)]}' { "log": [ { "commit": "ea7145477a461e09d8d194cac4b996dc4f449107", "tree": "aa45594b6303b854f5fa85f82ba5dca01e0aa8ac", "parents": [ "86cb2ec7b22a0a89b8660110dc03321fadbef45f" ], "author": { "name": "Jiri Olsa", "email": "jolsa@redhat.com", "time": "Mon Mar 07 19:10:39 2011 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Mar 08 17:22:11 2011 +0100" }, "message": "x86: Separate out entry text section\n\nPut x86 entry code into a separate link section: .entry.text.\n\nSeparating the entry text section seems to have performance\nbenefits - caused by more efficient instruction cache usage.\n\nRunning hackbench with perf stat --repeat showed that the change\ncompresses the icache footprint. The icache load miss rate went\ndown by about 15%:\n\n before patch:\n 19417627 L1-icache-load-misses ( +- 0.147% )\n\n after patch:\n 16490788 L1-icache-load-misses ( +- 0.180% )\n\nThe motivation of the patch was to fix a particular kprobes\nbug that relates to the entry text section, the performance\nadvantage was discovered accidentally.\n\nWhole perf output follows:\n\n - results for current tip tree:\n\n Performance counter stats for \u0027./hackbench/hackbench 10\u0027 (500 runs):\n\n 19417627 L1-icache-load-misses ( +- 0.147% )\n 2676914223 instructions # 0.497 IPC ( +- 0.079% )\n 5389516026 cycles ( +- 0.144% )\n\n 0.206267711 seconds time elapsed ( +- 0.138% )\n\n - results for current tip tree with the patch applied:\n\n Performance counter stats for \u0027./hackbench/hackbench 10\u0027 (500 runs):\n\n 16490788 L1-icache-load-misses ( +- 0.180% )\n 2717734941 instructions # 0.502 IPC ( +- 0.079% )\n 5414756975 cycles ( +- 0.148% )\n\n 0.206747566 seconds time elapsed ( +- 0.137% )\n\nSigned-off-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nCc: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: masami.hiramatsu.pt@hitachi.com\nCc: ananth@in.ibm.com\nCc: davem@davemloft.net\nCc: 2nddept-manager@sdl.hitachi.co.jp\nLKML-Reference: \u003c20110307181039.GB15197@jolsa.redhat.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "00afe029aab03bd95eba210b5e74a252017c4692", "tree": "4c9aa4a10d7fa5f813896c8c439df592aacf295d", "parents": [ "a9ece53c4089ef23d4002d34c4c7148d94622a40" ], "author": { "name": "Mike Frysinger", "email": "vapier@gentoo.org", "time": "Tue Sep 22 16:44:14 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 23 07:39:30 2009 -0700" }, "message": "asm/sections: add text/data checking functions for arches to override\n\nSome ports (like the Blackfin arch) have a discontiguous memory map which\nmeans there may be text or data that falls outside of the standard range\nof the start/end text/data symbols. Creating some helper functions allows\nthese non-standard ports to declare these regions without adversely\naffecting anyone else.\n\nSigned-off-by: Mike Frysinger \u003cvapier@gentoo.org\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Robin Getz \u003crgetz@blackfin.uclinux.org\u003e\nCc: Sam Ravnborg \u003csam@ravnborg.org\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b99b87f70c7785ab1e253c6220f4b0b57ce3a7f7", "tree": "ec5688052334448ec8edd3a1a9cb95cd68501ac7", "parents": [ "e24aca672ff06aff0e6a1045efab86043ea5f735" ], "author": { "name": "Peter Oberparleiter", "email": "oberpar@linux.vnet.ibm.com", "time": "Wed Jun 17 16:28:03 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 18 13:03:57 2009 -0700" }, "message": "kernel: constructor support\n\nCall constructors (gcc-generated initcall-like functions) during kernel\nstart and module load. Constructors are e.g. used for gcov data\ninitialization.\n\nDisable constructor support for usermode Linux to prevent conflicts with\nhost glibc.\n\nSigned-off-by: Peter Oberparleiter \u003coberpar@linux.vnet.ibm.com\u003e\nAcked-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nAcked-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nCc: Sam Ravnborg \u003csam@ravnborg.org\u003e\nCc: Jeff Dike \u003cjdike@addtoit.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Huang Ying \u003cying.huang@intel.com\u003e\nCc: Li Wei \u003cW.Li@Sun.COM\u003e\nCc: Michael Ellerman \u003cmichaele@au1.ibm.