)]}' { "log": [ { "commit": "15d5f8398311f565682959daaca30e3ca7aea600", "tree": "ba38e77782832191dd38e182e99525870f83c085", "parents": [ "3b171672831b9633c2ed8fa94805255cd4d5af19" ], "author": { "name": "Dmitriy Zavin", "email": "dmitriyz@google.com", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "message": "[PATCH] x86: Refactor thermal throttle processing\n\nRefactor the event processing (syslog messaging and rate limiting)\ninto separate file therm_throt.c. This allows consistent reporting\nof CPU thermal throttle events.\n\nAfter ACK\u0027ing the interrupt, if the event is current, the user\n(p4.c/mce_intel.c) calls therm_throt_process to log (and rate limit)\nthe event. If that function returns 1, the user has the option to log\nthings further (such as to mce_log in x86_64).\n\nAK: minor cleanup\n\nSigned-off-by: Dmitriy Zavin \u003cdmitriyz@google.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3b171672831b9633c2ed8fa94805255cd4d5af19", "tree": "42b64b8f45166e58347f32b4e0d381c888981542", "parents": [ "b89ebd0b0a65d5371aa9ad98e873c4616056ca68" ], "author": { "name": "Dmitriy Zavin", "email": "dmitriyz@google.com", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "message": "[PATCH] Add 64bit jiffies compares (for use with get_jiffies_64)\n\nThe current time_before/time_after macros will fail typechecks\nwhen passed u64 values (as returned by get_jiffies_64()). On 64bit\nsystems, this will just result in a warning about mismatching types\nwithout explicit casts, but since unsigned long and u64\n(unsigned long long) are of same size, it will still work.\nOn 32bit systems, a long is 32bits, so the value from get_jiffies_64()\nwill be truncated by the cast and thus lose all the precision gained by\n64bit jiffies.\n\nSigned-off-by: Dmitriy Zavin \u003cdmitriyz@google.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b89ebd0b0a65d5371aa9ad98e873c4616056ca68", "tree": "7205fc3c9174f4f48d3fd59b86d99dfd66d5dfd3", "parents": [ "0637a70a5db98182d9ad3d6ae1ee30acf20afde9" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:42 2006 +0200" }, "message": "[PATCH] Fix unwinder warning in traps.c\n\nFix\n\nlinux/arch/x86_64/kernel/traps.c: In function \u0027dump_trace\u0027:\nlinux/arch/x86_64/kernel/traps.c:275: warning: cast to pointer from integer of different size\n\nwith allnoconfig\n\nCc: jbeulich@novell.com\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "0637a70a5db98182d9ad3d6ae1ee30acf20afde9", "tree": "36b625e24f3fe11a97cd9926ca2be6b2df1cbf89", "parents": [ "8f60774a116ced9b73ae3913d511687889efe725" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "message": "[PATCH] x86: Allow disabling early pci scans with pci\u003dnoearly or disallowing conf1\n\nSome buggy systems can machine check when config space accesses\nhappen for some non existent devices. i386/x86-64 do some early\ndevice scans that might trigger this. Allow pci\u003dnoearly to disable\nthis. Also when type 1 is disabling also don\u0027t do any early\naccesses which are always type1.\n\nThis moves the pci\u003d configuration parsing to be a early parameter.\nI don\u0027t think this can break anything because it only changes\na single global that is only used by PCI.\n\nCc: gregkh@suse.de\nCc: Trammell Hudson \u003chudson@osresearch.net\u003e\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "8f60774a116ced9b73ae3913d511687889efe725", "tree": "4c719703d138998a5eed9aabb2fe216b79790687", "parents": [ "f157cbb1eb9ce3f33a401ec6d20eb3eb852351a3" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "message": "[PATCH] x86: Move direct PCI scanning functions out of line\n\nSaves about 200 bytes of code space.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "658fdbef66e5e9be79b457edc2cbbb3add840aa9", "tree": "931fa9bac154671ab96762fee73b3641b1191bb8", "parents": [ "adf1423698f00d00b267f7dca8231340ce7d65ef" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "message": "[PATCH] Don\u0027t leak NT bit into next task\n\nSYSENTER can cause a NT to be set which might cause crashes on the IRET\nin the next task.\n\nFollowing similar i386 patch from Linus.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "adf1423698f00d00b267f7dca8231340ce7d65ef", "tree": "ec2e28a32d467cc3c618c8ee0cedd3d1a49f9c43", "parents": [ "ab2e0b46cb9a197fab7d98e147cac7cd41a14047" ], "author": { "name": "Jan Beulich", "email": "jbeulich@novell.com", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:41 2006 +0200" }, "message": "[PATCH] i386/x86-64: Work around gcc bug with noreturn functions in unwinder\n\nCurrent gcc generates calls not jumps to noreturn functions. When that happens the\nreturn address can point to the next function, which confuses the unwinder.\n\nThis patch works around it by marking asynchronous exception\nframes in contrast normal call frames in the unwind information. Then teach\nthe unwinder to decode this.\n\nFor normal call frames the unwinder now subtracts one from the address which avoids\nthis problem. The standard libgcc unwinder uses the same trick.\n\nIt doesn\u0027t include adjustment of the printed address (i.e. for the original\nexample, it\u0027d still be kernel_math_error+0 that gets displayed, but the\nunwinder wouldn\u0027t get confused anymore.\n\nThis only works with binutils 2.6.17+ and some versions of H.J.Lu\u0027s 2.6.16\nunfortunately because earlier binutils don\u0027t support .cfi_signal_frame\n\n[AK: added automatic detection of the new binutils and wrote description]\n\nSigned-off-by: Jan Beulich \u003cjbeulich@novell.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "2817716ace8c397685bd702223cc5a8a42c7e997", "tree": "901fecc20019b1a8dd294bd70a155b8bd4cd6814", "parents": [ "de09bddb9d6f96785be470c832b881e6d72d589f" ], "author": { "name": "Jeremy Fitzhardinge", "email": "jeremy@goop.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] i386: Fix pack_descriptor()\n\nFix pack_descriptor:\n 1. flags are bits 20-23 in the high word\n 2. limit\u0027s 4 msb are bits 16-19 in the high word\n\nThese haven\u0027t mattered so far, because all users have had small limits\nand a flags setting of 0.\n\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@xensource.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n" }, { "commit": "a15da49debaf7f09460a886b0ecd08588410715e", "tree": "685ed022695ff03cec167c6690815e08191a803c", "parents": [ "1c9c0a6ca35e9325cea811d734d6ab7352be086b" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] Fix idle notifiers\n\nPreviously exit_idle would be called more often than enter_idle\n\nNow instead of using complicated tests just keep track of it\nusing the per CPU variable as a flip flop. I moved the idle state into the\nPDA to make the access more efficient.\n\nOriginal bug report and an initial patch from Stephane Eranian,\nbut redone by AK.\n\nCc: Stephane Eranian \u003ceranian@hpl.hp.com\u003e\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "73bb5117a448bdf0b56232ca28451fe4c534cb3a", "tree": "687012771cd1f12bc5c2ce9c798b6933ca9b3574", "parents": [ "fd167e42b237e0688005b3dec380eb5a6e5f3585" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] Remove unused asm-x86_64/mmx.h\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "fd167e42b237e0688005b3dec380eb5a6e5f3585", "tree": "628687df15662dd47781eebd6b227fa823e9924d", "parents": [ "c1a9d41f4f103bfef2ed0bea1e95b3190e39e448" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] Define __bad_pda_field as noreturn\n\nThis quietens so warnings about uninitialized use of the return\nvalue of the pda read operations.\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "c1a9d41f4f103bfef2ed0bea1e95b3190e39e448", "tree": "1b11b8fa27c5b4308a5014453b944ec15c22b1a8", "parents": [ "383d079bfdfcfccd6a720405a510fe325b3e6576" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] Reindent macros in pda.h\n\nReindent the macros in x86-64 pda.h, making them much more readable.\nFollows Jeremy\u0027s i386 version of this.\n\nNo functional changes\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "383d079bfdfcfccd6a720405a510fe325b3e6576", "tree": "0ca82b2f60b1cf47c027601b4a3771909462aebb", "parents": [ "26c13f2b5bbb03f798f8907db20296347e6c7ca6" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:40 2006 +0200" }, "message": "[PATCH] Fix some stylistic issues in uaccess.h\n\n- Replace some broken white space.\n- Replace __ keywords with standard names\n\nNo functional changes.