparis@redhat.com\u003e\nTested-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "13fca640bb8ab611a50e0ba120b186faa2994d6c", "tree": "0a40a28e498b2e9abdef8c86aa0aef7f82096f00", "parents": [ "67de0162fbb78713fcb23cb2502b380faa8bde73" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jun 15 21:53:52 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jun 15 21:53:52 2011 -0700" }, "message": "Revert \"fs/exec.c: use BUILD_BUG_ON for VM_STACK_FLAGS \u0026 VM_STACK_INCOMPLETE_SETUP\"\n\nThis reverts commit 7f81c8890c15a10f5220bebae3b6dfae4961962a.\n\nIt turns out that it\u0027s not actually a build-time check on x86-64 UML,\nwhich does some seriously crazy stuff with VM_STACK_FLAGS.\n\nThe VM_STACK_FLAGS define depends on the arch-supplied\nVM_STACK_DEFAULT_FLAGS value, and on x86-64 UML we have\n\n arch/um/sys-x86_64/shared/sysdep/vm-flags.h:\n\n\t#define VM_STACK_DEFAULT_FLAGS \\\n\t\t(test_thread_flag(TIF_IA32) ? vm_stack_flags32 : vm_stack_flags)\n\n\t#define VM_STACK_DEFAULT_FLAGS vm_stack_flags\n\n(yes, seriously: two different #define\u0027s for that thing, with the first\none being inside an \"#ifdef TIF_IA32\")\n\nIt\u0027s possible that it is UML that should just be fixed in this area, but\nfor now let\u0027s just undo the (very small) optimization.\n\nReported-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nAcked-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Michal Hocko \u003cmhocko@suse.cz\u003e\nCc: Richard Weinberger \u003crichard@nod.at\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7f81c8890c15a10f5220bebae3b6dfae4961962a", "tree": "264cb58ffbadb340ce713420a44c00d59fda4e7a", "parents": [ "b0825ee3a8c570df4873ee397fa453e67fdad5d7" ], "author": { "name": "Michal Hocko", "email": "mhocko@suse.cz", "time": "Wed Jun 15 15:08:11 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jun 15 20:03:59 2011 -0700" }, "message": "fs/exec.c: use BUILD_BUG_ON for VM_STACK_FLAGS \u0026 VM_STACK_INCOMPLETE_SETUP\n\nCommit a8bef8ff6ea1 (\"mm: migration: avoid race between shift_arg_pages()\nand rmap_walk() during migration by not migrating temporary stacks\")\nintroduced a BUG_ON() to ensure that VM_STACK_FLAGS and\nVM_STACK_INCOMPLETE_SETUP do not overlap. The check is a compile time\none, so BUILD_BUG_ON is more appropriate.\n\nSigned-off-by: Michal Hocko \u003cmhocko@suse.cz\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dac853ae89043f1b7752875300faf614de43c74b", "tree": "2f5732cd1197a1930a32eaa1ece1446ea73f1625", "parents": [ "06e86849cf4019945a106913adb9ff0abcc01770" ], "author": { "name": "Mathias Krause", "email": "minipli@googlemail.com", "time": "Thu Jun 09 20:05:18 2011 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 09 12:50:05 2011 -0700" }, "message": "exec: delay address limit change until point of no return\n\nUnconditionally changing the address limit to USER_DS and not restoring\nit to its old value in the error path is wrong because it prevents us\nusing kernel memory on repeated calls to this function. This, in fact,\nbreaks the fallback of hard coded paths to the init program from being\never successful if the first candidate fails to load.\n\nWith this patch applied switching to USER_DS is delayed until the point\nof no return is reached which makes it possible to have a multi-arch\nrootfs with one arch specific init binary for each of the (hard coded)\nprobed paths.\n\nSince the address limit is already set to USER_DS when start_thread()\nwill be invoked, this redundancy can be safely removed.\n\nSigned-off-by: Mathias Krause \u003cminipli@googlemail.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "57cc083ad9e1bfeeb4a0ee831e7bb008c8865bf0", "tree": "03b491fd20b4ca3e58354b15067f6961f32a016b", "parents": [ "3864601387cf4196371e3c1897fdffa5228296f9" ], "author": { "name": "Jiri Slaby", "email": "jslaby@suse.cz", "time": "Thu May 26 16:25:46 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 26 17:12:36 2011 -0700" }, "message": "coredump: add support for exe_file in core name\n\nNow, exe_file is not proc FS dependent, so we can use it to name core\nfile. So we add %E pattern for core file name cration which extract path\nfrom mm_struct-\u003eexe_file. Then it converts slashes to exclamation marks\nand pastes the result to the core file name itself.\n\nThis is useful for environments where binary names are longer than 16\ncharacter (the current-\u003ecomm limitation). Also where there are binaries\nwith same name but in a different path. Further in case the binery itself\nchanges its current-\u003ecomm after exec.\n\nSo by doing (s/$/#/ -- # is treated as git comment):\n\n $ sysctl kernel.core_pattern\u003d\u0027core.%p.%e.%E\u0027\n $ ln /bin/cat cat45678901234567890\n $ ./cat45678901234567890\n ^Z\n $ rm cat45678901234567890\n $ fg\n ^\\Quit (core dumped)\n $ ls core*\n\nwe now get:\n\n core.2434.cat456789012345.!root!cat45678901234567890 (deleted)\n\nSigned-off-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nReviewed-by: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3864601387cf4196371e3c1897fdffa5228296f9", "tree": "1c517a970194f9e49ef98ef434c650771ffa31e1", "parents": [ "63ab25ebbc50f74550bd8d164a34724b498f6fb9" ], "author": { "name": "Jiri Slaby", "email": "jslaby@suse.cz", "time": "Thu May 26 16:25:46 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 26 17:12:36 2011 -0700" }, "message": "mm: extract exe_file handling from procfs\n\nSetup and cleanup of mm_struct-\u003eexe_file is currently done in fs/proc/.\nThis was because exe_file was needed only for /proc/\u003cpid\u003e/exe. Since we\nwill need the exe_file functionality also for core dumps (so core name can\ncontain full binary path), built this functionality always into the\nkernel.\n\nTo achieve that move that out of proc FS to the kernel/ where in fact it\nshould belong. By doing that we can make dup_mm_exe_file static. Also we\ncan drop linux/proc_fs.h inclusion in fs/exec.c and kernel/fork.c.\n\nSigned-off-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d16dfc550f5326a4000f3322582a7c05dec91d7a", "tree": "8ee963542705cbf2187777f1d3f2b209cbda827a", "parents": [ "d05f3169c0fbca16132ec7c2be71685c6de638b5" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue May 24 17:11:45 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:12 2011 -0700" }, "message": "mm: mmu_gather rework\n\nRework the existing mmu_gather infrastructure.\n\nThe direct purpose of these patches was to allow preemptible mmu_gather,\nbut even without that I think these patches provide an improvement to the\nstatus quo.\n\nThe first 9 patches rework the mmu_gather infrastructure. For review\npurpose I\u0027ve split them into generic and per-arch patches with the last of\nthose a generic cleanup.\n\nThe next patch provides generic RCU page-table freeing, and the followup\nis a patch converting s390 to use this. I\u0027ve also got 4 patches from\nDaveM lined up (not included in this series) that uses this to implement\ngup_fast() for sparc64.\n\nThen there is one patch that extends the generic mmu_gather batching.\n\nAfter that follow the mm preemptibility patches, these make part of the mm\na lot more preemptible. It converts i_mmap_lock and anon_vma-\u003elock to\nmutexes which together with the mmu_gather rework makes mmu_gather\npreemptible as well.\n\nMaking i_mmap_lock a mutex also enables a clean-up of the truncate code.\n\nThis also allows for preemptible mmu_notifiers, something that XPMEM I\nthink wants.\n\nFurthermore, it removes the new and universially detested unmap_mutex.\n\nThis patch:\n\nRemove the first obstacle towards a fully preemptible mmu_gather.\n\nThe current scheme assumes mmu_gather is always done with preemption\ndisabled and uses per-cpu storage for the page batches. Change this to\ntry and allocate a page for batching and in case of failure, use a small\non-stack array to make some progress.\n\nPreemptible mmu_gather is desired in general and usable once i_mmap_lock\nbecomes a mutex. Doing it before the mutex conversion saves us from\nhaving to rework the code by moving the mmu_gather bits inside the\npte_lock.\n\nAlso avoid flushing the tlb batches from under the pte lock, this is\nuseful even without the i_mmap_lock conversion as it significantly reduces\npte lock hold times.\n\n[akpm@linux-foundation.org: fix comment tpyo]\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Russell King \u003crmk@arm.linux.org.uk\u003e\nCc: Paul Mundt \u003clethal@linux-sh.org\u003e\nCc: Jeff Dike \u003cjdike@addtoit.com\u003e\nCc: Richard Weinberger \u003crichard@nod.at\u003e\nCc: Tony Luck \u003ctony.luck@intel.com\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nCc: Namhyung Kim \u003cnamhyung@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d05f3169c0fbca16132ec7c2be71685c6de638b5", "tree": "37d82004869fa4e530617883f12cab7538dbd4a6", "parents": [ "248ac0e1943ad1796393d281b096184719eb3f97" ], "author": { "name": "Michal Hocko", "email": "mhocko@suse.cz", "time": "Tue May 24 17:11:44 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:12 2011 -0700" }, "message": "mm: make expand_downwards() symmetrical with expand_upwards()\n\nCurrently we have expand_upwards exported while expand_downwards is\naccessible only via expand_stack or expand_stack_downwards.\n\ncheck_stack_guard_page is a nice example of the asymmetry. It uses\nexpand_stack for VM_GROWSDOWN while expand_upwards is called for\nVM_GROWSUP case.\n\nLet\u0027s clean this up by exporting both functions and make those names\nconsistent. Let\u0027s use expand_{upwards,downwards} because expanding\ndoesn\u0027t always involve stack manipulation (an example is\nia64_do_page_fault which uses expand_upwards for registers backing store\nexpansion). expand_downwards has to be defined for both\nCONFIG_STACK_GROWS{UP,DOWN} because get_arg_page calls the downwards\nversion in the early process initialization phase for growsup\nconfiguration.\n\nSigned-off-by: Michal Hocko \u003cmhocko@suse.cz\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "99dff5856220a02b8711f2e8746413ea6e53ccf6", "tree": "d8bae7ccdf8f7ad5221b053bb74a6220df996b3a", "parents": [ "bb74e8ca352eecefdc5c1a9ebab01f18aa2f6f3c", "d9a0fbfd7bc5d2c42f0fa9bcbdab62c4942d0388" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 23 12:23:20 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 23 12:23:20 2011 -0700" }, "message": "Merge branch \u0027tty-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty-2.6\n\n* \u0027tty-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty-2.6: (48 commits)\n serial: 8250_pci: add support for Cronyx Omega PCI multiserial board.\n tty/serial: Fix break handling for PORT_TEGRA\n tty/serial: Add explicit PORT_TEGRA type\n n_tracerouter and n_tracesink ldisc additions.\n Intel PTI implementaiton of MIPI 1149.7.\n Kernel documentation for the PTI feature.\n export kernel call get_task_comm().\n tty: Remove to support serial for S5P6442\n pch_phub: Support new device ML7223\n 8250_pci: Add support for the Digi/IBM PCIe 2-port Adapter\n ASoC: Update cx20442 for TTY API change\n pch_uart: Support new device ML7223 IOH\n parport: Use request_muxed_region for IT87 probe and lock\n tty/serial: add support for Xilinx PS UART\n n_gsm: Use print_hex_dump_bytes\n drivers/tty/moxa.c: Put correct tty value\n TTY: tty_io, annotate locking functions\n TTY: serial_core, remove superfluous set_task_state\n TTY: serial_core, remove invalid test\n Char: moxa, fix locking in moxa_write\n ...\n\nFix up trivial conflicts in drivers/bluetooth/hci_ldisc.c and\ndrivers/tty/serial/Makefile.\n\nI did the hci_ldisc thing as an evil merge, cleaning things up.\n" }, { "commit": "4d9dec4db2efbd7edb549bd02373995b67496983", "tree": "a0ddbecea4823a7f3e20ab6eacb37a086a72dc36", "parents": [ "34b064569eba3bec65bf98efe057b0578fe13297", "ae6b585eeb74670a2dec1fe4394bdfbdb9395cc2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 23 08:28:34 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 23 08:28:34 2011 -0700" }, "message": "Merge branch \u0027exec_rm_compat\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/oleg/misc\n\n* \u0027exec_rm_compat\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/oleg/misc:\n exec: document acct_arg_size()\n exec: unify do_execve/compat_do_execve code\n exec: introduce struct user_arg_ptr\n exec: introduce get_user_arg_ptr() helper\n" }, { "commit": "7d74f492e4dd0034a61458eb80f70b1d2862ed07", "tree": "1be37b81f6cc0260314012a1e06b8779df0a4ec6", "parents": [ "e0f263d719cc70245406cc26d86a480925e6a3c9" ], "author": { "name": "J Freyensee", "email": "james_p_freyensee@linux.intel.com", "time": "Fri May 06 16:56:47 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri May 13 16:30:59 2011 -0700" }, "message": "export kernel call get_task_comm().\n\nThis allows drivers who call this function to be compiled modularly.\nOtherwise, a driver who is interested in this type of functionality\nhas to implement their own get_task_comm() call, causing code\nduplication in the Linux source tree.\n\nSigned-off-by: J Freyensee \u003cjames_p_freyensee@linux.intel.com\u003e\nAcked-by: David Rientjes \u003crientjes@google.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "ae6b585eeb74670a2dec1fe4394bdfbdb9395cc2", "tree": "ea1642cc0abf3b73573769bdb15c8d390bc3e7c8", "parents": [ "0e028465d18b7c6797fcbdea632299d16097c5cd" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sun Mar 06 18:03:11 2011 +0100" }, "committer": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sat Apr 09 15:53:57 2011 +0200" }, "message": "exec: document acct_arg_size()\n\nAdd the comment to explain acct_arg_size().\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\n" }, { "commit": "0e028465d18b7c6797fcbdea632299d16097c5cd", "tree": "d03a1f0f688e9c4a780b2a1a3ef8354378b0ad42", "parents": [ "ba2d01629d0d167598cfea85adc7926822bbfc45" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sun Mar 06 18:02:54 2011 +0100" }, "committer": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sat Apr 09 15:53:56 2011 +0200" }, "message": "exec: unify do_execve/compat_do_execve code\n\nAdd the appropriate members into struct user_arg_ptr and teach\nget_user_arg_ptr() to handle is_compat \u003d T case correctly.\n\nThis allows us to remove the compat_do_execve() code from fs/compat.c\nand reimplement compat_do_execve() as the trivial wrapper on top of\ndo_execve_common(is_compat \u003d\u003e true).\n\nIn fact, this fixes another (minor) bug. \"compat_uptr_t str\" can\noverflow after \"str +\u003d len\" in compat_copy_strings() if a 64bit\napplication execs via sys32_execve().\n\nUnexport acct_arg_size() and get_arg_page(), fs/compat.c doesn\u0027t\nneed them any longer.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nTested-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\n" }, { "commit": "ba2d01629d0d167598cfea85adc7926822bbfc45", "tree": "703a3f297df7cb4c0317504a58121f9958c0d708", "parents": [ "1d1dbf8135ab2f3603cc72e39e0f68784f453c39" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sun Mar 06 18:02:37 2011 +0100" }, "committer": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sat Apr 09 15:53:53 2011 +0200" }, "message": "exec: introduce struct user_arg_ptr\n\nNo functional changes, preparation.\n\nIntroduce struct user_arg_ptr, change do_execve() paths to use it\ninstead of \"char __user * const __user *argv\".\n\nThis makes the argv/envp arguments opaque, we are ready to handle the\ncompat case which needs argv pointing to compat_uptr_t.