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Heiko Carstens \u003cheicars2@linux.vnet.ibm.com\u003e\nCc: Martin Schwidefsky \u003cmschwid2@linux.vnet.ibm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3e5d8f978435bb9ba4dfe3f4514e65e7885db1a9", "tree": "1ce55b2ec16a0bd59a29857e05215960d463a1d8", "parents": [ "a698c823e15149941b0f0281527d0c0d1daf2639" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Jan 13 20:41:35 2009 +0900" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Jan 16 14:19:14 2009 +0100" }, "message": "x86: make percpu symbols zerobased on SMP\n\n[ Based on original patch from Christoph Lameter and Mike Travis. ]\n\nThis patch makes percpu symbols zerobased on x86_64 SMP by adding\nPERCPU_VADDR() to vmlinux.lds.h which helps setting explicit vaddr on\nthe percpu output section and using it in vmlinux_64.lds.S. A new\nPHDR is added as existing ones cannot contain sections near address\nzero. PERCPU_VADDR() also adds a new symbol __per_cpu_load which\nalways points to the vaddr of the loaded percpu data.init region.\n\nThe following adjustments have been made to accomodate the address\nchange.\n\n* code to locate percpu gdt_page in head_64.S is updated to add the\n load address to the gdt_page offset.\n\n* __per_cpu_load is used in places where access to the init data area\n is necessary.\n\n* pda-\u003edata_offset is initialized soon after C code is entered as zero\n value doesn\u0027t work anymore.\n\nThis patch is mostly taken from Mike Travis\u0027 \"x86_64: Base percpu\nvariables at zero\" patch.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "deac93df26b20cf8438339b5935b5f5643bc30c9", "tree": "8e67edd505e4a8bde380c724b67ce9ca98b8ee91", "parents": [ "7ae115b4f50d3c5824f1a15e572b5de9d1b06d35" ], "author": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Wed Sep 03 20:43:36 2008 -0500" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Sep 09 11:51:15 2008 -0700" }, "message": "lib: Correct printk %pF to work on all architectures\n\nIt was introduced by \"vsprintf: add support for \u0027%pS\u0027 and \u0027%pF\u0027 pointer\nformats\" in commit 0fe1ef24f7bd0020f29ffe287dfdb9ead33ca0b2. However,\nthe current way its coded doesn\u0027t work on parisc64. For two reasons: 1)\nparisc isn\u0027t in the #ifdef and 2) parisc has a different format for\nfunction descriptors\n\nMake dereference_function_descriptor() more accommodating by allowing\narchitecture overrides. I put the three overrides (for parisc64, ppc64\nand ia64) in arch/kernel/module.c because that\u0027s where the kernel\ninternal linker which knows how to deal with function descriptors sits.\n\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: Tony Luck \u003ctony.luck@intel.com\u003e\nAcked-by: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a3b81113fb6658629f4ebaabf8dd3067cd341020", "tree": "c6823e84d410f18b110ce07726643d7dc841553d", "parents": [ "0a5dcb51770be3cd0202d6b90a07996fb40130b6" ], "author": { "name": "Robin Getz", "email": "rgetz@blackfin.uclinux.org", "time": "Wed Feb 06 01:36:26 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Feb 06 10:41:01 2008 -0800" }, "message": "remove support for un-needed _extratext section\n\nWhen passing a zero address to kallsyms_lookup(), the kernel thought it was\na valid kernel address, even if it is not. This is because is_ksym_addr()\ncalled is_kernel_extratext() and checked against labels that don\u0027t exist on\nmany archs (which default as zero). Since PPC was the only kernel which\ndefines _extra_text, (in 2005), and no longer needs it, this patch removes\n_extra_text support.\n\nFor some history (provided by Jon):\n http://ozlabs.org/pipermail/linuxppc-dev/2005-September/019734.html\n http://ozlabs.org/pipermail/linuxppc-dev/2005-September/019736.html\n http://ozlabs.org/pipermail/linuxppc-dev/2005-September/019751.html\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Robin Getz \u003crgetz@blackfin.uclinux.org\u003e\nCc: David Woodhouse \u003cdwmw2@infradead.org\u003e\nCc: Jon Loeliger \u003cjdl@freescale.com\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Sam Ravnborg \u003csam@ravnborg.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a581c2a4697ee264699b364399b73477af408e00", "tree": "972d8dd6b6a90de49b080340806487c30a07cc1a", "parents": [ "e2c2770096b686b4d2456173f53cb50e01aa635c" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Sat Jul 01 04:36:30 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Jul 01 09:56:03 2006 -0700" }, "message": "[PATCH] add __[start|end]_rodata sections to asm-generic/sections.h\n\nAdd __start_rodata and __end_rodata to sections.h to avoid extern\ndeclarations. Needed by s390 code (see following patch).