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "95912008ba1fb9d0677c1ce5930aeb0e85ba5710", "tree": "e84c614f6aff984d105ce9f340f9474b613d567d", "parents": [ "3022d734a54cbd2b65eea9a024564821101b4a9a" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] Add __must_check to copy_*_user\n\nFollowing i386.\n\nAnd also fix the two occurrences that caused warnings in arch/x86_64/*\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3022d734a54cbd2b65eea9a024564821101b4a9a", "tree": "a1445aeaf8fbdf84898c8cdc13dd98c550b5be0f", "parents": [ "f0f4c3432e5e1087b3a8c0e6bd4113d3c37497ff" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] Fix zeroing on exception in copy_*_user\n\n- Don\u0027t zero for __copy_from_user_inatomic following i386.\nThis will prevent spurious zeros for parallel file system writers when\none does a exception\n- The string instruction version didn\u0027t zero the output on\nexception. Oops.\n\nAlso I cleaned up the code a bit while I was at it and added a minor\noptimization to the string instruction path.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "78be3706b21a232310590fe00258b224177ac05f", "tree": "57dbe444e57241613067145b04c0e15c98278cd6", "parents": [ "0da5db313317e3195482d3e660a1074857374a89" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] i386: Allow a kernel not to be in ring 0\n\nWe allow for the fact that the guest kernel may not run in ring 0. This\nrequires some abstraction in a few places when setting %cs or checking\nprivilege level (user vs kernel).\n\nThis is Chris\u0027 [RFC PATCH 15/33] move segment checks to subarch, except rather\nthan using #define USER_MODE_MASK which depends on a config option, we use\nZach\u0027s more flexible approach of assuming ring 3 \u003d\u003d userspace. I also used\n\"get_kernel_rpl()\" over \"get_kernel_cs()\" because I think it reads better in\nthe code...\n\n1) Remove the hardcoded 3 and introduce #define SEGMENT_RPL_MASK 3 2) Add a\nget_kernel_rpl() macro, and don\u0027t assume it\u0027s zero.\n\nAnd:\n\nClean up of patch for letting kernel run other than ring 0:\n\na. Add some comments about the SEGMENT_IS_*_CODE() macros.\nb. Add a USER_RPL macro. (Code was comparing a value to a mask\n in some places and to the magic number 3 in other places.)\nc. Add macros for table indicator field and use them.\nd. Change the entry.S tests for LDT stack segment to use the macros\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Zachary Amsden \u003czach@vmware.com\u003e\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@xensource.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "0da5db313317e3195482d3e660a1074857374a89", "tree": "9322a8eb60da97ae5f3a199cec13afd9b34c202d", "parents": [ "7b0bda74f7e77f362eaeee837e7911238acf4c76" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] i386: Abstract sensitive instructions\n\nAbstract sensitive instructions in assembler code, replacing them with macros\n(which currently are #defined to the native versions). We use long names:\nassembler is case-insensitive, so if something goes wrong and macros do not\nexpand, it would assemble anyway.\n\nResulting object files are exactly the same as before.\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@xensource.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "85691f135db78f3548107a0abe383dfab3bc38fa", "tree": "5bde2c801824b4c3d7e7d3def97b6a89ab178a1e", "parents": [ "baf5695dd1a49bb48a3daf08726d7f243f42e97e" ], "author": { "name": "Jeremy Fitzhardinge", "email": "jeremy@goop.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] Type checking for write_pda()\n\nI just added type checking for assignments the PDA in the i386 PDA code.\nHere\u0027s the x86-64 equivalent. (Obviously this doesn\u0027t contain the latest\nx86-64 PDA change.)\n\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "baf5695dd1a49bb48a3daf08726d7f243f42e97e", "tree": "7479fb8562395e168654d931db0ad89ba8ccd39c", "parents": [ "96e540492ab54423f3693958329e095878f1f12b" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:39 2006 +0200" }, "message": "[PATCH] Use %c instead of %P modifier in pda access\n\nApparently that is the more official way to get numbers without $ in inline\nassembly\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "0a4254058037eb172758961d0a5b94f4320a1425", "tree": "e9e64daf007952322c7ab0671b2f2f841085c04d", "parents": [ "b62a5c740df1e3d49a97349fce0c6a23f633d7fe" ], "author": { "name": "Arjan van de Ven", "email": "arjan@linux.intel.com", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] Add the canary field to the PDA area and the task struct\n\nThis patch adds the per thread cookie field to the task struct and the PDA.\nAlso it makes sure that the PDA value gets the new cookie value at context\nswitch, and that a new task gets a new cookie at task creation time.\n\nSigned-off-by: Arjan van Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nCC: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "29a9af60e2120f874d0c600bf9e27617254a0488", "tree": "90b4309599fa1081d58d0312c94e899eaf18f115", "parents": [ "027a51cef330433ba5924fc92fb25ef48bcfc1a3" ], "author": { "name": "Arjan van de Ven", "email": "arjan@linux.intel.com", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] Add comments to the PDA structure to annotate offsets\n\nChange the comments in the pda structure to make the first fields to have\ntheir offset documented and to have the comments aligned.\nThe stack protector series needs a field at offset 40 (gcc ABI); annotate\nupto 40 for that reason.\n\nSigned-off-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nCC: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3566561bfadffcb5dbc85d576be80c0dbf2cccc9", "tree": "eaafaf68156f9f2f1a2af8a7ffed95653fea8aa5", "parents": [ "4bfaaef01a1badb9e8ffb0c0a37cd2379008d21f" ], "author": { "name": "Magnus Damm", "email": "magnus@valinux.co.jp", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] i386: Avoid overwriting the current pgd (V4, i386)\n\nkexec: Avoid overwriting the current pgd (V4, i386)\n\nThis patch upgrades the i386-specific kexec code to avoid overwriting the\ncurrent pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used\nto start a secondary kernel that dumps the memory of the previous kernel.\n\nThe code introduces a new set of page tables. These tables are used to provide\nan executable identity mapping without overwriting the current pgd.\n\nSigned-off-by: Magnus Damm \u003cmagnus@valinux.co.jp\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "4bfaaef01a1badb9e8ffb0c0a37cd2379008d21f", "tree": "652effe9d02c80d93357512233f4ee962320ac3b", "parents": [ "f574164491d00d28b727d713685fb5edc9138200" ], "author": { "name": "Magnus Damm", "email": "magnus@valinux.co.jp", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] Avoid overwriting the current pgd (V4, x86_64)\n\nkexec: Avoid overwriting the current pgd (V4, x86_64)\n\nThis patch upgrades the x86_64-specific kexec code to avoid overwriting the\ncurrent pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used\nto start a secondary kernel that dumps the memory of the previous kernel.\n\nThe code introduces a new set of page tables. These tables are used to provide\nan executable identity mapping without overwriting the current pgd.\n\nSigned-off-by: Magnus Damm \u003cmagnus@valinux.co.jp\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "f574164491d00d28b727d713685fb5edc9138200", "tree": "7ab1a5a2257e27824cf3762fcf276840259deceb", "parents": [ "575400d1b483fbe9e03c68758059bfaf4e4768d1" ], "author": { "name": "Keith Owens", "email": "kaos@ocs.com.au", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] Remove most of the special cases for the debug IST stack\n\nRemove most of the special cases for the debug IST stack. This is a\nfollow on clean up patch, it requires the bug fix patch that adds\norig_ist.\n\nSigned-off-by: Keith Owens \u003ckaos@ocs.com.au\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "575400d1b483fbe9e03c68758059bfaf4e4768d1", "tree": "9839d343d235dabfdb27e8ad2e48511ae542b015", "parents": [ "53ee11ae0d73f28029a5f0d991bc4dcd7c817e7a" ], "author": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] i386: Fix the EDD code misparsing the command line\n\nThe EDD code would scan the command line as a fixed array, without\ntaking account of either whitespace, null-termination, the old\ncommand-line protocol, late overrides early, or the fact that the\ncommand line may not be reachable from INITSEG.\n\nThis should fix those problems, and enable us to use a longer command\nline.