\n\nSuggested-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nTested-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\n" }, { "commit": "1d1dbf8135ab2f3603cc72e39e0f68784f453c39", "tree": "e30b04b64f92a55200aab62d414fb1378f7bfaab", "parents": [ "bb3c90f0de7b34995b5e35cf5dc97a3d428b3761" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sun Mar 06 18:02:21 2011 +0100" }, "committer": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sat Apr 09 15:53:45 2011 +0200" }, "message": "exec: introduce get_user_arg_ptr() helper\n\nIntroduce get_user_arg_ptr() helper, convert count() and copy_strings()\nto use it.\n\nNo functional changes, preparation. This helper is trivial, it just\nreads the pointer from argv/envp user-space array.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nTested-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\n" }, { "commit": "39efa3ef3a376a4e53de2f82fc91182459d34200", "tree": "f75a44fa77396edcf74e41686912554f96681340", "parents": [ "e5c1902e9260a0075ea52cb5ef627a8d9aaede89" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Mar 23 10:37:00 2011 +0100" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Mar 23 10:37:00 2011 +0100" }, "message": "signal: Use GROUP_STOP_PENDING to stop once for a single group stop\n\nCurrently task-\u003esignal-\u003egroup_stop_count is used to decide whether to\nstop for group stop. However, if there is a task in the group which\nis taking a long time to stop, other tasks which are continued by\nptrace would repeatedly stop for the same group stop until the group\nstop is complete.\n\nConversely, if a ptraced task is in TASK_TRACED state, the debugger\nwon\u0027t get notified of group stops which is inconsistent compared to\nthe ptraced task in any other state.\n\nThis patch introduces GROUP_STOP_PENDING which tracks whether a task\nis yet to stop for the group stop in progress. The flag is set when a\ngroup stop starts and cleared when the task stops the first time for\nthe group stop, and consulted whenever whether the task should\nparticipate in a group stop needs to be determined. Note that now\ntasks in TASK_TRACED also participate in group stop.\n\nThis results in the following behavior changes.\n\n* For a single group stop, a ptracer would see at most one stop\n reported.\n\n* A ptracee in TASK_TRACED now also participates in group stop and the\n tracer would get the notification. However, as a ptraced task could\n be in TASK_STOPPED state or any ptrace trap could consume group\n stop, the notification may still be missing. These will be\n addressed with further patches.\n\n* A ptracee may start a group stop while one is still in progress if\n the tracer let it continue with stop signal delivery. Group stop\n code handles this correctly.\n\nOleg:\n\n* Spotted that a task might skip signal check even when its\n GROUP_STOP_PENDING is set. Fixed by updating\n recalc_sigpending_tsk() to check GROUP_STOP_PENDING instead of\n group_stop_count.\n\n* Pointed out that task-\u003egroup_stop should be cleared whenever\n task-\u003esignal-\u003egroup_stop_count is cleared. Fixed accordingly.\n\n* Pointed out the behavior inconsistency between TASK_TRACED and\n RUNNING and the last behavior change.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\n" }, { "commit": "1bef82917c74249ff21982127e57defd6ca2bb1b", "tree": "dd288483397881b5753e53fdf7ba193c69f475eb", "parents": [ "ff38c083ad3bb8dbbed80aa9090fcc96bc4af7db" ], "author": { "name": "Holger Hans Peter Freyther", "email": "holger@freyther.de", "time": "Thu Feb 24 17:46:49 2011 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 21 00:16:09 2011 -0400" }, "message": "Small typo fix...\n\nHi,\n\nI was backporting the coredump over pipe feature and noticed this small typo,\nI wish I would have something bigger to contribute...\n\n\u003eFrom 15d6080e0ed4267da103c706917a33b1015e8804 Mon Sep 17 00:00:00 2001\nFrom: Holger Hans Peter Freyther \u003cholger@moiji-mobile.com\u003e\nDate: Thu, 24 Feb 2011 17:42:50 +0100\nSubject: [PATCH] fs: Fix a small typo in the comment\n\nThe function is called umh_pipe_setup not uhm_pipe_setup.\n\nSigned-off-by: Holger Hans Peter Freyther \u003cholger@moiji-mobile.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "47c805dc2d2dff686962f5f0baa6bac2d703ba19", "tree": "b596e0a1aadb1550ef2099df95832196cd0eeda1", "parents": [ "c3e380b0b3cfa613189fb91513efd88a65e1d9d8" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Feb 23 17:44:09 2011 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 14 09:15:25 2011 -0400" }, "message": "switch do_filp_open() to struct open_flags\n\ntake calculation of open_flags by open(2) arguments into new helper\nin fs/open.c, move filp_open() over there, have it and do_sys_open()\nuse that helper, switch exec.c callers of do_filp_open() to explicit\n(and constant) struct open_flags.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "3cd90ea42f2c15f928b70ed66f6d8ed0a8e7aadd", "tree": "2e27c807baaec2e6528cd5431d57308a4d874dea", "parents": [ "1a44bc8c7cfe69756a116d38aef992d50fc1969d" ], "author": { "name": "Namhyung Kim", "email": "namhyung@gmail.com", "time": "Tue Feb 01 15:52:46 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 02 16:03:19 2011 -0800" }, "message": "vfs: sparse: add __FMODE_EXEC\n\nFMODE_EXEC is a constant type of fmode_t but was used with normal integer\nconstants. This results in following warnings from sparse. Fix it using\nnew macro __FMODE_EXEC.\n\n fs/exec.c:116:58: warning: restricted fmode_t degrades to integer\n fs/exec.c:689:58: warning: restricted fmode_t degrades to integer\n fs/fcntl.c:777:9: warning: restricted fmode_t degrades to integer\n\nSigned-off-by: Namhyung Kim \u003cnamhyung@gmail.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "462e635e5b73ba9a4c03913b77138cd57ce4b050", "tree": "6ff0e84eecc6252d41d7c08730018c0149e7227f", "parents": [ "0fcdcfbbc98f70f559e4b36773a69972489a6d8f" ], "author": { "name": "Tavis Ormandy", "email": "taviso@cmpxchg8b.com", "time": "Thu Dec 09 15:29:42 2010 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Dec 15 12:30:36 2010 -0800" }, "message": "install_special_mapping skips security_file_mmap check.\n\nThe install_special_mapping routine (used, for example, to setup the\nvdso) skips the security check before insert_vm_struct, allowing a local\nattacker to bypass the mmap_min_addr security restriction by limiting\nthe available pages for special mappings.\n\nbprm_mm_init() also skips the check, and although I don\u0027t think this can\nbe used to bypass any restrictions, I don\u0027t see any reason not to have\nthe security check.\n\n $ uname -m\n x86_64\n $ cat /proc/sys/vm/mmap_min_addr\n 65536\n $ cat install_special_mapping.s\n section .bss\n resb BSS_SIZE\n section .text\n global _start\n _start:\n mov eax, __NR_pause\n int 0x80\n $ nasm -D__NR_pause\u003d29 -DBSS_SIZE\u003d0xfffed000 -f elf -o install_special_mapping.o install_special_mapping.s\n $ ld -m elf_i386 -Ttext\u003d0x10000 -Tbss\u003d0x11000 -o install_special_mapping install_special_mapping.o\n $ ./install_special_mapping \u0026\n [1] 14303\n $ cat /proc/14303/maps\n 0000f000-00010000 r-xp 00000000 00:00 0 [vdso]\n 00010000-00011000 r-xp 00001000 00:19 2453665 /home/taviso/install_special_mapping\n 00011000-ffffe000 rwxp 00000000 00:00 0 [stack]\n\nIt\u0027s worth noting that Red Hat are shipping with mmap_min_addr set to\n4096.\n\nSigned-off-by: Tavis Ormandy \u003ctaviso@google.com\u003e\nAcked-by: Kees Cook \u003ckees@ubuntu.com\u003e\nAcked-by: Robert Swiecki \u003cswiecki@google.com\u003e\n[ Changed to not drop the error code - akpm ]\nReviewed-by: James Morris \u003cjmorris@namei.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "114279be2120a916e8a04feeb2ac976a10016f2f", "tree": "866a5dfab411d18941d58f8796edab6da760213e", "parents": [ "3c77f845722158206a7209c45ccddc264d19319c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Tue Nov 30 20:56:02 2010 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 30 17:56:38 2010 -0800" }, "message": "exec: copy-and-paste the fixes into compat_do_execve() paths\n\nNote: this patch targets 2.6.37 and tries to be as simple as possible.\nThat is why it adds more copy-and-paste horror into fs/compat.c and\nuglifies fs/exec.c, this will be cleanuped later.\n\ncompat_copy_strings() plays with bprm-\u003evma/mm directly and thus has\ntwo problems: it lacks the RLIMIT_STACK check and argv/envp memory\nis not visible to oom killer.\n\nExport acct_arg_size() and get_arg_page(), change compat_copy_strings()\nto use get_arg_page(), change compat_do_execve() to do acct_arg_size(0)\nas do_execve() does.\n\nAdd the fatal_signal_pending/cond_resched checks into compat_count() and\ncompat_copy_strings(), this matches the code in fs/exec.c and certainly\nmakes sense.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3c77f845722158206a7209c45ccddc264d19319c", "tree": "9eace97a8b88eb68b7d5d3127041b14c202421ae", "parents": [ "37a09f07459753e7c98d4e21f1c61e8756923f81" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Tue Nov 30 20:55:34 2010 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 30 17:56:37 2010 -0800" }, "message": "exec: make argv/envp memory visible to oom-killer\n\nBrad Spengler published a local memory-allocation DoS that\nevades the OOM-killer (though not the virtual memory RLIMIT):\nhttp://www.grsecurity.net/~spender/64bit_dos.c\n\nexecve()-\u003ecopy_strings() can allocate a lot of memory, but\nthis is not visible to oom-killer, nobody can see the nascent\nbprm-\u003emm and take it into account.\n\nWith this patch get_arg_page() increments current\u0027s MM_ANONPAGES\ncounter every time we allocate the new page for argv/envp. When\ndo_execve() succeds or fails, we change this counter back.\n\nTechnically this is not 100% correct, we can\u0027t know if the new\npage is swapped out and turn MM_ANONPAGES into MM_SWAPENTS, but\nI don\u0027t think this really matters and everything becomes correct\nonce exec changes -\u003emm or fails.\n\nReported-by: Brad Spengler \u003cspender@grsecurity.net\u003e\nReviewed-and-discussed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "98391cf4dcf893e9e74e1c14189851dbc9c5ad0d", "tree": "f76b582402e03a74628071b8190b93da274f59f4", "parents": [ "478735e38887077ac77a9756121b6ce0cb956e2f" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Wed Oct 27 15:34:16 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Oct 27 18:03:13 2010 -0700" }, "message": "exec: don\u0027t turn PF_KTHREAD off when a target command was not found\n\nPresently do_execve() turns PF_KTHREAD off before search_binary_handler().\n THis has a theorical risk of PF_KTHREAD getting lost. We don\u0027t have to\nturn PF_KTHREAD off in the ENOEXEC case.\n\nThis patch moves this flag modification to after the finding of the\nexecutable file.\n\nThis is only a theorical issue because kthreads do not call do_execve()\ndirectly. But fixing would be better.\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nAcked-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1b0d300bd0f047e2edaf9d4b6784189e6c67c3d1", "tree": "ce3777527c949add56658e4d5b2ab02781590931", "parents": [ "9b1bf12d5d51bca178dea21b04a0805e29d60cf1" ], "author": { "name": "Xiaotian Feng", "email": "dfeng@redhat.com", "time": "Wed Oct 27 15:34:08 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Oct 27 18:03:12 2010 -0700" }, "message": "core_pattern: fix truncation by core_pattern handler with long parameters\n\nWe met a parameter truncated issue, consider following:\n\u003e echo \"|/root/core_pattern_pipe_test %p /usr/libexec/blah-blah-blah \\\n%s %c %p %u %g 11 12345678901234567890123456789012345678 %t\" \u003e \\\n/proc/sys/kernel/core_pattern\n\nThis is okay because the strings is less than CORENAME_MAX_SIZE. \"cat\n/proc/sys/kernel/core_pattern\" shows the whole string. but after we run\ncore_pattern_pipe_test in man page, we found last parameter was truncated\nlike below:\n\n argc[10]\u003d\u003c12807486\u003e\n\nThe root cause is core_pattern allows % specifiers, which need to be\nreplaced during parse time, but the replace may expand the strings to\nlarger than CORENAME_MAX_SIZE. So if the last parameter is % specifiers,\nthe replace code is using snprintf(out_ptr, out_end - out_ptr, ...), this\nwill write out of corename array.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Xiaotian Feng \u003cdfeng@redhat.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReviewed-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9b1bf12d5d51bca178dea21b04a0805e29d60cf1", "tree": "902a2c0e5882a2152da59a589958d6426dc84643", "parents": [ "b84011508360d6885a9d95a235ec77d56f133377" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Wed Oct 27 15:34:08 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Oct 27 18:03:12 2010 -0700" }, "message": "signals: move cred_guard_mutex from task_struct to signal_struct\n\nOleg Nesterov pointed out we have to prevent multiple-threads-inside-exec\nitself and we can reuse -\u003ecred_guard_mutex for it. Yes, concurrent\nexecve() has no worth.\n\nLet\u0027s move -\u003ecred_guard_mutex from task_struct to signal_struct. It\nnaturally prevent multiple-threads-inside-exec.\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReviewed-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nAcked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3d5992d2ac7dc09aed8ab537cba074589f0f0a52", "tree": "0fbb51026e8aa13ddd7a6516c04471ceb97bc277", "parents": [ "0f4d208f1975f16f269134cee5f44c1f048581da" ], "author": { "name": "Ying Han", "email": "yinghan@google.com", "time": "Tue Oct 26 14:21:23 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:05 2010 -0700" }, "message": "oom: add per-mm oom disable count\n\nIt\u0027s pointless to kill a task if another thread sharing its mm cannot be\nkilled to allow future memory freeing. A subsequent patch will prevent\nkills in such cases, but first it\u0027s necessary to have a way to flag a task\nthat shares memory with an OOM_DISABLE task that doesn\u0027t incur an\nadditional tasklist scan, which would make select_bad_process() an O(n^2)\nfunction.\n\nThis patch adds an atomic counter to struct mm_struct that follows how\nmany threads attached to it have an oom_score_adj of OOM_SCORE_ADJ_MIN.\nThey cannot be killed by the kernel, so their memory cannot be freed in\noom conditions.\n\nThis only requires task_lock() on the task that we\u0027re operating on, it\ndoes not require mm-\u003emmap_sem since task_lock() pins the mm and the\noperation is atomic.\n\n[rientjes@google.com: changelog and sys_unshare() code]\n[rientjes@google.com: protect oom_disable_count with task_lock in fork]\n[rientjes@google.com: use old_mm for oom_disable_count in exec]\nSigned-off-by: Ying Han \u003cyinghan@google.com\u003e\nSigned-off-by: David Rientjes \u003crientjes@google.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8fd01d6cfbf75465d84a4e533ed70c5f57b3ff51", "tree": "9a03e08d879f14d5c0413fdc9f3e70bcf213279d", "parents": [ "cd07202cc8262e1669edff0d97715f3dd9260917" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 14 19:15:28 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 14 19:15:28 2010 -0700" }, "message": "Export dump_{write,seek} to binary loader modules\n\nIf you build aout support as a module, you\u0027ll want these exported.\n\nReported-by: Tetsuo Handa \u003cpenguin-kernel@i-love.sakura.ne.jp\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3aa0ce825ade0cf5506e32ccf51d01fc8d22a9cf", "tree": "77ca43d663b62922a743de3729a8b8b47eafb887", "parents": [ "ae42d8d44195a614c78a636683fe65ed31744cbd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 14 14:32:06 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 14 14:32:06 2010 -0700" }, "message": "Un-inline the core-dump helper functions\n\nTony Luck reports that the addition of the access_ok() check in commit\n0eead9ab41da (\"Don\u0027t dump task struct in a.out core-dumps\") broke the\nia64 compile due to missing the necessary header file includes.