)]}' { "log": [ { "commit": "6588c1e3ff01418acafd938db0740e3477dc8cb7", "tree": "b58beed5813667619be47c5d064f8f78804e92d1", "parents": [ "b3bfa0cba867f23365b81658b47efd906830879b" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Thu Apr 02 16:58:09 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: SI_USER: Masquerade si_pid when crossing pid ns boundary\n\nWhen sending a signal to a descendant namespace, set -\u003esi_pid to 0 since\nthe sender does not have a pid in the receiver\u0027s namespace.\n\nNote:\n\t- If rt_sigqueueinfo() sets si_code to SI_USER when sending a\n\t signal across a pid namespace boundary, the value in -\u003esi_pid\n\t will be cleared to 0.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b3bfa0cba867f23365b81658b47efd906830879b", "tree": "e54d9a9c4c23fc00bb2ff3ed57ef5317a0660ea5", "parents": [ "e4da026f980df125a4918c3bb9fe93185c7ef12a" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Thu Apr 02 16:58:08 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: protect cinit from blocked fatal signals\n\nNormally SIG_DFL signals to global and container-init are dropped early.\nBut if a signal is blocked when it is posted, we cannot drop the signal\nsince the receiver may install a handler before unblocking the signal.\nOnce this signal is queued however, the receiver container-init has no way\nof knowing if the signal was sent from an ancestor or descendant\nnamespace. This patch ensures that contianer-init drops all SIG_DFL\nsignals in get_signal_to_deliver() except SIGKILL/SIGSTOP.\n\nIf SIGSTOP/SIGKILL originate from a descendant of container-init they are\nnever queued (i.e dropped in sig_ignored() in an earler patch).\n\nIf SIGSTOP/SIGKILL originate from parent namespace, the signal is queued\nand container-init processes the signal.\n\nIOW, if get_signal_to_deliver() sees a sig_kernel_only() signal for global\nor container-init, the signal must have been generated internally or must\nhave come from an ancestor ns and we process the signal.\n\nFurther, the signal_group_exit() check was needed to cover the case of a\nmulti-threaded init sending SIGKILL to other threads when doing an exit()\nor exec(). But since the new sig_kernel_only() check covers the SIGKILL,\nthe signal_group_exit() check is no longer needed and can be removed.\n\nFinally, now that we have all pieces in place, set SIGNAL_UNKILLABLE for\ncontainer-inits.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "921cf9f63089c7442d44083477620132f4cea066", "tree": "2a79eb0f7328ba2f05759d1b2c1e141bcfe4aafc", "parents": [ "7978b567d31555fc828b8f945c605ad29e117b22" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Thu Apr 02 16:58:05 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: protect cinit from unblocked SIG_DFL signals\n\nDrop early any SIG_DFL or SIG_IGN signals to container-init from within\nthe same container. But queue SIGSTOP and SIGKILL to the container-init\nif they are from an ancestor container.\n\nBlocked, fatal signals (i.e when SIG_DFL is to terminate) from within the\ncontainer can still terminate the container-init. That will be addressed\nin the next patch.\n\nNote:\tTo be bisect-safe, SIGNAL_UNKILLABLE will be set for container-inits\n \tin a follow-on patch. Until then, this patch is just a preparatory\n\tstep.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7978b567d31555fc828b8f945c605ad29e117b22", "tree": "a66d7880f3031e417c2afb76c990ed5340565816", "parents": [ "f008faff0e2777c8b3fe853891b774ca465938d8" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Thu Apr 02 16:58:04 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: add from_ancestor_ns parameter to send_signal()\n\nsend_signal() (or its helper) needs to determine the pid namespace of the\nsender. But a signal sent via kill_pid_info_as_uid() comes from within\nthe kernel and send_signal() does not need to determine the pid namespace\nof the sender. So define a helper for send_signal() which takes an\nadditional parameter, \u0027from_ancestor_ns\u0027 and have kill_pid_info_as_uid()\nuse that helper directly.\n\nThe \u0027from_ancestor_ns\u0027 parameter will be used in a follow-on patch.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f008faff0e2777c8b3fe853891b774ca465938d8", "tree": "d2f325995473a33652f7f7ead71e63d5298fbd01", "parents": [ "43918f2bf4806675943416d539d9d5e4d585ebff" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Thu Apr 02 16:58:02 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: protect init from unwanted signals more\n\n(This is a modified version of the patch submitted by Oleg Nesterov\nhttp://lkml.org/lkml/2008/11/18/249 and tries to address comments that\ncame up in that discussion)\n\ninit ignores the SIG_DFL signals but we queue them anyway, including\nSIGKILL. This is mostly OK, the signal will be dropped silently when\ndequeued, but the pending SIGKILL has 2 bad implications:\n\n - it implies fatal_signal_pending(), so we confuse things\n like wait_for_completion_killable/lock_page_killable.\n\n - for the sub-namespace inits, the pending SIGKILL can\n mask (legacy_queue) the subsequent SIGKILL from the\n parent namespace which must kill cinit reliably.\n (preparation, cinits don\u0027t have SIGNAL_UNKILLABLE yet)\n\nThe patch can\u0027t help when init is ptraced, but ptracing of init is not\n\"safe\" anyway.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "43918f2bf4806675943416d539d9d5e4d585ebff", "tree": "1de2ea08eed647b181b7f008f95e4bc4ec34c343", "parents": [ "90bc8d8b1a38f1ab131a2399a202e1889db95de8" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Thu Apr 02 16:58:00 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:58 2009 -0700" }, "message": "signals: remove \u0027handler\u0027 parameter to tracehook functions\n\nContainer-init must behave like global-init to processes within the\ncontainer and hence it must be immune to unhandled fatal signals from\nwithin the container (i.e SIG_DFL signals that terminate the process).\n\nBut the same container-init must behave like a normal process to processes\nin ancestor namespaces and so if it receives the same fatal signal from a\nprocess in ancestor namespace, the signal must be processed.\n\nImplementing these semantics requires that send_signal() determine pid\nnamespace of the sender but since signals can originate from workqueues/\ninterrupt-handlers, determining pid namespace of sender may not always be\npossible or safe.\n\nThis patchset implements the design/simplified semantics suggested by\nOleg Nesterov. The simplified semantics for container-init are:\n\n\t- container-init must never be terminated by a signal from a\n\t descendant process.\n\n\t- container-init must never be immune to SIGKILL from an ancestor\n\t namespace (so a process in parent namespace must always be able\n\t to terminate a descendant container).\n\n\t- container-init may be immune to unhandled fatal signals (like\n\t SIGUSR1) even if they are from ancestor namespace. SIGKILL/SIGSTOP\n\t are the only reliable signals to a container-init from ancestor\n\t namespace.\n\nThis patch:\n\nBased on an earlier patch submitted by Oleg Nesterov and comments from\nRoland McGrath (http://lkml.org/lkml/2008/11/19/258).\n\nThe handler parameter is currently unused in the tracehook functions.\nBesides, the tracehook functions are called with siglock held, so the\nfunctions can check the handler if they later need to.\n\nRemoving the parameter simiplifies changes to sig_ignored() in a follow-on\npatch.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Daniel Lezcano \u003cdaniel.lezcano@free.fr\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "53da1d9456fe7f87a920a78fdbdcf1225d197cb7", "tree": "eccd5357ceff25a9a07be802ac0161c8c1842e64", "parents": [ "b0dcb4a91ddb79f2e213205cf8d86b467f8559c7" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Mon Mar 23 16:07:24 2009 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 23 09:22:31 2009 -0700" }, "message": "fix ptrace slowness\n\nThis patch fixes bug #12208:\n\n Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id\u003d12208\n Subject : uml is very slow on 2.6.28 host\n\nThis turned out to be not a scheduler regression, but an already\nexisting problem in ptrace being triggered by subtle scheduler\nchanges.\n\nThe problem is this:\n\n - task A is ptracing task B\n - task B stops on a trace event\n - task A is woken up and preempts task B\n - task A calls ptrace on task B, which does ptrace_check_attach()\n - this calls wait_task_inactive(), which sees that task B is still on the runq\n - task A goes to sleep for a jiffy\n - ...\n\nSince UML does lots of the above sequences, those jiffies quickly add\nup to make it slow as hell.\n\nThis patch solves this by not rescheduling in read_unlock() after\nptrace_stop() has woken up the tracer.\n\nThanks to Oleg Nesterov and Ingo Molnar for the feedback.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCC: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "32bd671d6cbeda60dc73be77fa2b9037d9a9bfa0", "tree": "591c941b9f12ce9f3caefd112f0ada3c5fcc53ab", "parents": [ "83895147b702434e6f236deeca75211fd0e3da3a" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Thu Feb 05 12:24:15 2009 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Feb 05 13:04:33 2009 +0100" }, "message": "signal: re-add dead task accumulation stats.\n\nWe\u0027re going to split the process wide cpu accounting into two parts:\n\n - clocks; which can take all the time they want since they run\n from user context.\n\n - timers; which need constant time tracing but can affort the overhead\n because they\u0027re default off -- and rare.\n\nThe clock readout will go back to a full sum of the thread group, for this\nwe need to re-add the exit stats that were removed in the initial itimer\nrework (f06febc9: timers: fix itimer/many thread hang).\n\nFurthermore, since that full sum can be rather slow for large thread groups\nand we have the complete dead task stats, revert the do_notify_parent time\ncomputation.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nReviewed-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "3a9f84d354ce1e19956083c8e691727dea33bd5a", "tree": "abe8c280d3adb7a8e5fc838c59b122a46e6fd464", "parents": [ "bf50c903faba4ec7686ee8a570ac384b0f20814d" ], "author": { "name": "Ed Swierk", "email": "eswierk@aristanetworks.com", "time": "Mon Jan 26 15:33:31 2009 -0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Jan 27 00:36:19 2009 +0100" }, "message": "signals, debug: fix BUG: using smp_processor_id() in preemptible code in print_fatal_signal()\n\nWith print-fatal-signals\u003d1 on a kernel with CONFIG_PREEMPT\u003dy, sending an\nunexpected signal to a process causes a BUG: using smp_processor_id() in\npreemptible code.\n\nget_signal_to_deliver() releases the siglock before calling\nprint_fatal_signal(), which calls show_regs(), which calls\nsmp_processor_id(), which is not supposed to be called from a\npreemptible thread.\n\nMake sure show_regs() runs with preemption disabled.\n\nSigned-off-by: Ed Swierk \u003ceswierk@aristanetworks.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d4e82042c4cfa87a7d51710b71f568fe80132551", "tree": "202c311b52f4e4db9fbbbd80607744e2aa2e5885", "parents": [ "836f92adf121f806e9beb5b6b88bd5c9c4ea3f24" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:34 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:31 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 32\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "a5f8fa9e9ba5ef3305e147f41ad6e1e84ac1f0bd", "tree": "a9bedc6890d6379f0839c280a20c5da3bd9f2352", "parents": [ "17da2bd90abf428523de0fb98f7075e00e3ed42e" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:11 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:21 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 09\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "17da2bd90abf428523de0fb98f7075e00e3ed42e", "tree": "4f2daf3cb792b7da62e6ae771b58902cc5ea24ab", "parents": [ "754fe8d297bfae7b77f7ce866e2fb0c5fb186506" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:10 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:21 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 08\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "754fe8d297bfae7b77f7ce866e2fb0c5fb186506", "tree": "b2650f728d50791fa86ecd8c49806f51db714bd4", "parents": [ "5add95d4f7cf08f6f62510f19576992912387501" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:09 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:20 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 07\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "b290ebe2c46d01b742b948ce03f09e8a3efb9a92", "tree": "4419602f911a8205de2160c56a0ff844f350557e", "parents": [ "ae1251ab785f6da87219df8352ffdac68bba23e4" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:06 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:19 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 04\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "2ed7c03ec17779afb4fcfa3b8c61df61bd4879ba", "tree": "4e0fefd574bab5470a02edf439727f472a9663c6", "parents": [ "4c696ba7982501d43dea11dbbaabd2aa8a19cc42" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:13:54 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:14 2009 +0100" }, "message": "[CVE-2009-0029] Convert all system calls to return a long\n\nConvert all system calls to return a long. This should be a NOP since all\nconverted types should have the same size anyway.