\n\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "53ee11ae0d73f28029a5f0d991bc4dcd7c817e7a", "tree": "c0a025cc592d516674bb1d8a0046bf281b6d1724", "parents": [ "80d2679cbc8e170011c9649fb8fb684ffd7e5c8f" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] Optimize PDA accesses slightly\n\nBased on a idea by Jeremy Fitzhardinge:\n\nReplace the volatiles and memory clobbers in the PDA access with\ntelling gcc about access to a proxy PDA structure that doesn\u0027t\nactually exist. But the dummy accesses give a defined ordering for\nread/write accesses.\n\nAlso add some memory barriers to the early GS initialization to\nmake sure no PDA access is moved before it.\n\nAdvantage is some .text savings (probably most from better\ncode for accessing \"current\"):\n\n text data bss dec hex filename\n4845647 1223688 615864 6685199 66020f vmlinux\n4837780 1223688 615864 6677332 65e354 vmlinux-pda\n\n1.2% smaller code\n\nCc: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "80d2679cbc8e170011c9649fb8fb684ffd7e5c8f", "tree": "7e0f6f42aacbfd4c2bcb26178d9f81b03c39a7d4", "parents": [ "f2a9e1dec27189a09ff642f2648e49ad9e76607b" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] x86: Remove incorrect comment about ACPI e820 entries\n\nThey cannot be actually freed because the FACS table has a\nshared-with-the-BIOS lock.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "0136611c62e8650e354b95c76dff6d2ce6030eff", "tree": "9ba66105bccc4d83b84663b8dda7e51962c22a04", "parents": [ "8380aabb99719af583447133f19a4d8074b5c337" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:38 2006 +0200" }, "message": "[PATCH] optimize hweight64 for x86_64\n\nBased on patch from David Rientjes \u003crientjes@google.com\u003e, but\nchanged by AK.\n\nOptimizes the 64-bit hamming weight for x86_64 processors assuming they\nhave fast multiplication. Uses five fewer bitops than the generic\nhweight64. Benchmark on one EMT64 showed ~25% speedup with 2^24\nconsecutive calls.\n\nDefine a new ARCH_HAS_FAST_MULTIPLIER that can be set by other\narchitectures that can also multiply fast.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "8380aabb99719af583447133f19a4d8074b5c337", "tree": "e26bcd8130199db5c7b978532534a006503b9a95", "parents": [ "b3698c03eb6d4581e879d6bb0f183ed8dda96d37" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Remove non e820 fallbacks in high level code\n\nDrop support for non e820 BIOS calls to get the memory map.\n\nThe boot assembler code still has some support, but not the C code now.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "e4251e130deef9de5226cc36faa70a1c6671d3c5", "tree": "22d75691bbcd93b88624ceaa19d8d4fd5ec5900a", "parents": [ "f2c2cca3acef8b253a36381d9b469ad4fb08563a" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Remove some cruft in apic id checking during processor setup\n\n- Remove a define that was used only once\n- Remove the too large APIC ID check because we always support\nthe full 8bit range of APICs.\n- Restructure code a bit to be simpler.\n\nCc: len.brown@intel.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "5e6b0bfe5b452957b7be4b6ef181cd41880f8359", "tree": "58fe6fa071e8a6870f4241217b18e181481c5c1e", "parents": [ "df992848f5aa803fcacd2c5e7d67034bb89e3fa3" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Use proper accessors to change PSE bits in change_page_attr()\n\nUse normal pte accessors in change_page_attr() to access the PSE\nbits.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "df992848f5aa803fcacd2c5e7d67034bb89e3fa3", "tree": "649ee5888f71bdd4243599f73d9725118262d368", "parents": [ "d3cf7f061521c78ad62e275eb6fbdc8f43fc75a7" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Fix pte_exec/mkexec and use it in change_page_attr()\n\nFix the pte_exec/mkexec page table accessor functions to really\nuse the NX bit. Previously they only checked the USER bit, but\nweren\u0027t actually used for anything.\n\nThen use them in change_page_attr() to manipulate the NX bit\nproperly.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "151f8cc1169f9052095b2be36183ab132d75c6c2", "tree": "f3aae7a7e2fc15e88a8471256b2aa43c4850ad0f", "parents": [ "34464a5b8937b79801776dfb6970c1b949fed4be" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Remove safe_smp_processor_id()\n\nAnd replace all users with ordinary smp_processor_id. The function\nwas originally added to get some basic oops information out even\nif the GS register was corrupted. However that didn\u0027t\nwork for some anymore because printk is needed to print the oops\nand it uses smp_processor_id() already. Also GS register corruptions\nare not particularly common anymore.\n\nThis also helps the Xen port which would otherwise need to\ndo this in a special way because it can\u0027t access the local APIC.\n\nCc: Chris Wright \u003cchrisw@sous-sol.org\u003e\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "a549b86dd0f3cbffcd5f9343f4ae7fcd59f7e756", "tree": "4405a0465a631a45945954f1ce1efa7d8fd24676", "parents": [ "1164c9994fe37d5b7035a5cf9328c98dd38af7b1" ], "author": { "name": "Chuck Ebbert", "email": "76306.1226@compuserve.com", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] i386: annotate FIX_STACK() and the rest of nmi()\n\nIn i386\u0027s entry.S, FIX_STACK() needs annotation because it\nreplaces the stack pointer. And the rest of nmi() needs\nannotation in order to compile with these new annotations.\n\nSigned-off-by: Chuck Ebbert \u003c76306.1226@compuserve.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "1bb4996bcebca1cde49d964b4e012699ce180e61", "tree": "ba761ac65661370b0ed5ce6b7dacd263afcef3c8", "parents": [ "dbf9272e863bf4b17ee8e3c66c26682b2061d40d" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:37 2006 +0200" }, "message": "[PATCH] Move compiler check for modules to ia64 only\n\nApparently IA64 needs it, but i386/x86-64 don\u0027t anymore\nsince gcc 2.95 support was dropped. Nobody else on linux-arch\nrequested keeping it generically\n\nCc: tony.luck@intel.com\nCc: kaos@sgi.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "f704cb935006580db0495e54d3c82631f6e2a984", "tree": "54c0dac91f938d2ee848a0ab05e7484790759b2f", "parents": [ "06039754d775d3e48e4a292e4f353321205eff53" ], "author": { "name": "Dave Jones", "email": "davej@redhat.com", "time": "Tue Sep 26 10:52:36 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:36 2006 +0200" }, "message": "[PATCH] x86: remove config.h includes from asm-i386 \u0026 asm-x86_64\n\nThis is now automatically included by kbuild.\n\nSigned-off-by: Dave Jones \u003cdavej@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "e07e23e1fd3000289fc7ccc6c71879070d3b19e0", "tree": "1290385cacd89e39b7bc1b12b7515b68423d78d3", "parents": [ "73fea175303926055440c06bc8894f0c5c58afc8" ], "author": { "name": "Arjan van de Ven", "email": "arjan@linux.intel.com", "time": "Tue Sep 26 10:52:36 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:36 2006 +0200" }, "message": "[PATCH] non lazy \"sleazy\" fpu implementation\n\nRight now the kernel on x86-64 has a 100% lazy fpu behavior: after *every*\ncontext switch a trap is taken for the first FPU use to restore the FPU\ncontext lazily. This is of course great for applications that have very\nsporadic or no FPU use (since then you avoid doing the expensive\nsave/restore all the time). However for very frequent FPU users... you\ntake an extra trap every context switch.\n\nThe patch below adds a simple heuristic to this code: After 5 consecutive\ncontext switches of FPU use, the lazy behavior is disabled and the context\ngets restored every context switch. If the app indeed uses the FPU, the\ntrap is avoided. (the chance of the 6th time slice using FPU after the\nprevious 5 having done so are quite high obviously).\n\nAfter 256 switches, this is reset and lazy behavior is returned (until\nthere are 5 consecutive ones again). The reason for this is to give apps\nthat do longer bursts of FPU use still the lazy behavior back after some\ntime.\n\n[akpm@osdl.org: place new task_struct field next to jit_keyring to save space]\nSigned-off-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\n" }, { "commit": "73fea175303926055440c06bc8894f0c5c58afc8", "tree": "466af16058837ab4f8a83210848ecddac20705d9", "parents": [ "40bee2ee73c745922e9b2d5595c46f19d1cf1b6f" ], "author": { "name": "Ashok Raj", "email": "ashok.raj@intel.com", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "message": "[PATCH] i386: Support physical cpu hotplug for x86_64\n\nThis patch enables ACPI based physical CPU hotplug support for x86_64.\nImplements acpi_map_lsapic() and acpi_unmap_lsapic() to support physical cpu\nhotplug.