\n\nRather than add yet another include (\u003casm/unistd.h\u003e) to make everything\nhappy, just uninline the silly core dump helper functions and move the\nbodies to fs/exec.c where they make a lot more sense.\n\ndump_seek() in particular was too big to be an inline function anyway,\nand none of them are in any way performance-critical. And we really\ndon\u0027t need to mess up our include file headers more than they already\nare.\n\nReported-and-tested-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9aea5a65aa7a1af9a4236dfaeb0088f1624f9919", "tree": "c0e36c74dc511912f59e207c70893dc6bd13b543", "parents": [ "7993bc1f4663c0db67bb8f0d98e6678145b387cd" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Tue Sep 07 19:37:06 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 10 08:10:26 2010 -0700" }, "message": "execve: make responsive to SIGKILL with large arguments\n\nAn execve with a very large total of argument/environment strings\ncan take a really long time in the execve system call. It runs\nuninterruptibly to count and copy all the strings. This change\nmakes it abort the exec quickly if sent a SIGKILL.\n\nNote that this is the conservative change, to interrupt only for\nSIGKILL, by using fatal_signal_pending(). It would be perfectly\ncorrect semantics to let any signal interrupt the string-copying in\nexecve, i.e. use signal_pending() instead of fatal_signal_pending().\nWe\u0027ll save that change for later, since it could have user-visible\nconsequences, such as having a timer set too quickly make it so that\nan execve can never complete, though it always happened to work before.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7993bc1f4663c0db67bb8f0d98e6678145b387cd", "tree": "d283304f635cc5c36d8fa3319dd663209138988e", "parents": [ "1b528181b2ffa14721fb28ad1bd539fe1732c583" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Tue Sep 07 19:36:28 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 10 08:10:26 2010 -0700" }, "message": "execve: improve interactivity with large arguments\n\nThis adds a preemption point during the copying of the argument and\nenvironment strings for execve, in copy_strings(). There is already\na preemption point in the count() loop, so this doesn\u0027t add any new\npoints in the abstract sense.\n\nWhen the total argument+environment strings are very large, the time\nspent copying them can be much more than a normal user time slice.\nSo this change improves the interactivity of the rest of the system\nwhen one process is doing an execve with very large arguments.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1b528181b2ffa14721fb28ad1bd539fe1732c583", "tree": "c6c11dd4fb1845ce1a7d5c048fba40a62d5d94de", "parents": [ "be6200aac985e0a3db56ec636763a32f3e32e7f1" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Tue Sep 07 19:35:49 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 10 08:10:26 2010 -0700" }, "message": "setup_arg_pages: diagnose excessive argument size\n\nThe CONFIG_STACK_GROWSDOWN variant of setup_arg_pages() does not\ncheck the size of the argument/environment area on the stack.\nWhen it is unworkably large, shift_arg_pages() hits its BUG_ON.\nThis is exploitable with a very large RLIMIT_STACK limit, to\ncreate a crash pretty easily.\n\nCheck that the initial stack is not too large to make it possible\nto map in any executable. We\u0027re not checking that the actual\nexecutable (or intepreter, for binfmt_elf) will fit. So those\nmappings might clobber part of the initial stack mapping. But\nthat is just userland lossage that userland made happen, not a\nkernel problem.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "145c3ae46b37993b0debb0b3da6256daea4a6ec5", "tree": "0dbff382ce36b23b3d2dbff87d3eaab73a07a2a4", "parents": [ "81ca03a0e2ea0207b2df80e0edcf4c775c07a505", "99b7db7b8ffd6bb755eb0a175596421a0b581cb2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 18 09:35:08 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 18 09:35:08 2010 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:\n fs: brlock vfsmount_lock\n fs: scale files_lock\n lglock: introduce special lglock and brlock spin locks\n tty: fix fu_list abuse\n fs: cleanup files_lock locking\n fs: remove extra lookup in __lookup_hash\n fs: fs_struct rwlock to spinlock\n apparmor: use task path helpers\n fs: dentry allocation consolidation\n fs: fix do_lookup false negative\n mbcache: Limit the maximum number of cache entries\n hostfs -\u003efollow_link() braino\n hostfs: dumb (and usually harmless) tpyo - strncpy instead of strlcpy\n remove SWRITE* I/O types\n kill BH_Ordered flag\n vfs: update ctime when changing the file\u0027s permission by setfacl\n cramfs: only unlock new inodes\n fix reiserfs_evict_inode end_writeback second call\n" }, { "commit": "2a4419b5b2a77f3f4537c14f7ad7df95770655dd", "tree": "ad66519a92b995920ecada788e4a08e265747545", "parents": [ "44672e4fbd40e2dda8bbce7d0f71d24dbfc7e00e" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Wed Aug 18 04:37:33 2010 +1000" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Aug 18 08:35:46 2010 -0400" }, "message": "fs: fs_struct rwlock to spinlock\n\nfs: fs_struct rwlock to spinlock\n\nstruct fs_struct.lock is an rwlock with the read-side used to protect root and\npwd members while taking references to them. Taking a reference to a path\ntypically requires just 2 atomic ops, so the critical section is very small.\nParallel read-side operations would have cacheline contention on the lock, the\ndentry, and the vfsmount cachelines, so the rwlock is unlikely to ever give a\nreal parallelism increase.\n\nReplace it with a spinlock to avoid one or two atomic operations in typical\npath lookup fastpath.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "d7627467b7a8dd6944885290a03a07ceb28c10eb", "tree": "a18c83468418e878cfb2d44e4310d81b8db84ad7", "parents": [ "da5cabf80e2433131bf0ed8993abc0f7ea618c73" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Tue Aug 17 23:52:56 2010 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 17 18:07:43 2010 -0700" }, "message": "Make do_execve() take a const filename pointer\n\nMake do_execve() take a const filename pointer so that kernel_execve() compiles\ncorrectly on ARM:\n\narch/arm/kernel/sys_arm.c:88: warning: passing argument 1 of \u0027do_execve\u0027 discards qualifiers from pointer target type\n\nThis also requires the argv and envp arguments to be consted twice, once for\nthe pointer array and once for the strings the array points to. This is\nbecause do_execve() passes a pointer to the filename (now const) to\ncopy_strings_kernel(). A simpler alternative would be to cast the filename\npointer in do_execve() when it\u0027s passed to copy_strings_kernel().\n\ndo_execve() may not change any of the strings it is passed as part of the argv\nor envp lists as they are some of them in .rodata, so marking these strings as\nconst should be fine.\n\nFurther kernel_execve() and sys_execve() need to be changed to match.\n\nThis has been test built on x86_64, frv, arm and mips.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nTested-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\nAcked-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8c8946f509a494769a8c602b5ed189df01917d39", "tree": "dfd96bd6ca5ea6803c6d77f65ba37e04f78b2d3b", "parents": [ "5f248c9c251c60af3403902b26e08de43964ea0b", "1968f5eed54ce47bde488fd9a450912e4a2d7138" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 10 11:39:13 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 10 11:39:13 2010 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.infradead.org/users/eparis/notify\n\n* \u0027for-linus\u0027 of git://git.infradead.org/users/eparis/notify: (132 commits)\n fanotify: use both marks when possible\n fsnotify: pass both the vfsmount mark and inode mark\n fsnotify: walk the inode and vfsmount lists simultaneously\n fsnotify: rework ignored mark flushing\n fsnotify: remove global fsnotify groups lists\n fsnotify: remove group-\u003emask\n fsnotify: remove the global masks\n fsnotify: cleanup should_send_event\n fanotify: use the mark in handler functions\n audit: use the mark in handler functions\n dnotify: use the mark in handler functions\n inotify: use the mark in handler functions\n fsnotify: send fsnotify_mark to groups in event handling functions\n fsnotify: Exchange list heads instead of moving elements\n fsnotify: srcu to protect read side of inode and vfsmount locks\n fsnotify: use an explicit flag to indicate fsnotify_destroy_mark has been called\n fsnotify: use _rcu functions for mark list traversal\n fsnotify: place marks on object in order of group memory address\n vfs/fsnotify: fsnotify_close can delay the final work in fput\n fsnotify: store struct file not struct path\n ...\n\nFix up trivial delete/modify conflict in fs/notify/inotify/inotify.c.\n" }, { "commit": "78417334b5cb6e1f915b8fdcc4fce3f1a1b4420c", "tree": "8e3fce9f27fd058904c0a3247a5c5ebb8df9fbe8", "parents": [ "cd816a0d84377c4e87f55cbe934a23417f9f5743", "5f202bd5ca64132cdd7f186656bc0221f257733d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Aug 07 17:06:54 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Aug 07 17:06:54 2010 -0700" }, "message": "Merge branch \u0027bkl/core\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/random-tracing\n\n* \u0027bkl/core\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/random-tracing:\n do_coredump: Do not take BKL\n init: Remove the BKL from startup code\n" }, { "commit": "2a12a9d7814631e918dec93abad856e692d5286d", "tree": "12817004ae9667bf83f869606f38050636edeb61", "parents": [ "8112e2d6a7356e8c3ff1f7f3c86f375ed0305705" ], "author": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Thu Dec 17 21:24:21 2009 -0500" }, "committer": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Wed Jul 28 09:58:32 2010 -0400" }, "message": "fsnotify: pass a file instead of an inode to open, read, and write\n\nfanotify, the upcoming notification system actually needs a struct path so it can\ndo opens in the context of listeners, and it needs a file so it can get f_flags\nfrom the original process. Close was the only operation that already was passing\na struct file to the notification hook. This patch passes a file for access,\nmodify, and open as well as they are easily available to these hooks.\n\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\n" }, { "commit": "5f202bd5ca64132cdd7f186656bc0221f257733d", "tree": "e53eac24bf18052955565370f3327da088fa1be7", "parents": [ "5e3d20a68f63fc5a310687d81956c3b96e488b84" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Sun Jul 04 00:02:25 2010 +0200" }, "committer": { "name": "Frederic Weisbecker", "email": "fweisbec@gmail.com", "time": "Sat Jul 10 01:21:49 2010 +0200" }, "message": "do_coredump: Do not take BKL\n\ncore_pattern is not actually protected and hasn\u0027t been\never since we introduced procfs support for sysctl -- a\n_long_ time. Don\u0027t take it here either.\n\nAlso nothing inside do_coredump appears to require bkl\nprotection.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n[ remove smp_lock.h headers ]\nSigned-off-by: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\n" }, { "commit": "3af9e859281bda7eb7c20b51879cf43aa788ac2e", "tree": "cddb7fc788abc826551e12bba5892c6b26942392", "parents": [ "8ed92280be013180e24c84456ab6babcb07037cc" ], "author": { "name": "Eric B Munson", "email": "ebmunson@us.ibm.com", "time": "Tue May 18 15:30:49 2010 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Jun 09 11:12:34 2010 +0200" }, "message": "perf: Add non-exec mmap() tracking\n\nAdd the capacility to track data mmap()s. This can be used together\nwith PERF_SAMPLE_ADDR for data profiling.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\n[Updated code for stable perf ABI]\nSigned-off-by: Eric B Munson \u003cebmunson@us.ibm.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Mike Galbraith \u003cefault@gmx.de\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nLKML-Reference: \u003c1274193049-25997-1-git-send-email-ebmunson@us.ibm.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d344193a05da89c97e965da2c5cbf687d7385eae", "tree": "6267b68c31c4547f1749acd35d2a5bfaff919839", "parents": [ "09faef11df8c559a23e2405d123cb2683733a79a" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed May 26 14:43:11 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:46 2010 -0700" }, "message": "exit: avoid sig-\u003ecount in de_thread/__exit_signal synchronization\n\nde_thread() and __exit_signal() use signal_struct-\u003ecount/notify_count for\nsynchronization. We can simplify the code and use -\u003enotify_count only.\nInstead of comparing these two counters, we can change de_thread() to set\n-\u003enotify_count \u003d nr_of_sub_threads, then change __exit_signal() to\ndec-and-test this counter and notify group_exit_task.\n\nNote that __exit_signal() checks \"notify_count \u003e 0\" just for symmetry with\nexit_notify(), we could just check it is !\u003d 0.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Veaceslav Falico \u003cvfalico@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "269b005a28e124a341df4adef2c3661cf7371fcc", "tree": "226b105cc4c1bbe0b3fe92f2733708a7dc2dedb8", "parents": [ "5e43aef530ba39206f7923295388f7ec3c5a7d93" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed May 26 14:43:08 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:45 2010 -0700" }, "message": "coredump: shift down_write(mmap_sem) into coredump_wait()\n\n- move the cprm.mm_flags checks up, before we take mmap_sem\n\n- move down_write(mmap_sem) and -\u003ecore_state check from do_coredump()\n to coredump_wait()\n\nThis simplifies the code and makes the locking symmetrical.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5e43aef530ba39206f7923295388f7ec3c5a7d93", "tree": "b29b706cff64eb47cc881a793566e38a511843f1", "parents": [ "d5bf4c4f5f9dcc90b7e25dbb2f7c4436cf6e7ed0" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed May 26 14:43:07 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:45 2010 -0700" }, "message": "coredump: factor out put_cred() calls\n\nGiven that do_coredump() calls put_cred() on exit path, it is a bit ugly\nto do put_cred() + \"goto fail\" twice, just add the new \"fail_creds\" label.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d5bf4c4f5f9dcc90b7e25dbb2f7c4436cf6e7ed0", "tree": "8c07402f685245fdf3397fac47bc7194f686066d", "parents": [ "c713541125002b8bc9e681af3b09118e771e2d8a" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed May 26 14:43:06 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:45 2010 -0700" }, "message": "coredump: cleanup \"ispipe\" code\n\n- kill \"int dump_count\", argv_split(argcp) accepts argcp \u003d\u003d NULL.\n\n- move \"int dump_count\" under \" if (ispipe)\" branch, fail_dropcount\n can check ispipe.\n\n- move \"char **helper_argv\" as well, change the code to do argv_free()\n right after call_usermodehelper_fns().\n\n- If call_usermodehelper_fns() fails goto close_fail label instead\n of closing the file by hand.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c713541125002b8bc9e681af3b09118e771e2d8a", "tree": "6c85885dc2403b0310ab13e1c21dcb30960ced83", "parents": [ "04b1c384fbc4e0209e5c1affb67050886376d44b" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed May 26 14:43:05 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:45 2010 -0700" }, "message": "coredump: factor out the not-ispipe file checks\n\ndo_coredump() does a lot of file checks after it opens the file or calls\nusermode helper. But all of these checks are only needed in !ispipe case.