)]}' { "log": [ { "commit": "17f272d8e4349543de855ea63e4eea4a67a7abc5", "tree": "aa8904eecaea5c48ecead4bac953fd69eba18d62", "parents": [ "e358073331debaab9e14b3139bc469184288aa48" ], "author": { "name": "Yegor Yefremov", "email": "yegor_sub1@visionsystems.de", "time": "Mon Jan 23 08:32:23 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Feb 13 11:06:05 2012 -0800" }, "message": "ARM: OMAP2+: GPMC: fix device size setup\n\ncommit 8ef5d844cc3a644ea6f7665932a4307e9fad01fa upstream.\n\nfollowing statement can only change device size from 8-bit(0) to 16-bit(1),\nbut not vice versa:\n\nregval |\u003d GPMC_CONFIG1_DEVICESIZE(wval);\n\nso as this field has 1 reserved bit, that could be used in future,\njust clear both bits and then OR with the desired value\n\nSigned-off-by: Yegor Yefremov \u003cyegorslists@googlemail.com\u003e\nSigned-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "e358073331debaab9e14b3139bc469184288aa48", "tree": "f2230b09275a747f596c0a175fb768e18ad269f8", "parents": [ "ae3939e12cc597be0ba28d5f3b3d8f158f2d6d70" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Mon Jan 30 20:23:29 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Feb 13 11:06:05 2012 -0800" }, "message": "ARM: 7308/1: vfp: flush thread hwstate before copying ptrace registers\n\ncommit 8130b9d7b9d858aa04ce67805e8951e3cb6e9b2f upstream.\n\nIf we are context switched whilst copying into a thread\u0027s\nvfp_hard_struct then the partial copy may be corrupted by the VFP\ncontext switching code (see \"ARM: vfp: flush thread hwstate before\nrestoring context from sigframe\").\n\nThis patch updates the ptrace VFP set code so that the thread state is\nflushed before the copy, therefore disabling VFP and preventing\ncorruption from occurring.\n\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "ae3939e12cc597be0ba28d5f3b3d8f158f2d6d70", "tree": "626aec44e1fdf2316ec58c675d31aea43db92dce", "parents": [ "886d462b0c92357a2f0ca0fc19f5877efbb1f638" ], "author": { "name": "Dave Martin", "email": "dave.martin@linaro.org", "time": "Mon Jan 30 20:22:28 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Feb 13 11:06:05 2012 -0800" }, "message": "ARM: 7307/1: vfp: fix ptrace regset modification race\n\ncommit 247f4993a5974e6759606c4d380748eecfd273ff upstream.\n\nIn a preemptible kernel, vfp_set() can be preempted, causing the\nhardware VFP context to be switched while the thread vfp state is\nbeing read and modified. This leads to a race condition which can\ncause the thread vfp state to become corrupted if lazy VFP context\nsave occurs due to preemption in between the time thread-\u003evfpstate\nis read and the time the modified state is written back.\n\nThis may occur if preemption occurs during the execution of a\nptrace() call which modifies the VFP register state of a thread.\nSuch instances should be very rare in most realistic scenarios --\nnone has been reported, so far as I am aware. Only uniprocessor\nsystems should be affected, since VFP context save is not currently\nlazy in SMP kernels.\n\nThe problem was introduced by my earlier patch migrating to use\nregsets to implement ptrace.\n\nThis patch does a vfp_sync_hwstate() before reading\nthread-\u003evfpstate, to make sure that the thread\u0027s VFP state is not\nlive in the hardware registers while the registers are modified.\n\nThanks to Will Deacon for spotting this.\n\nSigned-off-by: Dave Martin \u003cdave.martin@linaro.org\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "886d462b0c92357a2f0ca0fc19f5877efbb1f638", "tree": "509ca4ebcb9b9a250e394cbe6bd946627fb3b4a6", "parents": [ "a2eeb4b984210336662a9ebdbecb96efd8823ff2" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Mon Jan 30 20:21:42 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Feb 13 11:06:04 2012 -0800" }, "message": "ARM: 7306/1: vfp: flush thread hwstate before restoring context from sigframe\n\ncommit 2af276dfb1722e97b190bd2e646b079a2aa674db upstream.\n\nFollowing execution of a signal handler, we currently restore the VFP\ncontext from the ucontext in the signal frame. This involves copying\nfrom the user stack into the current thread\u0027s vfp_hard_struct and then\nflushing the new data out to the hardware registers.\n\nThis is problematic when using a preemptible kernel because we could be\ncontext switched whilst updating the vfp_hard_struct. If the current\nthread has made use of VFP since the last context switch, the VFP\nnotifier will copy from the hardware registers into the vfp_hard_struct,\noverwriting any data that had been partially copied by the signal code.\n\nDisabling preemption across copy_from_user calls is a terrible idea, so\ninstead we move the VFP thread flush *before* we update the\nvfp_hard_struct. Since the flushing is performed lazily, this has the\neffect of disabling VFP and clearing the CPU\u0027s VFP state pointer,\ntherefore preventing the thread from being updated with stale data on\nthe next context switch.\n\nTested-by: Peter Maydell \u003cpeter.maydell@linaro.org\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "d020b1d3d3379d183d0649cdc2f6de9131268419", "tree": "63d4be0201b59623b2380bc5241b1bd280434326", "parents": [ "1334533665277ccc5568c5104cd2358788a02e02" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Wed Jan 18 07:21:42 2012 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Feb 03 09:19:04 2012 -0800" }, "message": "net: bpf_jit: fix divide by 0 generation\n\n[ Upstream commit d00a9dd21bdf7908b70866794c8313ee8a5abd5c ]\n\nSeveral problems fixed in this patch :\n\n1) Target of the conditional jump in case a divide by 0 is performed\n by a bpf is wrong.\n\n2) Must \u0027generate\u0027 the full function prologue/epilogue at pass\u003d0,\n or else we can stop too early in pass\u003d1 if the proglen doesnt change.\n (if the increase of prologue/epilogue equals decrease of all\n instructions length because some jumps are converted to near jumps)\n\n3) Change the wrong length detection at the end of code generation to\n issue a more explicit message, no need for a full stack trace.\n\nReported-by: Phil Oester \u003ckernel@linuxace.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "95086de856ca703588967f63cc54b8059db55888", "tree": "164fe7a1fc35a4a2ff10bc2ed46f30a2c6eef119", "parents": [ "4e29fa93520b401b81a719fe053f06c6937d5b66" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Fri Jan 20 12:10:18 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Feb 03 09:18:58 2012 -0800" }, "message": "ARM: 7296/1: proc-v7.S: remove HARVARD_CACHE preprocessor guards\n\ncommit 612539e81f655f6ac73c7af1da8701c1ee618aee upstream.\n\nOn v7, we use the same cache maintenance instructions for data lines\nas for unified lines. This was not the case for v6, where HARVARD_CACHE\nwas defined to indicate the L1 cache topology.\n\nThis patch removes the erroneous compile-time check for HARVARD_CACHE in\nproc-v7.S, ensuring that we perform I-side invalidation at boot.\n\nReported-and-Acked-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\n\nAcked-by: Catalin Marinas \u003cCatalin.Marinas@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "4e29fa93520b401b81a719fe053f06c6937d5b66", "tree": "dbb58eab921e0b22ef218f5908e4df58548c0191", "parents": [ "e21ed1cebed09df11b4bf7809fd6d69b25457127" ], "author": { "name": "Srinidhi KASAGAR", "email": "srinidhi.kasagar@stericsson.com", "time": "Thu Jan 12 11:07:43 2012 +0530" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Feb 03 09:18:57 2012 -0800" }, "message": "mach-ux500: enable ARM errata 764369\n\ncommit d65015f7c5c5be9fd3f5e567889c844ba81bdc9c upstream.\n\nThis applies ARM errata 764369 for all ux500 platforms.\n\nSigned-off-by: Srinidhi Kasagar \u003csrinidhi.kasagar@stericsson.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "231f0496129d5d41ae77cc2410e3cea97540cd7b", "tree": "01500d2d505370278459bf430b6645eac67eb639", "parents": [ "8c6f009ebfcdae4196be7e456c99ad44eb013420" ], "author": { "name": "Andreas Herrmann", "email": "andreas.herrmann3@amd.com", "time": "Fri Jan 20 17:44:12 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Feb 03 09:18:56 2012 -0800" }, "message": "x86/microcode_amd: Add support for CPU family specific container files\n\ncommit 5b68edc91cdc972c46f76f85eded7ffddc3ff5c2 upstream.\n\nWe\u0027ve decided to provide CPU family specific container files\n(starting with CPU family 15h). E.g. for family 15h we have to\nload microcode_amd_fam15h.bin instead of microcode_amd.bin\n\nRationale is that starting with family 15h patch size is larger\nthan 2KB which was hard coded as maximum patch size in various\nmicrocode loaders (not just Linux).\n\nContainer files which include patches larger than 2KB cause\ndifferent kinds of trouble with such old patch loaders. Thus we\nhave to ensure that the default container file provides only\npatches with size less than 2KB.\n\nSigned-off-by: Andreas Herrmann \u003candreas.herrmann3@amd.com\u003e\nCc: Borislav Petkov \u003cborislav.petkov@amd.com\u003e\nCc: \u003cstable@kernel.org\u003e\nLink: http://lkml.kernel.org/r/20120120164412.GD24508@alberich.amd.com\n[ documented the naming convention and tidied the code a bit. ]\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "8c6f009ebfcdae4196be7e456c99ad44eb013420", "tree": "1f2ab47dc7991a1e5cca9163efef3c6b5a5aa0ca", "parents": [ "810b80a70cd24a682239d82370725cc3b847bab8" ], "author": { "name": "Russ Anderson", "email": "rja@sgi.com", "time": "Wed Jan 18 20:07:54 2012 -0600" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Feb 03 09:18:55 2012 -0800" }, "message": "x86/uv: Fix uv_gpa_to_soc_phys_ram() shift\n\ncommit 5a51467b146ab7948d2f6812892eac120a30529c upstream.\n\nuv_gpa_to_soc_phys_ram() was inadvertently ignoring the\nshift values. This fix takes the shift into account.\n\nSigned-off-by: Russ Anderson \u003crja@sgi.com\u003e\nLink: http://lkml.kernel.org/r/20120119020753.GA7228@sgi.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "72a82010500dea88b2d786a76b063a871e2a4603", "tree": "1076db01011cc5bf5bd52e530a44f1e849f9b1ba", "parents": [ "a6007c036e0a0e10c0b4dbd533576d201b822a90" ], "author": { "name": "Dan Rosenberg", "email": "drosenberg@vsecurity.com", "time": "Fri Jan 20 14:34:27 2012 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:25:04 2012 -0800" }, "message": "score: fix off-by-one index into syscall table\n\ncommit c25a785d6647984505fa165b5cd84cfc9a95970b upstream.\n\nIf the provided system call number is equal to __NR_syscalls, the\ncurrent check will pass and a function pointer just after the system\ncall table may be called, since sys_call_table is an array with total\nsize __NR_syscalls.\n\nWhether or not this is a security bug depends on what the compiler puts\nimmediately after the system call table. It\u0027s likely that this won\u0027t do\nanything bad because there is an additional NULL check on the syscall\nentry, but if there happens to be a non-NULL value immediately after the\nsystem call table, this may result in local privilege escalation.\n\nSigned-off-by: Dan Rosenberg \u003cdrosenberg@vsecurity.com\u003e\nCc: Chen Liqin \u003cliqin.chen@sunplusct.com\u003e\nCc: Lennox Wu \u003clennox.wu@gmail.com\u003e\nCc: Eugene Teo \u003ceugeneteo@kernel.sg\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\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "415b95df641c998a7cf15c916047f338ab2ce87b", "tree": "97fb6c5954537a10629a9c7484a01bd3829c4159", "parents": [ "6c8e76e158a9bc849ec946ab030595e06ad89210" ], "author": { "name": "Cliff Wickman", "email": "cpw@sgi.com", "time": "Mon Jan 16 15:18:48 2012 -0600" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:25:01 2012 -0800" }, "message": "x86/UV2: Fix BAU destination timeout initialization\n\ncommit d059f9fa84a30e04279c6ff615e9e2cf3b260191 upstream.\n\nMove the call to enable_timeouts() forward so that\nBAU_MISC_CONTROL is initialized before using it in\ncalculate_destination_timeout().\n\nFix the calculation of a BAU destination timeout\nfor UV2 (in calculate_destination_timeout()).\n\nSigned-off-by: Cliff Wickman \u003ccpw@sgi.com\u003e\nLink: http://lkml.kernel.org/r/20120116211848.GB5767@sgi.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0076d42a31bac7853d3c77c3858d2685f6fe8f2d", "tree": "55a28b5b5ee2d120f3769dddb399c7ad175409d2", "parents": [ "ef9a04d5b336853aeb3f3975c4e52b07c1c4d3ec" ], "author": { "name": "Kurt Garloff", "email": "kurt@garloff.de", "time": "Tue Jan 17 04:21:49 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:58 2012 -0800" }, "message": "ACPI, ia64: Use SRAT table rev to use 8bit or 16/32bit PXM fields (ia64)\n\ncommit 9f10f6a520deb3639fac78d81151a3ade88b4e7f upstream.\n\nIn SRAT v1, we had 8bit proximity domain (PXM) fields; SRAT v2 provides\n32bits for these. The new fields were reserved before.\nAccording to the ACPI spec, the OS must disregrard reserved fields.\n\nia64 did handle the PXM fields almost consistently, but depending on\nsgi\u0027s sn2 platform. This patch leaves the sn2 logic in, but does also\nuse 16/32 bits for PXM if the SRAT has rev 2 or higher.\n\nThe patch also adds __init to the two pxm accessor functions, as they\naccess __initdata now and are called from an __init function only anyway.\n\nNote that the code only uses 16 bits for the PXM field in the processor\nproximity field; the patch does not address this as 16 bits are more than\nenough.\n\nSigned-off-by: Kurt Garloff \u003ckurt@garloff.de\u003e\nSigned-off-by: Len Brown \u003clen.brown@intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ef9a04d5b336853aeb3f3975c4e52b07c1c4d3ec", "tree": "2f2472a300dd97cffed5a5018bd3721431fe75c7", "parents": [ "643147c50fde7eb0456953f468cc277d621f629e" ], "author": { "name": "Kurt Garloff", "email": "kurt@garloff.de", "time": "Tue Jan 17 04:20:31 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:57 2012 -0800" }, "message": "ACPI, x86: Use SRAT table rev to use 8bit or 32bit PXM fields (x86/x86-64)\n\ncommit cd298f60a2451a16e0f077404bf69b62ec868733 upstream.\n\nIn SRAT v1, we had 8bit proximity domain (PXM) fields; SRAT v2 provides\n32bits for these. The new fields were reserved before.\nAccording to the ACPI spec, the OS must disregrard reserved fields.\n\nx86/x86-64 was rather inconsistent prior to this patch; it used 8 bits\nfor the pxm field in cpu_affinity, but 32 bits in mem_affinity.\nThis patch makes it consistent: Either use 8 bits consistently (SRAT\nrev 1 or lower) or 32 bits (SRAT rev 2 or higher).