\n\nSigned-off-by: Ashok Raj \u003cashok.raj@intel.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nCc: \"Brown, Len\" \u003clen.brown@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\n" }, { "commit": "ba4d40bb5c465f0a4dcc30d02dab80c2cb7e1ff3", "tree": "12b10f26f5efddb15ef8a75a1eff38771334365d", "parents": [ "522e93e3fcdbf00ba85c72fde6df28cfc0486a65" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "message": "[PATCH] Auto size the per cpu area.\n\nNow for a completely different but trivial approach.\nI just boot tested it with 255 CPUS and everything worked.\n\nCurrently everything (except module data) we place in\nthe per cpu area we know about at compile time. So\ninstead of allocating a fixed size for the per_cpu area\nallocate the number of bytes we need plus a fixed constant\nfor to be used for modules.\n\nIt isn\u0027t perfect but it is much less of a pain to\nwork with than what we are doing now.\n\nAK: fixed warning\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "522e93e3fcdbf00ba85c72fde6df28cfc0486a65", "tree": "c6d1d0fd3109a667ca4ee0c0f8dc8a2fe4767240", "parents": [ "02ba1a32dbd3d406530a17a2643a8f0f8cbf3acc" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "message": "[PATCH] i386: Descriptor and trap table cleanups.\n\nThe implementation comes from Zach\u0027s [RFC, PATCH 10/24] i386 Vmi\ndescriptor changes:\n\nDescriptor and trap table cleanups. Add cleanly written accessors for\nIDT and GDT gates so the subarch may override them. Note that this\nallows the hypervisor to transparently tweak the DPL of the descriptors\nas well as the RPL of segments in those descriptors, with no unnecessary\nkernel code modification. It also allows the hypervisor implementation\nof the VMI to tweak the gates, allowing for custom exception frames or\nextra layers of indirection above the guest fault / IRQ handlers.\n\nSigned-off-by: Zachary Amsden \u003czach@vmware.com\u003e\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3d08a256da8aed5300bd0752200ece426f49b050", "tree": "26c7d5362b440fc0b855ff9b6ccace795125b309", "parents": [ "e8924acb2ef46b96c93f97025815ef3843cb67a2" ], "author": { "name": "Adrian Bunk", "email": "bunk@stusta.de", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:35 2006 +0200" }, "message": "[PATCH] i386: Make enable_local_apic static\n\nenable_local_apic can now become static.\n\nCc: len.brown@intel.com\n\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "d28c4393a7bf558538e9def269c1caeab6ec056f", "tree": "e5319c9b9c8a75d2290f7429ec84884d7e1a91fe", "parents": [ "3ca113ea74836a80645c79adba24caaa7a74120c" ], "author": { "name": "Prasanna S.P", "email": "prasanna@in.ibm.com", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] x86: error_code is not safe for kprobes\n\nThis patch moves the entry.S:error_entry to .kprobes.text section,\nsince code marked unsafe for kprobes jumps directly to entry.S::error_entry,\nthat must be marked unsafe as well.\nThis patch also moves all the \".previous.text\" asm directives to \".previous\"\nfor kprobes section.\n\nAK: Following a similar i386 patch from Chuck Ebbert\nAK: Also merged Jeremy\u0027s fix in.\n\n+From: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\n\nKPROBE_ENTRY does a .section .kprobes.text, and expects its users to\ndo a .previous at the end of the function.\n\nUnfortunately, if any code within the function switches sections, for\nexample .fixup, then the .previous ends up putting all subsequent code\ninto .fixup. Worse, any subsequent .fixup code gets intermingled with\nthe code its supposed to be fixing (which is also in .fixup). It\u0027s\nsurprising this didn\u0027t cause more havok.\n\nThe fix is to use .pushsection/.popsection, so this stuff nests\nproperly. A further cleanup would be to get rid of all\n.section/.previous pairs, since they\u0027re inherently fragile.\n\n+From: Chuck Ebbert \u003c76306.1226@compuserve.com\u003e\n\nBecause code marked unsafe for kprobes jumps directly to\nentry.S::error_code, that must be marked unsafe as well.\nThe easiest way to do that is to move the page fault entry\npoint to just before error_code and let it inherit the same\nsection.\n\nAlso moved all the \".previous\" asm directives for kprobes\nsections to column 1 and removed \".text\" from them.\n\nSigned-off-by: Chuck Ebbert \u003c76306.1226@compuserve.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "a32cf3975bed3b84491f8ffeb24abe8c45d86ab0", "tree": "f327b09bc2b17e66645f1b7b97c3b7ac11133e1b", "parents": [ "950fee84557416a3427dd404a13addc4be7b3e6c" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] i386: Get ebp from unwinder state when continuing fallback backtrace\n\nCc: jbeulich@novell.com\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "2b14a78cd07a52001b8c3865ed615d8b9b905b78", "tree": "415682b4b8a65322ed881fce5ae04fcb36f55930", "parents": [ "be7a91709b90825990e571b2f20cea937d5eef6c" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] i386: Do stacktracer conversion too\n\nFollowing x86-64 patches. Reuses code from them in fact.\n\nConvert the standard backtracer to do all output using\ncallbacks. Use the x86-64 stack tracer implementation\nthat uses these callbacks to implement the stacktrace interface.\n\nThis allows to use the new dwarf2 unwinder for stacktrace\nand get better backtraces.\n\nCc: mingo@elte.hu\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "c0b766f13d8e1189ce4d00e54700c9d96b543b9a", "tree": "7acb6b7a46b3b0b53328f7d72949f75a57509064", "parents": [ "c9538ed49272fb244ac06ba643ff076a68a77e12" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] Merge stacktrace and show_trace\n\nThis unifies the standard backtracer and the new stacktrace\nin memory backtracer. The standard one is converted to use callbacks\nand then reimplement stacktrace using new callbacks.\n\nThe main advantage is that stacktrace can now use the new dwarf2 unwinder\nand avoid false positives in many cases.\n\nI kept it simple to make sure the standard backtracer stays reliable.\n\nCc: mingo@elte.hu\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b7f5e3c7742d5332b78b831131f43fc3630e6322", "tree": "4334c97a3a8e2ac409444ecb2d7844d3a63e4118", "parents": [ "3fa7c794fe4dc127f7fac3fad4d13628e68f89ce" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] Don\u0027t access the APIC in safe_smp_processor_id when it is not mapped yet\n\nLockdep can call the dwarf2 unwinder early, and the dwarf2 code\nuses safe_smp_processor_id which tries to access the local APIC page.\nBut that doesn\u0027t work before the APIC code has set up its fixmap.\n\nCheck for this case and always return boot cpu then.\n\nCc: jbeulich@novell.com\nCc: mingo@elte.hu\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "5a1b3999d6cb7ab87f1f3b1700bc91839fd6fa29", "tree": "ec969406fe68d875d7904c364ea9a5b74c084567", "parents": [ "4ea8a5d8b57cd504b4b2de1212523848e7ab50cf" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:34 2006 +0200" }, "message": "[PATCH] x86: Some preparationary cleanup for stack trace\n\n- Remove unused all_contexts parameter\nNo caller used it\n- Move skip argument into the structure (needed for\nfollowon patches)\n\nCc: mingo@elte.hu\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "91cd444e56ebe0c2acd9576a045d77490b26f607", "tree": "647c37392e41541efb67d8656703f36cf15b5e35", "parents": [ "2b94ab2fd55768030e177a6ec224dedd031ad06b" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:33 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:33 2006 +0200" }, "message": "[PATCH] x86: Remove unneeded externs in acpi/boot.c\n\nAnd move one into proto.h\n\nCc: len.brown@intel.com\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "2c8c0e6b8d7700a990da8d24eff767f9ca223b96", "tree": "b5477ee2558efff235c483d1619ce892db479d18", "parents": [ "1a3f239ddf9208f2e52d36fef1c1c4518cbbbabe" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] Convert x86-64 to early param\n\nInstead of hackish manual parsing\n\nRequires earlier i386 patchkit, but also fixes i386 early_printk again.\n\nI removed some obsolete really early parameters which didn\u0027t do anything useful.\nAlso made a few parameters that needed it early (mostly oops printing setup)\n\nAlso removed one panic check that wasn\u0027t visible without\nearly console anyways (the early console is now initialized after that\npanic)\n\nThis cleans up a lot of code.