\n\nMove this code into the \"else\" branch and kill the ugly repetitive ispipe\nchecks.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "898b374af6f71041bd3bceebe257e564f3f1d458", "tree": "b1be6fd3cca69becfc75787eab906338f363d3cd", "parents": [ "a06a4dc3a08201ff6a8a958f935b3cbf7744115f" ], "author": { "name": "Neil Horman", "email": "nhorman@tuxdriver.com", "time": "Wed May 26 14:42:59 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 09:12:44 2010 -0700" }, "message": "exec: replace call_usermodehelper_pipe with use of umh init function and resolve limit\n\nThe first patch in this series introduced an init function to the\ncall_usermodehelper api so that processes could be customized by caller.\nThis patch takes advantage of that fact, by customizing the helper in\ndo_coredump to create the pipe and set its core limit to one (for our\nrecusrsion check). This lets us clean up the previous uglyness in the\nusermodehelper internals and factor call_usermodehelper out entirely.\nWhile I\u0027m at it, we can also modify the helper setup to look for a core\nlimit value of 1 rather than zero for our recursion check\n\nSigned-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nReviewed-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a8bef8ff6ea15fa4c67433cab0f5f3484574ef7c", "tree": "3ea9c122e02f523379d4560ee2134124d097895c", "parents": [ "e9e96b39f932a065e14f5d5bab0797ae261d03b5" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon May 24 14:32:24 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 25 08:06:59 2010 -0700" }, "message": "mm: migration: avoid race between shift_arg_pages() and rmap_walk() during migration by not migrating temporary stacks\n\nPage migration requires rmap to be able to find all ptes mapping a page\nat all times, otherwise the migration entry can be instantiated, but it\nis possible to leave one behind if the second rmap_walk fails to find\nthe page. If this page is later faulted, migration_entry_to_page() will\ncall BUG because the page is locked indicating the page was migrated by\nthe migration PTE not cleaned up. For example\n\n kernel BUG at include/linux/swapops.h:105!\n invalid opcode: 0000 [#1] PREEMPT SMP\n ...\n Call Trace:\n [\u003cffffffff810e951a\u003e] handle_mm_fault+0x3f8/0x76a\n [\u003cffffffff8130c7a2\u003e] do_page_fault+0x44a/0x46e\n [\u003cffffffff813099b5\u003e] page_fault+0x25/0x30\n [\u003cffffffff8114de33\u003e] load_elf_binary+0x152a/0x192b\n [\u003cffffffff8111329b\u003e] search_binary_handler+0x173/0x313\n [\u003cffffffff81114896\u003e] do_execve+0x219/0x30a\n [\u003cffffffff8100a5c6\u003e] sys_execve+0x43/0x5e\n [\u003cffffffff8100320a\u003e] stub_execve+0x6a/0xc0\n RIP [\u003cffffffff811094ff\u003e] migration_entry_wait+0xc1/0x129\n\nThere is a race between shift_arg_pages and migration that triggers this\nbug. A temporary stack is setup during exec and later moved. If\nmigration moves a page in the temporary stack and the VMA is then removed\nbefore migration completes, the migration PTE may not be found leading to\na BUG when the stack is faulted.\n\nThis patch causes pages within the temporary stack during exec to be\nskipped by migration. It does this by marking the VMA covering the\ntemporary stack with an otherwise impossible combination of VMA flags.\nThese flags are cleared when the temporary stack is moved to its final\nlocation.\n\n[kamezawa.hiroyu@jp.fujitsu.com: idea for having migration skip temporary stacks]\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Christoph Lameter \u003ccl@linux.com\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "34441427aab4bdb3069a4ffcda69a99357abcb2e", "tree": "3beebde910f25b0945e9105017fd743c9e5241a5", "parents": [ "3c904afd7358e9ef515eb5df36b6f25c2b7fc2da" ], "author": { "name": "Robin Holt", "email": "holt@sgi.com", "time": "Tue May 11 14:06:46 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 11 17:33:41 2010 -0700" }, "message": "revert \"procfs: provide stack information for threads\" and its fixup commits\n\nOriginally, commit d899bf7b (\"procfs: provide stack information for\nthreads\") attempted to introduce a new feature for showing where the\nthreadstack was located and how many pages are being utilized by the\nstack.\n\nCommit c44972f1 (\"procfs: disable per-task stack usage on NOMMU\") was\napplied to fix the NO_MMU case.\n\nCommit 89240ba0 (\"x86, fs: Fix x86 procfs stack information for threads on\n64-bit\") was applied to fix a bug in ia32 executables being loaded.\n\nCommit 9ebd4eba7 (\"procfs: fix /proc/\u003cpid\u003e/stat stack pointer for kernel\nthreads\") was applied to fix a bug which had kernel threads printing a\nuserland stack address.\n\nCommit 1306d603f (\u0027proc: partially revert \"procfs: provide stack\ninformation for threads\"\u0027) was then applied to revert the stack pages\nbeing used to solve a significant performance regression.\n\nThis patch nearly undoes the effect of all these patches.\n\nThe reason for reverting these is it provides an unusable value in\nfield 28. For x86_64, a fork will result in the task-\u003estack_start\nvalue being updated to the current user top of stack and not the stack\nstart address. This unpredictability of the stack_start value makes\nit worthless. That includes the intended use of showing how much stack\nspace a thread has.\n\nOther architectures will get different values. As an example, ia64\ngets 0. The do_fork() and copy_process() functions appear to treat the\nstack_start and stack_size parameters as architecture specific.\n\nI only partially reverted c44972f1 (\"procfs: disable per-task stack usage\non NOMMU\") . If I had completely reverted it, I would have had to change\nmm/Makefile only build pagewalk.o when CONFIG_PROC_PAGE_MONITOR is\nconfigured. Since I could not test the builds without significant effort,\nI decided to not change mm/Makefile.\n\nI only partially reverted 89240ba0 (\"x86, fs: Fix x86 procfs stack\ninformation for threads on 64-bit\") . I left the KSTK_ESP() change in\nplace as that seemed worthwhile.\n\nSigned-off-by: Robin Holt \u003cholt@sgi.com\u003e\nCc: Stefani Seibold \u003cstefani@seibold.net\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Michal Simek \u003cmonstr@monstr.eu\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "76595f79d76fbe6267a51b3a866a028d150f06d4", "tree": "eff36adfb339f8cb57b07fd9f09d37d982aa354e", "parents": [ "5c99cbf49a6e1a1efd25b11f4604c65c455e1612" ], "author": { "name": "Neil Horman", "email": "nhorman@tuxdriver.com", "time": "Fri Mar 05 13:44:16 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:46 2010 -0800" }, "message": "coredump: suppress uid comparison test if core output files are pipes\n\nModify uid check in do_coredump so as to not apply it in the case of\npipes.\n\nThis just got noticed in testing. The end of do_coredump validates the\nuid of the inode for the created file against the uid of the crashing\nprocess to ensure that no one can pre-create a core file with different\nownership and grab the information contained in the core when they\nshouldn\u0027 tbe able to. This causes failures when using pipes for a core\ndumps if the crashing process is not root, which is the uid of the pipe\nwhen it is created.\n\nThe fix is simple. Since the check for matching uid\u0027s isn\u0027t relevant for\npipes (a process can\u0027t create a pipe that the uermodehelper code will open\nanyway), we can just just skip it in the event ispipe is non-zero\n\nReverts a pipe-affecting change which was accidentally made in\n\n: commit c46f739dd39db3b07ab5deb4e3ec81e1c04a91af\n: Author: Ingo Molnar \u003cmingo@elte.hu\u003e\n: AuthorDate: Wed Nov 28 13:59:18 2007 +0100\n: Commit: Linus Torvalds \u003ctorvalds@woody.linux-foundation.org\u003e\n: CommitDate: Wed Nov 28 10:58:01 2007 -0800\n:\n: vfs: coredumping fix\n\nSigned-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5c99cbf49a6e1a1efd25b11f4604c65c455e1612", "tree": "e2a3451cc87f83b15f23a0472339be81770e12da", "parents": [ "30736a4d43f4af7f1a7836d6a266be17082195c4" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Fri Mar 05 13:44:14 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:46 2010 -0800" }, "message": "coredump: set -\u003egroup_exit_code for other CLONE_VM tasks too\n\nUser visible change.\n\ndo_coredump() kills all threads which share the same -\u003emm but only the\ncoredumping process gets the proper exit_code. Other tasks which share\nthe same -\u003emm die \"silently\" and return status \u003d\u003d 0 to parent.\n\nThis is historical behaviour, not actually a bug. But I think Frank\nHeckenbach rightly dislikes the current behaviour. Simple test-case:\n\n\t#include \u003cstdio.h\u003e\n\t#include \u003cunistd.h\u003e\n\t#include \u003csignal.h\u003e\n\t#include \u003csys/wait.h\u003e\n\n\tint main(void)\n\t{\n\t\tint stat;\n\n\t\tif (!fork()) {\n\t\t\tif (!vfork())\n\t\t\t\tkill(getpid(), SIGQUIT);\n\t\t}\n\n\t\twait(\u0026stat);\n\t\tprintf(\"stat\u003d%x\\n\", stat);\n\t\treturn 0;\n\t}\n\nBefore this patch it prints \"stat\u003d0\" despite the fact the child was killed\nby SIGQUIT. After this patch the output is \"stat\u003d3\" which obviously makes\nmore sense.\n\nEven with this patch, only the task which originates the coredumping gets\n\"|\u003d 0x80\" if the core was actually dumped, but at least the coredumping\nsignal is visible to do_wait/etc.\n\nReported-by: Frank Heckenbach \u003cf.heckenbach@fh-soft.de\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "30736a4d43f4af7f1a7836d6a266be17082195c4", "tree": "54e388cdaf4795bdb7651be03c50bba67eb0b43a", "parents": [ "8d9032bbe4671dc481261ccd4e161cd96e54b118" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@redhat.com", "time": "Fri Mar 05 13:44:12 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:46 2010 -0800" }, "message": "coredump: pass mm-\u003eflags as a coredump parameter for consistency\n\nPass mm-\u003eflags as a coredump parameter for consistency.\n\n ---\n1787 if (mm-\u003ecore_state || !get_dumpable(mm)) { \u003c- (1)\n1788 up_write(\u0026mm-\u003emmap_sem);\n1789 put_cred(cred);\n1790 goto fail;\n1791 }\n1792\n[...]\n1798 if (get_dumpable(mm) \u003d\u003d 2) { /* Setuid core dump mode */ \u003c-(2)\n1799 flag \u003d O_EXCL; /* Stop rewrite attacks */\n1800 cred-\u003efsuid \u003d 0; /* Dump root private */\n1801 }\n ---\n\nSince dumpable bits are not protected by lock, there is a chance to change\nthese bits between (1) and (2).\n\nTo solve this issue, this patch copies mm-\u003eflags to\ncoredump_params.mm_flags at the beginning of do_coredump() and uses it\ninstead of get_dumpable() while dumping core.\n\nThis copy is also passed to binfmt-\u003ecore_dump, since elf*_core_dump() uses\ndump_filter bits in mm-\u003eflags.\n\n[akpm@linux-foundation.org: fix merge]\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5ef097dd7ba4eab8b4f0026d85fcef9fe23b821f", "tree": "5d2e580897795b8eae278f7548799119ac4803a0", "parents": [ "9728e5d6e6c432ee8487c63ce6e479e2474d9945" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Fri Mar 05 13:42:57 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:33 2010 -0800" }, "message": "exec: create initial stack independent of PAGE_SIZE\n\nCurrently we create the initial stack based on the PAGE_SIZE. This is\nunnecessary.\n\nThis creates this initial stack independent of the PAGE_SIZE.\n\nIt also bumps up the number of 4k pages allocated from 20 to 32, to\nalign with 64K page systems.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nCc: Helge Deller \u003cdeller@gmx.de\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Americo Wang \u003cxiyou.wangcong@gmail.com\u003e\nCc: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d554ed895dc8f293cc712c71f14b101ace82579a", "tree": "fd29ba33d640e156e2ca4267ffbdb9dc566fd3d6", "parents": [ "221e3ebf6d5f2625373573155924e39f196c5d3d" ], "author": { "name": "Jiri Slaby", "email": "jslaby@suse.cz", "time": "Fri Mar 05 13:42:42 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:29 2010 -0800" }, "message": "fs: use rlimit helpers\n\nMake sure compiler won\u0027t do weird things with limits. E.g. fetching them\ntwice may return 2 different values after writable limits are implemented.\n\nI.e. either use rlimit helpers added in commit 3e10e716abf3 (\"resource:\nadd helpers for fetching rlimits\") or ACCESS_ONCE if not applicable.\n\nSigned-off-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5beb49305251e5669852ed541e8e2f2f7696c53e", "tree": "46457450a22f23938b24904aeba5d4ada2f53b20", "parents": [ "648bcc771145172a14bc35eeb849ed08f6aa4f1e" ], "author": { "name": "Rik van Riel", "email": "riel@redhat.com", "time": "Fri Mar 05 13:42:07 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:26 2010 -0800" }, "message": "mm: change anon_vma linking to fix multi-process server scalability issue\n\nThe old anon_vma code can lead to scalability issues with heavily forking\nworkloads. Specifically, each anon_vma will be shared between the parent\nprocess and all its child processes.\n\nIn a workload with 1000 child processes and a VMA with 1000 anonymous\npages per process that get COWed, this leads to a system with a million\nanonymous pages in the same anon_vma, each of which is mapped in just one\nof the 1000 processes. However, the current rmap code needs to walk them\nall, leading to O(N) scanning complexity for each page.\n\nThis can result in systems where one CPU is walking the page tables of\n1000 processes in page_referenced_one, while all other CPUs are stuck on\nthe anon_vma lock. This leads to catastrophic failure for a benchmark\nlike AIM7, where the total number of processes can reach in the tens of\nthousands. Real workloads are still a factor 10 less process intensive\nthan AIM7, but they are catching up.\n\nThis patch changes the way anon_vmas and VMAs are linked, which allows us\nto associate multiple anon_vmas with a VMA. At fork time, each child\nprocess gets its own anon_vmas, in which its COWed pages will be\ninstantiated. The parents\u0027 anon_vma is also linked to the VMA, because\nnon-COWed pages could be present in any of the children.\n\nThis reduces rmap scanning complexity to O(1) for the pages of the 1000\nchild processes, with O(N) complexity for at most 1/N pages in the system.\n This reduces the average scanning cost in heavily forking workloads from\nO(N) to 2.\n\nThe only real complexity in this patch stems from the fact that linking a\nVMA to anon_vmas now involves memory allocations. This means vma_adjust\ncan fail, if it needs to attach a VMA to anon_vma structures. This in\nturn means error handling needs to be added to the calling functions.\n\nA second source of complexity is that, because there can be multiple\nanon_vmas, the anon_vma linking in vma_adjust can no longer be done under\n\"the\" anon_vma lock. To prevent the rmap code from walking up an\nincomplete VMA, this patch introduces the VM_LOCK_RMAP VMA flag. This bit\nflag uses the same slot as the NOMMU VM_MAPPED_COPY, with an ifdef in mm.h\nto make sure it is impossible to compile a kernel that needs both symbolic\nvalues for the same bitflag.\n\nSome test results:\n\nWithout the anon_vma changes, when AIM7 hits around 9.7k users (on a test\nbox with 16GB RAM and not quite enough IO), the system ends up running\n\u003e99% in system time, with every CPU on the same anon_vma lock in the\npageout code.\n\nWith these changes, AIM7 hits the cross-over point around 29.7k users.\nThis happens with ~99% IO wait time, there never seems to be any spike in\nsystem time. The anon_vma lock contention appears to be resolved.