\n\ncc: x86@kernel.org\nSigned-off-by: Kurt Garloff \u003ckurt@garloff.de\u003e\nSigned-off-by: Len Brown \u003clen.brown@intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "15274414badc6ef0a893454e7102f3753f2c1ccd", "tree": "7bd66c9a3bebcfc7118ae98ac15f64bae1202c66", "parents": [ "a9680ece8e3874382fb2db1b8e247452c306b735" ], "author": { "name": "Jack Steiner", "email": "steiner@sgi.com", "time": "Fri Jan 06 13:19:00 2012 -0600" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:49 2012 -0800" }, "message": "x86, UV: Update Boot messages for SGI UV2 platform\n\ncommit da517a08ac5913cd80ce3507cddd00f2a091b13c upstream.\n\nSGI UV systems print a message during boot:\n\n\tUV: Found \u003cnum\u003e blades\n\nDue to packaging changes, the blade count is not accurate for\non the next generation of the platform. This patch corrects the\ncount.\n\nSigned-off-by: Jack Steiner \u003csteiner@sgi.com\u003e\nLink: http://lkml.kernel.org/r/20120106191900.GA19772@sgi.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "59c43b2c3ef410e585646825ea552507cd51ccb1", "tree": "a3841562c5bb1ee5aaa3244b3c250528d5c6cdcc", "parents": [ "ea1c62778121f6ece5e0120250716b45e204cb13" ], "author": { "name": "Ludwig Nussel", "email": "ludwig.nussel@suse.de", "time": "Tue Nov 15 14:46:46 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:43 2012 -0800" }, "message": "x86: Fix mmap random address range\n\ncommit 9af0c7a6fa860698d080481f24a342ba74b68982 upstream.\n\nOn x86_32 casting the unsigned int result of get_random_int() to\nlong may result in a negative value. On x86_32 the range of\nmmap_rnd() therefore was -255 to 255. The 32bit mode on x86_64\nused 0 to 255 as intended.\n\nThe bug was introduced by 675a081 (\"x86: unify mmap_{32|64}.c\")\nin January 2008.\n\nSigned-off-by: Ludwig Nussel \u003cludwig.nussel@suse.de\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: harvey.harrison@gmail.com\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/201111152246.pAFMklOB028527@wpaz5.hot.corp.google.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "72ce943013baaba4af7afec76c3556b16b9f6de4", "tree": "c372c65ba8efbc72b870c4d7639814a68dc57d47", "parents": [ "45e7e24360e799f7753f37ca5642a0d6e29b9c62" ], "author": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Thu Jan 12 08:01:40 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:40 2012 -0800" }, "message": "x86/PCI: build amd_bus.o only when CONFIG_AMD_NB\u003dy\n\ncommit 5cf9a4e69c1ff0ccdd1d2b7404f95c0531355274 upstream.\n\nWe only need amd_bus.o for AMD systems with PCI. arch/x86/pci/Makefile\nalready depends on CONFIG_PCI\u003dy, so this patch just adds the dependency\non CONFIG_AMD_NB.\n\nCc: Yinghai Lu \u003cyinghai@kernel.org\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "45e7e24360e799f7753f37ca5642a0d6e29b9c62", "tree": "e02c560d68a4458c66e7598d458dc41e82396759", "parents": [ "0a4179971477550df61b9218e664eb9128abf2e3" ], "author": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Thu Jan 05 14:27:19 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:39 2012 -0800" }, "message": "x86/PCI: amd: factor out MMCONFIG discovery\n\ncommit 24d25dbfa63c376323096660bfa9ad45a08870ce upstream.\n\nThis factors out the AMD native MMCONFIG discovery so we can use it\noutside amd_bus.c.\n\namd_bus.c reads AMD MSRs so it can remove the MMCONFIG area from the\nPCI resources. We may also need the MMCONFIG information to work\naround BIOS defects in the ACPI MCFG table.\n\nCc: Borislav Petkov \u003cborislav.petkov@amd.com\u003e\nCc: Yinghai Lu \u003cyinghai@kernel.org\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0a4179971477550df61b9218e664eb9128abf2e3", "tree": "33309b9b7433ddd33dfc82b0de16df8644e8e181", "parents": [ "65d61b46700e059210b4c9f63355d57dc0e1a18e" ], "author": { "name": "Gary Hade", "email": "garyhade@us.ibm.com", "time": "Mon Nov 14 15:42:16 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Jan 25 17:24:39 2012 -0800" }, "message": "x86/PCI: Ignore CPU non-addressable _CRS reserved memory resources\n\ncommit ae5cd86455381282ece162966183d3f208c6fad7 upstream.\n\nThis assures that a _CRS reserved host bridge window or window region is\nnot used if it is not addressable by the CPU. The new code either trims\nthe window to exclude the non-addressable portion or totally ignores the\nwindow if the entire window is non-addressable.\n\nThe current code has been shown to be problematic with 32-bit non-PAE\nkernels on systems where _CRS reserves resources above 4GB.\n\nSigned-off-by: Gary Hade \u003cgaryhade@us.ibm.com\u003e\nReviewed-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nCc: Thomas Renninger \u003ctrenn@novell.com\u003e\nCc: linux-kernel@vger.kernel.org\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ec0d3233562d591c892841fcfa73edd3700a6ca6", "tree": "600c51c819116e9b79f8608570b1b0329bee06bd", "parents": [ "bfaebb8af017d443937022cf3d5735d726e4f711" ], "author": { "name": "Li Zhong", "email": "zhong@linux.vnet.ibm.com", "time": "Sun Dec 18 16:03:04 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Jan 12 11:35:03 2012 -0800" }, "message": "powerpc: Fix unpaired probe_hcall_entry and probe_hcall_exit\n\ncommit e4f387d8db3ba3c2dae4d8bdfe7bb5f4fe1bcb0d upstream.\n\nUnpaired calling of probe_hcall_entry and probe_hcall_exit might happen\nas following, which could cause incorrect preempt count.\n\n__trace_hcall_entry \u003d\u003e trace_hcall_entry -\u003e probe_hcall_entry \u003d\u003e\nget_cpu_var \u003d\u003e preempt_disable\n\n__trace_hcall_exit \u003d\u003e trace_hcall_exit -\u003e probe_hcall_exit \u003d\u003e\nput_cpu_var \u003d\u003e preempt_enable\n\nwhere:\nA \u003d\u003e B and A -\u003e B means A calls B, but\n\u003d\u003e means A will call B through function name, and B will definitely be\ncalled.\n-\u003e means A will call B through function pointer, so B might not be\ncalled if the function pointer is not set.\n\nSo error happens when only one of probe_hcall_entry and probe_hcall_exit\nget called during a hcall.\n\nThis patch tries to move the preempt count operations from\nprobe_hcall_entry and probe_hcall_exit to its callers.\n\nReported-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nSigned-off-by: Li Zhong \u003czhong@linux.vnet.ibm.com\u003e\nTested-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "bfaebb8af017d443937022cf3d5735d726e4f711", "tree": "8f88264306700916b41aa8b0521fbf6f57db6fae", "parents": [ "a7b8c32b67b60fe9e8b53bb86bb7a04631e6e262" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Nov 23 20:07:17 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Jan 12 11:35:02 2012 -0800" }, "message": "powerpc/time: Handle wrapping of decrementer\n\ncommit 37fb9a0231ee43d42d069863bdfd567fca2b61af upstream.\n\nWhen re-enabling interrupts we have code to handle edge sensitive\ndecrementers by resetting the decrementer to 1 whenever it is negative.\nIf interrupts were disabled long enough that the decrementer wrapped to\npositive we do nothing. This means interrupts can be delayed for a long\ntime until it finally goes negative again.\n\nWhile we hope interrupts are never be disabled long enough for the\ndecrementer to go positive, we have a very good test team that can\ndrive any kernel into the ground. The softlockup data we get back\nfrom these fails could be seconds in the future, completely missing\nthe cause of the lockup.\n\nWe already keep track of the timebase of the next event so use that\nto work out if we should trigger a decrementer exception.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "b3c5fb8252b04ec02654f80988e68852f1a14cb5", "tree": "78a9282fe0e11d9e9ef319d84aace7693fa5b59f", "parents": [ "2a89fc8b91abf1ba56daa23b05e6572b30331837" ], "author": { "name": "Markus Kötter", "email": "nepenthesdev@gmail.com", "time": "Sat Dec 17 11:39:08 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:05 2012 -0800" }, "message": "net: bpf_jit: fix an off-one bug in x86_64 cond jump target\n\n[ Upstream commit a03ffcf873fe0f2565386ca8ef832144c42e67fa ]\n\nx86 jump instruction size is 2 or 5 bytes (near/long jump), not 2 or 6\nbytes.\n\nIn case a conditional jump is followed by a long jump, conditional jump\ntarget is one byte past the start of target instruction.\n\nSigned-off-by: Markus Kötter \u003cnepenthesdev@gmail.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "2a89fc8b91abf1ba56daa23b05e6572b30331837", "tree": "39bb67a7a1075e7f8032b76c1295880a48382d06", "parents": [ "d4afed4d20e91a12de7cd1c64bb3451ee2236d19" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Dec 26 12:30:13 2011 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:05 2012 -0800" }, "message": "sparc: Fix handling of orig_i0 wrt. debugging when restarting syscalls.\n\n[ A combination of upstream commits 1d299bc7732c34d85bd43ac1a8745f5a2fed2078 and\n e88d2468718b0789b4c33da2f7e1cef2a1eee279 ]\n\nAlthough we provide a proper way for a debugger to control whether\nsyscall restart occurs, we run into problems because orig_i0 is not\nsaved and restored properly.\n\nLuckily we can solve this problem without having to make debuggers\naware of the issue. Across system calls, several registers are\nconsidered volatile and can be safely clobbered.\n\nTherefore we use the pt_regs save area of one of those registers, %g6,\nas a place to save and restore orig_i0.\n\nDebuggers transparently will do the right thing because they save and\nrestore this register already.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "d4afed4d20e91a12de7cd1c64bb3451ee2236d19", "tree": "3657ec1ce46e5e4427c4e3367cdcd1290657b4fa", "parents": [ "23a652b45554faf08ac5a7a86c176a95cce8ca83" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Oct 31 01:05:49 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:04 2012 -0800" }, "message": "sparc64: Fix masking and shifting in VIS fpcmp emulation.\n\n[ Upstream commit 2e8ecdc008a16b9a6c4b9628bb64d0d1c05f9f92 ]\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "23a652b45554faf08ac5a7a86c176a95cce8ca83", "tree": "f1edbdae2b8aa114ca04c45cf0065780f397dee9", "parents": [ "2588f7f219ab1e93be648ecf8bcb1599b1ecba09" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 19 15:31:55 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:04 2012 -0800" }, "message": "sparc32: Correct the return value of memcpy.\n\n[ Upstream commit a52312b88c8103e965979a79a07f6b34af82ca4b ]\n\nProperly return the original destination buffer pointer.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nTested-by: Kjetil Oftedal \u003coftedal@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "2588f7f219ab1e93be648ecf8bcb1599b1ecba09", "tree": "4df028200d5f4d50d14e05c6dbfed0d9946b548e", "parents": [ "9dd04b12d0588de337ff615f6991862956dd40de" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 19 15:30:14 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:03 2012 -0800" }, "message": "sparc32: Remove uses of %g7 in memcpy implementation.\n\n[ Upstream commit 21f74d361dfd6a7d0e47574e315f780d8172084a ]\n\nThis is setting things up so that we can correct the return\nvalue, so that it properly returns the original destination\nbuffer pointer.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nTested-by: Kjetil Oftedal \u003coftedal@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "9dd04b12d0588de337ff615f6991862956dd40de", "tree": "bd4a61e622e510efbba1da6e9a440a84f139821b", "parents": [ "2d2eb1d284257cbb7ebb29bd75a3cbbc9275e4f7" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 19 15:15:58 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:03 2012 -0800" }, "message": "sparc32: Remove non-kernel code from memcpy implementation.\n\n[ Upstream commit 045b7de9ca0cf09f1adc3efa467f668b89238390 ]\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nTested-by: Kjetil Oftedal \u003coftedal@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "2d2eb1d284257cbb7ebb29bd75a3cbbc9275e4f7", "tree": "98fba028d8615e534b2f5a81b219e25fc992aa89", "parents": [ "cff6d2096e9a57c2497dd5ee4aed3c97149bfc9e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 17 18:17:59 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:03 2012 -0800" }, "message": "sparc: Kill custom io_remap_pfn_range().\n\n[ Upstream commit 3e37fd3153ac95088a74f5e7c569f7567e9f993a ]\n\nTo handle the large physical addresses, just make a simple wrapper\naround remap_pfn_range() like MIPS does.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "cff6d2096e9a57c2497dd5ee4aed3c97149bfc9e", "tree": "a269f55f7d1043b3cfbbde550e080982be5b8fad", "parents": [ "fde939495571ffd22458e94745b0c2e6af33478d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 17 22:44:58 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:02 2012 -0800" }, "message": "sparc64: Patch sun4v code sequences properly on module load.\n\n[ Upstream commit 0b64120cceb86e93cb1bda0dc055f13016646907 ]\n\nSome of the sun4v code patching occurs in inline functions visible\nto, and usable by, modules.\n\nTherefore we have to patch them up during module load.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "fde939495571ffd22458e94745b0c2e6af33478d", "tree": "cea44545aaf882a8445c5fe9bf2d9a9e9c385ea0", "parents": [ "747b409502fe765784cda1135d806042beddaa89" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 14 10:05:22 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:02 2012 -0800" }, "message": "sparc32: Be less strict in matching %lo part of relocation.\n\n[ Upstream commit b1f44e13a525d2ffb7d5afe2273b7169d6f2222e ]\n\nThe \"(insn \u0026 0x01800000) !\u003d 0x01800000\" test matches \u0027restore\u0027\nbut that is a legitimate place to see the %lo() part of a 32-bit\nsymbol relocation, particularly in tail calls.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nTested-by: Sergei Trofimovich \u003cslyfox@gentoo.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "747b409502fe765784cda1135d806042beddaa89", "tree": "3318a10bc0876a40b107a63dffe0af95f8792b68", "parents": [ "fca54d03a85e883b813255c63bbf11049d2eeb7a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Dec 22 13:23:59 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:14:01 2012 -0800" }, "message": "sparc64: Fix MSIQ HV call ordering in pci_sun4v_msiq_build_irq().\n\n[ Upstream commit 7cc8583372a21d98a23b703ad96cab03180b5030 ]\n\nThis silently was working for many years and stopped working on\nNiagara-T3 machines.\n\nWe need to set the MSIQ to VALID before we can set it\u0027s state to IDLE.\n\nOn Niagara-T3, setting the state to IDLE first was causing HV_EINVAL\nerrors. The hypervisor documentation says, rather ambiguously, that\nthe MSIQ must be \"initialized\" before one can set the state.\n\nI previously understood this to mean merely that a successful setconf()\noperation has been performed on the MSIQ, which we have done at this\npoint. But it seems to also mean that it has been set VALID too.