\nWith the exception of sys_exit_group which returned void. But that doesn\u0027t\nmatter since the system call doesn\u0027t return.\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "9cd4fd10437dda6b520cb1410b28f36967a34de8", "tree": "1a1d5eaa0d650c1efe0e63dbadb9674cac209d9d", "parents": [ "09bca05c90c639f57aae057e0c28f287e61f5a07" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Tue Jan 06 14:42:46 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 06 15:59:29 2009 -0800" }, "message": "SEND_SIG_NOINFO: set si_pid to tgid instead of pid\n\nPOSIX requires the si_pid to be the process id of the sender, so -\u003esi_pid\nshould really be set to \u0027tgid\u0027. This change does have following changes\nin behavior:\n\n\t- When sending pdeath_signal on re-parent to a sub-thread, -\u003esi_pid\n\t cannot be used to identify the thread that did the re-parent since\n\t it will now show the tgid instead of thread id.\n\n\t- A multi-threaded application that expects to find the specific\n\t thread that encountered a SIGPIPE using the -\u003esi_pid will now\n\t break.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nAcked-By: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "09bca05c90c639f57aae057e0c28f287e61f5a07", "tree": "5a5d88930380da8994525da361c9c8114ebf82fc", "parents": [ "ecb08d81313a3c015225236775de259d99ab47fe" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Tue Jan 06 14:42:45 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 06 15:59:28 2009 -0800" }, "message": "SEND_SIG_NOINFO: masquerade si_pid when crossing pid-ns boundary\n\nFor SEND_SIG_NOINFO, si_pid is currently set to the pid of sender\nin sender\u0027s active pid namespace. But if the receiver is in a\nEg: when parent sends the \u0027pdeath_signal\u0027 to a child that is in\na descendant pid namespace, we should set si_pid 0.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nAcked-By: Roland McGrath \u003croland@redhat.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b0f4b285d7ed174804658539129a834270f4829a", "tree": "be7f8dca58075aba2c6a137fcfd4d44c5c333efc", "parents": [ "be9c5ae4eeec2e85527e95647348b8ea4eb25128", "5250d329e38cdf7580faeb9c53c17d3588d7d19c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Dec 28 12:21:10 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Dec 28 12:21:10 2008 -0800" }, "message": "Merge branch \u0027tracing-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027tracing-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (241 commits)\n sched, trace: update trace_sched_wakeup()\n tracing/ftrace: don\u0027t trace on early stage of a secondary cpu boot, v3\n Revert \"x86: disable X86_PTRACE_BTS\"\n ring-buffer: prevent false positive warning\n ring-buffer: fix dangling commit race\n ftrace: enable format arguments checking\n x86, bts: memory accounting\n x86, bts: add fork and exit handling\n ftrace: introduce tracing_reset_online_cpus() helper\n tracing: fix warnings in kernel/trace/trace_sched_switch.c\n tracing: fix warning in kernel/trace/trace.c\n tracing/ring-buffer: remove unused ring_buffer size\n trace: fix task state printout\n ftrace: add not to regex on filtering functions\n trace: better use of stack_trace_enabled for boot up code\n trace: add a way to enable or disable the stack tracer\n x86: entry_64 - introduce FTRACE_ frame macro v2\n tracing/ftrace: add the printk-msg-only option\n tracing/ftrace: use preempt_enable_no_resched_notrace in ring_buffer_time_stamp()\n x86, bts: correctly report invalid bts records\n ...\n\nFixed up trivial conflict in scripts/recordmcount.pl due to SH bits\nbeing already partly merged by the SH merge.\n" }, { "commit": "7e066fb870fcd1025ec3ba7bbde5d541094f4ce1", "tree": "52acda06de25c029b9834110d7bf6b4abc50353b", "parents": [ "32f85742778dfc2c74975cf0b9f5bdb13470cb32" ], "author": { "name": "Mathieu Desnoyers", "email": "mathieu.desnoyers@polymtl.ca", "time": "Fri Nov 14 17:47:47 2008 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Nov 16 09:01:36 2008 +0100" }, "message": "tracepoints: add DECLARE_TRACE() and DEFINE_TRACE()\n\nImpact: API *CHANGE*. Must update all tracepoint users.\n\nAdd DEFINE_TRACE() to tracepoints to let them declare the tracepoint\nstructure in a single spot for all the kernel. It helps reducing memory\nconsumption, especially when declaring a lot of tracepoints, e.g. for\nkmalloc tracing.\n\n*API CHANGE WARNING*: now, DECLARE_TRACE() must be used in headers for\ntracepoint declarations rather than DEFINE_TRACE(). This is the sane way\nto do it. The name previously used was misleading.\n\nUpdates scheduler instrumentation to follow this API change.\n\nSigned-off-by: Mathieu Desnoyers \u003cmathieu.desnoyers@polymtl.ca\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "d84f4f992cbd76e8f39c488cf0c5d123843923b1", "tree": "fc4a0349c42995715b93d0f7a3c78e9ea9b3f36e", "parents": [ "745ca2475a6ac596e3d8d37c2759c0fbe2586227" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Nov 14 10:39:23 2008 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Nov 14 10:39:23 2008 +1100" }, "message": "CRED: Inaugurate COW credentials\n\nInaugurate copy-on-write credentials management. This uses RCU to manage the\ncredentials pointer in the task_struct with respect to accesses by other tasks.\nA process may only modify its own credentials, and so does not need locking to\naccess or modify its own credentials.\n\nA mutex (cred_replace_mutex) is added to the task_struct to control the effect\nof PTRACE_ATTACHED on credential calculations, particularly with respect to\nexecve().\n\nWith this patch, the contents of an active credentials struct may not be\nchanged directly; rather a new set of credentials must be prepared, modified\nand committed using something like the following sequence of events:\n\n\tstruct cred *new \u003d prepare_creds();\n\tint ret \u003d blah(new);\n\tif (ret \u003c 0) {\n\t\tabort_creds(new);\n\t\treturn ret;\n\t}\n\treturn commit_creds(new);\n\nThere are some exceptions to this rule: the keyrings pointed to by the active\ncredentials may be instantiated - keyrings violate the COW rule as managing\nCOW keyrings is tricky, given that it is possible for a task to directly alter\nthe keys in a keyring in use by another task.\n\nTo help enforce this, various pointers to sets of credentials, such as those in\nthe task_struct, are declared const. The purpose of this is compile-time\ndiscouragement of altering credentials through those pointers. Once a set of\ncredentials has been made public through one of these pointers, it may not be\nmodified, except under special circumstances:\n\n (1) Its reference count may incremented and decremented.\n\n (2) The keyrings to which it points may be modified, but not replaced.\n\nThe only safe way to modify anything else is to create a replacement and commit\nusing the functions described in Documentation/credentials.txt (which will be\nadded by a later patch).\n\nThis patch and the preceding patches have been tested with the LTP SELinux\ntestsuite.\n\nThis patch makes several logical sets of alteration:\n\n (1) execve().\n\n This now prepares and commits credentials in various places in the\n security code rather than altering the current creds directly.\n\n (2) Temporary credential overrides.\n\n do_coredump() and sys_faccessat() now prepare their own credentials and\n temporarily override the ones currently on the acting thread, whilst\n preventing interference from other threads by holding cred_replace_mutex\n on the thread being dumped.\n\n This will be replaced in a future patch by something that hands down the\n credentials directly to the functions being called, rather than altering\n the task\u0027s objective credentials.\n\n (3) LSM interface.\n\n A number of functions have been changed, added or removed:\n\n (*) security_capset_check(), -\u003ecapset_check()\n (*) security_capset_set(), -\u003ecapset_set()\n\n \t Removed in favour of security_capset().\n\n (*) security_capset(), -\u003ecapset()\n\n \t New. This is passed a pointer to the new creds, a pointer to the old\n \t creds and the proposed capability sets. It should fill in the new\n \t creds or return an error. All pointers, barring the pointer to the\n \t new creds, are now const.\n\n (*) security_bprm_apply_creds(), -\u003ebprm_apply_creds()\n\n \t Changed; now returns a value, which will cause the process to be\n \t killed if it\u0027s an error.\n\n (*) security_task_alloc(), -\u003etask_alloc_security()\n\n \t Removed in favour of security_prepare_creds().\n\n (*) security_cred_free(), -\u003ecred_free()\n\n \t New. Free security data attached to cred-\u003esecurity.\n\n (*) security_prepare_creds(), -\u003ecred_prepare()\n\n \t New. Duplicate any security data attached to cred-\u003esecurity.\n\n (*) security_commit_creds(), -\u003ecred_commit()\n\n \t New. Apply any security effects for the upcoming installation of new\n \t security by commit_creds().\n\n (*) security_task_post_setuid(), -\u003etask_post_setuid()\n\n \t Removed in favour of security_task_fix_setuid().\n\n (*) security_task_fix_setuid(), -\u003etask_fix_setuid()\n\n \t Fix up the proposed new credentials for setuid(). This is used by\n \t cap_set_fix_setuid() to implicitly adjust capabilities in line with\n \t setuid() changes. Changes are made to the new credentials, rather\n \t than the task itself as in security_task_post_setuid().\n\n (*) security_task_reparent_to_init(), -\u003etask_reparent_to_init()\n\n \t Removed. Instead the task being reparented to init is referred\n \t directly to init\u0027s credentials.\n\n\t NOTE! This results in the loss of some state: SELinux\u0027s osid no\n\t longer records the sid of the thread that forked it.\n\n (*) security_key_alloc(), -\u003ekey_alloc()\n (*) security_key_permission(), -\u003ekey_permission()\n\n \t Changed. These now take cred pointers rather than task pointers to\n \t refer to the security context.\n\n (4) sys_capset().\n\n This has been simplified and uses less locking. The LSM functions it\n calls have been merged.\n\n (5) reparent_to_kthreadd().\n\n This gives the current thread the same credentials as init by simply using\n commit_thread() to point that way.\n\n (6) __sigqueue_alloc() and switch_uid()\n\n __sigqueue_alloc() can\u0027t stop the target task from changing its creds\n beneath it, so this function gets a reference to the currently applicable\n user_struct which it then passes into the sigqueue struct it returns if\n successful.\n\n switch_uid() is now called from commit_creds(), and possibly should be\n folded into that. commit_creds() should take care of protecting\n __sigqueue_alloc().\n\n (7) [sg]et[ug]id() and co and [sg]et_current_groups.\n\n The set functions now all use prepare_creds(), commit_creds() and\n abort_creds() to build and check a new set of credentials before applying\n it.\n\n security_task_set[ug]id() is called inside the prepared section. This\n guarantees that nothing else will affect the creds until we\u0027ve finished.\n\n The calling of set_dumpable() has been moved into commit_creds().\n\n Much of the functionality of set_user() has been moved into\n commit_creds().\n\n The get functions all simply access the data directly.\n\n (8) security_task_prctl() and cap_task_prctl().\n\n security_task_prctl() has been modified to return -ENOSYS if it doesn\u0027t\n want to handle a function, or otherwise return the return value directly\n rather than through an argument.\n\n Additionally, cap_task_prctl() now prepares a new set of credentials, even\n if it doesn\u0027t end up using it.\n\n (9) Keyrings.\n\n A number of changes have been made to the keyrings code:\n\n (a) switch_uid_keyring(), copy_keys(), exit_keys() and suid_keys() have\n \t all been dropped and built in to the credentials functions directly.\n \t They may want separating out again later.\n\n (b) key_alloc() and search_process_keyrings() now take a cred pointer\n \t rather than a task pointer to specify the security context.\n\n (c) copy_creds() gives a new thread within the same thread group a new\n \t thread keyring if its parent had one, otherwise it discards the thread\n \t keyring.\n\n (d) The authorisation key now points directly to the credentials to extend\n \t the search into rather pointing to the task that carries them.\n\n (e) Installing thread, process or session keyrings causes a new set of\n \t credentials to be created, even though it\u0027s not strictly necessary for\n \t process or session keyrings (they\u0027re shared).\n\n(10) Usermode helper.\n\n The usermode helper code now carries a cred struct pointer in its\n subprocess_info struct instead of a new session keyring pointer. This set\n of credentials is derived from init_cred and installed on the new process\n after it has been cloned.