\n\n[akpm@linux-foundation.org: cleanups]\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Larry Woodman \u003clwoodman@redhat.com\u003e\nCc: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Hugh Dickins \u003chugh.dickins@tiscali.co.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "34e55232e59f7b19050267a05ff1226e5cd122a5", "tree": "6b94e776e87d2a2fe1ceca7c5606901575323900", "parents": [ "d559db086ff5be9bcc259e5aa50bf3d881eaf1d1" ], "author": { "name": "KAMEZAWA Hiroyuki", "email": "kamezawa.hiroyu@jp.fujitsu.com", "time": "Fri Mar 05 13:41:40 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:24 2010 -0800" }, "message": "mm: avoid false sharing of mm_counter\n\nConsidering the nature of per mm stats, it\u0027s the shared object among\nthreads and can be a cache-miss point in the page fault path.\n\nThis patch adds per-thread cache for mm_counter. RSS value will be\ncounted into a struct in task_struct and synchronized with mm\u0027s one at\nevents.\n\nNow, in this patch, the event is the number of calls to handle_mm_fault.\nPer-thread value is added to mm at each 64 calls.\n\n rough estimation with small benchmark on parallel thread (2threads) shows\n [before]\n 4.5 cache-miss/faults\n [after]\n 4.0 cache-miss/faults\n Anyway, the most contended object is mmap_sem if the number of threads grows.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a17e18790a8c47113a73139d54a375dc9ccd8f08", "tree": "2f85b0265e004826d76614f27c133063499b855e", "parents": [ "89f3f2199084a160a3a45fa6d9af235696321758" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Mon Feb 22 12:44:24 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 22 19:50:34 2010 -0800" }, "message": "fs/exec.c: fix initial stack reservation\n\n803bf5ec259941936262d10ecc84511b76a20921 (\"fs/exec.c: restrict initial\nstack space expansion to rlimit\") attempts to limit the initial stack to\n20*PAGE_SIZE. Unfortunately, in attempting ensure the stack is not\nreduced in size, we ended up not changing the stack at all.\n\nThis size reduction check is not necessary as the expand_stack call does\nthis already.\n\nThis caused a regression in UML resulting in most guest processes being\nkilled.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nCc: Anton Blanchard \u003canton@samba.org\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Jouni Malinen \u003cj@w1.fi\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "803bf5ec259941936262d10ecc84511b76a20921", "tree": "ed159cfdf254f6b139ffc6b4a1e8437c6362b474", "parents": [ "4cfbafd33f5ae99688ab82525a1d449c1c1b198f" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Feb 10 13:56:42 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 11 13:59:43 2010 -0800" }, "message": "fs/exec.c: restrict initial stack space expansion to rlimit\n\nWhen reserving stack space for a new process, make sure we\u0027re not\nattempting to expand the stack by more than rlimit allows.\n\nThis fixes a bug caused by b6a2fea39318e43fee84fa7b0b90d68bed92d2ba (\"mm:\nvariable length argument support\") and unmasked by\nfc63cf237078c86214abcb2ee9926d8ad289da9b (\"exec: setup_arg_pages() fails\nto return errors\").\n\nThis bug means that when limiting the stack to less the 20*PAGE_SIZE (eg.\n80K on 4K pages or \u0027ulimit -s 79\u0027) all processes will be killed before\nthey start. This is particularly bad with 64K pages, where a ulimit below\n1280K will kill every process.\n\nTo test, do:\n\n \u0027ulimit -s 15; ls\u0027\n\nbefore and after the patch is applied. Before it\u0027s applied, \u0027ls\u0027 should\nbe killed. After the patch is applied, \u0027ls\u0027 should no longer be killed.\n\nA stack limit of 15KB since it\u0027s small enough to trigger 20*PAGE_SIZE.\nAlso 15KB not a multiple of PAGE_SIZE, which is a trickier case to handle\ncorrectly with this code.\n\n4K pages should be fine to test with.\n\n[kosaki.motohiro@jp.fujitsu.com: cleanup]\n[akpm@linux-foundation.org: cleanup cleanup]\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Americo Wang \u003cxiyou.wangcong@gmail.com\u003e\nCc: Anton Blanchard \u003canton@samba.org\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7ab02af428c2d312c0cf8fb0b01cc1eb21131a3d", "tree": "dcb85f8a77f7919d4701bddb485a32bdfbda5852", "parents": [ "ab658321f32770b903a4426e2a6fae0392757755" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 02 12:37:44 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 02 12:37:44 2010 -0800" }, "message": "Fix \u0027flush_old_exec()/setup_new_exec()\u0027 split\n\nCommit 221af7f87b9 (\"Split \u0027flush_old_exec\u0027 into two functions\") split\nthe function at the point of no return - ie right where there were no\nmore error cases to check. That made sense from a technical standpoint,\nbut when we then also combined it with the actual personality setting\ngoing in between flush_old_exec() and setup_new_exec(), it needs to be a\nbit more careful.\n\nIn particular, we need to make sure that we really flush the old\npersonality bits in the \u0027flush\u0027 stage, rather than later in the \u0027setup\u0027\nstage, since otherwise we might be flushing the _new_ personality state\nthat we\u0027re just setting up.\n\nSo this moves the flags and personality flushing (and \u0027flush_thread()\u0027,\nwhich is the arch-specific function that generally resets lazy FP state\netc) of the old process into flush_old_exec(), so that it doesn\u0027t affect\nany state that execve() is setting up for the new process environment.\n\nThis was reported by Michal Simek as breaking his Microblaze qemu\nenvironment.\n\nReported-and-tested-by: Michal Simek \u003cmichal.simek@petalogix.com\u003e\nCc: Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "221af7f87b97431e3ee21ce4b0e77d5411cf1549", "tree": "480126aada06d87c09cb62e7c8fa292572438c18", "parents": [ "64a028a6de08545a2c94f302bc7694bf48aee5b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 28 22:14:42 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 29 08:22:01 2010 -0800" }, "message": "Split \u0027flush_old_exec\u0027 into two functions\n\n\u0027flush_old_exec()\u0027 is the point of no return when doing an execve(), and\nit is pretty badly misnamed. It doesn\u0027t just flush the old executable\nenvironment, it also starts up the new one.\n\nWhich is very inconvenient for things like setting up the new\npersonality, because we want the new personality to affect the starting\nof the new environment, but at the same time we do _not_ want the new\npersonality to take effect if flushing the old one fails.\n\nAs a result, the x86-64 \u002732-bit\u0027 personality is actually done using this\ninsane \"I\u0027m going to change the ABI, but I haven\u0027t done it yet\" bit\n(TIF_ABI_PENDING), with SET_PERSONALITY() not actually setting the\npersonality, but just the \"pending\" bit, so that \"flush_thread()\" can do\nthe actual personality magic.\n\nThis patch in no way changes any of that insanity, but it does split the\n\u0027flush_old_exec()\u0027 function up into a preparatory part that can fail\n(still called flush_old_exec()), and a new part that will actually set\nup the new exec environment (setup_new_exec()). All callers are changed\nto trivially comply with the new world order.\n\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f6151dfea21496d43dbaba32cfcd9c9f404769bc", "tree": "092b73a1cd7edbd0e04fe2076ed678447446aa70", "parents": [ "9cd80bbb07fcd6d4d037fad4297496d3b132ac6b" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@redhat.com", "time": "Thu Dec 17 15:27:16 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Dec 17 15:45:31 2009 -0800" }, "message": "mm: introduce coredump parameter structure\n\nIntroduce coredump parameter data structure (struct coredump_params) to\nsimplify binfmt-\u003ecore_dump() arguments.\n\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@redhat.com\u003e\nSuggested-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9cd80bbb07fcd6d4d037fad4297496d3b132ac6b", "tree": "322216d34b6d5ba2366f53ba0d677950c4cde766", "parents": [ "0f67b0b0392ccca98459bf40b36c0037793b5f71" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Thu Dec 17 15:27:15 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Dec 17 15:45:31 2009 -0800" }, "message": "do_wait() optimization: do not place sub-threads on task_struct-\u003echildren list\n\nThanks to Roland who pointed out de_thread() issues.\n\nCurrently we add sub-threads to -\u003ereal_parent-\u003echildren list. This buys\nnothing but slows down do_wait().\n\nWith this patch -\u003echildren contains only main threads (group leaders).\nThe only complication is that forget_original_parent() should iterate over\nsub-threads by hand, and de_thread() needs another list_replace() when it\nchanges -\u003egroup_leader.\n\nHenceforth do_wait_thread() can never see task_detached() \u0026\u0026 !EXIT_DEAD\ntasks, we can remove this check (and we can unify do_wait_thread() and\nptrace_do_wait()).\n\nThis change can confuse the optimistic search in mm_update_next_owner(),\nbut this is fixable and minor.\n\nPerhaps badness() and oom_kill_process() should be updated, but they\nshould be fixed in any case.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Ratan Nalumasu \u003crnalumasu@gmail.com\u003e\nCc: Vitaly Mayatskikh \u003cvmayatsk@redhat.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "4614a696bd1c3a9af3a08f0e5874830a85b889d4", "tree": "0b363d89ce612ab8557d49ab8b0f49b8520ffa66", "parents": [ "7e1e0ef22c9ba9f797d7c4448feee722584bba5c" ], "author": { "name": "john stultz", "email": "johnstul@us.ibm.com", "time": "Mon Dec 14 18:00:05 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Dec 15 08:53:24 2009 -0800" }, "message": "procfs: allow threads to rename siblings via /proc/pid/tasks/tid/comm\n\nSetting a thread\u0027s comm to be something unique is a very useful ability\nand is helpful for debugging complicated threaded applications. However\ncurrently the only way to set a thread name is for the thread to name\nitself via the PR_SET_NAME prctl.\n\nHowever, there may be situations where it would be advantageous for a\nthread dispatcher to be naming the threads its managing, rather then\nhaving the threads self-describe themselves. This sort of behavior is\navailable on other systems via the pthread_setname_np() interface.\n\nThis patch exports a task\u0027s comm via proc/pid/comm and\nproc/pid/task/tid/comm interfaces, and allows thread siblings to write to\nthese values.\n\n[akpm@linux-foundation.org: cleanups]\nSigned-off-by: John Stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Arjan van de Ven \u003carjan@infradead.org\u003e\nCc: Mike Fulton \u003cfultonm@ca.ibm.com\u003e\nCc: Sean Foley \u003cSean_Foley@ca.ibm.com\u003e\nCc: Darren Hart \u003cdvhltc@us.ibm.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c84d6efd363a3948eb32ec40d46bab6338580454", "tree": "3ba7ac46e6626fe8ac843834588609eb6ccee5c6", "parents": [ "7539cf4b92be4aecc573ea962135f246a7a33401", "22763c5cf3690a681551162c15d34d935308c8d7" ], "author": { "name": "James Morris", "email": "jmorris@macbook.(none)", "time": "Thu Dec 03 12:03:40 2009 +0530" }, "committer": { "name": "James Morris", "email": "jmorris@macbook.(none)", "time": "Thu Dec 03 12:03:40 2009 +0530" }, "message": "Merge branch \u0027master\u0027 into next\n" }, { "commit": "fc63cf237078c86214abcb2ee9926d8ad289da9b", "tree": "3af8571390aec334b7f47c08ba0cff0a72d6fbb3", "parents": [ "93f40e6f1493ae0c1e04760b221c125ef5e47205" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Nov 11 14:26:48 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 12 07:25:58 2009 -0800" }, "message": "exec: setup_arg_pages() fails to return errors\n\nIn setup_arg_pages we work hard to assign a value to ret, but on exit we\nalways return 0.\n\nAlso remove a now duplicated exit path and branch to out_unlock instead.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nReviewed-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6c21a7fb492bf7e2c4985937082ce58ddeca84bd", "tree": "6cfe11ba4b8eee26ee8b02d2b4a5fcc6ea07e4bd", "parents": [ "6e8e16c7bc298d7887584c3d027e05db3e86eed9" ], "author": { "name": "Mimi Zohar", "email": "zohar@linux.vnet.ibm.com", "time": "Thu Oct 22 17:30:13 2009 -0400" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Sun Oct 25 12:22:48 2009 +0800" }, "message": "LSM: imbed ima calls in the security hooks\n\nBased on discussions on LKML and LSM, where there are consecutive\nsecurity_ and ima_ calls in the vfs layer, move the ima_ calls to\nthe existing security_ hooks.\n\nSigned-off-by: Mimi Zohar \u003czohar@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "801460d0cf5c5288153b722565773059b0f44348", "tree": "ff2542903572a546ac59e92f295e9a18e718afa0", "parents": [ "2fa4341074cd02fb39aa23410740764948755635" ], "author": { "name": "Hiroshi Shimamoto", "email": "h-shimamoto@ct.jp.nec.com", "time": "Wed Sep 23 15:57:41 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:21:05 2009 -0700" }, "message": "task_struct cleanup: move binfmt field to mm_struct\n\nBecause the binfmt is not different between threads in the same process,\nit can be moved from task_struct to mm_struct. And binfmt moudle is\nhandled per mm_struct instead of task_struct.\n\nSigned-off-by: Hiroshi Shimamoto \u003ch-shimamoto@ct.jp.nec.com\u003e\nAcked-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "964ee7df90d799e38fb1556c57cd5c45fc736436", "tree": "58ae7325f80394cb86409d5876c710376dde8e1a", "parents": [ "61be228a06dc6e8662f30e89eda3c12083c1f379" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed Sep 23 15:56:59 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:21:01 2009 -0700" }, "message": "exec: fix set_binfmt() vs sys_delete_module() race\n\nsys_delete_module() can set MODULE_STATE_GOING after\nsearch_binary_handler() does try_module_get(). In this case\nset_binfmt()-\u003etry_module_get() fails but since none of the callers\ncheck the returned error, the task will run with the wrong old\n-\u003ebinfmt.\n\nThe proper fix should change all -\u003eload_binary() methods, but we can\nrely on fact that the caller must hold a reference to binfmt-\u003emodule\nand use __module_get() which never fails.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Hiroshi Shimamoto \u003ch-shimamoto@ct.jp.nec.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "61be228a06dc6e8662f30e89eda3c12083c1f379", "tree": "be481a4f2dcb2ab4fded8d72d42601e360fc89df", "parents": [ "a293980c2e261bd5b0d2a77340dd04f684caff58" ], "author": { "name": "Neil Horman", "email": "nhorman@tuxdriver.com", "time": "Wed Sep 23 15:56:58 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:21:00 2009 -0700" }, "message": "exec: allow do_coredump() to wait for user space pipe readers to complete\n\nAllow core_pattern pipes to wait for user space to complete\n\nOne of the things that user space processes like to do is look at metadata\nfor a crashing process in their /proc/\u003cpid\u003e directory. this is racy\nhowever, since do_coredump in the kernel doesn\u0027t wait for the user space\nprocess to complete before it reaps the crashing process. This patch\ncorrects that. Allowing the kernel to wait for the user space process to\ncomplete before cleaning up the crashing process. This is a bit tricky to\ndo for a few reasons:\n\n1) The user space process isn\u0027t our child, so we can\u0027t sys_wait4 on it\n2) We need to close the pipe before waiting for the user process to complete,\nsince the user process may rely on an EOF condition\n\nI\u0027ve discussed several solutions with Oleg Nesterov off-list about this,\nand this is the one we\u0027ve come up with. We add ourselves as a pipe reader\n(to prevent premature cleanup of the pipe_inode_info), and remove\nourselves as a writer (to provide an EOF condition to the writer in user\nspace), then we iterate until the user space process exits (which we\ndetect by pipe-\u003ereaders \u003d\u003d 1, hence the \u003e 1 check in the loop). When we\nexit the loop, we restore the proper reader/writer values, then we return\nand let filp_close in do_coredump clean up the pipe data properly.