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "1a3f239ddf9208f2e52d36fef1c1c4518cbbbabe", "tree": "f2ad21f766102540e15ea61145e94be65901e272", "parents": [ "33df0d19ea425d28bd5afb48898af32237fe81af" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] i386: Replace i386 open-coded cmdline parsing with\n\nThis patch replaces the open-coded early commandline parsing\nthroughout the i386 boot code with the generic mechanism (already used\nby ppc, powerpc, ia64 and s390). The code was inconsistent with\nwhether it deletes the option from the cmdline or not, meaning some of\nthese will get passed through the environment into init.\n\nThis transformation is mainly mechanical, but there are some notable\nparts:\n\n1) Grammar: s/linux never set\u0027s it up/linux never sets it up/\n\n2) Remove hacked-in earlyprintk\u003d option scanning. When someone\n actually implements CONFIG_EARLY_PRINTK, then they can use\n early_param().\n[AK: actually it is implemented, but I\u0027m adding the early_param it in the next\nx86-64 patch]\n\n3) Move declaration of generic_apic_probe() from setup.c into asm/apic.h\n\n4) Various parameters now moved into their appropriate files (thanks Andi).\n\n5) All parse functions which examine arg need to check for NULL,\n except one where it has subtle humor value.\n\nAK: readded acpi_sci handling which was completely dropped\nAK: moved some more variables into acpi/boot.c\n\nCc: len.brown@intel.com\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "fb2e28485679418e459583605f9b19807a72ceca", "tree": "c3e6bf7f75fb9c6ed286ef09eebf281388761cfe", "parents": [ "8b059d2373c16b6d32787a49daf8ccf72dc61b71" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] i386: Clean up spin/rwlocks\n\n- Inline spinlock strings into their inline functions\n- Convert macros to typesafe inlines\n- Replace some leftover __asm__ __volatile__s with asm volatile\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "8b059d2373c16b6d32787a49daf8ccf72dc61b71", "tree": "4814940e77c8956bd94d35edf765ff1698e4c57c", "parents": [ "a752d7194c4fb5a3e767c95542d04fc5decb1d52" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] Clean up spin/rwlocks\n\n- Inline spinlock strings into their inline functions\n- Convert macros to typesafe inlines\n- Replace some leftover __asm__ __volatile__s with asm volatile\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "8578bdf835d6d6ece6718858d351456367b8a5cf", "tree": "5bf4aa72bda1f5c508d955a2d32b5d292f8bd5a7", "parents": [ "7ca2b49b06a6d26e89e3535653889f1d7892b085" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] Don\u0027t use lock section for mutexes and semaphores\n\nLock sections cannot be handled by the dwarf2 unwinder.\n\nDisadvantage is a taken branch in the hot path.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "7ca2b49b06a6d26e89e3535653889f1d7892b085", "tree": "99e460b4da18b0d841267082219d9b390c13a649", "parents": [ "add659bf8aa92f8b3f01a8c0220557c959507fb1" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:32 2006 +0200" }, "message": "[PATCH] i386: Remove lock section support in semaphore.h\n\nLock sections don\u0027t work the new dwarf2 unwinder\nThis generates slightly smaller code. It adds one more taken\njump to the fast path.\n\nCc: jbeulich@novell.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "add659bf8aa92f8b3f01a8c0220557c959507fb1", "tree": "ef140ba27d91ee45ee0a01ab7d92de5138a681ca", "parents": [ "01215ad8d83e18321d99e9b5750a6f21cac243a2" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "message": "[PATCH] i386: Remove lock section support in rwsem.h\n\nLock sections don\u0027t work the new dwarf2 unwinder\nThis generates slightly smaller code. It adds one more taken\njump to the fast path.\n\nAlso move the trampolines into semaphore.S and add proper CFI\nannotations.\n\nCc: jbeulich@novell.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "01215ad8d83e18321d99e9b5750a6f21cac243a2", "tree": "aad34bac0001d2b2788dcb5c68566372db7968c3", "parents": [ "0577f148b5e9a773020e3da1e6332a7c6df9d601" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "message": "[PATCH] i386: Remove lock section support in mutex.h\n\nLock sections don\u0027t work the new dwarf2 unwinder\nThis generates slightly smaller code. It adds one more taken\njump to the fast path.\n\nCc: jbeulich@novell.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "f38db651d5da5e10235fd7dd31095969fb7ef6fb", "tree": "4df8d0921b5c4e392adad94eec8407e39e9ffae0", "parents": [ "5f4a7a93886ce1a4327f6028cc05d423f39eebf0" ], "author": { "name": "Muli Ben-Yehuda", "email": "muli@il.ibm.com", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "message": "[PATCH] Calgary IOMMU: consolidate per bus data structures\n\nMove the tce_table_kva array, disabled bitmap and bus_to_phb array\ninto a new per bus \u0027struct calgary_bus_info\u0027. Also slightly reorganize\nbuild_tce_table and tce_table_setparms to avoid exporting bus_info to\ntce.c.\n\nSigned-off-by: Muli Ben-Yehuda \u003cmuli@il.ibm.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@us.ibm.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "5f4a7a93886ce1a4327f6028cc05d423f39eebf0", "tree": "0c1ecfe880ce775fadce814f024412adc7196511", "parents": [ "352f7bae81faa2befa2a3c02b84478dce16b8fd6" ], "author": { "name": "Muli Ben-Yehuda", "email": "muli@il.ibm.com", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "message": "[PATCH] Calgary IOMMU: rearrange \u0027struct iommu_table\u0027 members\n\nRearrange struct members loosely based on size for improved alignment\nand to save a few bytes.\n\nSigned-off-by: Muli Ben-Yehuda \u003cmuli@il.ibm.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@us.ibm.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3b94355c47e2b025a7ececa0a14180e8fce6b0f1", "tree": "365eaa79b1ef8bc862d3a13cd9f14489949a2d3d", "parents": [ "caff0710ebf6f2c44cbd2b8b31fd6329148bed2e" ], "author": { "name": "Jan Beulich", "email": "jbeulich@novell.com", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:31 2006 +0200" }, "message": "[PATCH] remove int_delivery_dest\n\nThe genapic field and the accessor macro weren\u0027t used anywhere.\n\nSigned-off-by: Jan Beulich \u003cjbeulich@novell.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "e2414910f212c52d9d7c64c99a22863488ac5b48", "tree": "5a7e55191ace4aa755c0684bcefc9cf1b124128c", "parents": [ "fe7414a2882c953788af13a7f2c9d570ed8f71c5" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "message": "[PATCH] x86: Detect CFI support in the assembler at runtime\n\n... instead of using a CONFIG option. The config option still controls\nif the resulting executable actually has unwind information.\n\nThis is useful to prevent compilation errors when users select\nCONFIG_STACK_UNWIND on old binutils and also allows to use\nCFI in the future for non kernel debugging applications.\n\nCc: jbeulich@novell.com\nCc: sam@ravnborg.org\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "55f05ffaa788e039df2f1ebe0d7bfbcb6f39d0b4", "tree": "9b4da733f1afd01fea6622468b41561522e62cfa", "parents": [ "dfa4698c50bf85b7927214b0e4a3dc4bc3b3c4a9" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "message": "[PATCH] Replace mp bus array with bitmap for bus not pci\n\nSince we only support PCI and ISA legacy busses now there is no need to\nhave an full array with checking.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "dfa4698c50bf85b7927214b0e4a3dc4bc3b3c4a9", "tree": "5273f10e463cef99b1a3fd4d369a8a6f2f4c6dce", "parents": [ "edd965229669f8f8dfddec8c863250440fb65ab3" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "message": "[PATCH] Move early chipset quirks out to new file\n\nThey did not really belong into io_apic.c. Move them into a new file\nand clean it up a bit.\n\nAlso remove outdated ATI quirk that was obsolete,\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "a8fcf1a24a16e1c735c795d99773b5dcefb71518", "tree": "c1edcfe3b21dcee6d5f25a4461607a3dd785419c", "parents": [ "e50991343488edf25d58820be1684322808c763c" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:30 2006 +0200" }, "message": "[PATCH] Remove obsolete PIC mode\n\nPIC mode is an outdated way to drive the APICs that was used on\nsome early MP boards. It is not supported in the ACPI model.\n\nIt is unlikely to be ever configured by any x86-64 system\n\nRemove it thus.