\n\n call_usermodehelper_setup() allocates the new credentials and\n call_usermodehelper_freeinfo() discards them if they haven\u0027t been used. A\n special cred function (prepare_usermodeinfo_creds()) is provided\n specifically for call_usermodehelper_setup() to call.\n\n call_usermodehelper_setkeys() adjusts the credentials to sport the\n supplied keyring as the new session keyring.\n\n(11) SELinux.\n\n SELinux has a number of changes, in addition to those to support the LSM\n interface changes mentioned above:\n\n (a) selinux_setprocattr() no longer does its check for whether the\n \t current ptracer can access processes with the new SID inside the lock\n \t that covers getting the ptracer\u0027s SID. Whilst this lock ensures that\n \t the check is done with the ptracer pinned, the result is only valid\n \t until the lock is released, so there\u0027s no point doing it inside the\n \t lock.\n\n(12) is_single_threaded().\n\n This function has been extracted from selinux_setprocattr() and put into\n a file of its own in the lib/ directory as join_session_keyring() now\n wants to use it too.\n\n The code in SELinux just checked to see whether a task shared mm_structs\n with other tasks (CLONE_VM), but that isn\u0027t good enough. We really want\n to know if they\u0027re part of the same thread group (CLONE_THREAD).\n\n(13) nfsd.\n\n The NFS server daemon now has to use the COW credentials to set the\n credentials it is going to use. It really needs to pass the credentials\n down to the functions it calls, but it can\u0027t do that until other patches\n in this series have been applied.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: James Morris \u003cjmorris@namei.org\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "c69e8d9c01db2adc503464993c358901c9af9de4", "tree": "bed94aaa9aeb7a7834d1c880f72b62a11a752c78", "parents": [ "86a264abe542cfececb4df129bc45a0338d8cdb9" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Nov 14 10:39:19 2008 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Nov 14 10:39:19 2008 +1100" }, "message": "CRED: Use RCU to access another task\u0027s creds and to release a task\u0027s own creds\n\nUse RCU to access another task\u0027s creds and to release a task\u0027s own creds.\nThis means that it will be possible for the credentials of a task to be\nreplaced without another task (a) requiring a full lock to read them, and (b)\nseeing deallocated memory.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: James Morris \u003cjmorris@namei.org\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "b6dff3ec5e116e3af6f537d4caedcad6b9e5082a", "tree": "9e76f972eb7ce9b84e0146c8e4126a3f86acb428", "parents": [ "15a2460ed0af7538ca8e6c610fe607a2cd9da142" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Nov 14 10:39:16 2008 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Nov 14 10:39:16 2008 +1100" }, "message": "CRED: Separate task security context from task_struct\n\nSeparate the task security context from task_struct. At this point, the\nsecurity data is temporarily embedded in the task_struct with two pointers\npointing to it.\n\nNote that the Alpha arch is altered as it refers to (E)UID and (E)GID in\nentry.S via asm-offsets.\n\nWith comment fixes Signed-off-by: Marc Dionne \u003cmarc.c.dionne@gmail.com\u003e\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: James Morris \u003cjmorris@namei.org\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "76aac0e9a17742e60d408be1a706e9aaad370891", "tree": "e873a000d9c96209726e0958e311f005c13b2ed5", "parents": [ "b103c59883f1ec6e4d548b25054608cb5724453c" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Nov 14 10:39:12 2008 +1100" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Fri Nov 14 10:39:12 2008 +1100" }, "message": "CRED: Wrap task credential accesses in the core kernel\n\nWrap access to task credentials so that they can be separated more easily from\nthe task_struct during the introduction of COW creds.\n\nChange most current-\u003e(|e|s|fs)[ug]id to current_(|e|s|fs)[ug]id().\n\nChange some task-\u003ee?[ug]id to task_e?[ug]id(). In some places it makes more\nsense to use RCU directly rather than a convenient wrapper; these will be\naddressed by later patches.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nReviewed-by: James Morris \u003cjmorris@namei.org\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: linux-audit@redhat.com\nCc: containers@lists.linux-foundation.org\nCc: linux-mm@kvack.org\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "d25141a818383b3c3b09f065698c544a7a0ec6e7", "tree": "fb2f868b42ccdde710ddf426e1c90f8ffcc83368", "parents": [ "e74481e23283fb080d4591c258de20785cc3b6c3" ], "author": { "name": "Sukadev Bhattiprolu", "email": "sukadev@linux.vnet.ibm.com", "time": "Wed Oct 29 14:01:11 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 30 11:38:46 2008 -0700" }, "message": "\u0027kill sig -1\u0027 must only apply to caller\u0027s namespace\n\nCurrently \"kill \u003csig\u003e -1\" kills processes in all namespaces and breaks the\nisolation of namespaces. Earlier attempt to fix this was discussed at:\n\n\thttp://lkml.org/lkml/2008/7/23/148\n\nAs suggested by Oleg Nesterov in that thread, use \"task_pid_vnr() \u003e 1\"\ncheck since task_pid_vnr() returns 0 if process is outside the caller\u0027s\nnamespace.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nAcked-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nTested-by: Daniel Hokka Zakrisson \u003cdaniel@hozac.com\u003e\nSigned-off-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": "92b29b86fe2e183d44eb467e5e74a5f718ef2e43", "tree": "1bac8a1aa11d47322b66d10ec3a370016d843d06", "parents": [ "b9d7ccf56be1ac77b71a284a1c0e6337f9a7aff0", "98d9c66ab07471006fd7910cb16453581c41a3e7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 20 13:35:07 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 20 13:35:07 2008 -0700" }, "message": "Merge branch \u0027tracing-v28-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027tracing-v28-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (131 commits)\n tracing/fastboot: improve help text\n tracing/stacktrace: improve help text\n tracing/fastboot: fix initcalls disposition in bootgraph.pl\n tracing/fastboot: fix bootgraph.pl initcall name regexp\n tracing/fastboot: fix issues and improve output of bootgraph.pl\n tracepoints: synchronize unregister static inline\n tracepoints: tracepoint_synchronize_unregister()\n ftrace: make ftrace_test_p6nop disassembler-friendly\n markers: fix synchronize marker unregister static inline\n tracing/fastboot: add better resolution to initcall debug/tracing\n trace: add build-time check to avoid overrunning hex buffer\n ftrace: fix hex output mode of ftrace\n tracing/fastboot: fix initcalls disposition in bootgraph.pl\n tracing/fastboot: fix printk format typo in boot tracer\n ftrace: return an error when setting a nonexistent tracer\n ftrace: make some tracers reentrant\n ring-buffer: make reentrant\n ring-buffer: move page indexes into page headers\n tracing/fastboot: only trace non-module initcalls\n ftrace: move pc counter in irqtrace\n ...\n\nManually fix conflicts:\n - init/main.c: initcall tracing\n - kernel/module.c: verbose level vs tracepoints\n - scripts/bootgraph.pl: fallout from cherry-picking commits.\n" }, { "commit": "0a16b6075843325dc402edf80c1662838b929aff", "tree": "124e49d8c0196f3c6eb7aa09a7c4c3c3157fff7a", "parents": [ "4a0897526bbc5c6ac0df80b16b8c60339e717ae2" ], "author": { "name": "Mathieu Desnoyers", "email": "mathieu.desnoyers@polymtl.ca", "time": "Fri Jul 18 12:16:17 2008 -0400" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Oct 14 10:30:52 2008 +0200" }, "message": "tracing, sched: LTTng instrumentation - scheduler\n\nInstrument the scheduler activity (sched_switch, migration, wakeups,\nwait for a task, signal delivery) and process/thread\ncreation/destruction (fork, exit, kthread stop). Actually, kthread\ncreation is not instrumented in this patch because it is architecture\ndependent. It allows to connect tracers such as ftrace which detects\nscheduling latencies, good/bad scheduler decisions. Tools like LTTng can\nexport this scheduler information along with instrumentation of the rest\nof the kernel activity to perform post-mortem analysis on the scheduler\nactivity.\n\nAbout the performance impact of tracepoints (which is comparable to\nmarkers), even without immediate values optimizations, tests done by\nHideo Aoki on ia64 show no regression. His test case was using hackbench\non a kernel where scheduler instrumentation (about 5 events in code\nscheduler code) was added. See the \"Tracepoints\" patch header for\nperformance result detail.\n\nChangelog :\n\n- Change instrumentation location and parameter to match ftrace\n instrumentation, previously done with kernel markers.\n\n[ mingo@elte.hu: conflict resolutions ]\nSigned-off-by: Mathieu Desnoyers \u003cmathieu.desnoyers@polymtl.ca\u003e\nAcked-by: \u0027Peter Zijlstra\u0027 \u003cpeterz@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "f06febc96ba8e0af80bcc3eaec0a109e88275fac", "tree": "46dba9432ef25d2eae9434ff2df638c7a268c0f1", "parents": [ "6bfb09a1005193be5c81ebac9f3ef85210142650" ], "author": { "name": "Frank Mayhar", "email": "fmayhar@google.com", "time": "Fri Sep 12 09:54:39 2008 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Sep 14 16:25:35 2008 +0200" }, "message": "timers: fix itimer/many thread hang\n\nOverview\n\nThis patch reworks the handling of POSIX CPU timers, including the\nITIMER_PROF, ITIMER_VIRT timers and rlimit handling. It was put together\nwith the help of Roland McGrath, the owner and original writer of this code.\n\nThe problem we ran into, and the reason for this rework, has to do with using\na profiling timer in a process with a large number of threads. It appears\nthat the performance of the old implementation of run_posix_cpu_timers() was\nat least O(n*3) (where \"n\" is the number of threads in a process) or worse.\nEverything is fine with an increasing number of threads until the time taken\nfor that routine to run becomes the same as or greater than the tick time, at\nwhich point things degrade rather quickly.\n\nThis patch fixes bug 9906, \"Weird hang with NPTL and SIGPROF.\"\n\nCode Changes\n\nThis rework corrects the implementation of run_posix_cpu_timers() to make it\nrun in constant time for a particular machine. (Performance may vary between\none machine and another depending upon whether the kernel is built as single-\nor multiprocessor and, in the latter case, depending upon the number of\nrunning processors.) To do this, at each tick we now update fields in\nsignal_struct as well as task_struct. The run_posix_cpu_timers() function\nuses those fields to make its decisions.\n\nWe define a new structure, \"task_cputime,\" to contain user, system and\nscheduler times and use these in appropriate places:\n\nstruct task_cputime {\n\tcputime_t utime;\n\tcputime_t stime;\n\tunsigned long long sum_exec_runtime;\n};\n\nThis is included in the structure \"thread_group_cputime,\" which is a new\nsubstructure of signal_struct and which varies for uniprocessor versus\nmultiprocessor kernels. For uniprocessor kernels, it uses \"task_cputime\" as\na simple substructure, while for multiprocessor kernels it is a pointer:\n\nstruct thread_group_cputime {\n\tstruct task_cputime totals;\n};\n\nstruct thread_group_cputime {\n\tstruct task_cputime *totals;\n};\n\nWe also add a new task_cputime substructure directly to signal_struct, to\ncache the earliest expiration of process-wide timers, and task_cputime also\nreplaces the it_*_expires fields of task_struct (used for earliest expiration\nof thread timers). The \"thread_group_cputime\" structure contains process-wide\ntimers that are updated via account_user_time() and friends. In the non-SMP\ncase the structure is a simple aggregator; unfortunately in the SMP case that\nsimplicity was not achievable due to cache-line contention between CPUs (in\none measured case performance was actually _worse_ on a 16-cpu system than\nthe same test on a 4-cpu system, due to this contention). For SMP, the\nthread_group_cputime counters are maintained as a per-cpu structure allocated\nusing alloc_percpu(). The timer functions update only the timer field in\nthe structure corresponding to the running CPU, obtained using per_cpu_ptr().\n\nWe define a set of inline functions in sched.h that we use to maintain the\nthread_group_cputime structure and hide the differences between UP and SMP\nimplementations from the rest of the kernel. The thread_group_cputime_init()\nfunction initializes the thread_group_cputime structure for the given task.\nThe thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the\nout-of-line function thread_group_cputime_alloc_smp() to allocate and fill\nin the per-cpu structures and fields. The thread_group_cputime_free()\nfunction, also a no-op for UP, in SMP frees the per-cpu structures. The\nthread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls\nthread_group_cputime_alloc() if the per-cpu structures haven\u0027t yet been\nallocated. The thread_group_cputime() function fills the task_cputime\nstructure it is passed with the contents of the thread_group_cputime fields;\nin UP it\u0027s that simple but in SMP it must also safely check that tsk-\u003esignal\nis non-NULL (if it is it just uses the appropriate fields of task_struct) and,\nif so, sums the per-cpu values for each online CPU. Finally, the three\nfunctions account_group_user_time(), account_group_system_time() and\naccount_group_exec_runtime() are used by timer functions to update the\nrespective fields of the thread_group_cputime structure.