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "4347b837ab999865063649cf801ba4f8d7e8748e", "tree": "a2ddb7c3212dbd98c6de2fdc1fc2b8c798011395", "parents": [ "9267a9e850bf6c52dd7c95f3978dc78a1d3ad1a1" ], "author": { "name": "Vladimir Zapolskiy", "email": "vladimir.zapolskiy@nokia.com", "time": "Thu Dec 22 16:15:40 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:13:57 2012 -0800" }, "message": "oprofile, arm/sh: Fix oprofile_arch_exit() linkage issue\n\ncommit 55205c916e179e09773d98d290334d319f45ac6b upstream.\n\nThis change fixes a linking problem, which happens if oprofile\nis selected to be compiled as built-in:\n\n `oprofile_arch_exit\u0027 referenced in section `.init.text\u0027 of\n arch/arm/oprofile/built-in.o: defined in discarded section\n `.exit.text\u0027 of arch/arm/oprofile/built-in.o\n\nThe problem is appeared after commit 87121ca504, which\nintroduced oprofile_arch_exit() calls from __init function. Note\nthat the aforementioned commit has been backported to stable\nbranches, and the problem is known to be reproduced at least\nwith 3.0.13 and 3.1.5 kernels.\n\nSigned-off-by: Vladimir Zapolskiy \u003cvladimir.zapolskiy@nokia.com\u003e\nSigned-off-by: Robert Richter \u003crobert.richter@amd.com\u003e\nCc: Will Deacon \u003cwill.deacon@arm.com\u003e\nCc: oprofile-list \u003coprofile-list@lists.sourceforge.net\u003e\nLink: http://lkml.kernel.org/r/20111222151540.GB16765@erda.amd.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "8048ac75378918eab0f68ba175915e72ef73e0da", "tree": "6aeb17f8c843fe352836db6f1512058e28764ef5", "parents": [ "f3545737cf06d342d34483b7a8421d0bb90b9c1b" ], "author": { "name": "Jason Chen", "email": "jason.chen@linaro.org", "time": "Mon Dec 19 11:23:28 2011 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:13:56 2012 -0800" }, "message": "ARM:imx:fix pwm period value\n\ncommit 5776ac2eb33164c77cdb4d2b48feee15616eaba3 upstream.\n\nAccording to imx pwm RM, the real period value should be\nPERIOD value in PWMPR plus 2.\n\nPWMO (Hz) \u003d PCLK(Hz) / (period +2)\n\nSigned-off-by: Jason Chen \u003cjason.chen@linaro.org\u003e\nSigned-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "3c681ec96dcc0acd6d5386773b44e6c129919394", "tree": "c4ea80e1369d685c00273087864083b2230adad4", "parents": [ "724c6ae2912dbfc5c6a9e309f92b786fbc141462" ], "author": { "name": "Robert Richter", "email": "robert.richter@amd.com", "time": "Mon Dec 19 16:38:30 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:13:51 2012 -0800" }, "message": "oprofile: Fix uninitialized memory access when writing to writing to oprofilefs\n\ncommit 913050b91eb94f194392dd797b1ff3779f606ac0 upstream.\n\nIf oprofilefs_ulong_from_user() is called with count equals\nzero, *val remains unchanged. Depending on the implementation it\nmight be uninitialized.\n\nChange oprofilefs_ulong_from_user()\u0027s interface to return count\non success. Thus, we are able to return early if count equals\nzero which avoids using *val uninitialized. Fixing all users of\noprofilefs_ulong_ from_user().\n\nThis follows write syscall implementation when count is zero:\n\"If count is zero ... [and if] no errors are detected, 0 will be\nreturned without causing any other effect.\" (man 2 write)\n\nReported-By: Mike Waychison \u003cmikew@google.com\u003e\nSigned-off-by: Robert Richter \u003crobert.richter@amd.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: oprofile-list \u003coprofile-list@lists.sourceforge.net\u003e\nLink: http://lkml.kernel.org/r/20111219153830.GH16765@erda.amd.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "43579d76bdd733c9baf131bbc29719c2a2821569", "tree": "a366b2775bd4c4c75999f6d9d7c7815586f9c5e9", "parents": [ "fcd02ab513c07dfc792fff5baa7ba7e930eba4b0" ], "author": { "name": "Jason Chen", "email": "jason.chen@linaro.org", "time": "Wed Nov 30 11:34:27 2011 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:13:49 2012 -0800" }, "message": "MXC PWM: should active during DOZE/WAIT/DBG mode\n\ncommit c0d96aed8c6dd925afe9ea35491a0cd458642a86 upstream.\n\nSigned-off-by: Jason Chen \u003cjason.chen@linaro.org\u003e\nSigned-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9ce53d23da31815c0d0ae0380f898da4b5aa7af6", "tree": "31bb30fa6e1f80b504e5b84a6367e8ffba6ebbb4", "parents": [ "c6d7c4dbff353eac7919342ae6b3299a378160a6" ], "author": { "name": "Felipe Contreras", "email": "felipe.contreras@gmail.com", "time": "Thu Dec 08 22:23:00 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jan 06 14:13:44 2012 -0800" }, "message": "ARM: OMAP: rx51: fix USB\n\ncommit e5fe29c7198a1f6616286dfc8602a69da165cb3f upstream.\n\nCommit 10299e2e4e3ed3b16503d4e04edd48b33083f4e2 (ARM: RX-51:\nEnable isp1704 power on/off) added power management for isp1704.\n\nHowever, the transceiver should be powered on by default,\notherwise USB doesn\u0027t work at all for networking during\nboot.\n\nAll kernels after v3.0 are affected.\n\nSigned-off-by: Felipe Contreras \u003cfelipe.contreras@gmail.com\u003e\nReviewed-by: Sebastian Reichel \u003csre@debian.org\u003e\n[tony@atomide.com: updated comments]\nSigned-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "37e8c2c97ab911f99cb75c0b5596150a34e06023", "tree": "f4fd3be237f7b80019b76fe495c1a394aaa2500f", "parents": [ "d8091fda7895ed0624d1641089f4080164e75a35" ], "author": { "name": "Ian Campbell", "email": "Ian.Campbell@citrix.com", "time": "Wed Dec 14 12:16:08 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:42 2011 -0800" }, "message": "xen: only limit memory map to maximum reservation for domain 0.\n\ncommit d3db728125c4470a2d061ac10fa7395e18237263 upstream.\n\nd312ae878b6a \"xen: use maximum reservation to limit amount of usable RAM\"\nclamped the total amount of RAM to the current maximum reservation. This is\ncorrect for dom0 but is not correct for guest domains. In order to boot a guest\n\"pre-ballooned\" (e.g. with memory\u003d1G but maxmem\u003d2G) in order to allow for\nfuture memory expansion the guest must derive max_pfn from the e820 provided by\nthe toolstack and not the current maximum reservation (which can reflect only\nthe current maximum, not the guest lifetime max). The existing algorithm\nalready behaves this correctly if we do not artificially limit the maximum\nnumber of pages for the guest case.\n\nFor a guest booted with maxmem\u003d512, memory\u003d128 this results in:\n [ 0.000000] BIOS-provided physical RAM map:\n [ 0.000000] Xen: 0000000000000000 - 00000000000a0000 (usable)\n [ 0.000000] Xen: 00000000000a0000 - 0000000000100000 (reserved)\n-[ 0.000000] Xen: 0000000000100000 - 0000000008100000 (usable)\n-[ 0.000000] Xen: 0000000008100000 - 0000000020800000 (unusable)\n+[ 0.000000] Xen: 0000000000100000 - 0000000020800000 (usable)\n...\n [ 0.000000] NX (Execute Disable) protection: active\n [ 0.000000] DMI not present or invalid.\n [ 0.000000] e820 update range: 0000000000000000 - 0000000000010000 (usable) \u003d\u003d\u003e (reserved)\n [ 0.000000] e820 remove range: 00000000000a0000 - 0000000000100000 (usable)\n-[ 0.000000] last_pfn \u003d 0x8100 max_arch_pfn \u003d 0x1000000\n+[ 0.000000] last_pfn \u003d 0x20800 max_arch_pfn \u003d 0x1000000\n [ 0.000000] initial memory mapped : 0 - 027ff000\n [ 0.000000] Base memory trampoline at [c009f000] 9f000 size 4096\n-[ 0.000000] init_memory_mapping: 0000000000000000-0000000008100000\n-[ 0.000000] 0000000000 - 0008100000 page 4k\n-[ 0.000000] kernel direct mapping tables up to 8100000 @ 27bb000-27ff000\n+[ 0.000000] init_memory_mapping: 0000000000000000-0000000020800000\n+[ 0.000000] 0000000000 - 0020800000 page 4k\n+[ 0.000000] kernel direct mapping tables up to 20800000 @ 26f8000-27ff000\n [ 0.000000] xen: setting RW the range 27e8000 - 27ff000\n [ 0.000000] 0MB HIGHMEM available.\n-[ 0.000000] 129MB LOWMEM available.\n-[ 0.000000] mapped low ram: 0 - 08100000\n-[ 0.000000] low ram: 0 - 08100000\n+[ 0.000000] 520MB LOWMEM available.\n+[ 0.000000] mapped low ram: 0 - 20800000\n+[ 0.000000] low ram: 0 - 20800000\n\nWith this change \"xl mem-set \u003cdomain\u003e 512M\" will successfully increase the\nguest RAM (by reducing the balloon).\n\nThere is no change for dom0.\n\nReported-and-Tested-by: George Shuklin \u003cgeorge.shuklin@gmail.com\u003e\nSigned-off-by: Ian Campbell \u003cian.campbell@citrix.com\u003e\nReviewed-by: David Vrabel \u003cdavid.vrabel@citrix.com\u003e\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "f4347eb6d2aae437d9552a26704bcb07f4626d6c", "tree": "e78b2d81528b5e143d8e17fd0fbf9cf83c56bb7e", "parents": [ "53824693961bf513fe152d35c887e83eb016eaea" ], "author": { "name": "Mark Langsdorf", "email": "mark.langsdorf@amd.com", "time": "Fri Nov 18 16:33:06 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:39 2011 -0800" }, "message": "x86, hpet: Immediately disable HPET timer 1 if rtc irq is masked\n\ncommit 2ded6e6a94c98ea453a156748cb7fabaf39a76b9 upstream.\n\nWhen HPET is operating in RTC mode, the TN_ENABLE bit on timer1\ncontrols whether the HPET or the RTC delivers interrupts to irq8. When\nthe system goes into suspend, the RTC driver sends a signal to the\nHPET driver so that the HPET releases control of irq8, allowing the\nRTC to wake the system from suspend. The switchover is accomplished by\na write to the HPET configuration registers which currently only\noccurs while servicing the HPET interrupt.\n\nOn some systems, I have seen the system suspend before an HPET\ninterrupt occurs, preventing the write to the HPET configuration\nregister and leaving the HPET in control of the irq8. As the HPET is\nnot active during suspend, it does not generate a wake signal and RTC\nalarms do not work.\n\nThis patch forces the HPET driver to immediately transfer control of\nthe irq8 channel to the RTC instead of waiting until the next\ninterrupt event.\n\nSigned-off-by: Mark Langsdorf \u003cmark.langsdorf@amd.com\u003e\nLink: http://lkml.kernel.org/r/20111118153306.GB16319@alberich.amd.com\nTested-by: Andreas Herrmann \u003candreas.herrmann3@amd.com\u003e\nSigned-off-by: Andreas Herrmann \u003candreas.herrmann3@amd.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "cd989fe1fe5da572e45468c6dcb361a7d3c63e5c", "tree": "fb9bc988a4d64bdef3fef30a2500b79198bd8c4b", "parents": [ "0c5a975756c0c111294d728c683d083a9649b541" ], "author": { "name": "Youquan Song", "email": "youquan.song@intel.com", "time": "Thu Dec 08 14:34:16 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:35 2011 -0800" }, "message": "thp: add compound tail page _mapcount when mapped\n\ncommit b6999b19120931ede364fa3b685e698a61fed31d upstream.\n\nWith the 3.2-rc kernel, IOMMU 2M pages in KVM works. But when I tried\nto use IOMMU 1GB pages in KVM, I encountered an oops and the 1GB page\nfailed to be used.\n\nThe root cause is that 1GB page allocation calls gup_huge_pud() while 2M\npage calls gup_huge_pmd. If compound pages are used and the page is a\ntail page, gup_huge_pmd() increases _mapcount to record tail page are\nmapped while gup_huge_pud does not do that.\n\nSo when the mapped page is relesed, it will result in kernel oops\nbecause the page is not marked mapped.\n\nThis patch add tail process for compound page in 1GB huge page which\nkeeps the same process as 2M page.\n\nReproduce like:\n1. Add grub boot option: hugepagesz\u003d1G hugepages\u003d8\n2. mount -t hugetlbfs -o pagesize\u003d1G hugetlbfs /dev/hugepages\n3. qemu-kvm -m 2048 -hda os-kvm.img -cpu kvm64 -smp 4 -mem-path /dev/hugepages\n\t-net none -device pci-assign,host\u003d07:00.1\n\n kernel BUG at mm/swap.c:114!\n invalid opcode: 0000 [#1] SMP\n Call Trace:\n put_page+0x15/0x37\n kvm_release_pfn_clean+0x31/0x36\n kvm_iommu_put_pages+0x94/0xb1\n kvm_iommu_unmap_memslots+0x80/0xb6\n kvm_assign_device+0xba/0x117\n kvm_vm_ioctl_assigned_device+0x301/0xa47\n kvm_vm_ioctl+0x36c/0x3a2\n do_vfs_ioctl+0x49e/0x4e4\n sys_ioctl+0x5a/0x7c\n system_call_fastpath+0x16/0x1b\n RIP put_compound_page+0xd4/0x168\n\nSigned-off-by: Youquan Song \u003cyouquan.song@intel.com\u003e\nReviewed-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "a43cc03a3c0c1f1eb783b4493ecbf15c7638b7f7", "tree": "632a747821eee61533045258c338059cfdd619bc", "parents": [ "ac565b516b8e9b4e8989642f0c41e9d7d230e0da" ], "author": { "name": "Hans Verkuil", "email": "hans.verkuil@cisco.com", "time": "Mon Nov 14 19:20:49 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:34 2011 -0800" }, "message": "ARM: davinci: dm646x evm: wrong register used in setup_vpif_input_channel_mode\n\ncommit 83713fc9373be2e943f82e9d36213708c6b0050e upstream.\n\nThe function setup_vpif_input_channel_mode() used the VSCLKDIS register\ninstead of VIDCLKCTL. This meant that when in HD mode videoport channel 0\nused a different clock from channel 1.\n\nClearly a copy-and-paste error.\n\nSigned-off-by: Hans Verkuil \u003chans.verkuil@cisco.com\u003e\nAcked-by: Manjunath Hadli \u003cmanjunath.hadli@ti.com\u003e\nSigned-off-by: Sekhar Nori \u003cnsekhar@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ac565b516b8e9b4e8989642f0c41e9d7d230e0da", "tree": "5ad69adecc00f329fcdcd7cb84ecacfb5765720a", "parents": [ "5f29f2e19e7c6231559992be25acb2bfb0fa37d9" ], "author": { "name": "Jean-Christophe PLAGNIOL-VILLARD", "email": "plagnioj@jcrosoft.com", "time": "Mon Nov 28 12:53:08 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:34 2011 -0800" }, "message": "ARM: at91: fix clock conid for atmel_tcb.1 on 9260/9g20\n\ncommit 1808958d27b1250295f01dff4997d8a8814adaab upstream.\n\nThe conid is supposed to be t0/t1/t2_clk.\n\nSigned-off-by: Jean-Christophe PLAGNIOL-VILLARD \u003cplagnioj@jcrosoft.com\u003e\nAcked-by: Nicolas Ferre \u003cnicolas.ferre@atmel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "5f29f2e19e7c6231559992be25acb2bfb0fa37d9", "tree": "ef5042cf78b84a3b38d299e9457f4511353f2928", "parents": [ "ff40cdaf4ca551c43c94da75ab99719781de123a" ], "author": { "name": "Wolfram Sang", "email": "w.sang@pengutronix.de", "time": "Sat Nov 26 11:39:14 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:34 2011 -0800" }, "message": "arm: mx23: recognise stmp378x as mx23\n\ncommit 11357be9246c7d1acf9b37ad54a18b29bbb734be upstream.\n\nAdding the machine_is_* line was forgotten when converting mach-stmp378x to\nmach-mxs.\n\nSigned-off-by: Wolfram Sang \u003cw.sang@pengutronix.de\u003e\nSigned-off-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ff40cdaf4ca551c43c94da75ab99719781de123a", "tree": "8cd4147525869bbcbfe3bab8c7ebfa2579f4d059", "parents": [ "612e5dbc550f30c787e40b958d351720e72091c6" ], "author": { "name": "Manjunathappa, Prakash", "email": "prakash.pm@ti.com", "time": "Thu Nov 10 11:43:21 2011 +0530" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Wed Dec 21 12:57:34 2011 -0800" }, "message": "ARM: davinci: da850 evm: change audio edma event queue to EVENTQ_0\n\ncommit f1b21c525693b0159aed83b5871f2d0f077f208e upstream.