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "107878bb14b1fb6bddc646f4d5e72e8beaa2f6a2", "tree": "9584b064a1d2b2fc4e0752404e5eb0a6c893b56b", "parents": [ "b1c78c0fcc29097567e1afc39701012e6d89adb7" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] i386: Minor fixes \u0026 cleanup to tlb flush\n\n(based on x86-64 changes)\n- Add a proper memory clobber to invlpg\n- Remove an unused extern\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b1c78c0fcc29097567e1afc39701012e6d89adb7", "tree": "9b9a2a302740c7a68003ade0536ab244f20cb08c", "parents": [ "3f14c746a61ec932c204aca820c02c293118c5df" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] Clean up and minor fixes to TLB flush\n\n- Convert CR* accesses to dedicated inline functions and rewrite\nthe rest as C inlines\n- Don\u0027t do a double flush for global flushes (pointed out by Zach Amsden)\nThis was a bug workaround for old CPUs that don\u0027t do 64bit and is obsolete.\n- Add a proper memory clobber to invlpg\n- Remove an unused extern\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "7f11d8a5efd625ffa41cde1d8472f85e885478ec", "tree": "4f80219fc2153166c89c54d6555498155fa2f3b2", "parents": [ "5ba5891d44a6acade44887a0f3195489d46c12de" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] Remove all ifdefs for local/io apic\n\nIO-APIC or local APIC can only be disabled at runtime anyways and\nKconfig has forced these options on for a long time now.\n\nThe Kconfigs are kept only now for the benefit of the shared acpi\nboot.c code.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "ecaf45ee5ce60afe7cc46e91d82c1b0cbda09387", "tree": "09ff150151dc6724f250da068c2653dc13b4e3c0", "parents": [ "07c9819b31eda7954feddc83f2fae035f31c11e1" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] i386: Redo semaphore and rwlock assembly helpers\n\n- Move them to a pure assembly file. Previously they were in\na C file that only consisted of inline assembly. Doing it in pure\nassembler is much nicer.\n- Add a frame.i include with FRAME/ENDFRAME macros to easily\nadd frame pointers to assembly functions\n- Add dwarf2 annotation to them so that the new dwarf2 unwinder\ndoesn\u0027t get stuck on them\n- Random cleanups\n\nIncludes feedback from Jan Beulich and a UML build fix from Andrew\nMorton.\n\nCc: jbeulich@novell.com\nCc: jdike@addtoit.com\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "07c9819b31eda7954feddc83f2fae035f31c11e1", "tree": "7eb34bdcd74075d0e01da0daa93d5123032bc7bb", "parents": [ "b06babac45e1546dfb504f1f25eb0495632bfc41" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] i386: add alternative-asm.h to allow LOCK_PREFIX replacement in .S files\n\nLOCK_PREFIX is replaced by nops on UP systems, so it has to be a special\nmacro. Previously this was only possible from C. Allow it for pure\nassembly files too. Similar to earlier x86-64 patch.\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b06babac45e1546dfb504f1f25eb0495632bfc41", "tree": "687c60b4db34301f3d96f50d8ee4577a7427f155", "parents": [ "1a015b5644ec6df0a2c4cbeff1f8a3d24ba0478e" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] Add proper alignment to ENTRY\n\nPreviously it didn\u0027t align. Use the same one as the C compiler\nin blended mode, which is good for K8 and Core2 and doesn\u0027t hurt\non P4.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "1a015b5644ec6df0a2c4cbeff1f8a3d24ba0478e", "tree": "776cc4bfdf3a6aa74269a5b8458759a2157c770a", "parents": [ "9a0b26e6bc4ae1979d9bcc6194e57a71b2b5cac6" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:29 2006 +0200" }, "message": "[PATCH] i386: Remove const case for rwlocks\n\nrwlocks are now out of line, so it near never triggers. Also it was\nincompatible with the new dwarf2 unwinder because it had unannotiatable\npush/pops.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "9a0b26e6bc4ae1979d9bcc6194e57a71b2b5cac6", "tree": "01601a502007b85d4fc88f79de7a93c257ef9d97", "parents": [ "b4062b16094038334d9bbadac0397a3fc9e981b0" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Clean up read write lock assembly\n\n- Move the slow path fallbacks to their own assembly files\nThis makes them much easier to read and is needed for the next change.\n- Add CFI annotations for unwinding (XXX need review)\n- Remove constant case which can never happen with out of line spinlocks\n- Use patchable LOCK prefixes\n- Don\u0027t use lock sections anymore for inline code because they can\u0027t\nbe expressed by the unwinder (this adds one taken jump to the lock\nfast path)\n\nCc: jbeulich@novell.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b4062b16094038334d9bbadac0397a3fc9e981b0", "tree": "e75506b979825010319f96a0868b71788ac856df", "parents": [ "538b5b419c7ae39a4b2deb15278da36102e42346" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Support patchable lock prefix for pure assembly files\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "0cb91a2293648507886563ccb91979cfc94d6a4b", "tree": "f2a994f8b874e3c81c5519d445a0c5fa5522e4b5", "parents": [ "c16b63e09d9d03158e0a92e961234e94c4862620" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] i386: Account spinlocks to the caller during profiling for !FP kernels\n\nThis ports the algorithm from x86-64 (with improvements) to i386.\nPreviously this only worked for frame pointer enabled kernels.\nBut spinlocks have a very simple stack frame that can be manually\nanalyzed. Do this.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "d3a4f48d4866b8623ca9adde8ce4e5fde979c132", "tree": "04413a1f18c6f49688d6008b583b20a287920a0f", "parents": [ "2f766d16062d0147edff91be15de4a950667ca42" ], "author": { "name": "Stephane Eranian", "email": "eranian@hpl.hp.com", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] x86-64 TIF flags for debug regs and io bitmap in ctxsw\n\nHello,\n\nFollowing my discussion with Andi. Here is a patch that introduces\ntwo new TIF flags to simplify the context switch code in __switch_to().\nThe idea is to minimize the number of cache lines accessed in the common\ncase, i.e., when neither the debug registers nor the I/O bitmap are used.\n\nThis patch covers the x86-64 modifications. A patch for i386 follows.\n\nChangelog:\n\t- add TIF_DEBUG to track when debug registers are active\n\t- add TIF_IO_BITMAP to track when I/O bitmap is used\n\t- modify __switch_to() to use the new TIF flags\n\n\u003csigned-off-by\u003e: eranian@hpl.hp.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "2f766d16062d0147edff91be15de4a950667ca42", "tree": "20c7f80312aedabf8659483b53f67afa8ccc51db", "parents": [ "3cfc348bf90ffaa777c188652aa297f04eb94de8" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Clean up asm/smp.h includes\n\nNo need to include it from entry.S\nDrop all the #ifdef __ASSEMBLY__\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3cfc348bf90ffaa777c188652aa297f04eb94de8", "tree": "8908d6a5a61e54ab422ec7f4800d6ac591695423", "parents": [ "c08c820508233b424deab3302bc404bbecc6493a" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] x86: Add portable getcpu call\n\nFor NUMA optimization and some other algorithms it is useful to have a fast\nto get the current CPU and node numbers in user space.\n\nx86-64 added a fast way to do this in a vsyscall. This adds a generic\nsyscall for other architectures to make it a generic portable facility.\n\nI expect some of them will also implement it as a faster vsyscall.\n\nThe cache is an optimization for the x86-64 vsyscall optimization. Since\nwhat the syscall returns is an approximation anyways and user space\noften wants very fast results it can be cached for some time. The norma\nmethods to get this information in user space are relatively slow\n\nThe vsyscall is in a better position to manage the cache because it has direct\naccess to a fast time stamp (jiffies). For the generic syscall optimization\nit doesn\u0027t help much, but enforce a valid argument to keep programs\nportable\n\nI only added an i386 syscall entry for now. Other architectures can follow\nas needed.\n\nAK: Also added some cleanups from Andrew Morton\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "c08c820508233b424deab3302bc404bbecc6493a", "tree": "d6ed79fb2d03513f6c71fc7e47705c4a19512f53", "parents": [ "a670fad0adb1cc6202a607d250f10bd380593905" ], "author": { "name": "Vojtech Pavlik", "email": "vojtech@suse.cz", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Add the vgetcpu vsyscall\n\nThis patch adds a vgetcpu vsyscall, which depending on the CPU RDTSCP\ncapability uses either the RDTSCP or CPUID to obtain a CPU and node\nnumbers and pass them to the program.\n\nAK: Lots of changes over Vojtech\u0027s original code:\nBetter prototype for vgetcpu()\nIt\u0027s better to pass the cpu / node numbers as separate arguments\nto avoid mistakes when going from SMP to NUMA.\nAlso add a fast time stamp based cache using a user supplied\nargument to speed things more up.