\n\nNon-SMP operation is trivial and will not be mentioned further.\n\nThe per-cpu structure is always allocated when a task creates its first new\nthread, via a call to thread_group_cputime_clone_thread() from copy_signal().\nIt is freed at process exit via a call to thread_group_cputime_free() from\ncleanup_signal().\n\nAll functions that formerly summed utime/stime/sum_sched_runtime values from\nfrom all threads in the thread group now use thread_group_cputime() to\nsnapshot the values in the thread_group_cputime structure or the values in\nthe task structure itself if the per-cpu structure hasn\u0027t been allocated.\n\nFinally, the code in kernel/posix-cpu-timers.c has changed quite a bit.\nThe run_posix_cpu_timers() function has been split into a fast path and a\nslow path; the former safely checks whether there are any expired thread\ntimers and, if not, just returns, while the slow path does the heavy lifting.\nWith the dedicated thread group fields, timers are no longer \"rebalanced\" and\nthe process_timer_rebalance() function and related code has gone away. All\nsumming loops are gone and all code that used them now uses the\nthread_group_cputime() inline. When process-wide timers are set, the new\ntask_cputime structure in signal_struct is used to cache the earliest\nexpiration; this is checked in the fast path.\n\nPerformance\n\nThe fix appears not to add significant overhead to existing operations. It\ngenerally performs the same as the current code except in two cases, one in\nwhich it performs slightly worse (Case 5 below) and one in which it performs\nvery significantly better (Case 2 below). Overall it\u0027s a wash except in those\ntwo cases.\n\nI\u0027ve since done somewhat more involved testing on a dual-core Opteron system.\n\nCase 1: With no itimer running, for a test with 100,000 threads, the fixed\n\tkernel took 1428.5 seconds, 513 seconds more than the unfixed system,\n\tall of which was spent in the system. There were twice as many\n\tvoluntary context switches with the fix as without it.\n\nCase 2: With an itimer running at .01 second ticks and 4000 threads (the most\n\tan unmodified kernel can handle), the fixed kernel ran the test in\n\teight percent of the time (5.8 seconds as opposed to 70 seconds) and\n\thad better tick accuracy (.012 seconds per tick as opposed to .023\n\tseconds per tick).\n\nCase 3: A 4000-thread test with an initial timer tick of .01 second and an\n\tinterval of 10,000 seconds (i.e. a timer that ticks only once) had\n\tvery nearly the same performance in both cases: 6.3 seconds elapsed\n\tfor the fixed kernel versus 5.5 seconds for the unfixed kernel.\n\nWith fewer threads (eight in these tests), the Case 1 test ran in essentially\nthe same time on both the modified and unmodified kernels (5.2 seconds versus\n5.8 seconds). The Case 2 test ran in about the same time as well, 5.9 seconds\nversus 5.4 seconds but again with much better tick accuracy, .013 seconds per\ntick versus .025 seconds per tick for the unmodified kernel.\n\nSince the fix affected the rlimit code, I also tested soft and hard CPU limits.\n\nCase 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer\n\trunning), the modified kernel was very slightly favored in that while\n\tit killed the process in 19.997 seconds of CPU time (5.002 seconds of\n\twall time), only .003 seconds of that was system time, the rest was\n\tuser time. The unmodified kernel killed the process in 20.001 seconds\n\tof CPU (5.014 seconds of wall time) of which .016 seconds was system\n\ttime. Really, though, the results were too close to call. The results\n\twere essentially the same with no itimer running.\n\nCase 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds\n\t(where the hard limit would never be reached) and an itimer running,\n\tthe modified kernel exhibited worse tick accuracy than the unmodified\n\tkernel: .050 seconds/tick versus .028 seconds/tick. Otherwise,\n\tperformance was almost indistinguishable. With no itimer running this\n\ttest exhibited virtually identical behavior and times in both cases.\n\nIn times past I did some limited performance testing. those results are below.\n\nOn a four-cpu Opteron system without this fix, a sixteen-thread test executed\nin 3569.991 seconds, of which user was 3568.435s and system was 1.556s. On\nthe same system with the fix, user and elapsed time were about the same, but\nsystem time dropped to 0.007 seconds. Performance with eight, four and one\nthread were comparable. Interestingly, the timer ticks with the fix seemed\nmore accurate: The sixteen-thread test with the fix received 149543 ticks\nfor 0.024 seconds per tick, while the same test without the fix received 58720\nfor 0.061 seconds per tick. Both cases were configured for an interval of\n0.01 seconds. Again, the other tests were comparable. Each thread in this\ntest computed the primes up to 25,000,000.\n\nI also did a test with a large number of threads, 100,000 threads, which is\nimpossible without the fix. In this case each thread computed the primes only\nup to 10,000 (to make the runtime manageable). System time dominated, at\n1546.968 seconds out of a total 2176.906 seconds (giving a user time of\n629.938s). It received 147651 ticks for 0.015 seconds per tick, still quite\naccurate. There is obviously no comparable test without the fix.\n\nSigned-off-by: Frank Mayhar \u003cfmayhar@google.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "1b04624f93bb1c4f9495b8476d1dd0200af019e2", "tree": "aa64cc4706f5dc396b1e9a74ce951b29aa294964", "parents": [ "1fca25427482387689fa27594c992a961d98768f" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Tue Aug 19 20:37:07 2008 -0700" }, "committer": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Tue Aug 19 20:37:07 2008 -0700" }, "message": "tracehook: fix SA_NOCLDWAIT\n\nI outwitted myself again in commit 2b2a1ff64afbadac842bbc58c5166962cf4f7664,\nand broke the SA_NOCLDWAIT behavior so it leaks zombies. This fixes it.\n\nReported-by: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\n" }, { "commit": "67a077dca4e648a662e32cbeaaba8094d2e30229", "tree": "2f8943838b73b0a8ea590b1aaad9294eb280ea1e", "parents": [ "9b4d0bab32e18e4f72781f9fa309a81495b2aff3", "2106b531eaa2edd0c2dfa735a0556c08c7ba3c86" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Aug 11 16:46:11 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Aug 11 16:46:11 2008 -0700" }, "message": "Merge branch \u0027timers-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027timers-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:\n posix-timers: fix posix_timer_event() vs dequeue_signal() race\n posix-timers: do_schedule_next_timer: fix the setting of -\u003esi_overrun\n" }, { "commit": "2106b531eaa2edd0c2dfa735a0556c08c7ba3c86", "tree": "423fcf0ac74f0dd5d88864b7bd99a1160129e064", "parents": [ "e338125b8a886923ba8367207c144764dc352584", "ba661292a2bc6ddd305a212b0526e5dc22195fe7" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Jul 27 23:15:26 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Sun Jul 27 23:15:26 2008 +0200" }, "message": "Merge branch \u0027timers/urgent\u0027 of ssh://master.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip into timers/urgent\n" }, { "commit": "b787f7ba677840da16a2228c16571ce8a1fcb799", "tree": "b1838c4ae7f38c00efc2375871a255275500f245", "parents": [ "2b2a1ff64afbadac842bbc58c5166962cf4f7664" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:55 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: force signal_pending()\n\nThis defines a new hook tracehook_force_sigpending() that lets tracing\ncode decide to force TIF_SIGPENDING on in recalc_sigpending().\n\nThis is not used yet, so it compiles away to nothing for now. It lays the\ngroundwork for new tracing code that can interrupt a task synthetically\nwithout actually sending a signal.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "2b2a1ff64afbadac842bbc58c5166962cf4f7664", "tree": "660640eb95117ce6f7436331f23f73805abced10", "parents": [ "fa00b80b3c41a845b3d56f866fb40a2e98754c51" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:54 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: death\n\nThis moves the ptrace logic in task death (exit_notify) into tracehook.h\ninlines. Some code is rearranged slightly to make things nicer. There is\nno change, only cleanup.\n\nThere is one hook called with the tasklist_lock write-locked, as ptrace\nneeds. There is also a new hook called after exit_state changes and\nwithout locks. This is a better place for tracing work to be in the\nfuture, since it doesn\u0027t delay the whole system with locking.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "fa00b80b3c41a845b3d56f866fb40a2e98754c51", "tree": "819d6966cad1fdeef3f32497c696afb203e3720b", "parents": [ "7bcf6a2ca5f639b038c48711ebe6c4eca2036641" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:54 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: job control\n\nThis defines the tracehook_notify_jctl() hook to formalize the ptrace\neffects on the job control notifications. There is no change, only\ncleanup.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "7bcf6a2ca5f639b038c48711ebe6c4eca2036641", "tree": "4219a0725d581310f729d4616361febd700f0574", "parents": [ "283d7559e7712f95a05331eb0a85394c6368101b" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:53 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: get_signal_to_deliver\n\nThis defines the tracehook_get_signal() hook to allow tracing code to slip\nin before normal signal dequeuing. This lays the groundwork for new\ntracing features that can inject synthetic signals outside the normal\nqueue or control the disposition of delivered signals. The calling\nconvention lets tracehook_get_signal() decide both exactly what will\nhappen and what signal number to report in the handler/exit.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "445a91d2fe3667fb8fc251433645f686933cf56a", "tree": "496f0715ba3079f3bf98e9808480fe7f25872b16", "parents": [ "35de254dc60f91004b3b5ebb1fc7b2c3093d6032" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:52 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: tracehook_consider_fatal_signal\n\nThis defines tracehook_consider_fatal_signal() has a fine-grained hook for\ndeciding to skip the special cases for a fatal signal, as ptrace does.\nThere is no change, only cleanup.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "35de254dc60f91004b3b5ebb1fc7b2c3093d6032", "tree": "101e174247bb3174c8abe0fb8d1405ed6a4db6d1", "parents": [ "c45aea27617d6a1e0aacddc3b0233f704222fcbd" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:51 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:09 2008 -0700" }, "message": "tracehook: tracehook_consider_ignored_signal\n\nThis defines tracehook_consider_ignored_signal() has a fine-grained hook\nfor deciding to prevent the normal short-circuit of sending an ignored\nsignal, as ptrace does. There is no change, only cleanup.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "ff1188646c6870f336e910fb894eeed74f50471f", "tree": "7e0c7ecad4ab3f9c2f20540b289b4cd932453e67", "parents": [ "6341c393fcc37d58727865f1ee2f65e632e9d4f0" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Fri Jul 25 19:45:45 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:08 2008 -0700" }, "message": "tracehook: unexport ptrace_notify\n\nThe ptrace_notify() function should not be called by any modules. It was\nonly ever exported to be called by binfmt exec functions. But that is no\nlonger necessary since fs/exec.c deals with that generically now. There\nshould be no calls to ptrace_notify() from outside the core kernel.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nReviewed-by: 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": "19b0cfcca41dd772065671ad0584e1cea0f3fd13", "tree": "6a8a627e2e2f3f26b02fc4228517c421e66a8043", "parents": [ "33166b1ffca5e1945246bcaa77d72a22b0d3e531" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Fri Jul 25 01:48:35 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:45 2008 -0700" }, "message": "pidns: remove now unused kill_proc function\n\nThis function operated on a pid_t to kill a task, which is no longer valid\nin a containerized system.\n\nIt has finally lost all its users and we can safely remove it from the\ntree.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "999d9fc1670bc082928b93b11d1f2e0e417d973c", "tree": "e540e7fd2fab970ba2be5e39ac9f8282a373bc24", "parents": [ "32ecb1f26dd50eeaac4e3f4dea4541c97848e459" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:41 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:39 2008 -0700" }, "message": "coredump: move mm-\u003ecore_waiters into struct core_state\n\nMove mm-\u003ecore_waiters into \"struct core_state\" allocated on stack. This\nshrinks mm_struct a little bit and allows further changes.