\n\nOn OMAP-L138 platform, EDMA event queue 0 should be used for audio\ntransfers so that they are not starved by video data moving on event queue 1.\n\nCommit 48519f0ae03bc7e86b3dc93e56f1334d53803770 (ASoC: davinci: let platform\ndata define edma queue numbers) had a side-effect of changing this behavior\nby making the driver actually honor the platform data passed.\n\nFix this now by passing event queue 0 as the queue to be used for audio\ntransfers.\n\nSigned-off-by: Manjunathappa, Prakash \u003cprakash.pm@ti.com\u003e\nSigned-off-by: Sekhar Nori \u003cnsekhar@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "a89c8adbd9813435ccc29699e3dd474c1b823058", "tree": "113677709f0c8427b10f61dbc5561f958f82c8bf", "parents": [ "b3e838f4ee691c0e4f4e1a90d951f2acd63ab0ea" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Thu Dec 01 13:32:17 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:45 2011 -0800" }, "message": "add missing .set function for NT_S390_LAST_BREAK regset\n\ncommit b934069c991355d27a053a932591c77960f4e414 upstream.\n\nThe last breaking event address is a read-only value, the regset misses the\n.set function. If a PTRACE_SETREGSET is done for NT_S390_LAST_BREAK we\nget an oops due to a branch to zero:\n\nKernel BUG at 0000000000000002 verbose debug info unavailable\nillegal operation: 0001 #1 SMP\n...\nCall Trace:\n(\u003c0000000000158294\u003e ptrace_regset+0x184/0x188)\n \u003c00000000001595b6\u003e ptrace_request+0x37a/0x4fc\n \u003c0000000000109a78\u003e arch_ptrace+0x108/0x1fc\n \u003c00000000001590d6\u003e SyS_ptrace+0xaa/0x12c\n \u003c00000000005c7a42\u003e sysc_noemu+0x16/0x1c\n \u003c000003fffd5ec10c\u003e 0x3fffd5ec10c\nLast Breaking-Event-Address:\n \u003c0000000000158242\u003e ptrace_regset+0x132/0x188\n\nAdd a nop .set function to prevent the branch to zero.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "b3e838f4ee691c0e4f4e1a90d951f2acd63ab0ea", "tree": "82dae770c9627c8781cfc0aa84050ed319d88a21", "parents": [ "1e52c65de4b0f35f79ffa0a692f0a1174c7cf7ab" ], "author": { "name": "Robert Richter", "email": "robert.richter@amd.com", "time": "Mon Oct 10 16:21:10 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:44 2011 -0800" }, "message": "oprofile, x86: Fix crash when unloading module (nmi timer mode)\n\ncommit 97f7f8189fe54e3cfe324ef9ad35064f3d2d3bff upstream.\n\nIf oprofile uses the nmi timer interrupt there is a crash while\nunloading the module. The bug can be triggered with oprofile build as\nmodule and kernel parameter nolapic set. This patch fixes this.\n\noprofile: using NMI timer interrupt.\nBUG: unable to handle kernel NULL pointer dereference at 0000000000000008\nIP: [\u003cffffffff8123c226\u003e] unregister_syscore_ops+0x41/0x58\nPGD 42dbca067 PUD 41da6a067 PMD 0\nOops: 0002 [#1] PREEMPT SMP\nCPU 5\nModules linked in: oprofile(-) [last unloaded: oprofile]\n\nPid: 2518, comm: modprobe Not tainted 3.1.0-rc7-00019-gb2fb49d #19 Advanced Micro Device Anaheim/Anaheim\nRIP: 0010:[\u003cffffffff8123c226\u003e] [\u003cffffffff8123c226\u003e] unregister_syscore_ops+0x41/0x58\nRSP: 0018:ffff88041ef71e98 EFLAGS: 00010296\nRAX: 0000000000000000 RBX: ffffffffa0017100 RCX: dead000000200200\nRDX: 0000000000000000 RSI: dead000000100100 RDI: ffffffff8178c620\nRBP: ffff88041ef71ea8 R08: 0000000000000001 R09: 0000000000000082\nR10: 0000000000000000 R11: ffff88041ef71de8 R12: 0000000000000080\nR13: fffffffffffffff5 R14: 0000000000000001 R15: 0000000000610210\nFS: 00007fc902f20700(0000) GS:ffff88042fd40000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b\nCR2: 0000000000000008 CR3: 000000041cdb6000 CR4: 00000000000006e0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400\nProcess modprobe (pid: 2518, threadinfo ffff88041ef70000, task ffff88041d348040)\nStack:\n ffff88041ef71eb8 ffffffffa0017790 ffff88041ef71eb8 ffffffffa0013532\n ffff88041ef71ec8 ffffffffa00132d6 ffff88041ef71ed8 ffffffffa00159b2\n ffff88041ef71f78 ffffffff81073115 656c69666f72706f 0000000000610200\nCall Trace:\n [\u003cffffffffa0013532\u003e] op_nmi_exit+0x15/0x17 [oprofile]\n [\u003cffffffffa00132d6\u003e] oprofile_arch_exit+0xe/0x10 [oprofile]\n [\u003cffffffffa00159b2\u003e] oprofile_exit+0x1e/0x20 [oprofile]\n [\u003cffffffff81073115\u003e] sys_delete_module+0x1c3/0x22f\n [\u003cffffffff811bf09e\u003e] ? trace_hardirqs_on_thunk+0x3a/0x3f\n [\u003cffffffff8148070b\u003e] system_call_fastpath+0x16/0x1b\nCode: 20 c6 78 81 e8 c5 cc 23 00 48 8b 13 48 8b 43 08 48 be 00 01 10 00 00 00 ad de 48 b9 00 02 20 00 00 00 ad de 48 c7 c7 20 c6 78 81\n 89 42 08 48 89 10 48 89 33 48 89 4b 08 e8 a6 c0 23 00 5a 5b\nRIP [\u003cffffffff8123c226\u003e] unregister_syscore_ops+0x41/0x58\n RSP \u003cffff88041ef71e98\u003e\nCR2: 0000000000000008\n---[ end trace 43a541a52956b7b0 ]---\n\nSigned-off-by: Robert Richter \u003crobert.richter@amd.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1e52c65de4b0f35f79ffa0a692f0a1174c7cf7ab", "tree": "f1bbc123d2128b7b037d2d9c205924de988b883b", "parents": [ "347736f0fa8ffc5f89df7bc507a0beb90a5127b5" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Oct 07 13:36:40 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:44 2011 -0800" }, "message": "perf/x86: Fix PEBS instruction unwind\n\ncommit 57d1c0c03c6b48b2b96870d831b9ce6b917f53ac upstream.\n\nMasami spotted that we always try to decode the instruction stream as\n64bit instructions when running a 64bit kernel, this doesn\u0027t work for\nia32-compat proglets.\n\nUse TIF_IA32 to detect if we need to use the 32bit instruction\ndecoder.\n\nReported-by: Masami Hiramatsu \u003cmasami.hiramatsu.pt@hitachi.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "347736f0fa8ffc5f89df7bc507a0beb90a5127b5", "tree": "e58e2739ff6c38d6984ce32bb378c216923ae685", "parents": [ "c060a3d5e9bba4271331b69b9e2c53105999b97f" ], "author": { "name": "Konrad Rzeszutek Wilk", "email": "konrad.wilk@oracle.com", "time": "Tue Nov 15 14:49:09 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:44 2011 -0800" }, "message": "x86/paravirt: PTE updates in k(un)map_atomic need to be synchronous, regardless of lazy_mmu mode\n\ncommit 2cd1c8d4dc7ecca9e9431e2dabe41ae9c7d89e51 upstream.\n\nFix an outstanding issue that has been reported since 2.6.37.\nUnder a heavy loaded machine processing \"fork()\" calls could\ncrash with:\n\nBUG: unable to handle kernel paging request at f573fc8c\nIP: [\u003cc01abc54\u003e] swap_count_continued+0x104/0x180\n*pdpt \u003d 000000002a3b9027 *pde \u003d 0000000001bed067 *pte \u003d 0000000000000000 Oops: 0000 [#1] SMP\nModules linked in:\nPid: 1638, comm: apache2 Not tainted 3.0.4-linode37 #1\nEIP: 0061:[\u003cc01abc54\u003e] EFLAGS: 00210246 CPU: 3\nEIP is at swap_count_continued+0x104/0x180\n.. snip..\nCall Trace:\n [\u003cc01ac222\u003e] ? __swap_duplicate+0xc2/0x160\n [\u003cc01040f7\u003e] ? pte_mfn_to_pfn+0x87/0xe0\n [\u003cc01ac2e4\u003e] ? swap_duplicate+0x14/0x40\n [\u003cc01a0a6b\u003e] ? copy_pte_range+0x45b/0x500\n [\u003cc01a0ca5\u003e] ? copy_page_range+0x195/0x200\n [\u003cc01328c6\u003e] ? dup_mmap+0x1c6/0x2c0\n [\u003cc0132cf8\u003e] ? dup_mm+0xa8/0x130\n [\u003cc013376a\u003e] ? copy_process+0x98a/0xb30\n [\u003cc013395f\u003e] ? do_fork+0x4f/0x280\n [\u003cc01573b3\u003e] ? getnstimeofday+0x43/0x100\n [\u003cc010f770\u003e] ? sys_clone+0x30/0x40\n [\u003cc06c048d\u003e] ? ptregs_clone+0x15/0x48\n [\u003cc06bfb71\u003e] ? syscall_call+0x7/0xb\n\nThe problem is that in copy_page_range() we turn lazy mode on,\nand then in swap_entry_free() we call swap_count_continued()\nwhich ends up in:\n\n map \u003d kmap_atomic(page, KM_USER0) + offset;\n\nand then later we touch *map.\n\nSince we are running in batched mode (lazy) we don\u0027t actually\nset up the PTE mappings and the kmap_atomic is not done\nsynchronously and ends up trying to dereference a page that has\nnot been set.\n\nLooking at kmap_atomic_prot_pfn(), it uses\n\u0027arch_flush_lazy_mmu_mode\u0027 and doing the same in\nkmap_atomic_prot() and __kunmap_atomic() makes the problem go\naway.\n\nInterestingly, commit b8bcfe997e4615 (\"x86/paravirt: remove lazy\nmode in interrupts\") removed part of this to fix an interrupt\nissue - but it went to far and did not consider this scenario.\n\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Jeremy Fitzhardinge \u003cjeremy.fitzhardinge@citrix.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "c060a3d5e9bba4271331b69b9e2c53105999b97f", "tree": "6b771a9f4cb64a0e1a136e9d7cd4751740646170", "parents": [ "a1b8e912f4a2381f8d38996a87b183bce59f1920" ], "author": { "name": "Peter Chubb", "email": "peter.chubb@nicta.com.au", "time": "Mon Dec 05 16:53:53 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:43 2011 -0800" }, "message": "x86: Fix \"Acer Aspire 1\" reboot hang\n\ncommit 1ef03890969932e9359b9a4c658f7f87771910ac upstream.\n\nLooks like on some Acer Aspire 1s with older bioses, reboot via bios\nfails. It works on my machine, (with BIOS version 0.3310) but\nnot on some others (BIOS version 0.3309).\n\nThere\u0027s a log of problems at:\n\n https://bbs.archlinux.org/viewtopic.php?id\u003d124136\n\nThis patch adds a different callback to the reboot quirk table,\nto allow rebooting via keybaord controller.\n\nReported-by: Uroš Vampl \u003cmobile.leecher@gmail.com\u003e\nTested-by: Vasily Khoruzhick \u003canarsoul@gmail.com\u003e\nSigned-off-by: Peter Chubb \u003cpeter.chubb@nicta.com.au\u003e\nCc: Don Zickus \u003cdzickus@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLink: http://lkml.kernel.org/r/1323093233-9481-1-git-send-email-anarsoul@gmail.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "a1b8e912f4a2381f8d38996a87b183bce59f1920", "tree": "95201a26338544ac0ec7a7c13ad760d90483cf8b", "parents": [ "1f076488aa92f7e26d34f385d3b4f738f975e2ff" ], "author": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Sun Sep 25 15:29:00 2011 -0600" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:43 2011 -0800" }, "message": "x86/mpparse: Account for bus types other than ISA and PCI\n\ncommit 9e6866686bdf2dcf3aeb0838076237ede532dcc8 upstream.\n\nIn commit f8924e770e04 (\"x86: unify mp_bus_info\"), the 32-bit\nand 64-bit versions of MP_bus_info were rearranged to match each\nother better. Unfortunately it introduced a regression: prior\nto that change we used to always set the mp_bus_not_pci bit,\nthen clear it if we found a PCI bus. After it, we set\nmp_bus_not_pci for ISA buses, clear it for PCI buses, and leave\nit alone otherwise.\n\nIn the cases of ISA and PCI, there\u0027s not much difference. But\nISA is not the only non-PCI bus, so it\u0027s better to always set\nmp_bus_not_pci and clear it only for PCI.\n\nWithout this change, Dan\u0027s Dell PowerEdge 4200 panics on boot\nwith a log indicating interrupt routing trouble unless the\n\"noapic\" option is supplied. With this change, the machine\nboots reliably without \"noapic\".\n\nFixes http://bugs.debian.org/586494\n\nReported-bisected-and-tested-by: Dan McGrath \u003ctroubledaemon@gmail.com\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nCc: Dan McGrath \u003ctroubledaemon@gmail.com\u003e\nCc: Alexey Starikovskiy \u003caystarik@gmail.com\u003e\n[jrnieder@gmail.com: clarified commit message]\nSigned-off-by: Jonathan Nieder \u003cjrnieder@gmail.com\u003e\nLink: http://lkml.kernel.org/r/20111122215000.GA9151@elie.hsd1.il.comcast.net\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1f076488aa92f7e26d34f385d3b4f738f975e2ff", "tree": "99ef1308a1dffe2485184df1f177b2dd8e5cbf29", "parents": [ "15f2701a1eaf135eaab75987e4517daf5d88b880" ], "author": { "name": "Salman Qazi", "email": "sqazi@google.com", "time": "Tue Nov 15 14:12:06 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:43 2011 -0800" }, "message": "sched, x86: Avoid unnecessary overflow in sched_clock\n\ncommit 4cecf6d401a01d054afc1e5f605bcbfe553cb9b9 upstream.\n\n(Added the missing signed-off-by line)\n\nIn hundreds of days, the __cycles_2_ns calculation in sched_clock\nhas an overflow. cyc * per_cpu(cyc2ns, cpu) exceeds 64 bits, causing\nthe final value to become zero. We can solve this without losing\nany precision.\n\nWe can decompose TSC into quotient and remainder of division by the\nscale factor, and then use this to convert TSC into nanoseconds.\n\nSigned-off-by: Salman Qazi \u003csqazi@google.com\u003e\nAcked-by: John Stultz \u003cjohnstul@us.ibm.com\u003e\nReviewed-by: Paul Turner \u003cpjt@google.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLink: http://lkml.kernel.org/r/20111115221121.7262.88871.stgit@dungbeetle.mtv.corp.google.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "341f278d7bfbe7233438a8512d183d62ddfae34e", "tree": "c22524b513fb146f44760a4d7a28f2903e1fd050", "parents": [ "0443648aa178dbf8090edfd6dc54f31f916a174b" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Mon Nov 14 17:24:58 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:25 2011 -0800" }, "message": "ARM: 7161/1: errata: no automatic store buffer drain\n\ncommit 11ed0ba1754841316d4095478944300acf19acc3 upstream.\n\nThis patch implements a workaround for PL310 erratum 769419. On\nrevisions of the PL310 prior to r3p2, the Store Buffer does not\nautomatically drain. This can cause normal, non-cacheable writes to be\nretained when the memory system is idle, leading to suboptimal I/O\nperformance for drivers using coherent DMA.\n\nThis patch adds an optional wmb() call to the cpu_idle loop. On systems\nwith an outer cache, this causes an explicit flush of the store buffer.\n\nAcked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nTested-by: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0443648aa178dbf8090edfd6dc54f31f916a174b", "tree": "1d54652baef10a41e49e25776429c15e61254a81", "parents": [ "e2dcdeea518813b728216fe52d94c3b0a5a0b599" ], "author": { "name": "Ming Lei", "email": "tom.leiming@gmail.com", "time": "Wed Nov 23 14:44:50 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:24 2011 -0800" }, "message": "ARM: OMAP2: select ARM_AMBA if OMAP3_EMU is defined\n\ncommit a8a6565c7615cab3608d75af95b5c8a3522cd7c4 upstream.\n\nThis patch selects ARM_AMBA if OMAP3_EMU is defined because\nOC_ETM depends on ARM_AMBA, so fix the link failure[1].