\nUse fast method from Chuck Ebbert to retrieve node/cpu from\nGDT limit instead of CPUID\nMade sure RDTSCP init is always executed after node is known.\nDrop printk\n\nSigned-off-by: Vojtech Pavlik \u003cvojtech@suse.cz\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "a670fad0adb1cc6202a607d250f10bd380593905", "tree": "697d27fe0c92a65829f727f17f7048d46a675e4d", "parents": [ "81af4449af9c9b686a4eeeb00112614621655704" ], "author": { "name": "Vojtech Pavlik", "email": "vojtech@suse.cz", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Add initalization of the RDTSCP auxilliary values\n\nThis patch adds initalization of the RDTSCP auxilliary values to CPU numbers\nto time.c. If RDTSCP is available, the MSRs are written with the respective\nvalues. It can be later used to initalize per-cpu timekeeping variables.\n\nAK: Some cleanups. Move externs into headers and fix CPU hotplug.\n\nSigned-off-by: Vojtech Pavlik \u003cvojtech@suse.cz\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "81af4449af9c9b686a4eeeb00112614621655704", "tree": "52dc5cac0124a87a6a323f415338c31631dbbf20", "parents": [ "248dcb2ffffe8f3e4a369556a68988788c208111" ], "author": { "name": "Vojtech Pavlik", "email": "vojtech@suse.cz", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:28 2006 +0200" }, "message": "[PATCH] Add macros for rdtscp\n\nThis patch adds macros for reading tsc via the RDTSCP instruction, as well\nas writing the auxilliary MSR read by RDTSCP to msr.h\n\n[AK: changed rdtscp definition for old binutils]\n\nSigned-off-by: Vojtech Pavlik \u003cvojtech@suse.cz\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "248dcb2ffffe8f3e4a369556a68988788c208111", "tree": "07517710662b99db976e74064bfe399681437e25", "parents": [ "1de84979dfc527c422abf63f27beabe43892989b" ], "author": { "name": "Venkatesh Pallipadi", "email": "venkatesh.pallipadi@intel.com", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] x86: i386/x86-64 Add nmi watchdog support for new Intel CPUs\n\nAK: This redoes the changes I temporarily reverted.\n\nIntel now has support for Architectural Performance Monitoring Counters\n( Refer to IA-32 Intel Architecture Software Developer\u0027s Manual\nhttp://www.intel.com/design/pentium4/manuals/253669.htm ). This\nfeature is present starting from Intel Core Duo and Intel Core Solo processors.\n\nWhat this means is, the performance monitoring counters and some performance\nmonitoring events are now defined in an architectural way (using cpuid).\nAnd there will be no need to check for family/model etc for these architectural\nevents.\n\nBelow is the patch to use this performance counters in nmi watchdog driver.\nPatch handles both i386 and x86-64 kernels.\n\nSigned-off-by: Venkatesh Pallipadi \u003cvenkatesh.pallipadi@intel.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "fac58550e80c307bf17cfa0dd544fca4eff120a5", "tree": "04eb287f3dd1bb2d7e9887e4d25073ca9794fa27", "parents": [ "4038f901cf102a40715b900984ed7540a9fa637f" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] Fix up panic messages for different NMI panics\n\nWhen a unknown NMI happened the panic would claim a NMI watchdog timeout.\nAlso it would check the variable set by nmi_watchdog\u003dpanic and panic then.\n\nFix up the panic message to be generic\nUnconditionally panic on unknown NMI when panic on unknown nmi is enabled.\n\nNoticed by Jan Beulich\n\nCc: jbeulich@novell.com\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "4038f901cf102a40715b900984ed7540a9fa637f", "tree": "e73261bee0e0856dba5a8bc447b18779a61fe235", "parents": [ "c41c5cd3b20a2d81c30498f13b1527847a8fdf69" ], "author": { "name": "Shaohua Li", "email": "shaohua.li@intel.com", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] i386/x86-64: Fix NMI watchdog suspend/resume\n\nMaking NMI suspend/resume work with SMP. We use CPU hotplug to offline\nAPs in SMP suspend/resume. Only BSP executes sysdev\u0027s .suspend/.resume\nmethod. APs should follow CPU hotplug code path.\n\nAnd:\n\n+From: Don Zickus \u003cdzickus@redhat.com\u003e\n\nMakes the start/stop paths of nmi watchdog more robust to handle the\nsuspend/resume cases more gracefully.\n\nAK: I merged the two patches together\n\nSigned-off-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nCc: Don Zickus \u003cdzickus@redhat.com\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\n" }, { "commit": "8da5adda91df3d2fcc5300e68da491694c9af019", "tree": "bae152dabd728ba2f7fead421276e3cc9a779141", "parents": [ "e33e89ab1a8d295de0500b697f4f31c3ceee9aa2" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] x86: Allow users to force a panic on NMI\n\nTo quote Alan Cox:\n\nThe default Linux behaviour on an NMI of either memory or unknown is to\ncontinue operation. For many environments such as scientific computing\nit is preferable that the box is taken out and the error dealt with than\nan uncorrected parity/ECC error get propogated.\n\nA small number of systems do generate NMI\u0027s for bizarre random reasons\nsuch as power management so the default is unchanged. In other respects\nthe new proc/sys entry works like the existing panic controls already in\nthat directory.\n\nThis is separate to the edac support - EDAC allows supported chipsets to\nhandle ECC errors well, this change allows unsupported cases to at least\npanic rather than cause problems further down the line.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "407984f1af259b31957c7c05075a454a751bb801", "tree": "1e9318b4255957c27a4dbacd84711604bf789393", "parents": [ "2fbe7b25c8edaf2d10e6c1a4cc9f8afe714c4764" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] x86: Add abilty to enable/disable nmi watchdog with sysctl\n\nAdds a new /proc/sys/kernel/nmi call that will enable/disable the nmi\nwatchdog.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "2fbe7b25c8edaf2d10e6c1a4cc9f8afe714c4764", "tree": "e4012ae3cd4519cba1836668237f077c60fb1086", "parents": [ "957dc87c1bd849440f0eef27e2ade67387001e13" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] i386/x86-64: Remove un/set_nmi_callback and reserve/release_lapic_nmi functions\n\nRemoves the un/set_nmi_callback and reserve/release_lapic_nmi functions as\nthey are no longer needed. The various subsystems are modified to register\nwith the die_notifier instead.\n\nAlso includes compile fixes by Andrew Morton.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "957dc87c1bd849440f0eef27e2ade67387001e13", "tree": "2a1a9cfc7b5eaa5bc8194c4c3ce0961db9401b36", "parents": [ "1d001df19d5323e642ba8ac821c713675ebccd82" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:27 2006 +0200" }, "message": "[PATCH] Add ppoll/pselect syscalls\n\nNeeded TIF_RESTORE_SIGMASK first\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "1d001df19d5323e642ba8ac821c713675ebccd82", "tree": "6eec46bca129524fc6e9ee55772a3943ff091a2f", "parents": [ "3adbbcce9a49b900d4cc118cdccfdefa78bf1afb" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] Add TIF_RESTORE_SIGMASK\n\nWe need TIF_RESTORE_SIGMASK in order to support ppoll() and pselect()\nsystem calls. This patch originally came from Andi, and was based\nheavily on David Howells\u0027 implementation of same on i386. I fixed a typo\nwhich was causing do_signal() to use the wrong signal mask.\n\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "3adbbcce9a49b900d4cc118cdccfdefa78bf1afb", "tree": "3163758ed23d973a54b396593957c63470f39a0d", "parents": [ "b7471c6da94d30d3deadc55986cc38d1ff57f9ca" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] x86: Cleanup NMI interrupt path\n\nThis patch cleans up the NMI interrupt path. Instead of being gated by if\nthe \u0027nmi callback\u0027 is set, the interrupt handler now calls everyone who is\nregistered on the die_chain and additionally checks the nmi watchdog,\nreseting it if enabled. This allows more subsystems to hook into the NMI if\nthey need to (without being block by set_nmi_callback).\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b7471c6da94d30d3deadc55986cc38d1ff57f9ca", "tree": "6aa23314273763acccbe9ddd0b8bd442edde0509", "parents": [ "f2802e7f571c05f9a901b1f5bd144aa730ccc88e" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] i386: Add SMP support on i386 to reservation framework\n\nThis patch includes the changes to make the nmi watchdog on i386 SMP aware.\nA bunch of code was moved around to make it simpler to read. In addition,\nit is now possible to determine if a particular NMI was the result of the\nwatchdog or not. This feature allows the kernel to filter out unknown NMIs\neasier.