\n\nThis patch mostly does s/core_waiters/core_state. The only essential\nchange is that coredump_wait() must clear mm-\u003ecore_state before return.\n\nThe coredump_wait()\u0027s path is uglified and .text grows by 30 bytes, this\nis fixed by the next patch.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "3d749b9e676b26584a47e75c235aa6f69d0697ae", "tree": "2db2dab84cde07c4872ac5e8d4078f2a3fc4271d", "parents": [ "364d3c13c17f45da6d638011078d4c4d3070d719" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:37 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:39 2008 -0700" }, "message": "ptrace: simplify ptrace_stop()-\u003esigkill_pending() path\n\n1. SIGKILL can\u0027t be blocked, remove this check from sigkill_pending().\n\n2. When ptrace_stop() sees sigkill_pending() \u003d\u003d T, it can just return.\n Kill \"int killed\" and simplify the code. This also is more correct,\n the tracer shouldn\u0027t see us in TASK_TRACED if we are not going to\n stop.\n\nI strongly believe this code needs further changes. We should do the \"was\nthis task killed\" check unconditionally, currently it depends on\narch_ptrace_stop_needed(). On the other hand, sigkill_pending() isn\u0027t\nvery clever. If the task was killed tkill(SIGKILL), the signal can be\nalready dequeued if the caller is do_exit().\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "bc64efd220dcd4449aef8dd2564d73127b583b09", "tree": "21739809b2ab9b13677cae99f175c3983114abc6", "parents": [ "d8878ba3f05ae5bbfad5a6e72e5121c0ea35f989" ], "author": { "name": "Gustavo Fernando Padovan", "email": "gustavo@las.ic.unicamp.br", "time": "Fri Jul 25 01:47:33 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:39 2008 -0700" }, "message": "kernel/signal.c: change vars pid and tgid types to pid_t\n\nChange the type of pid and tgid variables from int to the POSIX type\npid_t.\n\nSigned-off-by: Gustavo F. Padovan \u003cgustavo@las.ic.unicamp.br\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\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": "d8878ba3f05ae5bbfad5a6e72e5121c0ea35f989", "tree": "31fbf7bd9605c103d8ef9e538beb54da8df04540", "parents": [ "e4901f92a8dbe843e76651a50f7a2a6dd3d53474" ], "author": { "name": "Michael Kerrisk", "email": "mtk.manpages@googlemail.com", "time": "Fri Jul 25 01:47:32 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:39 2008 -0700" }, "message": "signals: make siginfo_t si_utime + si_sstime report times in USER_HZ, not HZ\n\nIn the switch to configurable HZ in 2.6, the treatment of the si_utime and\nsi_stime fields that are exposed to userland via the siginfo structure\nlooks to have been botched. As things stand, these fields report times in\nunits of HZ, so that userland gets information that varies depending on\nthe HZ that the kernel was configured with. This patch changes the\nreported values to use USER_HZ units.\n\nSigned-off-by: Michael Kerrisk \u003cmtk.manpages@gmail.com\u003e\nAcked-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\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": "2b201a9eddf509e8e935b45e573648e36f4b623f", "tree": "84cbb34667bcde5566e5c3d127c9089bb9792dab", "parents": [ "92413d771e7123304fb4b9efd2a00cccc946e383" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:31 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:38 2008 -0700" }, "message": "signals: do_signal_stop: kill the SIGNAL_UNKILLABLE check\n\nfae5fa44f1fd079ffbed8e0add929dd7bbd1347f changed do_signal_stop() to check\nSIGNAL_UNKILLABLE, this wasn\u0027t needed. If signal_group_exit() \u003d\u003d F, the\nsignal sent to SIGNAL_UNKILLABLE task must be already filtered out by the\ncaller, get_signal_to_deliver(). And if signal_group_exit() \u003d\u003d T we are\nnot going to stop.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "92413d771e7123304fb4b9efd2a00cccc946e383", "tree": "c973360a9bec6c965b9f4b41923357feec14aa3a", "parents": [ "3854a771821c970065e3203a0b40ddc4101538cc" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:30 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:38 2008 -0700" }, "message": "signals: dequeue_signal: don\u0027t check SIGNAL_GROUP_EXIT when setting SIGNAL_STOP_DEQUEUED\n\ndequeue_signal() checks SIGNAL_GROUP_EXIT before setting\nSIGNAL_STOP_DEQUEUED. This was added by\n788e05a67c343fa22f2ae1d3ca264e7f15c25eaf a long ago to avoid the\ncoredump/SIGSTOP race.\n\nSince then the related code was changed, and now this subtle check is both\nincomplete and unneeded at the same time. It is incomplete because\nnowadays exec() doesn\u0027t set SIGNAL_GROUP_EXIT, so in fact we should check\nsignal_group_exit() to avoid a similar race. Fortunately, we doesn\u0027t need\nthe check at all. The only function which relies on SIGNAL_STOP_DEQUEUED\nis do_signal_stop(), and it ignores this flag if signal_group_exit() \u003d\u003d T,\nthis covers the SIGNAL_GROUP_EXIT case.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "100360f03077663b7bef3af44805b6cf700c3bee", "tree": "338cedfd71101fdaeb317fc0692d1bf93bd3a1a7", "parents": [ "d4434207616980885205c605697868c0f07e4378" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:29 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:38 2008 -0700" }, "message": "signals: change collect_signal() to return void\n\nWith the recent changes collect_signal() always returns true. Change it\nto return void and update the single caller.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d4434207616980885205c605697868c0f07e4378", "tree": "82415d69b92557e09d57f42c49aba05ef4775a89", "parents": [ "6715ca451cfff1c9ce4b33ad9918a1dacf43997c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:28 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:38 2008 -0700" }, "message": "signals: collect_signal: simplify the \"still_pending\" logic\n\nFactor out sigdelset() calls and remove the \"still_pending\" variable.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "6715ca451cfff1c9ce4b33ad9918a1dacf43997c", "tree": "a66cdfd7026e7ad40ad0675a35ff35886216ed5f", "parents": [ "96347e7759e2e433c427defa0fa1adfc8cce6226" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Jul 25 01:47:27 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 25 10:53:38 2008 -0700" }, "message": "signals: collect_signal: remove the unneeded sigismember() check\n\ncollect_signal() checks sigismember(\u0026list-\u003esignal, sig), this is not\nneeded. This \"sig\" was just found by next_signal(), so it must be valid.\n\nWe have a (completely broken) call to -\u003enotifier in between, but it must\nnot play with sigpending-\u003esignal bits or unlock -\u003esiglock.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "ba661292a2bc6ddd305a212b0526e5dc22195fe7", "tree": "69d5749d967c67deffb6d804f61c359f5da928ad", "parents": [ "54da1174922cddd4be83d5a364b2e0fdd693f513" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Jul 23 20:52:05 2008 +0400" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Thu Jul 24 01:26:08 2008 +0200" }, "message": "posix-timers: fix posix_timer_event() vs dequeue_signal() race\n\nThe bug was reported and analysed by Mark McLoughlin \u003cmarkmc@redhat.com\u003e,\nthe patch is based on his and Roland\u0027s suggestions.\n\nposix_timer_event() always rewrites the pre-allocated siginfo before sending\nthe signal. Most of the written info is the same all the time, but memset(0)\nis very wrong. If -\u003esigq is queued we can race with collect_signal() which\ncan fail to find this siginfo looking at .si_signo, or copy_siginfo() can\ncopy the wrong .si_code/si_tid/etc.\n\nIn short, sys_timer_settime() can in fact stop the active timer, or the user\ncan receive the siginfo with the wrong .si_xxx values.\n\nMove \"memset(-\u003einfo, 0)\" from posix_timer_event() to alloc_posix_timer(),\nchange send_sigqueue() to set .si_overrun \u003d 0 when -\u003esigq is not queued.\nIt would be nice to move the whole sigq-\u003einfo initialization from send to\ncreate path, but this is not easy to do without uglifying timer_create()\nfurther.\n\nAs Roland rightly pointed out, we need more cleanups/fixes here, see the\n\"FIXME\" comment in the patch. Hopefully this patch makes sense anyway, and\nit can mask the most bad implications.\n\nReported-by: Mark McLoughlin \u003cmarkmc@redhat.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Mark McLoughlin \u003cmarkmc@redhat.com\u003e\nCc: Oliver Pinter \u003coliver.pntr@gmail.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: stable@kernel.org\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\n kernel/posix-timers.c | 17 +++++++++++++----\n kernel/signal.c | 1 +\n 2 files changed, 14 insertions(+), 4 deletions(-)\n\n" }, { "commit": "cbaffba12ce08beb3e80bfda148ee0fa14aac188", "tree": "b35f29814b46593d864e8c8921e9eccac5a5a173", "parents": [ "c8e85b4f4b9ee23bf0e79bdeb3da274a0f9c663f" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Mon May 26 20:55:42 2008 +0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 26 10:37:07 2008 -0700" }, "message": "posix timers: discard SI_TIMER signals on exec\n\nBased on Roland\u0027s patch. This approach was suggested by Austin Clements\nfrom the very beginning, and then by Linus.\n\nAs Austin pointed out, the execing task can be killed by SI_TIMER signal\nbecause exec flushes the signal handlers, but doesn\u0027t discard the pending\nsignals generated by posix timers. Perhaps not a bug, but people find this\nsurprising. See http://bugzilla.kernel.org/show_bug.cgi?id\u003d10460\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Austin Clements \u003camdragon+kernelbugzilla@mit.edu\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c8e85b4f4b9ee23bf0e79bdeb3da274a0f9c663f", "tree": "5eadfdc177f24261379f36798c3366df9be4a00a", "parents": [ "84a881657d391121cd88c37f0a312dec3528fa44" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Mon May 26 20:55:42 2008 +0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 26 10:37:06 2008 -0700" }, "message": "posix timers: sigqueue_free: don\u0027t free sigqueue if it is queued\n\nCurrently sigqueue_free() removes sigqueue from list, but doesn\u0027t cancel the\npending signal. This is not consistent, the task should either receive the\n\"full\" signal along with siginfo_t, or it shouldn\u0027t receive the signal at all.\n\nChange sigqueue_free() to clear SIGQUEUE_PREALLOC but leave sigqueue on list\nif it is queued.\n\nThis is a user-visible change. If the signal is blocked, it stays queued\nafter sys_timer_delete() until unblocked with the \"stale\" si_code/si_value,\nand of course it is still counted wrt RLIMIT_SIGPENDING which also limits\nthe number of posix timers.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Austin Clements \u003camdragon+kernelbugzilla@mit.edu\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "da7978b0348d497688541e2d2f5739aa2a2c334f", "tree": "f4f55bf4293ff203f598e7a57959da1c5a7ad295", "parents": [ "dfc7064500061677720fa26352963c772d3ebe6b" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri May 23 13:04:41 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 24 09:56:10 2008 -0700" }, "message": "signals: fix sigqueue_free() vs __exit_signal() race\n\n__exit_signal() does flush_sigqueue(tsk-\u003epending) outside of -\u003esiglock.\nThis can race with another thread doing sigqueue_free(), we can free the\nsame SIGQUEUE_PREALLOC sigqueue twice or corrupt the pending-\u003elist.\n\nNote that even sys_exit_group() can trigger this race, not only\nsys_timer_delete().\n\nMove the callsite of flush_sigqueue(tsk-\u003epending) under -\u003esiglock.\n\nThis patch doesn\u0027t touch flush_sigqueue(-\u003eshared_pending) below, it is\ncalled when there are no other threads which can play with signals, and\nsigqueue_free() can\u0027t be used outside of our thread group.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "4e4c22c71144c1b2e22c257ec6cf08ccb5be1165", "tree": "0a41fb405ab60654f70ead9698d5ce414a066c37", "parents": [ "80fe728d593e3a048a56610de932919f7d6d968a" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Wed Apr 30 00:53:06 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:37 2008 -0700" }, "message": "signals: add set_restore_sigmask\n\nThis adds the set_restore_sigmask() inline in \u003clinux/thread_info.h\u003e and\nreplaces every set_thread_flag(TIF_RESTORE_SIGMASK) with a call to it. No\nchange, but abstracts the details of the flag protocol from all the calls.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.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": "80fe728d593e3a048a56610de932919f7d6d968a", "tree": "ff8effec1d05d871bcbb1347947d2a0970a8ff32", "parents": [ "7a5e873f096e04e6d8719e4ecb7b70d2decca503" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:05 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:37 2008 -0700" }, "message": "signals: allow the kernel to actually kill /sbin/init\n\nCurrently the buggy /sbin/init hangs if SIGSEGV/etc happens. The kernel sends\nthe signal, init dequeues it and ignores, returns from the exception, repeats\nthe faulting instruction, and so on forever.