\n\n[1],\narch/arm/kernel/built-in.o: In function `etm_remove\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:609: undefined\nreference to `amba_release_regions\u0027\narch/arm/kernel/built-in.o: In function `etb_remove\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:409: undefined\nreference to `amba_release_regions\u0027\narch/arm/kernel/built-in.o: In function `etm_init\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:640: undefined\nreference to `amba_driver_register\u0027\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:646: undefined\nreference to `amba_driver_register\u0027\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:648: undefined\nreference to `amba_driver_unregister\u0027\narch/arm/kernel/built-in.o: In function `etm_probe\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:545: undefined\nreference to `amba_request_regions\u0027\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:595: undefined\nreference to `amba_release_regions\u0027\narch/arm/kernel/built-in.o: In function `etb_probe\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:347: undefined\nreference to `amba_request_regions\u0027\n/home/tom/git/omap/linux-2.6-omap/arch/arm/kernel/etm.c:392: undefined\nreference to `amba_release_regions\u0027\narch/arm/mach-omap2/built-in.o: In function `emu_init\u0027:\n/home/tom/git/omap/linux-2.6-omap/arch/arm/mach-omap2/emu.c:62:\nundefined reference to `amba_device_register\u0027\n/home/tom/git/omap/linux-2.6-omap/arch/arm/mach-omap2/emu.c:63:\nundefined reference to `amba_device_register\u0027\nmake: *** [.tmp_vmlinux1] Error 1\nmaking modules\n\nSigned-off-by: Ming Lei \u003ctom.leiming@gmail.com\u003e\nSigned-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "e2dcdeea518813b728216fe52d94c3b0a5a0b599", "tree": "3a3ac9d1d8e5ff757dc8f021b2943d622dce2811", "parents": [ "051a11b53377e312462be9423c083adfc44d600c" ], "author": { "name": "Felipe Balbi", "email": "balbi@ti.com", "time": "Wed Nov 23 14:43:37 2011 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:24 2011 -0800" }, "message": "ARM: OMAP: smartreflex: fix IRQ handling bug\n\ncommit 5a4f1844c2ba21f804d7729306d9b16eaeb724a8 upstream.\n\nFix a bug which has been on this driver since\nit was added by the original commit 984aa6db\nwhich would never clear IRQSTATUS bits.\n\nSigned-off-by: Felipe Balbi \u003cbalbi@ti.com\u003e\nSigned-off-by: Kevin Hilman \u003ckhilman@ti.com\u003e\nSigned-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "051a11b53377e312462be9423c083adfc44d600c", "tree": "d8326cbcef9d54d13fbf60fb387fd75ced8115da", "parents": [ "bfbcb8185dfbb5a9ecaa09bb816a96740e104c0f" ], "author": { "name": "Wolfram Sang", "email": "w.sang@pengutronix.de", "time": "Sat Sep 10 12:26:07 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:24 2011 -0800" }, "message": "arm: mx28: fix bit operation in clock setting\n\ncommit c2735391fbc68feae10d6d14e60956c8106e725f upstream.\n\nreg | (1 \u003c\u003c clk-\u003eenable_shift) always evaluates to true. Switch it\nto \u0026 which makes much more sense. Same fix as 13be9f00 (ARM i.MX28: fix\nbit operation) at a different location.\n\nSigned-off-by: Wolfram Sang \u003cw.sang@pengutronix.de\u003e\nCc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nCc: Shawn Guo \u003cshawn.guo@freescale.com\u003e\nSigned-off-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "bfbcb8185dfbb5a9ecaa09bb816a96740e104c0f", "tree": "eaa54ee9671c130d68f3d1a494d8cdd22cdbe13a", "parents": [ "47b52de3faa57f5849c71bdaee4cd041debcc1ff" ], "author": { "name": "Haojian Zhuang", "email": "haojian.zhuang@marvell.com", "time": "Thu Nov 10 07:13:07 2011 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Dec 09 08:52:24 2011 -0800" }, "message": "ARM: pxa: fix inconsistent CONFIG_USB_PXA27X\n\ncommit c0a39151a4055332897cba615623d3de2f3896df upstream.\n\nSince CONFIG_USB_GADGET_PXA27X and other macros are renamed to\nCONFIG_USB_PXA27X. Update them in arch/arm/mach-pxa and arch/arm/configs\nto keep consistent.\n\nSigned-off-by: Haojian Zhuang \u003chaojian.zhuang@marvell.com\u003e\nAcked-by: Felipe Balbi \u003cbalbi@ti.com\u003e\nSigned-off-by: Eric Miao \u003ceric.y.miao@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "a633bc89bb5626c6d700c3fc5c3bee01da79ce47", "tree": "6f3b8d27417dedb4ea8e769064a74c9f82929709", "parents": [ "6e99164ee37d513fb99c7b941e3eecbcd8ae8573" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Mon Nov 07 18:05:53 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat Nov 26 09:09:53 2011 -0800" }, "message": "ARM: 7150/1: Allow kernel unaligned accesses on ARMv6+ processors\n\ncommit 8428e84d42179c2a00f5f6450866e70d802d1d05 upstream.\n\nRecent gcc versions generate unaligned accesses by default on ARMv6 and\nlater processors. This patch ensures that the SCTLR.A bit is always\ncleared on such processors to avoid kernel traping before\nalignment_init() is called.\n\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nTested-by: John Linn \u003cJohn.Linn@xilinx.com\u003e\nAcked-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "214af5d355ce204cd0bcf630245cdc37853cb484", "tree": "bcd9f58dc10224eb5fad1d425410d6fdb5f97954", "parents": [ "9db74684bd2f8261cef34d2398f32d48c8ad8e2f" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Mon Nov 14 12:54:47 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:25 2011 -0800" }, "message": "powerpc: Copy down exception vectors after feature fixups\n\ncommit d715e433b7ad19c02fc4becf0d5e9a59f97925de upstream.\n\nkdump fails because we try to execute an HV only instruction. Feature\nfixups are being applied after we copy the exception vectors down to 0\nso they miss out on any updates.\n\nWe have always had this issue but it only became critical in v3.0\nwhen we added CFAR support (breaks POWER5) and v3.1 when we added\nPOWERNV (breaks everyone).\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9db74684bd2f8261cef34d2398f32d48c8ad8e2f", "tree": "b6848266e7c3985c55e05e1d44aafa8afa1f1211", "parents": [ "015be9f48be43814f14326555e9ea576cf0343d7" ], "author": { "name": "Geoff Levand", "email": "geoff@infradead.org", "time": "Tue Nov 08 12:37:26 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:25 2011 -0800" }, "message": "powerpc/ps3: Fix lost SMP IPIs\n\ncommit 72f3bea075287785ed32b777b6dd2636aa7002e8 upstream.\n\nFixes the PS3 bootup hang introduced in 3.0-rc1 by:\n\n commit 317f394160e9beb97d19a84c39b7e5eb3d7815a\n sched: Move the second half of ttwu() to the remote cpu\n\nMove the PS3\u0027s LV1 EOI call lv1_end_of_interrupt_ext() from ps3_chip_eoi()\nto ps3_get_irq() for IPI messages.\n\nIf lv1_send_event_locally() is called between a previous call to\nlv1_send_event_locally() and the coresponding call to\nlv1_end_of_interrupt_ext() the second event will not be delivered to the\ntarget cpu.\n\nThe PS3\u0027s SMP IPIs are implemented using lv1_send_event_locally(), so if two\nIPI messages of the same type are sent to the same target in a relatively\nshort period of time the second IPI event can become lost when\nlv1_end_of_interrupt_ext() is called from ps3_chip_eoi().\n\nSigned-off-by: Geoff Levand \u003cgeoff@infradead.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "66e85a1675dc6d4e05e0fd152ef1ad485b3fb985", "tree": "4d0b23c334830c32f1725e9dc1642a838ba45362", "parents": [ "11b8fc6ae54bf18a48c94e181c37ca135b858b42" ], "author": { "name": "Zhenzhong Duan", "email": "zhenzhong.duan@oracle.com", "time": "Thu Oct 27 22:28:59 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:23 2011 -0800" }, "message": "xen:pvhvm: enable PVHVM VCPU placement when using more than 32 CPUs.\n\ncommit 90d4f5534d14815bd94c10e8ceccc57287657ecc upstream.\n\nPVHVM running with more than 32 vcpus and pv_irq/pv_time enabled\nneed VCPU placement to work, or else it will softlockup.\n\nAcked-by: Stefano Stabellini \u003cstefano.stabellini@eu.citrix.com\u003e\nSigned-off-by: Zhenzhong Duan \u003czhenzhong.duan@oracle.com\u003e\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "54e6e8d507358facee8b7f5dca27231914676fd8", "tree": "4f3d1a345ca81cbccd1f78db8ed6b2b51590410a", "parents": [ "03ff90c0f9c9ea4ab0840ac21a393da859784d4d" ], "author": { "name": "Mika Westerberg", "email": "mika.westerberg@linux.intel.com", "time": "Thu Oct 13 12:04:20 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:21 2011 -0800" }, "message": "x86, mrst: use a temporary variable for SFI irq\n\ncommit 153b19a3b9fd8b9478495b9ee1f93f6a77c564f9 upstream.\n\nSFI tables reside in RAM and should not be modified once they are\nwritten. Current code went to set pentry-\u003eirq to zero which causes\nsubsequent reads to fail with invalid SFI table checksum. This will\nbreak kexec as the second kernel fails to validate SFI tables.\n\nTo fix this we use temporary variable for irq number.\n\nSigned-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nReviewed-by: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "03ff90c0f9c9ea4ab0840ac21a393da859784d4d", "tree": "e849164b2cfa2464a583f68e94e0f3891232fe2a", "parents": [ "987ccebc15b3f1256dbcdfbb81ad542d1e2ad6a8" ], "author": { "name": "Kirill A. Shutemov", "email": "kirill.shutemov@linux.intel.com", "time": "Fri Aug 26 12:20:59 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:20 2011 -0800" }, "message": "sfi: table irq 0xFF means \u0027no interrupt\u0027\n\ncommit a94cc4e6c0a26a7c8f79a432ab2c89534aa674d5 upstream.\n\nAccording to the SFI specification irq number 0xFF means device has no\ninterrupt or interrupt attached via GPIO.\n\nCurrently, we don\u0027t handle this special case and set irq field in\n*_board_info structs to 255. It leads to confusion in some drivers.\nAccelerometer driver tries to register interrupt 255, fails and prints\n\"Cannot get IRQ\" to dmesg.\n\nSigned-off-by: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nSigned-off-by: Alan Cox \u003calan@linux.intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "f5116ff776a68277c0b60ba57e74c7074b72ab65", "tree": "67f567e8f27cde1f4dcbba22325963e5161e7bb3", "parents": [ "6c43a5e42ca88c5770de53fb969a0b6f04d86982" ], "author": { "name": "Nobuhiro Iwamatsu", "email": "nobuhiro.iwamatsu.yj@renesas.com", "time": "Fri Nov 04 22:13:50 2011 +0900" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Nov 21 14:31:15 2011 -0800" }, "message": "sh: Fix cached/uncaced address calculation in 29bit mode\n\ncommit dfd3b596fbbfa48b8e7966ef996d587157554b69 upstream.\n\nIn the case of 29bit mode, CAC/UNCAC_ADDR does not return a right address.\nThis revises this problem by using P1SEGADDR and P2SEGADDR in 29bit mode.\n\nReported-by: Yutaro Ebihara \u003cebiharaml@si-linux.co.jp\u003e\nSigned-off-by: Nobuhiro Iwamatsu \u003cnobuhiro.iwamatsu.yj@renesas.com\u003e\nTested-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nTested-by: Simon Horman \u003chorms@verge.net.au\u003e\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "eaeafcd898108e4e026faf43ca41817f5de6ee66", "tree": "363bec39cbe790f0aa86c657214753ad4808e270", "parents": [ "71b1a9d35d6cbd2e37c1101ca5f779ca55f7fb03" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Sep 14 09:43:15 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:37:11 2011 -0800" }, "message": "powerpc: Fix deadlock in icswx code\n\ncommit 8bdafa39a47265bc029838b35cc6585f69224afa upstream.\n\nThe icswx code introduced an A-B B-A deadlock:\n\n CPU0 CPU1\n ---- ----\nlock(\u0026anon_vma-\u003emutex);\n lock(\u0026mm-\u003emmap_sem);\n lock(\u0026anon_vma-\u003emutex);\nlock(\u0026mm-\u003emmap_sem);\n\nInstead of using the mmap_sem to keep mm_users constant, take the\npage table spinlock.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "71b1a9d35d6cbd2e37c1101ca5f779ca55f7fb03", "tree": "0a09a33b08cae40a6b1d50999e22bfa5d06e3f00", "parents": [ "747613961425127da9e14e020500b184d4f7a42c" ], "author": { "name": "Thadeu Lima de Souza Cascardo", "email": "cascardo@linux.vnet.ibm.com", "time": "Fri Aug 26 10:36:31 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:37:11 2011 -0800" }, "message": "powerpc/eeh: Fix /proc/ppc64/eeh creation\n\ncommit 8feaa43494cee5e938fd5a57b9e9bf1c827e6ccd upstream.\n\nSince commit 188917e183cf9ad0374b571006d0fc6d48a7f447, /proc/ppc64 is a\nsymlink to /proc/powerpc/. That means that creating /proc/ppc64/eeh will\nend up with a unaccessible file, that is not listed under /proc/powerpc/\nand, then, not listed under /proc/ppc64/.\n\nCreating /proc/powerpc/eeh fixes that problem and maintain the\ncompatibility intended with the ppc64 symlink.\n\nSigned-off-by: Thadeu Lima de Souza Cascardo \u003ccascardo@linux.vnet.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "747613961425127da9e14e020500b184d4f7a42c", "tree": "20281c1591ada653853288b8ca2775af6ff1787a", "parents": [ "55a6bcf9d9a66921b7a7a854ca86356e901114fe" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Sun Aug 14 14:30:30 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:37:10 2011 -0800" }, "message": "powerpc/pseries: Avoid spurious error during hotplug CPU add\n\ncommit 9c740025c51a26ab00192cfc464064d4ccbfe3fc upstream.\n\nDuring hotplug CPU add we get the following error:\n\nUnexpected Error (0) returned from configure-connector\n\nibm,configure-connector returns 0 for configuration complete, so\ncatch this and avoid the error.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "55a6bcf9d9a66921b7a7a854ca86356e901114fe", "tree": "6f98695256440dcc786672daa11e796b3cba2cee", "parents": [ "2b7940d84f184ad4b71cf66465bb55212d3b3fd8" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Aug 10 20:44:24 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:37:10 2011 -0800" }, "message": "powerpc: Fix oops when echoing bad values to /sys/devices/system/memory/probe\n\ncommit a11940978bd598e65996b4f807cf4904793f7025 upstream.\n\nIf we echo an address the hypervisor doesn\u0027t like to\n/sys/devices/system/memory/probe we oops the box:\n\n# echo 0x10000000000 \u003e /sys/devices/system/memory/probe\n\nkernel BUG at arch/powerpc/mm/hash_utils_64.c:541!\n\nThe backtrace is:\n\ncreate_section_mapping\narch_add_memory\nadd_memory\nmemory_probe_store\nsysdev_class_store\nsysfs_write_file\nvfs_write\nSyS_write\n\nIn create_section_mapping we BUG if htab_bolt_mapping returned\nan error. A better approach is to return an error which will\npropagate back to userspace.\n\nRerunning the test with this patch applied:\n\n# echo 0x10000000000 \u003e /sys/devices/system/memory/probe\n-bash: echo: write error: Invalid argument\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "2b7940d84f184ad4b71cf66465bb55212d3b3fd8", "tree": "eb3914e63df8853e52c076dc331fc749aedb7afa", "parents": [ "656460cd7714653cf60a10c79e02cbfe2d9bf732" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Aug 10 20:44:21 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:37:09 2011 -0800" }, "message": "powerpc/numa: Remove double of_node_put in hot_add_node_scn_to_nid\n\ncommit 6083184269fd723affca4f6340e491950267622a upstream.\n\nDuring memory hotplug testing, I got the following warning:\n\nERROR: Bad of_node_put() on /memory@0\n\nof_node_release\nkref_put\nof_node_put\nof_find_node_by_type\nhot_add_node_scn_to_nid\nhot_add_scn_to_nid\nmemory_add_physaddr_to_nid\n...