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "f2802e7f571c05f9a901b1f5bd144aa730ccc88e", "tree": "68a2c4dd80dbf2e01f048b46460517b15cda63c0", "parents": [ "cb9c448c661d40ce2efbce8e9c19cc4d420d8ccc" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] Add SMP support on x86_64 to reservation framework\n\nThis patch includes the changes to make the nmi watchdog on x86_64 SMP\naware. A bunch of code was moved around to make it simpler to read. In\naddition, it is now possible to determine if a particular NMI was the result\nof the watchdog or not. This feature allows the kernel to filter out\nunknown NMIs easier.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "828f0afda123a96ff4e8078f057a302f4b4232ae", "tree": "a6f7398e0037f5c8f4cbd95ff11c5e4bf78a4c4d", "parents": [ "b07f8915cda3fcd73b8b68075ba1e6cd0673365d" ], "author": { "name": "Don Zickus", "email": "dzickus@redhat.com", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] x86: Add performance counter reservation framework for UP kernels\n\nAdds basic infrastructure to allow subsystems to reserve performance\ncounters on the x86 chips. Only UP kernels are supported in this patch to\nmake reviewing easier. The SMP portion makes a lot more changes.\n\nThink of this as a locking mechanism where each bit represents a different\ncounter. In addition, each subsystem should also reserve an appropriate\nevent selection register that will correspond to the performance counter it\nwill be using (this is mainly neccessary for the Pentium 4 chips as they\nbreak the 1:1 relationship to performance counters).\n\nThis will help prevent subsystems like oprofile from interfering with the\nnmi watchdog.\n\nSigned-off-by: Don Zickus \u003cdzickus@redhat.com\u003e\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "b07f8915cda3fcd73b8b68075ba1e6cd0673365d", "tree": "73bd68878518350322098ddf69572c3da6f1e360", "parents": [ "874c4fe389d1358f82c96dc9b5092fc5c7690604" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] x86: Temporarily revert parts of the Core 2 nmi nmi watchdog support\n\nThis makes merging easier. They are readded a few patches later.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "874c4fe389d1358f82c96dc9b5092fc5c7690604", "tree": "9d79c398041c8902cff3abb6c9495c3d63bf59f1", "parents": [ "9142e0c8396b25ed4cb549b5efa636065768ebe0" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "committer": { "name": "Andi Kleen", "email": "andi@basil.nowhere.org", "time": "Tue Sep 26 10:52:26 2006 +0200" }, "message": "[PATCH] i386: Allow to use GENERICARCH for UP kernels\n\nThere are some machines around (large xSeries or Unisys ES7000) that\nneed physical IO-APIC destination mode to access all of their IO\ndevices. This currently doesn\u0027t work in UP kernels as used in\ndistribution installers.\n\nThis patch allows to compile even UP kernels as GENERICARCH which\nallows to use physical or clustered APIC mode.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\n" }, { "commit": "7e4720201ad44ace85a443f41d668a62a737e7d0", "tree": "8b9118dffcfd9511d9ce31b87776e04095f9acf1", "parents": [ "7b29122f9ec54db5c38a66a11127d94db0817c17", "4cc6773508299377099aa30cf30e6a2196c5872d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Sep 25 17:39:55 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Sep 25 17:39:55 2006 -0700" }, "message": "Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\n* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:\n [NetLabel]: update docs with website information\n [NetLabel]: rework the Netlink attribute handling (part 2)\n [NetLabel]: rework the Netlink attribute handling (part 1)\n [Netlink]: add nla_validate_nested()\n [NETLINK]: add nla_for_each_nested() to the interface list\n [NetLabel]: change the SELinux permissions\n [NetLabel]: make the CIPSOv4 cache spinlocks bottom half safe\n [NetLabel]: correct improper handling of non-NetLabel peer contexts\n [TCP]: make cubic the default\n [TCP]: default congestion control menu\n [ATM] he: Fix __init/__devinit conflict\n [NETFILTER]: Add dscp,DSCP headers to header-y\n [DCCP]: Introduce dccp_probe\n [DCCP]: Use constants for CCIDs\n [DCCP]: Introduce constants for CCID numbers\n [DCCP]: Allow default/fallback service code.\n" }, { "commit": "3212fe1594e577463bc8601d28aa008f520c3377", "tree": "6f4270c825c2df9f4a361360dab0963ecaed34c1", "parents": [ "08992986497471ce575f23796268fb1b50b5c2ab" ], "author": { "name": "KAMEZAWA Hiroyuki", "email": "kamezawa.hiroyu@jp.fujitsu.com", "time": "Mon Sep 25 16:25:31 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Sep 25 17:38:36 2006 -0700" }, "message": "[PATCH] cpu to node relationship fixup: map cpu to node\n\nAssume that a cpu is *physically* offlined at boot time...\n\nBecause smpboot.c::smp_boot_cpu_map() canoot find cpu\u0027s sapicid,\nnuma.c::build_cpu_to_node_map() cannot build cpu\u003c-\u003enode map for\nofflined cpu.\n\nFor such cpus, cpu_to_node map should be fixed at cpu-hot-add.\nThis mapping should be done before cpu onlining.\n\nThis patch also handles cpu hotremove case.\n\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fcd48280643e92ec6cb29a04e9079dd7b6b5bfef", "tree": "c594e16a021262e97f8b41493529c95bd616529e", "parents": [ "4fe5d5c07ab615a52fd1b0ceba5aeed7c612821a" ], "author": { "name": "Paul Moore", "email": "paul.moore@hp.com", "time": "Mon Sep 25 15:56:09 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 25 15:56:09 2006 -0700" }, "message": "[NetLabel]: rework the Netlink attribute handling (part 1)\n\nAt the suggestion of Thomas Graf, rewrite NetLabel\u0027s use of Netlink attributes\nto better follow the common Netlink attribute usage.\n\nSigned-off-by: Paul Moore \u003cpaul.moore@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4fe5d5c07ab615a52fd1b0ceba5aeed7c612821a", "tree": "245e504bb1049e86f9f6dc369b679c5a5ca495ec", "parents": [ "22acb19a91d2b551ea37647747972e5286284b22" ], "author": { "name": "Paul Moore", "email": "paul.moore@hp.com", "time": "Mon Sep 25 15:54:03 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 25 15:54:03 2006 -0700" }, "message": "[Netlink]: add nla_validate_nested()\n\nAdd a new function, nla_validate_nested(), to validate nested Netlink\nattributes.\n\nSigned-off-by: Paul Moore \u003cpaul.moore@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "22acb19a91d2b551ea37647747972e5286284b22", "tree": "c37455de7df838e76b6f4279f084bce95c16e416", "parents": [ "df2115c3134d0d1a18c1f37f5192394e7f64d1e0" ], "author": { "name": "Paul Moore", "email": "paul.moore@hp.com", "time": "Mon Sep 25 15:53:37 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 25 15:53:37 2006 -0700" }, "message": "[NETLINK]: add nla_for_each_nested() to the interface list\n\nAt the top of include/net/netlink.h is a list of Netlink interfaces, however,\nthe nla_for_each_nested() macro was not listed. This patch adds this interface\nto the list at the top of the header file.\n\nSigned-off-by: Paul Moore \u003cpaul.moore@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "14a72f53fb1bb5d5c2bdd8cf172219519664729a", "tree": "95a077fb9289a95c352af77f18f12e5aba3313c6", "parents": [ "597811ec167fa01c926a0957a91d9e39baa30e64" ], "author": { "name": "Paul Moore", "email": "paul.moore@hp.com", "time": "Mon Sep 25 15:52:01 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 25 15:52:01 2006 -0700" }, "message": "[NetLabel]: correct improper handling of non-NetLabel peer contexts\n\nFix a problem where NetLabel would always set the value of \nsk_security_struct-\u003epeer_sid in selinux_netlbl_sock_graft() to the context of\nthe socket, causing problems when users would query the context of the\nconnection. This patch fixes this so that the value in\nsk_security_struct-\u003epeer_sid is only set when the connection is NetLabel based,\notherwise the value is untouched.\n\nSigned-off-by: Paul Moore \u003cpaul.moore@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a6d967a485c67ec8a1276261f39d81ace6a3e308", "tree": "3e46be7f279342316d4578995dfbfce99728c579", "parents": [ "7c250413e5b7c3dfae89354725b70c76d7621395" ], "author": { "name": "Jeff Garzik", "email": "jeff@garzik.org", "time": "Mon Sep 25 15:33:09 2006 -0400" }, "committer": { "name": "Jeff Garzik", "email": "jeff@garzik.org", "time": "Mon Sep 25 15:33:09 2006 -0400" }, "message": "[libata] No need for all those arch libata-portmap.h headers\n\nThey all contain the same thing. Instead, have a single generic one in\ninclude/asm-generic, and permit an arch to override as needed.\n\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\n" } ], "next": "3cc27547d6ee2d50ecdd11e9127bc3cd1947e8dd" }