\n\nSigned-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nReported-by: Earl Chew \u003cearl_chew@agilent.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a293980c2e261bd5b0d2a77340dd04f684caff58", "tree": "69f2b2747548deafb146c18953ebce21ae1f02d4", "parents": [ "725eae32df7754044809973034429a47e6035158" ], "author": { "name": "Neil Horman", "email": "nhorman@tuxdriver.com", "time": "Wed Sep 23 15:56:56 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:21:00 2009 -0700" }, "message": "exec: let do_coredump() limit the number of concurrent dumps to pipes\n\nIntroduce core pipe limiting sysctl.\n\nSince we can dump cores to pipe, rather than directly to the filesystem,\nwe create a condition in which a user can create a very high load on the\nsystem simply by running bad applications.\n\nIf the pipe reader specified in core_pattern is poorly written, we can\nhave lots of ourstandig resources and processes in the system.\n\nThis sysctl introduces an ability to limit that resource consumption.\ncore_pipe_limit defines how many in-flight dumps may be run in parallel,\ndumps beyond this value are skipped and a note is made in the kernel log.\nA special value of 0 in core_pipe_limit denotes unlimited core dumps may\nbe handled (this is the default value).\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nReported-by: Earl Chew \u003cearl_chew@agilent.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "725eae32df7754044809973034429a47e6035158", "tree": "96a66c65a085940a19374ee02c26fb8f1931a6be", "parents": [ "ae6d2ed7bb3877ff35b9569402025f40ea2e1803" ], "author": { "name": "Neil Horman", "email": "nhorman@tuxdriver.com", "time": "Wed Sep 23 15:56:54 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:21:00 2009 -0700" }, "message": "exec: make do_coredump() more resilient to recursive crashes\n\nChange how we detect recursive dumps.\n\nCurrently we have a mechanism by which we try to compare pathnames of the\ncrashing process to the core_pattern path. This is broken for a dozen\nreasons, and just doesn\u0027t work in any sort of robust way.\n\nI\u0027m replacing it with the use of a 0 RLIMIT_CORE value. Since helper apps\nset RLIMIT_CORE to zero, we don\u0027t write out core files for any process\nwith that particular limit set. It the core_pattern is a pipe, any\nnon-zero limit is translated to RLIM_INFINITY.\n\nThis allows complete dumps to be captured, but prevents infinite recursion\nin the event that the core_pattern process itself crashes.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nReported-by: Earl Chew \u003cearl_chew@agilent.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d899bf7b55f503ba7d3d07ed27c3a37e270fa7db", "tree": "32a5ee7816b2f0cb3261dcca8102b9cafe9251bd", "parents": [ "cba8aafe1e07dfc8bae5ba78be8e02883bd34d31" ], "author": { "name": "Stefani Seibold", "email": "stefani@seibold.net", "time": "Tue Sep 22 16:45:40 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 23 07:39:41 2009 -0700" }, "message": "procfs: provide stack information for threads\n\nA patch to give a better overview of the userland application stack usage,\nespecially for embedded linux.\n\nCurrently you are only able to dump the main process/thread stack usage\nwhich is showed in /proc/pid/status by the \"VmStk\" Value. But you get no\ninformation about the consumed stack memory of the the threads.\n\nThere is an enhancement in the /proc/\u003cpid\u003e/{task/*,}/*maps and which marks\nthe vm mapping where the thread stack pointer reside with \"[thread stack\nxxxxxxxx]\". xxxxxxxx is the maximum size of stack. This is a value\ninformation, because libpthread doesn\u0027t set the start of the stack to the\ntop of the mapped area, depending of the pthread usage.\n\nA sample output of /proc/\u003cpid\u003e/task/\u003ctid\u003e/maps looks like:\n\n08048000-08049000 r-xp 00000000 03:00 8312 /opt/z\n08049000-0804a000 rw-p 00001000 03:00 8312 /opt/z\n0804a000-0806b000 rw-p 00000000 00:00 0 [heap]\na7d12000-a7d13000 ---p 00000000 00:00 0\na7d13000-a7f13000 rw-p 00000000 00:00 0 [thread stack: 001ff4b4]\na7f13000-a7f14000 ---p 00000000 00:00 0\na7f14000-a7f36000 rw-p 00000000 00:00 0\na7f36000-a8069000 r-xp 00000000 03:00 4222 /lib/libc.so.6\na8069000-a806b000 r--p 00133000 03:00 4222 /lib/libc.so.6\na806b000-a806c000 rw-p 00135000 03:00 4222 /lib/libc.so.6\na806c000-a806f000 rw-p 00000000 00:00 0\na806f000-a8083000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0\na8083000-a8084000 r--p 00013000 03:00 14462 /lib/libpthread.so.0\na8084000-a8085000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0\na8085000-a8088000 rw-p 00000000 00:00 0\na8088000-a80a4000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2\na80a4000-a80a5000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2\na80a5000-a80a6000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2\nafaf5000-afb0a000 rw-p 00000000 00:00 0 [stack]\nffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]\n\nAlso there is a new entry \"stack usage\" in /proc/\u003cpid\u003e/{task/*,}/status\nwhich will you give the current stack usage in kb.\n\nA sample output of /proc/self/status looks like:\n\nName:\tcat\nState:\tR (running)\nTgid:\t507\nPid:\t507\n.\n.\n.\nCapBnd:\tfffffffffffffeff\nvoluntary_ctxt_switches:\t0\nnonvoluntary_ctxt_switches:\t0\nStack usage:\t12 kB\n\nI also fixed stack base address in /proc/\u003cpid\u003e/{task/*,}/stat to the base\naddress of the associated thread stack and not the one of the main\nprocess. This makes more sense.\n\n[akpm@linux-foundation.org: fs/proc/array.c now needs walk_page_range()]\nSigned-off-by: Stefani Seibold \u003cstefani@seibold.net\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1f10206cf8e945220f7220a809d8bfc15c21f9a5", "tree": "4f2c48ccf9cc52dad9610df54e301e631a351288", "parents": [ "b28cfd2c0616e1b42acc6ee3c77ef6cc3873c510" ], "author": { "name": "Jiri Pirko", "email": "jpirko@redhat.com", "time": "Tue Sep 22 16:44:10 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 23 07:39:30 2009 -0700" }, "message": "getrusage: fill ru_maxrss value\n\nMake -\u003eru_maxrss value in struct rusage filled accordingly to rss hiwater\nmark. This struct is filled as a parameter to getrusage syscall.\n-\u003eru_maxrss value is set to KBs which is the way it is done in BSD\nsystems. /usr/bin/time (gnu time) application converts -\u003eru_maxrss to KBs\nwhich seems to be incorrect behavior. Maintainer of this util was\nnotified by me with the patch which corrects it and cc\u0027ed.\n\nTo make this happen we extend struct signal_struct by two fields. The\nfirst one is -\u003emaxrss which we use to store rss hiwater of the task. The\nsecond one is -\u003ecmaxrss which we use to store highest rss hiwater of all\ntask childs. These values are used in k_getrusage() to actually fill\n-\u003eru_maxrss. k_getrusage() uses current rss hiwater value directly if mm\nstruct exists.\n\nNote:\nexec() clear mm-\u003ehiwater_rss, but doesn\u0027t clear sig-\u003emaxrss.\nit is intetionally behavior. *BSD getrusage have exec() inheriting.\n\ntest programs\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\ngetrusage.c\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003cstring.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/time.h\u003e\n #include \u003csys/resource.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/wait.h\u003e\n #include \u003cunistd.h\u003e\n #include \u003csignal.h\u003e\n #include \u003csys/mman.h\u003e\n\n #include \"common.h\"\n\n #define err(str) perror(str), exit(1)\n\nint main(int argc, char** argv)\n{\n\tint status;\n\n\tprintf(\"allocate 100MB\\n\");\n\tconsume(100);\n\n\tprintf(\"testcase1: fork inherit? \\n\");\n\tprintf(\" expect: initial.self ~\u003d child.self\\n\");\n\tshow_rusage(\"initial\");\n\tif (__fork()) {\n\t\twait(\u0026status);\n\t} else {\n\t\tshow_rusage(\"fork child\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\n\tprintf(\"testcase2: fork inherit? (cont.) \\n\");\n\tprintf(\" expect: initial.children ~\u003d 100MB, but child.children \u003d 0\\n\");\n\tshow_rusage(\"initial\");\n\tif (__fork()) {\n\t\twait(\u0026status);\n\t} else {\n\t\tshow_rusage(\"child\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\n\tprintf(\"testcase3: fork + malloc \\n\");\n\tprintf(\" expect: child.self ~\u003d initial.self + 50MB\\n\");\n\tshow_rusage(\"initial\");\n\tif (__fork()) {\n\t\twait(\u0026status);\n\t} else {\n\t\tprintf(\"allocate +50MB\\n\");\n\t\tconsume(50);\n\t\tshow_rusage(\"fork child\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\n\tprintf(\"testcase4: grandchild maxrss\\n\");\n\tprintf(\" expect: post_wait.children ~\u003d 300MB\\n\");\n\tshow_rusage(\"initial\");\n\tif (__fork()) {\n\t\twait(\u0026status);\n\t\tshow_rusage(\"post_wait\");\n\t} else {\n\t\tsystem(\"./child -n 0 -g 300\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\n\tprintf(\"testcase5: zombie\\n\");\n\tprintf(\" expect: pre_wait ~\u003d initial, IOW the zombie process is not accounted.\\n\");\n\tprintf(\" post_wait ~\u003d 400MB, IOW wait() collect child\u0027s max_rss. \\n\");\n\tshow_rusage(\"initial\");\n\tif (__fork()) {\n\t\tsleep(1); /* children become zombie */\n\t\tshow_rusage(\"pre_wait\");\n\t\twait(\u0026status);\n\t\tshow_rusage(\"post_wait\");\n\t} else {\n\t\tsystem(\"./child -n 400\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\n\tprintf(\"testcase6: SIG_IGN\\n\");\n\tprintf(\" expect: initial ~\u003d after_zombie (child\u0027s 500MB alloc should be ignored).\\n\");\n\tshow_rusage(\"initial\");\n\tsignal(SIGCHLD, SIG_IGN);\n\tif (__fork()) {\n\t\tsleep(1); /* children become zombie */\n\t\tshow_rusage(\"after_zombie\");\n\t} else {\n\t\tsystem(\"./child -n 500\");\n\t\t_exit(0);\n\t}\n\tprintf(\"\\n\");\n\tsignal(SIGCHLD, SIG_DFL);\n\n\tprintf(\"testcase7: exec (without fork) \\n\");\n\tprintf(\" expect: initial ~\u003d exec \\n\");\n\tshow_rusage(\"initial\");\n\texecl(\"./child\", \"child\", \"-v\", NULL);\n\n\treturn 0;\n}\n\nchild.c\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n #include \u003csys/types.h\u003e\n #include \u003cunistd.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/wait.h\u003e\n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003cstring.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/time.h\u003e\n #include \u003csys/resource.h\u003e\n\n #include \"common.h\"\n\nint main(int argc, char** argv)\n{\n\tint status;\n\tint c;\n\tlong consume_size \u003d 0;\n\tlong grandchild_consume_size \u003d 0;\n\tint show \u003d 0;\n\n\twhile ((c \u003d getopt(argc, argv, \"n:g:v\")) !\u003d -1) {\n\t\tswitch (c) {\n\t\tcase \u0027n\u0027:\n\t\t\tconsume_size \u003d atol(optarg);\n\t\t\tbreak;\n\t\tcase \u0027v\u0027:\n\t\t\tshow \u003d 1;\n\t\t\tbreak;\n\t\tcase \u0027g\u0027:\n\n\t\t\tgrandchild_consume_size \u003d atol(optarg);\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tbreak;\n\t\t}\n\t}\n\n\tif (show)\n\t\tshow_rusage(\"exec\");\n\n\tif (consume_size) {\n\t\tprintf(\"child alloc %ldMB\\n\", consume_size);\n\t\tconsume(consume_size);\n\t}\n\n\tif (grandchild_consume_size) {\n\t\tif (fork()) {\n\t\t\twait(\u0026status);\n\t\t} else {\n\t\t\tprintf(\"grandchild alloc %ldMB\\n\", grandchild_consume_size);\n\t\t\tconsume(grandchild_consume_size);\n\n\t\t\texit(0);\n\t\t}\n\t}\n\n\treturn 0;\n}\n\ncommon.c\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003cstring.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/time.h\u003e\n #include \u003csys/resource.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/wait.h\u003e\n #include \u003cunistd.h\u003e\n #include \u003csignal.h\u003e\n #include \u003csys/mman.h\u003e\n\n #include \"common.h\"\n #define err(str) perror(str), exit(1)\n\nvoid show_rusage(char *prefix)\n{\n \tint err, err2;\n \tstruct rusage rusage_self;\n \tstruct rusage rusage_children;\n\n \tprintf(\"%s: \", prefix);\n \terr \u003d getrusage(RUSAGE_SELF, \u0026rusage_self);\n \tif (!err)\n \t\tprintf(\"self %ld \", rusage_self.ru_maxrss);\n \terr2 \u003d getrusage(RUSAGE_CHILDREN, \u0026rusage_children);\n \tif (!err2)\n \t\tprintf(\"children %ld \", rusage_children.ru_maxrss);\n\n \tprintf(\"\\n\");\n}\n\n/* Some buggy OS need this worthless CPU waste. */\nvoid make_pagefault(void)\n{\n\tvoid *addr;\n\tint size \u003d getpagesize();\n\tint i;\n\n\tfor (i\u003d0; i\u003c1000; i++) {\n\t\taddr \u003d mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, -1, 0);\n\t\tif (addr \u003d\u003d MAP_FAILED)\n\t\t\terr(\"make_pagefault\");\n\t\tmemset(addr, 0, size);\n\t\tmunmap(addr, size);\n\t}\n}\n\nvoid consume(int mega)\n{\n \tsize_t sz \u003d mega * 1024 * 1024;\n \tvoid *ptr;\n\n \tptr \u003d malloc(sz);\n \tmemset(ptr, 0, sz);\n\tmake_pagefault();\n}\n\npid_t __fork(void)\n{\n\tpid_t pid;\n\n\tpid \u003d fork();\n\tmake_pagefault();\n\n\treturn pid;\n}\n\ncommon.h\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nvoid show_rusage(char *prefix);\nvoid make_pagefault(void);\nvoid consume(int mega);\npid_t __fork(void);\n\nFreeBSD result (expected result)\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nallocate 100MB\ntestcase1: fork inherit?\n expect: initial.self ~\u003d child.self\ninitial: self 103492 children 0\nfork child: self 103540 children 0\n\ntestcase2: fork inherit? (cont.)\n expect: initial.children ~\u003d 100MB, but child.children \u003d 0\ninitial: self 103540 children 103540\nchild: self 103564 children 0\n\ntestcase3: fork + malloc\n expect: child.self ~\u003d initial.self + 50MB\ninitial: self 103564 children 103564\nallocate +50MB\nfork child: self 154860 children 0\n\ntestcase4: grandchild maxrss\n expect: post_wait.children ~\u003d 300MB\ninitial: self 103564 children 154860\ngrandchild alloc 300MB\npost_wait: self 103564 children 308720\n\ntestcase5: zombie\n expect: pre_wait ~\u003d initial, IOW the zombie process is not accounted.\n post_wait ~\u003d 400MB, IOW wait() collect child\u0027s max_rss.\ninitial: self 103564 children 308720\nchild alloc 400MB\npre_wait: self 103564 children 308720\npost_wait: self 103564 children 411312\n\ntestcase6: SIG_IGN\n expect: initial ~\u003d after_zombie (child\u0027s 500MB alloc should be ignored).\ninitial: self 103564 children 411312\nchild alloc 500MB\nafter_zombie: self 103624 children 411312\n\ntestcase7: exec (without fork)\n expect: initial ~\u003d exec\ninitial: self 103624 children 411312\nexec: self 103624 children 411312\n\nLinux result (actual test result)\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nallocate 100MB\ntestcase1: fork inherit?\n expect: initial.self ~\u003d child.self\ninitial: self 102848 children 0\nfork child: self 102572 children 0\n\ntestcase2: fork inherit? (cont.)\n expect: initial.children ~\u003d 100MB, but child.children \u003d 0\ninitial: self 102876 children 102644\nchild: self 102572 children 0\n\ntestcase3: fork + malloc\n expect: child.self ~\u003d initial.self + 50MB\ninitial: self 102876 children 102644\nallocate +50MB\nfork child: self 153804 children 0\n\ntestcase4: grandchild maxrss\n expect: post_wait.children ~\u003d 300MB\ninitial: self 102876 children 153864\ngrandchild alloc 300MB\npost_wait: self 102876 children 307536\n\ntestcase5: zombie\n expect: pre_wait ~\u003d initial, IOW the zombie process is not accounted.\n post_wait ~\u003d 400MB, IOW wait() collect child\u0027s max_rss.