\n\nof_find_node_by_type() loop does the of_node_put for us so we only\nneed the handle the case where we terminate the loop early.\n\nAs suggested by Stephen Rothwell we can do the of_node_put\nunconditionally outside of the loop since of_node_put handles a\nNULL argument fine.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "621112ec06bbe3a6dbb6af5f4db3451d01b309f9", "tree": "1b8dbf57854f8bee13a88a7492d7ba8ca93ceb8c", "parents": [ "1a2417c30c86961992afae7eb8f803acca2f6b9b" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:36 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:48 2011 -0800" }, "message": "thp: share get_huge_page_tail()\n\ncommit b35a35b556f5e6b7993ad0baf20173e75c09ce8c upstream.\n\nThis avoids duplicating the function in every arch gup_fast.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1a2417c30c86961992afae7eb8f803acca2f6b9b", "tree": "ddadf43521cc0acf2d0409b8fcd1320b0b936d78", "parents": [ "7a093bf2f9356230afccb4b79091787feb5aed97" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:28 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:47 2011 -0800" }, "message": "s390: gup_huge_pmd() return 0 if pte changes\n\ncommit 0693bc9ce2cc4f6a1b9c3c05790fc149a74c0b87 upstream.\n\ns390 didn\u0027t return 0 in that case, if it\u0027s rolling back the *nr pointer it\nshould also return zero to avoid adding pages to the array at the wrong\noffset.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "7a093bf2f9356230afccb4b79091787feb5aed97", "tree": "b5cedbe1592a2df1c2effc9d21d48db7d523deea", "parents": [ "2429d1c2f20466effdc70ced613f591f8675059f" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:25 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:47 2011 -0800" }, "message": "s390: gup_huge_pmd() support THP tail recounting\n\ncommit 220a2eb228d032acde60e9fd044ca802706ff583 upstream.\n\nUp to this point the code assumed old refcounting for hugepages (pre-thp).\nThis updates the code directly to the thp mapcount tail page refcounting.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "2429d1c2f20466effdc70ced613f591f8675059f", "tree": "84471f244f56df8b482bf7c98959ebe6e0607aed", "parents": [ "87c2bd66eef0c5685bb29fd995f6b77fa4d07ed1" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:19 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:46 2011 -0800" }, "message": "powerpc: gup_huge_pmd() return 0 if pte changes\n\ncommit cf592bf768c4fa40282b8fce58a80820065de2cb upstream.\n\npowerpc didn\u0027t return 0 in that case, if it\u0027s rolling back the *nr pointer\nit should also return zero to avoid adding pages to the array at the wrong\noffset.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "87c2bd66eef0c5685bb29fd995f6b77fa4d07ed1", "tree": "99015d0281ec26b90fa4cd5e6004a22c35578685", "parents": [ "cf2f493ba5c49afe9efd3afa02a505e9b060832d" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:15 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:46 2011 -0800" }, "message": "powerpc: gup_hugepte() support THP based tail recounting\n\ncommit 3526741f0964c88bc2ce511e1078359052bf225b upstream.\n\nUp to this point the code assumed old refcounting for hugepages (pre-thp).\nThis updates the code directly to the thp mapcount tail page refcounting.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "cf2f493ba5c49afe9efd3afa02a505e9b060832d", "tree": "ae5d2e241aba0b9ae102dc3c5bf4d7bd22117cc8", "parents": [ "acafc792476742f042e2d7a3679dbaa097444ccc" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:11 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:46 2011 -0800" }, "message": "powerpc: gup_hugepte() avoid freeing the head page too many times\n\ncommit 8596468487e2062cae2aad56e973784e03959245 upstream.\n\nWe only taken \"refs\" pins on the head page not \"*nr\" pins.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "acafc792476742f042e2d7a3679dbaa097444ccc", "tree": "aff51c5740d08196da5e7f456265bd46f7488c1a", "parents": [ "45a2755f0fd17acf45c7cea183815c65f28154c3" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:08 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:45 2011 -0800" }, "message": "powerpc: get_hugepte() don\u0027t put_page() the wrong page\n\ncommit 405e44f2e312dd5dd63e5a9f459bffcbcd4368ef upstream.\n\n\"page\" may have changed to point to the next hugepage after the loop\ncompleted, The references have been taken on the head page, so the\nput_page must happen there too.\n\nThis is a longstanding issue pre-thp inclusion.\n\nIt\u0027s totally unclear how these page_cache_add_speculative and\npte_val(pte) !\u003d pte_val(*ptep) checks are necessary across all the\npowerpc gup_fast code, when x86 doesn\u0027t need any of that: there\u0027s no way\nthe page can be freed with irq disabled so we\u0027re guaranteed the\natomic_inc will happen on a page with page_count \u003e 0 (so not needing the\nspeculative check).\n\nThe pte check is also meaningless on x86: no need to rollback on x86 if\nthe pte changed, because the pte can still change a CPU tick after the\ncheck succeeded and it won\u0027t be rolled back in that case. The important\nthing is we got a reference on a valid page that was mapped there a CPU\ntick ago. So not knowing the soft tlb refill code of ppc64 in great\ndetail I\u0027m not removing the \"speculative\" page_count increase and the\npte checks across all the code, but unless there\u0027s a strong reason for\nit they should be later cleaned up too.\n\nIf a pte can change from huge to non-huge (like it could happen with\nTHP) passing a pte_t *ptep to gup_hugepte() would also require to repeat\nthe is_hugepd in gup_hugepte(), but that shouldn\u0027t happen with hugetlbfs\nonly so I\u0027m not altering that.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "45a2755f0fd17acf45c7cea183815c65f28154c3", "tree": "3b9347732e79a4150560a6e92270387aa5bdee3b", "parents": [ "8adc3d3df0562b8dc4008f458081dcc2d8b98863" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:37:03 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:45 2011 -0800" }, "message": "powerpc: remove superfluous PageTail checks on the pte gup_fast\n\ncommit 2839bdc1bfc0af76a2f0f11eca011590520a04fa upstream.\n\nThis part of gup_fast doesn\u0027t seem capable of handling hugetlbfs ptes,\nthose should be handled by gup_hugepd only, so these checks are\nsuperfluous.\n\nPlus if this wasn\u0027t a noop, it would have oopsed because, the insistence\nof using the speculative refcounting would trigger a VM_BUG_ON if a tail\npage was encountered in the page_cache_get_speculative().\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "68fe9d9c796303de600dbc622086768ca4d8408b", "tree": "4d7c0a4d52960df61e3c61e4a894a182de5ae839", "parents": [ "a00fb1451d630898c1380349ac3776688e58010f" ], "author": { "name": "Andrea Arcangeli", "email": "aarcange@redhat.com", "time": "Wed Nov 02 13:36:59 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:29 2011 -0800" }, "message": "mm: thp: tail page refcounting fix\n\ncommit 70b50f94f1644e2aa7cb374819cfd93f3c28d725 upstream.\n\nMichel while working on the working set estimation code, noticed that\ncalling get_page_unless_zero() on a random pfn_to_page(random_pfn)\nwasn\u0027t safe, if the pfn ended up being a tail page of a transparent\nhugepage under splitting by __split_huge_page_refcount().\n\nHe then found the problem could also theoretically materialize with\npage_cache_get_speculative() during the speculative radix tree lookups\nthat uses get_page_unless_zero() in SMP if the radix tree page is freed\nand reallocated and get_user_pages is called on it before\npage_cache_get_speculative has a chance to call get_page_unless_zero().\n\nSo the best way to fix the problem is to keep page_tail-\u003e_count zero at\nall times. This will guarantee that get_page_unless_zero() can never\nsucceed on any tail page. page_tail-\u003e_mapcount is guaranteed zero and\nis unused for all tail pages of a compound page, so we can simply\naccount the tail page references there and transfer them to\ntail_page-\u003e_count in __split_huge_page_refcount() (in addition to the\nhead_page-\u003e_mapcount).\n\nWhile debugging this s/_count/_mapcount/ change I also noticed get_page is\ncalled by direct-io.c on pages returned by get_user_pages. That wasn\u0027t\nentirely safe because the two atomic_inc in get_page weren\u0027t atomic. As\nopposed to other get_user_page users like secondary-MMU page fault to\nestablish the shadow pagetables would never call any superflous get_page\nafter get_user_page returns. It\u0027s safer to make get_page universally safe\nfor tail pages and to use get_page_foll() within follow_page (inside\nget_user_pages()). get_page_foll() is safe to do the refcounting for tail\npages without taking any locks because it is run within PT lock protected\ncritical sections (PT lock for pte and page_table_lock for\npmd_trans_huge).\n\nThe standard get_page() as invoked by direct-io instead will now take\nthe compound_lock but still only for tail pages. The direct-io paths\nare usually I/O bound and the compound_lock is per THP so very\nfinegrined, so there\u0027s no risk of scalability issues with it. A simple\ndirect-io benchmarks with all lockdep prove locking and spinlock\ndebugging infrastructure enabled shows identical performance and no\noverhead. So it\u0027s worth it. Ideally direct-io should stop calling\nget_page() on pages returned by get_user_pages(). The spinlock in\nget_page() is already optimized away for no-THP builds but doing\nget_page() on tail pages returned by GUP is generally a rare operation\nand usually only run in I/O paths.\n\nThis new refcounting on page_tail-\u003e_mapcount in addition to avoiding new\nRCU critical sections will also allow the working set estimation code to\nwork without any further complexity associated to the tail page\nrefcounting with THP.\n\nSigned-off-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nReported-by: Michel Lespinasse \u003cwalken@google.com\u003e\nReviewed-by: Michel Lespinasse \u003cwalken@google.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003cjweiner@redhat.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "5f0d9e03e7e0792fbd36ba9fb9859f0ef7956a20", "tree": "cd9ef9985579e0a316808981ba2d00c834eff8d2", "parents": [ "73e2fc5d34007574d25d00e7088bbd42e988bb3e" ], "author": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Wed Nov 02 13:17:27 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:19 2011 -0800" }, "message": "um: fix ubd cow size\n\ncommit 8535639810e578960233ad39def3ac2157b0c3ec upstream.\n\nubd_file_size() cannot use ubd_dev-\u003ecow.file because at this time\nubd_dev-\u003ecow.file is not initialized.\nTherefore, ubd_file_size() will always report a wrong disk size when\nCOW files are used.\nReading from /dev/ubd* would crash the kernel.\n\nWe have to read the correct disk size from the COW file\u0027s backing\nfile.\n\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ad417042d59a26bedf600ab162b20b79fdcd1fa1", "tree": "ebb8f524f2e56310c7c3adacfa78b5ec1a16d5d5", "parents": [ "513c4eb162aecea92e0b05a7c5e206474526eda2" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Fri Aug 12 13:54:42 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:17 2011 -0800" }, "message": "ARM: mach-ux500: unlock I\u0026D l2x0 caches before init\n\ncommit 1bf6d2c1bb23533af6930581cc39b74685bc29de upstream.\n\nApparently U8500 U-Boot versions may leave the l2x0 locked down\nbefore executing the kernel. Make sure we unlock it before we\ninitialize the l2x0. This fixes a performance problem reported\nby Jan Rinze.\n\nThe l2x0 core has been modified to unlock the l2x0 by default,\nbut it will not touch the locking registers if the l2x0 was\nalready enabled, as on the ux500, so we need this quirk to\nmake sure it is properly turned off.\n\nCc: Srinidhi Kasagar \u003csrinidhi.kasagar@stericsson.com\u003e\nCc: Rabin Vincent \u003crabin.vincent@stericsson.com\u003e\nCc: Adrian Bunk \u003cadrian.bunk@movial.com\u003e\nReported-by: Jan Rinze \u003cjanrinze@gmail.com\u003e\nTested-by: Robert Marklund \u003crobert.marklund@stericsson.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "513c4eb162aecea92e0b05a7c5e206474526eda2", "tree": "731dc0326955d5d568a13416799c7ff92d08b2fc", "parents": [ "b5e074a6ff4cfa090f7aa8a87f5079d19206d91c" ], "author": { "name": "Paul Fertser", "email": "fercerpav@gmail.com", "time": "Mon Oct 10 11:19:23 2011 +0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:16 2011 -0800" }, "message": "plat-mxc: iomux-v3.h: implicitly enable pull-up/down when that\u0027s desired\n\ncommit 6571534b600b8ca1936ff5630b9e0947f21faf16 upstream.\n\nTo configure pads during the initialisation a set of special constants\nis used, e.g.\n#define MX25_PAD_FEC_MDIO__FEC_MDIO IOMUX_PAD(0x3c4, 0x1cc, 0x10, 0, 0, PAD_CTL_HYS | PAD_CTL_PUS_22K_UP)\n\nThe problem is that no pull-up/down is getting activated unless both\nPAD_CTL_PUE (pull-up enable) and PAD_CTL_PKE (pull/keeper module\nenable) set. This is clearly stated in the i.MX25 datasheet and is\nconfirmed by the measurements on hardware. This leads to some rather\nhard to understand bugs such as misdetecting an absent ethernet PHY (a\nreal bug i had), unstable data transfer etc. This might affect mx25,\nmx35, mx50, mx51 and mx53 SoCs.\n\nIt\u0027s reasonable to expect that if the pullup value is specified, the\nintention was to have it actually active, so we implicitly add the\nneeded bits.\n\nSigned-off-by: Paul Fertser \u003cfercerpav@gmail.com\u003e\nSigned-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9bdb666778b0eb502554288ac3188139c3c45f0e", "tree": "1200b57e2ea573e87a68d2b4f44033e6ceab08f8", "parents": [ "cf541bb3f2532e7493720a208021e43b85156870" ], "author": { "name": "Joerg Roedel", "email": "joerg.roedel@amd.com", "time": "Tue Oct 11 17:41:32 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:09 2011 -0800" }, "message": "iommu/amd: Fix wrong shift direction\n\ncommit fcd0861db1cf4e6ed99f60a815b7b72c2ed36ea4 upstream.\n\nThe shift direction was wrong because the function takes a\npage number and i is the address is the loop.\n\nSigned-off-by: Joerg Roedel \u003cjoerg.roedel@amd.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "cd1ee66bb5156203711560ee73b00e70bb8450b5", "tree": "afa56c9e28806fa34e550dc300f34e3bcba86cbd", "parents": [ "0a85ef94c2af8630deb851e559c9d3bdf5530802" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Sun Oct 30 15:16:08 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:08 2011 -0800" }, "message": "memory leak with RCU_TABLE_FREE\n\ncommit e73b7fffe487c315fd1a4fa22282e3362b440a06 upstream.