\n\nImho, such a behaviour is not good. I think that the explicit loud death of\nthe buggy /sbin/init is better than the silent hang.\n\nChange force_sig_info() to clear SIGNAL_UNKILLABLE when the task should be\nreally killed.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "fae5fa44f1fd079ffbed8e0add929dd7bbd1347f", "tree": "8990ac958d29733cb61733ae69265472f5e1d13c", "parents": [ "193191035ad6268db9f561e81e3474b8be89a5ba" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:03 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:37 2008 -0700" }, "message": "signals: fix /sbin/init protection from unwanted signals\n\nThe global init has a lot of long standing problems with the unhandled fatal\nsignals.\n\n\t- The \"is_global_init(current)\" check in get_signal_to_deliver()\n\t protects only the main thread. Sub-thread can dequee the fatal\n\t signal and shutdown the whole thread group except the main thread.\n\t If it dequeues SIGSTOP /sbin/init will be stopped, this is not\n\t right too. Note that we can\u0027t use is_global_init(-\u003egroup_leader),\n\t this breaks exec and this can\u0027t solve other problems we have.\n\n\t- Even if afterwards ignored, the fatal signals sets SIGNAL_GROUP_EXIT\n\t on delivery. This breaks exec, has other bad implications, and this\n\t is just wrong.\n\nIntroduce the new SIGNAL_UNKILLABLE flag to fix these problems. It also helps\nto solve some other problems addressed by the subsequent patches.\n\nCurrently we use this flag for the global init only, but it could also be used\nby kthreads and (perhaps) by the sub-namespace inits.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: \"Eric W. Biederman\" \u003cebiederm@xmission.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": "193191035ad6268db9f561e81e3474b8be89a5ba", "tree": "ac4af28bf65dd7fc6c40d7260806725d9f3330ba", "parents": [ "2e2ba22ea4fd4bb85f0fa37c521066db6775cbef" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:02 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:37 2008 -0700" }, "message": "signals: check_kill_permission: remove tasklist_lock\n\nNow that task_session() can\u0027t return a false NULL, check_kill_permission()\ndoesn\u0027t need tasklist_lock.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "2e2ba22ea4fd4bb85f0fa37c521066db6775cbef", "tree": "aada1523affbcbe1301decad43ad0875e1f680c4", "parents": [ "53c30337f2c61aff6eecf2a446e839641172f9bd" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:01 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:37 2008 -0700" }, "message": "signals: check_kill_permission: check session under tasklist_lock\n\nThis wasn\u0027t documented, but as Atsushi Tsuji pointed out\ncheck_kill_permission() needs tasklist_lock for task_session_nr(). I missed\nthis fact when removed tasklist from the callers.\n\nChange check_kill_permission() to take tasklist_lock for the SIGCONT case.\nRe-order security checks so that we take tasklist_lock only if/when it is\nactually needed. This is a minimal fix for now, tasklist will be removed\nlater.\n\nAlso change the code to use task_session() instead of task_session_nr().\n\nAlso, remove the SIGCONT check from cap_task_kill(), it is bogus (and the\nwhole function is bogus. Serge, Eric, why it is still alive?).\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: Atsushi Tsuji \u003ca-tsuji@bk.jp.nec.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "53c30337f2c61aff6eecf2a446e839641172f9bd", "tree": "268c52124c19b379b7f9c32430414acb606699cb", "parents": [ "021e1ae3d85a76ce962a300c96813f04ae50c87c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: send_signal: be paranoid about signalfd_notify()\n\nsend_signal() shouldn\u0027t call signalfd_notify() if it then fails with -EAGAIN.\nHarmless, just a paranoid cleanup.\n\nAlso remove the comment. It is obsolete, signalfd_notify() was simplified and\ndoes a simple wakeup.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: Davide Libenzi \u003cdavidel@xmailserver.org\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": "021e1ae3d85a76ce962a300c96813f04ae50c87c", "tree": "1c3cfbf8151c339baa3b35a4a78fa50ccc7594ab", "parents": [ "7e695a5ef5c1c768d7feb75cc61e42f13d763623" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:53:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: document CLD_CONTINUED notification mechanics\n\nA couple of small comments about how CLD_CONTINUED notification works.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "7e695a5ef5c1c768d7feb75cc61e42f13d763623", "tree": "97bf6b5865e097a77b791342eccaeddc7132b9fc", "parents": [ "2dce81bff28dceb2153c901883a56f278d91db65" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:59 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: fold sig_ignored() into handle_stop_signal()\n\nRename handle_stop_signal() to prepare_signal(), make it return a boolean, and\nmove the callsites of sig_ignored() into it.\n\nNo functional changes for now. But it would be nice to factor out the \"should\nwe drop this signal\" checks as much as possible, before we try to fix the bugs\nwith the sub-namespace init\u0027s signals (actually the global /sbin/init has some\nproblems with signals too).\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "2dce81bff28dceb2153c901883a56f278d91db65", "tree": "831da312ea00c5c505d376098b5b1a90fe33fed7", "parents": [ "34c8f07b9ac499a807918eda377193a55f64f8df" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:58 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: cleanup the usage of print_fatal_signal()\n\nMove the callsite of print_fatal_signal() down, under \"if\n(sig_kernel_coredump(signr))\", so we don\u0027t need to check signr !\u003d SIGKILL.\n\nWe are only interested in the sig_kernel_coredump() signals anyway, and due to\nthe previous changes we almost never can see other fatal signals here except\nSIGKILL.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "34c8f07b9ac499a807918eda377193a55f64f8df", "tree": "042cf061406273dcfdb3fabe4b07425e4cb8f0aa", "parents": [ "ac5c215383f43a106ba4ef298126bf78c126f5e9" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:58 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: handle_stop_signal: don\u0027t worry about SIGKILL\n\nhandle_stop_signal() clears SIGNAL_STOP_DEQUEUED when sig \u003d\u003d SIGKILL. Remove\nthis nasty special case. It was needed to prevent the race with group stop\nand exit caused by thread-specific SIGKILL. Now that we use complete_signal()\nfor private signals too this is not needed, complete_signal() will notice\nSIGKILL and abort the soon-to-begin group stop.\n\nExcept: the target thread is dead (has PF_EXITING). But in that case we\nshould not just clear SIGNAL_STOP_DEQUEUED and nothing more. We should either\nkill the whole thread group, or silently ignore the signal.\n\nI suspect we are not right wrt zombie leaders, but this is another issue which\nand should be fixed separately. Note that this check can\u0027t abort the group\nstop if it was already started/finished, this check only adds a subtle side\neffect if we race with the thread which has already dequeued sig_kernel_stop()\nsignal and temporary released -\u003esiglock.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "ac5c215383f43a106ba4ef298126bf78c126f5e9", "tree": "77340eea86103df7bc3b4a6c6d83f4140004bc9c", "parents": [ "e62e6650e99a3dffcd0bf0d063cd818fbc13fa95" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:57 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: join send_sigqueue() with send_group_sigqueue()\n\nWe export send_sigqueue() and send_group_sigqueue() for the only user,\nposix_timer_event(). This is a bit silly, because both are just trivial\nhelpers on top of do_send_sigqueue() and because the we pass the unused\n.si_signo parameter.\n\nKill them both, rename do_send_sigqueue() to send_sigqueue(), and export it.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "e62e6650e99a3dffcd0bf0d063cd818fbc13fa95", "tree": "75f79d45eb89fb75014117b6fdc67bae1608f826", "parents": [ "4cd4b6d4e0372075f846feb85aea016cbdbfec4c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:56 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: unify send_sigqueue/send_group_sigqueue completely\n\nSuggested by Pavel Emelyanov.\n\nsend_sigqueue/send_group_sigqueue are only differ in how they lock -\u003esiglock.\nUnify them. send_group_sigqueue() uses spin_lock() because it knows the task\ncan\u0027t exit, but in that case lock_task_sighand() can\u0027t fail and doesn\u0027t hurt.\n\nNote that the \"sig\" argument is ignored, it is always equal to -\u003esi_signo.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "4cd4b6d4e0372075f846feb85aea016cbdbfec4c", "tree": "219dd1d4de6313bcc9dd232fe3186423875bea5e", "parents": [ "5fcd835bf8c2cde06404559b1904e2f1dfcb4567" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:55 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: fold complete_signal() into send_signal/do_send_sigqueue\n\nFactor out complete_signal() callsites. This change completely unifies the\nhelpers sending the specific/group signals.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "5fcd835bf8c2cde06404559b1904e2f1dfcb4567", "tree": "c370d13b34d7c516c449427aebfde5bd29c83489", "parents": [ "2ca3515aa57224edf0151e05a8c9f21a76bf5957" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:55 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:36 2008 -0700" }, "message": "signals: use __group_complete_signal() for the specific signals too\n\nBased on Pavel Emelyanov\u0027s suggestion.\n\nRename __group_complete_signal() to complete_signal() and use it to process\nthe specific signals too. To do this we simply add the \"int group\" argument.\n\nThis allows us to greatly simply the signal-sending code and adds a useful\nbehaviour change. We can avoid the unneeded wakeups for the private signals\nbecause wants_signal() is more clever than sigismember(blocked), but more\nimportantly we now take into account the fatal specific signals too.\n\nThe latter allows us to kill some subtle checks in handle_stop_signal() and\nmakes the specific/group signal\u0027s behaviour more consistent. For example,\ncurrently sigtimedwait(FATAL_SIGNAL) behaves differently depending on was the\nsignal sent by kill() or tkill() if the signal was not blocked.\n\nAnd. This allows us to tweak/fix the behaviour when the specific signal is\nsent to the dying/dead -\u003egroup_leader.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "2ca3515aa57224edf0151e05a8c9f21a76bf5957", "tree": "fd0154e8ecb2f0f39425cba207cf7b74bc18c0ff", "parents": [ "71f11dc025055cb2ef9226424f26b3287efadd26" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:54 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: change send_signal/do_send_sigqueue to take \"boolean group\" parameter\n\nsend_signal() is used either with -\u003epending or with -\u003esignal-\u003eshared_pending.\nChange it to take \"int group\" instead, this argument will be re-used later.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "71f11dc025055cb2ef9226424f26b3287efadd26", "tree": "d489b52d6d1779af1a325d4bed2354a07ca1cc60", "parents": [ "db51aeccd7097ce19a522a4c5ff91c320f870e2b" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:53 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: move the definition of __group_complete_signal() up\n\nMove the unchanged definition of __group_complete_signal() so that send_signal\ncan see it. To simplify the reading of the next patches.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "db51aeccd7097ce19a522a4c5ff91c320f870e2b", "tree": "373bdfb55d1d90c04d53d4a419bcd156fe7f054c", "parents": [ "08d2c30ce98d274137f12b0a9b9c74137455922c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:52 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: microoptimize the usage of -\u003ecurr_target\n\nSuggested by Roland McGrath.\n\nInitialize signal-\u003ecurr_target in copy_signal(). This way -\u003ecurr_target is\nnever \u003d\u003d NULL, we can kill the check in __group_complete_signal\u0027s hot path.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "08d2c30ce98d274137f12b0a9b9c74137455922c", "tree": "cc649fd249aef0c4d227406b17a1a75c9d327e50", "parents": [ "3547ff3aefbe092ca35506c60c02e2d17a4f2199" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:51 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: send_sig_info: don\u0027t take tasklist_lock\n\nThe comment in send_sig_info() is wrong, tasklist_lock can\u0027t help.\n\nThe caller must ensure the task can\u0027t go away, otherwise -\u003esighand can be NULL\neven before we take the lock.\n\np-\u003esighand could be changed by exec(), but I can\u0027t imagine how it is possible\nto prevent exit(), but not exec().\n\nSince the things seem to work, I assume all callers are correct. However,\ndrm_vbl_send_signals() looks broken. block_all_signals() which is solely used\nby drm is definitely broken.