\ninitial: self 102876 children 307536\nchild alloc 400MB\npre_wait: self 102876 children 307536\npost_wait: self 102876 children 410076\n\ntestcase6: SIG_IGN\n expect: initial ~\u003d after_zombie (child\u0027s 500MB alloc should be ignored).\ninitial: self 102876 children 410076\nchild alloc 500MB\nafter_zombie: self 102880 children 410076\n\ntestcase7: exec (without fork)\n expect: initial ~\u003d exec\ninitial: self 102880 children 410076\nexec: self 102880 children 410076\n\nSigned-off-by: Jiri Pirko \u003cjpirko@redhat.com\u003e\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Hugh Dickins \u003chugh.dickins@tiscali.co.uk\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "cdd6c482c9ff9c55475ee7392ec8f672eddb7be6", "tree": "81f98a3ab46c589792057fe2392c1e10f8ad7893", "parents": [ "dfc65094d0313cc48969fa60bcf33d693aeb05a7" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Sep 21 12:02:48 2009 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Sep 21 14:28:04 2009 +0200" }, "message": "perf: Do the big rename: Performance Counters -\u003e Performance Events\n\nBye-bye Performance Counters, welcome Performance Events!\n\nIn the past few months the perfcounters subsystem has grown out its\ninitial role of counting hardware events, and has become (and is\nbecoming) a much broader generic event enumeration, reporting, logging,\nmonitoring, analysis facility.\n\nNaming its core object \u0027perf_counter\u0027 and naming the subsystem\n\u0027perfcounters\u0027 has become more and more of a misnomer. With pending\ncode like hw-breakpoints support the \u0027counter\u0027 name is less and\nless appropriate.\n\nAll in one, we\u0027ve decided to rename the subsystem to \u0027performance\nevents\u0027 and to propagate this rename through all fields, variables\nand API names. (in an ABI compatible fashion)\n\nThe word \u0027event\u0027 is also a bit shorter than \u0027counter\u0027 - which makes\nit slightly more convenient to write/handle as well.\n\nThanks goes to Stephane Eranian who first observed this misnomer and\nsuggested a rename.\n\nUser-space tooling and ABI compatibility is not affected - this patch\nshould be function-invariant. (Also, defconfigs were not touched to\nkeep the size down.)\n\nThis patch has been generated via the following script:\n\n FILES\u003d$(find * -type f | grep -vE \u0027oprofile|[^K]config\u0027)\n\n sed -i \\\n -e \u0027s/PERF_EVENT_/PERF_RECORD_/g\u0027 \\\n -e \u0027s/PERF_COUNTER/PERF_EVENT/g\u0027 \\\n -e \u0027s/perf_counter/perf_event/g\u0027 \\\n -e \u0027s/nb_counters/nb_events/g\u0027 \\\n -e \u0027s/swcounter/swevent/g\u0027 \\\n -e \u0027s/tpcounter_event/tp_event/g\u0027 \\\n $FILES\n\n for N in $(find . -name perf_counter.[ch]); do\n M\u003d$(echo $N | sed \u0027s/perf_counter/perf_event/g\u0027)\n mv $N $M\n done\n\n FILES\u003d$(find . -name perf_event.*)\n\n sed -i \\\n -e \u0027s/COUNTER_MASK/REG_MASK/g\u0027 \\\n -e \u0027s/COUNTER/EVENT/g\u0027 \\\n -e \u0027s/\\\u003cevent\\\u003e/event_id/g\u0027 \\\n -e \u0027s/counter/event/g\u0027 \\\n -e \u0027s/Counter/Event/g\u0027 \\\n $FILES\n\n... to keep it as correct as possible. This script can also be\nused by anyone who has pending perfcounters patches - it converts\na Linux kernel tree over to the new naming. We tried to time this\nchange to the point in time where the amount of pending patches\nis the smallest: the end of the merge window.\n\nNamespace clashes were fixed up in a preparatory patch - and some\nstylistic fallout will be fixed up in a subsequent patch.\n\n( NOTE: \u0027counters\u0027 are still the proper terminology when we deal\n with hardware registers - and these sed scripts are a bit\n over-eager in renaming them. I\u0027ve undone some of that, but\n in case there\u0027s something left where \u0027counter\u0027 would be\n better than \u0027event\u0027 we can undo that on an individual basis\n instead of touching an otherwise nicely automated patch. )\n\nSuggested-by: Stephane Eranian \u003ceranian@google.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nAcked-by: Paul Mackerras \u003cpaulus@samba.org\u003e\nReviewed-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nCc: Mike Galbraith \u003cefault@gmx.de\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Steven Rostedt \u003crostedt@goodmis.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: \u003clinux-arch@vger.kernel.org\u003e\nLKML-Reference: \u003cnew-submission\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "a2a8474c3fff88d8dd52d05cb450563fb26fd26c", "tree": "455415da039a77627f47ed19318bb9c8aa17de4d", "parents": [ "dd5d241ea955006122d76af88af87de73fec25b4" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sat Sep 05 11:17:13 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 05 11:30:42 2009 -0700" }, "message": "exec: do not sleep in TASK_TRACED under -\u003ecred_guard_mutex\n\nTom Horsley reports that his debugger hangs when it tries to read\n/proc/pid_of_tracee/maps, this happens since\n\n\t\"mm_for_maps: take -\u003ecred_guard_mutex to fix the race with exec\"\n\t04b836cbf19e885f8366bccb2e4b0474346c02d\n\ncommit in 2.6.31.\n\nBut the root of the problem lies in the fact that do_execve() path calls\ntracehook_report_exec() which can stop if the tracer sets PT_TRACE_EXEC.\n\nThe tracee must not sleep in TASK_TRACED holding this mutex. Even if we\nremove -\u003ecred_guard_mutex from mm_for_maps() and proc_pid_attr_write(),\nanother task doing PTRACE_ATTACH should not hang until it is killed or the\ntracee resumes.\n\nWith this patch do_execve() does not use -\u003ecred_guard_mutex directly and\nwe do not hold it throughout, instead:\n\n\t- introduce prepare_bprm_creds() helper, it locks the mutex\n\t and calls prepare_exec_creds() to initialize bprm-\u003ecred.\n\n\t- install_exec_creds() drops the mutex after commit_creds(),\n\t and thus before tracehook_report_exec()-\u003eptrace_stop().\n\n\t or, if exec fails,\n\n\t free_bprm() drops this mutex when bprm-\u003ecred !\u003d NULL which\n\t indicates install_exec_creds() was not called.\n\nReported-by: Tom Horsley \u003ctom.horsley@att.net\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: David Howells \u003cdhowells@redhat.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6777d773a463ac045d333b989d4e44660f8d92ad", "tree": "17291984d71515cec0a1970ecce3d631ed2f7981", "parents": [ "3edf2fb9d80a46d6c32ba12547a42419845b4b76" ], "author": { "name": "Mimi Zohar", "email": "zohar@linux.vnet.ibm.com", "time": "Fri Aug 21 14:32:48 2009 -0400" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Mon Aug 24 14:58:23 2009 +1000" }, "message": "kernel_read: redefine offset type\n\nvfs_read() offset is defined as loff_t, but kernel_read()\noffset is only defined as unsigned long. Redefine\nkernel_read() offset as loff_t.\n\nCc: stable@kernel.org\nSigned-off-by: Mimi Zohar \u003czohar@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "793285fcafce4719a05e0c99fa74b188157fe7fe", "tree": "3395530c1d4e604beac506c703b3093bb13f23ee", "parents": [ "82e3310ace59794ecf0f531eca94647b2863dfda" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Sun Jul 05 12:08:26 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 06 13:57:04 2009 -0700" }, "message": "cred_guard_mutex: do not return -EINTR to user-space\n\ndo_execve() and ptrace_attach() return -EINTR if\nmutex_lock_interruptible(-\u003ecred_guard_mutex) fails.\n\nThis is not right, change the code to return ERESTARTNOINTR.\n\nPerhaps we should also change proc_pid_attr_write().\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8a1ca8cedd108c8e76a6ab34079d0bbb4f244799", "tree": "636c715524f1718599209cc289908ea44b6cb859", "parents": [ "b640f042faa2a2fad6464f259a8afec06e2f6386", "940010c5a314a7bd9b498593bc6ba1718ac5aec5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 14:01:07 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 14:01:07 2009 -0700" }, "message": "Merge branch \u0027perfcounters-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027perfcounters-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (574 commits)\n perf_counter: Turn off by default\n perf_counter: Add counter-\u003eid to the throttle event\n perf_counter: Better align code\n perf_counter: Rename L2 to LL cache\n perf_counter: Standardize event names\n perf_counter: Rename enums\n perf_counter tools: Clean up u64 usage\n perf_counter: Rename perf_counter_limit sysctl\n perf_counter: More paranoia settings\n perf_counter: powerpc: Implement generalized cache events for POWER processors\n perf_counters: powerpc: Add support for POWER7 processors\n perf_counter: Accurate period data\n perf_counter: Introduce struct for sample data\n perf_counter tools: Normalize data using per sample period data\n perf_counter: Annotate exit ctx recursion\n perf_counter tools: Propagate signals properly\n perf_counter tools: Small frequency related fixes\n perf_counter: More aggressive frequency adjustment\n perf_counter/x86: Fix the model number of Intel Core2 processors\n perf_counter, x86: Correct some event and umask values for Intel processors\n ...\n" }, { "commit": "2c9e703c618106f5383226fbb1f526cb11034f8a", "tree": "87d7548001ea82f655fede0640466fc16aabcdf7", "parents": [ "6470c077cae12227318f40f3e6d756caadcce4b0", "5805977e63a36ad56594a623f3bd2bebcb7db233" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri May 22 18:40:59 2009 +1000" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri May 22 18:40:59 2009 +1000" }, "message": "Merge branch \u0027master\u0027 into next\n\nConflicts:\n\tfs/exec.c\n\nRemoved IMA changes (the IMA checks are now performed via may_open()).\n\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "b9fc745db833bbf74b4988493b8cd902a84c9415", "tree": "45a15174efb3b1c3dcbe5f0dc503e790c4f6fd70", "parents": [ "932995f0ce52525b32ff5127b522c2c164de3810" ], "author": { "name": "Mimi Zohar", "email": "zohar@linux.vnet.ibm.com", "time": "Tue May 19 13:25:57 2009 -0400" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri May 22 09:43:41 2009 +1000" }, "message": "integrity: path_check update\n\n- Add support in ima_path_check() for integrity checking without\nincrementing the counts. (Required for nfsd.)\n- rename and export opencount_get to ima_counts_get\n- replace ima_shm_check calls with ima_counts_get\n- export ima_path_check\n\nSigned-off-by: Mimi Zohar \u003czohar@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "dc3f81b129b5439ba7bac265bbc6a51a39275dae", "tree": "216030731d911249496d2e97206cd61431e31c89", "parents": [ "d2517a49d55536b38c7a87e5289550cfedaa4dcc", "1406de8e11eb043681297adf86d6892ff8efc27a" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 18 07:37:44 2009 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon May 18 07:37:49 2009 +0200" }, "message": "Merge commit \u0027v2.6.30-rc6\u0027 into perfcounters/core\n\nMerge reason: this branch was on an -rc4 base, merge it up to -rc6\n to get the latest upstream fixes.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "5e751e992f3fb08ba35e1ca8095ec8fbf9eda523", "tree": "711b1b47622dc9661f1d3d9c67d55c0b21456e8c", "parents": [ "d254117099d711f215e62427f55dfb8ebd5ad011" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri May 08 13:55:22 2009 +0100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Mon May 11 08:15:36 2009 +1000" }, "message": "CRED: Rename cred_exec_mutex to reflect that it\u0027s a guard against ptrace\n\nRename cred_exec_mutex to reflect that it\u0027s a guard against foreign\nintervention on a process\u0027s credential state, such as is made by ptrace(). The\nattachment of a debugger to a process affects execve()\u0027s calculation of the new\ncredential state - _and_ also setprocattr()\u0027s calculation of that state.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "6e8341a11eb21826b7192d0bb88cb5b44900a9af", "tree": "f9e03050f0834c2c4a21f1e6a255b6b0d1262b25", "parents": [ "a44ddbb6d8a8ffe4e34e417048dfdd8f3dd1de4f" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Apr 06 11:16:22 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat May 09 10:49:42 2009 -0400" }, "message": "Switch open_exec() and sys_uselib() to do_open_filp()\n\n... and make path_lookup_open() static\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "a44ddbb6d8a8ffe4e34e417048dfdd8f3dd1de4f", "tree": "a5f9962f00d8a7b2640682fffd6aefd3544b1fd3", "parents": [ "e24977d45f45d1675e050dc1a0aaf4bfc4ca9866" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Apr 06 09:38:49 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat May 09 10:49:42 2009 -0400" }, "message": "Make open_exec() and sys_uselib() use may_open(), instead of duplicating its parts\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "74641f584da8eccf30becfbb5507ab457187db22", "tree": "da212b7ae1b5a2a50a4314b3a75df5763623295e", "parents": [ "77b4cf5cb0c9d6010a18030a0e0a8d2aaf6b43ec" ], "author": { "name": "Ivan Kokshaysky", "email": "ink@jurassic.park.msu.ru", "time": "Thu Apr 30 15:08:49 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 02 15:36:10 2009 -0700" }, "message": "alpha: binfmt_aout fix\n\nThis fixes the problem introduced by commit 3bfacef412 (get rid of\nspecial-casing the /sbin/loader on alpha): osf/1 ecoff binary segfaults\nwhen binfmt_aout built as module. That happens because aout binary\nhandler gets on the top of the binfmt list due to late registration, and\nkernel attempts to execute the binary without preparatory work that must\nbe done by binfmt_loader.\n\nFixed by changing the registration order of the default binfmt handlers\nusing list_add_tail() and introducing insert_binfmt() function which\nplaces new handler on the top of the binfmt list. This might be generally\nuseful for installing arch-specific frontends for default handlers or just\nfor overriding them.\n\nSigned-off-by: Ivan Kokshaysky \u003cink@jurassic.park.msu.ru\u003e\nCc: Al Viro \u003cviro@ZenIV.linux.org.uk\u003e\nCc: Richard Henderson \u003crth@twiddle.net\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e7fd5d4b3d240f42c30a9e3d20a4689c4d3a795a", "tree": "4ba588631dd8189a818a91c9e3976526071178b6", "parents": [ "1130b0296184bc21806225fd06d533515a99d2db", "56a50adda49b2020156616c4eb15353e0f9ad7de" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Apr 29 14:46:59 2009 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Apr 29 14:47:05 2009 +0200" }, "message": "Merge branch \u0027linus\u0027 into perfcounters/core\n\nMerge reason: This brach was on -rc1, refresh it to almost-rc4 to pick up\n the latest upstream fixes.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "437f7fdb607f32b737e4da9f14bebcfdac2c90c3", "tree": "2064149ad4406774bd3b1ade60782980206f2e82", "parents": [ "8c652f96d3852b97a49c331cd0bb02d22f3cb31b" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Fri Apr 24 01:02:45 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 24 07:39:45 2009 -0700" }, "message": "check_unsafe_exec: s/lock_task_sighand/rcu_read_lock/\n\nwrite_lock(\u0026current-\u003efs-\u003elock) guarantees we can\u0027t wrongly miss\nLSM_UNSAFE_SHARE, this is what we care about. Use rcu_read_lock()\ninstead of -\u003esiglock to iterate over the sub-threads. We must see\nall CLONE_THREAD|CLONE_FS threads which didn\u0027t pass exit_fs(), it\ntakes fs-\u003elock too.\n\nWith or without this patch we can miss the freshly cloned thread\nand set LSM_UNSAFE_SHARE, we don\u0027t care.