\n\nThe rcu page table free code uses a couple of bits in the page table\npointer passed to tlb_remove_table to discern the different page table\ntypes. __tlb_remove_table extracts the type with an incorrect mask which\nleads to memory leaks. The correct mask is ((FRAG_MASK \u003c\u003c 4) | FRAG_MASK).\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0a85ef94c2af8630deb851e559c9d3bdf5530802", "tree": "6cfcb6ae212901b27d7ae7ac2253254b0f7cc163", "parents": [ "d69540f85872860468e85ef8004f5580652b3335" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Sun Oct 30 15:16:07 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:08 2011 -0800" }, "message": "user per registers vs. ptrace single stepping\n\ncommit a45aff5285871bf7be1781d9462d3fdbb6c913f9 upstream.\n\ngit commit 5e9a2692 \"[S390] ptrace cleanup\" introduced a regression\nfor the case when both a user PER set (e.g. a storage alteration trace) and\nPTRACE_SINGLESTEP are active. The new code will overrule the user PER set\nwith a instruction-fetch PER set over the whole address space for ptrace\nsingle stepping. The inferior process will be stopped after each instruction\nwith an instruction fetch event. Any other events that may have occurred\nconcurrently are not reported (e.g. storage alteration event) because the\ncontrol bits for them are not set. The solution is to merge the PER control\nbits of the user PER set with the PER_EVENT_IFETCH control bit for\nPTRACE_SINGLESTEP.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "d69540f85872860468e85ef8004f5580652b3335", "tree": "7a7e66ade9a5be5c3a3aa12376a05a12af9e3012", "parents": [ "54a8a620cee91954666080a5525d7f43cad0a693" ], "author": { "name": "Carsten Otte", "email": "cotte@de.ibm.com", "time": "Tue Oct 18 12:27:12 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:36:08 2011 -0800" }, "message": "KVM: s390: check cpu_id prior to using it\n\ncommit 4d47555a80495657161a7e71ec3014ff2021e450 upstream.\n\nWe use the cpu id provided by userspace as array index here. Thus we\nclearly need to check it first. Ooops.\n\nSigned-off-by: Carsten Otte \u003ccotte@de.ibm.com\u003e\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nSigned-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "e747500485ddef175ac6694dcff4fd8088e62071", "tree": "92d35552dfc0930f313bc934f880ea9a86487c09", "parents": [ "c7f65094f7e2b8463a7c002e967e78db8f30468a" ], "author": { "name": "Jan Beulich", "email": "JBeulich@suse.com", "time": "Wed Sep 28 16:44:54 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:35:52 2011 -0800" }, "message": "apic, i386/bigsmp: Fix false warnings regarding logical APIC ID mismatches\n\ncommit 838312be46f3abfbdc175f81c3e54a857994476d upstream.\n\nThese warnings (generally one per CPU) are a result of\ninitializing x86_cpu_to_logical_apicid while apic_default is\nstill in use, but the check in setup_local_APIC() being done\nwhen apic_bigsmp was already used as an override in\ndefault_setup_apic_routing():\n\n Overriding APIC driver with bigsmp\n Enabling APIC mode: Physflat. Using 5 I/O APICs\n ------------[ cut here ]------------\n WARNING: at .../arch/x86/kernel/apic/apic.c:1239\n ...\n CPU 1 irqstacks, hard\u003df1c9a000 soft\u003df1c9c000\n Booting Node 0, Processors #1\n smpboot cpu 1: start_ip \u003d 9e000\n Initializing CPU#1\n ------------[ cut here ]------------\n WARNING: at .../arch/x86/kernel/apic/apic.c:1239\n setup_local_APIC+0x137/0x46b() Hardware name: ...\n CPU1 logical APIC ID: 2 !\u003d 8\n ...\n\nFix this (for the time being, i.e. until\nx86_32_early_logical_apicid() will get removed again, as Tejun\nsays ought to be possible) by overriding the previously stored\nvalues at the point where the APIC driver gets overridden.\n\nv2: Move this and the pre-existing override logic into\n arch/x86/kernel/apic/bigsmp_32.c.\n\nSigned-off-by: Jan Beulich \u003cjbeulich@suse.com\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nLink: http://lkml.kernel.org/r/4E835D16020000780005844C@nat28.tlf.novell.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "2f3f49d0afbeb318b096f09fccaf8747251afccf", "tree": "b96a9eab153a3a713e2d57b90493fd93bceb54e6", "parents": [ "0989d31d57762309445fa1469907b2e6c2cd4ae5" ], "author": { "name": "Josh Stone", "email": "jistone@redhat.com", "time": "Mon Oct 24 10:15:51 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:35:27 2011 -0800" }, "message": "x86: Fix compilation bug in kprobes\u0027 twobyte_is_boostable\n\ncommit 315eb8a2a1b7f335d40ceeeb11b9e067475eb881 upstream.\n\nWhen compiling an i386_defconfig kernel with gcc-4.6.1-9.fc15.i686, I\nnoticed a warning about the asm operand for test_bit in kprobes\u0027\ncan_boost. I discovered that this caused only the first long of\ntwobyte_is_boostable[] to be output.\n\nJakub filed and fixed gcc PR50571 to correct the warning and this output\nissue. But to solve it for less current gcc, we can make kprobes\u0027\ntwobyte_is_boostable[] non-const, and it won\u0027t be optimized out.\n\nBefore:\n\n CC arch/x86/kernel/kprobes.o\n In file included from include/linux/bitops.h:22:0,\n from include/linux/kernel.h:17,\n from [...]/arch/x86/include/asm/percpu.h:44,\n from [...]/arch/x86/include/asm/current.h:5,\n from [...]/arch/x86/include/asm/processor.h:15,\n from [...]/arch/x86/include/asm/atomic.h:6,\n from include/linux/atomic.h:4,\n from include/linux/mutex.h:18,\n from include/linux/notifier.h:13,\n from include/linux/kprobes.h:34,\n from arch/x86/kernel/kprobes.c:43:\n [...]/arch/x86/include/asm/bitops.h: In function ‘can_boost.part.1’:\n [...]/arch/x86/include/asm/bitops.h:319:2: warning: use of memory input\n without lvalue in asm operand 1 is deprecated [enabled by default]\n\n $ objdump -rd arch/x86/kernel/kprobes.o | grep -A1 -w bt\n 551:\t0f a3 05 00 00 00 00 \tbt %eax,0x0\n 554: R_386_32\t.rodata.cst4\n\n $ objdump -s -j .rodata.cst4 -j .data arch/x86/kernel/kprobes.o\n\n arch/x86/kernel/kprobes.o: file format elf32-i386\n\n Contents of section .data:\n 0000 48000000 00000000 00000000 00000000 H...............\n Contents of section .rodata.cst4:\n 0000 4c030000 L...\n\nOnly a single long of twobyte_is_boostable[] is in the object file.\n\nAfter, without the const on twobyte_is_boostable:\n\n $ objdump -rd arch/x86/kernel/kprobes.o | grep -A1 -w bt\n 551:\t0f a3 05 20 00 00 00 \tbt %eax,0x20\n 554: R_386_32\t.data\n\n $ objdump -s -j .rodata.cst4 -j .data arch/x86/kernel/kprobes.o\n\n arch/x86/kernel/kprobes.o: file format elf32-i386\n\n Contents of section .data:\n 0000 48000000 00000000 00000000 00000000 H...............\n 0010 00000000 00000000 00000000 00000000 ................\n 0020 4c030000 0f000200 ffff0000 ffcff0c0 L...............\n 0030 0000ffff 3bbbfff8 03ff2ebb 26bb2e77 ....;.......\u0026..w\n\nNow all 32 bytes are output into .data instead.\n\nSigned-off-by: Josh Stone \u003cjistone@redhat.com\u003e\nCc: Masami Hiramatsu \u003cmasami.hiramatsu.pt@hitachi.com\u003e\nCc: Jakub Jelinek \u003cjakub@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0989d31d57762309445fa1469907b2e6c2cd4ae5", "tree": "6756dbf730e179d9ebfa7db06faaceaef90c2325", "parents": [ "f2c1c3233aa7b1742c458b1eed929d675222e70c" ], "author": { "name": "Jack Steiner", "email": "steiner@sgi.com", "time": "Tue Sep 20 13:55:04 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 11 09:35:26 2011 -0800" }, "message": "x86: uv2: Workaround for UV2 Hub bug (system global address format)\n\ncommit 6a469e4665bc158599de55d64388861d0a9f10f4 upstream.\n\nThis is a workaround for a UV2 hub bug that affects the format of system\nglobal addresses.\n\nThe GRU API for UV2 was inadvertently broken by a hardware change. The\nformat of the physical address used for TLB dropins and for addresses used\nwith instructions running in unmapped mode has changed. This change was\nnot documented and became apparent only when diags failed running on\nsystem simulators.\n\nFor UV1, TLB and GRU instruction physical addresses are identical to\nsocket physical addresses (although high NASID bits must be OR\u0027ed into the\naddress).\n\nFor UV2, socket physical addresses need to be converted. The NODE portion\nof the physical address needs to be shifted so that the low bit is in bit\n39 or bit 40, depending on an MMR value.\n\nIt is not yet clear if this bug will be fixed in a silicon respin. If it\nis fixed, the hub revision will be incremented \u0026 the workaround disabled.\n\nSigned-off-by: Jack Steiner \u003csteiner@sgi.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "83643e5d134408e61504caa4e3e9dd776ddd4207", "tree": "50e31a4b43609ec480915363b5fad876b5da75ca", "parents": [ "f92a292abeee48559493433e4e947d0f49fcdd8d" ], "author": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Sun Oct 23 23:19:12 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Tue Oct 25 07:10:16 2011 +0200" }, "message": "x86: Fix S4 regression\n\ncommit 8548c84da2f47e71bbbe300f55edb768492575f7 upstream.\n\nCommit 4b239f458 (\"x86-64, mm: Put early page table high\") causes a S4\nregression since 2.6.39, namely the machine reboots occasionally at S4\nresume. It doesn\u0027t happen always, overall rate is about 1/20. But,\nlike other bugs, once when this happens, it continues to happen.\n\nThis patch fixes the problem by essentially reverting the memory\nassignment in the older way.\n\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\nCc: Rafael J. Wysocki \u003crjw@sisk.pl\u003e\nCc: Yinghai Lu \u003cyinghai.lu@oracle.com\u003e\n[ We\u0027ll hopefully find the real fix, but that\u0027s too late for 3.1 now ]\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9d3aaf6229361652a8ea6fe8a2fe359e055b43b9", "tree": "fe5e792226548270df57aa5da1eaea72058f4372", "parents": [ "1289deb9b5c151876feb594cd82984f3932982e8" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Mon Oct 03 18:30:53 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Tue Oct 25 07:10:13 2011 +0200" }, "message": "ARM: 7117/1: perf: fix HW_CACHE_* events on Cortex-A9\n\ncommit 29a541f6c1f6e4a85628bb86071b9e72c9f8be2c upstream.\n\nUsing COHERENT_LINE_{MISS,HIT} for cache misses and references\nrespectively is completely wrong. Instead, use the L1D events which\nare a better and more useful approximation despite ignoring instruction\ntraffic.\n\nReported-by: Alasdair Grant \u003calasdair.grant@arm.com\u003e\nReported-by: Matt Horsnell \u003cmatt.horsnell@arm.com\u003e\nReported-by: Michael Williams \u003cmichael.williams@arm.com\u003e\nCc: Jean Pihet \u003cj-pihet@ti.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1289deb9b5c151876feb594cd82984f3932982e8", "tree": "1b591cdac9159f4ddf4c4ae062e7a44e5b41d10f", "parents": [ "62bf7928bfe312ec6cf829b06893eb5649ceaec5" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Thu Sep 29 09:37:23 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Tue Oct 25 07:10:13 2011 +0200" }, "message": "ARM: 7113/1: mm: Align bank start to MAX_ORDER_NR_PAGES\n\ncommit 002ea9eefec98dada56fd5f8e432a4e8570c2a26 upstream.\n\nThe VM subsystem assumes that there are valid memmap entries from\nthe bank start aligned to MAX_ORDER_NR_PAGES.\n\nOn the Ux500 we have a lot of mem\u003dN arguments on the commandline\ntriggering this bug several times over and causing kernel\noops messages.\n\nCc: Michael Bohan \u003cmbohan@codeaurora.org\u003e\nCc: Nicolas Pitre \u003cnico@fluxnic.net\u003e\nSigned-off-by: Johan Palsson \u003cjohan.palsson@stericsson.com\u003e\nSigned-off-by: Rabin Vincent \u003crabin.vincent@stericsson.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "e35c99bc50ef74c02244d6e6d40868f9a49a59be", "tree": "2ec2a8bbae33db7a65cdfec3f7c100722e40c48c", "parents": [ "a1164796d1ed6196c037429d796db5ffdb2674e3" ], "author": { "name": "Rafael J. Wysocki", "email": "rjw@sisk.pl", "time": "Thu Jun 02 21:06:48 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sun Oct 16 14:14:55 2011 -0700" }, "message": "MIPS: PM: Use struct syscore_ops instead of sysdevs for PM (v2)\n\ncommit bd7100099a46b59f433dd15ad60adbb4d4f3d625 upstream.\n\nConvert some MIPS architecture\u0027s code to using struct syscore_ops\nobjects for power management instead of sysdev classes and sysdevs.\n\nThis simplifies the code and reduces the kernel\u0027s memory footprint.\nIt also is necessary for removing sysdevs from the kernel entirely in\nthe future.\n\nSigned-off-by: Rafael J. Wysocki \u003crjw@sisk.pl\u003e\nAcked-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\nAcked-and-tested-by: Lars-Peter Clausen \u003clars@metafoo.de\u003e\nSigned-off-by: Lars-Peter Clausen \u003clars@metafoo.de\u003e\nCc: linux-kernel@vger.kernel.org\nCc: \"Rafael J. Wysocki\" \u003crjw@sisk.pl\u003e\nPatchwork: http://patchwork.linux-mips.org/patch/2431/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "33ab02d45f2f238b72c8fb9735b32c58ee24fa73", "tree": "bd11cdf064acccaa5621e138748490c0b6b46ce1", "parents": [ "2b7eea63de50d738ae12a1bf84b76ef91c007a0e" ], "author": { "name": "srinidhi kasagar", "email": "srinidhi.kasagar@stericsson.com", "time": "Tue Sep 20 11:15:46 2011 +0530" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sun Oct 16 14:14:54 2011 -0700" }, "message": "ARM: mach-ux500: enable fix for ARM errata 754322\n\ncommit 98e87d57aab9b1594f9cc53a386fcb6f2f2ba6e2 upstream.\n\nThis applies ARM errata fix 754322 for all ux500 platforms.\n\nSigned-off-by: srinidhi kasagar \u003csrinidhi.kasagar@stericsson.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "36f1ce1fa2ee9e2d2608ca2629b8b1232b14a1dc", "tree": "99718a875a21547f248b84edb9056ea49bb32446", "parents": [ "8a60d75bbc5feca1291edd728d6696be9e8dd465" ], "author": { "name": "Paul Menzel", "email": "paulepanter@users.sourceforge.net", "time": "Wed Aug 31 17:07:10 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sun Oct 16 14:14:53 2011 -0700" }, "message": "x86/PCI: use host bridge _CRS info on ASUS M2V-MX SE\n\ncommit 29cf7a30f8a0ce4af2406d93d5a332099be26923 upstream.\n\nIn summary, this DMI quirk uses the _CRS info by default for the ASUS\nM2V-MX SE by turning on `pci\u003duse_crs` and is similar to the quirk\nadded by commit 2491762cfb47 (\"x86/PCI: use host bridge _CRS info on\nASRock ALiveSATA2-GLAN\") whose commit message should be read for further\ninformation.