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "3547ff3aefbe092ca35506c60c02e2d17a4f2199", "tree": "722222124ae06ef45947fae20913bbd71e179c49", "parents": [ "6e65acba7ca8169e38ab55d62d52f29a75fb141f" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:51 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: do_tkill: don\u0027t use tasklist_lock\n\nConvert do_tkill() to use rcu_read_lock() + lock_task_sighand() to avoid\ntaking tasklist lock.\n\nNote that we don\u0027t return an error if lock_task_sighand() fails, we pretend\nthe task dies after receiving the signal. Otherwise, we should fight with the\nnasty races with mt-exec without having any advantage.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "6e65acba7ca8169e38ab55d62d52f29a75fb141f", "tree": "a691ee9392b0f2ed8a434e5a5cd6e6a213b70f1b", "parents": [ "c99fcf28b87d8cab592db7571e3164f5cb54c5b3" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:50 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: move handle_stop_signal() into send_signal()\n\nMove handle_stop_signal() into send_signal(). This factors out a couple of\ncallsites and allows us to do further unifications.\n\nAlso, with this change specific_send_sig_info() does handle_stop_signal().\nNot that this is really important, we never send STOP/CONT via send_sig() and\nfriends, but still this looks more consistent.\n\nThe only (afaics) special case is get_signal_to_deliver(). If the traced task\ndequeues SIGCONT, it can re-send it to itself after ptrace_stop() if the\nsignal was blocked by debugger. In that case handle_stop_signal() is\nunnecessary, but hopefully not a problem.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "c99fcf28b87d8cab592db7571e3164f5cb54c5b3", "tree": "356ce5d2b0bfd33794a81bf4cd90abd7859776da", "parents": [ "f8c5b5c06f63fe9aaebefbf9f0b79909066b1b6c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:49 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: send_group_sigqueue: don\u0027t take tasklist_lock\n\nhandle_stop_signal() was changed, now send_group_sigqueue() doesn\u0027t need\ntasklist_lock.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "f8c5b5c06f63fe9aaebefbf9f0b79909066b1b6c", "tree": "58f7cde6d5b6c52ad927658521a32822661f048d", "parents": [ "5fc894bb4fb1de8373d1d5fb6db19204a16859e8" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:49 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: __group_complete_signal: cache the value of p-\u003esignal\n\nCosmetic, cache p-\u003esignal to make the code a bit more readable.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "5fc894bb4fb1de8373d1d5fb6db19204a16859e8", "tree": "167c513b50b8ceed48a8fb7e4c2e63397f0fa7d7", "parents": [ "5c193e8871b76f3bf8ed1e31f7af7c70890ebc4f" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:48 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: send_sigqueue: don\u0027t forget about handle_stop_signal()\n\nsend_group_sigqueue() calls handle_stop_signal(), send_sigqueue() doesn\u0027t.\nThis is not consistent and in fact I\u0027d say this is (minor) bug.\n\nMove handle_stop_signal() from send_group_sigqueue() to do_send_sigqueue(),\nthe latter is called by send_sigqueue() too.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "5c193e8871b76f3bf8ed1e31f7af7c70890ebc4f", "tree": "533b708c86908e8d1a439ef63a1c1515871715d7", "parents": [ "f6b76d4fb0039e077824be85ed4ac94e96beef86" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:48 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "signals: send_sigqueue: don\u0027t take rcu lock\n\nlock_task_sighand() was changed, send_sigqueue() doesn\u0027t need rcu_read_lock()\nany longer.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "f6b76d4fb0039e077824be85ed4ac94e96beef86", "tree": "70a9ccff1d0194ff3a5ab23837f5a2ad839af6ba", "parents": [ "ad16a4606939ce1bedb79c87e412467be803e990" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:47 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:35 2008 -0700" }, "message": "get_signal_to_deliver: use the cached -\u003esignal/sighand values\n\nCache the values of current-\u003esignal/sighand. Shrinks .text a bit and makes\nthe code more readable. Also, remove \"sigset_t *mask\", it is pointless\nbecause in fact we save the constant offset.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Jiri Kosina \u003cjkosina@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ad16a4606939ce1bedb79c87e412467be803e990", "tree": "d9b1e96c2520c9c08660c4dfd5258b2acdc8d1ef", "parents": [ "fc321d2e60d6f4eee17206612d0b50519f526daf" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:46 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "handle_stop_signal: use the cached p-\u003esignal value\n\nCache the value of p-\u003esignal, and change the code to use while_each_thread()\nhelper.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Jiri Kosina \u003cjkosina@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fc321d2e60d6f4eee17206612d0b50519f526daf", "tree": "79dd03d347ccee2f5a538823f7434dfd40e302be", "parents": [ "6ca25b551309eb1b1b41f83414a92f7472e0b23d" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:46 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "handle_stop_signal: unify partial/full stop handling\n\nNow that handle_stop_signal() doesn\u0027t drop -\u003esiglock, we can\u0027t see both\n-\u003egroup_stop_count \u0026\u0026 SIGNAL_STOP_STOPPED. Merge two \"if\" branches.\n\nAs Roland pointed out, we never actually needed 2 do_notify_parent_cldstop()\ncalls.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Jiri Kosina \u003cjkosina@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6ca25b551309eb1b1b41f83414a92f7472e0b23d", "tree": "a2c5785ec484ec997ac6a03d3fceae04e4d21795", "parents": [ "e442055193e4584218006e616c9bdce0c5e9ae5c" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:45 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "kill_pid_info: don\u0027t take now unneeded tasklist_lock\n\nPreviously handle_stop_signal(SIGCONT) could drop -\u003esiglock. That is why\nkill_pid_info(SIGCONT) takes tasklist_lock to make sure the target task can\u0027t\ngo away after unlock. Not needed now.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Jiri Kosina \u003cjkosina@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e442055193e4584218006e616c9bdce0c5e9ae5c", "tree": "5e444253953c2f48c0a194bf8c5799f03bb2e266", "parents": [ "3b5e9e53c6f31b5a5a0f5c43707503c62bdefa46" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:44 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "signals: re-assign CLD_CONTINUED notification from the sender to reciever\n\nBased on discussion with Jiri and Roland.\n\nIn short: currently handle_stop_signal(SIGCONT, p) sends the notification to\np-\u003eparent, with this patch p itself notifies its parent when it becomes\nrunning.\n\nhandle_stop_signal(SIGCONT) has to drop -\u003esiglock temporary in order to notify\nthe parent with do_notify_parent_cldstop(). This leads to multiple problems:\n\n\t- as Jiri Kosina pointed out, the stopped task can resume without\n\t actually seeing SIGCONT which may have a handler.\n\n\t- we race with another sig_kernel_stop() signal which may come in\n\t that window.\n\n\t- we race with sig_fatal() signals which may set SIGNAL_GROUP_EXIT\n\t in that window.\n\n\t- we can\u0027t avoid taking tasklist_lock() while sending SIGCONT.\n\nWith this patch handle_stop_signal() just sets the new SIGNAL_CLD_CONTINUED\nflag in p-\u003esignal-\u003eflags and returns. The notification is sent by the first\ntask which returns from finish_stop() (there should be at least one) or any\nother signalled thread from get_signal_to_deliver().\n\nThis is a user-visible change. Say, currently kill(SIGCONT, stopped_child)\ncan\u0027t return without seeing SIGCHLD, with this patch SIGCHLD can be delayed\nunpredictably. Another difference is that if the child is ptraced by another\nprocess, CLD_CONTINUED may be delivered to -\u003ereal_parent after ptrace_detach()\nwhile currently it always goes to the tracer which doesn\u0027t actually need this\nnotification. Hopefully not a problem.\n\nThe patch asks for the futher obvious cleanups, I\u0027ll send them separately.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Jiri Kosina \u003cjkosina@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3b5e9e53c6f31b5a5a0f5c43707503c62bdefa46", "tree": "1244b7cf2755c06a8a793149ce4717e4a1311218", "parents": [ "9e3bd6c3fb2334be171e69b432039cd18bce4458" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:42 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "signals: cleanup security_task_kill() usage/implementation\n\nEvery implementation of -\u003etask_kill() does nothing when the signal comes from\nthe kernel. This is correct, but means that check_kill_permission() should\ncall security_task_kill() only for SI_FROMUSER() case, and we can remove the\nsame check from -\u003etask_kill() implementations.\n\n(sadly, check_kill_permission() is the last user of signal-\u003esession/__session\n but we can\u0027t s/task_session_nr/task_session/ here).\n\nNOTE: Eric W. Biederman pointed out cap_task_kill() should die, and I think\nhe is very right.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\nCc: David Quigley \u003cdpquigl@tycho.nsa.gov\u003e\nCc: Eric Paris \u003ceparis@redhat.com\u003e\nCc: Harald Welte \u003claforge@gnumonks.org\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Stephen Smalley \u003csds@tycho.nsa.gov\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9e3bd6c3fb2334be171e69b432039cd18bce4458", "tree": "199963534d6379457d84e6f2ead2b104088182dd", "parents": [ "c5363d03637885310f1101b95cbbd26d067b4c8d" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:41 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "signals: consolidate send_sigqueue and send_group_sigqueue\n\nBoth functions do the same thing after proper locking, but with\ndifferent sigpending structs, so move the common code into a helper.\n\nAfter this we have 4 places that look very similar: send_sigqueue: calls\ndo_send_sigqueue and signal_wakeup send_group_sigqueue: calls\ndo_send_sigqueue and __group_complete_signal __group_send_sig_info:\ncalls send_signal and __group_complete_signal specific_send_sig_info:\ncalls send_signal and signal_wakeup\n\nBesides, send_signal performs actions similar to do_send_sigqueue\u0027s\nand __group_complete_signal - to signal_wakeup.\n\nIt looks like they can be consolidated gracefully.\n\nOleg said:\n\n Personally, I think this change is very good. But send_sigqueue() and\n send_group_sigqueue() have a very subtle difference which I was never able\n to understand.\n\n Let\u0027s suppose that sigqueue is already queued, and the signal is ignored\n (the latter means we should re-schedule cpu timer or handle overrruns). In\n that case send_sigqueue() returns 0, but send_group_sigqueue() returns 1.\n\n I think this is not the problem (in fact, I think this patch makes the\n behaviour more correct), but I hope Thomas can take a look and confirm.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c5363d03637885310f1101b95cbbd26d067b4c8d", "tree": "e22fd822f03dd5437c53bf459ae986723aaa1613", "parents": [ "93585eeaf3d42d608cd7232e7420c93fb676bba1" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:40 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "signals: clean dequeue_signal from excess checks and assignments\n\nThe signr variable may be declared without initialization - it is set ro the\nreturn value from __dequeue_signal() right at the function beginning.\n\nBesides, after recalc_sigpending() two checks for signr to be not 0 may be\nmerged into one. Both if-s become easier to read.\n\nThanks to Oleg for pointing out mistakes in the first version of this patch.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "93585eeaf3d42d608cd7232e7420c93fb676bba1", "tree": "7c87ab85480af514a055563f03aec6ef7b1e3f87", "parents": [ "06fffb1267c9d986687b69d74a46ee332a50575e" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:39 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:34 2008 -0700" }, "message": "signals: consolidate checks for whether or not to ignore a signal\n\nBoth sig_ignored() and do_sigaction() check for signr to be explicitly or\nimplicitly ignored. Introduce a helper for them.\n\nThis patch is aimed to help handling signals by pid namespace\u0027s init, and was\nderived from one of Oleg\u0027s patches\nhttps://lists.linux-foundation.org/pipermail/containers/2007-December/009308.html\nso, if he doesn\u0027t mind, he should be considered as an author.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1406f2d321bae5ac5ff729dcb773336d9c05ec74", "tree": "178f7e2d800cee7be211225952f397458707587a", "parents": [ "bfc4b0890af566940de6e7aeb4b5faf46d3c3513" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:37 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:33 2008 -0700" }, "message": "lock_task_sighand: add rcu lock/unlock\n\nMost of the callers of lock_task_sighand() doesn\u0027t actually need rcu_lock().\nlock_task_sighand() needs it only to safely play with tsk-\u003esighand, it can\ntake the lock itself.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@linux.vnet.ibm.