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\n[ Fixed lock/unlock typo - Hugh ]\nAcked-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8c652f96d3852b97a49c331cd0bb02d22f3cb31b", "tree": "9752969f2ae7312026bb4037ecd950dafd403d9b", "parents": [ "091069740304c979f957ceacec39c461d0192158" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Fri Apr 24 01:01:56 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 24 07:39:45 2009 -0700" }, "message": "do_execve() must not clear fs-\u003ein_exec if it was set by another thread\n\nIf do_execve() fails after check_unsafe_exec(), it clears fs-\u003ein_exec\nunconditionally. This is wrong if we race with our sub-thread which\nalso does do_execve:\n\n\tTwo threads T1 and T2 and another process P, all share the same\n\t-\u003efs.\n\n\tT1 starts do_execve(BAD_FILE). It calls check_unsafe_exec(), since\n\t-\u003efs is shared, we set LSM_UNSAFE but not -\u003ein_exec.\n\n\tP exits and decrements fs-\u003eusers.\n\n\tT2 starts do_execve(), calls check_unsafe_exec(), now -\u003efs is not\n\tshared, we set fs-\u003ein_exec.\n\n\tT1 continues, open_exec(BAD_FILE) fails, we clear -\u003ein_exec and\n\treturn to the user-space.\n\n\tT1 does clone(CLONE_FS /* without CLONE_THREAD */).\n\n\tT2 continues without LSM_UNSAFE_SHARE while -\u003efs is shared with\n\tanother process.\n\nChange check_unsafe_exec() to return res \u003d 1 if we set -\u003ein_exec, and change\ndo_execve() to clear -\u003ein_exec depending on res.\n\nWhen do_execve() suceeds, it is safe to clear -\u003ein_exec unconditionally.\nIt can be set only if we don\u0027t share -\u003efs with another process, and since\nwe already killed all sub-threads either -\u003ein_exec \u003d\u003d 0 or we are the\nonly user of this -\u003efs.\n\nAlso, we do not need fs-\u003elock to clear fs-\u003ein_exec.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nAcked-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8d1b2d9361b494bfc761700c348c65ebbe3deb5b", "tree": "284706b328133063bf530cb47c52a827b523b65a", "parents": [ "8740f9418c78dcad694b46ab25d1645d5aef1f5e" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Apr 08 15:01:30 2009 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Apr 08 19:05:47 2009 +0200" }, "message": "perf_counter: track task-comm data\n\nSimilar to the mmap data stream, add one that tracks the task COMM field,\nso that the userspace reporting knows what to call a task.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Corey Ashford \u003ccjashfor@linux.vnet.ibm.com\u003e\nLKML-Reference: \u003c20090408130409.127422406@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f541ae326fa120fa5c57433e4d9a133df212ce41", "tree": "bdbd94ec72cfc601118051cb35e8617d55510177", "parents": [ "e255357764f92afcafafbd4879b222b8c752065a", "0221c81b1b8eb0cbb6b30a0ced52ead32d2b4e4c" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Apr 06 09:02:57 2009 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Mon Apr 06 09:02:57 2009 +0200" }, "message": "Merge branch \u0027linus\u0027 into perfcounters/core-v2\n\nMerge reason: we have gathered quite a few conflicts, need to merge upstream\n\nConflicts:\n\tarch/powerpc/kernel/Makefile\n\tarch/x86/ia32/ia32entry.S\n\tarch/x86/include/asm/hardirq.h\n\tarch/x86/include/asm/unistd_32.h\n\tarch/x86/include/asm/unistd_64.h\n\tarch/x86/kernel/cpu/common.c\n\tarch/x86/kernel/irq.c\n\tarch/x86/kernel/syscall_table_32.S\n\tarch/x86/mm/iomap_32.c\n\tinclude/linux/sched.h\n\tkernel/Makefile\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "5ad4e53bd5406ee214ddc5a41f03f779b8b2d526", "tree": "b3dab5140284b3edf02bf2b13f74bfddb25aa62a", "parents": [ "ce3b0f8d5c2203301fc87f3aaaed73e5819e2a48" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sun Mar 29 19:50:06 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 31 23:00:27 2009 -0400" }, "message": "Get rid of indirect include of fs_struct.h\n\nDon\u0027t pull it in sched.h; very few files actually need it and those\ncan include directly. sched.h itself only needs forward declaration\nof struct fs_struct;\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "f1191b50ec11c8e2ca766d6d99eb5bb9d2c084a3", "tree": "7eae2387269bd3d2d4c6f334568f9fea6518c787", "parents": [ "498052bba55ecaff58db6a1436b0e25bfd75a7ff" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 30 07:35:18 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 31 23:00:26 2009 -0400" }, "message": "check_unsafe_exec() doesn\u0027t care about signal handlers sharing\n\n... since we\u0027ll unshare sighand anyway\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "498052bba55ecaff58db6a1436b0e25bfd75a7ff", "tree": "bd3644ac60737e3733995a203acebd70cfd1b21b", "parents": [ "3e93cd671813e204c258f1e6c797959920cf7772" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 30 07:20:30 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 31 23:00:26 2009 -0400" }, "message": "New locking/refcounting for fs_struct\n\n* all changes of current-\u003efs are done under task_lock and write_lock of\n old fs-\u003elock\n* refcount is not atomic anymore (same protection)\n* its decrements are done when removing reference from current; at the\n same time we decide whether to free it.\n* put_fs_struct() is gone\n* new field - -\u003ein_exec. Set by check_unsafe_exec() if we are trying to do\n execve() and only subthreads share fs_struct. Cleared when finishing exec\n (success and failure alike). Makes CLONE_FS fail with -EAGAIN if set.\n* check_unsafe_exec() may fail with -EAGAIN if another execve() from subthread\n is in progress.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e426b64c412aaa3e9eb3e4b261dc5be0d5a83e78", "tree": "c1528139b34fef3e4595576266c64068098fe211", "parents": [ "53e9309e01277ec99c38e84e0ca16921287cf470" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Sat Mar 28 23:20:19 2009 +0000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 28 17:30:00 2009 -0700" }, "message": "fix setuid sometimes doesn\u0027t\n\nJoe Malicki reports that setuid sometimes doesn\u0027t: very rarely,\na setuid root program does not get root euid; and, by the way,\nthey have a health check running lsof every few minutes.\n\nRight, check_unsafe_exec() notes whether the files_struct is being\nshared by more threads than will get killed by the exec, and if so\nsets LSM_UNSAFE_SHARE to make bprm_set_creds() careful about euid.\nBut /proc/\u003cpid\u003e/fd and /proc/\u003cpid\u003e/fdinfo lookups make transient\nuse of get_files_struct(), which also raises that sharing count.\n\nThere\u0027s a rather simple fix for this: exec\u0027s check on files-\u003ecount\nhas been redundant ever since 2.6.1 made it unshare_files() (except\nwhile compat_do_execve() omitted to do so) - just remove that check.\n\n[Note to -stable: this patch will not apply before 2.6.29: earlier\nreleases should just remove the files-\u003ecount line from unsafe_exec().]\n\nReported-by: Joe Malicki \u003cjmalicki@metacarta.com\u003e\nNarrowed-down-by: Michael Itz \u003cmitz@metacarta.com\u003e\nTested-by: Joe Malicki \u003cjmalicki@metacarta.com\u003e\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "703a3cd72817e99201cef84a8a7aecc60b2b3581", "tree": "3e943755178ff410694722bb031f523136fbc432", "parents": [ "df7f54c012b92ec93d56b68547351dcdf8a163d3", "8e0ee43bc2c3e19db56a4adaa9a9b04ce885cd84" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue Mar 24 10:52:46 2009 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Tue Mar 24 10:52:46 2009 +1100" }, "message": "Merge branch \u0027master\u0027 into next\n" }, { "commit": "f9ce1f1cda8b73a36f47e424975a9dfa78b7840c", "tree": "66138b33dbf627f99f5ec67e0e39ce15ab830080", "parents": [ "523979adfa0b79d4e3aa053220c37a9233294206" ], "author": { "name": "Kentaro Takeda", "email": "takedakn@nttdata.co.jp", "time": "Thu Feb 05 17:18:11 2009 +0900" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Thu Feb 12 15:15:03 2009 +1100" }, "message": "Add in_execve flag into task_struct.\n\nThis patch allows LSM modules to determine whether current process is in an\nexecve operation or not so that they can behave differently while an execve\noperation is in progress.\n\nThis patch is needed by TOMOYO. Please see another patch titled \"LSM adapter\nfunctions.\" for backgrounds.\n\nSigned-off-by: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "95fd4845ed0ffcab305b4f30ce1c12dc34f1b56c", "tree": "aa2aac22a5b329b778a6771a87bbf1945ad49bbd", "parents": [ "d278c48435625cb6b7edcf6a547620768b175709", "8e4921515c1a379539607eb443d51c30f4f7f338" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 11 09:22:04 2009 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 11 09:22:04 2009 +0100" }, "message": "Merge commit \u0027v2.6.29-rc4\u0027 into perfcounters/core\n\nConflicts:\n\tarch/x86/kernel/setup_percpu.c\n\tarch/x86/mm/fault.c\n\tdrivers/acpi/processor_idle.c\n\tkernel/irq/handle.c\n" }, { "commit": "0bf2f3aec5474da80a60e1baca629af87ecb67b6", "tree": "5c1a7733e24aaacbcf46e0434a11f033cfde43ca", "parents": [ "6cec50838ed04a9833fb5549f698d3756bbe7e72" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Feb 06 11:45:46 2009 +0000" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Sat Feb 07 08:46:18 2009 +1100" }, "message": "CRED: Fix SUID exec regression\n\nThe patch:\n\n\tcommit a6f76f23d297f70e2a6b3ec607f7aeeea9e37e8d\n\tCRED: Make execve() take advantage of copy-on-write credentials\n\nmoved the place in which the \u0027safeness\u0027 of a SUID/SGID exec was performed to\nbefore de_thread() was called. This means that LSM_UNSAFE_SHARE is now\ncalculated incorrectly. This flag is set if any of the usage counts for\nfs_struct, files_struct and sighand_struct are greater than 1 at the time the\ndetermination is made. All of which are true for threads created by the\npthread library.\n\nHowever, since we wish to make the security calculation before irrevocably\ndamaging the process so that we can return it an error code in the case where\nwe decide we want to reject the exec request on this basis, we have to make the\ndetermination before calling de_thread().\n\nSo, instead, we count up the number of threads (CLONE_THREAD) that are sharing\nour fs_struct (CLONE_FS), files_struct (CLONE_FILES) and sighand_structs\n(CLONE_SIGHAND/CLONE_THREAD) with us. These will be killed by de_thread() and\nso can be discounted by check_unsafe_exec().\n\nWe do have to be careful because CLONE_THREAD does not imply FS or FILES.\n\nWe _assume_ that there will be no extra references to these structs held by the\nthreads we\u0027re going to kill.\n\nThis can be tested with the attached pair of programs. Build the two programs\nusing the Makefile supplied, and run ./test1 as a non-root user. If\nsuccessful, you should see something like:\n\n\t[dhowells@andromeda tmp]$ ./test1\n\t--TEST1--\n\tuid\u003d4043, euid\u003d4043 suid\u003d4043\n\texec ./test2\n\t--TEST2--\n\tuid\u003d4043, euid\u003d0 suid\u003d0\n\tSUCCESS - Correct effective user ID\n\nand if unsuccessful, something like:\n\n\t[dhowells@andromeda tmp]$ ./test1\n\t--TEST1--\n\tuid\u003d4043, euid\u003d4043 suid\u003d4043\n\texec ./test2\n\t--TEST2--\n\tuid\u003d4043, euid\u003d4043 suid\u003d4043\n\tERROR - Incorrect effective user ID!\n\nThe non-root user ID you see will depend on the user you run as.\n\n[test1.c]\n#include \u003cstdio.h\u003e\n#include \u003cstdlib.h\u003e\n#include \u003cunistd.h\u003e\n#include \u003cpthread.h\u003e\n\nstatic void *thread_func(void *arg)\n{\n\twhile (1) {}\n}\n\nint main(int argc, char **argv)\n{\n\tpthread_t tid;\n\tuid_t uid, euid, suid;\n\n\tprintf(\"--TEST1--\\n\");\n\tgetresuid(\u0026uid, \u0026euid, \u0026suid);\n\tprintf(\"uid\u003d%d, euid\u003d%d suid\u003d%d\\n\", uid, euid, suid);\n\n\tif (pthread_create(\u0026tid, NULL, thread_func, NULL) \u003c 0) {\n\t\tperror(\"pthread_create\");\n\t\texit(1);\n\t}\n\n\tprintf(\"exec ./test2\\n\");\n\texeclp(\"./test2\", \"test2\", NULL);\n\tperror(\"./test2\");\n\t_exit(1);\n}\n\n[test2.c]\n#include \u003cstdio.h\u003e\n#include \u003cstdlib.h\u003e\n#include \u003cunistd.h\u003e\n\nint main(int argc, char **argv)\n{\n\tuid_t uid, euid, suid;\n\n\tgetresuid(\u0026uid, \u0026euid, \u0026suid);\n\tprintf(\"--TEST2--\\n\");\n\tprintf(\"uid\u003d%d, euid\u003d%d suid\u003d%d\\n\", uid, euid, suid);\n\n\tif (euid !\u003d 0) {\n\t\tfprintf(stderr, \"ERROR - Incorrect effective user ID!\\n\");\n\t\texit(1);\n\t}\n\tprintf(\"SUCCESS - Correct effective user ID\\n\");\n\texit(0);\n}\n\n[Makefile]\nCFLAGS \u003d -D_GNU_SOURCE -Wall -Werror -Wunused\nall: test1 test2\n\ntest1: test1.c\n\tgcc $(CFLAGS) -o test1 test1.c -lpthread\n\ntest2: test2.c\n\tgcc $(CFLAGS) -o test2 test2.c\n\tsudo chown root.root test2\n\tsudo chmod +s test2\n\nReported-by: David Smith \u003cdsmith@redhat.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: David Smith \u003cdsmith@redhat.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "cb5629b10d64a8006622ce3a52bc887d91057d69", "tree": "7c06d8f30783115e3384721046258ce615b129c5", "parents": [ "8920d5ad6ba74ae8ab020e90cc4d976980e68701", "f01d1d546abb2f4028b5299092f529eefb01253a" ], "author": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Feb 06 11:01:45 2009 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Feb 06 11:01:45 2009 +1100" }, "message": "Merge branch \u0027master\u0027 into next\n\nConflicts:\n\tfs/namei.c\n\nManually merged per:\n\ndiff --cc fs/namei.c\nindex 734f2b5,bbc15c2..0000000\n--- a/fs/namei.c\n+++ b/fs/namei.c\n@@@ -860,9 -848,8 +849,10 @@@ static int __link_path_walk(const char\n \t\tnd-\u003eflags |\u003d LOOKUP_CONTINUE;\n \t\terr \u003d exec_permission_lite(inode);\n \t\tif (err \u003d\u003d -EAGAIN)\n- \t\t\terr \u003d vfs_permission(nd, MAY_EXEC);\n+ \t\t\terr \u003d inode_permission(nd-\u003epath.dentry-\u003ed_inode,\n+ \t\t\t\t\t MAY_EXEC);\n +\t\tif (!err)\n +\t\t\terr \u003d ima_path_check(\u0026nd-\u003epath, MAY_EXEC);\n \t\tif (err)\n \t\t\tbreak;\n\n@@@ -1525,14 -1506,9 +1509,14 @@@ int may_open(struct path *path, int acc\n \t\tflag \u0026\u003d ~O_TRUNC;\n \t}\n\n- \terror \u003d vfs_permission(nd, acc_mode);\n+ \terror \u003d inode_permission(inode, acc_mode);\n \tif (error)\n \t\treturn error;\n +\n- \terror \u003d ima_path_check(\u0026nd-\u003epath,\n++\terror \u003d ima_path_check(path,\n +\t\t\t acc_mode \u0026 (MAY_READ | MAY_WRITE | MAY_EXEC));\n +\tif (error)\n +\t\treturn error;\n \t/*\n \t * An append-only file must be opened in append mode for writing.\n \t */\n\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "6146f0d5e47ca4047ffded0fb79b6c25359b386c", "tree": "edd792e52ad56d4a5d3ac6caa8437d3283fc157e", "parents": [ "659aaf2bb5496a425ba14036b5b5900f593e4484" ], "author": { "name": "Mimi Zohar", "email": "zohar@linux.vnet.ibm.com", "time": "Wed Feb 04 09:06:57 2009 -0500" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Feb 06 09:05:30 2009 +1100" }, "message": "integrity: IMA hooks\n\nThis patch replaces the generic integrity hooks, for which IMA registered\nitself, with IMA integrity hooks in the appropriate places directly\nin the fs directory.\n\nSigned-off-by: Mimi Zohar \u003czohar@us.ibm.com\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "77835492ed489c0b870f82f4c50687bd267acc0a", "tree": "d80903ce1b8dd30aa44ccfc756616ad4d6c74d63", "parents": [ "af37501c792107c2bde1524bdae38d9a247b841a", "1de9e8e70f5acc441550ca75433563d91b269bbe" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Jan 21 16:37:27 2009 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Jan 21 16:37:27 2009 +0100" }, "message": "Merge commit \u0027v2.6.29-rc2\u0027 into perfcounters/core\n\nConflicts:\n\tinclude/linux/syscalls.h\n" }, { "commit": "1e7bfb2134dfec37ce04fb3a4ca89299e892d10c", "tree": "99c676262e696754dcbfb2d6f59499972cd0c38c", "parents": [ "c4ea37c26a691ad0b7e86aa5884aab27830e95c9" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:29 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:29 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 27\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" } ], "next": "506c10f26c481b7f8ef27c1c79290f68989b2e9e" }