\n\nSince commit 3e3da00c01d0 (\"x86/pci: AMD one chain system to use pci\nread out res\") Linux gives the following oops:\n\n parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE]\n HDA Intel 0000:20:01.0: PCI INT A -\u003e GSI 17 (level, low) -\u003e IRQ 17\n HDA Intel 0000:20:01.0: setting latency timer to 64\n BUG: unable to handle kernel paging request at ffffc90011c08000\n IP: [\u003cffffffffa0578402\u003e] azx_probe+0x3ad/0x86b [snd_hda_intel]\n PGD 13781a067 PUD 13781b067 PMD 1300ba067 PTE 800000fd00000173\n Oops: 0009 [#1] SMP\n last sysfs file: /sys/module/snd_pcm/initstate\n CPU 0\n Modules linked in: snd_hda_intel(+) snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event tpm_tis tpm snd_seq tpm_bios psmouse parport_pc snd_timer snd_seq_device parport processor evdev snd i2c_viapro thermal_sys amd64_edac_mod k8temp i2c_core soundcore shpchp pcspkr serio_raw asus_atk0110 pci_hotplug edac_core button snd_page_alloc edac_mce_amd ext3 jbd mbcache sha256_generic cryptd aes_x86_64 aes_generic cbc dm_crypt dm_mod raid1 md_mod usbhid hid sg sd_mod crc_t10dif sr_mod cdrom ata_generic uhci_hcd sata_via pata_via libata ehci_hcd usbcore scsi_mod via_rhine mii nls_base [last unloaded: scsi_wait_scan]\n Pid: 1153, comm: work_for_cpu Not tainted 2.6.37-1-amd64 #1 M2V-MX SE/System Product Name\n RIP: 0010:[\u003cffffffffa0578402\u003e] [\u003cffffffffa0578402\u003e] azx_probe+0x3ad/0x86b [snd_hda_intel]\n RSP: 0018:ffff88013153fe50 EFLAGS: 00010286\n RAX: ffffc90011c08000 RBX: ffff88013029ec00 RCX: 0000000000000006\n RDX: 0000000000000000 RSI: 0000000000000246 RDI: 0000000000000246\n RBP: ffff88013341d000 R08: 0000000000000000 R09: 0000000000000040\n R10: 0000000000000286 R11: 0000000000003731 R12: ffff88013029c400\n R13: 0000000000000000 R14: 0000000000000000 R15: ffff88013341d090\n FS: 0000000000000000(0000) GS:ffff8800bfc00000(0000) knlGS:00000000f7610ab0\n CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b\n CR2: ffffc90011c08000 CR3: 0000000132f57000 CR4: 00000000000006f0\n DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400\n Process work_for_cpu (pid: 1153, threadinfo ffff88013153e000, task ffff8801303c86c0)\n Stack:\n 0000000000000005 ffffffff8123ad65 00000000000136c0 ffff88013029c400\n ffff8801303c8998 ffff88013341d000 ffff88013341d090 ffff8801322d9dc8\n ffff88013341d208 0000000000000000 0000000000000000 ffffffff811ad232\n Call Trace:\n [\u003cffffffff8123ad65\u003e] ? __pm_runtime_set_status+0x162/0x186\n [\u003cffffffff811ad232\u003e] ? local_pci_probe+0x49/0x92\n [\u003cffffffff8105afc5\u003e] ? do_work_for_cpu+0x0/0x1b\n [\u003cffffffff8105afc5\u003e] ? do_work_for_cpu+0x0/0x1b\n [\u003cffffffff8105afd0\u003e] ? do_work_for_cpu+0xb/0x1b\n [\u003cffffffff8105fd3f\u003e] ? kthread+0x7a/0x82\n [\u003cffffffff8100a824\u003e] ? kernel_thread_helper+0x4/0x10\n [\u003cffffffff8105fcc5\u003e] ? kthread+0x0/0x82\n [\u003cffffffff8100a820\u003e] ? kernel_thread_helper+0x0/0x10\n Code: f4 01 00 00 ef 31 f6 48 89 df e8 29 dd ff ff 85 c0 0f 88 2b 03 00 00 48 89 ef e8 b4 39 c3 e0 8b 7b 40 e8 fc 9d b1 e0 48 8b 43 38 \u003c66\u003e 8b 10 66 89 14 24 8b 43 14 83 e8 03 83 f8 01 77 32 31 d2 be\n RIP [\u003cffffffffa0578402\u003e] azx_probe+0x3ad/0x86b [snd_hda_intel]\n RSP \u003cffff88013153fe50\u003e\n CR2: ffffc90011c08000\n ---[ end trace 8d1f3ebc136437fd ]---\n\nTrusting the ACPI _CRS information (`pci\u003duse_crs`) fixes this problem.\n\n $ dmesg | grep -i crs # with the quirk\n PCI: Using host bridge windows from ACPI; if necessary, use \"pci\u003dnocrs\" and report a bug\n\nThe match has to be against the DMI board entries though since the vendor entries are not populated.\n\n DMI: System manufacturer System Product Name/M2V-MX SE, BIOS 0304 10/30/2007\n\nThis quirk should be removed when `pci\u003duse_crs` is enabled for machines\nfrom 2006 or earlier or some other solution is implemented.\n\nUsing coreboot [1] with this board the problem does not exist but this\nquirk also does not affect it either. To be safe though the check is\ntightened to only take effect when the BIOS from American Megatrends is\nused.\n\n 15:13 \u003c ruik\u003e but coreboot does not need that\n 15:13 \u003c ruik\u003e because i have there only one root bus\n 15:13 \u003c ruik\u003e the audio is behind a bridge\n\n $ sudo dmidecode\n BIOS Information\n Vendor: American Megatrends Inc.\n Version: 0304\n Release Date: 10/30/2007\n\n[1] http://www.coreboot.org/\n\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d30552\n\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: H. Peter Anvin \u003chpa@zytor.com\u003e\nCc: x86@kernel.org\nSigned-off-by: Paul Menzel \u003cpaulepanter@users.sourceforge.net\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nAcked-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ca64baea4dd045516046a84a8860c3f9809b8984", "tree": "f39657b72df9082586401e9e321fa6d42cfa305b", "parents": [ "a004e0962a10dfa7fc83dfa4ed4109d1cf84124b" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 29 12:18:59 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sun Oct 16 14:14:51 2011 -0700" }, "message": "sparc64: Force the execute bit in OpenFirmware\u0027s translation entries.\n\nIn the OF \u0027translations\u0027 property, the template TTEs in the mappings\nnever specify the executable bit. This is the case even though some\nof these mappings are for OF\u0027s code segment.\n\nTherefore, we need to force the execute bit on in every mapping.\n\nThis problem can only really trigger on Niagara/sun4v machines and the\nhistory behind this is a little complicated.\n\nPrevious to sun4v, the sun4u TTE entries lacked a hardware execute\npermission bit. So OF didn\u0027t have to ever worry about setting\nanything to handle executable pages. Any valid TTE loaded into the\nI-TLB would be respected by the chip.\n\nBut sun4v Niagara chips have a real hardware enforced executable bit\nin their TTEs. So it has to be set or else the I-TLB throws an\ninstruction access exception with type code 6 (protection violation).\n\nWe\u0027ve been extremely fortunate to not get bitten by this in the past.\n\nThe best I can tell is that the OF\u0027s mappings for it\u0027s executable code\nwere mapped using permanent locked mappings on sun4v in the past.\nTherefore, the fact that we didn\u0027t have the exec bit set in the OF\ntranslations we would use did not matter in practice.\n\nThanks to Greg Onufer for helping me track this down.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "28d5b74edcc0819cefa66256fd804c8420aff19f", "tree": "1f184e320f55982fab0fd6d953502744fb7dfd82", "parents": [ "85fd323003d5fe4d5c798688f016ab0eda0c9dcf" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Fri Sep 23 14:34:12 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:41:06 2011 -0700" }, "message": "ARM: 7099/1: futex: preserve oldval in SMP __futex_atomic_op\n\ncommit df77abcafc8dc881b6c9347548651777088e4b27 upstream.\n\nThe SMP implementation of __futex_atomic_op clobbers oldval with the\nstatus flag from the exclusive store. This causes it to always read as\nzero when performing the FUTEX_OP_CMP_* operation.\n\nThis patch updates the ARM __futex_atomic_op implementations to take a\ntmp argument, allowing us to store the strex status flag without\noverwriting the register containing oldval.\n\nReported-by: Minho Ban \u003cmhban@samsung.com\u003e\nReviewed-by: Nicolas Pitre \u003cnicolas.pitre@linaro.org\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "85fd323003d5fe4d5c798688f016ab0eda0c9dcf", "tree": "4917cc663c26b87a79413f3607efe29eb9b14d9a", "parents": [ "017a4b549759497c802c12c5982cf06f93806e60" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Sep 15 11:45:15 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:41:06 2011 -0700" }, "message": "ARM: 7091/1: errata: D-cache line maintenance operation by MVA may not succeed\n\ncommit f630c1bdfbf8fe423325beaf60027cfc7fd7c610 upstream.\n\nThis patch implements a workaround for erratum 764369 affecting\nCortex-A9 MPCore with two or more processors (all current revisions).\nUnder certain timing circumstances, a data cache line maintenance\noperation by MVA targeting an Inner Shareable memory region may fail to\nproceed up to either the Point of Coherency or to the Point of\nUnification of the system. This workaround adds a DSB instruction before\nthe relevant cache maintenance functions and sets a specific bit in the\ndiagnostic control register of the SCU.\n\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nTested-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "017a4b549759497c802c12c5982cf06f93806e60", "tree": "55e9627a7c51f19ae4c1d7a22f4e9162e18b94ee", "parents": [ "a0dbac4607ee6d97aba53f0d676acbdd416c1214" ], "author": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Thu Sep 22 10:32:25 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:41:06 2011 -0700" }, "message": "ARM: dma-mapping: free allocated page if unable to map\n\ncommit d8e89b47e00ee80e920761145144640aac4cf71a upstream.\n\nIf the attempt to map a page for DMA fails (eg, because we\u0027re out of\nmapping space) then we must not hold on to the page we allocated for\nDMA - doing so will result in a memory leak.\n\nReported-by: Bryan Phillippe \u003cbp@darkforest.org\u003e\nTested-by: Bryan Phillippe \u003cbp@darkforest.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "3b576ff8b0e8a70be60a0ee8aa7d637ba27db2c8", "tree": "b91ee712b1e4cc3df719d8528ead50e2a72ccc51", "parents": [ "e670d472aaaa51f93a5a49de00dc573eea8a8486" ], "author": { "name": "Youquan Song", "email": "youquan.song@intel.com", "time": "Tue Aug 02 14:01:35 2011 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:40:49 2011 -0700" }, "message": "perf, x86: Add model 45 SandyBridge support\n\ncommit a34668f6beb4ab01e07683276d6a24bab6c175e0 upstream.\n\nAdd support to Romely-EP SandyBridge.\n\nSigned-off-by: Youquan Song \u003cyouquan.song@intel.com\u003e\nSigned-off-by: Anhua Xu \u003canhua.xu@intel.com\u003e\nSigned-off-by: Lin Ming \u003cming.m.lin@intel.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nLink: http://lkml.kernel.org/r/1312264895-2010-1-git-send-email-youquan.song@intel.com\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "310fef97e2c781acbc6603f010c5e14e0e90913b", "tree": "0c5419b0f7e4cd6280c6e0451e154ea1441fbd9a", "parents": [ "0208b80455ff8d4a1d3df826098fbeb0dcf6dc05" ], "author": { "name": "David Vrabel", "email": "david.vrabel@citrix.com", "time": "Tue Sep 13 10:17:32 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:40:39 2011 -0700" }, "message": "xen/e820: if there is no dom0_mem\u003d, don\u0027t tweak extra_pages.\n\ncommit e3b73c4a25e9a5705b4ef28b91676caf01f9bc9f upstream.\n\nThe patch \"xen: use maximum reservation to limit amount of usable RAM\"\n(d312ae878b6aed3912e1acaaf5d0b2a9d08a4f11) breaks machines that\ndo not use \u0027dom0_mem\u003d\u0027 argument with:\n\nreserve RAM buffer: 000000133f2e2000 - 000000133fffffff\n(XEN) mm.c:4976:d0 Global bit is set to kernel page fffff8117e\n(XEN) domain_crash_sync called from entry.S\n(XEN) Domain 0 (vcpu#0) crashed on cpu#0:\n...\n\nThe reason being that the last E820 entry is created using the\n\u0027extra_pages\u0027 (which is based on how many pages have been freed).\nThe mentioned git commit sets the initial value of \u0027extra_pages\u0027\nusing a hypercall which returns the number of pages (if dom0_mem\nhas been used) or -1 otherwise. If the later we return with\nMAX_DOMAIN_PAGES as basis for calculation:\n\n return min(max_pages, MAX_DOMAIN_PAGES);\n\nand use it:\n\n extra_limit \u003d xen_get_max_pages();\n if (extra_limit \u003e\u003d max_pfn)\n extra_pages \u003d extra_limit - max_pfn;\n else\n extra_pages \u003d 0;\n\nwhich means we end up with extra_pages \u003d 128GB in PFNs (33554432)\n- 8GB in PFNs (2097152, on this specific box, can be larger or smaller),\nand then we add that value to the E820 making it:\n\n Xen: 00000000ff000000 - 0000000100000000 (reserved)\n Xen: 0000000100000000 - 000000133f2e2000 (usable)\n\nwhich is clearly wrong. It should look as so:\n\n Xen: 00000000ff000000 - 0000000100000000 (reserved)\n Xen: 0000000100000000 - 000000027fbda000 (usable)\n\nNaturally this problem does not present itself if dom0_mem\u003dmax:X\nis used.\n\nSigned-off-by: David Vrabel \u003cdavid.vrabel@citrix.com\u003e\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0208b80455ff8d4a1d3df826098fbeb0dcf6dc05", "tree": "790871bd7ab75e71389c0259d38df263666b231b", "parents": [ "e2d598ab82b589fc40c4656a10004e328a90b6dd" ], "author": { "name": "David Vrabel", "email": "david.vrabel@citrix.com", "time": "Fri Aug 19 15:57:16 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:40:39 2011 -0700" }, "message": "xen: use maximum reservation to limit amount of usable RAM\n\ncommit d312ae878b6aed3912e1acaaf5d0b2a9d08a4f11 upstream.\n\nUse the domain\u0027s maximum reservation to limit the amount of extra RAM\nfor the memory balloon. This reduces the size of the pages tables and\nthe amount of reserved low memory (which defaults to about 1/32 of the\ntotal RAM).\n\nOn a system with 8 GiB of RAM with the domain limited to 1 GiB the\nkernel reports:\n\nBefore:\n\nMemory: 627792k/4472000k available\n\nAfter:\n\nMemory: 549740k/11132224k available\n\nA increase of about 76 MiB (~1.5% of the unused 7 GiB). The reserved\nlow memory is also reduced from 253 MiB to 32 MiB. The total\nadditional usable RAM is 329 MiB.\n\nFor dom0, this requires at patch to Xen (\u0027x86: use \u0027dom0_mem\u0027 to limit\nthe number of pages for dom0\u0027) (c/s 23790)\n\nSigned-off-by: David Vrabel \u003cdavid.vrabel@citrix.com\u003e\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "b7290a21efb472272306d64686efa523f7e93bf6", "tree": "b41c45626e601ea2c0fe1e0f6cd7e64fd4513fca", "parents": [ "562960c731b9ffba8cd200be8d38f5f793b23d74" ], "author": { "name": "Nicolas Pitre", "email": "nicolas.pitre@linaro.org", "time": "Wed Sep 14 01:22:05 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:40:36 2011 -0700" }, "message": "ARM: Dove: fix second SPI initialization call\n\ncommit 72cc205611879525db0374d9831f84f787112b25 upstream.\n\nCommit 980f9f601a \"ARM: orion: Consolidate SPI initialization.\"\nbroke it by overwriting the SPI0 registration.\n\nSigned-off-by: Nicolas Pitre \u003cnicolas.pitre@linaro.org\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "054b93a673c2b39106b02cbef05bfdc3f5ba6150", "tree": "e086e31d8bb267eb6195adf2882657eb1d177612", "parents": [ "5c755fc21c8e84aff1230e27646fe4277a19451d" ], "author": { "name": "Joerg Roedel", "email": "joerg.roedel@amd.com", "time": "Fri Sep 02 14:10:32 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Oct 03 11:40:20 2011 -0700" }, "message": "iommu/amd: Make sure iommu-\u003eneed_sync contains correct value\n\ncommit f1ca1512e765337a7c09eb875eedef8ea4e07654 upstream.\n\nThe value is only set to true but never set back to false,\nwhich causes to many completion-wait commands to be sent to\nhardware. Fix it with this patch.\n\nSigned-off-by: Joerg Roedel \u003cjoerg.roedel@amd.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" } ], "next": "5c755fc21c8e84aff1230e27646fe4277a19451d" }