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": "573cf9ad72c13750e86c91de43477e9dfb440523", "tree": "7b7fab20559e2e96ec7e8aeefb156b35ee2c6c55", "parents": [ "2acb024d5524eda305523c1d6061fe5ef1949165" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Wed Apr 30 00:52:36 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:33 2008 -0700" }, "message": "signals: do_signal_stop(): use signal_group_exit()\n\ndo_signal_stop() needs signal_group_exit() but checks sig-\u003egroup_exit_task.\n This (optimization) is correct, SIGNAL_STOP_DEQUEUED and SIGNAL_GROUP_EXIT\nare mutually exclusive, but looks confusing. Use signal_group_exit(), this\nis not fastpath, the code clarity is more important.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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": "2acb024d5524eda305523c1d6061fe5ef1949165", "tree": "036d7a02f2e6db3be810e42947f4caaa3b290d7f", "parents": [ "af7fff9c13d56657dc328c75590f401c99bcecd9" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:35 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:33 2008 -0700" }, "message": "signals: consolidate checking for ignored/legacy signals\n\nTwo callers for send_signal() - the specific_send_sig_info and the\n__group_send_sig_info - both check for sig to be ignored or already queued.\n\nMove these checks into send_signal() and make it return 1 to indicate that the\nsignal is dropped, but there\u0027s no error in this.\n\nBesides, merge comments and spell-check them.\n\n[oleg@tv-sign.ru: simplifications]\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "af7fff9c13d56657dc328c75590f401c99bcecd9", "tree": "8adc8d2039910efb28429cb0aec0f1bde8ff98be", "parents": [ "e1401c6bbb289d154eb0d0c292cc9f8259e4af73" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:34 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:33 2008 -0700" }, "message": "signals: turn LEGACY_QUEUE macro into static inline function\n\nThis makes the code more readable, due to less brackets and small letters in\nname.\n\nI also move it above the send_signal() as a preparation for the 3rd patch.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e1401c6bbb289d154eb0d0c292cc9f8259e4af73", "tree": "64e4e7754c240c1c34c69e2975ccfb350eb4aa1f", "parents": [ "2deb1acc653cbd5384b107d050d2deba089db2bd" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 30 00:52:34 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:33 2008 -0700" }, "message": "signals: remove unused variable from send_signal()\n\nThis function doesn\u0027t change the ret\u0027s value and thus always returns 0, with a\nsingle exception of returning -EAGAIN explicitly.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Roland McGrath \u003croland@redhat.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f5264481c8049673e2cc8c7aca410931f571ba2d", "tree": "cfe54747b75be1c83d2102d9e7f80c650602fcdc", "parents": [ "1fb7c6e4cba35b5e7cef88b58d0a2ad164d801a5" ], "author": { "name": "Pavel Machek", "email": "pavel@suse.cz", "time": "Mon Apr 21 22:15:06 2008 +0000" }, "committer": { "name": "Jesper Juhl", "email": "juhl@hera.kernel.org", "time": "Mon Apr 21 22:15:06 2008 +0000" }, "message": "trivial: small cleanups\n\nThese are small cleanups all over the tree.\n\nTrivial style and comment changes to\n fs/select.c, kernel/signal.c, kernel/stop_machine.c \u0026 mm/pdflush.c\n\nSigned-off-by: Pavel Machek \u003cpavel@suse.cz\u003e\nSigned-off-by: Jesper Juhl \u003cjesper.juhl@gmail.com\u003e\n" }, { "commit": "18c98b65279c00c3c983a4525161207f1aa6a04b", "tree": "21ddeaf029407a08a81843bf35aeb338b2d67dd7", "parents": [ "0e04388f0189fa1f6812a8e1cb6172136eada87e" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Thu Apr 17 18:44:38 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 18 08:17:57 2008 -0700" }, "message": "ptrace_signal subroutine\n\nThis breaks out the ptrace handling from get_signal_to_deliver into a\nnew subroutine. The actual code there doesn\u0027t change, and it gets\ninlined into nearly identical compiled code. This makes the function\nsubstantially shorter and thus easier to read, and it nicely isolates\nthe ptrace magic.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nAcked-by: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "13b1c3d4b49bd83d861c775ca2db54e1692a1b07", "tree": "6cefdfef300d3431f2b2b32ec86000b0132bd762", "parents": [ "976dde010e513a9c7c3117a32b7b015f84b37430" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Mon Mar 03 20:22:05 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Mar 04 07:59:54 2008 -0800" }, "message": "freezer vs stopped or traced\n\nThis changes the \"freezer\" code used by suspend/hibernate in its treatment\nof tasks in TASK_STOPPED (job control stop) and TASK_TRACED (ptrace) states.\n\nAs I understand it, the intent of the \"freezer\" is to hold all tasks\nfrom doing anything significant. For this purpose, TASK_STOPPED and\nTASK_TRACED are \"frozen enough\". It\u0027s possible the tasks might resume\nfrom ptrace calls (if the tracer were unfrozen) or from signals\n(including ones that could come via timer interrupts, etc). But this\ndoesn\u0027t matter as long as they quickly block again while \"freezing\" is\nin effect. Some minor adjustments to the signal.c code make sure that\ntry_to_freeze() very shortly follows all wakeups from both kinds of\nstop. This lets the freezer code safely leave stopped tasks unmolested.\n\nChanging this fixes the longstanding bug of seeing after resuming from\nsuspend/hibernate your shell report \"[1] Stopped\" and the like for all\nyour jobs stopped by ^Z et al, as if you had freshly fg\u0027d and ^Z\u0027d them.\nIt also removes from the freezer the arcane special case treatment for\nptrace\u0027d tasks, which relied on intimate knowledge of ptrace internals.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b5606c2d4447e80b1d72406af4e78af1eda611d4", "tree": "ebdaa1a0aae4279b84af82651c16a8777f76bfe4", "parents": [ "fbf6bfca76d50abef478ba902b8597ecbadfd390" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Wed Feb 13 15:03:16 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Feb 13 16:21:18 2008 -0800" }, "message": "remove final fastcall users\n\nfastcall always expands to empty, remove it.\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@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": "146a505d498c36de98ec161d791dd50beca7f9a3", "tree": "f126c55b34488866ff179717f1b2a0c7cdd5df6b", "parents": [ "d5df763b81946a405837b80874516dfc2a8f7ebf" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Fri Feb 08 04:19:22 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:29 2008 -0800" }, "message": "Get rid of the kill_pgrp_info() function\n\nThere\u0027s only one caller left - the kill_pgrp one - so merge these two\nfunctions and forget the kill_pgrp_info one.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nReviewed-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d5df763b81946a405837b80874516dfc2a8f7ebf", "tree": "6fdf06b2cda81d91e8aef03af4d880fc4a6cad72", "parents": [ "56496c1d83dfae0c74e2f43adb45d2d95e16c0d5" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Fri Feb 08 04:19:22 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:29 2008 -0800" }, "message": "Clean up the kill_something_info\n\nThis is the first step (of two) in removing the kill_pgrp_info.\n\nAll the users of this function are in kernel/signal.c, but all they need is to\ncall __kill_pgrp_info() with the tasklist_lock read-locked.\n\nFortunately, one of its users is the kill_something_info(), which already\nneeds this lock in one of its branches, so clean these branches up and call\nthe __kill_pgrp_info() directly.\n\nBased on Oleg\u0027s view of how this function should look.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fea9d175545b38cb3e84569400419eb81bc90fa3", "tree": "0d43fe9ed2ea6104ee8b15a3eb8da081dd08fd35", "parents": [ "46f382d2b69d2221086b823f0dbc8f32c027cac2" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Feb 08 04:19:19 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:29 2008 -0800" }, "message": "ITIMER_REAL: convert to use struct pid\n\nsignal_struct-\u003etsk points to the -\u003egroup_leader and thus we have the nasty\ncode in de_thread() which has to change it and restart -\u003ereal_timer if the\nleader is changed.\n\nUse \"struct pid *leader_pid\" instead. This also allows us to kill now\nunneeded send_group_sig_info().\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Davide Libenzi \u003cdavidel@xmailserver.org\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nAcked-by: Roland McGrath \u003croland@redhat.com\u003e\nAcked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d36174bc2bce0372693a9cfbdef8b2689c9982cb", "tree": "cdb327c68844f105a8ac8b732e862f24edc35f24", "parents": [ "f374ada53bd1ca7c16d7607369fccc6769704956" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Feb 08 04:19:18 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:28 2008 -0800" }, "message": "uglify kill_pid_info() to fix kill() vs exec() race\n\nkill_pid_info()-\u003epid_task() could be the old leader of the execing process.\nIn that case it is possible that the leader will be released before we take\nsiglock. This means that kill_pid_info() (and thus sys_kill()) can return a\nfalse -ESRCH.\n\nChange the code to retry when lock_task_sighand() fails. The endless loop is\nnot possible, __exit_signal() both clears -\u003esighand and does detach_pid().\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Davide Libenzi \u003cdavidel@xmailserver.org\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\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": "5dee1707dfbfc55eb7569b9ae5abaf932bd4c377", "tree": "6e162f61f9645f790aa14cb4146882b918bb0286", "parents": [ "6806aac6d282d58b97763f5e17e2787e62c3b440" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Feb 08 04:19:13 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:27 2008 -0800" }, "message": "move the related code from exit_notify() to exit_signals()\n\nThe previous bugfix was not optimal, we shouldn\u0027t care about group stop\nwhen we are the only thread or the group stop is in progress. In that case\nnothing special is needed, just set PF_EXITING and return.\n\nAlso, take the related \"TIF_SIGPENDING re-targeting\" code from exit_notify().\n\nSo, from the performance POV the only difference is that we don\u0027t trust\n!signal_pending() until we take -\u003esiglock. But this in fact fixes another\n___pure___ theoretical minor race. __group_complete_signal() finds the\ntask without PF_EXITING and chooses it as the target for signal_wake_up().\nBut nothing prevents this task from exiting in between without noticing the\npending signal and thus unpredictably delaying the actual delivery.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Davide Libenzi \u003cdavidel@xmailserver.org\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\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": "d12619b5ff5664623524aef796514d1946ea3b4a", "tree": "23140ca0148f9154440d3056e2040d8df468a24b", "parents": [ "430c623121ea88ca80595c99fdc63b7f8a803ae5" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Feb 08 04:19:12 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:27 2008 -0800" }, "message": "fix group stop with exit race\n\ndo_signal_stop() counts all sub-thread and sets -\u003egroup_stop_count\naccordingly. Every thread should decrement -\u003egroup_stop_count and stop,\nthe last one should notify the parent.\n\nHowever a sub-thread can exit before it notices the signal_pending(), or it\nmay be somewhere in do_exit() already. In that case the group stop never\nfinishes properly.\n\nNote: this is a minimal fix, we can add some optimizations later. Say we\ncan return quickly if thread_group_empty(). Also, we can move some signal\nrelated code from exit_notify() to exit_signals().\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-by: Davide Libenzi \u003cdavidel@xmailserver.org\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\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": "20686a309aa98c518adbbd9b57cdbb1804143deb", "tree": "4aac811e67d777d3c0015ca51312aa93b4832d35", "parents": [ "9cbab8100538efdd93aeae6fc37787d986f2f558" ], "author": { "name": "Oleg Nesterov", "email": "oleg@tv-sign.ru", "time": "Fri Feb 08 04:19:03 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:26 2008 -0800" }, "message": "ptrace_stop: fix racy nonstop_code setting\n\nIf the tracer is gone and we are not going to stop, ptrace_stop() sets\n-\u003eexit_code \u003d nostop_code. However, the tracer could actually clear the\nexit code before detaching. In that case get_signal_to_deliver() \"resends\"\nthe signal which was cancelled by the debugger. For example, it is\npossible that a quick PTRACE_ATTACH + PTRACE_DETACH can leave the tracee in\nSTOPPED state.\n\nChange the behaviour of ptrace_stop(). If the caller is ptrace notify(),\nwe should always clear -\u003eexit_code. If the caller is\nget_signal_to_deliver(), we should not touch it at all. To do so, change\nthe nonstop_code parameter to \"bool clear_code\" and change the callers\naccordingly.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\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" } ], "next": "6405f7f4675884b671bee66678e1c2859bdb0e56" }