\n\n[akpm@osdl.org: update architectures]\nCc: Arjan van de Ven \u003carjan@infradead.org\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nAcked-by: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "2bc0414ee04fd8bb798760801f5d7476dff44241", "tree": "3ae48fd461dca1f7f46f440325c674717a36f693", "parents": [ "6b75aeedde1e8a8513393d3c1367bf81bc5b0c67" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Sat Nov 05 17:25:53 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Nov 14 19:55:14 2005 -0800" }, "message": "[PATCH] x86_64: Only use asm/sections.h to declare section symbols\n\nAdding __initdata_* to asm-generic/sections.h\nReplaces a lot of open coded externs in arch/x86_64/*\nI had to change __bss_end to __bss_stop to match the other architectures.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d0aaff9796c3310326d10da44fc0faed352a1d29", "tree": "591fd8dedf34464989d23bbb0e66a1ccb2fa18a6", "parents": [ "505db03639db34ca2c64fe7ee27190d324281f2c" ], "author": { "name": "Prasanna S Panchamukhi", "email": "prasanna@in.ibm.com", "time": "Tue Sep 06 15:19:26 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Sep 07 16:57:59 2005 -0700" }, "message": "[PATCH] Kprobes: prevent possible race conditions generic\n\nThere are possible race conditions if probes are placed on routines within the\nkprobes files and routines used by the kprobes. For example if you put probe\non get_kprobe() routines, the system can hang while inserting probes on any\nroutine such as do_fork(). Because while inserting probes on do_fork(),\nregister_kprobes() routine grabs the kprobes spin lock and executes\nget_kprobe() routine and to handle probe of get_kprobe(), kprobes_handler()\ngets executed and tries to grab kprobes spin lock, and spins forever. This\npatch avoids such possible race conditions by preventing probes on routines\nwithin the kprobes file and routines used by kprobes.\n\nI have modified the patches as per Andi Kleen\u0027s suggestion to move kprobes\nroutines and other routines used by kprobes to a seperate section\n.kprobes.text.\n\nAlso moved page fault and exception handlers, general protection fault to\n.kprobes.text section.\n\nThese patches have been tested on i386, x86_64 and ppc64 architectures, also\ncompiled on ia64 and sparc64 architectures.\n\nSigned-off-by: Prasanna S Panchamukhi \u003cprasanna@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "a940199f206dcf51c65fae27e2ce412f2c5a2b22", "tree": "ada838ac347214d20fcb6003cba1c4938dc7ebae", "parents": [ "de04f3220b9789cc40fd6b1d151dfa0a7e3b03b9" ], "author": { "name": "Andi Kleen", "email": "ak@suse.de", "time": "Thu Jul 28 21:15:30 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jul 28 21:45:58 2005 -0700" }, "message": "[PATCH] x86_64: Some cleanup in setup64.c\n\nMinor cleanup.\n\nMove things into their include files, remove obsolete includes, fix\nindentation, remove obsolete special cases etc.\n\nI also added the per cpu section to asm-generic/sections.h and fixed\ninit/main.c to use it.\n\nSigned-off-by: Andi Kleen \u003cak@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "075d6eb16d273dab7b7b4b83fcee8bce4ee387ed", "tree": "d5e2cfb21dd752410649a675f2bb2a7c0db05c8e", "parents": [ "becf3aec2608d6807a58d0677661cb23c388d67f" ], "author": { "name": "David Woodhouse", "email": "dwmw2@infradead.org", "time": "Thu May 05 16:15:09 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu May 05 16:36:31 2005 -0700" }, "message": "[PATCH] ppc32: platform-specific functions missing from kallsyms.\n\nThe PPC32 kernel puts platform-specific functions into separate sections so\nthat unneeded parts of it can be freed when we\u0027ve booted and actually\nworked out what we\u0027re running on today.\n\nThis makes kallsyms ignore those functions, because they\u0027re not between\n_[se]text or _[se]inittext. Rather than teaching kallsyms about the\nvarious pmac/chrp/etc sections, this patch adds \u0027_[se]extratext\u0027 markers\nfor kallsyms.\n\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2", "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d", "parents": [], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n" } ] }